(Oxford Textbooks in Psychiatry) Tom Dening, Alan Thomas-Oxford Textbook of Old Age Psychiatry-Oxford University Press (2013) PDF

Oxford Textbook of
Old Age Psychiatry

Oxford Textbooks in Psychiatry
Oxford Textbook of Community Mental Health

Edited by Graham Thornicroft, George Szmukler, Kim T. Mueser, and Robert E. Drake
Oxford Textbook of Suicidology and Suicide Prevention
Edited by Danuta Wasserman and Camilla Wasserman
Oxford Textbook of Women and Mental Health
Edited by Dora Kohen
Oxford Textbook of
Old Age
Psychiatry
SECOND EDITION
Edited by
Tom Dening
Institute of Mental Health, University of Nottingham, Nottingham, UK
Alan Thomas
Institute for Ageing and Health, Newcastle University,
Newcastle upon Tyne, UK
1
3
Great Clarendon Street, Oxford, OX2 6DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© Oxford University Press 2013
The moral rights of the authors have been asserted
First Edition published in 2008
Second Edition published in 2013
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2013936563
ISBN 978–0–19–964495–7
Printed in China by
C&C Offset Printing Co. Ltd
Oxford University press makes no representation, express or implied, that the
drug dosages in this book are correct. Readers must therefore always check
the product information and clinical procedures with the most up-to-date
published product information and data sheets provided by the manufacturers
and the most recent codes of conduct and safety regulations. The authors and
the publishers do not accept responsibility or legal liability for any errors in the
text or for the misuse or misapplication of material in this work. Except where
otherwise stated, drug dosages and recommendations are for the non-pregnant
adult who is not breast-feeding
To our families, with love.
Karen, Lizzy, Jon, and Alex;
and Shona, Elspeth, Douglas, and Bethan.
This page intentionally left blank
Preface
Although this is the second edition of the Oxford Textbook of Old another. Somewhat surprisingly, in countries such as the UK that
Age Psychiatry, it is also the fifth edition of the book that originally have specialists in old age psychiatry, there may paradoxically be
appeared as Psychiatry in the Elderly in 1991. It is the first edition not threats to the existence of the speciality at a time of ever-increasing
to have been edited by Robin Jacoby and Catherine Oppenheimer, demand. Nonetheless, the core of knowledge about the science of
whose brilliant conception 25 years ago of the need for this book and practice in relation to mental disorders in later life is robust and
has stood the test of time. Both of us were privileged to work along- continues to improve. It is the latest version of this canon that we
side Robin and Catherine in producing the last edition, which we are proud to share with you in this volume.
hope has allowed for handover and continuity as well as evolution- Whilst maintaining the tried and tested book structure, in this
ary change. It has nonetheless been daunting and exciting for us to edition all chapters have again been thoroughly revised and updated
take responsibility for what we believe is the leading textbook in and we have benefited from an increase in the number of authors to
our speciality across the world. 96, of whom 59 have written for this book for the first time. These
One of the central features of ‘Jacoby and Oppenheimer’ has new contributors have brought fresh perspectives, helped perhaps
always been its clear structure, leading from basic science (both by them bringing a larger breadth of international experience (rep-
neurobiological and social varieties), through clinical practice, spe- resenting Australia, Austria, Canada, China, France, Germany,
cific disorders, and service provision, to a collection of ethical and Ireland, the Netherlands, Norway, Sweden, Switzerland, the UK,
legal chapters on some of the hard challenges for old age psychia- and the US). Developments have led us to add some wholly new
trists. We think this structure was visionary, and we have retained chapters (on palliative care, ethics of caring, and of living and dying
it as we believe that old age psychiatrists and those in related dis- with dementia) and to ask for expansions and larger changes in
ciplines will appreciate the breadth of information and scholarship the chapters on brain stimulation therapies (rather than just ECT),
that is contained within it. memory clinics (now more broadly known as memory services),
Old age psychiatry has arisen because of the particular health and capacity (now widened to all mental capacity and decision-
and social issues that surround ageing, dementia, comorbidity, making from the narrower focus before on testamentary capacity).
dependency, and the end of life in ageing societies across the world. We are grateful to all our contributors, especially those who have
The global burden and cost of dementia is enormous; so too is that shared their personal experiences. It has been a pleasure working
of depression and other mental illnesses. Dementia has become a with you all. We would also like to thank our editors, Martin Baum
major public policy issue in many countries across the world, so and Charlotte Green, and Pete Stevenson and Eloise Moir-Ford, at
our ability to recognize the condition and to offer appropriate and Oxford University Press for their support and diligence.
affordable help and support is already really important and will
only become more so. However, the way in which mental health Tom Dening, Nottingham
services are organized for older people varies from one country to Alan Thomas, Newcastle upon Tyne
Acknowledgements
T.D. acknowledges the support of the NIHR Cambridgeshire A.T. expresses thanks to the NIHR Biomedical Research Centre in
and Peterborough CLAHRC (Collaboration for Leadership in Ageing and the Biomedical Research Unit in Lewy Body Dementia
Applied Health Research and Care) and the Cambridgeshire and at Newcastle University and to Gateshead Health NHS Foundation
Peterborough NHS Foundation Trust. Also thanks to Nottingham Trust for their support.
University, Nottinghamshire Healthcare NHS Trust, and Barchester
Healthcare for their encouragement.
Contents
Contributors xv 12 Neuroimaging 177

Claire E. Sexton, Verena Heise, and Klaus P. Ebmeier
1 Biological aspects of human ageing 1
Doug Gray, Carole Proctor, and Tom Kirkwood 13 Psychopharmacology in older people 191
Craig W. Ritchie
2 Psychometric assessment in older people 13
Karen Ritchie 14 Brain stimulation therapies 199
Daniel W. O’Connor and Chris Plakiotis
3 The sociology of ageing 23
Ricca Edmondson 15 Person- and relationship-centred dementia
4 Transforming concepts of ageing: three
care: past, present, and future 213
case studies from anthropology 39

John Keady and Mike Nolan
Sharon R. Kaufman, Julie Livingston, 16 Psychological treatments 229
Hong Zhang, and Margaret Lock Philip Wilkinson
5 Epidemiology of old age psychiatry: 17 Cognitive behaviour therapy 233
an overview of concepts and main Philip Wilkinson
studies 57
18 Interpersonal psychotherapy 243
Thais Minett, Blossom Stephan, and Carol Brayne
Philip Wilkinson
6 Neuropathology 87
19 Psychodynamic psychotherapy 247
Johannes Attems and Kurt A. Jellinger
Jane Garner
7 Neurochemical pathology of dementia 107
20 Family therapy 259
Margaret Ann Piggott
Jane Pearce
8 Molecular genetics and biology
21 Nonpharmacological interventions
of dementia 123
in care homes 269
Denise Harold and Julie Williams
Ian A. James and Jane Fossey
9 Psychiatric assessment of older people 141
22 Principles of service provision
Alan Thomas
in old age psychiatry 283
10 Clinical cognitive assessment 149 Tom Dening
Christopher Kipps and John Hodges
23 Primary care management of older people
11 Physical assessment of older patients 159 with mental health problems 301
Duncan Forsyth Louise Robinson and Carolyn Chew-Graham
xii contents
24 Memory assessment services 319 40 Delirium 529

Sube Banerjee Jenny Hogg
25 Liaison old age psychiatry 325 41 The experience of depression 543
John Holmes Judith Boast
26 Social care 335 42 Depression in older people 545
Jo Moriarty Alan Thomas
27 Care homes for older people 343 43 Suicide and attempted suicide in older people 571
Tom Dening and Alisoun Milne Helen Chiu and Joshua Tsoh
28 Palliative care and end of life care 359 44 Manic syndromes in old age 581
Elizabeth Sampson and Karen Harrison Dening Akshya Vasudev
29 The concept of dementia 371 45 Anxiety disorders in older people 589
Introduction 371 Gerard Byrne
Alan Thomas and Tom Dening
Part 1 Dementia: What Is It and Can We Diagnose It? 371 46 Late-onset schizophrenia 603
Peter Whitehouse and Danny George Sarah Brunelle, Ipsit V. Vahia, and Dilip V. Jeste
Part 2 A Brief History of the Concept of Dementia 374
47 Personal experience of lifelong illness 621
Alexander F. Kurz and Nicola T. Lautenschlager
Part 3 The DSM-5 Approach to Dementia 377 Anonymous in collaboration with Sue Green
Perminder S. Sachdev 48 Severe and enduring mental illness 623
Part 4 Should We Diagnose Subtypes of Dementia? 381 Catherine Hatfield and Tom Dening
Robert Stewart
49 Alcohol and substance abuse in
30 Hello, I’m me! Living well with dementia 385
older people 631
June and Brian Hennell
Henry O’Connell and Brian Lawlor
31 Epidemiology of dementia 389
50 Older people with learning disabilities 653
Laura Fratiglioni and Chengxuan Qiu
Maria Luisa Hanney
32 Mild Cognitive Impairment and
51 Sleep disorders 667
predementia syndromes 415
Urs Peter Mosimann and Bradley F. Boeve
John T. O’Brien and Louise Grayson
52 The effect of ageing on personality 677
33 Alzheimer’s disease 431
Bob Woods and Gill Windle
John-Paul Taylor and Alan Thomas
53 Personality in later life: personality disorder
34 Vascular and mixed dementias 457
and the effects of illness on personality 691
Robert Stewart
Catherine Oppenheimer
35 Dementia with Lewy bodies and
54 Sexuality in later life 703
Parkinson’s disease dementia 469
Walter Pierre Bouman
Arvid Rongve and Dag Aarsland
55 Ethics and old age psychiatry 725
36 Frontotemporal dementia 479
Julian C. Hughes
Vincent Deramecourt, Florence
Lebert, and Florence Pasquier 56 Mental capacity and decision-making 745
Julian C. Hughes and Christopher Heginbotham
37 Neurological dementias 491
Andrew Graham 57 Ethics of living and dying with dementia 761
Cees Hertogh and Jenny van der Steen
38 Pharmacological treatment of dementia 503
Roy W. Jones 58 The ethics of caring 769
Clive Baldwin and Brandi Estey-Burtt
39 Management of dementia 513
Sarah Cullum
contents xiii
59 Elder abuse 779 63 The law relating to mental capacity

Jill Manthorpe and mental health 815
Kay Wheat
60 Crime, mental illness, and older people 785
Graeme Yorston
Index 823
61 Testamentary capacity 797
Robin Jacoby
62 Driving and psychiatric illness in later life 805
Desmond O’Neill
Contributors
Editors Carol Brayne

Director, Institute of Public Health and Professor of Public Health
Tom Dening
Medicine, Cambridge Institute of Public Health, University of
Barchester and Nottinghamshire Healthcare Professor of Dementia
Cambridge, Cambridge, UK
Research, Institute of Mental Health, University of Nottingham,
UK. Formerly Consultant Psychiatrist, Cambridgeshire and Sarah Brunelle
Peterborough NHS Foundation Trust, Cambridge, UK Geriatric Psychiatry Fellow, University of California, San Diego,
USA
Alan Thomas
Professor in Old Age Psychiatry, Institute for Ageing and Health, Gerard J. Byrne
Newcastle University, Newcastle upon Tyne, UK Professor of Psychiatry, Academic Discipline of Psychiatry,
University of Queensland, St Lucia, Queensland, Australia
Contributors Caroline Chew Graham
Dag Aarsland Professor of General Practice Research, Research Institute,
Research Director, Centre for Age-Related Medicine, Stavanger Primary Care and Health Sciences, Keele University, Keele, UK
University Hospital, Stavanger, Norway
Helen F.K. Chiu
Vasu Akshya Professor of Psychiatry, Department of Psychiatry, Faculty of
Assistant Professor of Geriatric Psychiatry and Medicine, Medicine, Chinese University of Hong Kong, Hong Kong,
Western University, London, Ontario, Canada. Retired, China
Cambridge, UK
Sarah Cullum
Johannes Attems Consultant Psychiatrist, Avon and Wiltshire Mental Health
Professor of Neuropathology, Institute for Ageing and Health, Partnership NHS Trust, and Honorary Senior Clinical Lecturer,
Newcastle University, Newcastle upon Tyne, UK Academic Unit of Psychiatry, School of Social and Community
Medicine, University of Bristol, Bristol, UK
Clive Baldwin
Canada Research Chair in Narrative Studies School of Social Work, Vincent Deramecourt
St Thomas University, Fredericton, New Brunswick, Canada Neurologist and Assistant Professor of Histology, University Lille-
Nord de France, Lille, France
Sube Banerjee
Professor of Dementia & Associate Dean, Brighton and Sussex Klaus Ebmeier
Medical School, University of Sussex, Brighton Professor of Old Age Psychiatry, University of Oxford, Oxford, UK
Judith Boast Ricca Edmondson
Retired, Cambridge, UK Professor, School of Political Science and Sociology, National
University of Ireland, Galway, UK
Bradley Boeve
Professor of Neurology, Department of Neurology and Center Brandi Estey-Burtt
for Sleep Medicine College of Medicine, Mayo Clinic, Rochester, Research Assistant, St Thomas University, Fredericton, New
Minnesota, USA Brunswick, Canada
Walter Bouman Duncan Forsyth
Consultant Psychiatrist-Sexologist, Nottingham Gender Clinic, Consultant Geriatrician, Cambridge University Hospitals NHS
Nottingham, UK Foundation Trust, Cambridge, UK
xvi contributors
Jane Fossey John Holmes

Associate Director of Psychological Services, Oxford Health NHS Senior Lecturer in Liaison Psychiatry of Old Age, University of
Foundation Trust, Oxford, UK Leeds, Leeds Institute of Health Sciences, Leeds, UK
Laura Fratiglioni Julian Hughes
Professor of Epidemiology, Karolinska Institutet and Stockholm Consultant in Old Age Psychiatry and Honorary Professor of
Gerontology Research Center, Stockholm, Sweden Philosophy of Ageing, Northumbria Healthcare NHS Foundation
Trust and Institute for Ageing and Health, Newcastle upon Tyne,
Jane Garner UK
Consultant Psychiatrist, University College London, London UK.
Former Consultant NHS Old Age Psychiatrist. Robin Jacoby
Emeritus Professor of Old Age Psychiatry, University of Oxford,
Daniel R. George Oxford, UK
Assistant Professor, Penn State College of Medicine, Hershey,
Pennsylvania, USA Ian James
Consultant Clinical Psychologist, Northumberland Tyne and Wear
Andrew Graham NHS FT, Visiting Professor Northumbria University, Campus for
Consultant Neurologist, Addenbrooke’s Hospital, Cambridge, UK Ageing and Vitality, Newcastle, UK
Doug Gray Kurt Jellinger
Professor of Ageing Science, Institute for Ageing and Health, Professor of Neuropathology, Institute for Clinical Neurobiology,
Newcastle University, Newcastle upon Tyne, UK Vienna, Austria
Louise Grayson Dilip V. Jeste
Specialty Registrar in Old Age Psychiatry, Institute for Ageing and Estelle and Edgar Levi Chair in Aging and Director, Sam and Rose
Health, Newcastle University, Newcastle upon Tyne, UK Stein Institute for Research on Aging, University of California, San
Sue Green Diego, USA
Consultant in Old Age Psychiatry, Cambridgeshire and Roy W. Jones
Peterborough NHS Foundation Trust, Cambridge, UK Director, RICE (The Research Institute for the Care of
Marissa Hanney Older People), Royal United Hospital; Honorary Consultant
Consultant in Old Age Psychiatry for People with Learning Geriatrician; and Honorary Professor of Clinical Gerontology,
Disabilities, Northumberland, Tyne and Wear NHS Foundation University of Bath, Bath, UK
Trust, Newcastle upon Tyne, UK Sharon R. Kaufman
Denise Harold Professor of Medical Anthropology, University of California, San
Research Associate, Cardiff University MRC Centre for Francisco, USA
Neuropsychiatric Genetics and Genomics, Institute for John Keady
Psychological Medicine and Clinical Neurosciences, Cardiff, UK Professor of Mental Health Nursing and Older People, University
Karen Harrison Dening of Manchester/Greater Manchester West Mental Health NHS
Head of Admiral Nursing, Dementia UK, UK Foundation Trust, Manchester, UK
Catherine Hatfield Christopher Kipps

Consultant Psychiatrist, Cambridgeshire and Peterborough NHS Consultant Neurologist and Honorary Clinical Senior Lecturer,
Foundation Trust, Cambridge, UK Wessex Neurological Centre, University Hospitals Southampton
NHS Foundation Trust, Southampton, UK
Christopher Heginbotham
Professor of Mental Health Policy and Management, University of Tom Kirkwood
Central Lancashire, and Institute of Clinical Education, Warwick Associate Dean for Ageing, Institute for Ageing and Health
University, Medical School, Wawick, UK Newcastle University, Newcastle upon Tyne, UK
Verena Heise Alexander Kurz

DPhil-Student, Department of Psychiatry, University of Oxford, Professor of Psychiatry, Department of Psychiatry and
Oxford, UK Psychotherapy, Klinikum rechts der Isar, Technische Universität
München, Munich, Germany
Brian and June Hennell
Gloucestershire, UK Nicola Lautenschlager
Professor of Psychiatry of Old Age, Academic Unit for Psychiatry
Cees Hertogh of Old Age, St. Vincent’s Health, Department of Psychiatry,
Professor of Elderly Care Medicine and Geriatric Ethics, VU University of Melbourne, Melbourne, Australia
University Medical Center, Amsterdam, The Netherlands
Brian Lawlor
John Hodges Conolly Norman Professor of Old Age Psychiatry, Trinity College
Federation Fellow and Professor of Cognitive Neurology, Dublin, Dublin, Ireland
Neuroscience Research Australia, Randwick, New South Wales,
Australia Florence Lebert
Clinical Psychiatrist, The Memory Department, University
Jenny Hogg Hospital of Lille, Lille, France
Consultant Geriatrician, Gateshead Health NHS Foundation
Trust, Gateshead, UK
contributors xvii
Julie Livingston Carole Proctor

Associate Professor of History, History Department, Rutgers the Senior Research Associate, Institute for Ageing and Health,
State University of New Jersey, New Brunswick, USA Newcastle University, Newcastle upon Tyne, UK
Margaret Lock Chengxuan Qiu
Marjorie Bronfman Professor in Social Studies in Medicine, Associate Professor of Epidemiology, Karolinska Institutet,
Emerita, Department of Social Studies of Medicine, McGill Stockholm, Sweden
University, Montreal, Quebec, Canada
Craig Ritchie
Jill Manthorpe Senior Lecturer in Old Age Psychiatry, Centre for Mental Health,
Professor of Social Work, Director of the Social Care Workforce Imperial College London, London, UK
Research Unit, Associate Director of NIHR School for Social Care
Research, King’s College London, London, UK Karen Ritchie
Research Director and Professor of Public Health and
Alisoun Milne Epidemiology, French National Institute of Medical Research
Reader in Social Gerontology and Social Work, University of Kent, (INSERM), University of Montpellier, Montpellier, France;
Chatham Maritime, UK Imperial College London, London, UK
Thais Minett Louise Robinson
Senior Research Associate, Cambridge Institute of Public Health, Professor of Primary Care and Ageing, University of Newcastle,
University of Cambridge, Cambridge, UK Newcastle upon Tyne, UK
Jo Moriarty Arvid Rongve
Research Fellow, Social Care Workforce Research Unit, King’s Consultant in Old Age Psychiatry and Post-Doctoral Research
College London, London, UK and Fellow of the National Institute Fellow, Department of Psychiatry, Haugesund Hospital,
for Health Research (NIHR) School for Social Care Research Haugesund, Norway
Urs Peter Mosimann Perminder Sachdev
Professor of Old Age Psychiatry, Department of Old Age Professor of Neuropsychiatry and Co-Director, CHeBA (Centre
Psychiatry, University of Bern, Bern, Switzerland for Healthy Brain Ageing) School of Psychiatry, UNSW Medicine,
University of New South Wales, Australia and Clinical Director,
Mike Nolan Neuropsychiatric Institute (NPI), Prince of Wales Hospital,
Professor of Gerontological Nursing, School of Nursing and Randwick, Australia
Midwifery, University of Sheffield, Sheffield, UK
Elizabeth Sampson,
John O’Brien Clinical Senior Lecturer, Marie Curie Palliative Care Research
Professor of Old Age Psychiatry, Department of Psychiatry, Unit, London, UK
University of Cambridge, Cambridge, UK
Claire E. Sexton
Henry O’Connell Post-doctoral Research Fellow, Department of Psychiatry,
Consultant Psychiatrist, Laois-Offaly Mental Health Services, and University of Oxford, Oxford, UK
Adjunct Senior Clinical Lecturer, Graduate Entry Medical School,
University of Limerick, Limerick, Ireland Blossom Stephan
Lecturer, University of Newcastle, Newcastle upon Tyne, UK
Daniel O’Connor
Professor of Old Age Psychiatry, Monash University, Victoria, Rob Stewart
Australia Professor of Psychiatric Epidemiology and Clinical Informatics,
King’s College London (Institute of Psychiatry), London, UK
Desmond O’Neill
Professor of Medical Gerontology, Trinity College Dublin, John Paul Taylor
Ireland Wellcome Intermediate Clinical Fellow and Honorary Consultant
in Old Age Psychiatry, Institute for Ageing and Health, Newcastle
Catherine Oppenheimer University, Newcastle upon Tyne, UK
Former Consultant in Old Age Psychiatry, Oxford, UK
Joshua Tsoh
Florence Pasquier Consultant (Psychiatry), Department of Psychiatry, Prince of
Professor of Neurology, University Hospital of Lille, Wales Hospital, New Territories, Hong Kong, China
Lille, France
Ipsit Vahia
Jane Pearce Assistant Professor, Department of Psychiatry, Stein Institute for
Honorary Consultant in Psychological Services, Oxford Health Research on Aging, University of California, San Diego, USA
NHS Foundation Trust, Oxford, UK
Jenny van der Steen
Margaret Ann Piggott Epidemiologist, VU University Medical Center, Amsterdam, The
Researcher, Institute for Ageing and Health, Newcastle University, Netherlands
Newcastle upon Tyne, UK
Akshya Vasudev
Chris Plakiotis Assistant Professor of Geriatric Psychiatry and Medicine
Research Fellow, Aged Mental Health Research Unit, Monash University of Western Ontario and Associate Scientist Lawson
University, and Consultant Psychiatrist, MonashHealth, Victoria, Research Institute, Victoria Hospital, London Health Sciences
Australia Centre, London, UK
xviii contributors
Kay Wheat Gill Windle

Reader in Law, Nottingham Trent University, Nottingham, UK Senior Research Fellow, Dementia Services Development Centre
Wales, Bangor University, Bangor, UK
Peter Whitehouse
Professor of Neurology and Director, University Hospitals Case Bob Woods
Medical Center, Adult Learning, Cleveland, USA Professor of Clinical Psychology of the Elderly, Dementia Services
Development Centre Wales, Bangor University, Bangor, UK
Philip Wilkinson
Oxford Health R&D Research Fellow, Honorary Senior Clinical Graeme Yorston
Lecturer, Consultant Psychiatrist, Department of Psychiatry, Visiting Professor of Ageing and Mental Health and Consultant
University of Oxford, Warneford Hospital, Oxford, UK Old Age Forensic Psychiatrist, Staffordshire University and St
Andrew’s Healthcare, Stoke-on-Trent, UK
Julie Williams
Professor of Neuropsychiatric Genetics, Cardiff University MRC Hong Zhang
Centre for Neuropsychiatric Genetics and Genomics, Institute for Associate Professor of East Asian Studies, Colby College,
Psychological Medicine and Clinical Neurosciences, Cardiff, UK Waterville, Maine, USA
CHAPTER 1
Biological aspects of
human ageing
Doug Gray, Carole Proctor, and Tom Kirkwood
The persistence of childhood memories in the mind of a centenar- is a voracious consumer of oxygen and a prodigious producer of
ian is evidence enough of the remarkable durability of the brain ROS. The brain contains sufficient quantities of metal ions (notably
at the tissue, cellular, and molecular level, but it is beyond dispute those of iron) to ensure the production of reactive hydroxyl ani-
that the brain of the most cognitively favoured centenarian will ons through Fenton chemistry (the thermodynamically favoured
have suffered structural decline. For organisms in possession of process which through oxidation of metal ions promotes decom-
a nervous system such decline does not appear to be inevitable; position of hydrogen peroxide into hydroxyl anions and hydroxyl
though its nervous system is undoubtedly simpler the freshwater radicals), and in addition contains high levels of neurotransmitters
cnidarian Hydra is able to escape such decline, and barring mis- such as dopamine, norepinephrine, and serotonin whose metal
fortune is potentially immortal (Martinez, 1998). What separates ion-catalysed oxidation can promote a vicious cycle of toxicity and
Hydra from more familiar metazoans is not the simplicity of its further ROS release (Halliwell, 2006). The damage to biological
nervous system but rather its departure from the scheme of having molecules by ROS is not discriminating and will affect proteins,
specialized germline versus somatic cells. If (as in Hydra) all our nucleic acids, carbohydrates, and lipids. Without discounting the
cells had the qualities of multipotent stem cells, there would be no importance of other molecular targets of ROS damage, the empha-
need for regenerative medicine. In humans, most cells (including sis of the current chapter will primarily be on proteins and DNA,
neurons) are terminally differentiated somatic cells. Stem cells are and the organelles and pathways most dependent on the continued
rare and, for reproductive success, only the germ cells must be kept integrity of these molecules (see Fig. 1.2).
in pristine condition. Kirkwood’s ‘disposable soma’ theory posits
that with only finite resources available to an organism there must Molecular Aspects of Ageing in the Brain
be a trade-off between the resources allocated to reproduction and
the resources allocated to repair (Kirkwood, 1977). Any reduction Protein damage
in repair will be accompanied by the accumulation of molecular To appreciate the consequences of oxidative damage to proteins
damage, and it is the accumulation of damage that drives ageing and DNA within the ageing brain one must first consider the very
(Fig. 1.1). The inherent stochasticity of damage is sufficient to different fates of these entities. For the most part proteins have
explain the lack of uniformity in ageing in genetically identical limited half-lives and are constantly being replaced (an interesting
organisms (discussed in Kirkwood, 2008), and the inevitability exception being crystallin proteins within the lens; lens crystallins
of decline in the face of unrepaired molecular damage precludes are permanent and as such can be used for accurate carbon-dating
the requirement for any programme ‘for’ ageing (other arguments of individuals born during the era of atmospheric atomic tests!
have been marshalled towards the conclusion that no such pro- (Lynnerup et al., 2008)). DNA, on the other hand, must persist
gramme is required, or indeed has been demonstrated (Kirkwood for the lifetime of the cell (temporarily leaving aside the more
and Melov, 2011)). specialized case of mitochondrial DNA). Though neurons, being
The structural decline of the brain that is a hallmark of human postmitotic, do not require pristine templates for DNA replicative
ageing arises from unrepaired stochastic damage to its compo- purposes, the integrity of the genome is required for their active
nents, and it is the purpose of this chapter to deal with such dam- programme of ongoing transcription. It is therefore not surprising
age at the molecular and cellular levels. It is widely accepted that the that damaged proteins are generally eliminated from mammalian
source of much damage in the ageing brain is reactive oxygen spe- cells (again, excepting damaged lens crystallins, whose destiny
cies (ROS), a collective term for oxygen radicals and other entities is the lens cataract) whereas damaged nuclear DNA is actively
that cause oxidative damage (Halliwell, 2006). There is no escap- repaired.
ing ROS—it constitutes a necessary by-product of cellular metabo- There exist two cellular systems for the degradation of proteins,
lism and a necessary intermediate of a number of cell signalling one involving membrane-bound vesicular trafficking of sub-
cascades (Nathan, 2003). Because of its energy demands, the brain strates and the other providing access to substrates throughout the
2 Oxford Textbook of Old Age Psychiatry
for the degradation of proteins, and without the requirement for

Young neuron
membrane trafficking can operate on a much more rapid times-
cale. The UPS has remarkable specificity, discriminating between
short-lived and long-lived proteins, post-translationally modified
Tolerable ROS production versus unmodified proteins, and normally structured versus aber-
from pristine mitochondria Proteolytic reserve
rant and/or damaged proteins. This specificity is mediated by more
than a thousand ubiquitin ligase (E3) enzymes that in concert with
a lesser number of ubiquitin conjugating (E2) enzymes associ-
ate with particular proteins and assemble upon them a chain of
ubiquitin proteins (Hochstrasser, 1996). Once the chain reaches a
Efficient protein degradation
threshold length of four ubiquitin monomers the chain has physical
affinity for subunits of the 26S proteasome (Piotrowski et al., 1997),
and delivery of the substrate to this degradation machine results in
Efficient DNA repair
its proteolytic destruction.
The hallmark of age-related neurodegenerative diseases is the
Aged neuron appearance of protein conglomerates within the cytoplasm or
nucleus of neurons, and in some cases glia. These entities are of
Mitochondrial dysfunction
sufficient size to be visible by light microscopy, and in general are
designated inclusion bodies (Woulfe, 2007), though more spe-
elevated ROS production
cific designations may be associated with particular diseases (for
example, the Lewy body in Parkinson’s disease and Lewy body
dementia). The composition of the inclusion body may vary from
Protein
Aggregation one neuropathological state to the next, but there is often a major
Telomere constituent that typifies the disease and is its aetiologic agent (for
damage example, α-synuclein, in the case of the Lewy body (Eriksen et al.,
Proteasome 2003)). Most inclusion bodies are immunoreactive for ubiquitin,
inhibition
which may signify the failure of the UPS to eliminate their con-
stituent proteins (Gray et al., 2003). Indeed, the most parsimoni-
Loss of proteolytic reserve Inefficient DNA repair ous explanation for inclusion bodies is the passive agglomeration
Fig. 1.1 Molecular damage affects multiple systems in ageing neurons. of aberrant proteins which, as the result of damage or misfolding,
expose hydrophobic regions that would normally be concealed
within structural folds. While this may be true for nuclear inclu-
Random sions the situation in the cytoplasm may be more complex. There is
molecular evidence that proteins are actively transported along microtubules
damage
to aggresomes (cytoplasmic inclusions assembled in the vicinity
of the centriole) (Johnston et al., 1998). It has been suggested that
Lewy bodies may represent a form of aggresome (McNaught et al.,
Cellular defects
2002), though the behaviour and composition of aggresomes in cell
culture models are not in complete agreement with findings from
autopsy specimens (Waxman et al., 2009), and questions remain
about the relationship of these entities.
Tissue dysfunction Inclusion bodies can be detected in the brains of normal aged
individuals (those not diagnosed with neurodegenerative dis-
ease), and it is reasonable to suggest that their presence denotes
the age-related decline of proteolytic efficiency. Examples of such
inclusions would include the rodlike and skeinlike inclusions of the
AGEING aged neostriatum, the granular cytoplasmic inclusions of the infe-
rior olivary nucleus, and the colloid or hyaline inclusions of the
Fig. 1.2 Molecular damage underlies ageing. hypoglossal and spinal motor neurons of the elderly (discussed in
Gray et al., 2003). Though the protein constituents of these inclu-
sions are at present poorly characterized, their age-related deposi-
cytoplasm and nucleus. The autophagic system is reliant on the tion points to the existence of a ‘tipping point’ beyond which the
engulfment of cellular components into membrane delimited com- rate of accumulation of abnormal protein exceeds the capacity for
partments which upon fusion with lysosomes become autophago- its elimination, resulting in widespread protein aggregation and
somes. Through autophagy substrates are typically degraded on a the ultimate accretion of aggregates into inclusion bodies. In the
timescale of hours, but the system has the capacity to process entire absence of heritable DNA mutations the source of the abnormal
organelles (oxidatively damaged mitochondria, for example) and protein would be stochastic damage and misfolding from intrinsic
degrade all of the macromolecules therein. The ubiquitin/protea- ROS production or ROS from environmental sources, coupled with
some system (UPS), on the other hand, is a more specialized system the incessant production of misfolded protein from translational
chapter  biological aspects of human ageing 3
and post-translational errors (accounting for as much as a third of Accrual Removal

of protein of protein
total protein production by some estimates (Schubert et al., 2000). damage Protein
damage UPS
In the disease state, inclusions would occur precociously due to the aggregation
added burden of abnormal protein, either genetically encoded or of ROS UPS
idiopathic origin. The tipping point conjecture is surely consistent
ROS
with the observation that the symptoms of pernicious inherited dis-
eases (Huntington’s disease, for example) do not become apparent
until adulthood is attained, and often not until late in life. The cor- Healthy/young cell Stressed/old cell
relation (Gusella and MacDonald, 2006) between disease onset in
Huntington’s disease and the length of the expanded polyglutamine Fig. 1.3 Loss of balance in protein homeostasis typifies ageing at the cellular level.
tract in the affected protein also fits with the tipping point conjec-
ture; the longer the repeat the greater the burden to the UPS, and
the earlier it is likely to be overwhelmed. The greater prevalence of pharmacological or genetic inhibition of p38 (Tang et al., 2010), but
inclusions within neurons as opposed to glia of Huntington’s patients not other MAP kinases (Tsirigotis et al., 2008).
is also in agreement with the conjecture, or is at least in agreement For purely physical reasons the formation of an inclusion body
with findings from cell culture model systems of these cell types. would seem a potentially catastrophic event, and the long associa-
UPS activity was found to be lower in cultured mouse neurons than tion of inclusion bodies with neuropathological conditions has cast
in cultured glia, and shows a more pronounced age-related decline them in an understandably negative light, but there is a body of
(Tydlacka et al., 2008). Mutant huntingtin protein was efficiently work that would argue the converse: under some conditions inclu-
degraded in cultured glia unless proteasome activity was pharma- sion formation appears to correlate with neuronal survival. The
cologically inhibited (Tydlacka et al., 2008). best evidence for the neuroprotective effect of inclusions comes
The functional basis of UPS decline in ageing individuals has from live cell imaging of neurons made to express fluorescent ver-
yet to be fully elucidated. That UPS efficiency does decline is evi- sions of an aggregation-prone protein. In these experiments it is
dent from the accumulation of UPS substrates in mouse brain possible to track the fate of individual cells and, in so doing, to
lysates, typically visualized by probing western blots of such sam- document the enhanced survival of neurons that formed inclusions
ples with an antibody specific for ubiquitin. What is observed is early (Arrasate et al., 2004). The supposition (now widely accepted)
an age-dependent increase in the intensity of a high-molecular is that smaller order aggregates (oligomers) are the more toxic enti-
weight smear (Ohtsuka et al., 1995; Gray et al., 2003), also observed ties, perhaps because of their propensity to bind promiscuously to
in the brains of human Huntington’s disease patients and mouse essential cellular proteins. By sequestering such dangerous enti-
models thereof (Bennett et al., 2007). This smear arises from ubiq- ties into inclusions, their potential for promiscuous interactions
uitin chains linked to a range of protein substrates (Bennett et al., is diminished and neuroprotection is hence conferred. While the
2007). The proteolytic deficit that generates this backlog of undi- beneficial effect of inclusions in the cell culture model system is
gested UPS substrates could arise from the age-related decline in indisputable, the timeframe of such experiments is necessarily
the expression of some key UPS component or components, and short, and may not reflect the fate of neurons laden with inclusions
microarray studies (Lee et al., 2000; Blalock et al., 2003) have iden- in situ. Inclusions are known to concentrate iron, and may serve as
tified candidate UPS components whose levels were decreased in centres of Fenton-mediated ROS production (Firdaus et al., 2006).
aged mouse brains (at least at the RNA level). Proteasome activity The long-term effects of localized ROS production around inclu-
has been shown to decline in the ageing brain (Keller et al., 2002), sions are unlikely to be beneficial.
but this decline has not been strongly correlated with reduction of A final, rather terrifying aspect of the aberrant proteins and pro-
any of the proteasome subunits and may reflect damage to the protein aggregates that accumulate in the ageing brain is their potential
teasome rather than the loss of expression. to spread to adjacent and/or distant cells and wreak havoc in a pri-
The prospect of proteasome damage raises an interesting ‘chicken on-like manner. The spreading or ‘seeding’ of protein aggregates has
and egg’ problem. Does a growing burden of protein damage within been the subject of much recent interest, and has been discussed in
the brain ultimately overwhelm the capacity of the proteasome, the context of a growing list of aggregation-prone proteins includ-
or does age-related damage to the proteasome itself (from ROS ing β-amyloid, α-synuclein, tau, huntingtin, superoxide dismutase,
or other sources) eventually generate the backlog of undigested fused in sarcoma gene (FUS), and transactivation-responsive (TAR)
substrates? It is conceivable, perhaps likely, that a vicious cycle of DNA-binding protein 43 (TDP-43) (Brundin et al., 2011; Jucker
inhibitions exists, and regardless of the entry point the outcome and Walker, 2011; Polymenidou and Cleveland, 2011; Dunning
will be deleterious for the cells in which the vicious cycle occurs et al., 2012). Aggregates may be actively secreted through exocy-
(Fig. 1.3). Support for such an inhibitory cycle comes from time tosis into the extracellular milieu or within membrane delimited
lapse imaging and stochastic computer modelling of cells express- exosomes that could travel through the vasculature to distant sites.
ing aggregation-prone proteins (Tang et al., 2010), wherein recip- Aggregates may also be released passively during cell death and
rocal inhibition is apparently mediated by ROS. This study also be taken up by endocytic pathways of nearby cells, or may travel
highlighted the importance of the stress-activated p38 MAP kinase through tunnelling nanotubes to access adjacent cells, as has been
molecule in mediating the death of inclusion-bearing cells. Because documented for prions (Gousset et al., 2009). Aggregate seeding
it can be activated by oxidative conditions (via the redox-sensitive has been proposed as a possible mechanism for the propagation
upstream ASK1 kinase (Matsuzawa and Ichijo, 2008)), p38 MAPK of pathology in a number of age-related conditions in which local-
would be responsive to the operation of the proposed vicious cycle, ized neurodegeneration becomes more widespread. In Alzheimer’s
and in the experimental system cell survival can be enhanced by disease (AD) the tau tangles stereotypically occur first in the locus
ceruleus and transentorhinal area, then spread to the amygdala and isoform being the most recognized genetic risk factor for late-on-
connected cortical regions (Braak and Braak, 1991). In Parkinson’s set AD (Saunders et al., 1993). Conversely, the rarer E2 isoform is
disease there is a predictable spread of pathology through anatomi- associated with a protective role (Corder et al., 1994) and the most
cally connected regions of the brainstem, limbic system, and neo- frequent E3 isoform shows an intermediate effect in terms of both
cortex, but it has been hypothesized that this pathology is preceded disease risk and age of onset.
by degenerative changes in olfactory structures and the enteric There are many model systems to address the role of genetics in
plexus, perhaps due to exposure of these tissues to an environmen- ageing, including nematode worms (Caenorhabditis elegans), fruit
tal agent such as a virus (Jang et al., 2009). If with ageing there is a flies (Drosophila melanogaster), mouse, and yeast. These species are
relentless loss of protein degradative capacity as described above, short-lived and so it is feasible to measure the effects of different
it would follow that the aggregate ‘seeds’ would find increasingly genetic strains on maximum and average lifespan. The first ‘longev-
fertile soil in the neurons of the aged brain. ity gene’ found in C. elegans was AGE-1 (Friedman and Johnson,
1988) and since then approximately 750 genes have been discov-
DNA damage ered which affect lifespan in C. elegans. A set of genes which have
It is beyond the scope of this chapter to summarize the extensive attracted a lot of attention are those involved in an insulin-like sig-
literature on age-related damage to the DNA genome (a subject nalling pathway and which regulate the activity of the transcription
that merits book-length treatment on its own (Vijg, 2007)), but the factor DAF-16. Genes involved in insulin-signalling are also known
extensive reliance of DNA repair systems on ubiquitin-mediated to affect the lifespan of D. melanogaster and mice. In humans, a var-
proteolysis leads to the necessary conclusion that age-related decline iant at the FOXO3A locus has been linked with longevity (reviewed
of the latter will ultimately compromise the former. Ubiquitin is in Ziv and Hu, 2011). Another signalling pathway that has recently
employed in some forms of DNA repair in a non-proteolytic role as been shown to have effects on ageing is the mTOR pathway that
a scaffold or molecular switch (Chiu et al., 2006), but in postmitotic affects many cellular functions including nutrient sensing, protein
cells such as neurons it is the proteolytic function of the UPS in synthesis, and autophagy (Hands et al., 2009; Zoncu et al., 2011).
nucleotide excision repair (NER) that is of paramount importance Mutations that decrease TOR activity have been shown to extend
(Nouspikel, 2011). In particular, it is the transcription-coupled lifespan in C. elegans, D. melanogaster, yeast, and mice (Jia et al.,
repair (TCR) system, a subsystem of NER, that is of greatest rel- 2004; Kapahi et al., 2004; Powers et al., 2006; Selman et al., 2009).
evance. TCR is charged with maintaining the integrity of the tran- A role for epigenetics in ageing has recently gained in importance
scribed strand of DNA; for metabolically active cells such as neurons (Tollefsbol and SpringerLink (Online service), 2010). It involves
a lesion within a coding region can block transcription with disas- changes in gene expression which are not due to changes in the DNA
trous consequences and such lesions must be detected and rapidly sequence. Examples of epigenetic changes are DNA methylation
repaired. A second form of NER (global genome repair, or GGR) or histone deacetylation, both of which suppress gene expression.
maintains genomic integrity at sites throughout the genome, but for Epigenetic changes are essential during development and differen-
postmitotic cells is much less critical. The TCR and GGR pathways tiation but can also occur later in life in response to changes in the
have shared components and components that are unique, but both cellular environment or as a result of random errors. These changes
sense distortions of the double helix and ultimately utilize com- are passed on when cells divide but are generally not passed on
mon ubiquitin-mediated degradation machinery (based on cullin through the germline. DNA methylation is the most studied epige-
4A complexes) to remove multicomponent repair systems from the netic modification and is the main topic of a recent review (Gravina
repaired lesion. As a last resort, stalled RNA polymerase II can itself and Vijg, 2010). Approximately 40% of promoters in mammalian
be degraded by the ubiquitination machinery of the TCR system genes contain clusters of unmethylated CpG pairs, known as CpG
to ‘reboot’ gene transcription. The importance of TCR to neuronal islands. These islands are targets for transcription factors. During
homeostasis was suggested by the neuronal deficits of individuals ageing, changes in methylation status of CpG islands occur which
bearing mutations in TCR-related genes, and has been confirmed can lead to either hypermethylation or hypomethylation. A study
experimentally by the age-related neurodegenerative phenotypes of of monozygotic twins revealed that epigenetic differences increased
mice engineered to be deficient in TCR components (Jaarsma et al., with age (Fraga et al., 2005). This suggests that environmental fac-
2011). As the efficiency of ubiquitin-mediated proteolysis declines, tors and lifestyle may play a role and many studies have examined
so too should the efficiency of TCR. Should TCR efficiency fall the effects of diet, for example a deficiency in folate has been linked
below a critical threshold the subsequent shutdown of transcription to hypomethylation (Friso and Choi, 2002). A link between epige-
could only imperil the functioning of the ageing brain. netics and the senescence pathway (discussed in section Cellular
mechanisms of ageing) has received recent attention (Simboeck
Ageing at the Cellular and Systems Level et al., 2011).
Genetics and epigenetics Cellular Mechanisms of Ageing

There is strong evidence for the role of genetics in age-related
diseases especially cardiovascular diseases, diabetes, AD, and Mitochondria
Parkinson’s disease (Ruse and Parker, 2001). A systematic review Mitochondria are the powerhouse of the cell, producing most of
of published family, twin, linkage, and association studies found the cell’s supply of adenosine triphosphate (ATP). They also pro-
evidence for the role of several genes in successful ageing (Glatt duce ROS as a by-product of respiration which can damage cel-
et al., 2007). These genes included APOE, GSST1, IL6, IL10, PON1, lular components. Mitochondria are themselves particularly prone
and SIRT3. The role of APOE in AD is well known. It exists as to damage since they are close to the source of ROS. Mitochondria
three different alleles: APOE2, APOE3, and APOE4, with the E4 contain DNA (mtDNA) which is circular and codes for 37 genes.
However, they also need to import mitochondrial proteins which in human tissues (Cortopassi and Arnheim, 1990; Hayakawa et al.,
are encoded by nuclear DNA. The number of mitochondria within 1991, 1992). Conversely, targeting catalase to mitochondria led to
a cell depends on the type of tissue but is typically in the region of lower levels of mtDNA mutations and extended lifespan (Schriner
100–1,000. It was once thought that mitochondria are static solitary et al., 2005). There is evidence for an accumulation of mitochon-
organelles. However, it is now known that they are very dynamic in drial DNA (mtDNA) mutations with age (Wallace, 1999). In partic-
terms of movement through the cell and are involved in turnover of ular, high levels of deleted mtDNA were found in substantia nigra
both the organelle and mtDNA and the formation of mitochondrial neurons from both aged controls and individuals with Parkinson’s
networks. Movement of mitochondria is particularly important in disease (Bender et al., 2006; Kraytsberg et al., 2006). These mtDNA
neurons since they are required at synapses and so they are continu- mutations are somatic, with different clonally expanded deletions
ally transported from the cell body to axonal synapses (anterograde in individual cells, and high levels of these mutations are associ-
transport) and back again (retrograde transport) for degradation. ated with respiratory chain deficiency. Respiratory activity has
There is evidence for impaired movement in age-related neuro- been shown to be decreased with age in several human tissues, for
degenerative diseases. A mouse model of Huntington’s disease example in liver (Yen et al., 1989) and skeletal muscle (Trounce
reported impairment in both anterograde and retrograde transport et al., 1989; Conley et al., 2000). In addition, it has been shown that
in cells in specific neurons which are known to be susceptible to cell accelerating the mtDNA mutation rate can result in premature age-
death in the human disease (Her and Goldstein, 2008). ing. For example, mice with a mutant POLG, the gene that encodes
In AD it has been shown that amyloid beta impairs antero- the only mtDNA polymerase, prematurely exhibited many phe-
grade transport resulting in degeneration of synapses (Calkins notypes associated with human ageing and had shorter lifespans
and Reddy, 2011). Furthermore, the mitochondrial protein Pink1 (Trifunovic et al., 2004; Kujoth et al., 2005). Age-related increases
which is mutated in some familial forms of Parkinson’s disease has in the frequency of cytochrome c oxidase (COX)-deficient cells,
been shown to play a role in mitochondrial transport (Weihofen which are associated with mtDNA mutation, have been reported
et al., 2009). Mitochondria are degraded by lysosomes by a proc- in human muscle (Muller-Hocker, 1989; Brierley et al., 1998), brain
ess known as mitophagy (mitochondrial autophagy). For many (Cottrell et al., 2000, 2001), and gut (Taylor et al., 2003). Cells in
years this was considered a random process but there is increas- which mtDNA mutation reaches a high level are likely to suffer
ing evidence that mitophagy is selective (reviewed by Kim et al., from impaired ATP production, resulting in a decline in tissue
2007). For example, it has been shown that Pink1 is recruited to bioenergenesis. However, there is also controversy about the role
mitochondria with Parkin (an E3 ubiquitin ligase) and induces of mitochondrial damage in ageing and some studies in mice show
mitophagy (Kawajiri et al., 2010). Since lysosomal degradation that there is no effect of mitochondrial antioxidants on lifespan
pathways decrease in efficiency with age, there is also impaired deg- (reviewed by Jang and Remmen, 2009). Interestingly, although the
radation of mitochondria which could lead to increase in the levels POLG mice model demonstrated that an increase in mitochondrial
of dysfunctional mitochondria. Degradation of mitochondria must mutations was associated with a variety of ageing phenotypes the
be balanced by biogenesis of new mitochondria in order to main- mice did not show any signs of increased oxidative stress (Trifunovic
tain mitochondrial numbers. Mitochondria also need to replicate et al., 2005). This puts into question the idea that oxidative stress
when cells divide. Replication occurs via mitochondrial binary fis- plays a major role in the ageing process. A study that counters the
sion. The mtDNA is also replicated and randomly distributed to mitochondrial theory of ageing is a long-lived mouse, heterozygous
the two daughter mitochondria. Mitochondrial biogenesis has been for CLK1 (CDC-like kinase 1), a mitochondrial enzyme necessary
shown to be diminished with age (Fannin et al., 1999). In addition, for ubiquinone biosynthesis, despite having massive mitochondrial
PGC-1a, an intracellular mediator of biogenesis has been shown to dysfunction starting early in life (Lapointe and Hekimi, 2008).
be decreased in age-related diseases such as diabetes and neurode- Therefore the link between mitochondria and ageing is complex. In
generation (reviewed by Wenz, 2011). Mitochondria form networks particular, mitochondria cannot be considered alone as they affect
and continuously undergo fission and fusion whereby an individual many other cellular mechanisms. For example, a recent study sug-
mitochondrion either leaves or joins the network respectively. The gests interplay between mitochondrial and proteasome activity in
purpose of fission and fusion is still unclear but the general consen- skin ageing (Koziel et al., 2011).
sus is that it allows exchange of mtDNA and mitochondrial proteins
and helps to segregate damage prior to fission and degradation. In Telomeres
neurons, fission and fusion of mitochondria takes place in the cell Telomeres are repetitive sequences of noncoding DNA that pro-
body and also along axons and has been widely studied using live tect the ends of chromosomes from degradation and end-to-end
cell imaging (Jahani-Asl et al., 2007; Twig et al., 2008). Disturbances fusions. In humans telomeres end with a stretch of single stranded
in the balance of fission and fusion, resulting in abnormally long or DNA, known as G-overhangs which form a T-loop structure. This
short mitochondria, have been implicated in age-related neurode- structure is bound and stabilized by proteins to form a so-called
generation (Knott and Bossy-Wetzel, 2008). cap to distinguish it from a DNA break. Telomeres shorten with
The importance of mitochondria in ageing was highlighted each cell division due to the end-replication problem (Olovnikov,
in 1991 by the proposal of the mitochondrial theory of ageing 1971) and also oxidative stress which has been shown to acceler-
(Miquel, 1991), an extension of the free radical theory (Harman, ate telomere shortening by causing single strand breaks in telom-
1956), being based on the idea that ROS generated by mitochon- eric DNA (von Zglinicki, 2002). Further evidence for the role of
dria induce mutations in mtDNA which leads to more ROS and a oxidative stress is that the addition of antioxidants to cell cultures
vicious cycle ensues. There is a wealth of evidence to support the slows telomere shortening (Serra et al., 2003). Telomeres need a
mitochondrial theory of ageing. For example, studies have shown minimum length in order to maintain their capped structure; how-
that an excess production of ROS leads to more mtDNA damage ever, the exact threshold required is not known. Short uncapped
telomeres are seen by the cell as DNA damage leading to induc- p53 and cell cycle arrest CDK inhibitor (CDKN1A), also known as
tion of the DNA damage response. This can result in permanent p21. The role of p53 in initiating cell cycle arrest (and also apopto-
cell cycle arrest, also known as cellular or replicative senescence, sis) has been clearly shown in a p53 mutant mouse model where
which is discussed in more detail in section Cellular senescence. enhanced p53 activity led to reduced incidence of cancer but early
As telomere lengths vary considerably between different chromo- age-associated pathology (Tyner et al., 2002).
some arms, it would be expected that a small subset of telomeres It has been shown that ROS are involved in the establishment
which are the shortest would be the most important in trigger- and maintenance of senescence (Passos et al., 2010). ROS accelerate
ing cell arrest (Hemann et al., 2001). However, other data indicate telomere shortening (von Zglinicki, 2002) and can directly dam-
that the average telomere length is a better predictor of replicative age DNA, leading to induction of the DNA damage response and
senescence (Martens et al., 2000). cell arrest. It was recently shown that the DNA damage response
Telomerase is an enzyme that elongates telomeres. In humans it is triggers mitochondrial dysfunction leading to higher ROS pro-
active in most embryonic tissues during early development. However, duction through a signalling pathway involving p53, p21, p38,
it is down-regulated during cellular differentiation in most somatic and Tgfβ (Passos et al., 2010). In addition this study showed that
cells apart from endothelial cells and some adult stem cells. Therefore ROS contribute in a stochastic manner to the long-term mainte-
telomeres shorten with age in most human tissues and organs nance of DNA damage foci. Senescent cells differentially express
(Djojosubroto et al., 2003). Telomere shortening was not observed hundreds of genes (Shelton et al., 1999); most prominent are genes
in the brain during ageing in a study which analysed whole-organ involved in the proinflammatory response (Coppe et al., 2008)
biopsies (Allsopp et al., 1995); however, this does not preclude the and marker genes for a retrograde response (a signalling pathway
possibility that telomere shortening occurs during ageing in specific from the mitochondria to the nucleus) induced by mitochon-
regions of the brain or that telomere shortening in other cell types drial dysfunction (Passos et al., 2007). Other features of senes-
affects brain function (Jiang et al., 2007). For example, patients with cent cells include an increase in size (Hayflick, 1965), expression
AD have abnormally short telomeres in lymphocytes when com- of senescence-associated β-galactosidase (Dimri et al., 1995),
pared to age-matched controls (Panossian et al., 2003). and, in most senescent cells, an increase in the expression of the
There is no evidence of shorter telomeres in Parkinson’s disease cyclin-dependent kinase inhibitor p16INKa (e.g. Stein et al., 1999,
(Wang et al., 2008), although there have been some contrary results and reviewed by Rodier and Campisi, 2011). Senescent cells also
(Guan et al., 2008). Studies have investigated whether telomere have a detrimental effect on neighbouring cells and the surround-
length could be used as a biomarker of ageing. An inverse correla- ing extracellular matrix, a phenomenon known as the bystander
tion between telomere length in blood and mortality rate was found effect. This may be due to senescent cells secreting not only inflam-
in 60- to 75-year-olds (Cawthon et al., 2003). However, no signifi- matory cytokines and growth factors but also reactive species such
cant relationship between telomere length in white blood cells and as ROS and nitric oxide (NO) which then diffuse into neighbouring
survival was found in subjects aged over 85 years old (Martin-Ruiz cells and cause DNA damage (Sokolov et al., 2007). The bystander
et al., 2005). To conclude, there have been numerous studies inves- effect may be proinflammatory, procancerous, or pro-ageing.
tigating the relationship between telomere length and ageing but so
far the evidence to support the use of telomere length as a biomar- Apoptosis
ker of ageing is inconclusive (Mather et al., 2011). Apoptosis, or programmed cell death, is an essential process during
development. Apoptosis is also required to remove damaged cells
Cellular senescence which can then be replaced by the division of healthy cells to main-
The fact that human fibroblast cells can only divide a limited tain tissue homeostasis. The processes of cell death and prolifera-
number of times was first discovered by Leonard Hayflick and tion need to be finely balanced. Too much cell death leads to loss
Paul Moorhead in the early 1960s (Hayflick and Moorhead, 1961). of tissue and has been implicated in many age-related diseases such
The maximum proliferative lifespan of cells is called the Hayflick as neurodegenerative disorders, osteoporosis, and atherosclerosis.
limit and the irreversible loss of division potential of somatic cells On the other hand, too much cell proliferation can lead to cancer.
is known as replicative cellular senescence. Telomere shortening Apoptosis is highly regulated and can be activated by either intrinsic
has long been thought to be the biological counter for replicative or extrinsic signals. The extrinsic pathway is triggered by extracel-
senescence. However, it is not telomere length per se that triggers lular signal proteins binding to cell-membrane death receptors such
cell arrest but rather some property of the telomere such as uncap- as the tumour necrosis factor (TNF) receptor or the Fas death recep-
ping (von Zglinicki, 2003). Cellular senescence acts as a tumour tor. After binding, death receptors recruit intracellular adaptor pro-
suppressor mechanism (Bartek et al., 2007) but it also contributes teins which in turn recruit procaspases leading to a caspase cascade
to loss of tissue homeostasis in human ageing. There is evidence and finally cell death. There is evidence for altered caspase activity
for an increase in senescent cells with age in various tissues (Dimri with ageing which may contribute to dysregulation of the extrinsic
et al., 1995; Krishnamurthy et al., 2004; Herbig et al., 2006; Wang pathway (Zhang et al., 2003). Cells can also activate apoptosis from
et al., 2009) and in age-related diseases (reviewed in Burton, 2009), signals within the cell such as DNA damage, lack of nutrients, or
including atherosclerosis (Minamino and Komuro, 2007), some hypoxia. The intrinsic pathway requires the release of cytochrome c
cancers (e.g. prostate (Choi et al., 2000), lung (Muller et al., 2006), from mitochondria which binds to Apaf-1 and procaspase-9 to form
liver (Paradis et al., 2001)), and diabetes (Sone and Kagawa, 2005). the apoptosome. This pathway can be triggered directly or indi-
Cellular senescence is caused by a DNA damage response triggered rectly by ROS, which may explain the increase in apoptosis with age
by uncapped telomeres or nontelomeric DNA damage (d’Adda di (Simon et al., 2000). There are many proteins involved in regulating
Fagagna, 2008). It involves the formation of DNA damage foci, the the intrinsic pathway including Bcl-2, Bax, Jnk, p38MAPK, and p53
activation of kinases, and the activation of checkpoint proteins, e.g. and all these are known to play a role in ageing.
Age-related neurodegenerative diseases are associated with death Sirtuin-1 (Sirt-1) may play a pivotal role in the response to dietary
of neurons in specific regions of the brain caused by too much apop- restriction (Wakeling et al., 2009). Sirt-1 is an NAD-dependent
tosis (Mattson, 2000). It has been shown that p53 is up-regulated deacetylase and although its function is yet to be fully determined,
in AD which may contribute to the increase in apoptosis (Hooper its counterpart in yeast, Sir2, is known to be involved in epigenetic
et al., 2007). Since neurons are mostly postmitotic, cells cannot be gene silencing.
replenished and brain atrophy results. Skeletal muscle and heart
are also postmitotic tissues that are affected by age-related distur- Systems biology
bances leading to overactive apoptotic pathways (Pollack et al., The reader may rightly conclude that ageing is very complex and
2002). However, although there is evidence for a role for apoptosis involves multiple mechanisms. The overlap between mechanisms
in the ageing process, the mechanisms by which ageing modify the is very evident from the previous sections. For example, mitochon-
regulatory mechanisms of apoptosis are still far from clear (Zhang drial dysfunction has an impact on all energy-dependent cellular
and Herman, 2002). mechanisms. While reductionist approaches have predominated
in the past, the realization for the need of an integrative approach
Stochasticity was recognized over 10 years ago (von Zglinicki et al., 2001).
It has been observed that there is wide variation in lifespan in Reductionist approaches are still required to generate the detailed
genetically identical organisms raised in a constant environ- data but the emphasis is now on integrating new findings with
ment which therefore must be due to random effects (Herndon previous discoveries. In addition, powerful new technologies such
et al., 2002; Kirkwood et al., 2005). Finch and Kirkwood (2000) as functional genomics produce large volumes of data that need
proposed intrinsic chance as a third factor to the conventional sophisticated computer software packages for analysis. Therefore
two-factor model of ageing which attributes genetics and the envi- the multidisciplinary framework of systems biology is essential to
ronment as the main determinants of lifespan. The accumulation advance our understanding of ageing (Kirkwood, 2011). Systems
of molecular damage is random both in terms of when and where biology involves the close interaction and collaboration between
damage occurs. Chance events also occur in many other processes biologists, mathematicians, computer scientists, engineers, and
such as reproductive ageing, development, cell numbers, thresh- statisticians. The emphasis is on an iterative cycle of experiment,
olds for dysfunction, cell fates, telomere shortening, and control theory, and quantitative modelling (Fig. 1.4).
of gene expression (Finch and Kirkwood, 2000). For example, it Mathematical and computer modelling is now being increas-
is known that there is large stochastic variation in telomere length ingly used in biomedical research. However, many scientists are
(Martin-Ruiz et al., 2004), and epigenetic changes are also highly still unclear on the use of models in this area and so we summarize
stochastic and have been shown to explain why identical twins age the main advantages of using models as an additional research tool.
differently (Petronis, 2006). Model building requires a specific hypothesis and each element of
the model and how it interacts with other elements must be speci-
Dietary restriction fied. This often highlights unknown elements and uncertainties
Dietary restriction (DR) is the most robust intervention known to about specific reactions. In this case, more than one model can be
extend lifespan and reduce morbidity in a range of species includ- built to investigate different possibilities. The model can be used to
ing rodents, C. elegans, D. melanogaster, and yeast. The extensions make both qualitative and quantitative predictions which may sug-
in lifespan are quite substantial. For example, a study showed that gest further experiments. Modelling is relatively inexpensive and
the average lifespan of mice in the top decile was 53 months in the quick compared to laboratory experiments and so it can provide a
group fed on a diet with a 35% reduction of the ad libitum intake low-cost rapid test-bed for candidate interventions.
compared to 35 months for the control group (Weindruch et al., A wide range of models have already been developed to investi-
1986). Another rodent study attributed an increase in lifespan of gate the molecular mechanisms of ageing, such as telomere erosion
dietary restricted animals to a reduction in age-related diseases (Proctor and Kirkwood, 2002, 2003), the accumulation of defec-
such as diabetes, atherosclerosis, respiratory disease, and cancer tive mitochondria (Kowald and Kirkwood, 1999, 2000, 2011; Elson
(Fontana and Klein, 2007). However, DR was also shown to have et al., 2001), the breakdown of protein homeostasis (Proctor et al.,
effects on ageing per se. 2005; 2007; Proctor and Lorimer, 2011), and signalling pathways
DR lowers the rate of mitochondrial ROS generation which such as the Forkhead box protein O (FOXO) response (Smith and
results in slower rate of development of tissue oxidative stress and
damage with age in rodents (Lambert and Merry, 2004; Merry,
2004; Sanz et al., 2005). However, the theory remains controver-
sial (Sanz et al., 2006; Jang and Remmen, 2009). A recent study Model
showed that the effect of DR on the rate of ROS generation can be
explained and reversed in liver and skeletal muscle mitochondria
by the indirect effect of exogenous insulin (Ash and Merry, 2011). Iterative cycle
DR also results in decreased concentrations of cytokines in plasma, Hypothesis of modelling Predictions
up-regulation of molecular chaperones, and a reduction in protein and experiments
damage and oxidized lipids (Fontana and Klein, 2007). Another
review of the effects of DR also suggests that lower concentra-
tions of plasma glucose, insulin, and cholesterol and an increase Experiments
in insulin resistance and glucose tolerance play an important role
(Guarente and Picard, 2005). Recently it has been suggested that Fig. 1.4 An iterative cycle of modelling and experimentation.
Shanley, 2010), the DNA damage response (Passos et al., 2010), Braak, H. and Braak, E. (1991). Neuropathological stageing of
stem cells (Ro and Rannala, 2001; Taylor et al., 2003; Glauche et al., Alzheimer-related changes. Acta Neuropathologica, 82, 239–59.
2011), and network models to explore the interactions between dif- Brierley, E.J., et al. (1998). Role of mitochondrial DNA mutations in human
aging: implications for the central nervous system and muscle. Annals of
ferent mechanisms (Kowald and Kirkwood, 1996, 2000; Sozou and
Neurology, 43, 217–23.
Kirkwood, 2001). Models have also been developed to specifically
Brundin, P., et al. (2011). Prion-like transmission of protein aggregates in
examine the processes involved in brain ageing and age-related neurodegenerative diseases. Nature Reviews Molecular Cell Biology, 11,
neurodegenerative disorders (McAuley et al., 2009; Proctor and 301–7.
Gray, 2010; Proctor et al., 2010). Burton, D.G. (2009). Cellular senescence, ageing and disease. Age (Dordrecht,
As well as modelling, systems biology involves the use of bioin- Netherlands), 31, 1–9.
formatic tools in order to try to understand how the many com- Calkins, M.J. and Reddy, P.H. (2011). Amyloid beta impairs mitochondrial
ponents involved in ageing interact (e.g. genes, proteins). A recent anterograde transport and degenerates synapses in Alzheimer’s disease
review highlights some of the databases related to the ageing neurons. Biochimica et Biophysica Acta, 1812, 507–13.
process (Peysselon and Ricard-Blum, 2011). Vast amount of high Cawthon, R.M., et al. (2003). Association between telomere length in blood
and mortality in people aged 60 years or older. Lancet, 361, 393–5.
throughput omics data are being generated and this has led to new
Chiu, R.K, et al. (2006). Lysine 63-polyubiquitination guards against
challenges in terms of data storage and analysis. Bayesian statis- translesion synthesis-induced mutations. PLoS Genetics, 2, e116.
tical techniques are being increasingly used to extract the maxi- Choi, J., et al. (2000). Expression of senescence-associated beta-galactosidase
mum amount of information from data that is subject to errors and in enlarged prostates from men with benign prostatic hyperplasia.
intrinsic noise (Wilkinson, 2007). Data storage and analysis require Urology, 56, 160–6.
high computational power and need the use of parallel computer Conley, K.E., et al. (2000). Oxidative capacity and ageing in human muscle.
clusters or cloud computing. Software tools also need to be con- Journal of Physiology, 526 Pt 1, 203–10.
tinually upgraded and new tools developed to keep pace with new Coppe, J.P., et al. (2008). Senescence-associated secretory phenotypes reveal
technologies for data acquisition. cell-nonautonomous functions of oncogenic ras and the p53 tumor
suppressor. PLoS Biology, 6, 2853–68.
Corder, E.H., et al. (1994). Protective effect of apolipoprotein e type 2 allele
Conclusion for late onset Alzheimer disease. Nature Genetics, 7, 180–4.
This chapter has given an overview of some of the biological aspects Cortopassi, G.A. and Arnheim, N. (1990). Detection of a specific
mitochondrial-DNA deletion in tissues of older humans. Nucleic Acids
of human ageing, with a particular emphasis on damage to molecu-
Research, 18, 6927–33.
lar and cellular components by ROS. The mechanisms involve many
Cottrell, D.A., et al. (2000). Neuropathological and histochemical changes
complex interactions and, as we have seen, vicious cycles of damage in a multiple mitochondrial DNA deletion disorder. Journal of
leading to more damage are a common theme. Our understanding Neuropathology and Experimental Neurology, 59, 621–7.
of the processes involved in ageing are still far from complete, but Cottrell, D.A., et al. (2001). Cytochrome c oxidase deficient cells accumulate
the vast improvements in experimental techniques and comput- in the hippocampus and choroid plexus with age. Neurobiological Aging,
ing power has greatly advanced our knowledge over recent years. 22, 265–72.
However, this has brought new challenges with regards to making d’Adda di Fagagna, F. (2008). Living on a break: cellular senescence as a
sense of the increasing amounts of data. It is also necessary to be DNA-damage response. Nature Reviews Cancer, 8, 512–22.
able to integrate data and hypotheses in order to gain better under- Dimri, G.P., et al. (1995). A biomarker that identifies senescent human cells
in culture and in aging skin in vivo. Proceedings of the National Academy
standing of the ageing process.
of Sciences of the United States of America, 92, 9363–7.
Therefore research into ageing now requires a multidisciplinary Djojosubroto, M.W., et al. (2003). Telomeres and telomerase in aging,
team of bio-informaticians, statisticians, computer programmers, regeneration and cancer. Molecules and Cells, 15, 164–75.
and mathematical modellers, in addition to clinicians and experi- Dunning, C.J., et al. (2012). Can Parkinson’s disease pathology be propagated
mental scientists. from one neuron to another? Progress in Neurobiology, 97, 205–19.
Elson, J.L., et al. (2001). Random intracellular drift explains the clonal
References expansion of mitochondrial DNA mutations with age. American Journal
Allsopp, R.C., et al. (1995). Telomere shortening is associated with cell-division of Human Genetics, 68, 802–6.
in-vitro and in-vivo. Experimental Cell Research, 220, 194–200. Eriksen, J.L., et al. (2003). Caught in the act: alpha-synuclein is the culprit in
Arrasate, M., et al. (2004). Inclusion body formation reduces levels of Parkinson’s disease. Neuron, 40, 453–6.
mutant huntingtin and the risk of neuronal death. Nature, 431, Fannin, S.W., et al. (1999). Aging selectively decreases oxidative capacity
805–10. in rat heart interfibrillar mitochondria. Archives of Biochemistry and
Ash, C.E. and Merry, B.J. (2011). The molecular basis by which dietary Biophysics, 372, 399–407.
restricted feeding reduces mitochondrial reactive oxygen species Finch, C.E. and Kirkwood, T.B.L. (2000). Chance, development, and aging.
generation. Mechanisms of Ageing and Development, 132, 43–54. Oxford University Press, New York.
Bartek, J., et al. (2007). DNA damage signalling guards against activated Firdaus, W.J., et al. (2006). Huntingtin inclusion bodies are iron-dependent
oncogenes and tumour progression. Oncogene, 26, 7773–9. centers of oxidative events. FEBS Journal, 273, 5428–41.
Bender, A., et al. (2006). High levels of mitochondrial DNA deletions in Fontana, L. and Klein, S. (2007). Aging, adiposity, and calorie restriction.
substantia nigra neurons in aging and Parkinson disease. Nature Genetics, Journal of the American Medical Association, 297, 986–94.
38, 515–17. Fraga, M.F., et al. (2005). Epigenetic differences arise during the lifetime of
Bennett, E.J., et al. (2007). Global changes to the ubiquitin system in monozygotic twins. Proceedings of the National Academy of Sciences of
Huntington’s disease. Nature, 448, 704–8. the United States of America, 102, 10604–9.
Blalock, E.M, et al. (2003). Gene microarrays in hippocampal aging: statistical Friedman, D.B. and Johnson, T.E. (1988). A mutation in the age-1 gene in
profiling identifies novel processes correlated with cognitive impairment. Caenorhabditis elegans lengthens life and reduces hermaphrodite fertility.
Journal of Neuroscience, 23, 3807–19. Genetics, 118, 75–86.
Friso, S. and Choi, S.W. (2002). Gene-nutrient interactions and DNA Johnston, J.A., et al. (1998). Aggresomes: a cellular response to misfolded
methylation. Journal of Nutrition, 132, 2382S–7S. proteins. Journal of Cell Biology, 143, 1883–98.
Glatt, S.J., et al. (2007). Successful aging: from phenotype to genotype. Jucker, M. and Walker, L.C. (2011). Pathogenic protein seeding in Alzheimer
Biological Psychiatry, 62, 282–93. disease and other neurodegenerative disorders. Annals of Neurology, 70,
Glauche, I., et al. (2011). Cellular aging leads to functional heterogeneity of 532–40.
hematopoietic stem cells: a modeling perspective. Aging Cell, 10, 457–65. Kapahi, P., et al. (2004). Regulation of lifespan in drosophila by modulation
Gousset, K., et al. (2009). Prions hijack tunnelling nanotubes for intercellular of genes in the tor signaling pathway. Current Biology, 14, 885–90.
spread. Nature and Cell Biology, 11, 328–36. Kawajiri, S., et al. (2010). Pink1 is recruited to mitochondria with parkin and
Gravina, S. and Vijg, J. (2010). Epigenetic factors in aging and longevity. associates with lc3 in mitophagy. FEBS Letters, 584, 1073–9.
Pflugers Archives, 459, 247–58. Keller, J.N., et al. (2002). The proteasome in brain aging. Ageing Research
Gray, D.A., et al. (2003). Ubiquitin, proteasomes, and the aging brain. Science Review, 1, 279–93.
of Aging Knowledge Environment, 34, RE6. Kim, I., et al. (2007). Selective degradation of mitochondria by mitophagy.
Guan, J.Z., et al. (2008). A percentage analysis of the telomere length in Archives of Biochemistry and Biophysics, 462, 245–53.
Parkinson’s disease patients. Journal of Gerontology Series A Biological Kirkwood, T.B. (1977). Evolution of ageing. Nature, 270, 301–4.
Science and Medical Science, 63, 467–73. Kirkwood, T.B. (2008). Understanding ageing from an evolutionary
Guarente, L. and Picard, F. (2005). Calorie restriction—the sir2 connection. perspective. Journal of Internal Medicine, 263, 117–27.
Cell, 120, 473–82. Kirkwood, T.B. (2011). Systems biology of ageing and longevity. Philosophical
Gusella, J.F. and MacDonald, M.E. (2006). Huntington’s disease: seeing the transactions of the Royal Society of London. Series B, Biological sciences,
pathogenic process through a genetic lens. Trends in Biochemical Science, 366, 64–70.
31, 533–40. Kirkwood, T.B. and Melov, S. (2011). On the programmed/non-programmed
Halliwell, B. (2006). Oxidative stress and neurodegeneration: where are we nature of ageing within the life history. Current Biology, 21, R701–7.
now? Journal of Neurochemistry, 97, 1634–58. Kirkwood, T.B., et al. (2005). What accounts for the wide variation in life span
Hands, S.L., et al. (2009). mTOR’s role in ageing: protein synthesis or of genetically identical organisms reared in a constant environment?
autophagy? Aging (Albany, New York), 1, 586–97. Mechanisms of Ageing and Development, 126, 439–43.
Harman, D. (1956). Aging: a theory based on free radical and radiation Knott, A.B. and Bossy-Wetzel, E. (2008). Impairing the mitochondrial fission
chemistry. Journal of Gerontology, 11, 298–300. and fusion balance: a new mechanism of neurodegeneration. Annals of
Hayakawa, M., et al. (1991). Age-associated accumulation of the New York Academy of Sciences, 1147, 283–92.
8-hydroxydeoxyguanosine in mitochondrial DNA of human diaphragm. Kowald, A. and Kirkwood, T.B.L. (1996). A network theory of ageing: the
Biochemical and Biophysical Research Communications, 179, 1023–9. interactions of defective mitochondria, aberrant proteins, free radicals
Hayakawa, M., et al. (1992). Age-associated oxygen damage and mutations and scavengers in the ageing process. Mutation Research-Dnaging Genetic
in mitochondrial DNA in human hearts. Biochemical and Biophysical Instability and Aging, 316, 209–36.
Research Communications, 189, 979–85. Kowald, A. and Kirkwood, T.B.L. (1999). Modeling the role of mitochondrial
Hayflick, L. (1965). The limited in vitro lifetime of human diploid cell strains. mutations in cellular aging. Journal of Anti-Aging Medicine, 2, 243–53.
Experimental Cell Research, 37, 614–36. Kowald, A. and Kirkwood, T.B.L. (2000). Accumulation of defective
Hayflick, L. and Moorhead, P.S. (1961). The serial cultivation of human mitochondria through delayed degradation of damaged organelles and
diploid cell strains. Experimental Cell Research, 25, 585–621. its possible role in the ageing of post-mitotic and dividing cells. Journal
Hemann, M.T., et al. (2001). The shortest telomere, not average telomere length, of Theoretical Biology, 202, 145–60.
is critical for cell viability and chromosome stability. Cell, 107, 67–77. Kowald, A. and Kirkwood, T.B.L. (2011). Evolution of the mitochondrial
Her, L.S. and Goldstein, L.S. (2008). Enhanced sensitivity of striatal neurons fusion–fission cycle and its role in aging. Proceedings of the National
to axonal transport defects induced by mutant huntingtin. Journal of Academy of Sciences, 108, 10237–42.
Neuroscience, 28, 13662–72. Koziel, R., et al. (2011). Functional interplay between mitochondrial and
Herbig, U., et al. (2006). Cellular senescence in aging primates. Science, proteasome activity in skin aging. Journal of Investigative Dermatology,
311, 1257. 131, 594–603.
Herndon, L.A., et al. (2002). Stochastic and genetic factors influence Kraytsberg, Y., et al. (2006). Mitochondrial DNA deletions are abundant and
tissue-specific decline in ageing C. elegans. Nature, 419, 808–14. cause functional impairment in aged human substantia nigra neurons.
Hochstrasser, M. (1996). Ubiquitin-dependent protein degradation. Annual Nature Genetics, 38, 518–20.
Review of Genetics, 30, 405–39. Krishnamurthy, J., et al. (2004). Ink4a/arf expression is a biomarker of aging.
Hooper, C., et al. (2007). P53 is upregulated in Alzheimer’s disease and induces Journal of Clinical Investigations, 114, 1299–307.
tau phosphorylation in hek293a cells. Neuroscience Letters, 418, 34–7. Kujoth, G.C., et al. (2005). Mitochondrial DNA mutations, oxidative stress,
Jaarsma, D., et al. (2011). Age-related neuronal degeneration: complementary and apoptosis in mammalian aging. Science, 309, 481–4.
roles of nucleotide excision repair and transcription-coupled repair in Lambert, A.J. and Merry, B.J. (2004). Effect of caloric restriction on
preventing neuropathology. PLoS Genetics, 7, e1002405. mitochondrial reactive oxygen species production and bioenergetics:
Jahani-Asl, A., et al. (2007). Mitofusin 2 protects cerebellar granule neurons reversal by insulin. American Journal of Physiology. Regulatory, Integrative
against injury-induced cell death. Journal of Biological Chemistry, 282, and Comparative Physiology, 286, R71–9.
23788–98. Lapointe, J. and Hekimi, S. (2008). Early mitochondrial dysfunction
Jang, H., et al. (2009). Viral parkinsonism. Biochimica Biophysica Acta, 1792, in long-lived mclk1+/– mice. Journal of Biological Chemistry, 283,
714–21. 26217–27.
Jang, Y.C. and Remmen, V.H. (2009). The mitochondrial theory of aging: Lee, C.K., et al. (2000). Gene-expression profile of the ageing brain in mice.
insight from transgenic and knockout mouse models. Experimental Nature Genetics, 25, 294–7.
Gerontology, 44, 256–60. Lynnerup, N., et al. (2008). Radiocarbon dating of the human eye lens
Jia, K., et al. (2004). The tor pathway interacts with the insulin signaling crystallines reveal proteins without carbon turnover throughout life.
pathway to regulate C. elegans larval development, metabolism and life PLoS One, 3, e1529.
span. Development, 131, 3897–906. Martens, U.M., et al. (2000). Accumulation of short telomeres in human
Jiang, H., et al. (2007). Telomere shortening and ageing. Zeitschrift für fibroblasts prior to replicative senescence. Experimental Cell Research,
Gerontologie und Geriatrie, 40, 314–24. 256, 291–9.
Martin-Ruiz, C., et al. (2004). Stochastic variation in telomere shortening rate Powers, R.W., 3rd, et al. (2006). Extension of chronological life span in yeast
causes heterogeneity of human fibroblast replicative life span. Journal of by decreased tor pathway signaling. Genes and Development, 20, 174–84.
Biological Chemistry, 279, 17826–33. Proctor, C.J. and Gray, D.A. (2010). Gsk3 and p53—is there a link in
Martin-Ruiz, C., et al. (2005). Telomere length in white blood cells Alzheimer’s disease? Molecular Neurodegeneration, 5, 7.
is not associated with morbidity or mortality in the oldest old: a Proctor, C.J. and Kirkwood, T.B. (2002). Modelling telomere shortening and
population-based study. Aging Cell, 4, 287–90. the role of oxidative stress. Mechanisms of Ageing and Development, 123,
Martinez, D.E. (1998). Mortality patterns suggest lack of senescence in hydra. 351–63.
Experimental Gerontology, 33, 217–25. Proctor, C.J. and Kirkwood, T.B. (2003). Modelling cellular senescence as a
Mather, K.A., et al. (2011). Is telomere length a biomarker of aging? A review. result of telomere state. Aging Cell, 2, 151–7.
Journal of Gerontology Series A Biological Science and Medical Science, Proctor, C.J. and Lorimer, I.A.J. (2011). Modelling the role of the hsp70/
66, 202–13. hsp90 system in the maintenance of protein homeostasis. PLoS One, 6,
Matsuzawa, A. and Ichijo, H. (2008). Redox control of cell fate by map kinase: e22038.
physiological roles of ASK1-MAP kinase pathway in stress signaling. Proctor, C.J., et al. (2005). Modelling the actions of chaperones and their role
Biochimica Biophysica Acta, 1780, 1325–36. in ageing. Mechanisms of Ageing and Develpment, 126, 119–31.
Mattson, M.P. (2000). Apoptosis in neurodegenerative disorders. Nature Proctor, C.J., et al. (2007). An in silico model of the ubiquitin-proteasome
Reviews in Molecular and Cellular Biology, 1, 120–9. system that incorporates normal homeostasis and age-related decline.
McAuley, M.T., et al. (2009). A mathematical model of aging-related and BMC Systems Biology, 1, 17.
cortisol induced hippocampal dysfunction. BMC Neuroscience, 10, 26. Proctor, C.J., et al. (2010). Modelling the role of uch-l1 on protein aggregation
McNaught, K.S., et al. (2002). Aggresome-related biogenesis of Lewy bodies. in age-related neurodegeneration. PLoS One, 5, e13175.
European Journal of Neuroscience, 16, 2136–48. Ro, S. and Rannala, B. (2001). Methylation patterns and mathematical models
Merry, B.J. (2004). Oxidative stress and mitochondrial function with aging— reveal dynamics of stem cell turnover in the human colon. Proceedings
the effects of calorie restriction. Aging Cell, 3, 7–12. of the National Academy of Sciences of the United States of America, 98,
Minamino, T. and Komuro, I. (2007). Vascular cell senescence: contribution 10519–21.
to atherosclerosis. Circulation Research, 100, 15–26. Rodier F, Campisi J (2011). Four faces of cellular senescence. J Cell Biol, 192,
Miquel, J. (1991). An integrated theory of aging as the result of 547–56.
mitochondrial-DNA mutation in differentiated cells. Archives of Ruse, C.E. and Parker, S.G. (2001). Molecular genetics and age-related
Gerontology and Geriatrics, 12, 99–117. disease. Age and Ageing, 30, 449–54.
Muller, K.C., et al. (2006). Lung fibroblasts from patients with emphysema Sanz, A., et al. (2005). Dietary restriction at old age lowers mitochondrial
show markers of senescence in vitro. Respiration Research, 7, 32. oxygen radical production and leak at complex I and oxidative DNA
Muller-Hocker, J. (1989). Cytochrome-c-oxidase deficient cardiomyocytes damage in rat brain. Journal of Bioenergetics and Biomembranes, 37,
in the human heart—an age-related phenomenon. A histochemical 83–90.
ultracytochemical study. American Journal of Pathology, 134, 1167–73. Sanz, A., et al. (2006). Carbohydrate restriction does not change
Nathan, C. (2003). Specificity of a third kind: reactive oxygen and nitrogen mitochondrial free radical generation and oxidative DNA damage.
intermediates in cell signaling. Journal of Clinical Investigations, 111, Journal of Bioenergetics and Biomembranes, 38, 327–33.
769–78. Saunders, A.M., et al. (1993). Association of apolipoprotein e allele epsilon
Nouspikel, T. (2011). Multiple roles of ubiquitination in the control of 4 with late-onset familial and sporadic Alzheimer’s disease. Neurology,
nucleotide excision repair. Mechanisms of Ageing and Development, 132, 43, 1467–72.
355–65. Schriner, S.E., et al. (2005). Extension of murine life span by overexpression
Ohtsuka, H., et al. (1995). Age-related accumulation of high-molecular-weight of catalase targeted to mitochondria. Science, 308, 1909–11.
ubiquitin protein conjugates in mouse brains. Journal of Gerontology Schubert, U., et al. (2000). Rapid degradation of a large fraction of newly
Series A Biological Science and Medical Science, 50, B277–81. synthesized proteins by proteasomes. Nature, 404, 770–4.
Olovnikov, A.M. (1971). Principles of marginotomy in template synthesis of Selman, C., et al. (2009). Ribosomal protein S6 kinase 1 signaling regulates
polynucleotides. Doklady Akademii Nauk SSSR, 201, 1496–9. mammalian life span. Science, 326, 140–4.
Panossian, L.A., et al. (2003). Telomere shortening in T cells correlates with Serra, V., et al. (2003). Extracellular superoxide dismutase is a major
Alzheimer’s disease status. Neurobiological Aging, 24, 77–84. antioxidant in human fibroblasts and slows telomere shortening. Journal
Paradis, V., et al. (2001). Replicative senescence in normal liver, chronic of Biological Chemistry, 278, 6824–30.
hepatitis C, and hepatocellular carcinomas. Human Pathology, 32, Shelton, D.N., et al. (1999). Microarray analysis of replicative senescence.
327–32. Current Biology, 9, 939–45.
Passos, J.F., et al. (2007). Mitochondrial dysfunction accounts for the stochastic Simboeck, E., et al. (2011). Epigenetics and senescence: learning from the
heterogeneity in telomere-dependent senescence. PLoS Biology, 5, e110. INK4-ARF locus. Biochemical Pharmacology, 82(10), 1361–70.
Passos, J.F., et al. (2010). Feedback between p21 and reactive oxygen Simon, H.U., et al. (2000). Role of reactive oxygen species (ROS) in apoptosis
production is necessary for cell senescence. Molecular Systems Biology, induction. Apoptosis, 5, 415–18.
6, 347. Smith, G.R. and Shanley, D.P. (2010). Modelling the response of foxo
Petronis, A. (2006). Epigenetics and twins: three variations on the theme. transcription factors to multiple post-translational modifications made
Trends in Genetics, 22, 347–50. by ageing-related signalling pathways. PLoS One, 5, e11092.
Peysselon, F. and Ricard-Blum, S. (2011). Understanding the biology of aging Sokolov, M.V., et al. (2007). Gamma-H2AX in bystander cells: not just a
with interaction networks. Maturitas, 69, 126–30. radiation-triggered event, a cellular response to stress mediated by
Piotrowski, J., et al. (1997). Inhibition of the 26 s proteasome by polyubiquitin intercellular communication. Cell Cycle, 6, 2210–12.
chains synthesized to have defined lengths. Journal of Biological Sone, H. and Kagawa, Y. (2005). Pancreatic beta cell senescence contributes
Chemistry, 272, 23712–21. to the pathogenesis of type 2 diabetes in high-fat diet-induced diabetic
Pollack, M., et al. (2002). The role of apoptosis in the normal aging brain, mice. Diabetologia, 48, 58–67.
skeletal muscle, and heart. Annals of the New York Academy of Sciences, Sozou, P.D. and Kirkwood, T.B.L. (2001). A stochastic model of cell replicative
959, 93–107. senescence based on telomere shortening, oxidative stress, and somatic
Polymenidou, M. and Cleveland, D.W. (2011). The seeds of neurodegeneration: mutations in nuclear and mitochondrial DNA. Journal of Theoretical
prion-like spreading in ALS. Cell, 147, 498–508. Biology, 213, 573–86.
Stein, G.H., et al. (1999). Differential roles for cyclin-dependent kinase mimicked by other dietary interventions? Age (Dordrecht, The
inhibitors p21 and p16 in the mechanisms of senescence and differentiation Netherlands), 31, 327–41.
in human fibroblasts. Molecular Cell Biology, 19, 2109–17. Wallace DC (1999). Mitochondrial diseases in mouse and man. Science, 283,
Tang, M.Y., et al. (2010). Experimental and computational analysis of 1482–8.
polyglutamine-mediated cytotoxicity. PLoS Computational Biology, 6, Wang, C., et al. (2009). DNA damage response and cellular senescence in
e100094. tissues of aging mice. Aging Cell, 8, 311–23.
Taylor, R.W., et al. (2003). Mitochondrial DNA mutations in human colonic Wang, H., et al. (2008). Telomere length and risk of Parkinson’s disease.
crypt stem cells. Journal of Clinical Investigation, 112, 1351–60. Movement Disorders, 23, 302–5.
Tollefsbol, T.O.s and SpringerLink (Online service) (2010). Epigenetics of Waxman, E.A., et al. (2009). Leucine-rich repeat kinase 2 expression leads
aging. Springer, New York. to aggresome formation that is not associated with alpha-synuclein
Trifunovic, A., et al. (2004). Premature ageing in mice expressing defective inclusions. Journal of Neuropathology and Experimental Neurology, 68,
mitochondrial DNA polymerase. Nature, 429, 417–23. 785–96.
Trifunovic, A., et al. (2005). Somatic MTDNA mutations cause aging Weihofen, A., et al. (2009). Pink1 forms a multiprotein complex with
phenotypes without affecting reactive oxygen species production. Miro and Milton, linking Pink1 function to mitochondrial trafficking.
Proceedings of the National Academy of Science USA, 102, 17993–8. Biochemistry, 48, 2045–52.
Trounce, I., et al. (1989). Decline in skeletal muscle mitochondrial respiratory Weindruch, R., et al. (1986). The retardation of aging in mice by dietary
chain function: possible factor in ageing. Lancet, 1, 637–9. restriction: longevity, cancer, immunity and lifetime energy intake.
Tsirigotis, M., et al. (2008). Activation of p38MAPK contributes to expanded Journal of Nutrition, 116, 641–54.
polyglutamine-induced cytotoxicity. PLoS One, 3, e2130. Wenz, T. (2011). Mitochondria and PGC-1alpha in aging and age-associated
Twig, G., et al. (2008). Fission and selective fusion govern mitochondrial diseases. Journal of Aging Research, 810619 (Epub).
segregation and elimination by autophagy. EMBO Journal, 27, 433–46. Wilkinson, D.J. (2007). Bayesian methods in bioinformatics and computational
Tydlacka, S., et al. (2008). Differential activities of the ubiquitin-proteasome systems biology. Brief Bioinformatics, 8, 109–16.
system in neurons versus glia may account for the preferential Woulfe, J.M. (2007). Abnormalities of the nucleus and nuclear inclusions
accumulation of misfolded proteins in neurons. Journal of Neuroscience, in neurodegenerative disease: a work in progress. Neuropathology and
28, 13285–95. Applied Neurobiology, 33, 2–42.
Tyner, S.D., et al. (2002). P53 mutant mice that display early ageing-associated Yen, T.C., et al. (1989). Liver mitochondrial respiratory functions decline
phenotypes. Nature, 415, 45–53. with age. Biochemical and Biophysical Research Commununications, 165,
Vijg, J. (2007). Aging of the genome. Oxford University Press, Oxford. 944–1003.
von Zglinicki, T. (2002). Oxidative stress shortens telomeres. Trends in Zhang, J.-H., et al. (2003). Caspases, apoptosis and aging. Ageing Research
Biochemical Sciences, 27, 339–44. Review, 2, 357–66.
von Zglinicki, T. (2003). Replicative senescence and the art of counting. Zhang, Y. and Herman, B. (2002). Ageing and apoptosis. Mechanisms of
Experimental Gerontology, 38, 1259–64. Ageing and Development, 123, 245–60.
von Zglinicki, T., et al. (2001). Stress, DNA damage and ageing—an integrative Ziv, E. and Hu, D. (2011). Genetic variation in insulin/IGF-1 signaling
approach. Experimental Gerontology, 36, 1049–62. pathways and longevity. Ageing Research Review, 10, 201–4.
Wakeling, L.A., et al. (2009). Could SIRT1-mediated epigenetic effects Zoncu R, et al. (2011). mTOR: from growth signal integration to cancer,
contribute to the longevity response to dietary restriction and be diabetes and ageing. Nature Reviews Molecular Cell Biology, 12, 21–35.
CHAPTER 2
Psychometric assessment
in older people
Karen Ritchie
Psychometric assessment involves the quantification of observa- conceptualizations of dysfunction. Increasing emphasis on the
tions of behaviour, cognition, and affect, and as such is an impor- quality of life of older people and an increasingly optimistic view
tant adjunct to psychogeriatric assessment in both the clinical of what should constitute the normal health status of the older per-
and research setting. The step from observation to measurement son have undoubtedly led to a lowering of the threshold for what is
is also important in the contribution it frequently makes to fur- considered ‘acceptable’ discomfort.
thering our understanding of a given health problem at a con- Given that the biological mechanisms underlying mental dis-
ceptual level. As Blalock (1968) has pointed out, ‘measurement order are only partially understood, diagnosis commonly relies
considerations often enable us to clarify our theoretical thinking on observations of the non-specific behavioural consequences of
and to suggest new variables that should be considered . . . careful mental disorder, which are dimensional rather than categorical
attention to measurement may force a clarification of one’s basic variables. The situation thus frequently arises that mental disor-
concepts and theories’. This chapter will consider firstly some of ders now commonly considered to be discontinuous with normal
the theoretical issues specific to the psychometric evaluation of ageing (for example, Alzheimer’s disease, major depressive illness)
older populations, and secondly review the use which has been are commonly diagnosed by reference to nonspecific dimensional
made of psychometric techniques in the evaluation of cognitive variables such as sadness, motor speed, and memory performance.
disorder in older people. As a result, measures of mental health status in older people are
often based on both the psychological and medical models. For
Conceptual Considerations example, neuropsychological measures of cognitive functioning
commonly take the form of dimensional behavioural measures
Assessment models based on the psychological model, such as word fluency and visual
Two principal models have governed our conceptualization of recall. Such tests also permit the investigator to observe the exist-
mental disorder. On the one hand, the dichotomic medical model ence of dichotomous signs indicative of pathology according to
clearly distinguishes normal fluctuations in mental functioning the medical model, such as perseveration, dyskinesia, aphasia, and
(e.g. transient feelings of sadness or ageing-associated memory visual field neglect. Magnetic resonance spectroscopy, functional
impairment) from psychopathology (e.g. major depressive episode, magnetic resonance imaging, and diffusion tensor imaging draw
dementia). This model construes psychiatric disorder as a disease together the two approaches by the visualization of the functional
process whose aetiology is separate from that of ‘normal’ ageing. and connectivity anatomical correlates of performance on tests of
Measures based on the medical model refer to pathological behav- cognition in both normal and pathological ageing (Nyberg, 1998;
iours which are not seen in normal populations (for example, apha- Page, 2006; Minati et al., 2007).
sia, suicide attempts, hallucinations) and thus clearly differentiate When developing a measuring instrument for diagnosis or
healthy and unhealthy cohorts. screening of mental disorder in older people, some consideration
The psychological model, on the other hand, conceptualizes should be given to its underlying conceptual assumptions as these
mental functioning in terms of a normal distribution. This model will play an important part in the scoring of items and in assess-
assumes that affective and cognitive problems are to some degree ing validity. In the case of the medical model, discrimination
present in all older people; poor mental health being defined in may be improved by increasing the number of items relating to
terms of degree of discomfort or a statistically significant deviation disease-specific symptoms and reducing those relating to nonspe-
from an established norm. Measures based on this model are there- cific symptoms. In the case of measures based on the psychological
fore dimensional rather than categorical and present the problem paradigm, adjustment is more commonly required in the cut-off
of determining a suitable cut-off point for ‘abnormality’. The deter- point according to symptom prevalence and severity in the target
mination of an appropriate cut-off point for this type of measure is population. This point is discussed further in relation to screening
in part a statistical problem, but also depends on changing social instruments (see section Screening tests).
The definition of ‘normality’ in older populations impact on test performance (Jackson et al., 2011), as may medica-
A fundamental consideration in the development of measures of tions commonly taken by older people, particularly those with anti-
mental functioning in the older person has been the question of cholinergic effects (Ancelin et al., 2006). Older populations have
what is ‘normal’ at a given age. All too often ‘normal’ performance a high prevalence of sensory impairment, yet very few tests have
is taken to be the average performance of an age cohort in which been developed specifically for older persons who have visual or
older people with mental disorder have been excluded. This prac- auditory problems. The inventive clinician may consider, however,
tice has undoubtedly underestimated true normal performance due making use of the tactile tests included in child assessment batter-
to the inclusion in the so-called normal group of persons with sub- ies. Unfortunately, as many of the psychometric measures available
clinical pathologies and other conditions likely to mask true abil- for use with older people have been validated on ‘selected’ popu-
ity (notably sensory impairments and coexisting physical illness). lations free of impairment and disease, their validity on the large
Advances in medical technology have also permitted the identifi- proportion of older people with pathology remains unknown.
cation of previously unrecognized pathology in so-called normal A further problem is the high level of illiteracy and low levels of
older brains, for example white matter lesions and reduced grey education often found in older populations. Education differentials
matter volume (Raj et al., 2012). raise two major problems. The first has been the difficulties inher-
Rapid changes over the last century in environmental factors ent in the development of ‘education fair’ measures which do not
likely to have an important influence on mental functioning (educa- produce, for example, high false positive rates in the assessment of
tion, medical care, nutrition, protection from adverse environmen- cognitive deficit in the poorly educated or false negative rates in
tal exposure) have given rise to important age-cohort effects. That is, older people with high levels of education. A number of statistical
younger older people are likely to have benefited from more favoura- techniques have been developed which may assist in the evalua-
ble conditions than the oldest old—including greater familiarity with tion of item bias, such as the use of statistical weighting using, for
questionnaires and psychometric tests. For example, an early study example, a nonparametric or stratified regression method (Kittner
by Schaie (1983) noted significant cross-sectional age group differ- et al., 1986), multicomponent latent trait models (MLTM), or item
ences in mean cognitive performance, but over a 20-year follow-up response theory (IRT) (Embretson and Yang, 2006). An exam-
there was very little difference in longitudinal change before the ple of the application of IRT to a cognitive test battery is given by
age of 80 or so. Drawing on the example of cardiovascular disease, Lindeboom et al. (2004). The second problem is the question of
Manton and Stallard (1988) have also raised the point that disorders whether in adjusting for education effects in psychometric tests the
such as loss of respiratory capacity, once thought to be an inevitable researcher is not in fact removing the effects of a true risk factor.
feature of the ageing process, are often redefined as pathologies: ‘age High rates of institutionalization in older populations, particu-
criteria are tending to disappear and what is considered to be the larly amongst the oldest old and the socially isolated, raise further
normal state for an elderly person is not very different from that of difficulties in psychometric assessment. The imposition of institu-
younger adults’. A study by Deary et al. (2006) associating cogni- tional regimes makes it difficult, for example, to differentiate apti-
tive tests in the older person with white matter lesion density found tude (what the older person is actually able to do) from performance
that this association was significantly modified by IQ at age 11, sug- (that which he or she habitually does in everyday life). This factor is
gesting that cognitive dysfunction may be erroneously attributed to particularly likely to affect measures of the consequences of mental
ageing-related changes rather than inherent ability. illness such as activities of daily living (ADL) scales and informant
measures of performance. The stress associated with the move to
long-term care and the isolated nature of institutional life together
Measurement Issues in may have a significant effect on performance on both affective
and cognitive measures. Performance on cognitive tests has been
Geriatric Assessment shown to drop significantly immediately after entry into an institu-
In developing tests for older populations a number of specific prob- tion (Ward et al., 1990; Ritchie and Fuhrer, 1992), with only partial
lems arise. Perhaps the most important, and yet most neglected, is restitution after a 3-month period. Ward et al. (1990) report a mean
that of the heterogeneity observed within age cohorts. The perform- drop of four points on the Mini-Mental State Examination (Folstein
ance of children is so highly predictable at a given age that it has et al., 1975) and Ritchie and Fuhrer (1992) observed that older peo-
been possible to constitute normative developmental scales which ple with mild dementia living in the community performed bet-
rapidly detect social and cognitive delays and abnormalities. With ter on this test than normal older people living in institutions. The
age, however, standard errors on almost all behavioural measures principal difficulty lies in differentiating true changes in mental
fan out to such an extent that the ‘normal’ performance of older age status which may be due to institutionalization (or to have been the
cohorts is extremely difficult to characterize. This is partly due to cause of institutionalization) from transient adjustment effects.
interindividual differences in inherent ability, increasing variation A general problem has been that tests used with older people are
in physical health, and epigenetic effects. This implies that with age commonly tests developed for use with younger adults. Not only
normal levels of functioning should be established on increasingly is the problem one of content (adapting test materials to older
large samples. In most cases the opposite has been the case so that populations) but also, at a more fundamental level, little thought
normative data at advanced ages are usually unreliable. has been given to the ways in which information processing might
A second issue is the problem of the high prevalence of sensory evolve at higher ages. Theories of cognitive development are pri-
impairment and multiple pathologies in older populations and the marily concerned with childhood changes and it is assumed that
difficulties inherent in developing measures that are independ- cognitive processes once mature in early adolescence do not evolve
ent of these factors. It is known, for example, that sleep disorders, further. Research in this area is clearly needed in order to deter-
which show increasing prevalence with age, may have an important mine whether differences between younger and older adults are
chapter  psychometric assessment in older people 15
due to deterioration or adaptive evolution of cognitive processes. psychometric measures of cognitive performance which have been
For example, small children rely heavily on rote memory which used with older subjects. The table indicates the name of the test, its
requires no analysis of information content. With age there is an more commonly known acronym, the country and the language in
increasing ability to learn by association and condensation; new which it has been developed, and the purpose for which it has been
information is linked with existing information and retained in a developed. The three principal uses of cognitive measures (screen-
summarized form. This permits the retention of larger amounts ing, diagnosis, and assessment of consequences) are now discussed.
of information. Interestingly, assumptions that older people have
poorer memories than younger persons are often based on per- Screening tests
formance on tests of rote recall (for example list learning) rather In medieval Britain the Prerogativa Regis (a Crown document later
than précis recall (requesting the subject to retain a summary of a adopted as common law) established tribunals in 1392 for the
text) on which older persons perform better. screening of cognitive impairment in order to ensure protection of
the afflicted individual and provide assistance in the management
Psychometric Measures of Cognitive of his financial affairs (Tomlins, 1822). It is interesting to note that
this examination consisted of questions to the individual relating to
Functioning in Older People temporal and spatial orientation, memory, calculation, and reason-
Within the field of geriatric psychiatry, psychometric evaluation of ing. The content is in fact strikingly similar to the many screening
cognitive functioning has served three principal purposes: screening tests for dementia in current use.
for cognitive impairment, differential diagnosis of disorders affect- Screening tests for cognitive impairment in older people may
ing intellectual performance, and evaluation of the consequences generally be divided into three categories: (1) brief mental status
of cognitive impairment. Each of these shall be considered in turn. examinations consisting of single-item assessments of orientation,
Table 2.1 provides summary information on most of the validated memory, and reasoning such as the Mini-Mental State Examination
Table 2.1 Psychometric tests developed for the assessment of cognitive performance in older people. The tests are classified according to
function: screening of cognitive disorder (Sc), assessment of a specific cognitive function (S), differential diagnosis (D), assessment of the
impact of therapeutic intervention (T), estimation of premorbid intelligence level (P), or for the evaluation of the consequences of cognitive
disorder (C)
Test name Author Country Function
Activities of Daily Living (N-ADL) Nishimura et al. (1993) Japan C
AD8 Informant Interview Galvin et al. (2005) US Sc
Alzheimer Disease Assessment Scale (ADAS) Rosen et al. (1984) US T
Amsterdam Dementia Screening Test (ADS) De Jonghe et al. (1994) Netherlands Sc
Alters Konzentrations Test (AKT) Geiger-Kabisch and Weyerer (1993) Germany S, C
Behaviour Dyscontrol Scale (BDS) Grigsby et al. (1992) UK S, C
Behavioural Pathology in AD (Behave-AD) Harwood et al. (1998) US C
Behavioural and Emotional Activities in Dementia Sinha et al. (1992) US T, C
Cambridge Examination for Mental Disorders (CAMDEX) Roth et al. (1988) UK C, D
CAMDEX-N (Dutch version) Neri et al. (1994) UK C, D
Clifton Assessment Scale (CAPE) Clarke et al. (1991) UK C
Cognitive Abilities Screening Instrument (CASI) Liu et al. (1994) China Sc
Cambridge Contextual Reading Test (CCRT) Beardsall and Huppert (1994) UK P
Clock Drawing Test (CDT) Ainslie and Murden (1993) US S
CERAD Neuropsychological Battery Welsh et al. (1994) US D
Canberra Interview for the Elderly (CIE) Henderson et al. (1994) Australia D
Computerized Neuropsychological Test Battery (CNTB) Veroff et al. (1991) US C
Cognitive Performance Test (CPT) Burns et al. (1994) US C
Cognitive Screening Test (CST) Ponds et al. (1992) Netherlands Sc
Détérioration Cognitive Observéé (DECO) Ritchie and Fuhrer (1992) France Sc
Dementia Rating Scale (DRS) Rosser and Hodges (1994) UK C, D
East Boston Memory Test (EBMT) Albert et al. (1991) US S, C
(continued)
Table 2.1 (Continued)
Test name Author Country Function

Structural Interview for the Diagnosis of Alzheimer’s type and Morinigo et al. (1990) Spain D
multi-infarct dementias (ENEDAM)
Extended Scale for Dementia (ESD) Helmes et al. (1992) Canada C, D
Functional Assessment Staging (FAST) Sclan and Reisberg (1992) US Sc, C
Gedragsobservatieschool-geriatrie (GOS-G) Gorissen (1994) Netherlands D
GPCOG Dementia in General Practice Brodaty et al. (2002) Australia Sc
Guy Advanced Dementia Schedule (Guy-ADS) Ward et al. (1990) UK D, S
Global Deterioration Scale (GDS) Eisdorfer et al. (1992) US D, C
Hierarchic Dementia Scale Cole and Dastoor (1996) Canada Sc
Hierarchic Dementia Scale (HDS) Ronnberg and Ericsson (1994) Sweden D, C
Hasegawa Dementia Scale (HDS) Gao (1991) China Sc
Hodkinson’s Test Gomez de Caso et al. (1994) Spain D, C
Hodkinson’s Abbreviated Mental Test Rocca et al. (1992) Italy C, Sc
Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE) Jorm et al. (1991) Australia D, Sc
Iowa Screening Test Eslinger et al. (1985) US S, Sc
London Psychogeriatric Rating Scale (LPRS) Reid et al. (1991) UK Sc, C
Mattis Dementia Rating Scale Coblentz et al. (1973) UK Sc
Mémoire de Prose Capitani et al. (1994) Italy C
Memory Impairment Screen (MIS) Buschke et al. (1999) US S
Mini-Mental State Examination (MMSE) Folstein et al. (1975) US Sc
Mini-Object Test Still et al. (1983) US Sc
Modified Ordinal Scales for Psychological Development (M-OSPD) Auer and Reisberg (1996) US C
Mental Status Questionnaire (MSQ) Kahn et al. (1960) US Sc
Nurnberger Alters Inventar (NAI) Pek and Fulop (1991) Germany/ Hungary C, T
National Adult Reading Test (NART) Nelson (1982) UK P, C
NM Scale Nishimura et al. (1993) Japan C
Nurse’s Observation Scale for Geriatric Patients (NOSGER) Tremmel and Spiegel (1993) UK D, C
Neuropsychiatric Inventory (NPI ) Cummings et al. (1994) US D
Observation Psycho Geriatrics (OPG) Duine (1991) Netherlands C
Qualitative Evaluation of Dementia (QED) Royall et al. (1994) US C
R148 Test of cued recall Ivanoiu et al. (2005) Belgium S, Sc
Refined ADL Assessment Scale (R-ADL) Tappen (1994) US C, D
Structured Assessment of Independent Living Skills (SAILS) Mahurin et al. (1991) US C
Structured Interview for the Diagnosis of Dementia of Alzheimer Type, Zaudig (1992) Germany D
Multi-Infarct Dementia and Dementias of other Etiology (SIDAM)
Syndrom Kurztest (SKT) Kim et al. (1993) Germany C, T
Short Portable Mental Status Questionnaire (SPMSQ) Albert et al. (1991) US Sc
Telephone Assessed Mental State (TAMS) Lanska et al. (1993) US D, Sc
Troublesome Behaviour Scale (TBS) Asada et al. (1994) Japan D
Test Your Memory Test Hancock and Larner (2011) UK Sc
(Folstein et al., 1975), the Mental Status Questionnaire (Kahn et al., psychology applied to clinical practice in the field of neuropsychol-
1960), and the Abbreviated Mental Test (Qureshi and Hodkinson, ogy. Quantifiable tasks have thus been developed which are capable
1974), adapted to milder forms of cognitive impairment where of isolating the specific cognitive subsystems affected by diseases
self-administration is possible (Hancock and Larner, 2011); and clinical syndromes, such as working and semantic memory,
(2) abbreviated neuropsychological batteries designed to target spe- attention, and visuospatial organization. Despite the proliferation
cific cognitive functions known to be affected by dementia such as of focalized testing methods now available in the field of cognitive
the Iowa battery (Eslinger et al., 1985) and the Memory Impairment processing research in normal adults, and their demonstrated util-
Screen (Buschke et al., 1999); and (3) informant tests designed ity in differential diagnosis, surprisingly few of these tests are being
to estimate degree of cognitive decline from premorbid levels of carried over into everyday clinical practice. A survey across devel-
functioning such as the proxy questionnaire from the Blessed oped countries suggests that reliance is still predominantly placed
Scale (Blessed et al., 1968), Détérioration Cognitive Observéé on older tests such as the Wechsler Memory and Intelligence Scales
(DECO) (Ritchie and Fuhrer, 1992), the Informant Questionnaire (Sullivan and Bowden, 1997).
on Cognitive Decline in the Elderly (IQCODE) (Jorm and Korten, Neuropsychometric tests targeting specific cognitive processes
1988), and the Cambridge Examination for Mental Disorders of the have now been used in the differential diagnosis of senile demen-
Elderly (CAMDEX) family interview (Roth et al., 1988). tia of the Alzheimer type (Almkvist et al., 1993; Kertesz and
Preference has generally been given to the first type of test, Clydesdale, 1994; Rosser and Hodges, 1994; Park et al., 2011),
undoubtedly because of its high face validity, although the other and subtypes of Alzheimer’s disease (Mann et al., 1992; Richards
two methods have been found to be equally as discriminative. While et al., 1993; Stern et al., 1993; Lundervold et al., 1994), vascular
formerly considered an adjunct to the clinical examination, inform- dementia (Almkvist et al., 1993; Kertesz and Clydesdale, 1994),
ant report has now been demonstrated by a number of researchers frontotemporal degeneration and Lewy body disease (Grossman
to be as highly discriminant in screening for cognitive disorder as et al., 1998; Filley et al., 1994; Park et al., 2011), depression
direct examination of the older people themselves, and less subject to (Masserman et al ., 1992), Huntington’s disease (Masserman
education effects (Jorm and Korten, 1988; Ritchie and Fuhrer, 1992). et al., 1992); Lundervold et al., 1994; Rossor and Hodges, 1994,
Informant methods also appear unaffected by institutionalization progressive supranuclear palsy (Rosser and Hodges, 1994), and
(Ritchie and Fuhrer, 1992). A combination of informant and cogni- Parkinson’s disease (Stern et al., 1993; Lundervold et al., 1994;
tive screening tests has been shown to have better discrimination than Westwater et al., 1997; Park et al., 2011). Cognitive testing has
either method alone (Mackinnon and Mulligan, 1998). Screening for also been used to monitor the effects of adverse environmental
dementia commonly takes place in general practice, for which most exposure in older people such as surgery and anaesthesia (Moller
of the instruments described above have been considered inadequate. et al., 1998; Ancelin et al., 2001).
A number of brief screening tests specifically designed for the gen- The psychometric tests used in the diagnosis of pathologies in
eral practitioner such as the GP Assessment of Cognition (GPCOG), older subjects have varied widely between studies, thus making
the Memory Impairment Screen (MIS), and the Mini-Cognitive comparisons between clinical centres very difficult. In response to
Assessment Instrument (Mini-Cog) have now been validated in this this problem, psychometric tests targeting specific cognitive func-
context (Brodaty et al., 2002; Milne et al., 2008). tions have been incorporated into standardized comprehensive
Most screening tests show quite high levels of discrimination in diagnostic batteries designed for the differential diagnosis of psy-
case-control studies which are typically designed with an equal case chogeriatric illness such as the Cambridge Cognition Examination
to noncase ratio, using normal subjects free of likely confounding (CAMCOG) which forms part of CAMDEX (Roth et al., 1988), the
characteristics, and cases of cognitive impairment which are rela- mental status examination of the Canberra Interview for the Elderly
tively clear-cut. However, performance on these same tests is seen (Henderson et al., 1994), and the cognitive assessment compo-
to drop dramatically when used in the community setting. This is nent of the Structured Interview for the Diagnosis of Dementia of
partly due to the fact that prevalence rates of cognitive disorder in Alzheimer Type, Multi-Infarct Dementia and Dementias of other
the community are much lower than in case-control studies, and Etiology (SIDAM) (Zaudig et al., 1991).
the level of cognitive impairment is often much milder, giving
rise to poorer positive and negative predictive values. Brayne and Measurement of the consequences
Calloway (1991) have demonstrated, for example, that the positive of psychiatric disorder
predictive value of the Mini-Mental State Examination falls from Increasing interest in the impact of psychiatric illness on the
89% when the case–noncase ratio is 1 to 10 to only 59% when it is quality of life, on caregiving services, and on the caregivers them-
1 to 50. Weinstein and Fineberg (1980) pointed out that this prob- selves has led to the more recent development of psychometric
lem can to a large extent be overcome by adjustment of the cut-off tests designed to assess the consequences of cognitive disorder. In
point of a screening test according to the predicted prevalence of this context, terms such as ‘disability’ and ‘dependency’ are often
the disease within the target population. A downward adjustment used, but with little precision. The International Classification
on an informant questionnaire was found by Ritchie and Fuhrer of Functioning, Disability and Health (WHO, 2001) provides a
(1992), for example to improve discrimination in the community useful conceptual framework for the consideration of the con-
setting by 10%. sequences of disease by differentiating three levels: function-
ing (the consequences of disease at the level of body organs and
Diagnostic instruments systems); disability (interference with the activities performed
Psychometric tests may also be used as an adjunct to the differential by the individual); and participation (the social consequences
diagnosis of disorders responsible for cognitive impairment in older of disease). At the impairment level, psychometric tests meas-
people. These tests derive from experimental studies in cognitive ure changes in specific cognitive processes (memory, language,
attention), as discussed above. Disability and handicap scales, on several items rather than one. This gives fewer nodes but allows for
the other hand, describe the impact of cognitive dysfunction on re-entrant nodes in which only a portion of the items in a subtest
behaviour and social adaptation. Examples of this type of scale need to be administered before branching to another. Model-based
are the Neuropsychiatric Inventory, which assesses behavioural branching is based on item response, or latent trait theory, assum-
and emotional changes in dementia (Cummings et al., 1994); the ing that item responses are probabilistically related by a specified
Troublesome Behaviour Scale (Asada et al., 1994); and the Refined function to a continuous underlying trait or ability. Theoretical
ADL Assessment Scale (Tappen, 1994). Other scales assess deteri- models of branching systems and scoring methods for adaptive
oration in daily activities corresponding to specific changes in cog- testing are described in greater detail by Vale (1981). These articles
nitive processing, for example the Functional Assessment Staging also provide practical guidelines for the construction of branching
Scale (Sclan and Reisberg, 1992), the Behavioural Pathology in strategies.
Alzheimer’s Disease Scale (BEHAVE-AD) (Harwood et al., 1998), Reliable data recording has been a persistent problem in both
and the Cognitive Performance Test (Burns et al., 1994). Mixed research and clinical investigations, as it is at this point that both
measures incorporating cognitive, behavioural, and functional conscious and unconscious interviewer bias may exert a strong
domains are also used to provide a broader assessment of change influence. This problem has been repeatedly reported in the lit-
over time, notably in response to new treatments (Holthoff et al., erature relating to behavioural evaluation since the 1940s. Most
2011). Tests have also been developed to monitor residual func- of us are familiar with this type of problem and, no matter how
tioning in severely impaired subjects, based on the Piagetian well interviewers are trained, the investigator can never be sure if
model, such as the Modified Ordinal Scales for Psychological the coded response is truly an accurate representation of the sub-
Development Scale (M-OSPD) (Auer and Reisberg, 1996) and the ject’s behaviour. Computer testing has greatly alleviated this prob-
Hierarchic Dementia Scale (HDS) (Cole and Dastoor, 1996). More lem. Computerized testing provides an interactive environment in
global measures of quality of life have also been attempted: how- which the subject can respond directly to the stimulus via a key-
ever, these remain difficult to validate and often overlap with the board or tactile screen and the response is registered immediately
content of depression scales (for a review see Schölzel-Dorenbos by the programme without the intermediary of an interviewer or
et al., 2007). response coder. In this way, the investigator may incorporate into
his programme a control system through which he may check
Computerized cognitive assessment at the end of a session that all items have been presented by the
Although automated cognitive testing has been reported in the lit- interviewer.
erature since the late 1960s, it has only become popular as a rou- While earlier tests generally only recorded simple information
tine clinical procedure in the past decade, principally due to three such as ‘right’ or ‘wrong’, computer technology has permitted the
important developments: the ability to simulate complex imagery, development of complex automated decision-making. For exam-
the microcomputer, and the touch screen. The most important of ple, the programme may automatically record persistent persever-
these has undoubtedly been the development of the microproces- ation between tasks, where subjects continue to attend to stimuli
sor, which has not only dramatically decreased the cost of auto- relevant to a previous task, or visual-field neglect, where the sub-
mated testing, but also greatly increased its flexibility, such that ject responds only to items in one part of the screen. Direct inter-
users can design their own testing programmes with little expense action between the respondent and the testing apparatus permits
and transfer results to other software for analysis. the accurate recording of reaction times and response latencies.
The most obvious advantage of computerized cognitive test- The latter is of particular interest in follow-up studies, as increased
ing is the possibility of standardizing stimulus presentation—an response time in subsequent administrations of a test is often a
advantage that has led to the computerization of popular manual more sensitive indicator of early cognitive deficit than error rate.
tests such as the Progressive Matrices and Mill Hill Vocabulary In this way, complex observations may be recorded even where
Test (Watts et al., 1982). By the end of the 1960s a review of cog- the examination is carried out by lay interviewers. For example,
nitive tests that had been adapted for computer administration Fagot et al. (1993) described a haptic recognition task in which the
had already appeared in the literature (Gedye and Miller, 1969). computer records the number and duration of hand contacts with
A further advantage of computer administration is significant each stimulus, and the Examen Cognitif par Ordinateur (ECO)
reduction in test time. Computerization also permits the use of cognitive battery for older people (Ritchie et al., 1993) automati-
extremely complex administration procedures which may be cally records visual field neglect and rotation errors in a matching
tailored to suit individual needs. This possibility has led to the to sample task.
development of ‘adaptive’ or ‘tailored’ testing in which the test In the early years of computerized test development, investigators
content is determined for each individual as a function of each (and in particular clinicians) expressed doubts as to the feasibility
response that is made in the course of the testing period. In this of presenting older subjects with computer hardware, and thus fre-
way, difficulty levels can be adjusted according to the ability of quently rejected computerized testing as being detrimental to the
the subject. clinician–patient relationship. On the other hand, even early reports
Item selection algorithms generally follow one of three branch- of the use of computerized testing with older people suggested that
ing models: item to item via predetermined structures, subtest to there is in practice very little difficulty (Carr et al., 1986). In the first
subtest, or as a function of a complex rule specified by a math- place, many older persons now own their own microcomputers and
ematical testing model. Item to item branching strategies are the many others have had some experience with them. Additionally,
simplest forms of adaptive testing with a triangular or pyramidal older people generally find computer-generation of tests far more
structure when drawn graphically. Intersubtest branching strategies interesting and less threatening than paper and pencil tests admin-
are similar except that each node in the diagram now consists of istered by an interviewer—the latter situation is often negatively
associated with school experiences, and the older person often feels used this technology to explore cognitive load and mobility deci-
he or she is being judged by the younger interviewer. sions in older people at risk of falls.
For readers interested in the use of computerized cognitive When adapting existing paper and pencil cognitive tests for
assessment some points are perhaps worth noting. Development of computer administration, reliability and validity should be estab-
a computerized test or battery of tests involves, first, the selection of lished, even where this has previously been done for the manually
both hardware and software. The options available are presently so administered form. Watts et al. (1982) have shown, for example,
numerous that it is not possible to cover them all within the scope that the computerized version of the Ravens Progressive Matrices
of this chapter. Researchers are generally guided by practical limita- gave absolute levels that were approximately 5 points lower than
tions. If the tests are designed for multicentre use, then standardi- obtained by the paper and pencil version of the test. Furthermore,
zation may be an important consideration and preference may be normative data collected using one type of visual display unit may
given to widely used material such as IBM and Macintosh, which not apply if the display type and quality are altered—especially
have user-friendly software packages well suited to the develop- when changing from a cathode ray tube to the liquid crystal dis-
ment and rapid modification of adaptive cognitive tests. If the test plays used in laptop machines. It may also be necessary to test
is to be used in general population studies, light-weight portable alternative administration methods to reduce error due to test pres-
hardware should be considered. A limitation of this has been the entation method. Banderet et al. (1988) compared two versions of
poor quality of the screens, which has subsequently limited their a computerized addition task with the original paper and pencil
use to the scoring of responses. However, with the development of version. The first version required subjects to enter answers from
monochrome LCD displays, the quality is now greatly improved. If a keyboard, and, despite pretest typing practice, subjects were 35%
response latencies are to be recorded, a separate monitor has com- slower with the keyboard than with the paper and pencil version.
monly been used, although the manufacturers of Macintosh and Furthermore, scores obtained from the computer version were less
IBM PC have recently developed laptop models incorporating a stable over time. An alternative computerized multiple-choice ver-
touch screen. sion was found to be not only more stable than either the paper
If response latencies are to be recorded in different research sites, and pencil or original computer task, but also more sensitive to the
then care should be exercised to ensure standardization of hard- experimental condition.
ware. The evolution of computer technology has been accompa- Finally, while, as noted above, subject acceptance is generally not
nied by a rapid increase in the speed of microprocessor operation. a problem, Kane and Kay (1992) stress that previous experience
Variations are therefore likely to exist between computers in the with computers may constitute an important source of variance
accuracy of reaction time or response latency measures, especially in test performance in older people, especially when the subject
if the programme controls timing by ‘delay loops’, as is the case with is required to manipulate several keys on a keyboard. Variation
older computers such as the Apple II series. Alternatively, timing between subjects is likely to be even greater with crosscultural data
may be controlled by the software using the computer’s Time of collection and in groups with a wide age range. It is thus important
Day clock. Even so, with IBM-compatible computers absolute tim- to standardize subject familiarity as far as possible. This should not
ing can still thus only reach one-tenth of a second, which, while be left to the interviewer, who may introduce significant error vari-
generally adequate for the estimation of response time in clinical ance at this point. The test programme should incorporate stand-
studies, may not be adequate for experimental examination of reac- ardized practice trials which bring all subjects up to an equivalent
tion time. For Macintosh computers the ‘tic’ rate permits a timing pretest level of competency in manipulating response devices before
accuracy of 34 ms using software commands to the system clock. commencing the testing procedures.
Timing accuracy can be increased for IBM-compatible computers Table 2.2 provides a list of currently available computerized
through BIOS modifying software (see Graves and Bradley (1991) cognitive tests that are suitable for use with older persons and the
for a description of this procedure) and for Macintosh using spe- hardware required for their administration. A more complete dis-
cial public domain software timing routines as described by Westall cussion of the use of computerized tests in older people is given by
et al. (1986, 1989). The use of iPhones and iPads is currently being Wild et al. (2008).
explored and these have the advantage of far more accurate reac-
tion time and response latency measures.
Display clarity depends upon the graphic standard used by the Conclusion
software. The relative advantages of different standards should be A large number of psychometric tests have been developed for
taken into consideration in the selection of test software. Earlier the evaluation of the mental health status of the older person. In
standards such as CGA (Color Graphics Adapter) give figures with this chapter we have considered developments specifically in the
relatively poor resolution, so that stimuli requiring finer detail or field of cognitive dysfunction and its behavioural consequences.
portraying dimensionality are best programmed by more advanced Computerized testing methods have greatly expanded the func-
graphics standards such as EGA (Enhanced Graphics Adapter) or tions that may be measured, and also increased efficiency and reli-
VGA (Visual Graphics Array). On the other hand, with lower reso- ability. Perhaps the greatest shortcoming at this point in time is the
lution, graphics can be drawn more quickly on the screen and the assumption that information processing in normal older persons
display and response timing of the test is easier to coordinate. More is the same as that for young adults. Little consideration has been
recent developments in Virtual Reality software, now commonly given to the possibility that an upper extension to existing theo-
used in psychiatry for the treatment of phobias, may offer in the ries of cognitive and emotional development may be required (that
future more realistic assessments of complex cognitive processes, is, beyond childhood and adolescence to different phases of adult
decision-making, and the everyday consequences of cognitive life) if psychometric testing is to be adequately adapted to older
impairment. Nagamatsu et al. (2011) have, for example, recently populations.
Table 2.2 Computerized cognitive tests suitable for use with older populations
Test Author Hardware
Abbo Cognitive Performance Test (ACPT) Unpublished* MacIntosh
Adaptive Rate Continuous Performance Test (ARCPT) Buschbaum and Sosteck (1980) Apple II/IBM PC
Automated Portable Test System (APTS) Bittner et al. (1986) IBM compatible
Automated Psychological Screening (B-MAPS) Acker and Acker (1982) IBM/MacIntosh
Cambridge Neuropsychological Test (CANTAB) Sahakian and Owen (1992) IBM PC
Cambridge Mental Disorders of the Elderly (CAMDEX) Roth et al. (1988) IBM PC
Computergestützte Neuropsychologische Testanordnung (CNAT) Unpublished** Atari
CogState (young elderly) Fredrickson et al. (2010) PC
ECO (MacIntosh) COGNITO (PC) in French and English Ritchie et al. (1993) MacIntosh PC
Geriatric Mental State (GMS-AGECAT) Copeland et al. (1986) PDP 11/34
Memory Assessment Clinics Battery (MAC) Larrabee et al. (1991) AT&T 6300
NeuroTrax Mindstreams Dwolatzky et al. (2004) IBM PC
Psychomotor and Visuospatial Tasks Hofman et al. (2000) IBM PC
Selective Reminding Tests (SRTs) Kane and Perrine (1988) IBM PC
TDAS (computerized ADAS-COG) Inoue et al. (2011) IBM PC
Walter Reed Performance Assessment Battery (WRPAB) Thorne et al. (1985) IBM PC
* Abbo Enterprises, 7334 Girard Ave, La Jolla, CA 92037.
** Reischies and Wilms, Psychiatrische Klinik und Poliklinik der Freie Universität, Berlin, 1987.
References Blalock, H.M. (1968). The measurement problem. In: Methodology in social
research (eds H.M. Blalock and A. Blalock). McGraw-Hill, New York.
Acker, W. and Acker, C. (1982). Bexley Maudsley automated psychological
Blessed, G., Tomlinson, B.E., and Roth, M. (1968). The association between
screening and Bexley Maudsley category sorting test: manual .
quantitative measures of dementia and of senile change in the cerebral gray
NFER-Nelson, for the Institute of Psychiatry, London.
matter of elderly subjects. British Journal of Psychiatry, 114, 797–811.
Ainslie, N.K. and Murden, R.A. (1993). Effect of education on the
Brayne, C. and Calloway, P. (1991). The case identification of dementia in the
clock-drawing dementia screen in non-demented elderly persons.
community: a comparison of methods. International Journal of Geriatric
Journal of the American Geriatrics Society, 41, 249–52.
Psychiatry, 5, 309–16.
Albert, M., et al. (1991). Use of brief cognitive tests to identify individuals
Brodaty, H., Pond, D., and Kemp, N.M., et al. (2002). The GPCOG: a new
in the community with clinically diagnosed Alzheimer’s disease.
screening test for dementia designed for general practice. Journal of the
International Journal of Neuroscience, 57(3–4), 167–78.
American Geriatrics Society, 50, 530–4.
Almkvist, O., et al. (1993). Patterns of neuropsychological performance in
Burns, T., Mortimer, J.A., and Merchak, P. (1994). Cognitive performance
Alzheimer’s disease and vascular dementia. Cortex, 29, 661–73.
test: a new approach to functional assessment in Alzheimer’s disease.
Ancelin, M.L., et al. (2001). Exposure to anaesthetic agents, cognitive Journal of Geriatric Assessment and Neurology, 7, 46–54.
functioning and depressive symptomatology in the elderly. British
Buschbaum, M.S. and Sostek, A.J. (1980). An adaptive rate continuous
Journal of Psychiatry, 178, 360–6.
performance test: vigilance and reliability for 400 male students.
Ancelin, M.L., et al. (2006). Non-degenerative mild cognitive impairment Perceptual and Motor Skills, 51, 707–13.
in elderly people and use of anticholinergic drugs: longitudinal cohort
Buschke, H., et al. (1999). Screening for dementia with the memory
study. British Medical Journal, 332, 455–9.
impairment screen. Neurology, 15, 231–8.
Asada, T., et al. (1994). Development of a toublesome behaviour scale (TBS)
Capitani, E., et al. (1994). Standardization and use of a test of memory of
for elderly patients with dementia. Japanese Journal of Public Health,
prose. Bollettino di Psicologia applicata, 209, 47–63.
41(6), 518–27.
Carr, A.C., Woods, R.T., and Moore, B.J. (1986). Automated cognitive
Auer, S.R. and Reisberg, B. (1996). Reliability of the modified ordinal scales
assessment of elderly patients: a comparison of two types of response
of psychological development: a cognitive assessment battery for severe
device. British Journal of Clinical Psychology, 25, 305–6.
dementia. International Psychogeriatrics, 8, 225–31.
Clarke, M., et al. (1991). The prevalence of dementia in a total population: a
Banderet, L.E., et al. (1988). Psychometric properties of three addition tasks
comparison of two screening instruments. Age Ageing, 20, 396–403.
with different response requirements. Proceedings of the 30th Annual
Coblentz, J.M., et al. (1973). Presenile dementia: clinical aspects and evaluation
Meeting of the Military Testing Association, Arlington, VA.
of cerebrospinal fluid dynamics. Archives of Neurology, 29, 299–308.
Beardsall, L. and Huppert, F.A. (1994). Improvement in NART word reading
Cole, M.G. and Dastoor, D.P. (1996). The hierarchic dementia scale:
in demented and normal older persons using the Cambridge Contextual
conceptualization. International Psychogeriatrics, 8, 205–12.
Reading Test. Journal of Clinical and Experimental Neuropsychology, 16,
232–42. Copeland, J.R.M., Dewey, M.E., and Griffiths-Jones, H.M. (1986). A
computerized diagnostic system and case nomenclature for elderly
Bittner, A.C., et al. (1986). Performance evaluation tests for environmental
subjects: GMS and AGECAT. Psychological Medicine, 16, 89–99.
research (PETER): evaluation of 114 measures. Perceptual and Motor
Skills, 63, 683–708. Cummings, J.L., et al. (1994). The neuropsychiatric inventory: comprehensive
assessment of psychopathology in dementia. Neurology, 44, 2308–14.
De Jonghe, J.F., et al. (1994). Differentiation between dementia and functional Henderson, A.S., et al. (1994). A survey of dementia in the Canberra
psychiatric disorders in a geriatric ward of a general psychiatric hospital population: experience with ICD-10 and DSM-III criteria. Psychological
using the Amsterdam dementia screening test. Nederlands Tijdschrift Medicine, 24, 473–82.
voor Geneeskunde, 138, 1668–73. Hofman, M., et al. (2000). Alzheimer’s disease, depression and normal ageing:
Deary, J.J., et al. (2006). White matter integrity and cognition in childhood merit of simple psychomotor and visuospatial tasks. International Journal
and old age. Neurology, 28, 505–12. of Geriatric Psychiatry, 15, 31–9.
Duine, T.J. (1991). Validity of a new psychogeriatric behavior observation Holthoff, V.A., et al. (2011). Validity of the relative outcome scale for
scale for application in nursing homes and homes for the aged. Tijdschrift Alzheimer’s disease: a novel multidomain assessment for daily medical
voor Gerontologie en Geriatrie, 22, 228–33. practice. Alzheimer’s Research Therapy, 3, 3–29.
Dwolatzky, T., et al. (2004). Validity of the Mindstreams computerized Inoue, M., et al. (2011). Touch Panel-type Dementia Assessment Scale: a new
cognitive battery for mild cognitive impairment. Journal of Molecular computer-based rating scale for Alzheimer’s disease. Psychogeriatrics, 11,
Neuroscience, 1, 33–44. 28–33.
Eisdorfer, C., et al. (1992). An empirical evaluation of the global deterioration Ivanoiu, A., et al. (2005). Memory evaluation with a new cued recall test in
scale for staging Alzheimer’s disease. American Journal of Psychiatry, patients with mild cognitive impairment and Alzheimer’s disease Journal
149, 190–4. of Neurology, 252, 47–55.
Embretson, S.E. and Yang, X. (2006). Multicomponent latent trait models for Jackson, H.L., Howard, M.E., and Barnes, M. (2011). Cognition and day-time
complex tasks. Journal of Applied Measures, 7, 335–50. functioning in sleep-related breathing disorders. Progress in Brain
Eslinger, P.J., et al. (1985). Neuropsychologic detection of abnormal mental Research, 190, 53–68.
decline in older persons. Journal of the American Geriatrics Society, 25, Jorm, A.F., et al. (1991). Performance of the Informant Questionnaire on
670–4. Cognitive Decline in the Elderly (IQCODE ) as a screening test for
Fagot, J., Lacreuse, A., and Vauclair, J. (1993). Haptic discrimination of dementia. Psychological Medicine, 21, 785–90.
nonsense shapes : hand exploratory strategies but not accuracy reveal Jorm, A.S. and Korten, A. (1988). Assessment of cognitive decline in the
laterality effects. Brain and Cognition, 21, 212–25. elderly by informant interview. British Journal of Psychiatry, 152, 209–13.
Filley, C.M., Kleinschmidt-DeMasters, B.K., and Gross, K.F. (1994). Non- Kahn, R.L., Godfarb, A.I., and Pellack, M. (1960). Brief objective measures
Alzheimer fronto-temporal degenerative dementia. A neurobehavioural for the determination of mental status in the aged . British Journal of
and pathologic study. Clinical Neuropathology, 13, 109–16. Psychiatry, 117, 326–8.
Folstein, M.F., Folstein, S.E., and McHugh, P.R. (1975). ‘Mini Mental State’. Kane, R.L. and Kay, G.G. (1992). Computerized assessment in
A practical method for grading the cognitive state of patients for the neuropsychology: a review of tests and test batteries. Neuropsychology
clinician. Journal of Psychiatric Research, 12, 189–98. Review, 3, 1–117.
Fredrickson, J., et al. (2010). Evaluation of the usability of a brief Kane, R.L. and Perrine, K.R. (1988). Construct validity of a nonverbal analogue
cognitive screening test in older people for epidemiological studies. to the selective reminding verbal learning test. Paper presented at the meeting
Neuroepidemiology, 34, 65–75. of the International Neuropsychological Society, New Orleans, LA.
Galvin, J.E., et al. (2005). The AD8: a brief informant interview to detect Kertesz, A. and Clydesdale, S. (1994). Neuropsychological deficits in vascular
dementia. Neurology, 65, 559–64. dementia vs Alzheimer’s disease. Archives of Neurology, 51, 1226–31.
Gao, Z. (1991) Assessment of Hasegawa’s dementia scale for screening and Kim, Y.S., Nibbelink, D.W., and Overall, J.E. (1993). Factor structure and
diagnosis of dementia in the elderly. Chinese Journal of Neurology and scoring of the SKT test battery. Journal of Clinical Psychology, 49, 61–71.
Psychiatry, 24, 258–61, 316. Kittner, S.J., et al. (1986). Methodological issues in screening for dementia: the
Gedye, J.L. and Miller, E. (1969). The automation of psychological assessment. problem of education adjustment. Journal of Chronic Diseases, 39, 163–70.
International Journal of Man-Machine Studies, 2, 237–62. Lanska, D.J., et al. (1993). Telephone-assessed mental state. Dementia, 4,
Geiger-Kabisch, C. and Weyerer, S. (1993). The geriatric concentration test. 117–19.
Results of a study of patients over 65 years of age in Mannheim. Zeitschrift Larrabee, G.J., West, R.L., and Crook, T.H. (1991). The association of memory
für Gerontologie, 26, 81–5. complaint with computer-simulated everyday memory performance.
Gomez de Caso, J.A., et al. (1994). Value of Hodkinson’s test for detecting Journal of Clinical and Experimental Neuropsychology, 13, 466–78.
dementia and mild cognitive impairment in epidemiological surveys. Lindeboom, R., et al. (2004). Improved brief assessment of cognition in aging
Neuroepidemiology, 13, 64–8. and dementia. Neurology, 63, 543–6.
Gorissen, J.P. (1994). Structure of the Behavior Observation Scale-Geriatrics. Liu, H.C., et al. (1994). Assessing cognitive abilities and dementia in a
Tijdschrift voor Gerontologie en Geriatrie, 25, 58–62. predominantly illiterate population of older individuals in Kinmen.
Graves, R.E. and Bradley, R. (1991). Millisecond timing on the IBM PC/ Psychological Medicine, 24, 763–70.
XT and PS/2: a review of the options and corrections for the Graves Lundervold, A.J., Karlsen, N.R., and Reinvang, I. (1994). Assessment of
and Bradley algorithm. Behavior Research Methods Instruments and sub-cortical dementia in patients with Huntington’s disease, Parkinson’s
Computers, 23, 377–9. disease, multiple sclerosis and AIDS by a neuropsychological screening
Grigsby, J., Kaye, K., and Robbins, L.J. (1992). Reliabilities, norms and factor battery. Scandinavian Journal of Psychology, 35, 48–55.
structure of the Behavioral Dyscontrol Scale. Perceptual and Motor Skills, Mackinnon, A. and Mulligan, R. (1998). Combining cognitive testing
74, 883–92. and informant report to increase accuracy in screening for dementia.
Grossman, M., Payer, F., and Onishi, K (1998). Language comprehension and American Journal of Psychiatry, 155, 1529–35.
regional cerebral defects in frontotemporal degeneration and Alzheimer’s Mahurin, R.K., DeBettignies, B.H., and Pirozzolo, F.J. (1991). Structured
disease. Neurology, 50, 157–63. assessment of independent living skills: preliminary report of a
Hancock, P. and Larner, A.J. (2011). Test Your Memory test: diagnostic performance measure of functional abilities in dementia. Journal of
utility in a memory clinic population. International Journal of Geriatric Gerontology, 46, 58–66.
Psychiatry, 26, 976–80. Mann, U.M., et al. (1992). Heterogeneity in Alzheimer’s disease: progression
Harwood, D.G., et al. (1998). The Behavioural Pathology in Alzheimer’s rate segregated by distinct neuropsychological and cerebral metabolic
Disease Scale (BEHAVE-AD). International Journal of Geriatric profiles. Journal of Neurology, Neurosurgery and Psychiatry, 55, 956–9.
Psychiatry, 11, 793–800. Manton, K.G. and Stallard, E. (1988). Chronic disease modeling: measurement and
Helmes, E., et al. (1992). An examination of psychometric properties of the evaluation of the risks of chronic disease processes. Charles Griffin, London.
extended scale for dementia in three different populations. Alzheimer’s Masserman, P.J., et al. (1992). The subcortical dysfunction hypothesis of
Disease and Associated Disorders, 6, 236–46. memory deficits in depression: neuropsychological validation in a
sub-group of patients. Journal of Clinical Experimental Neuropsychology, studies of adult psychological development (ed. K.W. Schaie). Guilford
14, 687–706. Press, New York.
Milne, A., et al. (2008). Screening for dementia in primary care: a review Schölzel-Dorenbos, C.J., et al. (2007). Evaluating the outcome of intervention
of the use, efficacy and quality of measures. International Journal of on quality of life in dementia: selection of the appropriate scale
Psychogeriatrics, 20, 911–26. International Journal of Geriatrics and Psychiatry, 22, 511–19.
Minati, L., Grisoli, M., and Bruzzone, M.G. (2007). MR spectroscopy, Sclan, S.G. and Reisberg, B. (1992). Functional Assessment Staging (FAST)
functionalMRI and diffusion-tensor imaging in the aging brain: a in Alzheimer’s disease: reliability, validity and ordinality. International
conceptual review. Journal of Geriatric Psychiatry Neurology, 20, 3–21. Psychogeriatrics, 4, 55–69.
Moller, J.Y., Cluitmans, P., and Rasmussen, L.S. (1998). Long-term postoperative Sinha, D., et al. (1992). A new scale for assessing behavioral agitation in
cognitive dysfunction in the elderly: ISPOCD1 Study. Lancet, 351, 857–61. dementia. Psychiatric Research, 41, 73–8.
Morinigo, A., et al. (1990). Description y validez ‘Test-Retest’ de le ENEDAM. Stern, Y., et al. (1993). Comparison of cognitive changes in patients
Actas Luso Espanolas de Neurologia Psiquitria y Ciencias Afines, 18, with Alzheimer’s and Parkinson’s disease. Archives of Neurology, 50,
396–402. 1040–5.
Nagamatsu, L.S., et al. (2011). Increased cognitive load leads to impaired mobility Still, C., Goldsmith, T., and Mallin, R. (1983). Mini-object test: a new brief
decisions in seniors at risk of falls. Psychology and Aging, 26, 253–9. clinical assessment for aphasi-apraxia-agnosia. Southern Medical Journal,
Nelson, H.E. (1982). National Adult Reading Test NFER. Nelson, London. 76, 52–4.
Neri, M., et al. (1994). Validation of the full and short forms of the CAMDEX Sullivan, K. and Bowden, S. (1997). Which tests do neuropsychologists use?
interview for diagnosing dementia. Dementia, 5, 257–65. Journal of Clinical Psychology, 53, 657–61.
Nishimura, T., et al. (1993) Scales for mental state and daily living activities Tappen, R.M. (1994). Development of the Refined ADL Assessment Scale for
for the elderly: clinical behavioral scales for assessing demented patients. patients with Alzheimer’s and related disorders. Journal of Gerontological
International Psychogeriatrics, 5(2), 117–34. Nursing, 20, 36–42.
Nyberg, L. (1998). Mapping episodic memory. Behavioural Brain Research, Thorne, D., et al. (1985). The Walter Reed Performance Assessment Battery.
90, 107–14. Neurobehavioral Toxicology and Teratology, 7, 415–18.
Park, K.W., et al. (2011). Dementia with Lewy bodies versus Alzheimer’s Tomlins, T.E. (1822). Statutes of the realm. Eyre and Strahan, London.
disease and Parkinson’s disease dementia: a comparison of cognitive Tremmel, L. and Spiegel, R. (1993). Clinical experience with the NOSGER:
profiles. Journal of Clinical Neurology, 7, 19–24. tentative normative data and sensitivity to change. International Journal
Pek, G. and Fulop, T. (1991). The Hungarian version of the Nuremberg of Geriatric Psychiatry, 8, 311–17.
geronto-psychological inventory. Orvosi Hetilap, 132, 2319–22. Vale, C.D. (1981). Design and implementation of a microcomputer-based
Ponds, R.W., et al. (1992) Screening for dementia: validity of the cognitive adaptive testing system. Behaviour Research and Instrumentation, 13,
screening test and the Mini-Mental State Examination. Tijdschrift voor 399–406.
Gerontologie en Geriatrie, 23, 94–9. Veroff, A.E., et al. (1991). A new assessment tool for Neuropsychopharmacologic
Qureshi, K.N. and Hodkinson, H.M. (1974). Evaluation of a ten-question research : the computerized Neuropsychological test battery. Journal of
mental test in the institutionalized elderly. Age and Ageing, 3, 152–7. Geriatric Psychiatric Neurology, 4, 211–17.
Raj, C.A., et al. (2012). White matter lesions and brain gray matter volume Ward, H.W., et al. (1990). Cognitive function testing in comprehensive
in cognitively normal elders. Neurobiological Aging, 33(4), 834.e7-1. geriatric assessment: a comparison of cognitive test performance in
Reid, D.W., et al. (1991). On the clinical value of the London Psychogeriatric residential and clinical settings. Journal of the American Geriatric Society,
Rating Scale. Journal of the American Geriatrics Society, 39, 368–71. 38, 1088–92.
Richards, M., et al. (1993). Patterns of neuropsychological performance in Watts, K., Baddeley, A.D., and Williams, M. (1982). Automated tailored testing
Alzheimer’s disease patients with and without extrapyramidal signs. using Raven’s Matrices and Mill Hill Vocabulary Tests: a comparison with
Neurology, 43, 1708–11. manual administration. International Journal of Man-Machine Studies,
Ritchie, K. and Fuhrer, R. (1992). A comparative study of the performance of 17, 331–44.
screening tests for senile dementia using receiver operating characteristics Weinstein, M.C. and Fineberg, H.V. (1980). Clinical decision analysis. WB
analysis. Journal of Clinical Epidemiology, 45, 627–37. Saunders, Philadelphia.
Ritchie, K., et al. (1993). Computerized cognitive examination of the elderly Welsh, K.A., et al. (1994). The Consortium to Establish a Registry for
(ECO): the development of a neuropsychological examination for clinic Alzheimer’s Disease (CERAD). Part V. A normative study of the
and population use. International Journal of Geriatric Psychiatry, 8, 700–9. neuropsychological battery. Neurology, 44, 609–14.
Rocca, W.A., et al. (1992). Validation of the Hodkinson abbreviated mental Westall, R., Perkey, M.N., Chute, D.L. (1986). Accurate millisecond timing on
test as a screening instrument for dementia in an Italian population. Apple’s MacIntosh using Drexel’s millitimer. Behavior Research Methods,
Neuroepidemiology, 11, 288–95. Instruments and Computers, 18, 307–11.
Ronnberg, L. and Ericsson, K. (1994). Reliability and validity of the hierarchic Westall, R., Perkey, M.N., and Chute, D.L. (1989). Millisecond timing on
dementia scale. International Psychogeriatrics, 6, 87–94. the Apple MacIntosh: updating Drexel’s millitimer. Behavior Research
Rosen, W.G., Mohs, R.C., and Davis, K.L. (1984). A new rating scale for Methods, Instruments and Computers, 21, 540–47.
Alzheimer’s disease. American Journal of Psychiatrics, 4, 1356–64. Westwater, H., et al. (1997). Implicit learning in Parkinson’s disease: evidence
Rosser, A.E. and Hodges, J.R. (1994). The dementia rating scale in Alzheimer’s from a verbal version of the serial reaction time task. Journal of Clinical
disease, Huntington’s disease and progressive supranuclear palsy. Journal Experimental Neuropsychololgy, 20, 413–18.
of Neurology, 241, 531–6. Wild, K., et al. (2008). Status of computerized cognitive testing in aging: a
Roth, M., et al. (1988). CAMDEX: the Cambridge Examination for Mental systematic review. Alzheimer’s Dementia, 4, 428–37.
Disorders of the Elderly. Cambridge University Press, Cambridge. World Health Organization (2001). International classification of functioning
Royall, D.R., et al. (1994). Bedside assessment of dementia type using the disability and health. WHO, Geneva.
qualitative evaluation of dementia. Neuropsychiatrics, Neuropsychology Zaudig, M. (1992). A new systematic method of measurement and diagnosis
and Behavavioural Neurology, 6, 235–44. of ‘mild cognitive impairment’ and dementia according to ICD-10 and
Sahakian, B.J. and Owen, A.M. (1992). Computerized assessment in DSM-III-R criteria. International Psychogeriatrics, 4 Suppl 2, 203–19.
neuropsychiatry using CANTAB: discussion paper. Journal of the Royal Zaudig, M., et al. (1991). SIDAM: a structured interview for the diagnosis of
Society of Medicine, 8, 399–402. dementia of the Alzheimer type, multi-infarct dementia and dementias
Schaie, K.W. (1983). The Seattle Longitudinal Study: a twenty-one year of other aetiology according to ICD-10 and DSM III-R. Psychology and
exploration of psychometric intelligence in adulthood. In: Longitudinal Medicine, 21, 225–36.
CHAPTER 3
The sociology of ageing
Ricca Edmondson
Ageing as a Social Process This contributes to an ambiguity about ageing which colours
social reactions to it, leading to sharp fluctuations in public debate.
The sociology of ageing interrogates the social circumstances that Ageing may be portrayed in terms of new forms of risk and dan-
allow human groups to flourish, the social choices that shape, sup- ger, depicted as generating an alarming social burden. In reaction,
port, or distort life-course developments, through to the micro- modern science is sometimes envisaged as a source of potentially
interactions that make up people’s experiences of their own ageing, ageless vitality, the mythological ‘fountain of youth’. These debates
and the social and cultural practices generating shared expectations themselves are social phenomena, attesting to contrasting concep-
about what ageing involves. The field has evolved in response to tions of ageing itself in different arenas.
successive attempts to come to terms with the complexity of these Examining human ageing in different cultural and temporal
phenomena. contexts illuminates the depths of recent change. By the 2030s,
Social environments and relationships not only impinge on the shape of the human population ‘pyramid’ will have taken just
ageing in all its aspects, both positively and negatively, but also over a century to shift from the broad-based triangle typical of
in part constitute it. Physical, conceptual, and experiential aspects most mammals to an approximate rectangle, with roughly equal
of ageing depend in different ways on the social settings in which numbers of individuals at each life-stage (Treas, 1995). Olshansky
they occur. Healthy ageing is associated with status and relative et al. (1993) term this an unprecedented experiment by human-
power; the higher someone is in a social structure, the more likely ity on itself—in terms of both added longevity and bodily changes
that person is to live longer (Donkin et al., 2002), as well as to feel involved. Yet it does not mean that older adults in past societies
that ageing makes it possible to fulfil personal goals (McGee et al., were necessarily rare: the most drastic changes in survival rates
2011). Much of this applies too between societies and over time. have affected children under five. In the early sixteenth century, an
Standard life-courses in the West contrast with those before the English aristocrat who had survived to be 21 could expect to live
Industrial Revolution, partly because modern societies involve to 72—the ‘three score years and ten’ conventionally anticipated
new living conditions, not least practices relating to nutrition, (Lancaster, 1990). A profound but often-ignored impact of popu-
hygiene, and housing. ‘Viewed in the early 1950s as shaped almost lation ‘rectangularization’ is its dissociation of death from birth.
entirely by biological processes and medical care, physical health In the past, death occurred most often in connection with child-
and illness are now understood to be as much or more a func- birth or childhood; mothers and children who survived still had to
tion of social, psychological, and behavioural factors’ (House, contend with the deaths of other infants or siblings. Now, death is
2001: 125). But social resources, attitudes, and habits also directly envisaged as happening at the opposite end of life from birth. This
affect what it means to be an (older) person. If older people are is a massive shift in social perception, but its implications remain
envisaged as incompetent, not only may barriers be constructed to be explored.
that constrain their activities, but also they may ‘auto-stereotype’ Within these overall circumstances, contrasts between tradi-
themselves (cf. Stuckelberger et al., 2012). It may be true that tional and modern forms of ageing take contradictory forms.
a person’s chronological age itself tells us little about that indi- Welfare states offer older people degrees of protection never before
vidual, but social ideas about ageing impact decisively on older known; on the other hand, western societies tend to envisage age-
people and the possibilities open to them (Bytheway, 2011)—in ing as a homogeneous development, treating older individuals as
fluctuating ways. The fact that societies constantly change makes comparatively insignificant members of a mass process. In reaction,
ageing itself a variable process, in which life-courses are enacted the ‘life-course’ approach to the sociology of ageing distinguishes
in contrasting forms. interactions between work, health, family membership, and other
Ageing processes are thus multifaceted ‘collective achievements’ factors in the entwined experiences of individuals and groups over
(Schwanen and Ziegler, 2011: 273). But the social circumstances of time—highlighting both the variability of these pathways and their
late industrial societies are highly untypical as far as human life- connections to larger social contexts.
times are concerned. In consequence, it cannot yet be entirely clear This perspective on ‘population ageing’ draws attention to fluctu-
what to expect of ageing—that of societies or that of individuals. ations in the social interpretation of what ‘ageing societies’ involve.
Popular debate ignores the fact that they would not contain rela- restrict the capacities attributed to them (Kunow, 2010). The stere-
tively large proportions of older people without an equally striking otypical perception and treatment of older people may actually
development: dramatic decreases in the readiness to have children. have hardened during the twentieth century (Hareven, 1982; Katz,
If the UK birth rate returned to the level of the third quarter of 1996), affected by twentieth-century states’ bureaucratic practices
the twentieth century, ‘the population’ might no longer count as and their impacts on the ‘social imaginaries’ (Gaonkar, 2002) of
‘ageing’, whether longevity rates increased or not (Bytheway, 2011: entire societies. Age became treated as a ‘master trait’, a feature
163ff ). But social views on the desirability or otherwise of having whose possession dominates perception of the person who has it.
children fluctuate. In the 1930s, fears about European popula- It is hard to disguise, signalling a chronological trajectory against
tion decline bolstered official support for family growth in some which individuals’ achievements and qualifications can be meas-
countries. Now, public discourses supporting the choice to remain ured. A whole concatenation of official documents, birthdays, and
‘child-free’ compete with exhortations about the ‘social responsibil- time-related social expectations make age seem more important
ity’ to have children (Institut fűr Demoskopie Allensbach, 2003). than it has ever done (Bytheway, 2011). Yet this trope in social
Individuals’ views of their own life-courses, both prospectively in judgement grossly oversimplifies the ‘multidimensional’ ways in
terms of the choices they make and retrospectively when they look which life-courses are actually lived out, ‘structured by virtue of
back on them, change correspondingly. Older women who have the order and timing of multiple social roles over the life span’
raised families may look back on lives without careers and conclude (Macmillan, 2005: 6).
they have ‘done nothing’ (Edmondson, 2011). Social changes can ‘Ageing’ thus takes different forms, and is imagined and experi-
make it hard for individuals to interpret their own life-courses, and enced in contrasting ways, in social settings that change over time.
this applies at the level of whole societies too. It is a variable, life-long process intertwined with the life-courses
Attitudes to child-bearing itself may be ambiguous, but since followed by contemporaries and by earlier and later generations.
infant, child, and maternal mortality rates are now so low in the We shall explore some impacts on ageing made by a key feature
West, they offer little room for dramatic improvement; leaps in of contemporary change—globalization—before interrogating
population-wide life-expectancy seem unlikely to occur here in theories of ageing, and their impacts on the ways in which differ-
future. Increased life-expectancy among adults has changed sig- ent aspects of ageing can be explored sociologically.
nificantly, but more slowly. In Ireland, for example, where social
change was relatively limited between the 1920s and the 1980s, Global(ized) Ageing: Migration, Cultures,
especially for older generations, there was little improvement in the
age at death of older men at all (Fahey et al. 2007, Fig. 4). Adult
Policies, and Perceptions
life-expectancy rates needed appropriate circumstances and poli- Examining ageing in global terms underscores the contrasting sta-
cies if they were to improve. tuses, constraints, opportunities, and expectations associated with
Life-course analyses accentuate awareness that much that hap- being ‘old’ in different societies. In addition, contemporary glo-
pens to individuals in later life, as well as how they themselves balization itself, with its accentuated levels of social and economic
regard their own lifetimes, follows from what they have done, or interdependence and its distribution of work and workers beyond
what has happened to them, at earlier stages. These impacts range national boundaries, produces new patterns of location and dislo-
in scale from the effects of war or peace to educational habits for- cation (Urry, 2007). Global migration paths stretch family networks
merly prevalent, from health-and-safety conditions in previous and life-courses between continents; older people may be left with-
places of work to expectations about the role of family. To be a out family carers in their place of origin, or migrate to new destina-
single, middle-class woman in her twenties, ‘29 and unmarried’, ‘a tions and grow older there, contributing to the cultures of their new
failure’ without children—as Virginia Woolf put it in 1911 (Woolf, societies but affected by local cultural tensions and inequalities in
1975: 466)—was to be engaged on a different life-course trajectory social structure (HelpAge International, 2000; Yeates, 2005). Thus
from that of a woman of similar age a century later. ‘global care chains’ can arise in which
The social conditions of children and adolescents in a given soci- an older daughter from a poor family . . . cares for her siblings while
ety are related to the way adulthood and old age is perceived in that her mother works as a nanny caring for the children of a migrating
society; and, conversely, the role and position of adults and older nanny who, in turn, cares for the child of a family in a rich country.
people is affected by the treatment and roles of those in earlier stages (Hochschild, 2000: 131)
of life. Such ‘care chains’ impinge on the staffing of nursing homes, a
(Hareven, 1982: 3)
burgeoning industry that in the US alone employs more people
Hence the need for contextual approaches to understand- ‘than the auto and steel industries combined’ (Folbre, 2002: 186).
ing older people in the large society, older individuals, and the This brings together two potentially vulnerable groups, older peo-
different forms attributed to the phenomenon of human ageing ple and migrant, often female, workers, and adds the potential for
itself. ‘Against the backdrop of history, changes in people’s lives cultural and linguistic strain to dangers of financial exploitation
influence and are influenced by changes in social structures and (Walsh and O’Shea, 2010). These strains may be accentuated by
institutions,’ giving rise to ‘reciprocal changes’; these in turn are changes in family structures in nations with decreasing numbers
‘linked to the meanings of age, which vary over time’ (Riley et al., of children. In Japan, for example, a need is growing for migrant
1999: 327). labour to provide elder care; this does not sit easily with traditional
Having a life-course itself is changing; it is affected by changes in attitudes to immigration (Powell, 2011: 47).
global and local economies and patterns of work, family and com- ‘Host’ nations may receive ‘waves’ of migrant workers from dif-
munity relations, religious and other sources of meaning. Yet older ferent origins; the circumstances and lifestyles associated with
people often remain constrained by stigmatizing expectations that ageing in their countries of origin may contrast with possibilities
chapter  the sociology of ageing 25
available in their new settings (Wingens et al., 2011). In the UK, rates for the oldest women are not improving at all (Rau et al., 2008:
Bangladeshi migrants, among others, may feel torn between the 757, 764–5). The oldest forms of aging can in principle become
need for economic survival and health support and, on the other healthier. But trends do not continue automatically; they depend, in
hand, the wish to maintain ideals of family belonging vis-à-vis part, on social circumstances and public-policy decisions.
relatives many thousands of miles distant. Attitudes to space and Thus Christensen et al. (2009) stress that what used to be seen as
place have become ‘underlain by deep feelings of rift and division’ objective ‘ageing processes’ are ‘modifiable’ through policy choice.
(Gardner, 2002: 220). Silverstein and Attias-Donfut (2010: 185) They stress capacities to live both longer and better: ‘most evidence
argue that for people aged younger than 85 years suggests postponement of
older adults who migrate to new international locations with, or fol- limitations and disabilities’, despite increases in chronic conditions
lowing, their adult children, often lack the linguistic and cultural capi- (2009: 1204). They attribute this to progress in factors such as hous-
tal to integrate fully into their new surroundings and consequently ing standards, transport, education, gender roles, and social atti-
suffer from loneliness and depression. tudes to disability, stressing that, overall, 30–40% of people in their
Irrespective of possible net gains to intergenerational networks nineties still live independently (Christensen et al., 2009: 1204–5).
from such moves, these are complex phenomena with mixed, some- Lubitz et al. (2003) show that although older people aged 70 have
times contradictory impacts on individuals. A higher proportion of a longer life expectancy if they are in better health than if not, this
UK older people have chosen to live abroad for their retirement need not necessarily mean that their total healthcare requires more
rather than live in nursing homes (Warnes, 2009), but they may expenditure. Improving the health of middle-aged populations
regard these choices differently as they age. A new research agenda could, in appropriate policy settings, enable them to live longer
now explores how family dynamics may stretch across continents, without costing health services more (Evans et al., 2001).
with ties and networks maintained or lost (Phillipson, 2003a). But the globalized politics of economics impacts on local forms
Globalization itself can have highly stratified effects on older of ageing through affecting national policies. During the twentieth
generations (Estes and Phillipson, 2002), while local social policies century, many states took innovative steps to reduce risks histori-
clearly affect provisions and possibilities available to older people cally associated with age. Estes and Phillipson (2002) argue that
(Goodman and Harper, 2008). Welfare-state and other forms of this progress is now being eroded, as current forms of globaliza-
social support have become precarious, where governments and tion undermine institutions protecting older people; international
employers embrace neoliberal policies, often in response to expecta- agencies such as the World Bank and OECD currently favour pri-
tions from global institutions (Yeates, 2001); older people may more vatization in elder care. Such global policies penetrate directly to
often be expected to manage their own ageing, under conditions of individual level. If profit-based forms of older care invoke rigid
heightened insecurity (Phillipson, 2009). In such circumstances, temporal accountancy, for example, informal practices in care
cultural resources to which they have access may either exacerbate homes change, rapidly constraining the everyday practices crucial
or mitigate their predicaments. Fox (2005: 482), for example, argues in making residents feel at home.
that while economic and political structures strongly influence Against this backdrop, ‘global ageing’ itself is often portrayed as a
how ageing is experienced, this does not preclude taking account problem, triggering disquiet about increasing proportions of older
of ‘cultural and social contexts’ to examine ‘confluences of biology, people in modern populations:
culture, material resources and aspects of social organization’ in Alarmist demography and gerontological knowledge came together in
the ways individuals respond to ageing. He seeks ‘to illuminate the the social surveys of the late nineteenth and early twentieth centuries
active, constructive work’ older people do ‘to find ways to age in that decried the growth rate and poverty of the elderly generation as
their culture’ (2005: 484), contrasting Thailand with Australia. In the an economic and moral crisis.
strongly Buddhist context of Thailand, filial care is experienced as (Katz, 1996: 69)
a ‘normative expectation’; in Australia, views of the life-course are These developments are reflected in demographic calculations
much less strongly related to religious values, and older people in where age and ‘dependency’ are conflated. Global dependency
nursing homes may see care as ‘a service to be provided in a profes- ratios are themselves a form of social perception, ‘measured’ by
sional manner’ (2005: 493). In divergent ways, social settings offer counting proportions in a population who are (say) below 15 and
cultural resources affecting ageing (cf. Torres, 2011), here enabling above 65 years of age. This assumes, counterfactually, that all those
recipients of care to ascribe themselves a measure of autonomy in between are gainfully employed. It omits people who may be
rather than perceiving themselves solely as dependent. disabled, and those on ‘home duties’. It also equates ‘dependency’
This does not negate the impact of sociopolitical factors on ageing, with not earning a monetary wage, ignoring other forms of social
reflected in survival rates in different national settings. Comparing input such as childcare, or mutual care among family members or
figures for the Czech Republic, France, East Germany, West Germany, neighbours. This has the political effect of erasing perception of a
Japan, and the US, Rau et al. (2008) found that the chances that peo- wide variety of forms of social support and citizenship. Bytheway
ple who reach the age of 80 will also reach 90 on average more than (2011: 173) points out that people who are economically employed
doubled during the last half of the twentieth century. However, sur- also ‘depend’ on older people (and children) for unpaid domestic
vival rates are far from uniform internationally: by far the highest are and caring work.
found in Japan. In contrast, until 1990 the Czech Republic and East McGee (2002) contends that even if crude calculations of ‘depend-
Germany deviated from the overall pattern, with very low survival ency’ are made on the basis of age alone, dependency ratios are not
probabilities. After the fall of the Iron Curtain, figures for the Czech necessarily rising; in Canada in 1951, the total dependency ratio
Republic remained relatively low, but East Germans, inhabiting a reached a level expected to remain virtually unchanged in 2014.
transformed sociopolitical context, have now caught up with West Using age ratios to generate widespread ‘moral panic’ thus amounts
Germans; in the US, in contrast, the rising trend has halted: survival to an ‘ageist’ campaign against older people. From a sociological
viewpoint, this campaign itself demands analysis. Global variations for instance in activities such as grandparenting or belonging
in patterns of ageing, and contrasts in how ageing is understood, to a church (cf. James et al., 2006 for a contemporary applica-
expose deep-rooted contestation about how the life-course itself is tion). Identifying such contributions seemed to adjure older peo-
regarded. ple to keep involved and energized, using leisure time creatively.
Relatedly, the ‘continuity’ approach to ageing (Atchley, 1989a: 183)
Perspectives on Ageing: saw maintaining continuity with past subjective and social experi-
ences as ‘a grand adaptive strategy’ by which individuals can defend
Theories and Stances themselves from structural disadvantage and ageism; this was
Seeking insight into different facets of ageing, sociologists have interpreted as advice for preserving individual identity over time
explored not only its social circumstances, but also discourse about (Atchley, 1989b; cf. Nuttman-Shwartz, 2008, for contemporary
ageing itself—including their own past and present debates. The uses). These positions share three features. All contain prescriptive
wish to avoid, suppress, or disguise ageing can be accentuated by elements; all have implications on the micro- as well as the mac-
the ‘social imaginaries’ of the time: that is, the underlying roscale, with individuals expected to bear the burden of problems
ways in which people imagine their social existence, how they stemming from social structures. And each contains observations
fit together with others, how things go on between them and their that are convincing in part: ‘disengagement’ applies to some people
fellows, the expectations that are normally met, and the deeper nor- in some circumstances, while supporting ‘continuity’ of lifestyle for,
mative notions and images that underlie these expectations. say, nursing-home residents has considerable appeal. As heuristics
(Taylor, 2002: 106) directing attention in research practice, if not as universal accounts
Achenbaum (1995) stresses that the development of academic of ageing, each has productive potential.
attention to ageing, just after World War II, occurred at a time of The question of theories’ apparent scale may thus be less sig-
flight from trust in politics towards faith in science as a harbin- nificant than their positions on prescriptiveness, power, and social
ger of progress. This accentuated the ‘biomedicalization’ of ageing determinism. Modernization theory (Cowgill, 1974), for example,
(Estes and Binney, 1989), in which ageing is perceived not just as a contends that industrial societies devalue the experience and knowl-
problem, but also as a medical problem: the ‘anti-ageing’ movement edge of earlier generations, impressed instead by fast-changing new
remains a current attempt to ‘cure’ it through biomedical means. skills associated with younger people. This damage to older peo-
Since ageing changes in so many dimensions, from its physi- ple’s status is described so encompassingly as to appear irremedi-
cal parameters to the ways it impinges on experience, sociological able. This position too was soon criticised as unduly universalistic,
theories of ageing are constantly developed and adjusted in efforts homogenizing a vast range of different experiences and situations,
to capture evolving interactions of different kinds. They also reflect as well as idealizing the position of older people in earlier, rural
developing accents and methods within sociology itself (Phillipson societies. We might add that its large-scale imputations are heav-
and Baars, 2007; Bengtson et al., 2009). But underlying reasons for ily inspired by induction from individual experience. Exchange
adjustments in theoretical approach embody conflicting views on theory also relates a society-wide perspective to particular interac-
social causality—how social structures and power affect individual tions, connecting older people’s social status to resources they can
agency and vice versa—and the locus of responsibility for life-course offer others (Dowd, 1980). Neither of these approaches is intrinsi-
events. Theoretical stances differ in terms of their capacities to take cally prescriptive, but each can in principle illuminate certain phe-
account of these complexities, and in terms of the extent to which nomena. Though older people’s social resources frequently seem
they endorse critical stances vis-à-vis extant social impacts on age- depleted from an ‘exchange’ viewpoint, in the semitraditional rural
ing (cf. Biggs et al., 2003). The surface contrast between larger-scale society of Ireland in the 1930s, many (especially older men owning
and more individual-level approaches is thus to some extent farms or small businesses) did possess powerful resources in terms
misleading. of others’ life-courses, and had status to match (Arensberg and
An early attempt to theorize ageing, the ‘disengagement’ Kimball, 1940/2001). In contrast, ‘age stratification theory’ seri-
approach taken by Cumming and Henry (1961), adopted the func- ously interrogated the direction of social causation on individuals’
tionalist analysis common among sociologists at the time. The behaviour, identifying roles prescribed for people in different age
authors explained separate sections of society in terms of contri- ‘strata’ and the ideological pressures involved, but at the same time
butions to the ‘functioning’ of the whole, seeing individuals’ age- it advocated resistance to such pressures (Riley and Riley, 2000;
ing as adapting to changing functions over the life-course. Ageing Dannefer et al., 2005).
and retirement appeared as the secession of older generations from Such contrasting approaches cumulatively made clear how
productive social roles, making room for younger ones. This work strongly social processes affect ageing at all levels; increasingly,
attracted objections for methodological reasons, but its moral and therefore, possibilities for influencing these processes were explored.
political implications for individuals caused particular contention. The ‘political economy of old age’ (Estes, 1991) focused on state
Cumming and Henry were read as supporting an economistic view decisions on distributing economic resources, radically affecting
of relations between individuals and society, accepting that people the lives of older people—in effect creating ‘structured dependency’
of retirement age are of negligible social importance and ignoring (Townsend, 1981: 77). Inequality in old age needed to be under-
the meaning of older age to individuals (Hochschild, 1976). stood in terms of socioeconomic processes tending to marginalize
The career of disengagement theory illustrates the easy slippage older people (Walker, 1981, 1996); they should not be regarded as a
between theories of how social ageing arises, and prescriptions homogenous group sharing problems resulting from chronological
for individuals’ own ageing. In contrast to Cumming and Henry, age, for their different problems had different causes, to be tracked
‘activity theory’ (Havighurst and Albrecht, 1953; Havighurst et al., empirically. Poverty among older people resulted from experience
1963) focused on the positive social ‘roles’ played by older people, of the labour market and subsequent exclusion from it, or from low
levels of social benefits. Ageing in any given society was fundamen- human experience (Biggs, 2004), and older individuals’ capacities
tally affected by the economy, state, and labour market, but also to use social meanings in creative ways. Blaikie (2002) argues that,
class, gender, and ethnicity. This approach harnessed interdiscipli- as an alternative to cultures of ‘consolation’ or ‘incorporation’, older
nary resources to explore such impacts on ageing under the aegis of individuals can adopt ‘cultures of resistance’ to homogenizing social
‘critical gerontology’ (Phillipson, 1998). processes. For Kunow (2009), the ‘transformative power’ of ageing
In this tradition H.R. Moody (1988: 32) argues that gerontology sometimes depicted in literature highlights special forms of resist-
itself should offer an ‘emancipatory discourse’, proffering ‘a positive ance, in which older people may reach heights of courage to which
idea of human development: that is, ageing as a movement toward they were unable to aspire at earlier life-stages.
freedom beyond domination (autonomy, wisdom, transcendence)’. Such work signals eagerness to underline the potential for older
Thomas Cole (1992: 237) too contends that analysing ageing in people’s active intervention in their own predicaments. Gilleard
purportedly technical terms alone amounts to a moral and political and Higgs (2000) and Rees Jones et al. (2008: 1) view consumption,
position in itself, relegating ageing to a (doomed) attempt to evade what people buy, not as a mere side-effect of work, but as possess-
disease rather than confronting ‘conflict, mystery and suffering in ing ‘a significance of its own’; they see modern retirees not just as
late life’. Thus ‘ageing’ is an irreducibly moral concept. This ‘human- ‘pensioners’ but as ‘a putative leisure class’. This approach explores
istic’ understanding of ageing includes explicating how moral posi- whether at least some older people can situate themselves powerfully
tions are generated, endorsed, or precluded by particular social in society, with heightened resources for defence against ageism,
constellations. Theories of ‘moral economy’ show that social norms, profiting from bodily appearances that are socially acceptable, and
for instance prescribing obligations we owe each other within fami- health and resources enabling them to engage in leisure activities
lies or between generations, are themselves influenced by political formerly associated with younger generations. This approach also
economies of ageing (Kohli, 1987; Minkler and Cole, 1991). Public offers to modify Mannheim’s earlier work on generations, no longer
views on the legitimacy or desirability of certain ways of behav- seeing them as discrete, opposing horizontal groups, but postulat-
ing reflect ideas about power and justice, not least justice between ing that succeeding generations both take with them elements of
generations: to what extent should older generations sacrifice their previous generations’ views and contribute new, contrasting habits
interests for those of younger ones, for instance? and behaviour.
Consciousness of society-wide phenomena needed repeated Reference to ‘generations’ connects sociological approaches to
counterbalance, exploring variations among forms of ageing and ageing explicitly with conceptions of time (see Baars, 2009). Elder
their social settings. Hareven (1984) opposed seeking linear pat- (1998) stresses that trajectory over time gives their form and mean-
terns of development in accounts of ageing, given the differential ing to life-course transitions; ‘life-course theory’, which McDonald
impacts of race, class, ethnicity, and family form on intergenera- (2011: 1186) describes as the current ‘vanguard gerontological
tional relationships in different times and circumstances. Using theory’, tracks intertwining trajectories that intersect through indi-
the adaptation of Schutzian phenomenology introduced by Berger viduals’ lifetimes, both responding to social contexts and affecting
and Luckmann (1966), subsequent accounts of ageing stressed the them (see Wingens et al., 2011 for a life-course approach to eth-
effects of ‘the social construction of reality’: the impact of social nicity). Early work in this tradition deployed biographical methods
practices on how ageing and older people are envisaged, affect- to yield insights into the life-worlds of people living out compa-
ing both opportunities and constraints in practice. Hence much rable trajectories, notably artisan bakers in France (Bertaux and
sociological work ‘deconstructs’ everyday experiences to track how Kohli, 1984). Bengtson et al. (2005: 493–4) stress the life-course
social interaction shapes them. Such efforts by no means entail a approach’s potential not only to explore ‘linked lives’ within cohorts
relativistic approach to the philosophy of science; most sociologists and between generations, within their historical contexts, but also
of ageing show an assiduity in seeking optimal forms of evidence to highlight ‘agency and the idea that planfulness and effort can
and argument that is implicitly compatible with critical realism affect life outcomes’.
(Sayer, 2000) in respect of their views about the empirical world. In practice, life-course analyses have been combined with a
Thus Corner and Bond (2006), discussing effects of the diagno- range of positions on the issue of individual freedom to affect the
sis of ‘mild cognitive impairment’, point to potent social effects of course of ageing. Kohli (2007: 254), for example, sees the life-course
applying this ‘label’ to older individuals: they may both be treated approach as focusing ‘on the patterning of time in the various life
differently and come to regard themselves differently as a result. domains’, intended to result in ‘an account of the longue durée’—
As Phillipson’s exposition of critical gerontology stresses (1998; the way ‘typical’ life-courses have changed as between pre- and
Edmondson and von Kondratowitz, 2009), this means attending to late-industrial societies in the West, pre-eminently in response to
both economics and power, and meaning and the social processes changes in the structure of work. This analysis sees ‘the experience
by which it is created and changed. of the temporality of life’ as ‘a feature of an evolved structural order’
The variety of approaches summarized as ‘cultural’ gerontology in which individuals move through sequences of social positions,
hence accentuate attention to social origins of meaning and experi- prescribed to them via rules and norms in succeeding historical set-
ence, and their effects on ageing in contemporary societies. Symbols tings (Kohli, 2007: 255). Central questions concern the role of ‘indi-
such as stories and images relating to ageing (Hepworth, 2000) cast vidualization’ in contemporary societies and the extent to which
light on the ways in which people develop a sense of who they are as it ‘frees’ individuals from ‘the bonds of status, locality and family
they age, and how older people are seen by others. This ascription of origin’ (Kohli, 2007: 255)—or, on the contrary, merely exposes
of cultural meanings profoundly affects people’s experiences even of them to larger-scale social pressures.
their own bodies; culture, in other words, involves power (Andersson, Longitudinal data collected across national settings yield material
2002; Tulle, 2008). Yet exploring the operation of symbolic proc- tracking life-courses across ever larger numbers of birth cohorts,
esses in practice can also draw attention to the dialogical nature of adding information on the strong input of state decisions to
life-course developments (Mayer, 2004, 2009). Thus attempts can be for the worse, capacities for enacting larger-scale social policies had
made to isolate relatively stable socioeconomic characteristics and also expanded, and could be used with the aim of protecting them.
behaviour, as well as interactions between life-stages; how ‘set’ are It was therefore to be expected that older people should depend
trajectories by early life-circumstances? ‘Can events and conditions more than most other ‘groups’ on welfare states for economic and
in adulthood significantly alter further trajectories and outcomes?’ health security (cf. Komp and van Tilburg, 2010)—but in a range
(Mayer, 2009: 417). Laub and Sampson (2003) contest expectations of ways. Those most likely to be economically vulnerable in older
of childhood determinism, but Mayer (2009: 423) underlines the age include older women, members of minorities, and those with
continuing nature of this debate. Brűckner and Mayer (2005) warn life-long low incomes. When older people become poor they are
too against imputing overall, unidirectional trends to entire socie- likely to remain so in the absence of state intervention (Scharf,
ties: some social patterns may be destandardized, while others are 2009: 45). But contrasting forms of welfare-state provision affect
not. Using half a century’s data for West Germany, they confirm citizens differently (Motel-Klingebiel, 2006). Esping-Anderson
‘de-coupling’ between trajectories of family formation and those of (1999) suggested distinctions between welfare-regime types,
school, training, and work (2005: 48–9). But women’s life-courses according to shares of responsibility allocated to markets, states,
have converged with men’s in terms of education and participation and families. Liberal welfare states such as the US and UK provide
in the labour force, in that respect increasing life-course homogeni- means-tested benefits to a limited extent; social democratic regimes
zation. At the same time, Ferraro et al. (2009) emphasize that social such as Norway and Sweden offer much higher benefits (currently
structures generate inequalities that accumulate not only through under modification); conservative-corporatist regimes such as
individuals’ life-courses but also even between generations. Germany, France and Italy offer benefits on insurance principles;
These analyses of ageing and time are complemented by accounts ‘Mediterranean’ regimes such as Spain are in the process of evolving
derived from the work of Foucault, stressing fluctuating relations away from ‘residualist’ systems whose major reliance for older care
between power and social modes of discourse. These ways of imag- rested on the family. While these structures impact differentially on
ining the world extend deep into the ways individuals internalize social relationships, not least between older individuals and their
power structures, through to their experiences of their own bodies. families, these impacts are not always those anticipated (or feared).
A Foucauldian approach to ‘biopolitics’ explores the efforts of states Motel-Klingebiel et al. (2005), on the basis of a five-country study,
to monitor and control their citizens’ life processes, from birth to rebut suspicions that welfare states ‘crowd out’ family relationships;
death, supported by the power of ‘experts’ in modern societies, and they do not automatically cause families to withdraw from elder
the corresponding framing of older people as ‘dependent’ (Powell, care, abdicating responsibility to the state. As Kűnemund and Rein
2006: 98ff ). Foucault sees elements of social life otherwise supposed (1999) had predicted, high volumes of state help may encourage
exceptional, such as the constant surveillance exercised in prisons, family members to follow suit. Indeed, Daatland (2009) argues that
as core to processes involved in social power and control (Tulle, developed welfare-state provision offers opportunities for newly
2008). Thus Katz’s work on Disciplining Old Age (1996) interprets respectful and independent family relationships.
the rise of gerontology itself at a particular historical juncture in Relations between patterns of ageing and welfare states thus
terms of Foucauldian patterns of control. Sociological approaches depend on complex blends of social attitudes and opportunities.
that explicitly emphasize time, therefore, re-emphasize the ques- They also involve generational compacts: those who are currently
tion to what extent social processes have deterministic impacts on younger pay for older generations, assuming that they too will
the individuals they affect. eventually be supported. But these agreements themselves rely on
implicit or explicit gender contracts (Ginn and Arber, 2000); the
gender distribution of paid versus unpaid labour means that caring
Welfare States and Changing Societies for children and older people tends to fall to the share of women.
The circumstances and experience of ageing, and the ways it is Women’s unpaid work affects welfare systems directly, freeing men
socially conceptualized, represent entwined patterns of advance, to work, and assuming responsibility for those who are young, old,
regression, and unintended consequences. Support from welfare or sick. Ginn and Arber argue that while welfare states’ efforts to
states has saved millions from the rigours of poverty affecting older recompense for caring over the life-course can mitigate its effects
age, but, at the same time, work has become a dominant source on income, this cannot outweigh the results of employment pat-
of status. Preventing them from working thus marginalizes older terns. Especially if they are members of ethnic minorities, women’s
people. The development of European and American welfare states employment records are especially likely to involve part-time work,
illustrates such dilemmas; elsewhere, problems and potential solu- or include breaks for family-related reasons. Thus, reducing public
tions vary, since it cannot be assumed that welfare states of this in favour of private and occupational pensions leads to greater gen-
form are universally a practical option (Harper, 2006). der inequality of incomes in later life.
Bismarck’s efforts in Germany after the 1870s represented an Christensen et al. (2009) highlight the role of social choice in
attempt to curtail ills associated with developing industrialism, shaping pensions and work. They suggest that if in future people
mitigating social problems to which families, neighbours, churches, work to later ages, this could involve more part-time employment;
and private charities were becoming unequal. As Thane emphasizes they envisage reducing working hours throughout populations,
(2005: 9), past societies often contained large proportions of older making ours the century of the ‘redistribution of work’ (2009:
people: ‘At least 10% of the populations of England, France and 1205). This could, they argue, benefit both genders and all genera-
Spain were aged over 60 even in the 18th century.’ But their lives tions, militating against the effects of ‘long hours’ work-cultures on
were often precarious, even miserable, particularly if children did families. However, they do not imply that future worlds of ageing
not survive to adulthood or moved to far-flung locations in search in the West will be sufficiently idyllic not to need social support.
of work. While industrialization initially affected many older people The influence of parental generations on children’s life-chances in
countries like the UK is no longer decreasing, so that individuals’ nursing homes (2005: 531). Older women in Europe and the UK
freedom to move up the social structure during their life-courses are more likely to be poor, owing both to family (child-) care pat-
remains limited (Blanden and Machin, 2007); the health impact terns and the prevalence of part-time work during their working
of ‘socioeconomic position and race/ethnicity’ remains decisive lives (cf. Arber et al., 2007).
(House, 2001: 125). This does not amount to a portrait of social Gender relations are thus entwined with power relations, suf-
circumstances in which individuals no longer need welfare-state fusing cultural attitudes and images, and impact heavily on the
protection. ways in which women and men age (Bernard et al., 2000). Older
women’s experiences derive in part from expectations about their
social presence and roles; discourses about gender are entwined
Generations and Gender with those about age, and the two sets of expectations can reinforce
Welfare-state support for older generations has been hailed as each other (Krekula, 2007; Wilińska, 2010). Much work on women
resulting from public resolve to mitigate the insecurity to which old and ageing centres on the body and how women feel pressured to
age has been subject throughout history (Walker, 1996; Hudson, present themselves: the ways in which social identities are con-
1999). Against this, proponents of curtailing the welfare state argue structed and contested through choices and conventions involving
that ‘generational equity’ requires reducing entitlements for older clothing, for example (Twigg, 2012). Susan Sontag’s observations
people. This position does not appear to attract widespread public are still quoted on the absence of public dignity and status for older
support, with the ‘generational contract’ remaining relatively intact women (Teuscher and Teuscher, 2007):
(Frommert et al., 2009). Moreover, Albertini et al. (2007), among Growing older is mainly an ordeal of the imagination—a moral dis-
others, stress that, taking account of all forms of financial transfer ease, a social pathology—intrinsic to which is the fact that it afflicts
between generations, including family transfers, older generations women much more than men. It is particularly women who experi-
give more to younger ones than the other way round, contributing ence growing older . . . with such distaste and even shame.’
actively to local economies throughout their life-courses. (Sontag, 1972: 30)
Attias-Donfut and Arber (2000) explore the many different uses If this is often the case, it does not preclude complex variations.
of the term ‘generation’ itself, blending different approaches to his- In societies including the UK, older women’s comparatively promi-
torical timing and periodization, and to the genealogical rungs of nent role in supporting kin, though it imposes obligations, offers
family ladders. Bengtson (2001: 8) posits six interacting dimen- compensations (Finch and Mason, 1993; cf. Wenger, 1987), and
sions of intergenerational relations, distinguishing affectual and people of both genders continue to develop and change as they age
associational measures, and consensual, functional, normative, and (Fairhurst, 2003). Health-related and cultural changes also impact
structural forms of solidarity. Contemporary generations also share on age and gender roles. In Africa, traditional expectations of older
more years together than in the past (Hank and Buber, 2009). This people centred on spiritual roles such as ensuring communal har-
was highlighted by the term ‘the Third Age’, denoting the active, mony; many such expectations are currently being shattered, not
constructive later lives that now-healthier older generations might only by the impact of HIV/AIDS on parental generations, but also
anticipate, with new impacts on families and society (Laslett, 1987; by the effects of westernized modernization (Nhongo, 2004). More
cf. James and Wink, 2007). The notion of the ‘Fourth Age’ arose attention too needs to be paid to older men and their experiences
as a counterbalance, indicating an eventual period of infirmity of ageing; this deficit is beginning to be remedied (Davidson et al.,
remaining unavoidable for some. Sociologists such as Haim Hazan 2003), exploring ways in which older men support communal life.
(2009) warn against imposing demands for ‘successful’ ageing on In Ireland, Timonen et al. (2011: 58) report that men older than
this period of life especially: ‘positive’ forms of conformity should 75 are more socially integrated than women of a similar age. This
not be enforced on older individuals. Cavalli et al. (2007) confirm may attest to the old custom of ‘visiting’, ‘rambling’, or ‘bringing the
the importance of health in transitions between these life ‘stages’: it chat’ among men, when Ireland was still a semitraditional society
is in response to declining health that the octogenarians they study (Edmondson, 2011). It underlines the fact that men’s contributions
disengage from social interaction with family and friends, rather to community and care should not be ignored.
than life-events such as the loss of relatives or entering nursing
homes. Social choices supporting the health of older generations
thus remain key to their future capacities for well-being, but much Environments, Families, and Mutual Care
work remains to be done to understand and support the ‘Fourth Diversity in life-courses reflects variations in environment, but in
Age’ and processes surrounding death. ways whose effects cannot be anticipated en bloc. Rural settings in
Arber and Ginn (2005: 527–8) point out that the numerical particular are often pictured as offering excellent surroundings for
domination by women over older men is now diminishing as sur- later life, but this overlooks a variety of possible exigencies, includ-
vival rates begin to converge; but the ‘feminization of ageing’ is ing poverty, isolation, and local social decline (Keating, 2008).
still marked by women’s systematic disadvantages under current Rozanova et al. (2008: 75) acknowledge that ‘social participation
social conditions, for example with regard to pensions, health, has positive associations with better health and well-being in later
and access to care (Attias-Donfut and Arber, 2000). Even mar- life’ and that older people often invest considerable participation
riage affects the genders differently. If older men are widowed, in rural communities; but they underline the need to explore what
they tend to remarry; thus most older men have partners for type of participation is involved and how freely it is given. Rural
‘companionship, domestic service support and care should they areas may support specific forms of ‘social capital’, but these, how-
become physically disabled’ (Arber and Ginn, 2005: 529). But ever significant in their own terms, may not necessarily satisfy
nearly half of women over 65 are widowed, and over four-fifths at liberal expectations (Edmondson, 2001). Older people in the coun-
85 and over; in consequence, twice as many women as men live in tryside may be respected for their knowledge and skills, an effect
that may be enhanced in subcultures among groups who know the favour social care, and the south, seen as more familistic, relying
older individuals in question, and to whose lives their achievements heavily on family care owing to a dearth of effective social serv-
are relevant, as may be the case for music or sport. To what extent ices. This was interpreted as indicating weaker family ties in the
does this depend on rurality as such? Gallagher (2008) found that north (Reher, 1998). Dykstra and Fokkema (2011) contest this,
in Ireland—where, admittedly, the distinction between rural and arguing that each of four family patterns is fairly common through-
urban areas remains relatively fluid—patterns of mutual social sup- out European countries. They postulate two ‘familistic’ forms, both
port among older people in Dublin and those in the country were characterized by members living close to each other, with fre-
surprisingly similar. In both cases, neighbourly relations and local quent contact, and acknowledging norms of family obligation. In
networks, and the comfort of sharing the social landscape with ‘descending’ forms, help in kind is given more by parents to chil-
familiar others, contributed markedly to older people’s lives. dren, whereas in ‘ascending’ forms, more help in kind is given by
Such phenomena help explain the common desire to ‘age in children to parents. Then there are ‘supportive-at-a-distance’ fami-
place’, now recognized in much social policy for older people. But lies in which members neither live nearby nor support obligation
Sixsmith and Sixsmith (2008) point to possible complications: the norms, but do have frequent contact, with financial support given
homes in which individuals have grown old may no longer be easy mainly by parents to adult children; and ‘autonomous’ families
to inhabit, for reasons ranging from their physical structures to where members neither live in proximity, nor acknowledge obliga-
their locations (cf. Means, 2007). The expanding field of environ- tion norms, nor engage in frequent contact or assistance. Dykstra
mental gerontology explores complex interactions between social and Fokkema argue that ‘autonomous’ families are not most con-
and physical elements of place in relation to older people (Andrews centrated in the northern nations, which instead show high lev-
and Phillips, 2005; Peace et al., 2005), exploring how settings come els of ‘support at a distance’. They found high levels of autonomy
to be ‘enabling’ or ‘disabling’ to those who live in them (Smith, in France and Switzerland; these were also likelier among families
2009). characterized by divorce and high socioeconomic status. A single
Environments are in part experienced through the medium of family might move through different patterns at various stages in
family structures and expectations, which themselves are changing the family life-cycle—perhaps culminating in ‘ascending’ patterns
in contemporary societies, though Bengtson (2001) insists that this when parents were most vulnerable. Families are thus developing
does not make them less important. Rather, multigenerationality and responsive processes rather than fixed structures.
the role of grandparents in caring for grandchildren may be grow- While family ties are crucial in ageing, particularly in terms of
ing more significant, as grandparents survive longer. Families are care and life satisfaction, more research is needed on other types
often ‘long and thin’, ‘beanpoles’ (Bengtson, et al., 1990), rather than of relationship too. Bengtson (2001) argues that friendships may
networks with large numbers of siblings and cousins at each level. support or even replace family ties (cf. Allan, 2010). Phillipson
Bengtson himself points out that there is great diversity across fam- (2003b), though, underscores Sennett’s (1998) argument that
ily practices and no entirely dominant type (cf. Matthews and Sun, many contemporary pressures militate against maintaining
2006). Amid fears for the overall ‘decline of the family’, Stacey (1996) friendships. These include work- and life-course-related insecu-
argues that diverse, ‘postmodern’ family forms have become normal. rities, or professional pressures to move location often, encour-
But normality does not entail freedom from stress; Harper (2006) aging short-term rather than long-term relationships. In western
emphasizes that forming new sets of relationships after a divorce may societies, more people are living alone in later life, as a result not
be even more complex and challenging than after a death. only of widowhood but also of attitudes and choice: reluctance to
Fluid family forms do not automatically increase social and enter nursing homes or a preference for ‘intimacy at a distance’
familial isolation; comparatively large numbers of close relatives with family. This trend does not apply only to older individu-
may in principle be available to individuals in modern societies. als; between 26% of households in the US and 40% in northern
Uhlenberg (1996) calculates that for children born in 1900, the Europe contain only one person, with percentages expected to
chances of being orphaned before the age of 18 were 18%, and by grow. What this means in practice varies, from independence to
the age of 30 only 21% had any grandparents still living. But for pursue one’s own interests to feelings of helplessness, withdrawal,
children born in 2000, 68% will have four grandparents still alive and isolation (Portacolone, 2011: 805). Ziegler and Schwanen
by the time they are 18. Today’s 20-year-olds are more likely to (2011) emphasize that mobility through physical space—as in
have a grandmother living than 20-year-olds in 1900 were to have taking trips, both offering new experiences and heightening a
a mother. The social implications of such changes are diverse. As sense of agency—enhances wellbeing in later life. Key, they argue,
Hareven (1984) points out, transitions in individuals’ life-stages at is ‘mobility of the self ’, willingness ‘to connect with the world and
the start of the twentieth century depended not only on their own otherness, with people and places beyond oneself, one’s house-
circumstances, as when people would wait to find work before they hold and one’s residence’ (2011: 763).
married, but also on the circumstances of their families: marriage Yet it is part of the stereotype of older people to picture them
might be delayed by the obligation to support dependent kin. There as lonely; even in contemporary Sweden, nearly 90% of respond-
followed a period in the mid-twentieth century when transitions ents expected retired people to be lonely (Tornstam, 2007). In fact,
were expected to conform to more external norms; for example, highest rates of reporting loneliness ‘some of the time’ are found
only after marriage would people aim to have children. Such pat- among young people, and only then people older than 75; however,
terns are now reforming. Where in the nineteenth century families those who say they feel lonely ‘all or most of the time’ tend to be
were disrupted and reconstituted as a result of deaths, they are now older than 70 (Victor et al., 2009). The contributions of social cir-
reconstituted by factors such as divorce and remarriage. cumstances to loneliness are underlined by the finding that it var-
It has often been assumed that family forms contrast between the ies more by country than by age (Yang and Victor, 2011); forms
north of Europe, seen as more individualistic but more inclined to taken by loneliness and loss vary too according to social pathways
connected with stigmatization, health, and social estrangement own lives; exclusion from basic services, affecting the ability to
(Graham and Stephenson, 2010). manage everyday life; and neighbourhood exclusion, since imme-
Caring for both the self and other people, by contrast, happens diate residential settings affect people’s sense of self and quality of
predominantly at home: most elder care happens in the family, life. Dimensions of exclusion can cluster in particular places, with
regardless of country. For men and women who live alone, whether impacts on residents’ self-definitions as well as the ways they are
or not by choice, social and care options need different approaches regarded by other people. In some urban areas of England, over a
(de Jong Gierveld, 2003). Older people themselves not only receive third of residents are socially excluded according to two or more
but give care, both outside the family and within it. In the contem- of these measures; health and ethnic origin play parts here, as does
porary world, about three-quarters will become grandparents at educational level (Scharf et al., 2005: 83). Yet increasing numbers of
some time in their lives, and will enjoy this status for longer than the world’s population, including its older population, are living in
formerly. Often, maternal grandmothers have the closest relation- cities, not necessarily in environments suitable for ageing.
ships with grandchildren, though this depends on relative ages, While cities can be disabling and threatening environments at any
health, and geographical closeness (Harper, 2005), and grandmoth- age, . . . the associated risks increase with age. The key point is that
ers’ own employment obligations (Johnson and Lo Sasso, 2004). at 75 or 85, people may feel an even greater sense of being trapped or
Hagestad (2006) explores grandparents’ functions as ‘supporting’, disadvantaged by urban decay, and that this may limit their ability to
‘saving’, or ‘rescuing’ either grandchildren or their parents (often by maintain a sense of self-identity.
enabling mothers to go out to work). As Timonen and Arber (2012) (Scharf et al., 2005: 85)
point out, this develops Townsend’s (1955) East London research Public settings can enhance older people’s senses of self and
into grandmothers’ heavy involvement in rearing grandchildren mastery of their environments (for example, when it is taken for
and their close emotional relations with their daughters. granted that personal encounters can take place safely in familiar
In contrast, Lüscher and Pillemer (1998) indicate the possi- public spaces), or diminish it (as dependence on car transport may
bility of ‘ambivalence’ to grandparenthood, while Cherlin and do). The World Health Organization (2011) has urged ‘age-friendly’
Furstenberg (1986) describe detached, passive, supportive, authori- cities to develop social and material structures to heighten fit with
tative, or influential styles of grandparenting. Yet Timonen and older adults’ needs and preferences.
Arber (2012: 7) remark that such categories are ‘not well suited to In addition, a further key issue should be highlighted: the exclu-
examining possible differences in how a grandparent relates to dif- sion of (older) people from socially meaningful life-activities. This
ferent grandchildren, and also overlook the possibility that grand- effect is particularly pronounced in societies where social status is
children may exert an influence on the nature of the relationship’. intimately connected with work (notwithstanding arguments by
They underline grandparents’ roles in ‘transmitting knowledge and Gilleard and Higgs (2000, 2005) that leisure activities offer some
values to younger generations’, ‘providing a sense of family heritage counterbalance in terms of older people’s social participation).
and stability’, as well as influencing grandchildren’s ‘core moral val- Research is needed to explore types and circumstances of work
ues’ and religious orientations (2012: 5). To the extent that women from which people wish or do not wish to retire, and the access
are expected to play more active roles than men in sustaining family to social fulfilment it provides or fails to provide. High levels of
relations, they may be ‘more likely to become key figures in young participation in voluntary work indicate that older people wish to
grandchildren’s lives’ (2012: 8). Thus the impact of gender on family remain active, but demand some degree of control over what they
experiences of ageing remains salient. It is heightened in crisis situ- do. In societies where agriculture remains significant, individuals
ations such as those involving HIV/AIDS, when older women may in the lowest income quintiles, many of them farmers, as well as
be compelled to rise to heights of resilience and activity in defence in the highest, such as doctors, may continue to work until late in
of their grandchildren (Casale, 2011). This does not make intergen- life (Mosca and Barrett, 2011). Such whole-life activities may offer
erational understanding and solidarity automatic, particularly in access to meaningful activity and respected social participation; but
relation to the extremely old and infirm (Hazan, 2011). Lowenstein they underline rather than solve the dual problem of concentrating
et al. (2011) argue, indeed, that, under evolving contemporary cir- social meaning around work, and limiting access to it.
cumstances, entirely revised intercultural processes are needed to ‘Social exclusion’ refers to barriers from activities regarded as
support ‘sustainable generational relations’. normal in a given society; ‘ageism’, a term introduced by Robert
Butler (1975: 22), refers specifically to ‘systematic stereotyping and
discrimination against people because they are old’ (cf. Nelson,
Social Exclusion and Ageism 2004). Stereotyping may be so ingrained that it is hard to avoid.
Bowling (2005) considers a variety of factors enhancing the possibil- Faulkner (2001) shows that individuals exposed to care which is
ity of ‘ageing well’, including good health, adequate material stand- unintentionally disempowering, depriving them of control, develop
ards, good social relations, and an agreeable neighbourhood—not a ‘learned helplessness’: such risks to wellbeing can emanate from
omitting less tangible features such as ‘independence, control, and healthcare settings themselves. Perpetuating stereotypes can itself
autonomy’. How easy is it to achieve ageing like this? Poverty and ‘increase psychosocial risk factors for ill health’, including a cascade
class in particular construct perennial barriers; it is possible to dis- of confirmation bias, inattention to discordant data, self-fulfilling
aggregate the ways such influences exclude individuals from partic- prophecies among professionals, and ‘suboptimal’ environments
ipating in activities that are otherwise accepted as normal (Abrams for older adults—including the stress induced by membership of
et al., 2007). Scharf et al. (2005: 78) highlight exclusion from mate- a group that is discriminated against (Golub and Langer, 2007:
rial resources, income, and security; exclusion from social relations 9–10). Ageist stereotypes can impact strongly where older people
and meaningful interaction with others; exclusion from civic activi- are vulnerable, as in care homes. Knight et al. (2010) highlight the
ties, especially decision-making processes influencing individuals’ positive effects that residents’ participation in care homes’ decisions
can have; but Cooney and Murphy (2009) are not alone in find- Coleman (2009) contends that they face contradictory problems in
ing that residents’ capacities to exercise a modicum of control this regard. On the one hand, they are expected to transmit tradi-
over their lives range from broad to completely minimal. The UK tional views down the generations; on the other, they must adapt to
Equality and Human Rights Commission (2011) found that while change. Older people themselves may face religious problems and
many older people were satisfied with the quality of home care pro- doubts—in which, Coleman reports, churches themselves may take
vided to them, identifiably abusive forms of interaction were also surprisingly little interest. Dillon (2009), analysing life-course tra-
widespread. jectories of older people born in California in 1929, tracks their per-
While ageism is experienced as a problem in the population at sonal responses to religious issues. She distinguishes between those
large (Stuckelberger et al., 2012), the study of ageing itself is not who are ‘religious’ in the sense of belonging to religious communi-
immune to it. Though students of ageing purportedly reject ‘defi- ties or networks, attending church, and participating in connected
cit’ approaches to the field, questionnaires often assume that older social activities, and ‘spiritual seekers’, occupied with trying to work
people’s activities focus on personal relationships and immediate out spiritual meanings in their personal lives. Superficially, more
settings, ignoring activities of broader social significance. This is prescriptive settings can also offer constructive social resources:
despite the fact that ‘three fifths of men and women over 60 in Mehta (2009) examines approaches to ageing among Chinese,
Britain are members of one or more social, leisure, community, or Malay, and Indian ethnic groups in Singapore with allegiances to
religious group’ (Arber et al., 2002: 91). On the basis of 117 articles Confucianism, Islam, or Hinduism. Appropriate modes of life pre-
published in Ageing and Society between 1997 and 2000, Bytheway scribed for individuals at particular life-stages can enable rich and
(2002: 73) argues that ‘much of our research is embedded in ageist satisfying responses. Religion thus plays a variety of complex roles
assumptions about the societal implications of an aging popula- in later life, though finding appropriate languages for expressing
tion’, making it ‘difficult to dispute Townsend’s (1981) accusation and recording the trajectories involved poses a challenge for evi-
of acquiescent functionalism’. Social effects of gerontological theo- dence collection (Coleman, 2011).
ries themselves may have ageist effects. Jolanki (2009) reports that Lars Tornstam’s work on ‘gerotranscendence’ explores the
respondents in Finland felt compelled to demonstrate how ‘active’ increasing sympathy felt by some older people with ‘transcendental
they were, fearing that otherwise they might not count as compe- sources of happiness’ (2005: 59), also often difficult to express in
tent moral agents. conventional language. Frequently,
One aspect of researchers’ efforts to avoid ageism centres on treat- The experience of nature evokes the feeling of being at one with the
ing older people as active, specifically involving them in research universe, which is called at-one-ment in the Eastern tradition. The
processes (Barnes and Warren, 1999). There is debate about how increasing significance of these small everyday experiences of nature
to accomplish this; older volunteers helped Johnson et al. (2010) to could therefore be interpreted as a way in which the barrier between
collect local information on care homes, for example, but recruiting the self and the universe is transcended.
older researchers has considerable training, time, and cost impli- (Tornstam, 2005: 59–60)
cations (Peace and Hughes, 2010). Other approaches situate older —as does, for many, the experience of having grandchildren.
people as arbiters of research: Ranzijn’s (2002) ‘environmental fit’ Tornstam’s respondents show identity changes in later life, often
approach interrogates whether settings provide what older peo- achieving a less self-centred ‘modern asceticism’ and ‘broadmind-
ple actively need and prefer. Tanner (2010) commends ‘learning edness, tolerance and humility’ (Tornstam, 2005: 68–89). Tornstam
from older people’ how to support their strategies for negotiating contrasts this with wisdom paradigms remaining within ‘our ‘nor-
bodily and external changes while attempting to ‘stay themselves’. mal’, ‘paradigmatic world’ (2005: 77). Achenbaum and Orwoll (1991)
Enhancing older people’s social participation, for instance in also focus on increasingly transcendent attitudes in comparison
life-long learning institutions, is key to supporting this in powerful with more worldly aims in earlier life-stages; H.R. Moody (2002)
ways (Manheimer, 2005), intended to illuminate and explore the explores the holistic, transformational notion of ‘conscious aging’,
benefits to society at large of older people’s membership. and Schachter-Shalomi (1995: 5) stresses its ‘service to the commu-
nity’. Rothermund and Brandstädter (2003) support a ‘mindfulness’
Life-Course Development and Social approach to ageing, focusing on continual growth and change, with
evolving criteria for what individuals think of themselves as obliged
Meaning to do. Thus McKee and Barber (1999) argue that wisdom may be
It is a perennial complaint that as ageing has evolved in western developed in response to loss, potentially decreasing egocentricity
societies, social expectations rarely emphasize responsibilities for and heightening empathy.
older people or make good use of their skills and capacities (Riley Caspari and Lee (2006) portray older individuals as both offer-
et al., 1994). Yet Bond et al. (1993) suggest that people may grow ing links to personal and social networks and providing wisdom,
more individual as they age, as a result of long personal life-courses as well as contributing to the species’ survival by their work and
in specific environments. Rather than concentrating on loss in rela- in other ways. Wisdom is currently studied under the aegis of
tion to ageing, therefore, Levenson et al. (2001) argue for a ‘libera- positive psychology by authors including Sternberg (1998), Baltes
tive’ approach to it. There is evidence that people attach importance and Staudinger (2000), or with a social-psychological empha-
to learning through and from their life-courses; McGee et al. (2011) sis, notably by Ardelt (2004) (see Ferrari and Westlake, 2012).
find that older people in Ireland not uncommonly feel they have Understanding wisdom also demands reconstructing the types
increased in wisdom and maturity. The question is how contempo- of reasoning it involves (Edmondson et al., 2009), as well as its
rary societies respond to this. variety of social forms (Edmondson, 2005, 2009, 2012): here it is
Older people’s roles in relation to life-course meaning have tra- possible to use ethnographic methods to track contrasting forms
ditionally been conceptualized in connection with religion, but of wisdom-in-use, including significant but less transcendent
versions practised in ordinary social settings. These everyday ver- Atchley, R. (1989a). A continuity theory of normal aging. The Gerontologist,
sions of wisdom include the way in which some types of social 29(2), 183–90.
interaction heighten the insights of individuals who on their own Atchley, R. (1989b). Continuity and adaptation in old age. Johns Hopkins
might behave less wisely. Exploring older people’s roles in such University Press, Baltimore.
Attias-Donfut, C. and Arber, S. (2000). Equity and solidarity across the
processes has potential for impacting on ‘deficit’ models of what
generations. In: The myth of generational conflict. The family and state in
the life-course involves, both for individuals and for the ‘socio- ageing societies (eds S. Arber and C. Attias-Donfut), pp. 1–21. Routledge,
scapes’ to which they belong. London.
Older people’s interactions with their immediate environments Attias-Donfut, C. and Wolff, C. (2000). Complementarity between private
and their neighbourhoods, communities, and societies help them and public transfers. In: The myth of generational conflict. The family and
form, maintain, and change their roles and identities. These envi- state in ageing societies (eds S. Arber and C. Attias-Donfut), pp. 47–68.
ronments can be crucial sources of support in times of loss or Routledge, London.
transition; in turn, older people contribute decisively to them, Baars, J. (2009). Problematic foundations: theorizing time, age and ageing.
in ways that demand further emphasis and exploration. Ageing In: Theories of ageing (eds V. Bengston et al.), pp. 87–100. Springer,
New York.
therefore needs to be understood in terms of interacting practices
Baltes, P. and Staudinger, U. (2000). Wisdom: a metaheuristic (pragmatic) to
forming ‘social ecologies’, in which mutual relationships change orchestrate mind and virtue towards excellence. American Psychologist,
over time within layers of social and physical space and power. 55, 122–36.
As people grow older, their relations to where and how they live Barnes, M. and Warren, L. (1999). Paths to empowerment. Policy, Bristol.
may change, but becoming physically more fragile need not nec- Bengtson, V. (2001). Beyond the nuclear family: the increasing importance
essarily undermine these relations (Peace et al., 2005). Eales et of multigenerational bonds. Journal of Marriage and Family , 63 ,
al. (2008: 111) distinguish between the natural, built, and social 1–16.
resources that communities can offer; they include opportunities Bengtson, V., Elder, G., and Putney, N. (2005). The lifecourse perspective on
for maintaining relationships with family and friends, for ‘civic, ageing: linked lives, timing, and history. In: The Cambridge handbook
cultural, educational, and voluntary engagement’, and for activi- of age and ageing (eds M. Johnson et al.), pp. 493–501. Cambridge
University Press, Cambridge.
ties promoting health as well as ‘social and spiritual wellbeing’.
Bengtson, V., Rosenthal, C., and Burton, L. (1990). Families and ageing:
The fact that social structures, interpretations, and decisions so diversity and heterogeneity. In: Handbook of aging and the social sciences
strongly affect these opportunities makes the sociology of ageing (eds R. Binstock and L. George), 3rd edition, pp. 263–87. Academic
a reflection by society on its own processes. Since these processes Press, New York.
will extend into the future (Harper, 2006; Vincent et al., 2006), Bengtson, V., et al. (eds) (2009). Handbook of theories of aging, 2nd edition.
for many it also imposes the obligation to develop approaches to Springer, New York.
enhancing them. Berger, P. and Luckmann, T. (1966). The social construction of reality. Anchor
Books, Garden City, New York.
References Bernard, M., et al. (eds) (2000). Women ageing: changing identities, challenging
Abrams, D., Christian, J.N., and Gordon, D. (eds) (2007). Multidisciplinary myths. Routledge, London.
handbook of social exclusion research. Wiley-Blackwell, Oxford. Bertaux, D. and Kohli, M. (1984). The life-story approach: a continental view.
Achenbaum, A. (1995). Crossing frontiers: gerontology emerges as a science. Annual Review of Sociology, 10, 215–37.
Cambridge University Press, Cambridge. Biggs, S. (2004). Age, gender, narratives and masquerades. Journal of Aging
Achenbaum, A. and Orwoll, L. (1991). Becoming wise: a psycho-gerontological Studies, 18(1), 45–58.
study of the Book of Job. International Journal of Aging and Human Biggs, S., Hendricks, J., and Lowenstein, A. (eds) (2003). The need for theory:
Development, 32, 21–39. critical approaches to social gerontology. Baywood, Amityville.
Albertini, M., Kohli, M., and Voegel, C. (2007). Intergenerational transfers Blaikie, A. (2002). The secret world of sub-cultural ageing: what unites and
of time and money in European families: common patterns—different what divides? In: Cultural gerontology (ed. L. Andersson), pp. 95–110.
regimes? Journal of European Social Policy, 17(4), 319–34. Auburn House, Westport.
Allan, G. (2010). Friendship and ageing. In The Sage handbook of social Blanden, J. and Machin, S. (2007). Recent changes in intergenerational mobility
gerontology (eds D. Dannefer and C. Phillipson), pp. 239–47. Sage, in Britain. <http://thest57.demonweb.co.uk/reports/mainreport.pdf>
London. (accessed 20.06.2011).
Andersson, L. (ed.) (2002). Cultural gerontology. Auburn House, Westport. Bond, J., Briggs, R., and Coleman, P. (1993). The study of ageing. In: Ageing
Andrews, G. and Phillips, D. (2005). Ageing and place: perspectives, policy and in society: an introduction to social gerontology (eds J. Bond et al.),
practice. Routledge, London. pp. 19–52. Sage, London.
Arber, S. and Ginn, J. (2005). Gender dimensions of the age shift. In: Bowling, A. (2005). Ageing well: quality of life in older age. Open University
Cambridge handbook of age and ageing (eds M. Johnson et al.), pp. 527–37. Press, Maidenhead.
Cambridge University Press, Cambridge. Brűckner, H. and Mayer, K.U. (2005). Destandardization of the life course:
Arber, S., Perren, K., and Davidson, K. (2002). Involvement in social what it might mean? And if it means anything, whether it actually took
organisations in later life: variations by gender and class. In: Cultural place?. In: The structure of the life course: standardized? individualized?
gerontology (ed. L. Andersson), pp. 77–93. Auburn House, Westport. differentiated? (ed. R. Macmillan), pp. 27–53. Elsevier, Amsterdam.
Arber, S., Andersson, L., and Hoff, A. (2007). Gender, ageing and power: Butler, R. (1975). Why survive? Being old in America. Harper and Row,
changing dynamics across western societies. Current Sociology, 55(2), San Francisco.
147–312, Monograph 1. Butler, R. and Gleason, H. (1985). Productive ageing: enhancing vitality in
Ardelt, M. (2004). Wisdom as expert knowledge system: a critical review later life. Springer, New York.
of a contemporary operationalization of an ancient concept. Human Bytheway, B. (2002). Positioning gerontology in an ageist world. In: Cultural
Development, 47, 257–85. gerontology (ed. L. Andersson), pp. 59–76. Auburn House, Westport.
Arensberg, C. and Kimball, S. (1940/2001). Family and community in Ireland. Bytheway, B. (2011). Unmasking age: the significance of age for social research.
CLASP Press, Ennis. Policy, Bristol.
Casale, M. (2011). ‘I am living a peaceful life with my grandchildren. Nothing Edmondson, R. (2005). Wisdom in later life: ethnographic approaches.
else.’ Stories of adversity and ‘resilience’ of older women caring for Ageing and Society, 25, 339–56.
grandchildren in the context of HIV/AIDS and other stressors. Ageing Edmondson, R. (2009). Wisdom: a humanist approach to valuing older
and Society, 31(8), 1265–88. people. In: Valuing older people: a humanist approach to ageing (eds
Caspari, R. and Lee, S.-H. (2006). Is human longevity a consequence of R. Edmondson and H.J. von Kondratowitz), pp. 201–16. Policy, Bristol.
cultural change or modern human biology? American Journal of Physical Edmondson, R. (2011). Ageing in Ireland. In: Global aging (eds S. Chen and
Anthropology, 129, 512–17. J. Powell). Nova Science, New York.
Cavalli, S., Bickel, J.-F., and Lalive d’Epinay, C.J. (2007). Exclusion in very old Edmondson, R. (2012). A social interpretation of personal wisdom.
age: the impact of three critical life events. International Journal of Ageing In: Personal wisdom (eds M. Ferrari and N. Weststrate). Springer,
and Later Life, 2(1), 9–31. New York.
Cherlin, A. and Furstenberg, F. (1986). The new American grandparent: a Edmondson, R. and von Kondratowitz, H.-J. (eds) (2009). Valuing older
place in the family, a life apart. Basic Books, New York. people: a humanistic approach to ageing. Policy, Bristol.
Christensen, K., et al. (2009). Ageing populations: the challenges ahead. Edmondson, R., Pearce, J., and Woerner, M.H. (2009). When wisdom is called
Lancet, 374, 1196–208. for in clinical reasoning. Theoretical Medicine and Bioethics, 30, 231–47.
Cole, T. (1992). The journey of life. A cultural history of aging in America. Elder, G. (1974). Children of the Great Depression. University of Chicago
Cambridge University Press, Cambridge. Press, Chicago.
Cole, T., Van Tassel, D., and Kastenbaum, R. (eds) (1992). Handbook of the Elder, G. (1998). The life course as developmental theory. Child Development,
humanities and aging. Springer, New York. 69(1), 1–12.
Coleman, P. (2009). Religious belonging and spiritual questioning: a Western Equality and Human Rights Commission (2011). Close to home: an
European perspective. In: Valuing older people: a humanist approach enquiry into older people and human rights in home care. <http://www.
to ageing (eds R. Edmondson and H.J. von Kondratowitz), pp. 23–36. equalityhumanrights.com/uploaded_files/homecareFI/home_care_
Policy, Bristol. report.pdf> (accessed 06.12.2011).
Coleman, P. (2011). Belief and ageing: spiritual pathways in later life. Policy, Esping-Anderson, G. (1999). Social foundations of post-industrial economies.
Bristol. Oxford University Press, Oxford.
Cooney, A. and Murphy, K. (2009). Ethos of care and environment in Estes, C. (1991). The new political economy of aging: introduction and
long-stay care settings: impacts on residents’ lives. In: Valuing older critique. In: Critical perspectives on aging: the political and moral economy
people: a humanist approach to ageing (eds R. Edmondson and H.J. von of growing old (eds M. Minckler and C. Estes), pp. 19–36. Baywood,
Kondratowitz), pp. 161–76. Policy, Bristol. Amityville.
Corner, L. and Bond, J. (2006). The impact of the label of mild cognitive Estes, C. and Binney, E. (1989). The biomedicalization of aging: dangers and
impairment on the individual’s sense of self. Philosophy, Psychiatry, and dilemmas. The Gerontologist, 29(5), 587–96.
Psychology, 13(1), 3–12. Estes, C. and Phillipson, C. (2002). The globalization of capital, the welfare
Cowgill, D. (1974). Aging and modernization: a revision of the theory. state, and old age policy. International Journal of Health Services, 32(2),
In: Late life (ed. J. Gubruim ), pp. 124–46 . Charles C. Thomas, 279–97.
Springfield. Estes, C., Biggs, S., and Phillipson, C. (2003). Social theory, social policy and
Cumming, E. and Henry, W. (1961). Growing old: the process of disengagement. ageing: a critical introduction. Open University Press, Maidenhead.
Basic Books, New York. Evans, R.G., et al. (2001). APOCALYPSE NO: population aging and
Daatland, S.-O. (2009). How to balance generations: solidarity dilemmas in the future of health care systems. Canadian Journal of Ageing, 20,
a European perspective. In: Valuing older people: a humanist approach 160–91.
to ageing (eds R. Edmondson and H.J. von Kondratowitz), pp. 123–38. Fahey, T., et al. (2007). A social portrait of older people in Ireland. ESRI, Dublin.
Policy, Bristol. <http://www.socialinclusion.ie/publications/documents/Older_lowres2.
Dannefer, D., et al. (2005). On the shoulders of a giant: the legacy of Matilda pdf> (accessed 12.03.2009).
White Riley for gerontology. Journal of Gerontology (B), 60(6), 296–304. Fairhurst, E. (2003). New identities in ageing: perspectives on age, gender
Davidson, K., Daly, T., and Arber, S. (2003). Older men, social integration and life after work. In: Gender and aging: changing roles and relationships
and organisational activities. Social Policy and Society, 2(2), 81–9. (eds S. Arber et al.), pp. 31–46. Open University Press, Maidenhead.
de Jong Gierveld, J. (2003). Social networks and social well-being of older Faulkner, M. (2001). The onset and alleviation of learned helplessness in
men and women living alone. In: Gender and ageing: changing roles and older hospitalized people. Ageing and Mental Health, 5, 379–86.
relationships (eds S. Arber et al.), pp. 95–110. Open University Press, Featherstone, M. and Hepworth, M. (1995). Images of positive ageing: a
Maidenhead. case study of Retirement Choice magazine. In: Images of ageing: cultural
Dillon, M. (2009). Integrating the sacred in creative ageing. In: Valuing older representations of later life (eds M. Featherstone and M. Wernick),
people: a humanist approach to ageing (eds R. Edmondson and H.J. von pp. 29–47. Routledge, London.
Kondratowitz), pp. 51–72. Policy, Bristol. Ferrari, M. and Westlake, N. (eds) (2012). Personal wisdom. Springer,
Donkin, A., Goldblatt, P. and Lynch, K. (2002). Inequalities in life expectancy New York.
by social class 1972–1999. Health Statistics Quarterly, 15, 5–15. Ferraro, K., Pylypiv Shippee, T., and Schafer, M. (2009). Cumulative inequality
Dowd, J.J. (1980). Exchange rates and old people. Journal of Gerontology, theory for research on aging and the life-course. In: Theories of ageing
35(4), 596–602. (eds V. Bengtson et al.), pp. 413–34. Springer, New York.
Dykstra, P. and Fokkema, T. (2011). Relationships between parents and their Finch, J. and Mason, J. (1993). Negotiating family responsibilities. Routledge,
adult children: a West European typology of late-life families. Ageing and London.
Society, 31(4), 545–69. Fiori, K.L., Considine, N.S., and Magai, C. (2008). Ethnic differences in
Eales, J., Keefe, J. and Keating, N. (2008). Age-friendly rural communities. patterns of social exchange among adults: the role of resource context.
In: Rural ageing: a good place to grow old (ed. N. Keating), pp. 109–120. Ageing and Society, 28, 495–524.
Policy, Bristol. Folbre, N. (2002). Accounting for care in the United States. In: Care work: the
Edmondson, R. (2001). Tracing civic culture ethnographically and what it quest for security (ed. M. Daly), pp. 175–91. International Labour Office,
tells us about social capital: communities in the west of Ireland. In: The Geneva.
politics of everyday life (eds E. Uslaner and P. Dekker), pp. 59–72. Foucault, M. (1978) (trans. Hurley R). The history of sexuality. Allen Lane,
Routledge, London. London.
Fox, N. (2005). Cultures of ageing in Thailand and Australia (what can an Hochschild, A. (2000). Global care chains and emotional surplus value. In:
ageing body do?) Sociology, 39, 481–98. On the edge: living with global capitalism (eds W. Hutton and A. Giddens),
Frommert, D., et al. (2009). Pension systems and the challenge of population pp. 130–46. Cape, London.
ageing: what does the public think?. In: Valuing older people: a humanist House, J. (2001). Understanding social factors and inequalities in health: 20th
approach to ageing (eds R. Edmondson and H.J. von Kondratowitz), pp. century progress and 21st century prospects. Journal of Health and Social
139–60. Policy, Bristol. Behavior, 43, 125–42.
Gallagher, C. (2008). The community life of older people in Ireland. Peter Lang, Hudson, R. (1999). Conflict in today’s aging politics: new population
Oxford. encounters old ideology. Social Service Review, 73, 358–79.
Gaonkar, D.P. (2002). Toward new imaginaries: an introduction. Public Institut fűr Demoskopie Allensbach (2003). Einflußfaktoren auf die
Culture, 14(1), 1–19. Geburtenrate (IfD Umfrage 5177). <http://www.ifd-allensbach.de/main.
Gardner, K. (2002). Age, narrative and migration: the life course and life php?selection=73&rubrik=0> (accessed 12.10.2011).
histories of Bengali elders in London. Berg, Oxford and New York. James, J. and Wink, P. (2007). The crown of life: dynamics of the early
Gilleard, C. and Higgs, P. (2000). Cultures of ageing: self, citizen and the body. post-retirement period. Springer, New York.
Prentice-Hall, London. James, W., Witte, J., and Galbraith, M. (2006). Havighurst’s social roles
Gilleard, C. and Higgs, P. (2005). Contexts of ageing: class, cohort and revisited. Journal of Adult Development, 13(1), 52–60.
community. Polity, Cambridge. Jamieson, A. and Victor, C. (2002). Researching ageing and later
Ginn, J. (2001). Privatising pensions: new options and new risks for women. life: the practice of social gerontology . Open University Press,
In: Researching social life (ed. N. Gilbert), pp. 287–301. Sage, London. Maidenhead.
Ginn, J. and Arber, S. (2000). Gender, the generational contract and pension Johnson, J., Rolph, S., and Smith, R. (2010). Residential care transformed:
privatisation. In: The myth of generational conflict: the family and state revisiting ‘the last refuge’. Palgrave Macmillan, London.
in ageing societies (eds C. Attias-Donfut and S. Arber), pp. 133–53. Johnson, R. and Lo Sasso, A. (2004). Family support of the elderly and female
Routledge, London. labor supply: tradeoffs among caregiving, financial transfers and work. In:
Golub, S. and Langer, E. (2007). Challenging assumptions about adult Families in ageing societies: a multidisciplinary approach (ed. S. Harper),
development: implications for the health of older adults. In: Handbook pp. 114–42. Oxford University Press, Oxford.
of health psychology and aging (eds C. Aldwin et al.), pp. 9–29. Guildford Jolanki, O. (2009). Talk about old age, health and morality. In: Valuing older
Press, New York. people: a humanist approach to ageing (eds R. Edmondson and H.J. von
Goodman, R. and Harper, S. (eds) (2008). Ageing in Asia: Asia’s position in the Kondratowitz), pp. 261–74. Policy, Bristol.
new global demography. Routledge, London. Katz, S. (1996). Disciplining old age: the formation of gerontological knowledge.
Graham, J. and Stephenson, P. (eds) (2010). Contesting ageing and loss. University Press of Virginia, Charlottesville.
University of Toronto Press, Toronto. Katz, S. (2002). Busy bodies: activity, aging, and the management of everyday
Hagestad, G. (2006). Transfers between grandparents and grandchildren: life. Journal of Aging Studies, 14(2), 135–52 .
the importance of taking a three-generation perspective. Zeitschrift fűr Keating, N. (ed.) (2008). Rural ageing: a good place to grow old? Policy,
Familienforschung, 3, 315–32. Bristol.
Hank, K. and Buber, I. (2009). Findings from the 2004 Survey on Health, Knight, C., Haslam, A., and Haslam, C. (2010). In home or at home? How
Ageing and Retirement in Europe. Journal of Family Issues, 30(1), 53–73. collective decision-making in a new care facility enhances social
Hareven, T. (1982). The life course and aging in historical perspective. In: interaction and wellbeing amongst older adults. Ageing and Society,
Aging and life course transitions: an interdisciplinary perspective (ed. 30(8), 1393–418.
T. Harveven), pp. 1–26. Guildford, New York. Kohli, K. (1987). Retirement and the moral economy: an historical
Hareven, T. (1984). Family time and industrial time: the relationship between interpretation of the German case. Journal of Aging Studies, 1, 125–44.
the family and work in a New England industrial community. Cambridge Kohli, M. (2007). The institutionalization of the life course: looking back to
University Press, Cambridge. look ahead. Research in Human Development, 4, 253–71.
Harper, S. (ed.) (2004). Families in ageing societies: a multi-disciplinary Komp, K., and Van Tilburg, T. (2010). Ageing societies and the welfare state:
approach. Oxford University Press, Oxford. where the intergenerational contract is not breached. International
Harper, S. (2005). Grandparenthood. In: Cambridge handbook of age and Journal of Ageing and Later Life, 5(1), 7–11.
ageing (eds M. Johnson et al.), pp. 422–8., Cambridge University Press, Krekula, C. (2007). The intersection of age and gender: reworking gender
Cambridge. theory and social gerontology. Current Sociology, 55(2), 155–71.
Harper, S. (2006). Ageing societies: myths, challenges and opportunities. Künemund, H. and Rein, M. (1999). There is more to receiving than needing:
Hodder Arnold, London. theoretical arguments and empirical explorations of crowding in and
Havighurst, R. and Albrecht, R. (1953). Older people. Longmans, Green, crowding out. Ageing and Society, 19, 93–121.
London. Kunow, R. (2009). The coming of age: the descriptive organization of later life.
Havighurst, R., Neugarten, B., and Tobin, J. (1963). Disengagement and In: Representation and decoration in a postmodern age (eds A. Hornung
patterns of aging. Chicago University Press, Chicago. and R. Kunow), pp. 295–309. Winter, Heidelberg.
Hazan, H. (2009). Beyond dialogue: entering the fourth space in old age. In: ) Kunow, R. (2010). Old age and globalization. In: What does it mean to grow
Valuing older people: a humanist approach to ageing (eds R. Edmondson old? (eds T. Cole et al.) pp. 93–318. Johns Hopkins University Press,
and H.J. von Kondratowitz), pp. 91–104. Policy, Bristol. Baltimore.
Hazan, H. (2011). Gerontological autism: terms of accountability in the Lancaster, H.O. (1990). Expectations of life: a study in the demography,
cultural study of the category of the Fourth Age. Ageing and Society, statistics and history of world mortality. Springer, New York.
31(7), 1125–40. Laslett, P. (1987). The emergence of the Third Age. Ageing and Society, 7(2),
HelpAge International (2000). The mark of a noble society. HelpAge 133–60.
International, London. Laub, J.H. and Sampson, R.J. (2003). Shared beginnings, divergent lives:
Hepworth, M. (2000). Stories of ageing. Open University Press , delinquent boys to age 70. Harvard University Press, Cambridge.
Maidenhead. Levenson, M., Aldwin, C., and Cuperino, A. (2001). Adult development
Hochschild, A. (1976). Disengagement theory: a logical, empirical, as self-transcendence: extending the liberative model. In: Maturidade
and phenomenological critique. In: Time, roles, and self in old age e Velhice: Um enfoque multidisciplinario (ed. A.L. Neri), pp. 99–116.
(ed. J.F. Gubrium), pp. 53–87. Human Sciences Press, New York. Papirus, Sao Paulo.
Lowenstein, A., Katz, R., and Biggs, S. (2011). Rethinking theoretical and Nuttman-Shwartz, O. (2008). Bridging the gap: the creation of continuity by
methodological issues in intergenerational family relations research. men on the verge of retirement. Ageing and Society, 28, 185–202.
Ageing and Society, 31(7), 1077–84. Olshansky, S.J., Carnes, B., and Cassel, C. (1993). The aging of the human
Lubitz, J., et al. (2003). Health, life expectancy, and health care species. Scientific American, 286(4), 46–52.
spending among the elderly. New England Journal of Medicine, 394, Peace, S. and Hughes, J. (eds) (2010). Reflecting on user-involvement and
1048–55. participatory research. Centre for Policy on Ageing, London.
Lüscher, K. and Pillemer, K. (1998). Intergenerational ambivalence: a new Peace, S., Holland, C., and Kellaher, L. (2005). Environment and identity in
approach to the study of parent–child relations in later life. Journal of later life. Open University Press, Maidenhead.
Marriage and Family, 60(2), 413–25. Phillipson, C. (1998). Reconstructing old age: new agenda in social theory and
Macmillan, R. (2005). The structure of the life course: classic issues and practice. Sage, London.
current controversies. In: The structure of the life course: standardized? Phillipson, C. (2003a). Globalization and the future of ageing: Developing
individualized? differentiated (ed. R. Macmillan), pp. 3–23. Elsevier, a critical gerontology. Sociological Res, 8(4), <http://www.socresonline.
Amsterdam. org.uk/8/4/phillipson.html> (accessed 12.11.2011).
Manheimer, R. (2005). The older learner’s journey to an ageless society: Phillipson, C. (2003b). From family groups to personal communities.
lifelong learning on the brink of a crisis. Journal of Transformative In: Global aging and challenges to families (eds V. Bengtson and A.
Education, 3(3), 198–220. Lowenstein), pp. 54–69. Aldine de Gruyter, New York.
Matthews, S.H. and Sun, R. (2006). Incidence of four-generation family Phillipson, C. (2009). Social welfare, ageing and inequality in a post-industrial
lineages: is timing of fertility or mortality a better explanation? Journal of society. In: The welfare state in post-industrial society (eds J. Powell and J.
Gerontology(B), 61(2), 99–106. Hendricks), pp. 57–70. Springer, New York.
Mayer, K.-U. (2004). Whose lives? How history, societies and institutions define Phillipson, C. and Baars, J. (2007). Social theory and social ageing. In: Ageing
and shape life courses. Research in Human Development, 1, 161–87. in society: European perspectives on gerontology (eds J. Bond et al.), pp.
Mayer, K.-U. (2009). New directions in life course research. Annual Review 68–84. Sage London.
of Sociology, 35, 413–43. Portacolone, E. (2011). The myth of independence for older Americans living
McDonald, L. (2011). Theorising about ageing, family and immigration. alone in the Bay area of San Francisco. Ageing and Society, 31(5), 803–28.
Ageing and Society, 31(7), 1180–201. Powell, J. (2006). Social theory and aging. Rowman and Littlefield, Lanham.
McGee, E. (2002). Misconceptions and misapprehensions about population Powell, J. (2011). Aging, theory and globalization. Nova Science, New York.
aging. International Journal of Epidemiology, 31(4), 750–3. Ranzijn, R. (2002). The potential of older adults to enhance community
McGee, H., et al. (2011). Quality of life and beliefs about ageing. In: Fifty plus quality of life: links between positive psychology and productive aging.
in Ireland 2011: first results from the Irish Longitudinal Study on Ageing Ageing International, 27(2), 30–55.
(eds A. Barrett et al.), pp. 265–92. TILDA, Dublin. Rau, R., et al. (2008). Continued reductions in mortality at advanced ages.
McKee, P. and Barber, C. (1999). On defining wisdom. International Journal Population Development Review, 34(4): 747–86.
of Aging and Human Development, 49, 149–64. Razanova, R., Dosman, D., and Gierveld, J. de J. (2008). Participation in rural
Means, R. (2007). Safe as houses? Ageing in place and vulnerable people in contexts: community matters. In: (2008) Rural ageing: a good place to
the UK. Social Policy and Administration, 41(1), 65–85. grow old? (ed. N. Keating), pp. 75–86. Policy, Bristol.
Mehta, K. (2009). Spirituality: a means for achieving integration in Rees Jones, I., et al. (2008). Ageing in a consumer society: from passive to active
personal and community spheres in an ageing Singapore. In: Valuing consumption in Britain. Policy, Bristol.
older people: a humanist approach to ageing (eds R. Edmondson and Reher, D.S. (1998). Family ties in Western Europe: persistent contrasts.
H.J. von Kondratowitz), pp. 37–50. Policy, Bristol. Population and Development Review, 24(2), 203–34.
Minkler, M. and Cole, T. (1991). Political and moral economy: not such Riley, M., Foner, A., and Riley, J.W. (1999). The ageing and society paradigm.
strange bedfellows. In: Critical perspectives on aging: the political and In: Handbook of theories of ageing (eds V. Bengtson and K. Schaie), pp.
moral economy of growing old (eds M. Minkler and C. Estes), pp. 37–49. 327–43. Springer, New York.
Baywood, Amityville. Riley, M., Kahn, R., and Foner, A. (eds) (1994). Age and structural lag. Wiley,
Moody, H.R. (1988). Toward a critical gerontology: the contribution of the New York.
humanities to theories of ageing. In: Emergent theories of aging (eds Riley, M.W. and Riley, J.W. (2000). Age integration: conceptual and historical
J.E. Birren and V.L. Bengtson), pp. 19–40. Springer, New York. background. Journal of Gerontology (B), 40(3), 266–70.
Moody, H.R. (2002). Conscious ageing: a strategy for positive development Rothermund, K. and Brandstädter, J. (2003). Coping with deficits and losses
in later life. In: Mental wellness in aging: strength-based approaches in later life: from compensatory action to accommodation. Psychology
(eds J. Ronch and J. Goldfield), pp. 139–60. Health Professions Press, and Aging, 18, 896–905.
Baltimore. Sayer, A. (2000). Realism and social science. Sage, London.
Mosca, I. and Barrett, A. (2011). Retirement and labour market participation. Schachter-Shalomi, Z. and Miller, R. (1995). From age-ing to Sage-ing: a
In: Fifty plus in Ireland 2011: first results from the Irish Longitudinal Study profound new vision of growing older. Warner, New York.
on Ageing (eds G. Barrett et al.), pp. 219–42. TILDA, Dublin. Scharf, T. (2009). Too tight to mention: unequal income in older age. In:
Motel-Klingebiel, A. (2006). Quality of life in old age, in equality and welfare Unequal ageing: the untold story of exclusion in old age (eds P. Cann and
state reform: empirical comparisons between Norway, Germany and M. Dean), pp. 25–52. Policy, Bristol.
England. In: Quality of life in old age (eds H. Mollenkopf and A. Walker), Scharf, T., Phillipson, C., and Smith, A. (2005). Social exclusion of older
pp. 85–100. Kluwer, Boston. people in deprived urban communities of England. European Journal of
Motel-Klingebiel, A., Tesch-Römer, C., and von Kondratowitz, H.J. (2005). Ageing, 2, 76–87.
Welfare states do not crowd out the family: evidence for mixed Schwanen, T. and Ziegler, F. (2011). Wellbeing, independence and mobility:
responsibility from comparative analyses. Ageing Society, 25(6): 863–82. an introduction. Ageing & Society, 31(5), 719–33.
Nelson, T. (2004). Ageism: stereotyping and prejudice against older persons. Sennett, R. (1998). The corrosion of character: the personal consequences of
MIT Press, Cambridge, MA. work in the new capitalism. Norton, New York.
Nhongo, T.M. (2004). The changing role of older people in African households Silverstein, M. and Attias-Donfut, C. (2010). Intergenerational relationships
and the impact of ageing on African family structures. Presentation to of international migrants in developed nations: the United States and
Ageing in Africa Conference. <http://www.kubatana.net/docs/hivaid/ France. In: The Sage handbook of social gerontology (ed. D. Dannefer), pp.
helpage_impact_of_ageing_0408.pdf> (accessed 12.08.2011). 177–89. Sage, London.
Sixsmith, A. and Sixsmith, J. (2008). Ageing in place in the United Kingdom. Uhlenberg, P. (1996). Mutual attraction: demography and life-course analysis.
Ageing International, 32(3), 219–35. The Gerontologist, 36, 226–9.
Smith, A. (2009). Ageing in urban neighbourhoods: place attachment and Urry, J. (2007). Mobilities. Polity, Cambridge.
social exclusion. Policy, Bristol. Victor, C. (2008). The social world of older people: understanding
Sontag, S. (1972) The double standard of aging. Saturday Review of Literature, loneliness and social isolation in later life. Open University Press,
39, 29–38. Maidenhead.
Stacey, J. (1996). In the name of the family: rethinking family values in the Victor, C. (2010). Ageing, health and care. Policy, Bristol.
postmodern age. Beacon, Boston. Victor, C., Scambler, S., and Bond, J. (2009). The social world of older people:
Sternberg, R. (1998). A balance theory of wisdom. Review of General understanding loneliness and social isolation in later life. Open University
Psychology, 2, 347–65. Press, Maidenhead.
Stuckelberger, A., Abrams, D., and Chastonay, P. (2012). Age discrimination Vincent, J., Phillipson, C., and Downs, M. (eds) (2006). The futures of old age.
as a source of exclusion in Europe: the need for a human rights plan for Sage, London.
older persons. In: From exclusion to inclusion in old age: a global challenge Walker, A. (1981). Towards a political economy of old age. Ageing Society,
(eds N. Keating and T. Scharf ). Policy, Bristol. 1(1), 73–94.
Tanner, D. (2010). Managing the ageing experience: learning from older people. Walker, A. (ed.) (1996). The new generational contract: intergenerational
Policy, Bristol. relations, old age and welfare. University College London Press,
Taylor, C. (2002). Modern social imaginaries. Public Culture, 14(1), 91–124. London.
Teuscher, U. and Teuscher, C. (2007). Reconsidering the double standard of Walsh, K. and O’Shea, E. (2010). Marginalised care: migrant workers caring
aging: effects of gender and sexual orientation on attractiveness ratings. for older people in Ireland. Population Ageing, 3, 17–37.
Personality and Individual Differences, 42, 631–39. Warnes, A. (2009). Migration and age. In: The Sage handbook of gerontology
Thane, P. (2005). The age of old age. In: A history of old age (ed. P. Thanne), (eds D. Dannefer and C. Phillipson), pp. 389–404. Sage, London.
pp. 9–29. Oxford University Press, Oxford. Wenger, G.C. (1987). Dependence, interdependence and reciprocity after
Timonen, V. and Arber, S. (2012). Contemporary grandparenting: changing eighty. Journal of Aging Studies, 4(1), 355–77.
family relationships in global contexts. Policy, Bristol. Wilińska, M. (2010). Because women will always be women and men are
Timonen, V., Kamiya, Y., and Matym, S. (2011). Social engagement of older just getting older: intersecting discourses of age and gender. Curr Sociol,
people. In: Fifty plus in Ireland 2011: first results from the Irish Longitudinal 58(6), 879–96.
Study on Ageing (eds A. Barrett et al.), pp. 51–71. TILDA, Dublin. Wingens, M., et al. (eds) (2011). A life-course perspective on migration and
Tornstam, L. (2005). Gerotranscendence—a developmental theory of positive integration. Springer, New York.
aging. Springer, New York. Woolf, V. (1975). The letters of Virginia Woolf, vol. 1 (ed. Nicholson N).
Tornstam, L. (2006). The complexity of ageism: a proposed typology. Harcourt, Brace, Jovanovich, London.
International Journal of Ageing and Later Life, 1(1), 43–68. World Health Organization (2011). WHO global network of age-friendly cities
Tornstam, L. (2007). Stereotypes of old people persist: a Swedish ‘Facts on and communities. <http://www.who.int/ageing/age_friendly_cities_
Aging Quiz’ in a 23-year comparative perspective. International Journal network/en/index.html> (accessed 30.10.2011).
of Ageing and Later Life, 2(1), 33–59. Yang, K. and Victor, C. (2011). Age and loneliness in 25 European nations.
Torres, S. (2011). Cross-cultural differences in aging. In: An introduction to Ageing and Society, 31(8), 1368–88.
gerontology (ed. I. Stuart-Hamilton), pp. 340–62. Cambridge University Yeates, N. (2001). Globalisation and social policy. Sage, London.
Press, Cambridge. Yeates, N. (2005). A global political economy of care. Social Policy and Society,
Townsend, P. (1955). The family life of old people: an investigation in East 4(2), 227–34.
London. Sociological Review, 3(2), 175–95. Ylänne, V., Williams, A., and Wadleigh, P. (2009). Ageing well? Older people’s
Townsend, P. (1981). The structured dependency of the elderly: the creation health and well-being as portrayed in UK magazine advertisements.
of social policy in the twentieth century. Ageing and Society, 1(1), 5–28. International Journal of Ageing and Later Life, 4(2), 33–62.
Treas, J. (1995). Older Americans in the 1990s and beyond. Population Young, M. and Wilmott, P. (1957). Family and kinship in East London.
Bulletin, 50(2), 1–46. Penguin, Harmondsworth.
Tulle, E. (2008). Ageing, the body and social change. Palgrave, London. Ziegler, F. and Schwanen, T. (2011). ‘I like to go out and be energised by
Twigg, J. (2012). Fashion and age: dress, the body and later life. Berg, different people’: an exploratory analysis of mobility and wellbeing in
London. later life. Ageing and Society, 31(5), 758–81.
CHAPTER 4
Transforming concepts of
ageing: three case studies
from anthropology
Sharon R. Kaufman, Julie Livingston,
Hong Zhang, and Margaret Lock
Anthropological approaches to ageing have customarily addressed continues to be very unevenly distributed, and essentially unavail-
the phenomenon of growing old as subjectively experienced by indi- able for many people due largely to geographical isolation and/or
viduals and their families. Emphasis is usually given to the way in poverty.
which ageing in societies around the world is inevitably embedded Other variables are bringing about what promise to be long-lasting
in specific cultural, social, and political contexts that profoundly changes with respect to ageing and the local meanings attributed to
influence the latter part of the life cycle. Such findings are often sup- it. Kinsella (2009) reminds us: ‘the transition from a youthful to a
plemented with demographic and epidemiological data, and with more aged society has occurred gradually in some nations, but will
national and local healthcare policy information. In addition, use by be compressed in many others’. Whereas in several European coun-
patients and families of biomedical facilities and of the very many tries the demographic transition to an ‘ageing society’—that is, the
indigenous forms of traditional medicine, literate and non-literate, rapid increase in the over-60 population—took place over a period
have been well documented by anthropologists (Ikels, 1997, 2001; of 80–115 years, in contrast, first in Japan and now in China and
Cohen, 1998; Lamb, 2000; Traphagan, 2000; Barker, 2009). other parts of Asia, this transition is taking place over the course of
Increasingly, it has become clear that representing the ageing 25 years or less and is being driven by precipitous declines in fertil-
experience as timeless, unchanging, and fully informed by local ity levels, exacerbated greatly in the case of China by the one-child
cultures and social arrangements is highly misleading. This has policy.
become particularly evident over the past two decades. Although A further contributing factor is the surprising extension of life
local value systems without doubt play a role, other factors ensure expectancy among older people. The proportion of over-80s is
that the very ideas of what constitute ageing are being transformed, increasing at an unprecedented rate, and centenarians are today the
in many places radically and exceedingly rapidly. Among other most rapidly growing age group in many countries, including Japan,
variables, key factors are the increasing availability and distribution most of Europe, the Antipodes, Canada, and the US. It is this exten-
of biomedical technologies, simple and advanced, the global spread sion of life expectancy that has taken everyone by surprise, includ-
of biomedical knowledge, and, above all, the diffusion among ing demographers (Kertzer and Laslett, 1995). During the period
local populations of an understanding about what biomedical care 2002–2007, the net size of the world’s older population increased
makes possible. Understanding at the local level is facilitated by the by 925,000 persons per month; 75% of this change occurred in the
internet, and perhaps especially by cell phones that have increas- developing world (Kinsella, 2009). The absolute numbers of indi-
ingly become a staple in the everyday lives of people in large areas viduals in these countries is substantially greater than in the devel-
of the world. oped world, and very often poverty is rife, together with an absence
Generation of biomedical knowledge, including that resulting of medical care. The impact of these demographic changes on fam-
from randomized controlled trials (RCTs), as well as numerous ily structure is significant, and the normative expectation that older
technological innovations, today takes place in many parts of Asia, people will inevitably be taken care of by their families is no longer
Latin America, and Africa, in addition to the West, with the result realistic.
that the production of biomedical knowledge and its associated Further factors that have exacerbated the effects of recent demo-
technologies is global. Even so, it must be stressed that biomedicine graphic changes have been the move to a cash economy in all but
40 oxford textbook of old age psychiatry
the most remote of areas, accompanied often by a dramatic rise Despite these differences, a broad set of epidemiological, social,
in increased mobility worldwide. Labour migration, often involv- economic, and cultural changes have taken place across the region,
ing rural to urban removal or emigration to other countries and among them, migration and urbanization; demographic upheaval
continents, entails extended absence of younger adults from the caused by the AIDS epidemic; a rise in consumerism and personal
family, sometimes for years on end. These absent individuals are debt; the spread of formal education; steadily increasing access to
primarily men, but women who work as servants and in factories biomedicine, along with rising rates of chronic illness. Even though
are also often absent from their families, and it is these people who many people cope with chronic, seasonal, or occasional food short-
would formerly have been primarily responsible for care of the ages across much of southern Africa due to urbanization and the
older generation. In addition, extended families are today subject now ubiquitous cash economy, an increased consumption of proc-
to complete disruption when the entire younger generation moves essed foods is common, some of which have become dietary sta-
away from their natal home, leaving older people largely to fend for ples. Many of those are manufactured in South Africa but are sold
themselves. throughout the region. These changes in diet mean that hyperten-
Other local contingencies may well arise, including wars, fam- sion, diabetes, and obesity are on the rise, as is cancer. The numer-
ine, appropriation of land by those in power, and so on, to which ous transformations, together with the ageing of the population,
must be added the effects of epidemics, notably that of HIV. have resulted in new kinds of pressures on caregiving at both fam-
Government-controlled policies in connection with reproduction ily and institutional levels, making ageing a precarious and fraught
and care of both healthy and sick older individuals result in changes process for a good number of individuals.
that often set up unforeseen tensions among the interests of indi- One location within the region, southeastern Botswana, provides
viduals, families, communities, and society as a whole. examples of the complexities of ageing in everyday life, particularly
The vignettes that follow, set in Botswana, China, and the US, as they relate to caregiving within families and public institutions.
respectively, graphically illustrate how very different are the con- Two challenges facing Botswana resonate in other African countries
cepts of ageing in these three locations and, further, how these as well. First, how does the new demography of ageing in Botswana
concepts are debated and challenged within families and with affect family relationships and the family economy? Newly evident
healthcare professionals, as people struggle with how best to handle pressures result from the combined challenges of caring for young
their frail and sick ageing relatives. adults dying from AIDS, for the orphans they leave behind, and
for those with chronic diseases associated with ageing. All this
puts intense pressures on female caregivers, many of whom are old
Vignette 1—Botswana (Julie Livingston) themselves. These developments are transforming ideas about nor-
In one sense, sub-Saharan Africa is quite a youthful place. According mal ageing and about how the latter part of the life cycle should be
to a recent United Nations report (2009: 6), the median age on lived. Historically, older people almost without exception resided
the continent is 19.6 years, approximately half the median age for with their adult daughters and grandchildren and received care
Europe. And while the proportion of old people is not nearly as high from them and from extended family members. Families who left
as in other regions of the world, the actual number of people over their frail older relatives to care for themselves were vulnerable to
the age of 60 is significant and rising steadily. The same UN report community criticism. In recent years, however, a new pragmatics of
projected an average annual growth rate of 3.35% for Africans over care has emerged. Able older people are often called upon to pro-
60 years of age between 2009 and 2050, the highest such growth rate vide care for those younger than they are or, at the very least, are left
in the world. In sub-Saharan Africa in 2005, there were 34 million to manage more of their own daily needs.
people age 60 and over, and this number is projected to increase to Second, new challenges arise in the face of increased medicaliza-
over 67 million by 2030 (Velkoff and Kowal, 2006: 56). tion of chronic illnesses and the uneven distribution of biomedi-
In southern Africa, the growing population of older people cal technologies. Now that those over 60 are increasingly grappling
faces significant social, economic, and health-related challenges with chronic, often life-threatening illnesses, hospitals have become
as they age in a region shaped by pervasive poverty, unevenly deeply involved in caregiving and the prolongation of life. While
distributed infrastructure, and a profound epidemic of HIV/ more sophisticated hospitals, clinics, and technologies are emerg-
AIDS. Southern Africa is a region of great diversity. It includes ing, the availability of high-tech care is limited and triage decisions
the economic powerhouse of South Africa, which is struggling must be made. At the same time, basic medical resources continue
amid persistently high rates of unemployment to rectify the mas- to be in short supply and unevenly distributed. Sensitive social and
sive maldistribution of resources, opportunities, and infrastruc- ethical questions continually arise when impoverished older indi-
ture created during over a century of institutionalized racism. It viduals are discharged from the hospital and returned to their local
contains Zimbabwe, with its recent history of internal oppression, communities where even basic medications are often in short sup-
economic collapse, and massive out-migration of economic and ply, in order to make room for other patients in even greater need.
political migrants. The region also includes Mozambique, which Four-and-a-half decades of postindependence economic devel-
suffered through decades of civil war (covertly fomented by the opment and well-planned social services make Botswana a harbin-
South African apartheid government), and Botswana, marked by ger of sorts for future patterns of ageing and health in southern
the historical absence of violent conflict. The region incorporates Africa. As in other developing countries, in Botswana, tradition-
sparsely populated, arid Namibia and Botswana and the densely ally being an ‘elder’ was a time when one had both social authority
populated mountain nation of Lesotho. Southern Africa is home and economic security. However, today ageing increasingly brings
to numerous ethnic groups and language families, and to a small with it social disempowerment and bodily affliction. The rise of
but significant middle class, which coexists with entrenched rural chronic illnesses, particularly diabetes, cancer, and hypertension,
and urban poverty. means more debility in later life, at an historic moment when the
CHAPTER 4 transforming concepts of ageing: three case studies from anthropology 41
pervasiveness of HIV/AIDS drains caregivers, and when the social company and help them around the home. Their parents, in turn,
authority of many older people to command care and resources is come to visit for long weekends and holidays, bringing parcels of
diminishing. food, cash, clothing, and other contributions. Pregnant women
often return home during their confinement, as do sick adults who
Development, social welfare, and ageing require extensive nursing care.
Since its independence from British colonial rule in 1966, Botswana Village homes usually abut those of other family members, so that
has in many ways become a best-case scenario in the region for its residents of a ward of a village are often (at least distantly) related
relative political and economic stability, and for the strength and to one another. At one time, this meant that neighbours cared for
extent of its carefully planned infrastructure and social welfare older people living in other households by sending along their chil-
programmes. It is often held up as a model of successful develop- dren with cooked meals, helping to fetch water or firewood, and the
ment on the continent. At independence, Botswana was one of the like. Yet, over the past decades, socioeconomic changes stemming
poorest countries in the world, but within a few years vast diamond from the cash economy (such as the increased demand for migrant
wealth was discovered in several locations. The new government, labour and the limitations placed on children’s labour as a result of
less hindered by ethnic tensions than in neighbouring countries, formal education) have narrowed the scope of personal responsibil-
given the overwhelming majority of the Tswana ethnic group, pro- ity for others living nearby (Livingston, 2005). Once unthinkable,
tected much of that wealth as a public asset. Over the past four dec- it is now the case that older people sometimes live alone, although
ades, the state has prudently invested in infrastructure, education, this is still rare.
and social welfare programmes. Standards of living have been steadily rising in Botswana over the
Batswana (as citizens of Botswana are called) have a system of past decades, but, even so, poverty and economic insecurity remain
universal healthcare. Food baskets and other supports are available pervasive problems. While in 1966 the GDP per capita was only
for the destitute (including orphans). There is a ready access to clean around $70, by 2007 it had risen to $6,120, and Botswana has gone
water, a network of tarred roads, telecommunications, and various from one of the poorest countries in the world to a middle-income
other public goods and services. Batswana have experienced peace- nation (World Bank, 2008). Yet 47% of the population lives below
ful, democratic transitions in political leadership. But they have the poverty line, and the unemployment rate is 15.8%, a figure that
faced significant challenges in recent decades, many of which shape does not take into account the vast amount of underemployment
the experiences of ageing, similar to others across the region. (United Nations Development Programme, 2011a). Poverty has a
Botswana was at the epicentre of the HIV/AIDS epidemic on major impact on ageing, and so does the coexistence of wealth and
the continent. During the late 1990s, rates of HIV infection among poverty within the society. Rising wealth has led to rampant con-
pregnant women in some towns approached 40%. In 2007, nearly sumerism, often resulting in crushing personal debt (Livingston,
a quarter (23.9%) of people aged 15–49 were HIV positive, the sec- 2009). Increasingly, younger people feel torn between their desire
ond highest adult prevalence rate in the world, behind Swaziland for consumer goods and the social status such articles signify, and
(World Health Organization, 2011). The epidemic has significantly their responsibility to share their wages with family, including its
affected older people socially, economically, physically, and, of older members.
course, emotionally. AIDS has exacted a high death toll on individ- Unlike much of Africa, Botswana did not accumulate significant
uals in their prime, and older women, in particular, have increas- foreign debt and was spared the structural adjustment policies of
ingly found themselves providing care to orphaned grandchildren the IMF and World Bank that gutted social services and infrastruc-
(Thupayagale-Tshweneagae, 2008). Older parents have suffered the ture across much of Africa. Botswana, along with South Africa and
loss of crucial and expected economic support from wage-earning Namibia, is one of the few African countries to provide an old-age
adult children who became too sick to work and instead returned pension. Pensions represent crucial economic resources for poor
home, seeking care as they died. Many seniors have buried the chil- households, often forming the basis of the monthly household
dren they assumed would outlive them. Extended families struggle budget. Across southern Africa, the decline of peasant agriculture
with the care of children (often orphaned), frail older individuals, and the accompanying rise of wage labour during the past century
and those with AIDS. The national health system has been over- have eroded the former gerontocracy. In South Africa particularly,
whelmed by the HIV epidemic which has drained health resources means-tested pensions have helped reinstate the value of older fam-
away from older people. ily members, who, by virtue of their control over pension payments,
Although 61% of Batswana live in cities and large towns, most remain critical economic actors and enjoy the power and the con-
maintain a home in their natal village where they retire in later life. siderable responsibilities associated with income redistribution.
Others live part or all of their adult lives in their parents’ home In Botswana, the pension, while appreciated, is a much smaller
or a neighbouring compound ( United Nations Development amount than in South Africa, and its deployment exposes ongoing
Programme, 2011b). Most elderly people, therefore, reside in vil- moral negotiations over care for seniors in a changing socioeco-
lage households that incorporate three or four generations of nomic landscape. In 2011, the universal (not means-tested) pen-
extended family, including those who might not have spouses or sion is given to those aged 65 and above, who are paid 220 Pula
adult children to support them. Marriages are usually monoga- (US $28) per month. This money is a critical resource for older
mous, but the majority of Batswana women choose not to marry, people, but its effects within families reveal the fraught politics of
opting instead to participate in a series of long-term partnerships, ageing amid a changing economy. The pension was introduced in
which do not give men or the men’s parents control over children 1997 to supplement family resources, and to empower older peo-
from these unions, nor obligate women to the care of their partner’s ple by giving them some cash of their own for use in the market
parents. Children of adults working in town are often sent to their economy. Yet, almost immediately, many older people complained
village-based grandparents or aunts to keep their elderly relatives that the pension payments resulted in less care from family, which
left them ‘begging from our children’, as one older woman noted. the impression that today dementia associated with HIV and that
Older women might be told by other family members not to take associated with ageing among otherwise healthy older individuals
sugar or tea from the household larder with comments like, ‘after may not be well differentiated in the minds of many Batswana, and
all, you have your own money.’ For many adult daughters and sons clearly this calls for some careful research at a future date.
working menial jobs, a desire to pool the family resources is under-
standable but nonetheless fraught with moral implications. Thus, Biomedicine amid a changing epidemiology
a pension may provide an older mother with better food security The public health system in Botswana is extensively distributed at
and the pleasures of pocket money, but these gains are offset by the the primary level, but it is not well equipped to address the prob-
corresponding loss in some households of gifts of food and cash lems associated with ageing. Within Botswana, as across southern
from adult children that are important cultural manifestations of Africa as a whole, as already noted, rates of cancer, diabetes, and
love and affection (Van der Geest, 1997; Livingston, 2003). hypertension have been rising as the population ages, and changes
in diet, daily exercise levels, and other lifestyle factors are clearly
Pressures on care implicated. Obesity has become a significant problem, particu-
The rise of disabling chronic illnesses among adults in their 50s larly among older women. The national healthcare system was not
and 60s is remaking senescence, creating an earlier onset of ‘old designed to cope with these increasingly prevalent conditions, but
age’ in the popular imagination, and straining caregiving net- the Ministry of Health must attempt to do so, while at the same
works, which are increasingly rife with intergenerational tensions. time responding to the epidemic of HIV/AIDS.
Typically, daughters are expected to care for their parents as they At independence, Batswana found themselves with only a hand-
face the frailties of age, while mothers are expected to care for their ful of small mission hospitals and clinics peppered across a vast
children during times of illness. As a result, there is a great deal at country (roughly the size of France), alongside a significant net-
stake in determining the meanings of affliction among older peo- work of private indigenous healers practising Tswana therapeutics,
ple, because whether an affliction is regarded as a manifestation of who had been deregulated and driven underground during coloni-
either senescence or illness determines who is responsible for care. alism. Since then, the state has made a long-term, consistent invest-
Amid the turmoil of these simultaneous epidemics of HIV/AIDS ment in public health. They have created an extensive network of
and chronic illnesses, a triage is underway in many households to clinics and primary hospitals that then feed into the nation’s two
distinguish healthy older people from those who are sick. Older referral hospitals, and have provided healthcare as a right of citizen-
women, as both care providers and care seekers, occupy a pivotal ship. This model of developing extensive coverage in primary care,
position in the networks of caregiving. For example, when the and then later turning to develop specialty care at the tertiary level
60-year-old daughter of an 80-year-old mother becomes debilitated (a process that is just beginning now), is in accord with a national
after a stroke, her mother may attempt to ensure that her daughter’s emphasis on equity in the distribution of services also seen in the
condition is defined as old age rather than illness, thus freeing the pension scheme. It is also in step with the historical trajectory of
older woman from the potentially overwhelming responsibilities development, since services were planned in the 1970s, during the
of caring for her daughter, effectively displacing that responsibility heyday of the movement to set up primary healthcare as the corner-
onto the patient’s own daughters. Meanwhile, adult women, who stone of what was then called International (now Global) Health.
are often caring for children as well as sick adults, might in turn try This means that in Botswana, older people have had ready access to
to resist the responsibilities of caring for their mothers, and thereby primary care provided mainly by nurses and nursing assistants in
attempt to define their parents’ frailties as illness. Thus, even as the both rural villages and urban neighbourhoods (although older peo-
physical markers of senescence reflect changing epidemiological ple are more likely to live in villages than the city). But it is much
realities, their meanings (is this illness or is it old age?) are negoti- more difficult for them to access the very limited specialty services
ated in the context of a larger set of pragmatic and social concerns. designed to deal with the increasing chronic and life-threatening
Such negotiations inform the way in which older patients and their medical needs that are so often part of the ageing process today.1
caregivers approach healthcare institutions. Illness implies the Biomedicine is by no means the only type of medical serv-
need for intervention, and possible alleviation, while senescence ice available in southern Africa; it coexists with indigenous and
implies a natural and benign process of ageing, which may need Christian healing practices as well as other types of practice. But
little palliation. in Botswana, biomedicine is the system most used by older people.
It is of note that in Botswana, dementia was formerly understood This is in part because it is public and therefore free (or subject to
as a normal part of ageing and thought of as becoming baby-like very nominal fees), as opposed to traditional healing which can be
once again. Dementia also had a spiritual dimension, one of being quite costly. It is also because the frailties associated with ageing are
close to the ancestors, and people might even send children to go usually understood by Batswana as simply part of becoming old, or
and touch a demented person, or visit them for good fortune. Not alternatively thought of as novel diseases of modernization (such as
surprisingly, given the extent of the HIV epidemic in Botswana, diabetes, hypertension, and cancer). For these reasons, families are
AIDS-related dementia is strikingly apparent, and the English often hesitant or unable to marshal significant resources to pay for
term ‘dementia’ is now part of the local lexicon. Livingston has indigenous therapeutics that, everyone knows, neither ameliorate
1 Botswana is not alone in its desire for equity; this politics holds across the region. By contrast, however, neighbouring countries like South Africa, Namibia,
and Zimbabwe inherited highly inequitable colonial health systems that were built in large part to serve the needs of their significant European populations
(something Botswana did not have). As a result, there were well-developed (in some cases world-class) specialty care in tertiary hospitals, but very poor
quality and distribution of primary care.
senescence nor manage the new chronic diseases. Under the cir- in these new medical settings, and become entangled with shifting
cumstances, the cheaper biomedicine is thought of as offering bet- epidemiological and technological possibilities. There is no longer
ter, albeit still inadequate support. a formerly recognized pathway for managing the end of life. In a
Beginning in the mid-1990s, health facilities at all levels in country with no residential hospice care and no nursing homes,
Botswana (and across southern Africa) were overwhelmed with most older people die at home. On the one hand, this allows them
patients suffering the complications of HIV/AIDS. This created a to be in familiar surroundings in the care of their kin. On the other
crisis of confidence in both biomedical institutions and the practices hand, families may be overwhelmed as a result, and may not have
of indigenous healers. In response, over the past decade, the state access to some of the basic palliative technologies used in care of
has implemented the first-ever public antiretroviral programme on those dying in hospitals.
the continent, and it has also begun to contend more directly with The contrasting cases of end-of-life care for two cancer patients
the increased incidence of chronic diseases associated with ageing. witnessed by Livingston in 2007 highlight how patients and car-
Weekly hypertension clinics are now run in some provincial hospi- egivers have begun to seek out pain relief for terminally ill older
tals (staffed by general practitioners) and basic diabetes testing and patients, with markedly different outcomes. At present, opiate
management is increasingly available at primary hospitals as well. analgesics are available only at the two central referral hospitals in
But there are very limited mental health services in the country, the country, and access to these referral hospitals in great meas-
and no services targeting the mental health needs of older people. ure determines the burden of pain that patients dying from cancer
The success of antiretrovirals has helped to prompt a new and will have to undergo. These cases also illustrate the uneven distri-
growing biotechnical optimism among some Batswana as they bution of biomedical technologies within and among facilities. In
become increasingly aware of the host of biotechnologies now one facility, an older patient might be given high-tech interventions
available in the more sophisticated biomedical settings. Frequent like brachytherapy, but many of the usual scans, blood tests, and
media reports about organ transplantation, open-heart surgery, intravenous drugs routinely made use of in wealthier countries are
and other therapies available in Europe, the US, Asia, and South not available.
Africa’s elite research hospitals have meant that patients and rela- Mma Mosadi,2 a cervical cancer patient in her 60s, lived in a large
tives have begun to inquire about such possibilities during clinical village supplied with a primary care hospital and three clinics. She
encounters. However, in Botswana, specialty services are mainly lived with her seven grown children, none of whom had regular
concentrated in one or two sites throughout the country, are vastly employment, and several grandchildren, and was too ill to continue
overburdened, and can only be accessed through laborious and her work as a laundress. Three people in her household were tak-
uncertain referral processes. And, further, at this tertiary level ing antiretrovirals for HIV. The household received a food ration
the facilities are uneven. For example, since 2002 there has been from the local clinic, but it was not enough to go around. Often
a cancer ward and clinic in the nation’s central referral hospital in there was no food in the house, and meals had to be skipped. In
Gaborone that grapples with an ever-growing volume of patients. 2005, when it had become clear that there were no further curative
But this hospital lacks a nephrologist and a cardiologist, and has or life-extending interventions to be offered to Mma Mosadi, the
no dialysis or MRI machines. Clinical care is provided as a right of gynaecologist at the central hospital, where she had been looked
citizenship in Botswana. This means that a broad socioeconomic after, referred her case back to the primary hospital in her village
cross section of the population appears in primary and tertiary care for management. This meant that Mma Mosadi, like many older
settings alike. While the wealthy may be able to marshal critical people in Botswana, was left to die without adequate management
transportation resources to access hospitals, purchase special foods of the considerable pain and other symptoms of her condition that
or nursing supplies, or domestic labour to assist in home care, they are available in the central hospital.
cannot bribe their way into specialty services in Botswana. Referral By early 2007, Mma Mosadi’s cancer was at the terminal stage,
from a primary care site is a necessary gateway to a specialty con- and she had a fistula that had developed after radiation treatments.
sultation. In the cancer ward it is not unusual to see a banker teth- She spent her days lying on a mattress on the floor in her hut with
ered to the same intravenous pole in the chemotherapy room as an a blanket rolled up to support her on her side. Her oldest daughter
80-year-old woman in a tattered dress from a remote village. stayed up with her each night in her room attempting to massage
her and keep her mother company during her agony. Because of the
Palliation, futility, and end-of-life care fistula, faeces, urine, and blood were all expelled from her vagina,
Widespread knowledge about antiretroviral efficacy as well as vari- and Mma Mosadi had an extreme fear of eating, since defaecation
ous public health campaigns that seed biomedical knowledge into was now so agonizing. As she explained to the visiting anthropol-
the popular domain have engendered new forms of biotechnical ogist, ‘Whether it is day or night, I am always having pain … It
optimism, and families increasingly seek high-tech medical inter- is just the same to be at home or in the [local primary] hospital,
ventions for their older relatives. As a result, in Botswana, new because there isn’t anything they are doing to ease that pain. I asked
forms of end-of–life care and decision-making promise to become them to make another opening to pass the stool, but they said it is
more pervasive as epidemiological, institutional, and social changes impossible.’
combine to bring about what will be the norms of the future. Mma Mosadi’s clinic card showed that she was on ibuprofen.
Respirators, feeding tubes, dialysis, and portable oxygen are not yet Morphine had been ordered at the referral hospital upon her dis-
options for older patients in Botswana. However, questions about charge, but it was not in stock at the primary hospital—so her only
therapeutic futility arise repeatedly amid the idiosyncrasies of care hope for effective pain relief was to find a way back to the specialist
2 Personal names have been anonymized throughout this essay to protect privacy.
oncology ward for help. Yet access to oncology can be brokered hospital with highly restricted visiting hours. His son asked for sur-
only through referral. When she had asked at the primary hospital gery, radiation, chemotherapy, and scans, but he was told that there
for a referral back to the central hospital, she was told that nothing was nothing else to be done, other than to control the pain and
more could be done—exactly what had been written on her card hope to send him home. On 9 March, Rra Molefi could no longer
when she was first discharged from the referral hospital as a termi- swallow, became completely dependent on IV fluids, and was put
nal case. Since the family lacked money to make their way directly on continuous oxygen. A few days later the family held a prayer
to the referral hospital some hundred kilometres away to appeal meeting around his bed. By 15 March, he was unconscious, but the
directly, they were at the mercy of the triage logic of the overloaded clinical staff continued to focus on managing his pain. Four days
health system. Yet Mma Mosadi’s hopes remained focused on the later he died in the ward.
pain relief she knew was just out of reach. She said: These two cases suggest some of the new challenges of increas-
If I were rich, each time that I would go to the hospital I would go ingly medicalized ageing in southern Africa. Specialty services that
straight to [the referral hospital], because they are the ones who are have access to radiation treatments can extend life and ease pain,
actually doing something. Here locally they don’t do anything. Here I and those in extreme pain can be further palliated with medication.
will be in pain and try to tell the doctor and they would never say let However, patients and their families may not be able to pursue essen-
me see where the pain is [by scanning or x-raying]—they would be tial follow-up care to address the sequelae of previous medical inter-
talking to me from the chair only. No doctor here would be saying let ventions. Furthermore, caring for seriously ill older patients at home
me see where and what is happening with this pain. is extremely difficult, and requires a great deal of time and labour on
Meanwhile, one of her daughters who had HIV was seeking a the part of the family, but enables older patients to experience the
way to become enrolled in a clinical trial, since she would then have comfort and familiarity of family life. On the other hand, caring for
access to free transport and entry into the referral hospital where, seriously ill older patients in the hospital gives them access to some
hopefully, she could also bring her mother’s medical records to refill important palliative services and professional nursing care, but also
the long-expired prescription for oral morphine. leaves them socially isolated for long periods of time in institutional-
Rra Molefi, a farmer, was also in his 60s, and a terminally ill can- ized settings. In neither setting, however, is any mental health sup-
cer patient with metastatic abdominal squamous cell carcinoma port available to older patients as they face the existential crises and
when Julie Livingston met him. But, in contrast to Mma Mosadi, pain of serious illness. As more specialty services become available
he rotated in and out of a highly medicalized setting. In November and as ageing is increasingly biomedicalized throughout the country
2006 his family had brought Rra Molefi to the cancer ward of the and in the region, such dilemmas will only intensify amid debates
referral hospital asking that he be admitted, since his wife believed about therapeutic futility, equity, and the shifting debatable bounda-
him to be extremely ill and his care needs were too onerous for her. ries between normal senescence and disease.
But beds in the ward are rationed to prioritize patients on active Across southern Africa, new dilemmas associated with ageing
treatment, and so Rra Molefi was sent back home. pervade overloaded care networks, already reshaped by a cash econ-
By January 2007, Rra Molefi’s state had declined to the point where omy, uneven infrastructure, and shifting epidemiology that affect
he required a series of in-patient transfusions, and so his family was both familial and institutional care settings. These difficulties—in
able to secure a bed for him in the cancer ward of the referral hospi- large part the result of novel diseases and the uncertain promise of
tal. When he was admitted, his son, a soldier, informed the hospital biotechnologies—are experienced across extended family networks
staff that the local clinic did not have the right drugs to relieve his that link rural and urban households and directly affect caregiving.
father’s pain. He also asked if the doctor would scan his father’s New struggles over palliation, the limits of biomedical therapeutics,
shoulder to better identify the source and nature of the pain. His the rationing of high-tech medicine, and novel forms of dying are
relatives further stated that he was in such pain he would wake up already evident among Botswana’s growing population of older can-
in the night crying, underscoring that palliation was a priority for cer patients. Such dilemmas promise to become more pervasive over
them. The scan revealed metastases in Rra Molefi’s shoulders and time as the population ages, and chronic illnesses increase among
he was given oral morphine and scheduled for palliative radiation. those individuals now thought of among the Batswana as ‘elderly’.
This eased his agony somewhat and he went home.
By mid-February, Rra Molefi was readmitted. The metastatic
disease had entered his brain and he arrived with a fever and Vignette 2—China (Hong Zhang)
confusion. Like many older patients, the transition to the strange We turn now to China, where the effects of rapid demographic
environment of the hospital greatly heightened his confusion and changes are more graphically evident than anywhere else in the
anxiety and one night he absconded to the bus station and had to be world.
fetched back to the ward. Another night he crossed to the women’s Although China has the world’s largest population, for most of
side of the ward and tried to climb in bed with one of the patients its history it has been a demographically young society. As recently
whom he mistook for his wife, after which he was put on sedatives. as the early 1970s, China’s median age was 20, the fertility rate
As his father lay curled up in the fetal position dressed in a set of was 5.8, and people classified as old made up just about 7% of the
hospital diapers, his son was very upset and worried that his father population. But, in the past three decades, the population in China
was shivering. He asked the doctor if the staff was feeding his father has started to age at an accelerated rate, transforming it from a
properly, and requested that he be fed by a drip. A nurse explained predominantly young to an ageing society. According to the lat-
that they were indeed feeding him, and that it was not possible to est census in 2010, people 60 years old and over now account for
give proper nutrition by means of an IV. 13.26% of the total population, up from 10.3% in 2000. The result
By 1 March, Rra Molefi had become too sick to be sent back is that China now has a population of old people of 177.6 million,
home as the doctor had hoped, and so now he faced dying in a surpassing the number of senior citizens in all European countries
put together. At the same time, the population in the age group of labour-intensive care. The family tried to hire live-in helpers several
0–14 today accounts for 16.60%, down sharply from 23% in 2000, times, but in vain, because nobody was willing to care for a bedrid-
due to the continuation of China’s strict birth-control policy over den old man. In desperation, the family tried several times to place
the past three decades (He, 2011). In a single decade, China’s fertil- him in a nursing home, but they were turned away, because either
ity rate more than halved from 5.8 in 1970 to 2.2 in 1980, and it many nursing homes were not equipped to care for such patients,
further declined to 1.8 by 2000, and thence to 1.4 in 2010 (Jackson or they were afraid to take such a ‘high-risk’ patient. As a last resort,
et al., 2009: 8; The Economist, 2011). In the meantime, life expect- in 2007, each of his five daughters either gave up her job or took
ancy has continuously climbed from 41 in 1950, to 63 in 1975, and early retirement so that they could take turns to assume full-time
73.5 in 2010. An ageing population is usually taken to be a sign care for their father.
of an improvement in the standard of living and advancements in CCTV aired the Shi family story not only to show that providing
public health, but China’s ageing population is characterized by the long-term care for infirm parents is still principally a family matter
Chinese themselves as ‘premature ageing’ (wei fu xian lao), mean- in China, but also to call public attention to the stress and burden
ing literally ‘growing old before getting rich’, and as such it presents on family members who care for frail elderly parents—a problem
unprecedented challenges. that is exacerbated as China’s life expectancy rises and more older
One challenge is that China has become an ageing society while people suffer from prolonged illnesses and disability. One impor-
still a developing country. When developed countries first crossed tant feature of China’s population ageing is the sharp rise of the
a threshold in which more than 10% of the population was over oldest old, defined as 80 years and older. Between 2000 and 2010,
60 years of age, their per capita GDP ranged between $5,000 and the number of older people in this category has doubled to over
$10,000. But China’s GDP was $806 in 1999, when 10% of its 20 million. Caregiving for those in advanced old age presents new
population first became 60 years of age and over (Mu, 2011). Even challenges, as the prevalence of disability, frailty, chronic diseases,
though China’s per capita GDP has risen sharply in recent years, and ageing-related dementia and Alzheimer’s disease also increases
it still remains a low-income country. In 2007, China’s GDP per dramatically. To this day, China does not have any social policies
capita was $5,046; in comparison, that of South Korea was $23,364, or programmes for long-term care for a growing older population.
Japan was $31,607, and the US was $43,227 (Jackson et al., 2009: The Chinese government, exalting China’s cultural tradition of
3). As a low-income country with a huge rural population,3 China family care for the aged, holds to the position that the family has
has neither enough resources nor a strong pension system in place the primary responsibility to take care of older members, and the
to deal with an ageing population. Another challenge is due to the family responsibility is codified in the Law of the People’s Republic
compressed timeline of China’s demographic transition. While it of China on Protection of the Rights and Interests of the Elderly, first
took many countries in Europe and North America 50–100 years circulated in 1996.
for its population to transform from relatively young to old, it took However, whether or not Chinese families can continue to shoul-
China only 18–26 years (Kinsella, 2009; Mu, 2011). Consequently, der the burden of caring for their older members in the context of
China has not had enough time to develop comprehensive health- a sharp decline in fertility and an increase in longevity is now in
care and service programmes to deal with its rapidly ageing popu- serious doubt. Mr Shi is very fortunate, because in his generation
lation. Moreover, China’s rapid population ageing is taking place there was no restriction on family size, and his five daughters were
in the context of accelerated modernization, urbanization, and willing to sacrifice their own careers and family life to provide care
economic transformation. One result is that China’s traditional for him. His own daughters will not be so lucky when they in turn
family-based support for care of older people has been severely become old and need long-term care, because they are the genera-
weakened. Another is that adequate government-supported care tion of parents who grew up under China’s one-child policy and
systems are lacking. most have only one child to count on for care.
Three topics—the lack of long-term care, the decline of fam- Recently, China Daily, China’s official English-language news-
ily support, and the deepening rural–urban divide—illustrate the paper, published a rare, candid article entitled ‘No Country for
daunting task for China and other developing countries to cope Alzheimer’s Patients’ (Li, 2011),4 to illustrate how China lags
with a rapidly ageing society in the context of dramatic social and behind in meeting the population’s increasing need for long-term
economic transformations. care for older people. According to this article, 4.8% of Chinese
people 65 and older are diagnosed with Alzheimer’s disease,
The crisis of long-term care need and it is estimated that this number increases by a million each
On 26 October 2009, the Chinese state media, CCTV, aired a spe- year. Currently, China has an estimated 10 million Alzheimer’s
cial programme titled ‘Today’s Elderly, Tomorrow’s Us’, highlight- patients. And yet, Dr Tian, who is director of China’s premier
ing the problem of providing long-term care for incapacitated older Neurology Centre at Peking University of Chinese Medicine,
people. The programme began with the story of how five daugh- declares:
ters of a Shi family in Tianjin scrambled and struggled to care for
there is not a single department in any major Chinese hospital that
their sick and bedridden father. The father was 86 years old and treats Alzheimer’s exclusively. There are more than 300 Alzheimer’s
had suffered from mental decline and heart problems since 2003. disease centres in the United States that provide diagnosis, treatment,
Initially his wife and one daughter were his main caregivers. His rehabilitation, and daily nursing, China has none.
situation worsened over the years, and he needed more skilled and (Li, 2011)
3 The 2010 census indicates that China’s urban population is 665.57 million or 49.68%, while the rural population is 674.15 million or 50.32%.
4 This frank report is quite unusual. The Chinese state media usually reports positive news about China’s achievements with economic development.
Most nursing homes do not accept Alzheimer’s patients because Rise of empty-nest households and the decline of
they are not equipped to care for them. Those who are equipped family support
are often concerned about the high risk of lawsuits, because
Until relatively recently, in most countries, extended families and
Alzheimer’s patients are prone to harm themselves or other resi-
coresident adult children have been the primary means of support
dents in the facilities. Among the few nursing homes that accept
for the older generation. In China, intergenerational coresidence
Alzheimer’s patients, the cost is often prohibitively high and out of
and the cultural tradition of filial support for parental old age has
the reach of ordinary families. For example, in Beijing the average
played a major role in provision of care for older people. However,
monthly pension was 2,032 yuan ($322.5) in 2010, but the formal
population ageing is taking place at the same time as the country is
care cost for an Alzheimer’s patient can be as high as 8,000 yuan
undergoing rapid social and economic change, accelerated urbani-
($1,270) a month. As a result, most Alzheimer’s patients get no pro-
zation, and changes in family structure and gender roles. Such devel-
fessional care and are confined or ‘locked up’ at home, while others
opments have weakened traditional family-based support and have
end up in mental hospitals. The situation is so grave that the deputy
contributed to a steady rise in ‘empty-nest households’ (kongchao
secretary-general of China’s Ageing Development Foundation, Liu
jiating) in which the older generation lives alone, separated from
Hongchen, is reported to have said, ‘In China’s battle against ageing,
their adult children. According to the latest government data, the
the scariest part is not our enormous and rapidly increasing aged
number of single-generation families composed of old people has
population. It is how we deal with Alzheimer’s and other dementia
risen over 50% in both rural and urban China, and in some cities
diseases’ (Li, 2011).
the proportion is over 70% (Li, 2010). The rate of such empty-nest
Even though Chinese families feel obliged to take care of their
families is expected to climb further and more steeply when parents
old, many are not prepared or are unable to provide long-term
whose reproductive lives were limited due to the one-child policy
care for their chronically ill parents. An ancient Chinese saying
become old in the next two decades.
succinctly captures the intense dilemma between cultural expecta-
In industrial countries, a decision to live alone is often ‘a reflec-
tions and the difficult reality: ‘Even filial children cannot take care
tion of an economic demand for privacy or autonomy’, and is
of chronically sick parents forever.’ Clearly, China’s increased life
found to be positively correlated with income level—an increase
expectancy has posted huge challenges to a cultural tradition that
in income level is followed by an increased propensity to live alone
prides itself on filial care for the older generation. Until now the
(Michael et al., 1980; Becker, 1981). Modernization theories have
Chinese government has done very little to assist family caregivers
also shown that urbanization and industrialization alter the tradi-
or develop alternative care services to provide long-term care.
tional extended family living arrangements by eroding the authority
It was not until 2010 that the China National Committee on
of older persons and increasing the desire of children to live inde-
Ageing (CNCA) and the China Research Centre on Ageing (CRCA)
pendently (Goode, 1963; Levy 1966). In China, the recent increase
conducted the first nationwide survey on the extent and situation
in empty-nest families can be understood as an indication of a link
of Chinese who can no longer care for themselves. The objective
between rising prosperity and economic self-sufficiency, and it leads
of the survey was to lay down the groundwork for drafting China’s
to older people living without their adult children. For example, a
national polices for long-term care. In March 2011, the CNCA/
survey on living arrangements in Wuhan in the mid-1980s revealed
CRCA released the survey findings: China currently has a stagger-
that parents with state pensions were four times more likely to live
ing 33 million older people who have completely or partially lost
alone than were parents without pensions (Unger, 1993). A 2004
the ability to live independently and who are in need of long-term
survey of older individuals in Nanjing showed that the majority
care. In connection with providing care for these needy individu-
(63.3 %) chose to live in single-generation households, not only
als, 98% are cared for by family members at home, while only 2%
because they could afford to live separately because they received
receive care in nursing homes (CNCA/CRCA, 2011). The same
pensions and had savings, but also because they preferred ‘sepa-
report reveals that fewer than 60% of nursing homes in China are
rate living’ since it gave them ‘more freedom’. Moreover, about a
equipped with hospital facilities, less than half have access to doc-
third said they did not want to ‘burden their children’ (Zhang et al.,
tors, and less than 30% have trained nurses available.
2006). In her study of care of older people in Guangzhou in south
In this CNCA/CRCA report, the Chinese government for the
China, Charlotte Ikels (2004) notes that a housing construction
first time acknowledged that providing long-term care for depend-
boom that started in the late 1980s greatly contributed to the rise
ent older people is ‘not only the pressing need for the elderly them-
of empty-nest households in that city when adult children moved
selves and their families, but also a societal issue that needs to be
out of crowded, shared accommodations in a span of 10 years from
addressed urgently’. It calls for the setting up of long-term care
1987–1998.
insurance and an old-age security system in the years to come. The
The increase in empty-nest families often compromises the care
government also promises to provide more training for professional
of older people, gives rise to loneliness, and in some cases even leads
care, to facilitate the development of community and home-care
to parental abandonment and neglect. In China, a culture of filial
services, and to give tax incentives to build more private care homes
piety in a family system that emphasizes multigenerational coresi-
suitable for old people. However, the report continues to emphasize
dence has traditionally been the basis for support in old age; the
the role of the family as the primary support for its older members.
cultural ideal of having a fulfilling and happy old age was defined
It is argued that because ‘China is still a developing country with
as ‘sandai tongtang’ (three-generations living under one roof). Yet
few resources, it is impossible to expect the government to meet
despite drastic social and economic changes brought about, first by
the need for long-term care at the present time’. While government
Mao’s socialist revolution (1949–1978) and then by China’s current
recognition of the need for long-term care for China’s ageing popu-
market reform (1978 to the present), family obligation to provide
lation is welcome, it has come too late for the many older people
parental support remains robust, and coresidence is still considered
already in need of care.
the culturally preferred way in which parental support is main- in contemporary China. Today it is clear that young family mem-
tained and provided (Whyte, 2003; Silversteinet al., 2006; Zimmer bers often have to choose between work and parental care. Lack of
and Korinek, 2010). However, the rise of empty-nest families since resources and the need to succeed in a competitive global economy
the 1980s has meant that family-based parental support is weaken- have distracted the Chinese government, as in other emerging coun-
ing, and the need for care of older people is not being adequately tries, from developing a viable social safety net for the older popu-
met. For instance, Ikels’ longitudinal study in Guangzhou shows lation. Under these circumstances, encouraging the continuance of
that coresidence was high in 1987, prior to the housing boom, and the traditional form of family support for older people becomes an
the proportion of caregiving delivered by family members was expedient way for the state to deal with the ageing population.
86.6%. But in 1998, this proportion dropped to 48.1% because adult
children had moved out. None of the older people in Ikels’ survey A tale of two ageing worlds: the deepening
sample reported lack of care in 1987, but 7.8% did so in 1998 (Ikels, rural–urban divide in China
2004: 347–8). China has a huge rural population and uneven economic develop-
Even though Chinese parents may accept and adapt to the mod- ment within the country—a common situation in many developing
ern trend of separate living arrangements, they still hope that their countries that face rapid population ageing and concurrent changes
children will pay regular visits and provide emotional support for associated with modernization and urbanization. Despite China’s
them. When children do not visit regularly, empty-nest parents are impressive economic growth and prosperity in the past three dec-
likely to feel lonely, abandoned, and forgotten by their children. ades, resources continue to heavily favour and flow to cities, and
According to a news story carried in Beijing’s Metropolis Daily on economic development in rural China lags behind. Consequently,
25 November 2010, an 83-year-old man living in Beijing sued his the wealth and health gaps along the urban–rural divide have wid-
six children for failing to show emotional care for him. This father ened significantly. That deepening divide in the wake of China’s
was economically independent as he had a pension and access market reform has huge implications for the differential lot of older
to medical care. But he was seeking a court intervention so that people in rural and urban China. This inequality both stems from
his six children would be obliged to take turns to visit him once a and marks a major departure from the previous Mao era.
week (Zhang, 2010). Another local news story in the city of Dalian After Mao’s socialist revolution in 1949, China pursued a collec-
reported an empty-nest couple’s ‘innovative’ approach to secure tive economy aiming at redistribution of wealth and elimination of
more visits from their adult children. The couple enjoyed a pension income inequality and private property. The Chinese government
of 4,000 yuan ($635), but had to ‘bribe’ their three children, two sons also implemented a household registration system (hukou) that
and a daughter, for home visits. The couple offered their children divided the population into ‘non-agricultural’ (urban) and ‘agricul-
contracts that paid 1,000 yuan ($159) to those who visited at least tural’ (rural) and made it virtually impossible to change one’s hukou
twice a month, and bonuses would be added if they brought their in the name of a planned economy. Urban workers were mostly
grandchildren along (Fishman, 2010). A recent study on depression employed in large state enterprises/institutions (almost 80%) or in
among older Chinese individuals shows that the number suffering small-scale neighbourhood cooperatives (about 15%). They were
from feelings of depression is ‘30% higher than it was in the 1990s, provided with work-unit housing, medical care, a monthly wage,
when family ties were stronger and the country was less urbanized’ job security, and pensions upon retirement. In rural China, farm-
(Li, 2011). Receiving love and support from the family is often an ers were organized into production teams under the collectives
effective way of dealing with depression, but for many older people and communes. Production quotas were determined by the state,
such support is often hard to come by when many are living apart and no farmers or production teams could sell crops for profit.
from their children in ‘empty nests’ (Li, 2011). Although farmers’ income fluctuated from year to year depending
Recently, media reports about older people dying unnoticed in on the harvest and the quota set by the state, they would receive
their apartments have become frequent and an increase in lawsuits state subsidies or be waived from the production quota if natural
concerning parental neglect has been documented, with the result disasters hit that resulted in poor harvests. The collectives and com-
that the Chinese government passed a legal amendment in March munes also provided subsidized medical care and schools, giving
2011 to the 1996 Elder Protection Law on parental visits. Under all farming families access to basic healthcare and education.
this legal amendment, aptly referred to as the ‘chang huijia kankan Throughout the Mao era, little urbanization took place, and the
cao’an’ (visit-home-regularly amendment), adult children must visit urban and rural population maintained a steady 20/80 percentage
their parents ‘regularly’, but the amendment does not specify what rate from 1949–1978 due to the strict hukou system. Even though
constitutes ‘regularly’. For older people who feel ignored it is now urban older people enjoyed better fringe benefits and healthcare
possible to go to court to claim their legal rights to be physically than their rural counterparts, there was not a sharp divide between
and mentally looked after by their children. This new legislation them due to at least three factors (Zhang, 2009). First, as a result
has ignited an intense public debate about the wellbeing of older of enforcing the socialist egalitarian principle and a redistribution
people in China and their right to parental support. While many of wealth in the Mao era, the standard of living was low across the
welcome the amendment, and think of it as encouraging the whole country and did not distinguish sharply between the urban and the
society to pay more attention to and care for older people, others rural population, or between the rich and the poor. Second, even
argue that job-related migration and work pressures of the modern though older people in rural areas did not have state pensions, they
world as well as a competitive market economy have made it dif- had access to subsidized healthcare and a basic livelihood under
ficult for adult children to keep close contact with and provide care the collective economy. Third, the hukou system that prohibited
for their parents. It is unclear whether the law is enforceable, or how rural-to-urban migration in the Mao era also meant that most adult
effective it will be in helping older people to cope with the reality children lived in the same village if not the same housing compound
of empty-nest living and their diminished care by family members with their parents, and could provide care when needed.
However, China’s post-Mao market reform, development model, family support for the old, and left many older individuals in rural
and the hukou legacy have all contributed to substantially widen settings to fend for themselves and frequently to take care of their
the gap in the general wellbeing between older people living in grandchildren (Liu and Guo, 2007). A recent comparative study on
rural and urban environments. On the one hand, city-living older rural and urban mortality rates indicates that the mortality rate for
people have benefited from the urban boom and the prosperity Chinese rural old people is 30% higher than their urban counter-
resultant from China’s market reform. Owing to the legacy of the parts due to the postreform increased health disparities in rural set-
socialist era, most urban retirees receive pensions and have access tings (Zimmer et al., 2007).
to medical care and subsidized housing, giving them a certain level
of economic security and independence in old age. According to a Signs of change? From family-based care to social care
recent survey by Shanghai Municipal Statistics Bureau, ‘more than for older people
97% of seniors who are 60 or older said their pensions are their In postindustrialized countries where advanced healthcare and
main sources of income. … Only 14.7% received support from their social security systems are in place, governments have existing
adult children’ (Shanghai, 2011). The official retirement age is 60 for structures to cope with an ageing society. But China is rapidly age-
men, 55 for female professionals and government officials, and 50 ing while still a developing country. For the first two decades of
for rank-and-file female workers. This relative early retirement age China’s market reform from 1978 to the early 2000s, the govern-
has enabled many urban retirees to have more leisure time to pur- ment emphasized economic growth, production efficiency, and the
sue other interests such as dancing, singing, and doing exercises in attraction of foreign investment from multinational companies,
the parks, or playing ping pong, practising calligraphy, and learning thereby unleashing an abundant cheap rural labour force to power
arts and crafts in community centres (Zhang, 2009). China’s export industry. That strategy made China the ‘World
China’s market reform and the emerging private sector, together Factory’. This rush to embrace the market economy and join the
with a new focus on developing a consumer-oriented society, have world economy was accompanied by withdrawal of the state as the
spurred the rise of such new industries as fee-based household provider of social welfare and services. China’s market reform is
services and institutional care. Even though urban older people a stunning success, in that the country has witnessed double-digit
are increasingly living in empty-nest households, they now have GDP growth over the past three decades, but, as noted above, the
more choices to cope with the decline of family care, because they transition to a market economy has brought about dramatic social
can purchase care by living in a residential care facility or hiring a changes that have severely weakened family care for the older gen-
live-in or hourly helper (Zhang, 2006). The huge influx of young eration and fuelled worries about the future of care for older people
rural migrants to the cities has also kept nonfamilial care services in China. These changes are, above all, of concern in rural areas,
for older people relatively affordable for urban retirees, although where the deepening income gap along an urban–rural divide,
these migrants are rarely trained to work in nursing homes and/or the worsening plight of rural elderly people following economic
are reluctant to provide care for patients with dementia. reforms, and the decline in the number of children available for
In contrast, the wellbeing of older people in rural areas has parental support as a result of China’s governmentally orchestrated
become much more insecure and vulnerable since China embarked birth-control policy are most apparent. Married couples who have
on market reform in 1978. The abandonment of the collectives has been allowed to have only one child will start to become old en
meant the collapse of subsidized rural medical cooperatives and masse as of 2015.
schooling, resulting in farmers and their families having to pay Clearly, China has a daunting task ahead. However, it has certain
out of pocket for the increase in privatized fee-for-services. When advantages in coping with population ageing: the cultural tradition
Zhang conducted her ethnographic fieldwork in a rural village in of filial support and family care is still alive, even if weakened; China
the mid-1990s, she witnessed frequent instances of families caught has a strong central government that can implement new policies
between either paying medical bills for sick parents or spending the and redirect investment effectively into care for older people should
money on their children’s education. Because old people who are it so decide; and increased government revenue as a result of rapid
rural residents have no pensions, they have only their adult chil- economic growth has enabled the state to revisit its role in building
dren to turn to for old-age support. But the free-wheeling market a safety net for the poor and the older generation. Somewhat belat-
economy and intense competition have often forced adult sons to edly, the Chinese government has taken new steps toward develop-
put their conjugal family’s interest first, frequently at the expense ing a ‘social eldercare service’ system characterized as ‘at-home care
of doing their expected share for parental old-age support (Zhang, forming the basis, supplemented by community day-care centres
2004). A national survey conducted in 2000 showed that many peo- and supported by institutional care’.
ple living in rural areas continue to farm well past the age of 60 Beginning in 2003 the government began to address the widening
(67.6% of 60- to 69-year-olds and 19% aged 70 and over) (Pang income gap between rural and urban residents by adopting a series
et al., 2004). of prorural policies including reinstating government-subsidized
Moreover, with China’s urban-centred development model and rural medical cooperatives; removal of school fees for the 9-year
new and better job opportunities created in the coastal cities, tens compulsory education; abolishing the agricultural tax and levies;
of millions of rural young and able-bodied men and women from and, in 2009, piloting rural pension programmes, with the goal of
poor interior provinces have left their home villages for urban covering all rural older people by 2020. It appears that these new
employment. According to the 2010 census, the number of rural-to- polices have begun to work because the widening income ratio
urban migrants was over 220 million. Due to the hukou legacy, rural between urban and rural residents has started to narrow, from
migrants can only work in cities as cheap labour, and are excluded 3.33:1 in 2009 to 3.13:1 in 2011 (Xinhua News Agency, 2011). The
from enjoying urban residence rights and social services. Thus, the new rural medical cooperatives and the rural pension programme
mass exodus of the rural young has caused an accelerated decline of are a major step forward in aiding rural families with old-age
support, and in helping rural older people to obtain a reasonably the newest policy, particularly popular among older people, was the
secure later life. implementation of the rural pension system in 2009. Villagers 60
Mr Wei’s story provides cautious optimism about the future of years old and over now receive a monthly pension of 55 yuan ($8.7)
ageing in China, and the positive role that the state might well play from the government, still a meagre sum in comparison to the aver-
in facilitating care for older people in a rapidly transforming society age pensions of 1,000–3,000 yuan ($158.7–476.19) for urban reti-
by drawing on certain aspects of the cultural tradition. rees. But for tens of millions of rural-living older people who have
When Zhang met Mr Wei in his village in central China in 1993– never received any government pension before, this pension plan
1994, he was 72 years old and full of bitterness about young people’s gives them both relief and hope for the future.
‘unfilial’ behaviours. He chose to live alone even though he had two
adult sons living in the same village because he wanted more control Vignette 3—United States (Sharon
over his daily routine. In a typical family in rural China, the parents
who live with their married son’s family do not have savings of their Kaufman)
own, and it is the married son and his wife who control the family The third vignette, about the US, shifts attention to the use of sys-
budget. Thus, living alone gives the senior parents more freedom to tematic medical interventions into ageing bringing about life exten-
control their own budget, especially if they can raise chickens and sion, with dramatic effects on the experience of growing old and
sell them in the market, for example. If they live with their married significant implications for caregiving.
children, the money from chickens and other animals usually goes The populations of the industrial and postindustrial nations are
directly to the ‘common budget’ and is not in their control. ageing, and trends in healthcare delivery to older individuals are
In 1994 Mr Wei confided to Zhang that he had accumulated both a source and consequence of that demographic development.
an entire bottle of sleeping pills and was ready to end his life if he The US is unusual because clinical intervention plays a dominant
became too sick or his relationship with his sons turned sour. When role in the extension of older lives and, importantly, in how US
Zhang saw him again in 2002 he was still in despair about the dete- citizens think about old age and about health in later life (Gillick,
riorating life of rural-living older people and told her that there had 2007; Callahan, 2009). While surgery, drugs, and devices have
been three cases of suicide among the older generation in the vil- changed the actual life-expectancy for many older people around
lage since her previous visit in 2000. Mr Wei now suffered from the globe, an explosion in the varieties of preventive, life-saving,
prostatitis and said that he was simply waiting for his death, and and life-extending interventions for older individuals in the US is
that he did not seek treatment for fear of burdening his sons with changing citizen expectations about longevity and mortality and
mounting medical bills. changing the actual practices of many medical specialties (Kaufman
When Zhang visited the village again in 2010, she was surprised et al., 2004). From the medical management of cholesterol, blood
to find Mr Wei, at 89, with a sharp mind and a relatively happy and pressure, and heart disease to cardiac stents, valve replacement,
relaxed mood that she had not seen before. He told her that he was and bypass surgery; from aggressive cancer therapies to drugs for
still alive due to the actions of his two sons and the existence of the memory enhancement, depression, and sexual dysfunction; from
new rural medical cooperative. Mr Wei had been about to end his heroic life-extending procedures in hospital intensive care units to
life in 2005, he reported to her, when his prostatitis worsened and hospice care at home, older persons and their families have more
he felt that he had lived long enough. But he did not want to commit options than ever before among standard, alternative, and experi-
suicide in the village as it might give his sons a bad reputation, who, mental treatments. They have more responsibility about whether to
by village standards, were quite filial and never failed to provide extend life in the face of chronic disease and frailty. They are offered
food and assistance for him. Moreover, his younger son was the choices about the style and timing of death. While those choices
Communist party secretary of the village at that time, and Mr Wei’s are, for the most part, wanted, they also bring with them anxie-
suicide could potentially jeopardize his son’s political career. So Mr ties and ethical dilemmas about how much medical intervention to
Wei told his two sons that he had left some cash under his bed and employ as one grows older.
that he now wanted to go on a long journey, and they should not go There is a growing medical literature on the justification and ben-
looking for him. Apparently, his two sons stopped him and insisted efits of performing many kinds of procedures on persons older than
taking him to the county hospital, where he underwent a success- age 80. Indeed, treatments now routine for older US citizens, such
ful operation and fully recovered. Mr Wei’s younger son later told as renal dialysis, organ transplants, cardiac implants, and aggressive
Zhang that the cost for his father’s surgery and hospital fees was cancer therapies, are changing the patterns of ‘end-stage’ disease.
about 10,000 yuan ($1,587), but 30% of the cost was reimbursed In many cases, formerly terminal diseases have become chronic
by the rural medical cooperative and the rest was equally shared illnesses that require ongoing medical management, including
among the three siblings (the two brothers and their elder sister).5 routine diagnostic tests and therapies to prevent or thwart spread
When Zhang commented to Mr Wei’s younger son that she saw of disease. Great attention paid to risk, prevention, and ongoing
a noticeable improvement in intergenerational relations in the vil- medical surveillance has become the norm in what is considered
lage, he told her that family disputes had substantially decreased optimal medical care for older people. These developments both
in the village in the past few years, once villagers had seen their contribute to and result from a growing societal expectation: doc-
incomes increase following the removal of the agricultural tax and tors and patients alike expect medical interventions to prevent,
levies in 2005, and the new rural medical cooperative made health- thwart, and cure disease at all ages, including advanced age. And
care more accessible and affordable to the villagers. He added that these developments influence family and medical responsibility in
5 The split self-pay cost, 2,300 yuan ($365), was still substantial, considering the per capita income was 4,700 yuan ($746) in the village in 2005.
ways that could not have been predicted even a decade ago. Many pointed this out more than two decades ago, showing that the shift
factors contribute to extensive and intensive geriatric intervention in meaning occurs because new technologies almost immediately
in the US, including, for example: the high societal value placed on ‘feel’ routine to practitioners and then quickly become standard of
new technologies; the fact that physicians are paid by procedure; care. ‘Once a new technology is developed,’ she noted, ‘the forces
and the federal reimbursement scheme for older adults that does favoring its adoption and continued use as a standard therapy are
not take cost into consideration. formidable’ (Koenig, 1988: 467).
The growing array of life-extending therapies, together with the In the culture of medicine today, the technological impera-
ratcheting up of the age for aggressive treatments, have intensified tive is bolstered by the value given to evidence-based studies and,
the already recalcitrant and well-known tension in the US—between mostly, by the value given to clinical trial results. Evidence-based
the desire and the ability to cure disease and extend life by any assessments of the overall risks and benefits of a drug, device, pro-
means, on the one hand, and the widespread societal cry to resist cedure, or service are the most important factor in determining
interventions that prolong dying and suffering, on the other. That reimbursement decisions. Importantly, series of clinical trials that
tension is becoming more deeply entrenched in the US because show benefit for the use of a specific intervention on younger adults
when patients and their families are faced with life-threatening dis- increase the pressure on Medicare to expand payment coverage for
ease and told by their doctors that they may benefit from certain ever-older adults—even without evidence of effectiveness for older
treatments (even if the chances are small), it is difficult to say ‘no’. persons. An assumption of benefit, which may or may not be true,
To reject therapies that have become, or are quickly becoming, the drives that extension. Medicare coverage decisions influence pri-
standard of care is to deny the authority of medical knowledge and vate insurers. Together, those insurance mechanisms authorize
medicine’s progress in curing and preventing disease. Such rejec- what are now called ‘best practices’ through the acknowledgment
tion would deny the assumption that doctors are considering what that the evidence produced in clinical trials is now scientifically
is best for this patient—an individual with particular diseases and adequate to show safety and positive outcomes. Thus Medicare
symptoms. reimbursement policy strongly influences physician, patient, and
The role of technology and the role of Medicare policy strongly family expectations for life-extending treatments. Because treat-
determine how the structure and financing of US healthcare deliv- ments for life-threatening conditions are common among older
ery makes available and promotes the medical interventions that people, Medicare policies become fundamental to how life is lived
have become so routine. Medicare is the US government pro- for a growing segment of the population.
gramme that pays for acute medical treatment for persons age 65 The heart of the matter is that new technologies (mostly created
and over, without cost restrictions. Medicare reimbursement for by the multinational for-profit drug, device, and biotechnology
new technologies (drugs, devices, and procedures) and for the industries) come to be reimbursed by Medicare, and then, covered
expanded use of existing technologies is the fundamental driver of by private payers, quickly become standard of care. ‘Standard of
the explosion in the availability and use of life-extending interven- care becomes a moral, as well as technical, obligation,’ Koenig noted
tions for ever-older US citizens. Importantly, citizens who could (1988), and it is exceptionally difficult for clinicians to decide not
not afford private health insurance when they were younger qualify to offer what quickly has become the standard of care, and then for
for government-sponsored Medicare coverage when they turn 65. patients and families to say ‘no’ to standard interventions.
And they express relief and enormous peace of mind when they Medicare policy is also the means through which the technologi-
know that their acute healthcare needs, at the very least, will be cov- cal imperative (via drugs, devices, and procedures) becomes ethi-
ered for the rest of their lives. Government reimbursement for new cal necessity. In the ageing society of the US, patients and families,
medical tools influences what physicians advise and what patients like health providers, understand today’s technologies as ethically
and families come to need. necessary and think of them, as well, within the parameters of
Because it has come to include more diseases and treatments under risk awareness and reduction that characterize so much activity
its reimbursement umbrella, Medicare insurance has enabled older in healthcare delivery and in contemporary life. For everyone, as
people to live longer and, as a result, to develop other problems that the risks associated with different therapies diminish—and cardiac
require more intervention later—some of which Medicare also pays surgeries and organ transplantation are good examples of that—the
for. Yet the problematic ramifications of medicine’s life-extending social and medical perception of risk shifts to the risks of death, and
capabilities are highly visible—in the ‘epidemics’ of Alzheimer’s dis- doing everything possible to reduce those risks, even, in the US,
ease and heart failure that exist because medical interventions have regardless of age and disease state.
enabled people to live long enough to suffer those consequences; in Two recent examples of therapies that have shifted from ‘unthink-
the growing and now largely untenable waiting lists for organs; in able’ even a decade ago to routine and standard treatment for older
the profound responsibilities families face for keeping patients alive persons in the US today are liver transplantation for primary liver
and well after complex surgeries; in the ongoing care requirements cancer and the expanding use of the implantable cardiac devices.
of the frail, dependent, and chronically ill older people, often for Need for any individual patient (and family) emerges in dialogue
years; in the growth of the assisted-living, life-care, long-term care, with health professionals, but it is established, first, by what becomes
and memory-care housing and support industries; and in the crisis standard, reimbursable treatment. One cannot need a therapy that
facing the solvency of the government Medicare programme itself. has not been proven effective. Need is felt strongly if the interven-
tion is considered standard of care. Need, of course, affects patients’
Role of technology and Medicare policy and families’ lives. Two case studies from Kaufman’s ethnographic
For clinicians, the unavoidable ‘technological imperative’ in medi- research in the US illustrate the emergence of need, first in the case
cine, first described by health economist Victor Fuchs (1975), of liver transplantation, and then in the case of the implantable
becomes, also, a moral imperative. Anthropologist Barbara Koenig cardiac devices. They show that need, standard-of-care medicine,
clinical appropriateness, and ethical necessity have become inex- expectation. It is only ‘thinkable’ because clinical evidence paved
tricable. Those entanglements drive and give shape to medical the way for Medicare coverage of liver transplants, which can cure
responsibility in geriatric care and to patient and family delibera- lethal disease and extend life. The survival statistics are compelling.
tion about treatment. The surgeon’s evidence, encouraging the patient and family to con-
sider living 5, 10, or 15 years longer without liver disease, inspired
Organ transplantation the daughter’s question and positioned the family to consider an
Over time, Medicare policy has broadened the eligibility criteria open-ended future for Mrs Dang, as though that potentially ‘added
for liver transplantation (as it has for other treatments) so that, time’ would automatically result from treatment.
for persons aged 65 and beyond, previously fatal liver diseases are At the second clinic visit 6 months later, the liver specialist pre-
now objects of treatment. By 2001, studies showed that outcomes sented the family with the numerical evidence: 90% of patients sur-
for patients with cancer that originates in the liver (hepatocellular vive the first year, and most live at least 3 years. And he said, ‘I think
carcinoma) improved with transplantation. Medicare coverage for she would benefit from a liver transplant’. The family walked out of
eligible patients began that year. Transplantation for liver cancer the clinic extremely ambivalent.
has grown steadily ever since, and there is no doubt that it saves Two months after that, Mrs Dang had turned 73. In the clinic
lives. Liver transplants are performed in 127 US medical centres. waiting room, the daughters explained that the idea of a transplant
They are the second most common organ transplant operation was a huge dilemma for them. They worried that, at age 73, their
(after kidneys) in the US. The percentage of liver transplants for mother would suffer complications and become frailer and that
older persons is rising. More than 6,300 (6,320) liver transplants the surgery would not prolong her life but rather shorten it. They
were performed in 2009, 11% of them on adults over 65 years of were ambivalent because age mattered to them. Was it worth it at
age (according to the Organ Procurement and Transplantation her age? Their worries were part of the emotional work and ethical
Network, 2011). Older patients Kaufman met in different centres responsibility that have been transferred, in the US, to families as
in the US who were candidates for transplant or who had received they respond to the prospect of this and other life-extending inter-
transplants may be considered the leading edge of these numbers ventions, as they respond to the technological imperative, the risk
because many liver diseases that begin earlier in life, such as hepa- of death, and the value placed on clinical evidence. Families do not
titis C, take years to become end stage, and so it is older adults who often discuss those worries with their physicians because the lure of
‘naturally’ come to need a transplant to survive. In an ageing soci- the evidence for life extension is so powerful. The ethical respon-
ety, more older persons will come to need liver transplants in the sibilities that rest on their shoulders thus often remain invisible to
years ahead. clinicians.
Transplantation becomes ethically necessary to avoid death. The One of Mrs Dang’s daughters also pondered out loud a
individual decision-making that takes place ‘downstream’, in medi- now-frequently debated question: ‘If you have cancer and decide
cal clinics and among patients and their family members, is already not to treat it, is that suicide? I don’t think so, but I wonder. If I think
prefigured by the confluence of clinical evidence, Medicare cover- my mother shouldn’t be listed for transplant, is that murder?’ Those
age, standard technology use, and, therefore, the creation of need. reflections—in which families feel a huge burden of guilt and com-
The determinative connection between Medicare cost coverage and plicity, as though they could be ‘killing’ or ‘saving’ a loved one—are
standard of care is the critical variable because standard of care common in the US today. Those reflections are a downstream effect
means best, appropriate practice. of technological innovation and its legitimacy, first by Medicare
reimbursement, and then by what becomes standard, ethical prac-
The story of Mrs Dang: the logic of transplantation tice at ever-older ages. Three doctors had by now advised Mrs Dang
Mrs Dang’s story is illustrative (Kaufman with Fjord, 2011). Her to have the transplant and the daughters were inclined to follow
daughters brought her, at age 72, to the liver clinic at a major medi- that advice. Mrs Dang said she didn’t know what she would do, but
cal centre liver clinic because her chronic liver disease was becom- she was not completely opposed to a transplant.
ing more advanced, and her local doctor suspected cancer. In three A few minutes later in the examination room, the doctor said, ‘I
clinic visits, over an 8-month period, the patient and family moved feel strongly that a transplant is the best chance to save her life. The
from not wanting a liver transplant, to ambivalence, and finally to odds are that she’s not going to live very long without it. She has an
acceptance of it—in order for Mrs Dang to live. 80– 90% chance of making it through the first year. She may have
At the first clinic visit, the daughters asked: Will a transplant a little more trouble because she is older.’ Mrs Dang replied at that
extend her lifespan or shorten it? Will it make her life worse? How moment, ‘I’ve made up my mind just now. It’s okay. I’ll do it to live.’
would it complicate their own lives, if she didn’t do well? What if one The doctor asked the daughters if they agree, and they said ‘yes’. Thus
of them was a donor and had complications? The surgeon guided Mrs Dang moved toward liver transplant because standard research
the family to think about the future when he said that he thought and clinical pathways, expanded payment criteria, and professional
Mrs Dang would be ‘in good enough’ shape to withstand the stress and familial obligation all led toward that outcome. The doctors
of transplant surgery. He urged the family to make a decision about were guided by clinical evidence and Medicare guidelines. For the
moving forward. He said, ‘I think she would have a tough year, family, saying ‘no’ to transplant would not be rational or ethical in a
getting the transplant, and then she could live 9–15 years with no system in which treatment can, most likely, stave off death.
problems.’ Mrs Dang did not want a transplant. Regarding the uses of ever-more technology on greater num-
Walking out of the clinic building, one of Mrs Dang’s daughters bers of older persons, age matters in relation to cost, allocation,
said, ‘I need to ask my mother if she wants to live ten more years’. and scarcity because there are more older people than ever before
This kind of statement has become ordinary. This kind of thinking and because more of them seek, demand, and agree to potentially
reflects a new relationship among time, technology, and life-course life-extending, high-tech medical treatments into very old age. Yet,
in tension with the pressures of cost and allocation brought about these devices go to persons over age 90. Yet, in treating a poten-
by an ageing population, published evidence in medicine and val- tially lethal arrhythmia, the ICD prevents sudden death, the kind of
ues in US society show that transplant and other procedures suc- death many say they actually want in late life.
cessfully prolong the lives of older persons ( Lipshutz et al., 2007; Although some US physicians ponder the ethics and practical
Chan et al., 2009). In those studies, advanced age per se does not appropriateness of implanting this device in patients in their late 80s
indicate ineffectiveness of the therapy. Geriatricians and other spe- and 90s, several electrophysiologists echoed the statement of one
cialists know this well. who reported, ‘I don’t even blink when I have a patient that comes
To complicate matters, in the clinic, where doctors, patients, and in who is in the late 80s, because that has become the standard. I’d
families seek to prolong one individual’s precious life, actual out- say the number I think twice about is 90 or above. But we have many
comes for any one person cannot be predicted. Many have com- patients over the age of 90 now.’ And from another, ‘Now we’ve come
mented on the fact that clinical trials mostly exclude the very old to realize that you can put an ICD in someone who’s never had an
and underrepresent those between the ages of 70 and 80. Yet it is event at all, without doing any other testing, but just bring them in
clinical trial results that pave the way for Medicare coverage, pri- from the office and put it in. Because at some point, they may face
vate insurance coverage, and then what becomes standard practice. this arrhythmia risk and, scientifically, they’ll be better if they have
In addition, what is effective in clinical studies is, always, a mov- this than someone who doesn’t have it. We’ve all grown to accept
ing target. So, while age per se may be no indicator of successful that. So I think I’ve changed in terms of my thinking about what’s
outcome for any one patient, the use of scarce, limited, and costly treatable or when it should be treated’ (Shim et al., 2008).
medical resources on an expanding population of older individu- For practitioners and patients alike, the trend towards more
als complicates discussions of appropriate therapy and the goals of sophisticated interventions at older ages influences deliberations
medicine in an ageing society. about whether to treat. The use of one cardiac treatment along a
The enormous caregiving and emotional burdens on families, continuum makes additional procedures with the newest devices
their work necessary to help extend older lives, is not considered conceivable and appropriate (Shim et al., 2008). Older patients and
part of the scientific evidence base that determines so much about their families then must ponder an individual ethic of life exten-
life-prolonging treatments today. Yet families have enormous respon- sion. For patients, it often goes like this: Given my current age, that
sibility—for the decision to pursue life-extending therapies and for is, how long I have already lived, how much longer do I want to try
caring for postsurgical patients. The full extent of the responsibilities to live, given the options that medicine provides? The story of Mr
of families in an ageing society is invisible to healthcare policymak- Albert (Kaufman et al., 2011) illustrates this treatment trend and
ers and sometimes to clinicians as well, and so discussion of those common patient response. It is a story that takes place every day in
responsibilities has been largely absent from the sociopolitical con- clinics across the US.
versations about the multiple costs and burdens of high-technology
healthcare delivery to the oldest citizens in the US. The story of Mr Albert: insuring risk reduction;
treating ageing
Cardiac devices The cardiologist at a major medical clinic greeted Mr and Mrs
The growing normalization of cardiac treatments for the oldest Albert and said, ‘I have spoken with your local doctor. I want to
citizens is made possible by the decreasing risks of the procedures talk to you about a defibrillator and a pacemaker. The question is
themselves. As devices such as automatic implantable cardioverter whether you might benefit from an ICD with or without pacing of
defibrillators (AICDs or ICDs) become smaller, as techniques for the heart all the time. The defibrillator is a special pacemaker that
implanting them become safer, and as less invasive procedures are has the ability to shock the heart in a rhythm that would lead to
being used with greater frequency and success, physicians and the death. It can be thought of as an insurance policy to prevent that
public have learned to view them as standard interventions that one kind of arrhythmia. It’s important to think about the defibrillator as
does not easily refuse. In the US, reduced risks produce a sense that an insurance policy. Do we want to insure the cost—for something
life extension is open-ended as long as one treats risk. That is the we may not need? It’s hard to predict which individuals will actually
prevailing logic that drives so much treatment (Shim et al., 2008). benefit from the device.’
Hundreds of thousands of Medicare recipients fit clinical trial ‘Really,’ he continued, ‘that’s all the defibrillator is. It’s not going to
criteria for the implantable cardioverter defibrillator. The device make you feel better. In fact, sometimes, it gives ‘inappropriate’ shocks,
was used sparingly up to 2002–2003 for those who had already when it doesn’t need to. Over a 5-year period, 5–10% of patients will
survived a potentially lethal heart rhythm but were at high risk for experience that kind of shock. It’s extremely painful. Like a kick in
another life-threatening cardiac event. In the past few years, use of the chest. Also, there’s risk of infection. And you’ve heard about the
the device has risen substantially for two reasons. First, following a recalls, the faulty devices. So, it’s that type of decision.’
series of clinical trials, Medicare committees in 2005 expanded the The doctor then offered an additional procedure because there
eligibility criteria to include primary prevention for those who have is newer technology. The newer, resynchronizer pacer (CRT) could
never suffered a potentially fatal rhythm disturbance (Redberg, improve the symptoms of Mr Albert’s advancing heart failure. The
2007; Tung and Swerdlow, 2009). Second, the ICD is used rou- doctor continued, ‘If we decide to do the ICD, should we do a more
tinely now along with the cardiac resynchronizer (CRT) in more extensive procedure at the same time? Putting in an extra lead in
sophisticated multifunction devices. In 2008, more than 340,000 the heart, to better synchronize the two chambers, to treat the heart
Americans received an ICD, up from 34,000 in 2000 (Grant, 2010). failure. It is a more complex procedure. We have to inject dye in the
Currently, one-fifth of ICD and CRT devices are implanted in per- heart, go into a small vein. This pacer, the cardiac resynchronizer,
sons over 80 in the US (Swindle et al., 2010). Kaufman learned from is designed to make you feel better. The problem is, we don’t know
several medical centres in 2008–2009 that approximately 10% of who will feel better. About two-thirds of patients will feel better;
but one-third won’t. So, you could undergo the surgery, and not feel Conclusion
better.’ Though he clearly invoked what some US clinicians refer to
as the ‘technology parade’, he did not paint an unduly rosy picture. Until the late twentieth century, anthropological literature on age-
Mr Albert and his wife asked common questions: Is it worth it ing devoted considerable space to discussion of gerontocracies and
when you’re in your 80s? What would you do? And of course it was the benefits that such a societal arrangement gave to older people.
impossible for the doctor to answer definitively. After more discus- One of the most striking features about a gerontocracy, of course,
sion, the doctor summarized the rather complex decision tree the is the power invested in older people. The reason for this invest-
patient now faced. ment of power, above all, is that the memories of the older genera-
He said, ‘There are two possibilities. First, the defibrillator—you tion are valued, and not merely their long experience. In nonliterate
do qualify for it. You are eligible.’ Kaufman heard this exact language societies, old people are living libraries—repositories of the past
repeatedly, and it is important. The physician is referring to the fact and upholders of the moral order, as it was and always should be.
that the patient’s medical condition fits both the clinical trial evi- Of course, there are also accounts of aged individuals who were
dence for a good outcome and the Medicare reimbursement crite- abused and abandoned, even when in theory respect was their due
ria developed from the clinical trials data. To the patient, however, until death. There is evidence from the remnants of the oral tradi-
this language sounds as though he has won something in a lottery. tions of preliterate societies as to how revered singers of tales used
‘Second,’ the doctor noted, ‘we could go for the ICD and the resyn- mnemonic devices to assist them as an aide memoire in chanting
chronizer, in hopes of making you feel better in terms of symptoms. the epic histories of their societies. Such performances frequently
But this is an unknown. And if we do that, then we have to have a lasted for hours on end without interruption; today these singers,
plan—to stop if it’s too complex, if the vein is blocked.’ often blind, are sometimes asked to perform publicly at conferences
He concluded, ‘Considering your risk, it would be appropriate devoted to ancient and medieval history.
to buy the insurance. It’s not black and white. I’m not the one who But times have changed: with modernity we look less and less to
is paying the premium, having to live with infections, shocks, et the past for guidance and increasingly to futures enabled by tech-
cetera. It’s up to you. I do think it might benefit you, that’s why we nologies, over which the young have the greatest mastery. Libraries
are offering it.’ Mr Albert’s reply was a common one, based on the full of books are no longer valued to the extent that was formerly
clinical expectation that the symptoms of heart failure in later life the case, let alone the memories of older people. The old are now a
can be reduced, and on the societal expectation that the signs of burden to society, and the approaching pandemic of ageing will be,
ageing and approaching death can be pushed further away by medi- we are assured, a drain on the global economy. The above vignettes
cal technique. He replied, ‘I’m wearing out. Things are degenerat- make clear how three societies, struggling to cope with bourgeon-
ing, deteriorating. That’s why I’m here. I think I should have it.’ He ing populations of older people, attempt to balance humane care
gave his consent, and the doctor scheduled the procedure. with finite resources, bringing about mixed social consequences for
Medicine has always pushed the boundaries of what is possi- individuals and their families
ble. What is different today in industrialized and postindustrial-
ized societies, markedly so in the US, is the ready availability of
References
life-extending technologies, increasing life-expectancy, higher Barker, J. (2009). Between humans and ghosts: the decrepit elderly in a
expectations about good health, and greater longevity, and the Polynesian society. In: Sokolovsky, J. (ed.) The cultural context of aging:
worldwide perspectives, 3rd edition, pp. 606–22. Praeger, Westport.
new kinds of clinical and emotional burdens that the technological
Becker, G.S. (1981). A treatise on the family. Harvard University Press,
imperative, standards, and ethical necessity foster. For clinicians, Cambridge.
those burdens include weighing the evidence against the ‘technol- Callahan, D. (2009). Taming the beloved beast: how medical technology
ogy parade’, that is, the overuse or inappropriate use of medical costs are destroying our health care system. Princeton University Press,
interventions. For patients, those burdens include experiencing the Princeton.
need to pursue treatments to stay alive, sometimes for the sake of CCTV (2009). Today’s elderly, tomorrow’s us (News 1+1 article).<http://
their families. For families, the burden is living with questions that news.sina.com.cn/c/sd/2009–10–26/234218912411.shtml> (accessed
are becoming so common—should I encourage her to have this 26.10.2009).
treatment? What does it mean if I do not? Am I a good enough Chan, P.S., et al. (2009). Impact of age and medical comorbidity on the
effectiveness of implantable cardioverter-defibrillators for primary
spouse or child if I do not advocate for aggressive intervention?
prevention. Circulation: Cardiovascular Quality and Outcomes, 2, 16–24.
Clinicians need to be aware that these questions are now inescap-
China Daily (2011). Number of ‘empty-nest families’ increasing in city. <http://
able, though they are not often articulated. www.china.org.cn/china/2011–08/19/content_23242107.htm> (accessed
In the US, clinicians are aware that some treatments, especially for 19.08.2011).
the very old, can be a double-edged endeavour, yet they want—and CNCA/CRCA (2011). Quanguo Chengxiang Shineng Laonianren
citizens want them—to provide potentially life-extending thera- Zhuangkuang Yanjiu Xinwen Fabugao (Press release on the National
peutic options. Older persons, many of whom are ambivalent about Survey on Infirm Elderly in Urban and Rural China). <http://www.
undergoing aggressive treatments, mostly do not want to authorize cncaprc.gov.cn/info/13085.html> (accessed 01.03.2011).
their own deaths by rejecting a potentially life-prolonging therapy. Cohen, L. (1998). No aging in India: Alzheimer’s, the Bad family, and other
modern things. University of California Press, Berkeley.
Finally, families do not want the responsibility of saying ‘no’ to
Economist. (2011). China’s Population: The most surprising demographic
life-extending therapies for their loved ones and, of course, they crisis. Available at: http://www.economist.com/node/18651512 (accessed
hope that treatments can extend meaningful life. Thus, the science, 20.01.2012).
Medicare policy which expands payment coverage for new technol- Fishman, T.C. (2010). Shock of Gray. Scribner, New York.
ogies without cost limits, doctors, patients, and families, and their Fuchs, V.R. (1975). Who shall live? Health, economics, and social choice. Basic
responsibilities, all shape the making of ‘old age’ today. Books, New York.
Gillick, M.R. (2007). The technological imperative and the battle for the Michael, R.T., Fuchs, V.R., and Scott, S.R. (1980). Changes in the propensity
hearts of America. Perspectives in Biology and Medicine, 50, 276–94. to live alone: 1950–1976. Demography, 17, 39–53.
Goode, W. J. (1963). World revolution and family patterns. Free Press, New Mu, G.Z. (2011). Senior citizens need better treatment (China Daily op-ed
York. article, Debate: Ageing Population). <http://www.chinadaily.com.cn/
Grant, M. (2010). Ethical and attitudinal considerations for critical care opinion/2011–01/10/content_11815790.htm> (accessed 29.12.2011).
nurses regarding deactivation of implantable cardioverter-defibrillators. Organ Procurement and Transplantation Network (2011). The Organ
AACN Advanced Critical Care, 21, 222–6. Procurement and Transplantation Network Data. <http://optn.transplant.
He, B.L. (2011). Aging population a challenge (China Daily op-ed hrsa.gov/data/>.
article). <http://www.chinadaily.com.cn/opinion/2011–05/19/ Pang, L.H., de Brauw, A., and Rozelle, S. (2004). ‘Working till you drop’: the
content_12537334.htm> (accessed 20.01.2012). elderly of rural China. The China Journal, 52, 73–94.
He, D. (2011). Nursing homes close doors to many (China Daily, Redberg, R.F. (2007). Disparities in use of implantable
Business: Infographic article). <http://europe.chinadaily.com.cn/ cardioverter-defibrillators: moving beyond process measures to outcomes
business/2011–03/02/content_12104612.htm> (accessed 27.01.2012). data. Journal of the American Medical Association, 298, 1564–6.
Ikels, C. (1997). Long-term care and the disabled elderly in urban China. Shanghai (2011). Number of ‘empty-nest’ increasing in city. <http://www.
In: Sokolovsky, J. (ed.) (2009). The cultural context of aging: worldwide china.org.cn/china/2011–08/19/content_23242107.htm>.
perspectives, 2nd edition, pp. 452–71. Bergin and Garvey, Westport. Shim, J.K., Russ, A.J., and Kaufman, S.R. (2008). Late-life cardiac interventions
Ikels, C. (2001). The experience of dementia in China. Culture, Medicine and and the treatment imperative. PLoS Medicine, 5, e7.
Psychiatry, 22, 257–83. Silverstein, M., Cong, Z., and Li, S. (2006). Intergenerational transfers and living
Ikels, C. (2004). The impact of housing policy on China’s urban elderly. arrangements of older people in rural China: consequences for psychological
Journal of Urban Anthropology, 33, 321–55. well-being Journal of Gerontology: Social Sciences, 61, S256–66.
Jackson, R., Nakashima, K., and Howe, N. (2009). China’s long march to Sokolovsky, J. (ed.) (2009). The cultural context of aging: worldwide perspectives,
retirement reform: the graying of the Middle Kingdom revisited (Centre for 3rd edition, pp.13–29. Praeger, Westport.
Strategic and International Studies report). <http://csis.org/files/media/ Swindle, J.P., et al. (2010). Implantable cardiac device procedures in older
csis/pubs/090422_gaI_chinareport_En.pdf> (accessed 26.01.2012). patients: use and in-hospital outcomes. Archives of Internal Medicine,
Kaufman, S.R. with Fjord, L. (2011). Medicare, ethics, and reflexive longevity: 170, 631–7.
governing time and treatment in an aging society. Medical Anthropology The Economist (2011). China’s population: the most surprising demographic
Quarterly, 25, 209–31. crisis . <http://www.economist.com/node/18651512> (accessed
Kaufman, S.R., Shim, J.K., and Russ, A.J. (2004). Revisiting the 20.01.2012).
biomedicalization of aging: clinical trends and ethical challenges. The Thupayagale-Tshweneagae, G. (2008). Psychosocial effects experienced
Gerontologist, 44, 731–8. by grandmothers as primary caregivers in rural Botswana. Journal of
Kaufman, S.R., et al. (2011). Ironic technology: old age and the implantable Psychiatric and Mental Health Nursing, 15, 351–6.
cardioverter defibrillator in US health care. Social Science Medicine, 72, Traphagan, J. W. (2000). Taming oblivion: aging bodies and the fear of senility
6–14. in Japan. State University of New York Press, Albany.
Kertzer, D. and Laslett, P. (eds) (1995). Aging in the past: demography, society Tung, R. and Swerdlow, C.D. (2009). Refining patient selection for primary
and old age. University of California Press, Berkeley. prevention implantable cardioverter-defibrillator therapy: reeling in a
Kinsella, K. (2009). Global perspectives on the demography of aging. In: net cast to widely. Circulation, 120, 825–7.
Koenig, B. (1988). The technological imperatives in medical practice. In: Unger, J. (1993). Urban families in the eighties: an analysis of Chinese surveys.
Lock, M.M. and Gordon, D. (eds) Biomedicine examined, pp. 465–96. In: Davis, D. and Harrell, S. (eds) (1993). Chinese families in post-Mao
Kluwer Academic, Boston. era. University of California Press, Berkeley.
Lamb, S. (2000). White saris and sweet mangoes: aging, gender and body in United Nations, Department of Economic and Social Affairs, Population
north India. University of California Press, Berkeley. Division (2009). World population prospects: the 2008 revision. Highlights.
Levy, M.J. (1966). Modernization and the structure of societies. Princeton Working Paper No. ESA/P/WP.210. United Nations, New York.
University Press, Princeton. United Nations Development Programme (2011a). Poverty reduction
Li, L. (2011). No country for Alzheimer’s patients (China Daily: USA, China: (Botswana) . <http://www.unbotswana.org.bw/undp/poverty.html>
Society article). <http://www.chinadaily.com.cn/usa/china/2011–03/22/ (accessed 17.12.2011).
content_12206776.htm> (accessed 22.03.2011). United Nations Development Programme (2011b). Country profile:
Li, L.G. (2010). To prioritize policies and speed up the social eldercare system international human development indicators (Botswana). <http://hdrstats.
suitable for an aging population (Ministry of Civil Affairs of the People’s undp.org/en/countries/profiles/BWA.html> (accessed 07.012012).
Republic of China article). <http://mzzt.mca.gov.cn/article/yltjh/ Van der Geest, S. (1997). Money and respect: the changing value of old age in
ldjh/201011/20101100112983.shtml> (accessed 29.12.2011). rural Ghana. Africa, 67, 534–59.
Lipshutz, G.S., et al. (2007). Outcome of liver transplantation in Velkoff, V.A. and Kowal, P.R. (2006). Aging in Sub-Saharan Africa: the
septuagenarians: a single-center experience. Archives of Surgery, 142, changing demography of the region. In: Cohen, B. and Menken, J. (eds)
775–81; discussion 781–4. Aging in Sub-Saharan Africa: recommendations for furthering research.
Liu, L.J. and Guo, Q. (2007). Loneliness and health-related quality of life Panel on Policy Research and Data Needs to Meet the Challenge of Aging
for the empty nest elderly in the rural area of a mountainous county in in Africa. Committee on Population, Division of Behavioral and Social
China. Quality Life Research, 16(8), 1275–80. Sciences and Education. National Research Council. National Academies
Livingston, J. (2003). Reconfiguring old age: elderly women and concerns Press, Washington.
over care in southeastern Botswana. Medical Anthropology, 22, 205–31. Whyte, M. K. (2003). The persistence of family obligations in Baoding. In:
Livingston, J. (2005). Debility and the moral imagination in Botswana. Indiana Whyte, M.K. (ed.) China’s revolutions and intergenerational relations.
University Press, Bloomington. University of Michigan Press, Ann Arbor.
Livingston, J. (2009). Suicide, risk, and investment in the heart of the African World Bank (2008). Botswana: country brief. <http://go.worldbank.
miracle. Cultural Anthropology, 24, 652–80. org/8P9IVY6270> (accessed 31.03.2009).
Li, Xinzhu (2011). Depressed emotions on rise in older people (China Daily, World Health Organization (2011). Health situation analysis in the African
China: Society article). <http://usa.chinadaily.com.cn/china/2011– region: atlas of health statistics, 2011. WHO Regional Office for Africa,
05/25/content_12574175.htm> (accessed 25.05.2005). Brazzaville.
Xinhua News Agency (2011). Income gap between China’s urban, rural Zhang, H. (2010). An 83 year old father sues his six children, his request—
residents narrows in 2011. <http://news.xinhuanet.com/english/ coming to visit him regularly (Xin Jing Bao, Beijing News article). <http://
china/2012–01/20/c_131371091.htm> (accessed 20.01.2012). society.people.com.cn/GB/13309064.html> (accessed 06.012012).
Zhang, H. (2004). ‘Living alone’ and the rural elderly: strategy and agency Zhang, H., et al. (2006). Jiangqiang yu wunai: Nanjingshi duju laoren
in post-Mao rural China. In: Filial piety: practice and discourse in shengcun zhuangkuang yu fuwu xuqiu diaocha baogao (Perseverance
contemporary East Asian countries (ed. C. Ikels). Stanford University and compromise: research report on the living conditions of the elderly
Press, California. who live alone in Nanjing). In: Yingdui renkou loaling hua (Coping with
Zhang, H. (2006). Family care or residential care? The moral and practical population aging). Social Science Document Press, Beijing.
dilemmas facing the elderly in urban China. Asian Anthropology, 5, Zimmer, Z. and Korinek, K.. (2010). Shifting coresidence near the end of
57–83. life: comparing decedents and survivors of a follow-up study in China.
Zhang, H. (2009). The new realities of aging in contemporary China: coping Demography, 47, 537–54.
with the decline of family care. In: (ed.). J. Sokolovsky. Cultural Context Zimmer, Z., Kaneda, T., and Spess, L. (2007). An examination of urban versus
of Aging: World-Wide Perspective. Greenwood Publishing Group, rural mortality in China using community and individual data. Journal
Westport, CT. of Gerontology, 62, 349–57.
CHAPTER 5
Epidemiology of old age
psychiatry: an overview of
concepts and main studies
Thais Minett, Blossom Stephan,
and Carol Brayne
transition’, the older population is growing proportionally faster

Part 1 than the other segments. Indeed, the number of older people has
Epidemiology and Neuropsychiatric tripled over the last 50 years and will more than triple again over
the next 50 years.
Disorders in Old Age According to the United Nations, in 1950 there were 205 mil-
This chapter has two parts, the first where we discuss some of the lion people aged 60 or over worldwide. Fifty-nine years later, this
major concepts and methodological issues in the epidemiology of number increased to 737 million. In 2050, it is projected that there
mental disorders in old age, and the second containing a summary will be nearly 2 billion people aged 60+ (Department of Economic
of many of the most important studies in this field. and Social Affairs—Population Division, 2010). This demographic
Epidemiology is the foundation of public health and rational transition is a global phenomenon. It was first experienced by
planning of services. In the field of old age psychiatry, the informa- high-income countries (HIC), though it has recently become appar-
tion provided by epidemiological research has been highly influen- ent in many of the low- and medium-income countries (LAMIC)
tial. This chapter reviews the methods used and the more important (Department of Economic and Social Affairs—Population Division,
population-based and combined studies that inform current think- 2010). Although the proportion of older aged individuals living in
ing. However, this work is a dynamic process, as cohorts now and HIC is higher, most of the global older population have been liv-
in the future may differ in relation to cultural, psychosocial, and ing in LAMIC. The projection for 2050 is that six countries will
medical factors, which need to be understood as we move forward have more than 10 million people aged 80 years or over: China (101
in time. million), India (43 million), US (32 million), Japan (16 million),
In this chapter we will present some of the world demographic Brazil (14 million), and Indonesia (12 million). Together they will
data emphasizing the changes in the age distribution witnessed in account for 55% of all those 80 or over in the world (Department of
the last century. Following this, some of the basic epidemiologi- Economic and Social Affairs—Population Division, 2010).
cal concepts will be covered. Diving further into the fields of epi- To describe the demographic transition, four phases have been
demiology and neuropsychology, we will raise the burden of the identified. In the first phase, fertility was high and populations grew
neuropsychiatric disorders under the new demographic scenario, slowly even in the face of a high level of mortality. Periodic epi-
methodological issues in calculating the epidemiological measures, demics of plague, cholera, typhoid, and other infectious diseases
and the diagnostic methods of identifying such neuropsychiatric would in one or two years almost wipe out the gains made over dec-
conditions. We include an historical perspective, the evolution of ades. Because overall both rates were in balance, population growth
the neuropathological findings of the dementias, as well as some of was typically very slow in stage one. Many human populations are
the most influential epidemiological longitudinal studies and com- believed to have had this balance until the late eighteenth century,
bined studies that focus on old age neuropsychological conditions. when this balance ended in Western Europe.
The second stage begins when epidemics were better controlled
Demographic transition and improvements to food supply and sanitation started to emerge.
The world population is ageing, with extraordinary reductions in As a consequence, death rates dropped quickly, increasing lifespan.
mortality and fertility rates. As a consequence of this ‘demographic In Europe, the decline in death rates started in the late eighteenth
century and carried on for approximately 100 years. As fertility

Box 5.1 Basic epidemiological concepts
remained high at first, there was an excess of births over deaths
which led to population growth. ◆ Prevalence:
In the third phase, the imbalance between births over deaths was
Number of cases in a population at a given period of time
reduced, as fertility rates dropped for a variety of reasons includ-
Population at the same given time
ing access to contraception, urbanization, a reduction in the value
of children’s work, greater security for the older populations, and ◆ Incidence:
an increase in parental investment in the education of children, in ◆ Cumulative incidence:
addition to other social changes. Number of new cases in a population at a given period of
In the fourth stage, mortality and fertility are in balance again, time
but at lower levels. In some countries such as Germany, Italy, and Number of individuals at risk of developing the condition at
Japan, birth rates have dropped below replacement level, leading to the same given time
a shrinking of the population. ◆ Incidence density rate:
Fertility rates are now well below the replacement level in many Number of new cases in a population at a given period of
HIC and in 31 LAMIC. The level considered to ensure the replace- time
ment of generations is about 2.1 children per woman. Total fertil- Sum of the follow-up times for each individual at risk of
ity rate in the HIC has dropped from 2.8 children per woman in developing the condition
1950–1955 to 1.6 children per woman in 2005–2010. In LAMIC the ◆ Lifetime risk:
decline has started later, in the last three decades of the twentieth ◆ Probability of someone of a given age and sex developing a
century, but has progressed faster. From 1950–1955 to 2005–2010, condition during their remaining lifespan
total fertility in the LAMIC dropped by over half from 6.0 to 2.7
◆ Disability adjusted life years (DALY):
children per woman. However, great disparities between HIC and
◆ Years of life lost due to premature mortality + years lost due
LAMIC still persist (Department of Economic and Social Affairs—
to disability for incident cases
Population Division, 2010).
Life-expectancy at birth increased globally by 21 years from
1990–1995 to 2005–2010. The average gain in life-expectancy at
birth is 24.6 years in LAMIC and 11.1 years in HIC (Department and it may be regarded as a snapshot view of the number of affected
of Economic and Social Affairs—Population Division, 2010). HIC cases in a population. Prevalence is determined by the duration of
have experienced an epidemiological transition in parallel with a disease and the quantity of new cases. It is important for develop-
their demographic transition. In the epidemiological transition, ing management and health service planning as well as eradication
pandemics of infection are replaced by degenerative, neoplastic, and programmes. As prevalence is measured in cross-sectional settings,
man-made diseases caused by the adoption of unhealthy behaviours it is used to reveal associations with other variables. However, the
such as tobacco consumption, poor dietary habits, and a decline in main disadvantage is that, as it does not account for the existence
physical activity. This transition has also resulted in a change in the of a temporal sequence of events, it cannot be used to establish
landscape of diagnosis and investigation of the modalities of dis- causation and effect of events. Furthermore, in neuropsychiatric
ease, particularly in the dementias including Alzheimer’s disease syndromes, the prevalence figure is greatly influenced by case defi-
(AD), Lewy body disease, and frontal temporal dementia, begin- nitions and the operational criteria underlying these definitions.
ning in the early 1900s. Chronic diseases have become the primary
causes of not only mortality but also morbidity. This applies for Incidence
HIC as well as for LAMIC, where this process is underway. Incidence relates to new cases and it is a measure of the risk of
With the shift in the world age demographic, the population that developing a new condition within a specified period of time (Box
will experience the greatest increase in risk of disease and disability 5.1). The incidence rate is expressed either as a cumulative inci-
will no longer be infants, but individuals aged 60 years and older. dence with the number of new cases over some period of time, or as
Nonetheless, according to the Save the Children report (2009), a density rate when the denominator is the sum of the person-time
child hunger and malnutrition are persistent problems worldwide: of the at-risk population. The latter is a more precise estimate of the
one child in three in LAMIC is malnourished, and many of those rate of occurrence of a particular disease, especially in cohort stud-
who survive will experience impaired cognitive development. Both ies where the length of time that the participants stay in the study
HIC and LAMIC have to face the challenge of increases in chronic varies. The density rate contrasts the number of new cases with the
disorders, especially those disorders that are prevalent in the older sum of the time that each person remained under observation and
population. This raises issues regarding prevention and treatment free from disease.
as behavioural and genetic risk factors are a hallmark of this era; As incidence studies are more expensive and time consuming,
preventive measurements are less effective; and treatments are most of our knowledge about the occurrence of the neuropsy-
more complex, and can be lifelong and therefore expensive. chiatric conditions is based on prevalence rather than incidence
studies.
Basic epidemiological measures Incidence can be used to determine the impact of management
Prevalence and treatment strategies, to determine risk factors and natural his-
The prevalence of a condition in a population is defined as the total tory, and to appropriately allocate future resources (for example,
number of cases over the number of individuals in a population at public health initiatives, prevention and intervention strategies,
a given time (Box 5.1). It indicates how widespread the condition is funding for research). It is measured from data obtained in cohort
CHAPTER 5 epidemiology of old age psychiatry: an overview of concepts and main studies 59
studies: therefore, the temporal sequence of events is implicit, which might be incorporated into the calculation of DALYs. Unequal
allows us to infer not only associations, but also causation. However, age-weights are an attempt to capture different social roles at differ-
estimates may be influenced by unexpected changes in the environ- ent ages. The young, and often older individuals too, depend on the
ment, the population tested (e.g. population-specific differences rest of society for physical, emotional, and financial support. Given
in risk and protective factors), and can be affected by difficulties different roles and changing levels of dependency with age, it may
in certainty of distinguishing new and old cases. The collection of be appropriate to consider valuing the time lived at a particular age
data needed to calculate prevalence and incidence can be challeng- unequally (Murray, 1994).
ing and costly. Incidence is difficult to define in neuropsychiatric
conditions, as these are syndromal diagnoses, and therefore there is The epidemiology of neuropsychiatric disorders
no sudden moment of incidence as onset is almost always gradual, Neuropsychology aims to better understand the relationship
over a period of time. between the brain and behaviour. This section explores the epide-
Lifetime risk miology of neuropsychiatric conditions in ageing populations, with
a focus on the dementias and depression, including their diagnosis,
The definition of lifetime risk of a condition is the probability that a
prevalence, incidence, and risk factors (where available). Evidence
person who is currently free of the condition will develop it at some
will be gathered from longitudinal population-based studies that
time during the remainder of their expected lifespan (Seshadri
have included a focus on ageing from across the world.
and Wolf, 2007) (Box 5.1). The lifetime risk of a disease is the risk
over a long period rather than the risk over a year, which is the The global burden of old age neuropsychiatric conditions
age-specific annual incidence of the disease (Seshadri and Wolf, Disability adjusted life years (DALY)
2007). Operationally, the lifetime risk is the conditional probability From the psychiatric point of view, neuropsychiatric disorders are
of developing a disease when a person has reached the index age estimated to contribute to 7.5% of the total DALY of the global
and is free of that disease. Its estimation enables a long-term per- population aged 60 years and older. More specifically, AD and other
spective, which is especially useful for chronic conditions such as dementias correspond to 4.1% of the total DALY among the older
dementias, where exposure to risk factors in midlife can alter the population. This figure is just behind the contribution of ischaemic
incidence of disease in later life. heart disease (14.4%), cerebrovascular diseases (11.7 %), and chronic
obstructive pulmonary disease (7.1%) (Department of Health
Disability adjusted life years (DALY)
Statistics and Informatics of World Health Organization, 2004).
The concept of DALY (Murray, 1994) was designed to measure
Table 5.1 shows the global disability burden of the neuropsychi-
the global burden of a disease. It has the advantage of aggregating
atric disorders among people over 60 years old. Unipolar depressive
mortality and morbidity in a single tool (Box 5.1). Consequently, it
disorders, AD, and other dementias together correspond to 75% of
allows the evaluation of burden of diseases with low mortality but
the burden of the neuropsychiatric disorders among the population
which are highly incapacitating. The DALY corresponds to the time
aged 60 or over.
lived with disability and the time lost due to premature mortality.
It is important to highlight that all estimates presented by the
This methodology uses epidemiological data and vital statistics,
World Health Organization are based on systematic assessments of
which are normally available even in LAMIC, facilitating interna-
the available data on incidence, prevalence, duration, and severity
tional comparisons and evaluations of the impact of international
of a wide range of conditions, which are themselves often based on
investments and health policies.
inconsistent, fragmented, and partial data from different studies.
DALY is calculated as the sum of the years of life lost due to pre-
This means that there are substantial data gaps and uncertainties
mature mortality (YLL) in the population and the years lost due to
(Department of Health Statistics and Informatics of World Health
disability (YLD) for incident cases of the health condition:
Organization, 2004).
DALY = YLL + YLD Mortality of dementia
The years of life lost (YLL) is calculated from the number of deaths Guehne et al. (2005) conducted a review study of the mortal-
multiplied by the standard life-expectancy at the age at which death ity risk in dementia and potential influencing factors, based on
occurs, which is derived from the model life-table West Level 26, population-based samples. They found that apart from the meth-
which has a life-expectancy at birth of 82.5 for women and 80.0 for odological differences between longitudinal population-based
men. The basic formula for YLL for a given cause, age, and sex is: studies, which this chapter will later cover intensively, all types of
dementia, without exception, were associated with a considerably
YLL = number of deaths × standard life-expectancy
increased mortality risk. Moreover, the risk of death rises with
at age of death in years
advancing severity of the disorder. The authors recommended the
To estimate YLD for a particular cause in a particular time use of incident dementia cases to calculate the course of dementia
period, the number of incident cases in that period is multiplied by and the mean survival time more precisely. For incident cases of
the average duration of the disease and a weight factor that reflects dementia, the time of onset of the disorder could be assumed to be
the severity of the disease on a scale from 0 (perfect health) to 1 the midpoint between baseline and follow-up interview or date of
(dead): death, provided that the follow-ups were conducted at short inter-
vals. However, only few studies have based their finding on incident
YLD = number of incident cases × disability weight × average
cases. Aguero-Torres et al. (1999) calculated a mean survival time
duration of the case until remission or death (years)
of 3.0 years among 75-year-old patients with incident dementia, in
Additionally, 3% time discounting and nonuniform age weights contrast to a mean survival time of 4.2 years for persons without
which give less weight to years lived at young and older ages dementia. Helmer et al. (2001) reported a mean survival time in
Table 5.1 Disability adjusted life years (DALYs) (thousands) in the global population aged 60 or older by sex, 2004
Cause Men Women Total
Unipolar depressive disorders 1,011 2,257 3,268
Bipolar affective disorder 4 6 10
Schizophrenia 42 67 109
Epilepsy 155 149 304
Alcohol use disorders 465 75 540
Alzheimer and other dementias 3,366 5,878 9,244
Parkinson disease 509 540 1,049
Multiple sclerosis 22 38 60
Drug use disorders 24 8 32
Post-traumatic stress disorder 5 19 24
Obsessive-compulsive disorder 25 42 67
Panic disorder 11 26 37
Insomnia (primary) 155 267 422
Migraine 0 0 0
Total neuropsychiatric disorders 6,465 10,172 16,637
Total DALYs (all causes) 109,688 112,817 222,505
(Source: Department of Health Statistics and Informatics of World Health Organization, 2004.)
incident cases of 4.5 years among 65-year-olds and a relative risk controlled for. With increasing sample sizes, higher rates of mor-
of 1.8 (95% CI = 1.8, 2.7). Xie et al. (2008) observed median sur- tality are found. For measures of depression, there are higher rates
vival time of 4.1 years (interquartile range 2.5–7.6) for men and 4.6 of mortality when depression is assessed by structured diagnostic
years (2.9–7.0) for women. Regarding the influence of education interview rather than psychiatric examination, and in turn these
and occupation, some studies have found individuals with demen- are both higher than self-report measures. Interestingly, for com-
tia and lower education having a higher survival time (Helmer et parison groups, matched control groups had higher rates than
al., 2001; Qiu et al., 2001). A shorter survival time for highly edu- cohorts, and both more than general population studies. Numerous
cated individuals with dementia could be explained by their abil- factors have been controlled for, including age, sex, physical illness,
ity to continue living an effective and independent life for longer, smoking, alcohol, and suicidal behaviour. Of the studies that were
resulting in the late recognition of the disease. It is presumed that restricted entirely to late-life population samples, 10 (67%) dis-
there is a similarly fast progression of the underlying pathological closed positive reports and 5 (33%), negative reports. Among those
processes, which means that, because of later recognition, highly positive studies, the relative risks for depression as a predictor of
educated persons may live with this condition for a shorter period mortality varied from 1.2– 4.0, with the majority of studies fall-
of time, and, therefore, present an apparently higher mortality rate ing in the 1.5–2.5 range. Depression might increase the likelihood
(Guehne et al., 2005). of dying through several factors, such as poor adherence to treat-
ment of comorbidities, poor maintenance of cognitive and physi-
Mortality of depression cal functioning capabilities, and alienation from social networks.
There is a well-established risk of suicide among individuals with However, these associations do not point to a causal factor. Schulz
depression and this risk is recognized not only among younger et al. (2002) proposed that a bidirectional model might be more
patients suffering from depression but also among older people. promising to account for this relationship.
However, depression, especially in older people, can be a conse-
Methodology
quence of medical illness and disability, and it may also influence
Prevalence and incidence are heavily influenced by study design,
morbidity and mortality through a variety of behavioural and bio-
methodological differences, population uptake, and cultural fac-
logical mediators (Schulz et al., 2002). A systematic review examin-
tors (Brayne, 1993). This has hampered worldwide comparisons.
ing the relationship between depression and nonsuicidal mortality
However, there are strong examples of standardized methods across
reported that there is evidence supporting an association between
combined studies, such as used in the 10/66 study discussed later
the two and the mechanisms that might account for this relation-
in this chapter.
ship (Schulz et al., 2002). They assessed the strength of the studies
to enter in the revision according to criteria published previously Case definition
(Wulsin et al., 1999) which involves four components: sample size, To study disease incidence, prevalence, duration, and severity, the
measure of depression, choice of comparison group, and factors definition of what is a case is important. For some conditions, such
as most infectious and parasitic diseases, this is a straightforward case-finding techniques should be based on symptoms. Moreover,
process. By isolating the aetiological agent or antibodies against the the data collected should then be further organized into syndromes,
agent, the presence of disease can be determined. In some individu- and, in a third stage, into recognized diagnostic entities (WHO
als this may manifest in clinical symptoms, while in others it may Expert Committee on Mental Health, 1960).
not. However, with neuropsychiatric disorders the diagnostic proc- Attempts have been made to deal systematically with the con-
ess is less clear since diagnosis is mainly based on a combination cepts and methods of assessment. However, depending on the
of symptoms, their quantity, and intensity. Therefore, there is more system of diagnostic classification used for case definition, an indi-
difficulty in conceptualizing what is a mental disorder and in deter- vidual can be identified as being a case according to one system, but
mining whether it is present or not. This is particularly difficult not a case on another. For most neuropsychiatric disorders, differ-
among the older population, since medical comorbidity is com- ent diagnostic classification systems include different combinations
mon and may affect symptom profiles and interfere with function- of symptoms that reflect impairments in cognitive, emotional, and
ing through, for example, polypharmacy and disease-related effects social abilities to inform the process of diagnosis. Therefore, the
on mood and other mental functions. use of different criteria jeopardizes comparisons between studies,
Cultural aspects as they may classify the participants differently (e.g. diseased vs.
not-diseased) (Erkinjuntti et al., 1997).
Cultural concepts of diseases may lead to different expectations of
Erkinjuntti et al. (1997) examined the effects of six commonly
what is considered mentally normal. World-views towards ageing
used classification schemes on the prevalence of dementia in a
vary immensely and as a consequence cultural concepts influence
population-based cohort of older people. The classification schemes
the awareness of diseases, especially mental disorders, and the uti-
used were the Third Edition of the Diagnostic and Statistical Manual
lization of services.
of Mental Disorders (DSM-III) (American Psychiatric Association,
There have been examples over time of studies examining native
1980), the DSM-III-R (American Psychiatric Association, 1987),
and migrant populations. In a study of Cree Indians living on two
the DSM-IV (American Psychiatric Association, 1994), the Ninth
reserves in Manitoba (Canada), age-adjusted prevalence of dementia
Edition of the international classification of diseases (ICD-9)
was equivalent to whites living in Winnipeg (4.2% both groups), but
(World Health Organization, 1977), the ICD-10 (World Health
AD prevalence was lower (0.5% vs. 3.5% in whites) (Hendrie et al.,
Organization, 1992), and Cambridge Mental Disorders of the
1993). In the UK, vascular dementia has been found to be more
Older Population Examination (CAMDEX) (Roth et al., 1986).
prevalent than AD in individuals of African-Caribbean origin (vs.
They found that although there was substantial overlap among the
British whites) (Richards et al., 2000; Livingston et al., 2001). Paraiso
groups of persons identified by the various systems, the frequency
et al. (2011) found that dementia in an urban area of Benin (West
of dementia varied depending on the scheme used. The frequency
Africa) was slightly more prevalent than in a rural area of Benin, but
of dementia was 3.1% using the ICD-10, 4.9% with CAMDEX, 5.0%
the rate was similar to that recorded in other cities in LAMIC.
with ICD-9, 13.7% with DSM-IV, 17.3% with DSM-III-R, 20.9%
Besides genetic factors, other features could explain different
according to the Canadian Study of Health and Aging (CSHA)
prevalence estimates for dementia between different populations.
clinical-consensus method (1994), and 29.1% with the DSM-III
First, symptoms have to be perceived as an abnormal feature.
criteria. However, while dementia prevalence varied depending on
Variation in the environment, including education and social
the diagnostic criteria used, the frequency of dementia increased
meaning of cognitive changes with ageing, all influence how soon
with increasing age for each classification method.
individuals seek medical help. Furthermore, different approaches
The ICD has existed for more than a century; it is global, multi-
to cognitive testing may direct towards a diagnosis.
disciplinary, and multilingual and is ratified by all 193 WHO
In a study of impressions of the onset and diagnosis of demen-
member countries. On the other hand, the DSM developed by the
tia among African-American, Chinese, and Latino family car-
American Psychiatric Association focuses specifically on mental
egivers in the US, minority ethnic groups were found to convey
disorders and it does not have such an international approach. As
striking crosscultural similarities in the characterization of ini-
diagnosis in psychiatry is mostly descriptive and based on a collec-
tial memory changes as normal ageing (Mahoney et al., 2005).
tion of symptoms, the ICD and DSM diagnostic classifications are
As dementia symptoms progressed, however, cultural differences
functionally equivalent in clinical settings. However, in epidemio-
emerged. Normalization of cognitive symptoms until the precipita-
logical studies where structured interviews are usually employed,
tion of one critical event appeared to be most prolonged among
such differences become more important (Andrews et al., 1999).
African-Americans; Chinese seemed to be the most concerned
Another methodological issue regarding the diagnosis of demen-
about stigmatization.
tia or its subtypes is that diagnostic criteria have changed over the
In the clinical setting, patients or carers have to perceive a symp-
years as new investigations became available. To illustrate this,
tom as an abnormal feature to seek help and for the person to be
the diagnosis of probable dementia with Lewy bodies (DLB) was
defined as a case. In contrast, in most of the epidemiological stud-
initially based simply on clinical criteria, such as the presence of
ies, a case is defined by means of structured questionnaires in which
dementia, visual hallucinations, fluctuations, and parkinsonism
responses will be balanced to check against a set of operational cri-
(McKeith et al., 1996). However, the third report of the consortium
teria; therefore, it does not rely solely on an informant’s opinion,
on DLB international workshop (McKeith et al., 2005) included
but on informant standardized information, which in essence can
dopamine transporter imaging among the diagnostic criteria. This
minimize cultural bias.
technique is a functional imaging of the dopamine transporter
Diagnostic methods and defines integrity of the nigrostriatal dopaminergic system.
The World Health Organization (WHO) suggested that the most Regarding neuropathological criteria, originally the only require-
standardizable, countable and comparable units of observation in ment for DLB was the presence of Lewy bodies somewhere in the
brain of a patient with a clinical history of dementia, whereas the symptoms of dementia. Otherwise, study validity might be jeop-
new criteria take into account the extent of Lewy body-related ardized by between-clinician variability and changes in diagnostic
pathology and AD-type pathology in assessing the degree of cer- criteria over time (Brayne et al., 2011).
tainty that the neuropathological findings explain the DLB clinical In the context of psychiatric studies, the Present State Examination
syndrome. Moreover, the use of new immunohistochemical staining was developed as a standardized instrument (Wing et al., 1967). This
techniques such as alpha-synuclein have been proposed to replace instrument was adapted by Copeland et al. (1976) to become the
the former ubiquitin immunohistochemistry, as alpha-synuclein Geriatric Mental State (GMS) Examination, which was designed to
has been shown to be a more sensitive and specific method for detect dementia, depression, and other mental illness in the older
detecting Lewy bodies. population by generating diagnostic algorithms (Copeland et al.,
Diagnostic criteria for vascular dementia and AD have also 1986) validated against the clinical diagnostic process based on the
changed according to the influence of available technology. DSM-III-R. The aim was to introduce a more structured approach
Roman et al. (1993) emphasized the importance of brain imag- to diagnosis that could be used by nonclinicians. Unlike clinicians
ing to support clinical findings for vascular dementia. For AD, as who quickly narrow down to the presenting diagnosis, structured
the amyloid-imaging positron emission tomography (PET) tracer, interviews systematically explore each diagnostic criterion before
termed Pittsburgh Compound-B (PIB) (Klunk et al., 2004), pro- assigning a diagnosis. Studies with a large number of participants
vides quantitative information on amyloid deposits in living sub- make a quasiclinical diagnosis possible by adopting this diagnostic
jects, it is likely that fairly soon this technique will be introduced algorithm based on the assumption that interviewers are trained
into diagnostic criteria for AD. to administer questions or conduct examinations in a standard-
This instrumentalization of medicine impacts on epidemiology as ized manner (Brayne et al., 2011). Recently there has been renewed
medical diagnosis, theoretically, becomes more accurate and effec- interest in the algorithm approach (Weir et al., 2011).
tive, but at the same time these technologies might label individu-
Interviews
als as diseased without them having experienced any symptoms at
all, and without knowing if they will ever present symptoms of the Data collection can be made by means of a written questionnaire,
given disease. From an epidemiological perspective, the critical fact telephone interview, and face-to-face interview with participants or
here is how these newly suggested criteria perform in samples repre- their informants, or even extracted from case records and death
sentative of the general population. It is important to highlight that certificates. Semistructured interviews are preferred when there
although dementia is considered a progressive disorder, in popula- is the need to gather qualitative data, whereas in structured inter-
tion settings, the course of the deterioration of the symptoms is not views standardization is easily obtained in detriment of qualitative
always inexorable, especially when the symptoms are mild. information. In structured interviews, there is a tightly scripted text
Classification problems also emerge when attempts to define the from which the interviewers are not supposed to deviate. Examples
transitional state between ‘normal’ ageing, pathological decline, of highly structured interviews are the CAMDEX (Roth et al., 1986)
and progression into dementia are made. Although several clas- and the Geriatric Mental State Examination/Automated Geriatric
sification systems claim to represent this intermediate stage, their Examination Computer Assisted Taxonomy (GMS/AGECAT). The
criteria differ slightly, so different outcomes are to be expected. To neurological examination in the CAMDEX is more comprehensive
illustrate this, Matthews et al. (2008) compared the 2-year outcome than the GMS. The 10/66 study, which demands that the data are
of 16 different classifications in the same population-based setting. highly standardization due to cross-country data collection, has
They found that the overall progression was highest in classifica- used two approaches for dementia diagnosis: one based on GMS/
tions in which impairment extended to memory and nonmemory AGECAT and the other based directly on DSM-IV criteria.
domains, such as multiple domain mild cognitive impairment Interviewers
(m-MCI) (14.3%), mild neurocognitive disorder (MNCD) (31%), In large-scale surveys, there is a tendency to recruit nonclinical
and mild cognitive disorder (MCD) (29%). On the other hand, the interviewers to collect the data, provided the interview is structured.
conversion rate for age-consistent memory impairment (ACMI) Naïve trained interviewers are known to be very adept at adminis-
was 0.3% and for age-related cognitive decline (ARCD) was 4%, tering structured interviews in a consistent manner. Depending on
showing that they captured a lower-risk group in the population what is included in the assessment protocol, there may still be a
with greater stability and reversion to normality. need for qualified professionals. For example, if a physical exami-
In many clinical research centres, the diagnosis of dementia is nation is intended, a doctor would normally be required (Butler
made based on a consensus approach, which relies on a multidis- and Brayne, 1998). In some multistage studies, lay interviewers are
ciplinary panel of expert clinicians who meet to review detailed used in the first stage and medical doctors in subsequent stages, as
information on various aspects of a given person, such as clinical in the ALPHA Liverpool study (Saunders et al., 1993) and in the
examination, informant reports of cognition, behaviour, functional Nakayama study (Ikeda et al., 2001). In the Soham study, a clini-
impairment, and neuropsychological diagnosis. This process allows cian administered the CAMDEX to all participants (Brayne et al.,
each study participant to be individually considered in detail (Weir 1997).
et al., 2011). However, in such cases, the diagnosis process is inevita-
bly influenced by the clinicians’ philosophy, personality, discipline, Sampling
culture, and inherent biases. Although clinical examination is of To define the sampling frame of a study, the site of the study has to
extreme importance, in the context of epidemiological research, to be defined, as well as the sample procedure. The study site might be
assure cross-study comparisons, it is crucial to standardize the proc- defined geographically or, for example, be composed of individuals
ess of data collection, such as cognitive function, activities of daily registered in a specific organization, such as a primary care trust,
living (ADL), physical health, and behavioural and psychological hospitals, healthcare system, retirement community, nursing home,
or participants of an electoral register. The sample procedure might stage, a brief and inexpensive screening instrument (such as the
include the entire population above a certain age, or use a sampling Mini-Mental State Examination (MMSE)) is administered, fol-
strategy to select a representative subgroup of manageable size, lowed by more complex, time–consuming, and expensive tests in
such as random selection or systematic selection. Furthermore, the second stage. Typically, all those people who screen positive
stratification according to age groups, sex, and other variables can (e.g. below a cut-off score on the MMSE), along with a random
be used so as not to overrepresent groups. sample of the rest, are seen again using more comprehensive diag-
Population-based studies nostic tools, which usually include a structured clinical assessment,
a more extensive multidomain test of cognitive function, and a
Few studies are truly representative of the whole population. Many
structured interview with an informant. The second-stage assess-
others would be more accurately described as population derived:
ment may also include a physical examination, brain scan, and col-
for example, many exclude individuals in institutions. There are
lection of biological samples. Although a multistage design makes
studies where a general population defined by geographical bound-
diagnostic procedures more efficient, it carries a serious problem
aries is the sampling frame (Poels and Last, 2008). It is well known
of refusal or nonavailability. Furthermore, it is more prone to
that referral of patients can cause population bias, affecting the
bias due to incorrect analysis of partially verified data, where the
results of epidemiological studies (Sackett, 1979). According to
screened negative participants who were not selected for the sec-
the Goldberg–Huxley model of the pathway from the community
ond stage are considered as true negatives or are simply excluded
to the hospital in terms of psychiatric care, there are five levels
from the analysis. As a consequence, sensitivity and specificity are
and four filters (Goldberg and Huxley, 1980). People with severe
poorly estimated (Yu and Zhou, 2012). Other studies, to overcome
illnesses pass more easily through the filters to secondary profes-
these biases, randomly select the participants for the second stage,
sional care than do people with common mental disorders.
weighting towards cognitive impaired persons to allow for infer-
Referral not only is influenced by the condition itself but also
ring the sample results back to the whole population (The Medical
may vary according to burden of symptoms, family recognition of a
Research Council Cognitive Function and Ageing Study (MRC
problem, access to healthcare, popularity of the condition, and the
CFAS), 1998). But the analyses are rarely indicated with full atten-
presence of specialized centres nearby (Brayne, 1993). Furthermore,
tion to the multiple stages and rates of dropout (Matthews, 2005).
specialized patient research groups derived from referrals typically
This overinflates power in that the reported confidence intervals are
use stringent selection criteria so that patients are usually selected
in fact too narrow.
to have fewer comorbid conditions. As an example, the occurrence
of behavioural symptoms among patients with dementia is often Screening tool: a focus on the MMSE
the triggering event for recognition and referral to healthcare rather The MMSE is one of the most widely applied tools in clinical and
than the cognitive impairment itself (Lawlor, 2002). It is therefore research practice to screen for cognitive impairment and dementia.
important to differentiate between population-based studies and However, there are disagreements regarding the best cut-off score to
community-based studies. Where the first is potentially of use for distinguish impaired versus not-impaired (e.g. 18, 21, 22, 24, or 26)
generalizing the findings to the whole population, the latter often is and whether cut-off scores should be adjusted for age and educa-
restricted to members of a selected community that does not neces- tion levels. In high-functioning samples, the MMSE is found to suf-
sarily represent the population. An example is the Nun study, where fer from ceiling effects, and in low-functioning samples, from floor
the entire community of nuns was involved in a long-term longi- effects too. Both extended, such as the extended mental state exam-
tudinal study (Snowdon et al., 1996). However, it is likely that this ination, (Huppert et al., 2005) and abbreviated versions, such as the
religious community is very different from any general population 11-item version used in the European Prospective Investigation
sample and therefore that the nature of risk exposure and outcome into Cancer (EPIC-Norfolk (Matthews et al., 2011)), have been cre-
may be different from the general population. However, depend- ated to overcome these issues. Further, due to copyright law restric-
ing on the research question, some findings might be suitable for tions, new brief and free screening tests that also measure cognitive
generalizing purposes. function, such as the Sweet-16 (Fong et al., 2011), are becoming
Another point that should be highlighted is that some studies may available and are likely to become widely used.
exclude individuals who live in institutions. Many countries do not
have the possibility of sampling from both general and institution- Influence of other variables
alized populations. This may lead to underestimation of prevalence Performance on cognitive tests may be influenced by many fac-
and, in longitudinal terms, underestimation of decline or mortal- tors other than cognitive impairment, such as educational back-
ity. Larson et al. (2004) reported findings from a survival study of ground, cultural experiences, prior testing experience, emotional
community-dwelling patients with AD, where men had a median and physical states, the testing environment, use of medicines, and
survival of 4.2 years from their initial diagnosis and women had measurement error. This makes it difficult to control and compare
a median survival of 5.7 years. These estimates were longer than such measurements in different studies even if the same tool is used
the ones of the Canadian Study of Aging (Wolfson et al., 2001), to assess cognition. Furthermore, most tests currently used are
where the median survival was 3.17 years for men and 3.36 years subject to ceiling and floor effects (The Medical Research Council
for women. These differences may well be due to the fact that the Cognitive Function and Ageing Study (MRC CFAS), 1998; Morris
Canadian Study of Aging included nursing-home residents, who et al., 1999).
were doubtless at a later stage of the disease. Missing data
Stratification Almost all studies have some missing observations. Missing data
Some epidemiological studies of dementia have used more than can arise for two main reasons: missing values due to nonresponse
one-stage diagnostic procedures to optimize resources. In the first at baseline, death, or dropout of the study, and item nonresponse.
Longitudinal and multistage studies are especially prone to be sub- Age effect: let us suppose that the prevalence of a condition was
ject to missing data due to death or dropout. The main concern is measured in a single group of people born in the same year (sin-
that in those cases, missing data are not at random and do affect gle birth cohort) in wave-1. Five years later, the prevalence of the
final estimations. Participants with cognitive impairment are more same condition was measured again (wave-2), and 10 years later
likely to drop out from a study than healthy individuals. Chatfield (wave-3) in the same original group of people. If the prevalence
et al. (2005) performed a systematic review to investigate large of this condition changes between waves, this is a representation
population-based studies of the older population that report on of an age effect.
attrition between waves of a follow-up in a systematic manner. The Cohort effect: this can be obtained when a study is interested in
review concentrated on dropout due to refusal, sickness, inability looking at a prevalence of a condition only at a given age (let us
to locate individuals, and individuals moving away from a defined say 60 years) and does so by recruiting 60-year-old individuals in
study area. They found that increasing age and cognitive impair- wave-1, a different group of 60-year-old individuals in wave-2,
ment were the two main independent factors related to increased and a third 60-year-old group in wave-3. If the prevalence of this
attrition. People who were very ill or frail had higher dropout rates, condition changes between waves, this is a representation of a
and people in worse health were less likely to be contactable on a cohort effect, as, in fact, the different birth cohorts were com-
second occasion. As these factors are not preventable, the authors pared longitudinally. The cohort effect can be seen as a result of
suggested that oversampling of these groups in the initial phases the risk factors and environmental exposures that are present in
could be used to ensure that sufficient numbers of participants early life or are typical for a given generation.
remain at follow-up. Also, the follow-up methods can be adjusted
to ensure maximum participation in those individuals with cogni- Period effects: these are the consequences of any phenomenon
tive impairment. The length of the interview can be modified and a occurring in a specific point in time, which affects the entire pop-
proxy interview can be used in a wider range of situations (Chatfield ulation, reflecting in changes in prevalence of a given condition
et al., 2005). Saxton et al. (2009) compared rates of MCI and rates across all age groups and birth cohorts (Szklo and Nieto, 2007).
of progression to dementia using different MCI diagnostic systems This is the case of wars, new curative treatments, vaccines, etc.
and, regarding dropouts, they found that MCI status at baseline was Regarding dementia, age effects are well established across stud-
significantly associated with dropouts lost to follow-up. Obviously, ies, but cohort effects are controversial. In the Lundby study, from
loss of participants also reduces the power of a study. 1947–1972 to 1972–1997, a decrease in the incidence of dementia
Missing data are even more common in retrospective studies, in was found in all age groups except in the 40–49 interval (Bogren
which routinely collected data are subsequently used for a different et al., 2007). Whether or not it is a true cohort effect or an artefact
purpose. When information is gathered from participants’ medi- is not known. On the other hand, Rocca et al. (2011) analysed data
cal records, the notes often do not point to whether or not a par- from four studies to verify if declines in age-specific prevalence and
ticipant has the aimed risk factor. It is unwise to assume that the incidence rates for dementia have occurred in recent years. Three of
answer is not present when there is no indication that the risk factor those were community-based studies: one conducted in Rochester,
was present (Altman and Bland, 2007). There no really satisfactory Minnesota (1975–1994) (Knopman et al., 2006), the Indianapolis
solution to the problem of missing data. The main ways of handling branch from the Indianapolis-Ibadan Dementia Project (1992–
missing data in analysis are omitting variables or individuals who do 2001) (Hendrie et al., 1995), the Chicago Health and Aging Project
not have complete data; or imputation, whereby missing values are (1997–2008) (Hebert et al., 2010), and a national survey, the Health
estimated from that individual’s available data (Altman and Bland, and Retirement Study (<http://hrsonline.isr.umich.edu>) (Juster
2007). Ignoring missing data in the analysis is a common approach; and Suzman, 1995). In the Minnesota cohort, there were no dif-
however, it might bias the results, as the data are rarely missing at ran- ferences in incidence across the 10 birth cohorts, but a marginal
dom in cohort studies. Although imperfect, multiple imputation is decline for dementia was observed. The Chicago Health and Aging
recommended for handling missing data. Such models not only can Project did not find a relationship between calendar year of evalua-
control for the fact that dropout has occurred, but also, where avail- tion and disease incidence, which suggests no change in incidence
able, may include information on the reason for dropout. Following over time. In the Indianapolis-Ibadan Dementia Project, although
imputation, sensitivity analysis can be run to check for similarity in they found that the 2001 cohort had higher levels of hypertension,
observed associations in the restricted (e.g. complete-case analysis) diabetes, and stroke, there were no differences in the prevalence of
compared to the imputation-derived dataset. dementia. The Health and Retirement Study did not assess demen-
tia directly, but ‘cognitive impairment consistent with dementia’
Age, period, and cohort effects (CI-D) based on a 35-point cognitive scale or proxy’s assessment of
Age, period, and cohort effects are closely interrelated, as age is the respondent’s memory for those who were not self-respondents.
the result of a given year (period) minus the year of birth (cohort). The authors reported that the prevalence of CI-D had an absolute
Szklo and Nieto (2007) proposed a definition of age, period, and decrease of 3.5% points, and a relative decrease of nearly 30.0%.
cohort effects, where age effect is the change in the rate of a con- Furthermore, the prevalence of some cardiovascular risk factors
dition according to age, irrespective of birth cohort and calendar increased significantly, but higher levels of education were found in
time; cohort effect is the change in the rate of a condition according the most recent cohort.
to year of birth, irrespective of age and calendar time; and period The ZARADEMP project (Lobo et al., 2007) investigated possible
effect is the change in the rate of a condition affecting an entire pop- cohort effects regarding the prevalence of dementia in Zaragoza,
ulation at some point in time, irrespective of age and birth cohort. Spain, for 1988 versus 1994. The authors found that there was sta-
To better understand age and cohort effects definitions, it is easier bility in the global prevalence of dementia over time, but a decrease
to approach them by fixing one parameter. among men aged between 70 and 84 years. It is interesting to note
that the prevalence of dementia in men found in the first cohort neuropathology underlying no cognitive impairment, MCI, and
was also higher than in European, pooled data reported dur- dementia from two community-based cohorts and one clinic-based
ing the EURODEM Concerted Action study (Lobo et al., 2000). cohort. They found that community-based participants with prob-
Although in ZARADEMP the two time points were probably too able AD showed less severe AD pathology and more often had inf-
close together to detect differences in environmental exposure, arcts and mixed pathologies; while those with MCI more often had
the authors pointed out that the Instituto Nacional de Estadística infarcts and mixed pathologies. Also, clinic-based individuals had
(<http://www.ine.es/>) had reported improved control of potential more Lewy bodies and atypical pathologies. Based on these results,
risk factors for dementia, such as smoking habits, diabetes, and car- the authors concluded that the spectrum of pathologies underly-
diovascular disorders in Spain. ing cognitive impairment in clinic-based cohorts differs from
Despite these methodological challenges, considerable attempts community-based cohorts (Schneider et al., 2009).
have been made to compare and combine findings from epidemi- For extrapolation of results to the population to be valid, research
ological studies that focus on the most common and debilitating must be conducted on a true population sample, or on groups with
neuropsychiatric disorders in old age psychiatry, including demen- well-characterized biases.
tia and depression. These studies are presented in this chapter. As dementia is a chronic and progressive disorder, it is impossible
to determine the exact point when an individual became demented.
The importance of neuropathological studies in the In addition to uncertainty regarding symptom onset is a lack of cer-
epidemiology of dementia and its subtypes tainty in diagnosis. In interobserver studies, the clinical diagnosis
‘Plaques’ in cerebral grey matter were first described by Blocq and of AD is not 100% correct in all cases (Holmes et al., 1999; Xuereb
Marinesco in 1892 and related to the pathology of senile dementia et al., 2000; Richards and Brayne, 2010; Scheltens and Rockwood,
by Simchowicz, who named them as ‘senile plaques’ (Blessed et al., 2011). Some classification criteria, such as the NINCDS—ADRDA,
1968). In 1907, the neuropathological findings of neurofibrillary rely on histopathological confirmation to diagnose AD as a definite
tangles in the degenerated neurons and senile plaques deposited condition (McKhann et al., 1984). As a consequence, AD is there-
in the cortex were first linked to a presenile dementia syndrome fore framed as a neuropathological entity in spite of the fact that the
(Alzheimer et al., 1995). At first, this combination of presenile diagnosis of AD in living patients is made on the basis of clinical
dementia, neurofibrillary tangles, and senile plaques, described by information (Richards and Brayne, 2010). In contrast, dementia is
Alois Alzheimer, was recognized as a rare disease separated as a dis- a syndrome rather than a neuropathological diagnosis (Xuereb et
tinct entity from senile dementia, and was baptized as Alzheimer’s al., 2000), and so a clinical diagnosis of AD assumes dominance of
disease (Berrios, 1990; Zilka and Novak, 2006). These ancient Alzheimer pathology as the cause of dementia in those patients.
reports were based on case reports and small groups of patients. The pathological criteria for AD were derived from brains origi-
It was not until 1966 that an investigation of the nature and extent nally from a highly selective clinical sample. However, the brains of
of the relationship between plaque formation and mental deteriora- people included in truly population-based neuropathological stud-
tion in old age was undertaken in a clinic-based study (Roth et al., ies of dementia have a high percentage of mixed pathologies, which
1966) conducted on patients admitted to a psychiatric hospital, challenges the pathology-led model of diagnosing definitive AD.
a geriatric hospital, and several wards in a general hospital. The The first study to disclose such a pattern in a systematic way reported
severity of dementia in individuals during life was ascertained to on autopsied brains of 101 participants (Xuereb et al., 2000). These
determine whether this bore any relationship with mean plaque findings were ratified when the sample was augmented to 213 brains
counts in cerebral grey matter. These assessments were repeated at (Brayne et al., 2009), where 22% of the brains of participants clini-
6-monthly intervals on survivors. Roth et al. concluded from the cally diagnosed as having dementia had mixed pathologies. White
results of 37 patients that, far from being irrelevant for the pathol- et al. (2002) in the Honolulu-Asia Aging study also presented
ogy of old-age mental disorder, the density of plaque formation in similar findings on 285 donated brains, where 16% of the clinically
the brain was highly correlated with quantitative measures of intel- demented decedents had mixed pathologies. The Hisayama study,
lectual functioning. When the sample was expanded to 60 patients based on 275 autopsied brains, reported an even higher proportion
the results were consistent, showing a highly significant correlation of mixed pathologies (34%) among those with clinical dementia
between mean plaque counts and scores for dementia and perform- (Noda et al., 2006). Also, the neuropathology group from the CFAS
ance in psychological tests (Blessed et al., 1968). The plaques seen cohort have emphasized the high prevalence of vascular pathology
in authenticated cases of AD were indistinguishable on light micro- in this population and the common occurrence of a mixture of both
scopy from those investigated in their study. This observation led AD and vascular pathology (Neuropathology Group, 2001).
to the realization that half or more cases of senile dementia were
associated with the same neuropathology that characterized the
early onset dementia of AD. Part 2
Significant knowledge of the clinical features and the neuropa- Review of Major Studies in Old Age
thology of the different types of dementia has come from obser-
vations on the brains of individuals with dementia from nursing Psychiatry
homes, acute medical units, hospitals, and ordinary postmortem This section summarizes some of the most important and informa-
series (Zaccai et al., 2006). However, the nature of these services tive epidemiological studies on cognitive and other mental changes
can lead to selection bias, which influences the findings of the in old age. Three main types of study are discussed: combined stud-
studies. Schneider et al. (2009) investigated the differences in neu- ies; studies that have used a synthesis of the literature with data
ropathological findings from persons with and without dementia from several sources; and longitudinal studies. There are several
in clinical versus community-based settings. They compared the difficulties in examining epidemiological studies in the context of
dementia; for example, some studies report on overall dementia study. Response proportions varied between 72% and 98% (Llibre
estimators, whereas others restrict their reports to AD, using this Rodriguez et al., 2008). Two approaches were used for dementia
term almost as a synonym of dementia, disregarding the existence diagnosis. The first was based on a regression equation developed
of mixed pathologies. in the 10/66 international pilot study that uses coefficients derived
from the GMS/AGECAT (Copeland et al., 1986), Community
Combined studies Screening Instrument for Dementia, and ten word-list learning
In this section, the following combined studies for the epidemiol- tasks (Prince et al., 2003). This approach has been validated for
ogy of dementia are presented: the EURODEM initiative and the crosscultural settings and is sensitive to educational status (a large
10/66 Dementia Research Group. proportion of older aged individuals in LAMIC have little or no
education), and regional variation, particularly associated with
The EURODEM initiative diversity in language and culture. The second approach applied
In the early 1990s, there was a collaborative initiative of all major DSM-IV criteria directly (American Psychiatric Association,
European groups working on the epidemiology of dementia, organ- 1994). For diagnosis of dementia subtypes the following criteria
ized by EURODEM, EC Concerted Action on the Epidemiology of were used: National Institute of Neurological and Communicative
Dementia (Hofman et al., 1991). A total of 20 centres took part and Disorders and Stroke and the Alzheimer’s Disease and Related
contributed original data of 23 population studies published between Disorders Association (NINCDS-ADRDA) AD criteria (McKhann
1980 and 1990, from which 12 were included in the analysis. Those et al., 1984), National Institute of Neurological Disorders and
studies that were methodologically similar and suitable for compari- Stroke and Association Internationale pour la Recherché et
son were selected to describe geographical differences and to pro- l’Enseignement en Neurosciences (NINDS-AIREN), vascular
vide an overall estimate of the prevalence of dementia in Europe. dementia criteria (Roman et al., 1993), and Lewy body dementia
The study confirmed the steep rise of dementia prevalence with age, criteria (McKeith et al., 1992).
showing that within 5-year age groups the prevalence of dementia
almost doubles from the age group 65–69 onwards. No major dif- 10/66 study: dementia
ferences were noticed regarding sex; however, the overall estimates Across 10/66 regions, dementia prevalence varied between 5.6%
were somewhat larger for men than for women until the age of 75 and 11.7% using the regression equation. When DSM-IV criteria
years and somewhat larger for women over the age of 75 years. These were applied directly, prevalence varied between 0.4% and 6.4%.
patterns were similar across the 12 studies (Hofman et al., 1991). Prevalence determined by the 10/66 equation was higher at every
Only 2% of cases were found in those less than 65 years. site, and generally double that estimated using direct application of
An incidence phase of the EURODEM initiative was conducted the DSM-IV criteria (Llibre Rodriguez et al., 2008). Comparison
and included results of analyses based on pooling the data from the of the prevalence of dementia reported in the EURODEM
studies conducted in Denmark, France, the Netherlands, and the meta-analysis with prevalence reported using the DSM-IV criteria
UK (Launer et al., 1999). In Denmark, the Odense Study (Andersen in the 10/66 sites found large variation: the prevalence in urban
et al., 1997) had a baseline cohort of 3,346 persons; in France, the Latin American sites was about four-fifths of that in Europe, the
PAQUID study (Letenneur et al., 1994) included 3,777 individu- prevalence in the Chinese sites was just over half, and that in rural
als at baseline; in the Netherlands, the Rotterdam study (Ott et al., Latin American and Indian sites only between a quarter and a fifth
1995) had a baseline cohort starting from 65 years of 5,265 persons; (Llibre Rodriguez et al., 2008).
and in the UK, the MRC-ALPHA study (Saunders et al., 1992) had 10/66 study: depression
a baseline cohort of 5,222 participants. In this pooled analysis of Depression, anxiety, and the co-occurrence of anxiety and depres-
individuals 65 years and older, 528 incident dementia patients were sive syndromes have been investigated amongst the 10/66 centres
reported and 28,768 person-years of follow-up. Incidence rates for (Prina et al., 2011). Anxiety was measured by using the GMS/
dementia were similar across studies and increased with age; at 65 AGECAT and depression was assessed according to ICD-10 and
years of age, the incidence rate for dementia was 2.5 (95% CI = 1.6, EURO-D criteria.
4.1), and at 90+ years the rate was 85.6 (95% CI = 70.4, 104.0) per The prevalence of depression according to the ICD-10 was
1,000 person-years. The study investigated potential risk factors for reasonably consistent across Latin America and India (range
dementia, and found that current smoking and low levels of educa- 4.9–13.8%) but was much lower in the Chinese site. A similar pat-
tion increased the risk of dementia significantly. A history of head tern was also found for the distribution of anxiety (range exclud-
trauma with unconsciousness, female gender, and family history of ing China: 2.3–8.9%). The prevalence of co-occurring anxiety and
dementia, did not increase risk significantly. depression ranged between 0.9% and 4.2% across sites. Having both
The 10/66 Dementia Research Group disorders was linked to higher disability scores than having anxi-
The 10/66 Dementia Research Group (<www.als.co.uk/1066/>) ety or depression alone. This has major implications for treatment
brings together researchers with an interest in ageing in LAMIC, outcome, which is found to be worse in individuals with comor-
including determination of the prevalence and incidence of demen- bid anxiety and mood disorders compared to individuals suffering
tia and noncommunicable diseases (Prince et al., 2007). from depression alone (Prina et al., 2011).
Cross-sectional comprehensive one-phase surveys have been 10/66 study: summary
conducted of all residents aged 65 and over of geographically The 10/66 Dementia Research Group focuses on the study
defined catchment areas in ten LAMIC (India, China, Nigeria, of dementia diagnosis among populations in LAMIC. Their
Cuba, Dominican Republic, Brazil, Venezuela, Mexico, Peru, population-based surveys will provide a unique resource for com-
and Argentina), with a sample size of between 1,000 and 3,000 in parative descriptive research of not only prevalence and incidence,
each site. Overall, 14,960 individuals completed the prevalence but also its effects, risk factors and costs, interventions, estimations
of need, roles of racial mixture, micronutrient deficiency, and car- of dementia had to be reported for age groups spanning 10 years or
diovascular disease. less; the study had to involve a population-based sample rather than
volunteers; incidence rates had to be reported for mild or moderate
Synthesis of literature dementia; and data needed to calculate the standard errors of inci-
In this section, the following studies for the epidemiology of dence rates had to be available or accessible from the authors. The
dementia are presented: Worldwide Prevalence and Incidence of final analysis was based on data from 23 studies and showed that
Dementia; Delphi consensus study; Meta-analysis of dementia the incidence of both dementia and AD rise exponentially up to the
incidence; World Alzheimer Report 2011. age of 90 years, with no sign of levelling off, whereas the incidence
of vascular dementia varied greatly. There was no sex difference in
Worldwide Prevalence and Incidence of Dementia dementia incidence, but women tended to have a higher incidence
The Worldwide Prevalence and Incidence of Dementia (Fratiglioni of AD in very old age and men tended to have a higher incidence
et al., 1999) paper reviews the prevalence and incidence data for of vascular dementia at younger ages. East Asian countries had a
dementia reported in the international literature in the last 10 years. lower incidence of dementia than European countries.
Results from 36 prevalence and 15 incidence studies reported from
1989–1999 found an increase in dementia with age. Worldwide World Alzheimer Report 2011
dementia prevalence was estimated to be 0.3–1.0 per 100 people The Alzheimer’s Disease International World Alzheimer Report
in individuals aged 60–64 years and 42.3–68.3 per 100 people in 2011 was based on a series of systematic reviews conducted by an
individuals aged 95 years and older. Incidence rates ranged from independent research group to collate and review all of the avail-
0.8–4.0 per 1,000 person-years in people aged 60–64 years, and able evidence relating early diagnosis of and early intervention in
increased to 49.8–135.7 per 1,000 person-years when the popula- dementia (Alzheimer’s Disease International, 2011). They reported
tion was older than 95 years. Geographical variation in prevalence the crude prevalence estimates of dementia in individuals aged
and incidence was low, with differences between countries largely 60+ in 2010 ranging from 2.7% in Africa to 6.5% in the Americas
reflecting methodological rather than real differences. Regarding (world prevalence estimate = 4.7%). They highlighted that current
the dementia subtypes, AD was always the leading type of demen- evidence suggests that available drug treatments and psychologi-
tia in all continents. However, the relative proportions attributed cal and psychosocial interventions can be effective in ameliorating
to AD and vascular dementia seemed to differ among continents symptoms for people with dementia and in reducing strain among
and multiethnic communities of western countries. North America their carers during the early stages of the disease. The report also
had the highest relative proportion of AD among all the dementing stresses that interventions for carers may be more effective in allow-
disorders (74.5%), whereas Asia had the least relative proportion ing them to continue to provide care at home, avoiding or delaying
of AD (46.5%). The prevalence of vascular dementia ranged from institutionalization of the person with dementia, when applied ear-
10.0% in North America to 38.1% in Asia. Differences in diagnostic lier in the disease.
criteria and procedures might account for these inconsistencies.
Longitudinal studies
Delphi consensus study Numerous longitudinal studies on the epidemiology of depres-
Alzheimer’s Disease International convened an international group sion and dementia, including its subtypes, have been undertaken
of experts to generate up-to-date evidence-based estimates for the in the last 50 years. We present a selection of the main, truly
prevalence of dementia for each world region (Ferri et al., 2005). population-based, longitudinal studies throughout the years to
The authors used the Delphi consensus method, which in essence illustrate methods and findings:
derives quantitative estimates through the qualitative assessment of
evidence where studies of widely different design and quality can be ◆ Lundby study
assessed. When published information is scarce, experts can make ◆ Iceland birth cohort
inferences using other data from comparable contexts. Although ◆ Reykjavik study
studies varied widely in quality, methodology, and dementia out-
come definition, the only inclusion criterion was that the study ◆ Gothenburg study
should be population-based. They found that the seven countries ◆ Cambridge City over-75s Cohort Study (CC75C)
with the largest number of people with dementia in 2001 were: ◆ Framingham study
China (5.0 million), the European Union (5.0 million), the US (2.9
million), India (1.5 million), Japan (1.1 million), Russia (1.1 mil- ◆ Established Populations for Epidemiologic Studies of the Elderly
lion), and Indonesia (1.0 million). They estimated that 24 million (EPESE)
people had dementia in 2005 and that this amount would double ◆ Gospel Oak study
every 20 years, to 42 million by 2020 and 81 million by 2040, assum-
◆ Cognitive Function and Ageing Study (CFAS)
ing no changes in mortality and no effective prevention strategies
or curative treatments. ◆ Rotterdam study
Meta-analysis ◆ Vantaa 85+
Jorm and Jolley (1998) conducted a meta-analysis of the age-specific ◆ Personnes âgées QUID (Paquid)
incidence of all dementias, including AD and vascular demen- ◆ Italian Longitudinal Study of Ageing (ILSA)
tia. The inclusion criteria were: case finding should be based on
a field survey or studies of medical and social agencies, excluding ◆ The Three-City study (3C)
purely hospital-based case register studies; age-specific incidence ◆ The English Longitudinal Study of Ageing (ELSA)
◆ Newcastle 85+ study Male and female age-standardized first-incidence rates of demen-
◆ Epidemiological Clinicopathological Studies in Europe tia every fifth year from 1947–1997 are shown in Fig. 5.1. In both
(EClipSE) men and women, a trend of decreasing dementia incidence can be
observed from 1947–1997 (Bogren et al., 2007). The data between
This cannot be comprehensive. In all of these studies we will focus waves were gathered from all available sources, such as registers
on the papers that disclosed figures for prevalence and incidence and case notes.
of dementia and depression, and comment on possible additional This finding was unexpected and may be explained by the fact
findings in those papers. that the study ran when different principles of classification were
Lundby study used, possibly affecting diagnostic conclusions, or even because
The first and the longest comprehensive prospective study of an there was a difference in study waves regarding the number of sup-
entire community with a focus on psychiatric epidemiology was plementary sources of information concerning outpatient care and
conducted by Essen-Möller in 1947 (Essen-Möller et al., 1956). This key informants, which in 1947–1972 was greater than in 1972–1997
study had a unique place in the field of epidemiology of not only (Bogren et al., 2007). An alternative explanation is that it might be
old age psychiatry but also psychiatry overall, since it involved par- a genuine cohort effect, as the incidence of dementia could have
ticipants aged 15 years and over. The aim was to observe the entire fallen due to factors such as less medical comorbidity or healthier
population of a community in the south of Sweden, notionally lifestyle.
called Lundby, to study individual traits and morbidity in a general Lundby study: depression
population, not in patients. All but 1% of the 2,550 adult inhabit- The median age at first onset of depression was reported to be
ants of Lundby aged 15 years and over were examined (Henderson around 35 years and the recurrence rate was about 40%. Transition
and Jablensky, 2010). This cohort was further re-examined in 1957, to other diagnoses was registered in 21% of the total sample, alco-
1972, and in 1997. In addition to the psychiatric interview, and hol disorders in 7%, and bipolar disorder in 2%. Five percent com-
unlike most other surveys, information was obtained from mul- mitted suicide, with increased risk of suicide associated with male
tiple sources: face-to-face interviews, informants and community sex and increased depression severity. Although figures are not spe-
nurses, general practitioners, death registers, the Swedish psychi- cific for older individuals, the Lundby study showed that after the
atric register, the national hospital inpatient register, and the local first onset of depression, 6% of men and 10% of women developed
outpatient register. In 1972, the investigators were able to obtain organic disorder or dementia (Mattisson et al., 2007).
sufficient information to reach a diagnosis on 99% of the cohort,
and in 1997 on 94%.
The Lundby strategy is a marked contrast to today’s large-scale 20
surveys. Information was obtained by psychiatrists, who were free
18
Incidence rate per 1000 years at risk
to explore the respondents’ symptoms and behaviour at interview, Dementia Men

16
rather than by lay interviewers who were required to complete a
symptom checklist and follow a tightly scripted text from which 14
they must not deviate. 12
Lundby study: dementia 10
The prevalence of all-cause dementia in both sexes in the Lundby 8
study according to age group are shown in Table 5.2 (Hofman et al., 6
1991). 4
2
Table 5.2 Prevalence of dementia (all types) in both sexes in the 0
Lundby study according to age group 20
Incidence rate per 1000 years at risk
18
Age group Prevalence (%) Number of cases Number of groups Dementia Women
studied 16
14
30–59 0.0 0 971
12
60–64 0.5 1 191
10
65–69 1.7 3 177 8
70–74 4.8 6 126 6
75–79 7.9 6 76 4
2
80–84 17.8 8 45
0
85–89 15.8 3 19 47–52 52–57 57–62 62–67 67–72 72–77 77–82 82–87 87–92 92–97
90–94 27.3 3 11 Fig. 5.1 First-incidence rates of dementia in the Lundby study.
(Adapted with permission from Bogren, M., Mattisson, C., Horstmann, V., Bhugra,
95–99 0.0 0 1 D., Munk-Jorgensen, P., and Nettelbladt, P. (2007), ‘Lundby revisited: first incidence
of mental disorders 1947–1997’, Australian and New Zealand Journal of Psychiatry,
(Source: data extracted from Hofman et al. 1991.)
41 (2), 178–86. (c) Sage, 2007.)
Lundby study: summary When examining the course of mild dementia in this cohort,
The Lundby cohort has been followed prospectively during a Magnússon and Helgason (1993) found that many cases diagnosed
period of great transition in society, which encompasses increased as mild dementia by the indirect method had no or very few cog-
welfare, urbanization, and a change of societal structures, includ- nitive symptoms when the AGECAT was applied. Almost 30% of
ing development of public healthcare and education, entrance of cases of mild dementia diagnosed by the indirect method before
women into the labour market, birth control, changes in family the age of 75 years had no symptoms of dementia at the age of
structure, lessening of the cohesive power of family, church, and 81 years and more than 10% continued to have mild symptoms.
community, and changing roles of men and women. Biological and Similar results were found at the age of 87 years.
physical factors have also undergone change, including changes Iceland birth cohort: depression
of lifestyle (diet and tobacco use) and the availability of medical The prevalence of depression according to the indirect method was
care, such as new drug therapies, as well as changes in the physical 8.7% and it was unaffected by age (Magnusson, 1989).
environment (Bogren et al., 2007). Moreover, with the increase in
life-expectancy, it is possible to distinguish two different profiles for Iceland birth cohort: summary
the older old and the recent old. The study demonstrates how prevalence can be influenced substan-
tially by information sources. While indirect methods have their
Iceland birth cohort
advantages (e.g. avoids retrieval problems in patient groups), within
The Iceland birth cohort consisted of all Icelanders born in the the context of psychiatric illness, reliability of diagnoses derived
years 1895, 1896, and 1897, and focused on the investigation of from indirect data is questionable. Indeed, depression and demen-
psychiatric diagnosis. The study was completed in four phases: the tia were in many cases not known to the family doctor (Magnússon
first spanning from 14–61 years (Helgason, 1964); the second from and Helgason, 1993).
61–75 years; the third from 75–81 years; and the concluding phase
extending from 81–87 years (Magnusson, 1989). Information on Reykjavik study (Age, Gene/Environment Susceptibility Study:
the mental health of each proband was collected from several differ- AGES-Reykjavik)
ent sources. In each phase, every general practitioner in the country The Reykjavik study started in 1967 and comprised a random sam-
was interviewed in a systematic manner by a psychiatrist, asking ple of 30,795 participants born in 1907–1935 in Reykjavik, Iceland
about mental symptoms of the probands in the cohort. Next, all (Harris et al., 2007). The study was performed in six waves—1967–
records from every hospital and nursing home in the country were 1969, 1970–1972, 1974–1979, 1979–1984, 1985–1991, and 1991–
studied and crosschecked with the data provided by the general 1996—and the study sample was divided into six groups by birth
practitioners. When it was not possible to collect sufficient infor- date within month. Each group was invited to participate in spe-
mation to make a psychiatric diagnosis, other key informants were cific waves of the study. One group attended at all waves, another at
contacted, such as relatives, local nursing staff, and neighbours. two, and the remaining only once. In 2002, 11,549 participants were
This method of data collection is referred to as the indirect method still alive and 5,764 were re-examined during 2002–2006, as a part
since the information on mental and physical health did not come of the Age, Gene/Environment Susceptibility (AGES)-Reykjavik
directly from the persons. study. In this substudy, the oldest group of the original study was
Diagnoses of psychiatric conditions were divided into three not recruited. The AGES-Reykjavik study aimed to evaluate the
groups: the dementia syndrome, affective disorders, and other common mechanisms leading to diseases in neurological, car-
mental disorders. The accuracy of the information in the initial diovascular, musculoskeletal, and metabolic systems. Participants
phases of this study depended on how well the informants knew underwent comprehensive assessments which included a question-
the proband. In the last phase of the study, however, an interview naire, clinical examination, cognitive battery, and images of the
scheme that covered the major symptoms and signs of mental dis- brain and retina (Harris et al., 2007). The response rate was 72%.
orders in the aged population was implemented to allow validation Reykjavik study: dementia
of the indirect method. Dementia case ascertainment was based on the AGES-Reykjavik
Iceland birth cohort: dementia study cohort and followed a three-step procedure (Qiu et al., 2010).
Dementia syndrome was initially diagnosed using information All participants were screened on the MMSE (Folstein et al., 1975)
acquired from probands and records. Individuals were categorized and Digit Symbol Substitution Test (DSST) (Wechsler, 1981).
according to severity as either mildly or severely demented. At later Screened positives on either of the tests were administered another
study stages, the participant was interviewed using the shortened more complete diagnostic test battery. Those who screened posi-
version of the Geriatric Mental State Schedule. This allowed psy- tively on the Trails A and B (Reitan, 1992) or the Rey Auditory
chiatric diagnoses based on the computerized program AGECAT Verbal Learning Test (Rey, 1958) went for a final assessment that
(Copeland et al., 1986). Comparison of dementia prevalence across included a proxy interview and a neurological examination. The
methods indicated a tendency of the indirect method to overdiag- diagnosis of dementia and subtypes was made by means of a con-
nose. The average age was 87 years. Indeed, 46% of participants consensus including a geriatrician, neurologist, neuropsychologist, and
sidered to be demented by the indirect method were not demented neuroradiologist. Dementia was diagnosed according to the guide-
according to AGECAT, whereas only 3% of the probands diagnosed lines of the DSM-IV (American Psychiatric Association, 1994).
as not having dementia by the indirect method were diagnosed as AD, according to the criteria of the NINCDS-ADRDA (McKhann
having dementia by AGECAT (Magnusson, 1989). The prevalence et al., 1984), and vascular dementia followed the criteria of the
of dementia according to the indirect method was 27% and accord- State of California AD Diagnostic and Treatment Centers (Chui
ing to AGECAT was 17%. Most disagreement was in the group of et al., 1992). Of the 3,906 participants, 132 (3%) were diagnosed
participants considered to have mild dementia. with dementia, including 66 with AD, 31 with vascular dementia,
and 20 with both. The group also explored whether microvascular (deceased and refusals) was obtained from medical records or other
damage, indicated by cerebral microbleeds and retinal microvas- sources. Sufficient information was thus obtained on 320 partici-
cular signs, was associated with cognitive impairment and demen- pants (92% of the population at risk). Sixty-three (20%) developed
tia. People with multiple cerebral microbleeds had lower scores on dementia during the study period. Of these, 42 cases were diag-
tests of processing speed and executive function, and this difference nosed from the neuropsychiatric examination, and 21 deceased or
was greater if there were multiple cerebral microbleeds in the deep refusals from medical records or other information. The incidence
hemispheric or infratentorial areas. All these associations were of dementia was 90/1,000 per year, within which the incidence of
independent of major cardiovascular factors, white matter hyper- AD was 36/1,000 per year and vascular dementia 39/1,000 per year
intensity, and cerebral infarcts. (Aevarsson and Skoog, 1996).
As the AGES-Reykjavik study is a single-wave substudy based on a The group also investigated the role of blood pressure on demen-
survival cohort, there are no incidence data available. Furthermore, tia at 75, 79, and 85 years (Skoog et al., 1996) in participants free
the temporal relationship of retinal and cerebrovascular lesions to from dementia at 70 years. They found that participants who
cognitive dysfunction could not be established, nor a possible pro- developed dementia at age 79–85 had higher systolic and diastolic
tective role of the cerebral microbleeds. blood pressures at 70 years of age. However, just before dementia
Reykjavik study: depression onset, blood pressure declined, and was then similar to or lower
than that of participants without dementia. Although the sample
Data on depressive symptoms were collected during the
was representative of survivors at age 79 years, only 11 participants
AGES-Reykjavik study and included the Geriatric Depression
had dementia and, at age 85 years, 18, which makes it difficult to
Scale (GDS) (Yesavage, 1988), depression history, and medications.
extrapolate these findings. The group further investigated this rela-
However, reports of these findings are yet to be published.
tionship in a different but larger cohort and published similar find-
Reykjavik study: summary ings (Joas et al., 2012).
This large population-based cohort has made several contributions
to the understanding of risk factors for myocardial infarcts and Gothenburg study: depression
cancers (Harris et al., 2007). Data on genetic and other new risk The contribution of this cohort in terms of depression was based on
factors and their relationship with more specific themes for old age an investigation of the relation between depression and the 3-year
psychiatry have been collected and publication is awaited. incidence of first-ever stroke. The diagnosis of depression was made
according to DSM-III-R (American Psychiatric Association, 1987)
Gothenburg study criteria and included the categories major depression, dysthymia,
In 1986–1987, all 85-year-old people born between 1 July 1901 and and depression not otherwise specified. The diagnoses were based
30 June 1902 in Gothenburg, Sweden, and registered for census on symptoms during the month preceding the examination and
purposes in Gothenburg were invited to take part in a health sur- observed symptoms during the psychiatric examination. Among
vey. This study was conducted as part of a series of longitudinal all 85-year-olds, 93 had a history of stroke and 19% were diagnosed
gerontological population studies (Rinder et al., 1975). Participants with depression. Depression at baseline (hazard ratio (HR) = 2.7,
from both the community and institutions were invited. Some of 95% CI = 1.5, 4.7) was related to increased incidence of first-ever
the later publications from this group disclose amalgamated results stroke during follow-up. Depression increased stroke risk among
from this cohort and the study of women in Gothenburg, 1968– not only participants without dementia but also those with demen-
1969 (Bengtsson et al., 1973). tia. However, stroke history at age 85 years (baseline) was not asso-
Gothenburg study: dementia ciated with clinical depression.
A systematic subsample of 826 individuals underwent psycho- Gothenburg study: summary
logical and psychiatric examination comprising questions about This study emphasizes the magnitude of dementia in the very old,
background factors, ratings psychiatric symptoms and signs, rat- showing that almost 10% of persons between the ages of 85 and 88
ings of signs common in dementia, and tests of mental functioning. develop dementia each year. This cohort showed a high prevalence
Response rate was 63% (n = 494). After examination, an inter- and incidence of vascular dementia with the caveat that clinically
view with a close informant was carried out (Skoog et al., 1993). defined mixed dementias were included in the vascular dementia
There were 147 cases of dementia, according to the DSM-III-R diagnosis. Depression was quoted as a predictor of stroke.
criteria (American Psychiatric Association, 1987), a prevalence
of 30%. Subjects with dementia were further investigated with CT Cambridge City over-75s cohort study (CC75C)
scans and classified into subtypes of dementia: AD, according to The first cohort study especially devoted to investigating dementia
the NINCDS-ADRDA classification (McKhann et al., 1984), was from a population perspective and determining the clinicopatho-
present in 43% of the participants, vascular dementia based on the logical correlates of dementia was the Cambridge City over-75s
criteria proposed by Erkinjuntti et al. (1988) was found in 47%, and Cohort Study (CC75C) (<http://www.cc75c.group.cam.ac.uk>).
dementia due to other causes was diagnosed in the remaining 9%. The original prevalence phase of the study is known as the Hughes
The 3-year mortality rate was 23% in subjects without dementia, Hall Project for Later Life, and the incidence survey, which was
42% in patients with AD, and 67% in patients with vascular demen- launched 2 years later, is known as the Cambridge Project for
tia. Individuals diagnosed as having mixed dementia were included Later Life (Fleming et al., 2007). The study began in 1985. A rep-
in the vascular dementia category. The population at risk comprised resentative sample of 2,609 individuals aged 75 years and over liv-
of 347 individuals was re-evaluated 3 years later (Aevarsson and ing in Cambridge city, UK, were surveyed (O’Connor et al., 1989),
Skoog, 1996). Among them, 188 (54%) took part in a neuropsychi- approximately one-third of all residents in this age range. All people
atric examination at the age of 88. Information on 132 withdrawals in this age group in six family practices were approached and one in
three from a seventh practice. Respondent rate was 95% and 40% of Results after over 200 donations had occurred have also been
the oldest old in Cambridge took part (Brayne et al., 1992), so the reported (Brayne et al., 2009). The overall burden of pathology was
sample was representative of the whole population in terms of age generally high across all participants, with most brains showing suf-
distribution, sex, and accommodation, according to the Office of ficiently extensive lesions to suggest pathological classification of
Population Censuses and Surveys (OPCS). dementia, whether or not the donor had been clinically demented
Each survey has included a detailed cognitive assessment, before death. However, participants with a clinical diagnosis of
including at least the MMSE (Folstein et al., 1975), usually its dementia did generally have a greater burden of pathology than
extended version, and, in the majority of interviews, the Cambridge those without dementia.
Cognitive Exam (CAMCOG) (Roth et al., 1986). Subsamples have CC75C: summary
had detailed psychiatric assessment using the CAMDEX, detailed
The main contribution for our understanding of this genuinely
neuropsychological assessment, informant interview, and addi-
population-based study with substantial numbers of brains was
tional tests.
the heterogeneity of lesions found in very old populations. Further,
A consensus diagnosis for dementia status at death was made
there was considerable overlap in the pathologies found in the
consistent with DSM-IV (American Psychiatric Association, 1994)
demented and nondemented, and the pathological overlap between
criteria using postmortem review of all interviews, including proxy
those diagnosed as having different subtypes of clinical dementia.
informant data, death certificates, and retrospective informant data
This indicates that when assessing dementia during life, particularly
after death, but blinded to neuropathology findings. Dementia
in the older old, a focus on a single pathology has limited utility.
was rated by severity and, where possible, subtypes were identi-
fied (Brayne et al., 2009). The study is still ongoing, representing an Framingham study
exceptional duration for a population-based cohort, although there Established in 1948, the Framingham study (Dawber et al., 1951)
are now very few survivors, all aged over 100. began as a longitudinal population-based cohort study of cardio-
CC75C: dementia vascular disease and associated risk factors, which enrolled 5,209
volunteers (55% woman). Comprehensive physical examinations
The total prevalence of all grades of dementia was 10.5%. Estimates
of the Framingham cohort have been obtained on a biennial basis.
of prevalence increased with age including: 4.1% (75–79 years),
During biennial examination 14 or 15 (January 1976 through
11.3% (80–84 years), 19.1% (85–89 years), and 32.6% (90 years or
March 1978), a dementia-free inception cohort was established.
over) (O’Connor et al., 1989). Annual incidence rates for demen-
Among 2,828 persons who were seen for physical examination,
tia also increased with age, approximately doubling every 5 years:
response rate was 75% and complete data were available for 1,085
2.3% for participants initially aged 75–79 years, 4.6% for partici-
participants aged 65 years or older, who agreed to complete a brief
pants aged 80–84 years, and 8.5% for participants aged 85–89 years
neuropsychological screening battery (Linn et al., 1995). Starting
(Paykel et al., 1994).
with examination 17 (1982/1983) and on all successive biennial
CC75C: depression examinations, an MMSE was administered. Participants falling
Using questionnaire information, diagnoses according to below age-education adjusted levels were evaluated by a neurologist
DSM-III-R criteria (American Psychiatric Association, 1987) and neuropsychologist to determine if dementia was present and, if
were made. In addition, the interviewing clinician rated each per- so, to ascertain dementia type. By using the criteria of Cummings
son for ‘severity of depressive symptoms’ on a 5-point scale: none, and Benson (1986), dementia was considered to be present by the
minimal, mild, moderate, and severe (Girling et al., 1995). The neurologist if the participant demonstrated a compromise in at
population-estimated prevalence of major depressive disorder, least three areas of mental activity, including language, memory,
based on DSM-III-R criteria, was 2.4% (95% CI = 0.9%, 4.0%) and, visuospatial skills, personality or behaviour, and cognition, and if
based on the CAMDEX criteria, was 3.0% (95% CI = 0.7%, 5.3%). there was no disturbance in consciousness. Dementia severity was
Five percent of persons diagnosed as having dementia according judged by criteria similar to those in the DSM-III-R (American
to the CAMDEX criteria also received a diagnosis of depression. Psychiatric Association, 1980). A neuropsychologist who was
Around one in five people with dementia were rated as mildly or unaware of the dementia diagnosis made by the neurologist also
moderately depressed (Girling et al., 1995). saw the participants. If a person’s performance was more than one
CC75C: neuropathology standard deviation (SD) below published age-adjusted normative
values on at least three of the seven neuropsychological tests used,
Using data from the brain donation programme, neuropathological
they were considered to have cognitive impairment consistent
analyses on tissue from a representative sample of the older popu-
with dementia. Dementia was considered present by the panel if
lation have been undertaken. The neuropathological protocol was
the following three criteria were met: (1) cognitive impairment—if
based on the Consortium to Establish a Registry for Alzheimer’s
the neurologist found moderate or severe dementia to be present,
Disease (CERAD) method, with additional features to allow
and the neuropsychologist independently confirmed the presence
Braak staging of neurofibrillary pathology. The first 101 brains
of cognitive impairment consistent with dementia; (2) cognitive
were reported by Xuereb et al. (2000), with half having developed
decline—evidence that the person had definite deterioration in
dementia by death. The median age group was 86–90 years. This
cognition from a pre-existing level of functioning; and (3) duration
study found that tangles and neuritic plaques were highly inter-
of cognitive impairment of at least 1 year (Linn et al., 1995).
correlated. Significant, but weaker, correlations were also found
between plaques and vascular amyloid, plaques and white-matter Framingham study: dementia
pallor, tangles and vascular amyloid, and between tangles and The prevalence of dementia was 30.5/1,000 for men and 48.2/1,000
white-matter pallor. Microvascular infarcts showed no relationship for women and increased with advancing age. Cases of probable AD
to any of these measures (Xuereb et al., 2000). constituted 55.6% of all dementia cases. The prevalence of AD was
11.7/1,000 for men and 30.1/1,000 for women and also increased of functioning. The survey elicited information from persons 65
with advancing age. Prevalence of dementia and probable AD were years of age and older in four different geographic locations in the
greater for women than men. The women to men ratio of preva- US: East Boston, New Haven, Iowa, and North Carolina. The base-
lence for cohort members 75 years of age and older was 1.8 for all line data cover demographic characteristics (age, sex, race, income,
cases of dementia and 2.8 for cases of probable AD (Bachman et al., education, marital status, number of children, employment, and
1992). religion), height, weight, social and physical functioning, chronic
To determine the incidence of dementia and AD (Bachman et al., conditions, related health problems, health habits, self-reported use
1993), people previously free of dementia and falling below the 1 of dental, hospital, and nursing home services, and depression.
SD cut-off score on screening tests were evaluated further to verify In East Boston, individuals eligible for the study were identified
whether dementia was present and, if so, the type of dementia, fol- through a total community census performed concurrently with
lowing the procedure described above. All new cases arising in this the baseline interview in 1982. In Iowa, the sample was not truly
cohort over a maximum of 10 years of follow-up were ascertained. populational since interviews were attempted with all eligible indi-
The incidence of dementia increased with age, doubling in succes- viduals enumerated using a list from the area’s Agency on Aging,
sive 5-year age groups. Dementia incidence rose from 7.0/1,000 per supplemented by additional listings from local informants. New
year at ages 65–69 to 118.0/1,000 at ages 85–89 for men and women Haven used a stratified random sample of clusters of households.
combined. The incidence of probable AD also doubled with succes- The sample was stratified for three types of residence, including
sive quinquennia, from 3.5 at ages 65–69 to 72.8/1,000 at ages 85–89 public housing for the older population, private housing for the
years. Incidence of dementia and of probable AD did not level off older population, and elsewhere in the community. Men were over-
with advanced age and was not different in men and women. sampled to attempt to achieve balance in the sex distribution of the
sample. In North Carolina, area sampling was used at the first stage
Framingham study: depression
of the design to obtain a sample of 1,980 census blocks, block clus-
In 1990, at the start of the 22nd biennial examination cycle, 1,753 ters, and enumeration districts. The sample was designed so that it
people from the original cohort were still alive. Of these, 1,166 would consist of at least 50% black older persons. The investigation
(67%) attended the 22nd biennial examination, and among them, of dementia was conducted using the sampled persons members
949 (81%; 604 women, 345 men) were dementia free and were of the Duke EPESE (Established Populations for Epidemiologic
assessed for depressive symptoms. These participants were followed Studies of the Elderly) (Heyman et al., 1991), which focused on five
for up to 17 years (average follow-up 8 years) for incident dementia adjacent counties, one primarily urban and the other four mostly
to examine the association between depressive symptoms at base- rural, in the Piedmont area of North Carolina. The last stage of
line and risk of incident dementia (Saczynski et al., 2010). the sampling procedure consisted of selecting one older person at
During the 17-year follow-up period, 164 participants devel- random from each household in which there were residents aged
oped dementia; 136 of these cases were AD. A total of 21.6% of 65 years and older. Of 5,223 persons selected for the sample, 4,164
participants who were depressed at baseline developed dementia, were successfully interviewed, yielding a response rate of 80%. Of
compared with 16.6% of those who were not depressed. Depressed these, 2,259 (54%) were black and 1,905 (46%) were nonblack. The
participants, defined by scores of at least 16 on the Center for sampling design permitted the development of weights, which took
Epidemiologic Studies Depression Scale (CES-D) (Radloff, 1977), into account age, sex, race, geographic location, number of old peo-
had more than a 50% increased risk for dementia (HR = 1.72, 95% ple in the household, and nonresponse, so that it was possible to
CI = 1.04, 2.84) and AD (HR = 1.76, 95% CI = 1.03, 3.01). Results project data from the Duke EPESE sample to the same age popula-
were similar when participants taking antidepressant medications tion of other areas.
were included. For each 10-point increase on the CES-D, there was A brief screen of cognitive function, the Short Portable Mental
significant increase in the risk of dementia (HR = 1.46, 95% CI = Status Questionnaire (SPMSQ) (Pfeiffer, 1975), was administered
1.18, 1.79) and AD (HR = 1.39, 95% CI = 1.11, 1.75). Results were at baseline, and, according to the authors by using the adjusted
similar when persons with possible MCI were excluded. scores, little race or education bias is found. The case finding pro-
Framingham study: summary cedure involved a trichotomization of the SPMSQ scores, that
Another important contribution of the Framingham study to the is, the total EPESE sample was divided into persons with scores
field of dementia was the estimation of the lifetime risk of demen- below the cut-off for cognitive impairment, those with one point
tia. A group of 2,794 participants without dementia who were 65 better than the cut-off, and, finally, those with two points better
years or older were followed up for a maximum of 29 years (42,233 than the cut-off. The selected participants were seen by a neurolo-
person-years). There were 400 cases of incident dementia of all gist who administered a semistructured interview which involved
types and 292 cases of incident AD. The lifetime risk of any demen- a medical and psychiatric history and physical and neurological
tia was estimated at more than 1 in 5 in women and 1 in 6 in men, examinations. Based on these measures, a diagnosis of demen-
and the lifetime risk of AD about 1 in 5 for women and 1 in 10 for tia and, specifically, of AD was made, using the DSM-III and
men (Seshadri et al., 2006). NINCDS-ADRDA (McKhann et al., 1984). The severity of the
dementia, when present, was rated on the Clinical Dementia Rating
Established Populations for Epidemiologic Studies of the scale, which includes information for rating the individual in six
Elderly (EPESE) cognitive and behavioural categories: memory, orientation, judge-
The goals of the Established Populations for Epidemiologic Studies ment and problem solving, community affairs, home and hobbies,
of the Elderly (EPESE) project were to describe and identify predic- and personal care (Hughes et al., 1982). The project attempted to
tors of mortality, hospitalization, and placement in long-term care compare the prevalence and the incidence of dementia between
facilities and to investigate risk factors for chronic diseases and loss black and all other participants.
EPESE: dementia impairment and depressed mood severe enough for further clini-
For the 1986 and 1987 survey, when the baseline data were collected, cal intervention (Kay et al., 1964). The final sample for interview
the estimated prevalence of dementia was 8.9% for black men, consisted of 932 participants (women over 60, men over 65), who
19.9% for black women, 3.3% for white men, and 2.9% for white represented 15.2% of the ward population. The response rate of
women. The overall estimated prevalences for blacks and whites available people was 87.2%.
were 16% (95% CI = 7.9, 24.1) and 3.0% (95% CI = 0, 6.9), respec-
Gospel Oak study: dementia
tively (Heyman et al., 1991). The overall prevalence of dementia
On the six screening scales, 8% were identified as cases of organic
among blacks was significantly higher than in whites. However, the
brain syndrome, 27% as depressive cases, 25% with subjective
same difference was not found in the 1989 and 1990 survey, where
memory complaints, 33% had sleep disorders, 32% had limitation
the prevalence of dementia was 7.0% (95% CI = 2.1–11.9) for blacks
in performing activities, and 24% had somatic symptoms. After
and 7.2% (95% CI = 2.2, 12.2) for whites. Prevalence for black men
the refinement for identifying syndromes of cognitive impairment
(7.8%, 95% CI = 0.1, 15.5) exceeded that for black women (6.6%,
(the Dementia Diagnostic Syndrome scale), dementia prevalence at
95% CI = 0.3, 12.9), but gender prevalence values for whites were
screening was estimated as 4.7%. However, this value increased to
reversed (men: 4.4%, 95% CI = 0.0, 10.3; women: 8.7%, 95% CI =
7% when the residents of the local authority home were included.
1.5, 16.0) (Fillenbaum et al., 1998).
Eighty percent of the 60 people initially identified by screen-
The 3 year incidence of dementia between the two surveys, 1986–
ing were further assessed using the GMS/AGECAT instrument
1987 and 1989–1990, was found to be 5.8% (95% CI = 2.6, 9.0) for
(Copeland et al., 1986), a psychiatric interview, and neuropsycho-
blacks and 6.2% (95% CI = 2.7, 9.7) for whites. Neither race nor
logical testing (Livingston et al., 1990a). Among those participants,
gender differences were significant (Fillenbaum et al., 1998).
43 were diagnosed as having dementia of any type based on the
EPESE: depression psychiatric diagnosis. For the whole Gospel Oak population, the
To address the influence of socioeconomic variables in the asso- overall prevalence for dementia of any type was 6.1% (43/705). AD
ciation between racial differences and late-life depression, basic prevalence, according to the NINCDS-ADRDA criteria (McKhann
needs, income, and education variables, controlling for sex, age, et al., 1984), was 3.1% (22/705). Multi-infarct dementia (0.01%,
and functional status were examined (Sachs-Ericsson et al., 2005). 1/705), mixed dementia (0.7%, 5/705), and secondary dementia
Data for this analysis were derived from the Duke EPESE. Before (0.7%, 5/705) were rare. In contrast, diagnosis based on the GMS/
adjusting for socioeconomic variables, African-American older AGECAT resulted in a prevalence of organic case of 4% (28/705).
people had more depressive symptoms than white older partici- GMS/AGECAT was more likely to diagnose as ‘organic’ those par-
pants in the cross-sectional analyses. However, after the inclusion ticipants whom the psychiatrists diagnosed as having AD. The total
of socioeconomic variables, the relationship was inverted, such that dementia prevalence as diagnosed by the psychiatrists was higher
white-Americans were significantly more likely to endorse depres- (6.1%) than reported at screening (4.7%), where the prevalence was
sive symptoms than African-Americans. In the longitudinal analy- calculated according to the Dementia Diagnostic Syndrome scale.
ses, after controlling for baseline depressive symptoms, race was In 1990, 502 participants were successfully rescreened using the
unrelated to depressive symptoms 3 years later. Short CARE. Six percent of the rescreened participants were identi-
fied as incident dementia cases based on screening (Boothby et al.,
EPESE: summary
1994) and 1.6% according to the Dementia Diagnostic Syndrome
The study highlights the importance of race and demographic scale. However, by means of clinical diagnosis, the overall annual
and sociodemographic factors in determining estimates of disease incidence of dementia in persons over 65 years was 2%.
prevalence and incidence. These findings reinforce that there is
room for improvement of chronic conditions by reducing socio- Gospel Oak study: depression
economic disparities. Depression varied according to place of residence; 17.3% (122) of
the population living at home were classed as probably suffering
Gospel Oak study from pervasive depression as measured using the depression diag-
This was a longitudinal prospective study based on the Gospel Oak nostic scale. Excluding those individuals who also were diagnosed
electoral ward in north London (Livingston et al., 1990b). Its main as having dementia reduced the prevalence of depression to 15.9%.
purpose was to detect those persons likely to be suffering from On the other hand, prevalence increased to 18.5% when the resi-
dementia or depression or to be impaired in performing ADL. The dents of the local authority home were included. Depression was
electoral ward had an estimated population of 6,136 living in 3,000 not associated with age (Livingston et al., 1990b).
households and had higher rates of most indices of deprivation
than the average for England and Wales at the time. Gospel Oak study: summary
Interviews took place in 1987. The sample consisted of all women Additional research questions within the framework of the Gospel
aged over 60 and men over 65 who were residents of this area. All Oak study have focused on the association between cigarette
participants were interviewed using the Standard Comprehensive smoking and alcohol drinking and incident cognitive impairment
Assessment and Referral Evaluation (CARE) (Gurland et al., 1984). (Cervilla et al., 2000). Participants were asked whether they had
This instrument is composed of six screening scales that assess ever smoked, and among those who smoked, currently or in the
depression, organic brain syndrome, subjective memory impair- past, information was obtained on the average number of ciga-
ment, sleep disorder, somatic symptoms, and activity limitation. rettes smoked a day and on the number of years they had smoked
The depression and dementia scales have been further refined to for. Participants were also asked about the amount of alcohol used
become depression and dementia diagnostic scales (the Dementia before or after the age of 65. Current smokers were nearly four times
Diagnostic Syndrome scale), which refer to syndromes of cognitive more likely to be cognitively impaired than non (never) smokers or
ex-smokers, after adjusting for baseline cognitive function, depres- population aged 65 years and over, the standardized (to the England
sion, occupational class, education, handicap, and alcohol con- and Wales population estimates for 1991) dementia prevalence was
sumption before and after the age of 65. Alcohol drinking was not estimated as 6.6% (95% CI = 5.9, 7.3). This was based on demen-
a risk factor for incident cognitive impairment. The finding that tia defined as an AGECAT organicity rating scale of 3 and above
current smokers but not ex-smokers are at higher risk of develop- (1998). Dementia prevalence across the centres did not vary greatly,
ing cognitive decline is of great public health relevance for smoking although there was some evidence of nonsystematic fluctuation in
prevention and smoking cessation campaigns and polices targeting individual age- and sex-specific groups.
prevention of cognitive impairment. Dementia incidence estimated between the first two waves of
interviews increased with age, from 7.4 (95% CI = 3.6, 16.1) per
Cognitive Function and Ageing Study (CFAS)
1,000 person-years at age 65–69 years to 84.9 (95% CI = 63.0,
CFAS is a multidisciplinary, multiphase, population-based study, 107.8) per 1,000 person-years at age 85 years and above. The rate
which involved five identical sites including Cambridgeshire, of increase for both sexes was marked, and continued into the old-
Gwynedd, Newcastle, Nottingham, and Oxford. An additional site est age groups. It was estimated that approximately 180,000 new
was based in Liverpool, but the sampling and interview structure cases of dementia occur in England and Wales each year. There
base were different. The study was designed to cover three main is no convincing evidence of variation across sites, and incidence
areas related to dementia and ageing: epidemiology, neuropathol- rates did not reflect the variations in the prevalence of possible risk
ogy, and policy (Brayne et al., 2006). The fieldwork began in 1991 factors in these sites (Matthews and Brayne, 2005). CFAS has also
(Chadwick, 1992). provided profiles of cognition weighed back to the UK popula-
Background information on the demographics of the population based on the MMSE, extended MMSE, and CAMCOG scores
tions sampled was collected from the OPCS 1990–1991 census, to (including total score and subdomain scores of orientation, lan-
enable comparison with regional and national data. The popula- guage (expression and comprehension), memory (learning, recent,
tion sample was drawn from the Family Health Service Authorities remote), praxis, attention, calculation, and perception) (Williams
lists. These are registers of the general practitioners, which provide et al., 2003; Huppert et al., 2005).
a nearly total population enumeration in the areas chosen for study, The scope and operability of 16 different terms reflecting cog-
including individuals living in institutions. Individuals aged 65 nitive decline intermediate to normal ageing and dementia, and
years and over were selected for participation. quantification of their prevalence and longitudinal course of dis-
Trained interviewers administered structured interviews at the ease, have also been undertaken (Matthews et al., 2008). Across the
respondents’ homes, which included the GMS, the MMSE (Folstein 16 definitions, prevalence estimates were found to vary substan-
et al., 1975), and basic information on residence, marital status, tially (range 0.1–42%). As expected, prevalence tended to increase
social class, and main occupation during working life (Elias et al., for those definitions that capture a broader state of impairment,
1993); social and service contacts (Wenger, 1989); physical health including, for example, Subjective Memory Complaint (SMC) and
and wellbeing, including vascular risk factors (Launer et al., 1992); Cognitive Impairment No Dementia (CIND), and was less frequent
ADL as measured by the Townsend scale (Townsend, 1979); and for more restrictive definitions including Amnestic Mild Cognitive
regular medication use (prescribed and over the counter). CFAS Impairment (A-MCI). Rates of progression to dementia also varied
also includes a brain donation programme, with a total of 456 brains and tended to be low. Overall, dementia progression was highest
donated up to August 2004 (Matthews and Brayne, 2005). It is esti- for more broadly defined concepts, and those that capture greater
mated that in 2012 around 500 brains will have been donated. levels of cognitive decline, particularly in older individuals. Across
From the five identical sites, 13,004 individuals were screened definitions, at 2 years’ follow-up, most individuals had remained
(85% response from eligible sample). Following screening, a 20% stable, reverted to normal, or developed impairment outside the
subsample was selected based on age and cognition, weighted intermediate range. In the neuropathological analysis, MCI was
towards the older and more cognitively frail, to complete a more found to be associated with an increased risk of neurodegenerative
in-depth baseline interview shortly afterwards, with a repeat at 2 and vascular pathologies (Stephan et al., 2012a, 2012b).
years. Further screen and assessment interviews with the whole In terms of risk factor analyses, CFAS has shown that a simple
sample who remained in the study were carried out (see <www.cfas. measure of self-rated health (SRH) was associated with a higher
ac.uk> for full details on the study design) (Brayne et al., 2006). risk of death and functional and cognitive impairment. The asso-
The ALPHA Liverpool study fits within the framework of the CFAS ciations remained after adjustment for age, gender, functional abil-
but started earlier than the other five centres (Saunders et al., 1993). ity, and MMSE at baseline: comparing those who rated their health
The Liverpool Family Practitioner Committee central computerized as excellent and good, hazard ratios for risk of death and functional
list of general practice patients was used as a sample frame. From and cognitive impairment were 0.8 (95% CI = 0.8, 0.9), 0.6 (95%
this list, all patients aged 65 years or over with Liverpool addresses CI = 0.5, 0.7), and 0.7 (95% CI = 0.5, 0.9), respectively (Bond et al.,
were selected. To provide a check on diagnosis and to enable sub- 2006).
classification of GMS/AGECAT organic disorder into dementia Results from the neuropathological analyses have found that
types, a subsample of individuals received a second-stage assessment multiple neuropathological features determine the overall bur-
and informant interview conducted by psychiatrists. The informant den of dementia, including mixed vascular and Alzheimer lesions
interview consisted of the History and Aetiology Schedule (HAS) together with other changes such as atrophy (Neuropathology
which is drawn upon by the HAS/AGECAT. Group, 2001; Matthews et al., 2009; Savva et al., 2009; Wharton et
CFAS: dementia al., 2011). Furthermore, the relationship between the clinical mani-
CFAS has reported on the UK population prevalence and incidence festations of dementia and the neuropathological findings varies
of dementia as well as extensively on cognition and MCI. In the with age, such that there is considerable overlap in the burden of
neuropathological features of AD between groups of the oldest old This new study is called the Cognitive Function and Ageing Study
persons with dementia and those without dementia. Even in partic- II (CFAS-II) and includes centres in Cambridgeshire, Newcastle,
ipants who died without dementia, the burden of Alzheimer’s-type Nottingham, and CFAS Wales: Gwynedd and Swansea. This study
disease in the population increased with increasing age. will provide baseline information on approximately 12,500 people
In the Liverpool study, GMS interviews were obtained with 5,222 aged 65 and over in 2008–2011, and will follow them up over time,
(87%) of the 6,035 in the study area. A total of 444 GMS/AGECAT with a 2-year phase confirmed (2014). The data collected will pro-
organic cases were identified at phase 1 and a sample of them was ran- vide important information on generational and geographical dif-
domly selected for reinterview at phase 2. Among the available 205 ferences, including details on those living in institutions. CFAS-II
participants who were assessed by psychiatrists to verify the clinical will allow the estimation of new patterns of the number of people
diagnosis, dementia was diagnosed in 84% (n = 172) of the organic with dementia and disease comorbidity.
cases. At wave 2, 328 were diagnosed as cases of organic disorder
Rotterdam study
(of whom 120 were from wave 1 and 208 were new cases), and 232
in wave 3 (of whom 54 were from wave 1 and 53 from wave 2, leav- The Rotterdam study (<http://www.epib.nl/research/ergo.htm>) is
ing 125 new cases). Comparison of the confidence intervals for the a single-centre population-based prospective dynamic cohort study
age-specific rates by sex showed no significant sex difference for the of individuals aged 55 and over that started in 1990 in Ommoord,
incidence rates of undifferentiated dementia (Copeland et al., 1999). a suburb of Rotterdam. The main objective has been to investigate
the prevalence and incidence of risk factors for chronic diseases in
CFAS: depression the older population, including cardiovascular, neurological, loco-
The age and sex prevalence of depression standardized to the 1991 motor, and ophthalmologic diseases.
population of England and Wales was 8.7% (95% CI = 7.3, 10.2). Baseline measurements were obtained between 1990 and 1993
No relationship between depression prevalence and age was found. and all participants were subsequently examined every 2–3 years.
However, high deprivation, high disability, and two or more comor- In total, 7,983 (78%) persons took part, including 897 persons liv-
bid illnesses were associated with a greater prevalence of depression ing in one of the six homes for older people. In 2002, another 3,011
(McDougall et al., 2007). Previous neuropathology studies have participants (55 years of age since 1990) were added to the cohort,
found an association between depression and markers of neurode- which comprised a total of 10,994 persons. In 2006, the cohort was
generative and nonAD pathology, including neurofibrillary tangles, further expanded by 3,932 persons aged 45 years and over. The total
diffuse and neuritic plaques, Lewy bodies, brain atrophy, and cer- Rotterdam study population encompasses 14,926 participants. In
ebrovascular disease (Thomas et al., 2001; Wilson et al., 2003; Sweet 1995 and 1999 random subsets of the Rotterdam study underwent
et al., 2004; Jellinger, 2009). In the CFAS neuropathology resource, neuroimaging, and from 2005 onwards magnetic resonance imag-
depression (n = 36 out of 153 nondemented participants at death) ing (MRI) has been implemented into the core protocol of the
has been associated with subcortical Lewy bodies (Tsopelas et al., Rotterdam study. Up to January 2011, a total of 5,886 brain MRI
2011). In contrast to early findings from other studies, no asso- scans have been obtained, which includes multiple scans from the
ciation was found between depression and cerebrovascular or same person (Ikram et al., 2011).
Alzheimer pathology, although depression was associated with Regarding old age psychiatric conditions, the Rotterdam study
neuronal loss in the hippocampus as well as in some of the sub- addresses questions about the prevalence and incidence of vari-
cortical structures investigated (nucleus basalis, substantia nigra, ous types of dementia and of Parkinson’s disease, and also verifies
raphe nucleus) (Tsopelas et al., 2011). It is important to note that the determinants of such conditions. Participants were screened
despite the strong statistical association with depression in the sam- for dementia using a brief cognitive test including the MMSE and
ples where subcortical Lewy bodies or neuronal loss were detected, GMS. Once screened positively, participants were then seen by a
these neuropathological features were relatively rare, and most physician with the CAMDEX, and those who were still suspected of
cases of depression found in this sample were not associated with dementia were examined by a neurologist, had a MRI of the brain,
any subcortical pathology. This suggests that subcortical pathology and were tested by a neuropsychologist. No interviews were done
may account for a small number of cases of late-life depression in on those who screened negatively (MMSE 26–30).
the population.
Rotterdam study: dementia
In the ALPHA Liverpool study, the relationship between depres-
Of the 10,275 eligible individuals, 7,528 (73%) were screened for
sion and risk of incident dementia was also investigated (Chen
dementia, and 6.3% of them were diagnosed as having dementia.
et al., 2008). The risk of dementia was significantly increased with
Overall, 72% of the dementias were of Alzheimer type, 16% were
level 4 depressive syndromes derived from the GMS/AGECAT. The
vascular dementia, 6% were Parkinson’s disease dementia, and 5%
multiple adjusted HR is 2.47 (95% CI = 1.25, 4.89) and 2.62 (95%
were other dementias (Ott et al., 1995). At follow-up, 5,571 (79%)
CI = 1.18, 5.80) at 2- and 4-year follow-up, respectively. The effect
participants were rescreened for dementia. The overall incidence
was greater in younger participants.
was 10.7/1,000 person-years (Ott et al., 1998). Dementia subtype
CFAS: summary clinical diagnosis was determined in 98% of the cases. AD was
Estimates of prevalence and incidence of dementia, cognitive diagnosed in 61%, mixed dementia of Alzheimer with cerebrovas-
decline, and depression from the CFAS represent true population cular disease was detected in 12%, vascular dementia in 14%, and
estimates due to study design. This has important implications for other dementias in 13%.
gauging the consequence of each disease (e.g. burden of care, cost) Results from the neuroimaging data show that white matter
from a true population perspective. The CFAS collaborators have lesions are related to impairment of subcorticofrontal functions
begun collecting information on a new sample of individuals aged (Breteler et al., 1994), but also that changes affecting the micro-
65 years and older that builds on the original CFAS study design. structural integrity of normal white matter before these can be
visualized using conventional MRI are also associated with cogni- when defined neuropathologically according to the CERAD proto-
tive function (Vernooij et al., 2009b). Those changes, such as mean col (Mirra et al., 1991). Forty-one (55%) of the 74 individuals with
diffusivity and fractional anisotropy, correlate directly with the neuropathological AD were either in the no-dementia group or
amount of myelin in the white matter and to a lesser extent also had dementia of nonAlzheimer’s type clinically defined (Polvikoski
to axonal count. They can be measured using the diffusion tensor et al., 2001). The incidence of dementia was 8.1/100 person-years
imaging technique in the MRI. (Ahtiluoto et al., 2010).
Regarding cerebral microbleeds, their spatial distribution was In this study, the authors also investigated the relation of dia-
found to follow the known topographic distribution of amyloid betes to dementia, AD, and vascular dementia based on the fact
angiopathy implicated in AD (Vernooij et al., 2009a). They were that population-based longitudinal studies have shown contro-
present in 1 in 5 persons over the age of 60 and in over 1 in 3 in versial findings regarding diabetes as an independent risk factor
persons aged 80 years and older (Vernooij et al., 2008; Poels et al., for dementia. When the prevalence of dementia was split among
2010). Furthermore, the presence of these numerous microbleeds, those with and without diabetes, no association was found between
especially in a strictly lobar location, was associated with worse dementia and diabetes after adjusting for sex, age, education, car-
performance on cognitive tests, even after adjustment for vascu- diovascular conditions, and apolipoprotein E (APOE) e4 allele fre-
lar risk factors and other imaging markers of small vessel disease quency. However, the incidence of clinically defined dementia in
(Ikram et al., 2011). These results suggest an independent role for participants free of dementia at baseline (n = 355) was almost twice
microbleed-associated vasculopathy in cognitive impairment. as high in patients with diabetes (12.1 person-years, 95% CI = 8.5,
Rotterdam study: depression 7.2) than nondiabetic (7.2 person-years, 95% CI = 5.7, 9.0) indi-
viduals, even after adjustments for age, sex, education, and APOE
Depressive disorders were assessed by a two-step procedure that
e4 status (Ahtiluoto et al., 2010). One possible explanation for the
included completion of the Dutch version of the original Center for
difference between the findings for prevalence and incidence is that
Epidemiological Studies Depression Scale (CES-D), and for those who
the dementia duration was shorter in diabetic compared to nondia-
screened positive for depressive symptoms a clinical psychiatric eval-
betic participants, probably due to the diabetes-related increase in
uation was undertaken. A strong relationship was observed between
mortality.
severe coronary and aortic calcifications and depressive disorders
Regarding the neuropathological findings, the proportion of
(odds ratio (OR) = 3.89, 95% CI = 1.55, 9.77; and OR = 2.00, 95%
individuals with beta-amyloid and neurofibrillary tangles was
CI = 1.02, 3.96, respectively). Although the analysis cannot establish
lower in diabetic compared to nondiabetic individuals. In contrast,
a causal role of atherosclerosis due to the cross-sectional nature of the
the proportion of participants with cerebral infarctions was signifi-
study, it provides evidence that a generalized atherosclerotic process is
cantly higher in diabetic compared to nondiabetic persons, even
associated with late-life depression (Tiemeier et al., 2004).
after adjustments for age at death, gender, education, APOE, and
Rotterdam study: summary dementia status.
The Rotterdam study has extensive information from a wide variety
of sources, such as biological and MRI-based data, that can be used Vantaa 85+: depression
to help untangle the association between health and health-related Depressive symptoms were investigated using the Zung Depression
risk factors and poor mental health. The high prevalence with MRI Status Inventory (DSI) (Zung, 1972), which is a 20-item semistruc-
scans has helped with understanding the association between white tured, interviewer-rated depression instrument with scores ranging
matter lesions and cognitive impairment. from 25 to 100. Higher scores indicate more depressive symptoms.
In the general population aged 85+ there was a very low prevalence
Vantaa 85+ of depression (1.1% for clearcut clinical depression and 4.1% for
The Vantaa 85+ study is a prospective population-based study minimal to mild depression). Among participants with dementia,
which was established in 1991 (Polvikoski et al., 1995). The study vascular dementia was significantly more common in individuals
population included all individuals who were born before 1 April with higher depression scores (40 points or more on the DSI). There
1906 and were aged 85 years or over living in the city of Vantaa in was no association between DSI score and dementia severity.
southern Finland (n = 601). Clinical examination was possible of
553 (92%) participants and a neuropathological examination of 304 Vantaa 85+: summary
(51%). A neurologist and a trained public health nurse performed The Vantaa 85+ study is one of the few autopsy-controlled, pro-
the clinical evaluations, which included a structured general and spective, and population-based studies available on the prevalence
neurological examination. Data were systematically collected of dementia in the very old population.
on health, health-related behaviour, and medication. Cognitive Personnes âgées QUID (Paquid)
function, depression, and functional abilities were also assessed.
Paquid was the first large French epidemiological study on demen-
Survivors were re-examined in 1994, 1996, 1999, and 2001. The
tia. It is an interdisciplinary study designed to investigate cerebral
entire study population is now deceased.
and functional ageing. Residents living in two administrative areas
Dementia was diagnosed using DSM-III-R criteria, AD by
of southwestern France (2,797 in Gironde and 1,504 in Dordogne)
NINCDS-ADRDA criteria, and vascular dementia using the
were randomly chosen from the electoral lists. A cohort of 3,777
NINDS-AIREN criteria. Initial blood samples for DNA analysis
participants aged 65 or over was interviewed. This cohort was com-
were obtained from 550 of the 553 persons examined.
plemented by a random sample of 380 institutionalized persons
Vantaa 85+: dementia (Letenneur et al., 1993a).
AD was clinically diagnosed in 16% of the Vantaa 85+ cohort. In Trained psychologists administered a standardized question-
contrast, the prevalence of AD was 33% (Polvikoski et al., 2001) naire at home. Baseline variables included sociodemographic
factors, living conditions and habits, subjective health measures, Italian Longitudinal Study of Ageing (ILSA)
dependence in ADL (Katz et al., 1970) and instrumental activities ILSA is a population-based, longitudinal study aimed at determin-
of daily living (IADL) (Lawton and Brody, 1969), and the Rosow ing the health status of people aged 65–84 years (Maggi et al., 1994).
and Breslow scale (Rosow and Breslau, 1966). Depressive symp- Common chronic conditions were investigated along with poten-
tomatology was also assessed. Intellectual functioning was exam- tial risk and protective factors. ILSA was also designed to assess
ined through an extensive test battery that included an evaluation age-associated physical and mental functional changes. A random
of global mental status, visual memory, verbal memory, verbal flu- sample of 5,632 people, stratified by age and sex using an equal
ency, visuospatial attention, and simple logical reasoning. allocation strategy, was gathered from the demographic lists of the
After the psychometric evaluation, the psychologists completed registry office of eight municipalities including: Genoa, Segrate
systematically a standardized questionnaire, allowing determina- (Milan), Selvazzano-Rubano (Padua), Impruneta (Florence),
tion of the DSM-III criteria for dementia. Patients who met the Fermo (Ascoli Piceno), Napoli, Casamassima (Bari), and Catania.
DSM-III criteria for dementia were seen by a neurologist who The baseline examination started in March 1992. The case identifi-
applied the NINCDS-ADRDA criteria to indicate the aetiology of cation for all conditions was based on a two-phase procedure, con-
the deterioration. Persons were re-evaluated following the same sisting of a screening phase, administered by lay interviewers, and,
procedure as used for the baseline screening at 1, 3, and 5 years after for those who screened positive, a clinical assessment run by spe-
the initial visit in Gironde and 3 and 5 years after the initial visit cialists according to the condition. The final diagnosis of demen-
in Dordogne. However, to improve the sensitivity of the detection tia was made according to the DSM-III-R criteria, for AD based
of incident cases, respondents were selected for the neurological on the NINCDS-ADRDA criteria, and for vascular dementia and
examination if they met the criteria for DSM-III-R dementia or if other dementias according to ICD-10. From the total population
they had experienced a cognitive decline of more than two points sampled, 5,462 were eligible and 84% took part in the home inter-
on the MMSE. view and 64% participated to the clinical examination (The Italian
Paquid: dementia Longitudinal Study on Aging Working Group, 1997)
The overall prevalence of dementia was estimated based on the ILSA: dementia
Gironde data and was 4.3%. No difference was found between men Among the 3,497 persons included in the baseline wave, the prev-
and women (Letenneur et al., 1993b). Of the 5,554 contacted sub- alence of dementia was 7.2% for women and 5.3% for men (The
jects, 3,777 (68%) agreed to participate in the study. Risk factor Italian Longitudinal Study on Aging Working Group, 1997). The
analysis found greater risk for AD in people with fewer years of follow-up phase was conducted on 2,498 individuals free of demen-
formal education. The overall incidences of dementia and AD were tia who were reassessed after a mean period of 3.8 years. The inci-
estimated at 1.59/100 person-years and 1.17/100 person-years, dence rate of overall dementia was of 12.5/1,000 person-years (Di
respectively. The incidence of AD was higher in women than men Carlo et al., 2002).
after 80 years of age, whereas the incidence was higher in men
before the age of 80. This different progression of the incidence ILSA: depression
according to sex was not found when other dementias were ana- Depressive symptoms were investigated using the Italian version
lysed (Letenneur et al., 1999). of the 30-item GDS (Yesavage et al., 1982). Further investigation
sought to determine the possible impact of depressive symptoms
Paquid: depression on the rate of progression to dementia in individuals diagnosed
To estimate the predictive relationship between depressive symp- with MCI (Panza et al., 2008). Among the 2,963 participants, 139
toms and incident dementia, Fuhrer et al. (2003) investigated prevalent patients with MCI were diagnosed at baseline. During
16,373 person-years of observation. Baseline prevalence of depres- 3.5-year follow-up, 14 patients with MCI progressed to dementia,
sive symptomatology was 12.9% for men and 14.7% for women. of whom nine had a GDS over 10. The association between depres-
The OR for the age-adjusted association between elevated depressive symptoms and rate of progression to dementia in MCI was not
sive symptoms and onset of dementia was 2.0 (95% CI = 1.4, 2.8). significant (Relative Risk (RR) = 1.42, 95% CI = 0.48, 4.23).
However, after adjusting for gender, education, and cognition, the
risk was reduced to 1.3 (95% CI = 0.8, 2.0), but these associations ILSA: summary
were significantly different between men and women. Men who This cohort has produced results on the prevalence of not only
had high depressive symptoms were more than three times as likely neuropsychological conditions, such as dementia, but also general
(OR = 3.5, 95% CI = 1.9, 6.5) than men with low depressive symp- medical disorders, such as myocardial infarction, angina, arrhyth-
toms to develop incident dementia. This effect was not replicated mia, congestive heart failure, peripheral arterial disease, hyperten-
for women. A possible explanation for this finding could be due sion, diabetes, stroke, parkinsonism, distal symmetric neuropathy
to sex differences in vascular disease. It was found that the risk of of lower limbs, and disability. These data are of great importance
dementia for men with hypertension who were depressed was 50% for planning of health services, to evaluate the real needs for assist-
higher than for normotensive depressed men. ance, and to study those factors that determine the transition from
independence to loss of autonomy.
Paquid: summary
Paquid is an interdisciplinary study on cerebral and functional age- The Three-City study (3C)
ing, made up of a cohort of 3,777 community residents living in two The main objective of the 3C study is to estimate the risk of
administrative areas of southwestern France. The epidemiological dementia attributable to vascular diseases or vascular risk factors,
basis of the programme focused on the incidence, natural history, and to provide data for modelling the expected impact of vascu-
and nongenetic risk factors of dementia. lar risk reduction on the incidence and prevalence of dementia
(<http://www.three-city-study.com/>). Other objectives are to circular process of the diagnosis of vascular dementia, which might
study incidence and risk factors of stroke, to provide data on result in overemphasizing the strength of relationships.
incidence and risk factors of coronary diseases, and to analyse 3C: depression
temporal trends in the incidence and prevalence of incapacities
The relationship between metabolic syndrome and depression was
and loss of autonomy (Alperovitch et al., 2002). Recruitment was
also investigated. Both metabolic syndrome and depressive symp-
undertaken based on the electoral registries of three French cit-
toms increased during old age (Akbaraly et al., 2011). Over the
ies (Bordeaux, Dijon, Montpellier) and included individuals aged
4-year follow-up, 827 (18.6%) new cases of depression measured by
65 years, between 1999 and 2011. The acceptance rate was 37%
the CES-D (Radloff, 1977) were observed. Participants with meta-
(9,693) of the selected people who could be contacted.
bolic syndrome were more likely to develop depressive symptoms
Data were collected during face-to-face interviews using stand-
(OR = 1.73, 95% CI = 1.02, 2.95) compared to participants without
ardized questionnaires. Baseline workup included extensive
metabolic syndrome, even after adjusting for sociodemographic
assessment of vascular risk factors including blood pressure meas-
characteristics, smoking, alcohol consumption, and health status
urements; ultrasound examination of the carotid arteries; meas-
factors such as treatment, cognitive deficit, disability, BMI, and
urement of biological parameters such as blood glucose, urea and
self-report history of cerebrovascular disease at baseline. However,
electrolytes, and lipids; cognitive functioning; and a clinical diagno-
this association was not significant within the older age groups
sis of dementia. Cerebral MRI examinations were also performed
(70–75; 75–80; and 80–91). Regarding the specific components
in a subsample of 3,442 persons aged between 65 and 79 years.
of metabolic syndrome, low HDL cholesterol was associated with
Participants have been re-examined, on average, every 2 years. The
increased odds of new-onset depressive symptoms in those aged
third wave of follow-up examinations started in 2006 and is due for
65–69 years. These results suggest that onset of late-life depression
completion in 2012.
(after 70 years old) does not share the same aetiology and risk fac-
Dementia was diagnosed using a three-step procedure (3C Study
tors as onset of depressive symptoms in middle aged and ‘young’ old
Group, 2003). First, screening was based on a thorough neuropsy-
people. However, the results could also be due to different response
chological examination by trained psychologists. From the initial
rates or bioresources at different ages.
9,693 cohort, 7 participants had to be excluded as they aged less
than 65 years. Second, the participants who were suspected of hav- 3C: summary
ing dementia on the basis of their neuropsychological performance This is a population-based study that focuses on the investigation
were examined by a neurologist. Four percent of the participants of the association between vascular health, cerebrovascular disease,
refused to take part in the medical interview. Finally, all suspected and risk of cognitive decline and dementias. New data collected
dementia cases were analysed by a common independent commit- from the third wave of follow-up will be important for looking at
tee of neurologists according to DSM-IV criteria. The committee the long-term impact of vascular disease and its risk factors on cog-
reviewed by teleconference all potential cases of dementia of the nitive health.
three study centres to obtain a consensus on diagnosis and aeti- The English Longitudinal Study of Ageing (ELSA)
ology based on all existing information. With regard to the dif-
ELSA is an interdisciplinary data resource on health, economic posi-
ferent subtypes of dementia, AD was diagnosed according to the
tion, and quality of life as people age (<http://www.ifs.org.uk/elsa>/).
NINCDS-ADRDA criteria, and vascular dementia based on history
The aim is to explore the relationships between health, functioning,
of vascular disease, Hachinski score (Hachinski, 1994), and MRI
social networks, and economic position. The sample was drawn from
whenever possible (Raffaitin et al., 2009). To date, baseline vascu-
households that had previously responded to the Health Survey for
lar risk has been investigated and prevalence of vascular disease at
England (HSE). The HSE is an annual cross-sectional household
baseline explored (3C Study Group, 2003).
survey that collects a wide range of health data and biometric meas-
3C: dementia ures. The main HSE samples were designed to be representative of
Baseline prevalence of dementia was 2.2% (3C Study Group, 2003). the English population living in private households. Fieldwork for
During 4 years of follow-up, 0.84 incident dementia cases per 100 the first wave of ELSA began in March 2002 and spanned 12 months,
person-years (95% CI = 0.72, to 0.95) were validated (Raffaitin being completed in March 2003. All households with one or more
et al., 2009). The authors investigated the relationship between inci- 50+-year-old individuals were eligible for participation. In the first
dent dementia and metabolic syndrome, defined according to the wave, 12,100 individuals were interviewed. The survey achieved a
National Cholesterol Education Programme Adult Treatment Panel household response rate of 70%; approximately 96% of individuals
III (NCEP ATP III) criteria (Grundy et al., 2005). Metabolic syn- responded within households.
drome was present in 15.8% of the study participants, and its pres- Topic areas covered at wave 1 included: individual and house-
ence increased the risk of incident vascular dementia but not AD hold characteristics; physical, cognitive, mental, and psychological
over 4 years (Raffaitin et al., 2009). High triglyceride level was the health; social participation and social support; housing, work, pen-
only component of metabolic syndrome that was significantly associ- sions, income, and assets; and expectations for the future. A shorter
ated with the incidence of all-cause (HR = 1.45, 95% CI = 1.05, 2.00) interview was attempted with a proxy informant if the eligible sam-
and vascular (HR = 2.27, 95% CI = 1.16, 4.42) dementia. Diabetes, ple member was unable to respond because of physical or mental
but not impaired fasting glycaemia, was significantly associated ill health, or cognitive impairment. All those interviewed in person
with all-cause (HR = 1.58, 95% CI = 1.05, 2.38) and vascular (HR were asked for permission to link their responses to administrative
= 2.53, 95% CI = 1.15, 5.66) dementia. It is worth highlighting that data sources. Respondents at wave 1 comprise the baseline study
one of the challenges in studies that use all the available information and individuals have been reapproached every 2 years, including
to diagnose subtypes of dementia is that it is difficult to avoid the those in institutions.
Langa et al. (2009) reported a crosscultural comparison between angiopathy and atrophy, lacunes, infarcts, white matter pallor,
cognitive performance of older adults in the Health and Retirement Braak stage, and brain weight. This study is the largest dataset of its
Study (HRS) in the US and the ELSA. As cognitive function is a key type in the world, with brain donor sample sizes of 241 (CC75C),
determinant of independence and quality of life among older adults, 304 (Vantaa), and 548 (CFAS).
the authors sought to identify sociodemographic and medical fac- EClipSE: dementia
tors associated with differences in cognitive function between the
The potential protective role of education for dementia was explored
two countries. The overall response rate among all eligible respond-
within the EClipSE study (Brayne et al., 2010). Although almost all
ents was 87% for the 2002 HRS and 67% for ELSA. The final study
older people have some pathology in their brain at death but have not
samples included 8,299 individuals from the HRS and 5,276 indi-
necessarily died with dementia, the relationship between education
viduals from the ELSA. The main findings were that despite a higher
and brain pathology at death was investigated, testing the hypoth-
prevalence of cardiovascular risks and cardiovascular disease
esis that greater exposure to education reduces the risk of demen-
among older US adults, they performed significantly better than
tia through either protection from pathology or compensatory
their English counterparts on tests of memory. While the authors
mechanism. Education during earlier life was recorded in number
were unable to confidently identify the cause or causes of this US
of years. Incident dementia was detected through follow-up inter-
‘cognitive advantage’, higher levels of education and wealth, lower
views, complemented by all the available sources of information,
levels of depressive symptoms, and more aggressive treatment of
such as retrospective informant interviews, death certificate data,
cardiovascular risks such as hypertension were pointed as possible
and linked health/social records after death. Dementia-related neu-
important contributing factors (Langa et al., 2009).
ropathologies were assessed based on the Consortium to Establish
ELSA: dementia a Registry for Alzheimer’s Disease (CERAD) protocol (Mirra et al.,
Although there are no data available on dementia in the ELSA 1991). Included were 872 brain donors, of whom 56% were diag-
study yet, some interesting results regarding cognition have been nosed as having dementia at death. The main findings in this study
published. Llewellyn et al. (2008) investigated whether psychologi- were that longer years in education were associated with decreased
cal wellbeing is associated with cognitive function. They found that dementia risk at death (OR = 0.89, 95% CI = 0.83, 0.94). Moreover,
higher levels of psychological wellbeing were associated with better education did not protect individuals from developing neurode-
global cognitive function and performance in multiple cognitive generative and vascular neuropathology by the time they died, but
domains, after controlling for the influence of depressive symptoms it did appear to mitigate the impact of pathology on the clinical
and a wide range of additional potential confounders. expression of dementia before death. In other words, for a specific
ELSA: depression pathological burden, those participants who remained in education
for longer, earlier in life, were at reduced dementia risk in older age.
The associations between dual sensory loss (hearing and vision)
These results support the ‘brain reserve hypothesis’ (Stern, 2002;
with onset and persistence of depression were investigated (Chou,
Valenzuela, 2008) where greater exposure to education reduces the
2008). There were 469 new cases of depression out of the 2,844
risk of clinical dementia by compensating for the pathological bur-
older participants who were not depressed at baseline (16.5%);
den later in life, rather than being protective against the accumula-
and among the 938 depressed at baseline, 549 (58.5%) were also
tion of pathology.
depressed at follow-up. Visual loss was found to be a robust predic-
Geographical and cross-cultural differences such as education
tor of both onset and persistence of depression, but dual sensory
should be taken into account when interpreting the study results.
loss was not.
During the early mid-twentieth century the educational systems
ELSA: summary differed considerably between the UK (CFAS and CC75C) and
ELSA data are being used to explore the dynamics of ageing, to Finland (Vantaa 85+). Participants from CFAS and CC75C com-
inform policy debates and for comparative analysis with the HRS pleted an average of 9 years of formal education, while participants
in the US and the Survey of Health and Retirement in Europe from the Vantaa 85+ study completed approximately four. However,
(SHARE). ELSA is still ongoing. even after controlling the analysis regarding education and demen-
Epidemiological Clinicopathological Studies in Europe tia for study site differences, age, and sex, the results remained
(EClipSE) strongly associated.
EClipSE harmonizes the neuropathological and longitudinal clini- EClipSE also investigated the significance of rarer and ‘disre-
cal data of brain donors from three population-based prospec- garded’ pathologies, such as Pick bodies, severe neuronal loss,
tive longitudinal studies of ageing in Europe that included a brain gliosis, and granulovacuolar degeneration, along with brainstem
donation programme: the CFAS (baseline 1989–1993), the CC75C plaques, tangles, neuronal loss, gliosis, pigmentary incontinence,
(baseline 1985), and the Vantaa 85+ (baseline 1991) study (EClipSE and Lewy bodies in relation to dementia in the population. A total
Collaborative Members, 2009) (<www.eclipsestudy.eu>). The of 627 individuals with clinical dementia were assessed. All pathol-
project was created to address the lack of statistical power within ogies were associated with dementia when controlling for plaques
individual studies for specific analyses assessing relationships and tangles, except Hirano bodies, granulovacuolar degeneration,
between data collected during life and neuropathology at death. and brainstem plaques, which shows that dementia in old age is
All three studies interviewed participants at regular intervals, gath- associated with a broad range of pathological and anatomical sub-
ering information on sociodemographic details and health status strates (Keage et al., 2012).
in addition to cognitive function, including dementia diagnosis. EClipSE: depression
Neuropathological parameters include neocortical and hippoc- Depression has not yet been studied within the EClipSE data
ampal neuritic plaques, diffuse plaques, tangles, cerebral amyloid resource.
EClipSE: summary on resources and care worldwide. Indeed, the cost of dementia is
The EClipSE study represents a unique resource of neuropathological significant, not only in terms of personal cost, but also on soci-
and longitudinal clinical data of brain donors in Europe. Data from etal resources (Alzheimer’s Disease International, 2011). As high-
three studies were combined: CFAS in England and Wales, CC75C in lighted in this chapter, many large population-based cohort studies
England, and the Vantaa 85+ study in Finland. The finding that edu- have been undertaken in the ageing population, worldwide and in
cation can mitigate the impact of pathology on the clinical expression Europe, in order better to identify patterns of disease and disease
of dementia combined with the understanding of lifecourse factors risk factors.
supports investments in health and education in early life. There is, however, no standardized approach to the diagnosis of
dementia, cognitive dysfunction, and depression. As a result, esti-
Newcastle 85+ study mates of disease prevalence and incidence that have been reported
The focus of the Newcastle 85+ study is the oldest old. The main across studies may vary, not due to true differences but as a result of
aim is to examine health trajectories and outcomes as the cohort methodology. Reported estimates could also vary across studies as
ages, and their associations with underlying biological, medical, a result of differences in the presence of risk and protective factors
and social factors. All individuals born in 1921 who were perma- across individuals and as a result of cultural variation. Such vari-
nently registered with a participating general practice in Newcastle ability is important to identify, especially for the development of
upon Tyne or North Tyneside primary care trusts in the UK were preventative strategies (such as those linked to modifiable factors
sampled (Collerton et al., 2007). The baseline sample will be fol- such as diet, educational exposure, and health-related comorbid-
lowed until the last participant has died. Of the 1,470 people eligible ity). Furthermore, the type of information collected and method
to participate, 1,042 participated, three having health assessment of assessment is variable across the studies. For example, some
only, 188 having general practice record review only, and 851 hav- studies report cardiovascular disease comorbidity (selfreported or
ing both (Collerton et al., 2009). objectively measured), brain-related changes (determined through
For most diseases, prevalence was determined on the basis of a neuroimaging), psychiatric-related comorbidity (e.g. behavioural
review of data from general practice records alone. The targeted and psychological symptoms of dementia), blood-based biomark-
common old age chronic diseases include: hypertension, ischaemic ers (including nutritional and genetic risk markers), and neuropa-
heart disease, cerebrovascular disease, peripheral vascular disease, thology. No single study has included all the measures required to
heart failure, atrial flutter or fibrillation, arthritis, osteoporo- completely map the full spectrum of risk/protective factors and
sis, chronic obstructive pulmonary disease or asthma, diabetes, the determinants of disease progression. Summarizing the results
hypothyroidism, hyperthyroidism, cancer diagnosed within the across all studies suggests that with the age demographic transi-
last 5 years, eye disease, dementia, Parkinson’s disease, and renal tion, the impact of dementia and age-related conditions is and will
impairment. continue to be considerable. Better methodology for extracting and
Newcastle 85+ study: dementia weighting data across studies will be important for future synthesis
Screening for dementia was based on the standardized MMSE of cross-study analysis and for informing the design of new cohort
score. Using the sMMSE score, moderate or severe cognitive studies focused on the life-course of ageing populations.
impairment was found in 12.5% (105/840) of participants, of whom
53% (56/105) had no diagnosis of dementia in their general prac- References
tice records. Therefore, the prevalence of undiagnosed cognitive 3C Study Group (2003). Vascular factors and risk of dementia: design of the
impairment was estimated at 7%. Three-City study and baseline characteristics of the study population.
Newcastle 85+ study: depression Neuroepidemiology, 22 (6), 316–25.
Aevarsson, O. and Skoog, I. (1996). A population-based study on the
Depression was assessed using the GDS. Similar results were dis-
incidence of dementia disorders between 85 and 88 years of age. Journal
closed for depression as for dementia: 8.4% of the participants of the American Geriatrics Society, 44 (12), 1455–60.
(65/772) had a GDS score suggestive of severe depression, of whom Aguero-Torres, H., et al. (1999). Mortality from dementia in advanced age:
82% (53/65) had no diagnosis recorded in their general practice a 5-year follow-up study of incident dementia cases. Journal of Clinical
records in the previous year. Epidemiology, 52 (8), 737–43.
Newcastle 85+ study: summary Ahtiluoto, S., et al. (2010). Diabetes, Alzheimer disease, and vascular
dementia: a population-based neuropathologic study. Neurology, 75
Overall, this large cohort of 85-year-olds showed good levels of (13), 1195–202.
both selfrated health and functional ability despite significant lev- Akbaraly, T.N., et al. (2011). Metabolic syndrome and onset of depressive
els of disease and impairment. Hypertension, ischaemic heart dis- symptoms in the elderly: findings from the three-city study. Diabetes
ease, atrial fibrillation, depression, and dementia may, however, be Care, 34 (4), 904–9.
underdiagnosed among this age group. Alperovitch, A., et al. (2002). Epidemiological studies on aging in France:
from the PAQUID study to the Three-City study. Comptes Rendus
Biologies, 325 (6), 665–72.
Conclusion Altman, D.G. and Bland, J.M. (2007). Missing data. British Medical Journal,
There is a continuous need for further epidemiological research 334 (7590), 424.
Alzheimer, A., et al. (1995). An English translation of Alzheimer’s 1907 paper,
in old age psychiatry as the world population is ageing.
‘Uber eine eigenartige Erkankung der Hirnrinde’. Clinical Anatomy,
Neuropsychiatric conditions, such as depression and dementia, 8 (6), 429–31.
cognitive impairment, and behavioural and functional decline, Alzheimer’s Disease International (2011). World Alzheimer Report 2011.
place a considerable onus on the health, social, and economic The benefits of early diagnosis and intervention. Alzheimer’s Disease
systems. Of particular importance is the potentially large impact International, London.
American Psychiatric Association (1980). Diagnostic and statistical manual Chadwick, C. (1992). The MRC Multicentre Study of Cognitive Function and
of mental disorders (DSM-III), 3rd edition. American Psychiatric Ageing: a EURODEM incidence study in progress. Neuroepidemiology,
Association, Washington, DC. 11 Suppl 1, 37–43.
American Psychiatric Association (1987). Diagnostic and statistical manual of Chatfield, M.D., Brayne, C.E., and Matthews, F.E. (2005). A systematic
mental disorders (DSM-III-R), 3rd review edition. American Psychiatric literature review of attrition between waves in longitudinal studies in the
Association, Washington, DC. elderly shows a consistent pattern of dropout between differing studies.
American Psychiatric Association (1994). Diagnostic and statistical manual Journal of Clinical Epidemiology, 58 (1), 13–19.
of mental disorders (DSM-IV), 4th edition. American Psychiatric Chen, R., et al. (2008). Severity of depression and risk for subsequent
Association, Washington, DC. dementia: cohort studies in China and the UK. British Journal of
Andersen, K., et al. (1997). Prevalence of very mild to severe dementia in Psychiatry, 193 (5), 373–7.
Denmark. Acta Neurologica Scandinavica, 96 (2), 82–7. Chou, K.L. (2008). Combined effect of vision and hearing impairment on
Andrews, G., Slade, T., and Peters, L. (1999). Classification in psychiatry: depression in older adults: evidence from the English Longitudinal Study
ICD-10 versus DSM-IV. British Journal of Psychiatry, 174, 3–5. of Ageing. Journal of Affective Disorders, 106 (1–2), 191–6.
Bachman, D.L., et al. (1992). Prevalence of dementia and probable senile Chui, H.C., et al. (1992). Criteria for the diagnosis of ischemic vascular
dementia of the Alzheimer type in the Framingham Study. Neurology, dementia proposed by the State of California Alzheimer’s Disease
42 (1), 115–19. Diagnostic and Treatment Centers. Neurology, 42 (3 Pt 1), 473–80.
Bachman, D.L., et al. (1993). Incidence of dementia and probable Alzheimer’s Collerton, J., et al. (2007). The Newcastle 85+ study: biological, clinical and
disease in a general population: the Framingham Study. Neurology, 43 psychosocial factors associated with healthy ageing: study protocol. BMC
(3 Pt 1), 515–19. Geriatrics, 7, 14.
Bengtsson, C., et al. (1973). The study of women in Gothenburg 1968–1969— Collerton, J., et al. (2009). Health and disease in 85 year olds: baseline
a population study. General design, purpose and sampling results. Acta findings from the Newcastle 85+ cohort study. British Medical Journal,
Medica Scandinavica, 193 (4), 311–18. 339, b4904.
Berrios, G.E. (1990). Alzheimer’s disease: a conceptual history. International Copeland, J.R., et al. (1976). A semi-structured clinical interview for the
Journal of Geriatric Psychiatry, 5 (6), 355–65. assessment of diagnosis and mental state in the elderly: the Geriatric
Blessed, G., et al. (1968). The association between quantitative measures Mental State Schedule. I. Development and reliability. Psychological
of dementia and of senile change in the cerebral grey matter of elderly Medicine, 6 (3), 439–49.
subjects. British Journal of Psychiatry, 114 (512), 797–811. Copeland, J.R., Dewey, M.E., and Griffiths-Jones, H.M. (1986). A computerized
Bogren, M., et al. (2007). Lundby revisited: first incidence of mental disorders psychiatric diagnostic system and case nomenclature for elderly subjects:
1947–1997. Australian and New Zealand Journal of Psychiatry, 41 (2), GMS and AGECAT. Psychological Medicine, 16 (1), 89–99.
178–86. Copeland, J.R., et al. (1999). Undifferentiated dementia, Alzheimer’s disease
Bond, J., et al. (2006). Self-rated health status as a predictor of death, functional and vascular dementia: age- and gender-related incidence in Liverpool.
and cognitive impairment: a longitudinal cohort study. European Journal The MRC-ALPHA Study. British Journal of Psychiatry, 175, 433–8.
of Ageing, 3 (4), 193–206. Cummings, J.L. and Benson, D.F. (1986). Dementia of the Alzheimer type.
Boothby, H., et al. (1994). The Gospel Oak Study stage III: the incidence of An inventory of diagnostic clinical features. Journal of the American
dementia. Psychological Medicine, 24 (1), 89–95. Geriatrics Society, 34 (1), 12–19.
Brayne, C. (1993). Research and Alzheimer’s disease: an epidemiological Dawber, T.R., Meadors, G.F., and Moore, F.E., Jr. (1951). Epidemiological
perspective. Psychological Medicine, 23 (2), 287–96. approaches to heart disease: the Framingham Study. American Journal of
Brayne, C., et al. (1992). The Cambridge Project for Later Life: design and Public Health and the Nations Health, 41 (3), 279–81.
preliminary results. Neuroepidemiology, 11 Suppl 1, 71–5. Department of Economic and Social Affairs—Population Division (2010).
Brayne, C., et al. (1997). Five-year incidence and prediction of dementia World population ageing 2009. United Nations, New York.
and cognitive decline in a population sample of women aged 70–79 at Department of Health Statistics and Informatics of World Health
baseline. International Journal of Geriatric Psychiatry, 12 (11), 1107–18. Organization (2004). The global burden of disease—2004 update. World
Brayne, C., McCracken, C., and Matthews, F.E. (2006). Cohort profile: the Health Organization, Geneva.
Medical Research Council Cognitive Function and Ageing Study (CFAS). Di Carlo, A., al. (2002). Incidence of dementia, Alzheimer’s disease, and
International Journal of Epidemiology, 35 (5), 1140–5. vascular dementia in Italy. The ILSA Study. Journal of the American
Brayne, C., et al. (2009). Neuropathological correlates of dementia in Geriatrics Society, 50 (1), 41–8.
over-80-year-old brain donors from the population-based Cambridge EClipSE Collaborative Members (2009). Cohort profile: Epidemiological
city over-75s cohort (CC75C) study. Journal of Alzheimer’s Disease, Clinicopathological Studies in Europe (EClipSE). Journal of Alzheimer’s
18 (3), 645–58. Disease, 18 (3), 659–63.
Brayne, C., et al. (2010). Education, the brain and dementia: neuroprotection Elias, P., Halstead, K., and Prady, K. (1993). Computer assisted standard
or compensation? Brain, 133 (Pt 8), 2210–16. occupational coding. HMSO, London.
Brayne, C., Stephan, B.C., and Matthews, F.E. (2011). A European perspective Erkinjuntti, T., et al. (1988). Accuracy of the clinical diagnosis of vascular
on population studies of dementia. Alzheimer’s Dementia, 7 (1), 3–9. dementia: a prospective clinical and post-mortem neuropathological
Breteler, M.M., et al. (1994). Cognitive correlates of ventricular enlargement study. Journal of Neurology, Neurosurgery and Psychiatry, 51 (8), 1037–44.
and cerebral white matter lesions on magnetic resonance imaging. The Erkinjuntti, T., et al. (1997). The effect of different diagnostic criteria on the
Rotterdam Study. Stroke, 25 (6), 1109–15. prevalence of dementia. New England Journal of Medicine, 337 (23),
Butler, R. and Brayne, C. (1998). Epidemiology. In: Butler, R., and Pitt, B. 1667–74.
(eds.) Seminars in old age psychiatry. Gaskell, London. Essen-Möller, E., et al. (1956). Individual traits and morbidity in a Swedish
Canadian Study of Health and Aging Working Group (1994). Canadian rural population. Acta Psychiatrica et Neurologica Scandinavica
study of health and aging: study methods and prevalence of dementia. Supplementum, 100, 1–160.
Canadian Medical Association Journal, 150 (6), 899–913. Ferri, C.P., et al. (2005). Global prevalence of dementia: a Delphi consensus
Cervilla, J.A., Prince, M., and Mann, A. (2000). Smoking, drinking, and study. Lancet, 366 (9503), 2112–17.
incident cognitive impairment: a cohort community based study Fillenbaum, G.G., et al. (1998). The prevalence and 3-year incidence of
included in the Gospel Oak project. Journal of Neurology, Neurosurgery dementia in older black and white community residents. Journal of
and Psychiatry, 68 (5), 622–6. Clinical Epidemiology, 51 (7), 587–95.
Fleming, J., et al. (2007). Cohort profile: the Cambridge City over-75s Cohort Jellinger, K.A. (2009). Lewy body/alpha-synucleinopathy in schizophrenia
(CC75C). International Journal of Epidemiology, 36 (1), 40–6. and depression: a preliminary neuropathological study. Acta
Folstein, M.F., Folstein, S.E., and McHugh, P.R. (1975). ‘Mini-mental state’. Neuropathologica, 117 (4), 423–7.
A practical method for grading the cognitive state of patients for the Joas, E., et al. (2012). Blood pressure trajectories from midlife to late life in relation
clinician. Journal of Psychiatric Research, 12 (3), 189–98. to dementia in women followed for 37 years. Hypertension, 59 (4), 796–801.
Fong, T.G., et al. (2011). Development and validation of a brief cognitive Jorm, A.F. and Jolley, D. (1998). The incidence of dementia: a meta-analysis.
assessment tool: the sweet 16. Archives of Internal Medicine, 171 (5), Neurology, 51 (3), 728–33.
432–7. Juster, F.T. and Suzman, R. (1995). An overview of the Health and Retirement
Fratiglioni, L., De Ronchi, D., and Aguero-Torres, H. (1999). Worldwide Study. Journal of Human Resources, 30 (5), S7–56.
prevalence and incidence of dementia. Drugs and Aging, 15 (5), 365–75. Katz, S., et al. (1970). Progress in development of the index of ADL. The
Fuhrer, R., Dufouil, C., and Dartigues, J.F. (2003). Exploring sex differences in Gerontologist, 10 (1), 20–30.
the relationship between depressive symptoms and dementia incidence: Kay, D.W., Beamish, P., and Roth, M. (1964). Old age mental disorders
prospective results from the PAQUID Study. Journal of the American in Newcastle upon Tyne. I. A study of prevalence. British Journal of
Geriatrics Society, 51 (8), 1055–63. Psychiatry, 110, 146–58.
Girling, D.M., et al. (1995). The prevalence of depression in a cohort of the Keage, H.A., et al. (2012). Impact of less common and ‘disregarded’
very elderly. Journal of Affective Disorders, 34 (4), 319–29. neurodegenerative pathologies on dementia burden in a population-based
Goldberg, D. and Huxley, P. (1980). Mental illness in the community: the cohort. Journal of Alzheimer’s Disease, 28 (2), 485–93.
pathway to psychiatric care. Tavistock Press, London. Klunk, W.E., et al. (2004). Imaging brain amyloid in Alzheimer’s disease with
Grundy, S.M., et al. (2005). Diagnosis and management of the metabolic Pittsburgh Compound-B. Annals of Neurology, 55 (3), 306–19.
syndrome: an American Heart Association/National Heart, Lung, and Knopman, D.S., et al. (2006). Incidence and causes of nondegenerative
Blood Institute scientific statement. Circulation, 112 (17), 2735–52. nonvascular dementia: a population-based study. Archives of Neurology,
Guehne, U., Riedel-Heller, S., and Angermeyer, M.C. (2005). Mortality in 63 (2), 218–21.
dementia. Neuroepidemiology, 25 (3), 153–62. Langa, K.M., et al. (2009). Cognitive health among older adults in the United
Gurland, B., et al. (1984). The SHORT-CARE: an efficient instrument for the States and in England. BMC Geriatrics, 9, 23.
assessment of depression, dementia and disability. Journal of Gerontology, Larson, E.B., et al. (2004). Survival after initial diagnosis of Alzheimer disease.
39 (2), 166–9. Annals of Internal Medicine, 140 (7), 501–9.
Hachinski, V. (1994). Vascular dementia: a radical redefinition. Dementia, 5 Launer, L.J., Brayne, C., and Breteler, M.M. (1992). Epidemiologic approach to
(3–4), 130–2. the study of dementing diseases: a nested case-control study in European
Harris, T.B., et al. (2007). Age, gene/environment susceptibility-Reykjavik incidence studies of dementia. Neuroepidemiology, 11 Suppl 1, 114–18.
study: multidisciplinary applied phenomics. American Journal of Launer, L.J., et al. (1999). Rates and risk factors for dementia and Alzheimer’s
Epidemiology, 165 (9), 1076–87. disease: results from EURODEM pooled analyses. EURODEM Incidence
Hebert, L.E., et al. (2010). Change in risk of Alzheimer disease over time. Research Group and Work Groups. European Studies of Dementia.
Neurology, 75 (9), 786–91. Neurology, 52 (1), 78–84.
Helgason, T. (1964). Epidemiology of mental disorders in Iceland. A Lawlor, B. (2002). Managing behavioural and psychological symptoms in
psychiatric and demographic investigation of 5395 Icelanders. Acta dementia. British Journal of Psychiatry, 181, 463–5.
Psychiatrica Scandinavica, 40 Suppl 173, 1+. Lawton, M.P. and Brody, E.M. (1969). Assessment of older people:
Helmer, C., et al. (2001). Mortality with dementia: results from a French self-maintaining and instrumental activities of daily living. The
prospective community-based cohort. American Journal of Epidemiology, Gerontologist, 9 (3), 179–86.
154 (7), 642–8. Letenneur, L., et al. (1993a). Cerebral and functional aging: first results on
Henderson, S. and Jablensky, A. (2010). The Lundby Study. Australian and prevalence and incidence of the PAQUID cohort. Methods of Information
New Zealand Journal of Psychiatry, 44 (1), 1–3. in Medicine, 32 (3), 249–51.
Hendrie, H.C., et al. (1993). Alzheimer’s disease is rare in Cree. International Letenneur, L., et al. (1993b). Prevalence of dementia in Gironde (France).
Psychogeriatrics, 5 (1), 5–14. Revue d‘Epidemiologie et de Sante Publique, 41 (2), 139–45.
Hendrie, H.C., et al. (1995). Prevalence of Alzheimer’s disease and dementia in Letenneur, L., et al. (1994). Incidence of dementia and Alzheimer’s disease
two communities: Nigerian Africans and African Americans. American in elderly community residents of south-western France. International
Journal of Psychiatry, 152 (10), 1485–92. Journal of Epidemiology, 23 (6), 1256–61.
Heyman, A., et al. (1991). Estimated prevalence of dementia among elderly Letenneur, L., et al. (1999). Are sex and educational level independent
black and white community residents. Archives of Neurology, 48 (6), predictors of dementia and Alzheimer’s disease? Incidence data from the
594–8. PAQUID project. Journal of Neurology, Neurosurgery and Psychiatry, 66
Hofman, A., et al. (1991). The prevalence of dementia in Europe: a (2), 177–83.
collaborative study of 1980–1990 findings. Eurodem Prevalence Research Linn, R.T., et al. (1995). The ‘preclinical phase’ of probable Alzheimer’s
Group. International Journal of Epidemiology, 20 (3), 736–48. disease. A 13-year prospective study of the Framingham cohort. Archives
Holmes, C., et al. (1999). Validity of current clinical criteria for Alzheimer’s of Neurology, 52 (5), 485–90.
disease, vascular dementia and dementia with Lewy bodies. British Livingston, G., et al. (1990a). The Gospel Oak Study stage II: the diagnosis of
Journal of Psychiatry, 174, 45–50. dementia in the community. Psychological Medicine, 20 (4), 881–91.
Hughes, C.P., et al. (1982). A new clinical scale for the staging of dementia. Livingston, G., et al. (1990b). The Gospel Oak Study: prevalence rates of
British Journal of Psychiatry, 140, 566–72. dementia, depression and activity limitation among elderly residents in
Huppert, F.A., Cabelli, S.T., and Matthews, F.E. (2005). Brief cognitive inner London. Psychological Medicine, 20 (1), 137–46.
assessment in a UK population sample—distributional properties and Livingston, G., et al. (2001). Mental health of migrant elders—the Islington
the relationship between the MMSE and an extended mental state study. British Journal of Psychiatry, 179, 361–6.
examination. BMC Geriatrics, 5, 7. Llewellyn, D.J., et al. (2008). Framingham Stroke Risk Profile and poor
Ikeda, M., et al. (2001). Increased prevalence of vascular dementia in Japan: a cognitive function: a population-based study. BMC Neurology, 8, 12.
community-based epidemiological study. Neurology, 57 (5), 839–44. Llibre Rodriguez, J.J., et al. (2008). Prevalence of dementia in Latin America,
Ikram, M.A., et al. (2011). The Rotterdam Scan Study: design and update up India, and China: a population-based cross-sectional survey. Lancet, 372
to 2012. European Journal of Epidemiology, 26 (10), 811–24. (9637), 464–74.
Lobo, A., et al. (2000). Prevalence of dementia and major subtypes in Europe: O’Connor, D.W., et al. (1989). The prevalence of dementia as measured
a collaborative study of population-based cohorts. Neurologic Diseases by the Cambridge Mental Disorders of the Elderly Examination. Acta
in the Elderly Research Group. Neurology, 54 11 Suppl 5, S4–9. Psychiatrica Scandinavica, 79 (2), 190–8.
Lobo, A., et al. (2007). Prevalence of dementia in a southern European Ott, A., et al. (1995). Prevalence of Alzheimer’s disease and vascular dementia:
population in two different time periods: the ZARADEMP Project. Acta association with education. The Rotterdam Study. British Medical Journal,
Psychiatrica Scandinavica, 116 (4), 299–307. 310 (6985), 970–3.
Maggi, S., et al. (1994). The Italian Longitudinal Study on Aging (ILSA): Ott, A., et al. (1998). Incidence and risk of dementia. The Rotterdam Study.
design and methods. Aging, 6 (6), 464–73. American Journal of Epidemiology, 147 (6), 574–80.
Magnusson, H. (1989). Mental health of octogenarians in Iceland. An Panza, F., et al. (2008). Impact of depressive symptoms on the rate of
epidemiological study. Acta Psychiatrica Scandinavica Supplementum, progression to dementia in patients affected by mild cognitive
349, 1–112. impairment. The Italian Longitudinal Study on Aging. International
Magnússon, H. and Helgason, T. (1993). The course of mild dementia in a Journal of Geriatric Psychiatry, 23 (7), 726–34.
birth cohort. International Journal of Geriatric Psychiatry, 8 (8), 639–47. Paraiso, M.N., et al. (2011). Prevalence of dementia among elderly people living
Mahoney, D.F., et al. (2005). African American, Chinese, and Latino in Cotonou, an urban area of Benin (West Africa). Neuroepidemiology,
family caregivers’ impressions of the onset and diagnosis of dementia: 36 (4), 245–51.
cross-cultural similarities and differences. The Gerontologist, 45 (6), Paykel, E.S., et al. (1994). Incidence of dementia in a population older than
783–92. 75 years in the United Kingdom. Archives of General Psychiatry, 51 (4),
Matthews, F. and Brayne, C. (2005). The incidence of dementia in England 325–32.
and Wales: findings from the five identical sites of the MRC CFA Study. Pfeiffer, E. (1975). A short portable mental status questionnaire for the
PLoS Medicine, 2 (8), e193. assessment of organic brain deficit in elderly patients. Journal of the
Matthews, F.E. (2005). Incidence estimation in studies of complex design: an American Geriatrics Society, 23 (10), 433–41.
example using the Medical Research Council Cognitive Function and Poels, M. and Last, J.M. (2008). A dictionary of epidemiology, 5th edition.
Ageing Study. University of Cambridge, Cambridge. Oxford University Press, New York.
Matthews, F.E., et al. (2008). Two-year progression from mild cognitive Poels, M.M., et al . (2010). Prevalence and risk factors of cerebral
impairment to dementia: to what extent do different definitions agree?. microbleeds: an update of the Rotterdam scan study. Stroke, 41 10
Journal of the American Geriatrics Society, 56 (8), 1424–33. Suppl, S103–6.
Matthews, F.E., et al. (2009). Epidemiological pathology of dementia: Polvikoski, T., et al. (1995). Apolipoprotein E, dementia, and cortical
attributable-risks at death in the Medical Research Council Cognitive deposition of beta-amyloid protein. New England Journal of Medicine,
Function and Ageing Study. PLoS Medicine, 6 (11), e1000180. 333 (19), 1242–7.
Matthews, F. E., et al. (2011). Full-scale scores of the Mini Mental State Polvikoski, T., et al. (2001). Prevalence of Alzheimer’s disease in very elderly
Examination can be generated from an abbreviated version. Journal of people: a prospective neuropathological study. Neurology, 56 (12),
Clinical Epidemiology, 64 (9), 1005–13. 1690–6.
Mattisson, C., et al. (2007). The long-term course of depressive disorders in Prina, A.M., et al. (2011). Co-occurrence of anxiety and depression amongst
the Lundby Study. Psychological Medicine, 37 (6), 883–91. older adults in low- and middle-income countries: findings from the
McDougall, F.A., et al. (2007). Prevalence of depression in older people in 10/66 study. Psychological Medicine, 41 (10), 2047–56.
England and Wales: the MRC CFA Study. Psychological Medicine, 37 Prince, M., et al. (2003). Dementia diagnosis in developing countries: a
(12), 1787–95. cross-cultural validation study. Lancet, 361 (9361), 909–17.
McKeith, I.G., et al. (1992). Operational criteria for senile dementia of Lewy Prince, M., et al. (2007). The protocols for the 10/66 dementia research group
body type (SDLT). Psychological Medicine, 22 (4), 911–22. population-based research programme. BMC Public Health, 7, 165.
McKeith, I.G., et al. (1996). Consensus guidelines for the clinical and pathologic Qiu, C., et al. (2001). The influence of education on clinically diagnosed
diagnosis of dementia with Lewy bodies (DLB): report of the Consortium dementia incidence and mortality data from the Kungsholmen project.
on DLB international workshop. Neurology, 47 (5), 1113–24. Archives of Neurology, 58 (12), 2034–9.
McKeith, I.G., et al. (2005). Diagnosis and management of dementia with Qiu, C., et al. (2010). Cerebral microbleeds, retinopathy, and dementia: the
Lewy bodies: third report of the DLB Consortium. Neurology, 65 (12), AGES-Reykjavik Study. Neurology, 75 (24), 2221–8.
1863–72. Radloff, L.S. (1977). The CES-D scale: a self-report depression scale for
McKhann, G., et al. (1984). Clinical diagnosis of Alzheimer’s disease: report research in the general population. Applied Psychological Measurement,
of the NINCDS-ADRDA Work Group under the auspices of Department 1, 385–401.
of Health and Human Services Task Force on Alzheimer’s Disease. Raffaitin, C., et al. (2009). Metabolic syndrome and risk for incident
Neurology, 34 (7), 939–44. Alzheimer’s disease or vascular dementia: the Three-City Study. Diabetes
Mirra, S.S., et al. (1991). The Consortium to Establish a Registry for Alzheimer’s Care, 32 (1), 169–74.
Disease (CERAD). Part II. Standardization of the neuropathologic Reitan, R.M (1992). Trail making test: manual for administration and scoring.
assessment of Alzheimer’s disease. Neurology, 41 (4), 479–86. Reitan Neuropsychology Laboratory, South Tucson, AZ.
Morris, M.C., et al. (1999). Methodological issues in the study of cognitive Rey, A. (1958). Léxamen clinique en psychologie. Presses Universitaire de
decline. American Journal of Epidemiology, 149 (9), 789–93. France, Paris.
Murray, C.J. (1994). Quantifying the burden of disease: the technical basis for Richards, M. and Brayne, C. (2010). What do we mean by Alzheimer’s
disability-adjusted life years. Bulletin of the World Health Organization, disease? British Medical Journal, 341, c4670.
72 (3), 429–45. Richards, M., et al. (2000). Cognitive function in UK community-dwelling
Neuropathology Group (2001). Pathological correlates of late-onset dementia African Caribbean and white elders: a pilot study. International Journal
in a multicentre, community-based population in England and Wales. of Geriatric Psychiatry, 15 (7), 621–30.
Neuropathology Group of the Medical Research Council Cognitive Rinder, L., et al. (1975). Seventy-year-old people in Gothenburg. A population
Function and Ageing Study (MRC CFAS). Lancet, 357 (9251), 169–75. study in an industrialized Swedish city. Acta Medica Scandinavica, 198
Noda, K., et al. (2006). Quantitative analysis of neurofibrillary pathology in (5), 397–407.
a general population to reappraise neuropathological criteria for senile Rocca, W.A., et al. (2011). Trends in the incidence and prevalence of
dementia of the neurofibrillary tangle type (tangle-only dementia): the Alzheimer’s disease, dementia, and cognitive impairment in the United
Hisayama Study. Neuropathology, 26 (6), 508–18. States. Alzheimer’s Dementia, 7 (1), 80–93.
Roman, G.C., et al. (1993). Vascular dementia: diagnostic criteria for The Italian Longitudinal Study on Aging Working Group (1997). Prevalence
research studies. Report of the NINDS-AIREN International Workshop. of chronic diseases in older Italians: comparing self-reported and clinical
Neurology, 43 (2), 250–60. diagnoses. International Journal of Epidemiology, 26 (5), 995–1002.
Rosow, I. and Breslau, N. (1966). A Guttman health scale for the aged. Journal The Medical Research Council Cognitive Function and Ageing Study (MRC
of Gerontology, 21 (4), 556–9. CFAS) (1998). Cognitive function and dementia in six areas of England
Roth, M., Tomlinson, B.E., and Blessed, G. (1966). Correlation between and Wales: the distribution of MMSE and prevalence of GMS organicity
scores for dementia and counts of ‘senile plaques’ in cerebral grey matter level in the MRC CFA Study. The Medical Research Council Cognitive
of elderly subjects. Nature, 209 (5018), 109–10. Function and Ageing Study (MRC CFAS). Psychological Medicine, 28 (2),
Roth, M., et al. (1986). CAMDEX. A standardised instrument for the 319–35.
diagnosis of mental disorder in the elderly with special reference to the Thomas, A.J., et al. (2001). A neuropathological study of vascular factors in
early detection of dementia. British Journal of Psychiatry, 149, 698–709. late-life depression. Journal of Neurology, Neurosurgery and Psychiatry,
Sachs-Ericsson, N., Plant, E.A., and Blazer, D.G. (2005). Racial differences 70 (1), 83–7.
in the frequency of depressive symptoms among community dwelling Tiemeier, H., et al. (2004). Relationship between atherosclerosis and late-life
elders: the role of socioeconomic factors. Aging and Mental Health, 9 (3), depression: the Rotterdam Study. Archives of General Psychiatry, 61 (4),
201–9. 369–76.
Sackett, D.L. (1979). Bias in analytic research. Journal of Chronic Diseases, 32 Townsend, P. (1979). Poverty in the United Kingdom. Pelican, London.
(1–2), 51–63. Tsopelas, C., et al. (2011). Neuropathological correlates of late-life depression
Saczynski, J.S., et al. (2010). Depressive symptoms and risk of dementia: the in older people. British Journal of Psychiatry, 198, 109–14.
Framingham Heart Study. Neurology, 75 (1), 35–41. Valenzuela, M.J. (2008). Brain reserve and the prevention of dementia.
Saunders, P.A., et al. (1992). Alpha: the Liverpool MRC Study of the incidence Current Opinions in Psychiatry, 21 (3), 296–302.
of dementia and cognitive decline. Neuroepidemiology, 11 Suppl 1, Vernooij, M.W., et al. (2008). Prevalence and risk factors of cerebral
44–7. microbleeds: the Rotterdam Scan Study. Neurology, 70 (14), 1208–14.
Saunders, P.A., et al. (1993). The prevalence of dementia, depression and Vernooij, M.W., et al. (2009a). Use of antithrombotic drugs and the presence
neurosis in later life: the Liverpool MRC-ALPHA Study. International of cerebral microbleeds: the Rotterdam Scan Study. Archives of Neurology,
Journal of Epidemiology, 22 (5), 838–47. 66 (6), 714–20.
Save the Children (2009). Hungry for change: an eight-step, costed plan of Vernooij, M.W., et al. (2009b). White matter microstructural integrity and
action to tackle global child hunger. Save the Children Fund, London. cognitive function in a general elderly population. Archives of General
Savva, G.M., et al. (2009). Age, neuropathology, and dementia. New England Psychiatry, 66 (5), 545–53.
Journal of Medicine, 360 (22), 2302–9. Wechsler, D. (1981). Wechsler Adult Intelligence Scale—revised . The
Saxton, J., et al. (2009). Functional and cognitive criteria produce different Psychological Corporation, New York.
rates of mild cognitive impairment and conversion to dementia. Journal Weir, D.R., et al. (2011). Reducing case ascertainment costs in U.S. population
of Neurology, Neurosurgery and Psychiatry, 80 (7), 737–43. studies of Alzheimer’s disease, dementia, and cognitive impairment—
Scheltens, P. and Rockwood, K. (2011). How golden is the gold standard of Part 1. Alzheimer’s Dementia, 7 (1), 94–109.
neuropathology in dementia? Alzheimer’s Dementia, 7 (4), 486–9. Wenger, C.G. (1989). Support networks in old age: constructing a typology.
Schneider, J.A., et al. (2009). The neuropathology of older persons with and In: Jeffreys, M. (ed.) Growing old in the 20th century, pp. 166–85.
without dementia from community versus clinic cohorts. Journal of Routledge, London.
Alzheimer’s Disease, 18 (3), 691–701. Wharton, S.B., et al. (2011). Epidemiological neuropathology: the MRC
Schulz, R., Drayer, R.A., and Rollman, B.L. (2002). Depression as a risk factor Cognitive Function and Aging Study experience. J Alzheimer’s Disease,
for non-suicide mortality in the elderly. Biological Psychiatry, 52 (3), 25 (2), 359–72.
205–25. White, L., et al. (2002). Cerebrovascular pathology and dementia in
Seshadri, S. and Wolf, P.A. (2007). Lifetime risk of stroke and dementia: autopsied Honolulu-Asia Aging Study participants. Annals of the New
current concepts, and estimates from the Framingham Study. Lancet York Academy of Sciences, 977, 9–23.
Neurology, 6 (12), 1106–14. Williams, J.G., et al. (2003). Performance and normative values of a concise
Seshadri, S., et al. (2006). The lifetime risk of stroke: estimates from the neuropsychological test (CAMCOG) in an elderly population sample.
Framingham Study. Stroke, 37 (2), 345–50. International Journal of Geriatric Psychiatry, 18 (7), 631–44.
Skoog, I., et al. (1993). A population-based study of dementia in 85-year-olds. Wilson, O., et al. (2003). Depressive symptoms, clinical AD, and cortical
New England Journal of Medicine, 328 (3), 153–8. plaques and tangles in older persons. Neurology, 61 (8), 1102–7.
Skoog, I., et al. (1996). 15-year longitudinal study of blood pressure and Wing, J.K., et al. (1967). Reliability of a procedure for measuring and
dementia. Lancet, 347 (9009), 1141–5. classifying ‘present psychiatric state’. British Journal of Psychiatry, 113
Snowdon, D.A., et al. (1996). Linguistic ability in early life and cognitive (498), 499–515.
function and Alzheimer’s disease in late life. Findings from the Nun Wolfson, C., et al. (2001). A reevaluation of the duration of survival after
Study. Journal of the American Medical Association, 275 (7), 528–32. the onset of dementia. New England Journal of Medicine, 344 (15),
Stephan, B.C., et al. (2012a). Neuropathological profile of mild cognitive 1111–16.
impairment from a population perspective. Alzheimer’s Disease and World Health Organization (1977). International classification of diseases:
Associated Disorders, 26(3), 205–12. manual of the international statistical classification of diseases, injuries,
Stephan, B.C., et al. (2012b). Alzheimer and vascular neuropathological and causes of death: based on recommendations of the Ninth Revision
changes associated with different cognitive states in a non-demented Conference, 1975. WHO, Geneva.
sample. Journal of Alzheimer’s Disease, 29 (2), 309–18. World Health Organization (1992). International classification of diseases,
Stern, Y. (2002). What is cognitive reserve? Theory and research application 10th review. WHO, Geneva.
of the reserve concept. Journal of the International Neuropsychological World Health Organization Expert Committee on Mental Health (1960).
Society, 8 (3), 448–60. Epidemiology of mental disorders: eighth report of the Expert Committee
Sweet, R.A., et al. (2004). Neuropathologic correlates of late-onset major on Mental Health, p. 29. WHO, Geneva.
depression. Neuropsychopharmacology, 29 (12), 2242–50. Wulsin, L.R., Vaillant, G.E., and Wells, V.E. (1999). A systematic review
Szklo, M. and Nieto, F.J. (2007). Epidemiology: beyond the basics. Jones and of the mortality of depression. Psychosomatic Medicine, 61 (1),
Bartlett, Burlington, MA. 6–17.
Xie, J., Brayne, C., and Matthews, F.E. (2008). Survival times in people with Yu, B. and Zhou, C. (2012). Assessing the accuracy of a multiphase diagnosis
dementia: analysis from population based cohort study with 14 year procedure for dementia. Journal of the Royal Statistical Society: Series C
follow-up. British Medical Journal, 336 (7638), 258–62. (Applied Statistics), 61 (1), 67–81.
Xuereb, J.H., et al. (2000). Neuropathological findings in the very old. Results from Zaccai, J., Ince, P., and Brayne, C. (2006). Population-based neuropathological
the first 101 brains of a population-based longitudinal study of dementing studies of dementia: design, methods and areas of investigation—a
disorders. Annals of the New York Academy of Sciences, 903, 490–6. systematic review. BMC Neurology, 6, 2.
Yesavage, J.A. (1988). Geriatric Depression Scale. Psychopharmacology Zilka, N. and Novak, M. (2006). The tangled story of Alois Alzheimer.
Bulletin, 24 (4), 709–11. Bratislavske Lekarske Listy, 107 (9–10), 343–5.
Yesavage, J.A., et al. (1982). Development and validation of a geriatric Zung, W.W. (1972). The Depression Status Inventory: an adjunct to the
depression screening scale: a preliminary report. Journal of Psychiatric Self-Rating Depression Scale. Journal of Clinical Psychology, 28 (4),
Research, 17 (1), 37–49. 539–43.
CHAPTER 6
Neuropathology
Johannes Attems and Kurt A. Jellinger
This chapter mainly describes the neuropathological lesions of What are the causes and mechanisms of neuronal dysfunction
age-associated neurodegenerative diseases and the related patho- and cell death in neurodegeneration? Apoptosis or ‘programmed
logical classifications of these, to assist the reader in interpreting cell death’ is an attractive mechanism to explain selective neuro-
neuropathological reports. It is beyond the scope of this chapter to nal vulnerability (Dickson, 2011), while the term necrosis refers to
provide a comprehensive review of the neuropathology of all neu- the morphological changes that take place in a living organism in
rodegenerative diseases. Cerebrovascular disease, which is com- response to a noxious stimulus that causes cell death. In the cen-
mon and strongly age-associated, is covered and the chapter also tral nervous system (CNS), necrosis is commonly observed after
reviews the increasing evidence on cerebral multimorbidity, that is, trauma, infection, and infarction, conditions that by definition do
the presence of multiple pathologies in post-mortem brains. It is not fall into the category of neurodegeneration.
now recognized that multimorbidity in the aged brain is rather the Apoptosis, on the other hand, is regulated by extrinsic (receptor-
rule, while pure pathologies are the exception. Finally, we give a mediated) or intrinsic (mitochondria-mediated) pathways that via
view on new and currently emerging neuropathological methods the activation of caspases lead to degradation of both DNA and
that should more accurately reflect the burden of pathology seen in cytoskeletal proteins (Wyllie, 1997). In several neurodegenerative
post-mortem brains. diseases, damage resulting from reactive oxygen species has been
identified as a major cytopathological feature, suggesting that oxi-
dative stress plays an important role in neurodegeneration (Sayre
Neurodegeneration: General Considerations et al., 2001). Cells that fail to compensate for oxidative stress enter
Neurodegenerative diseases are characterized by progressive neu- apoptosis (Perry et al., 1998). However, apoptosis leads to cell
ronal dysfunction with consecutive neuronal loss of specific popu- death within hours, while neurodegenerative diseases typically
lations of neurons that frequently involve distinct anatomically have a course of years, suggesting increased compensatory cellu-
related systems. Thus, selective neuronal vulnerability is a character- lar response in neurodegeneration in living organisms. In addition,
istic feature of neurodegenerative diseases. It should be emphasized failure in the maintenance of mitochondria has been suggested to
that diseases with known vascular, toxic, metabolic, infectious, or play a role in neurodegeneration (for review see Karbowski and
immunologically determined causes are by definition not classified Neutzner, 2012).
as (primary) neurodegenerative diseases. The clinical presentation While the respective roles of apoptosis, oxidative stress, and
of neurodegenerative diseases depends on the system or region mitochondria in neurodegeneration remain unclear, aggregation of
affected by the disease rather than on the molecular nature of the misfolded proteins (that might be neurotoxic) is a well described
characteristic neuropathological lesion per se; e.g. severe neuronal and unifying feature of neurodegeneration. Various proteins (e.g.
loss in the substantia nigra will clinically present as parkinsonism amyloid-β, tau) lose their native structure and form fibrils rich in
irrespective of the associated neuropathological lesion that poten- β-sheets that accumulate intra- or extracellularly (for review see
tially could be aggregates of α-synuclein (Lewy bodies), hyperphos- Jellinger, 2010, 2012a). Indeed, the nature of the respective mis-
phorylated tau (neurofibrillary tangles), cerebrovascular lesions, or folded protein in conjunction with the topographical localization
other pathologies. Admittedly, the most likely neuropathological of the affected regions is the basis for the classification of neurode-
lesion associated with clinical parkinsonism is α-synuclein deposi- generative diseases (Table 6.1).
tion in the substantia nigra, in which case the neuropathological Despite considerable advances in our understanding of some
diagnosis would be Lewy body disease with brainstem preponder- patho-mechanisms that ultimately lead to sporadic age-associated
ance, i.e. Parkinson’s disease (Dickson et al., 2009; Jellinger, 2011c). neurodegeneration, the exact causes for protein accumulation await
On the other hand, in patients with advanced Alzheimer’s disease further elucidation and are likely to be multifactorial. However, in
(AD) the substantia nigra is often affected by tau pathology, result- hereditary neurodegenerative diseases the mechanisms that cause
ing in clinical parkinsonism (Attems et al., 2007), and vascular accumulation of misfolded proteins are in most cases reasonably
lesions in the substantia nigra are well recognized as causing vascu- well understood, thereby providing helpful information to fur-
lar parkinsonism, which is a rare form of parkinsonism (Jellinger, ther our understanding of patho-mechanisms in sporadic neuro-
2008). degeneration; e.g. in trisomy 21 (Down syndrome) an increased
Table 6.1 Major protein aggregates in neurodegenerative diseases

Disease Protein aggregate Characteristic form Localization
Alzheimer’s disease Tau (3R, 4R) NFT, NT Neuronal cell bodies (NFT) and processes (NT)
Aβ (1–40, 1–42) Aβ-plaque Extracellular
Aβ (1–40, 1–42) and tau (3R, 4R) Neuritic plaque Aβ, extracellular; tau, neuronal processes
Lewy body diseases
Parkinson’s disease α-Synuclein LB, LN Neuronal cell bodies (LB) and processes (LN)
Dementia with Lewy bodies α-Synuclein LB, LN Neuronal cell bodies (LB) and processes (LN)
Multiple system atrophy α-Synuclein GCI Cytoplasm of glial cells
Frontotemporal lobar degeneration
Pick’s disease Tau (3R) Pick bodies Neuronal cell bodies
Corticobasal degeneration Tau (4R) Astrocytic plaque Distal segments of astrocytes
Progressive supranuclear palsy Tau (4R) Globose NFT Neuronal cell body
Tau (4R) Tufted astrocyte Astrocytic cell body
Agyrophilic grain disease Tau (4R) Grains Neuronal processes (dendrites)
Neurofibrillary tangle dominant dementia Tau (3R, 4R) NFT, NT Neuronal cell bodies (NFT) and processes (NT)
FTLD-TDP TDP-43 NCI (NII) Neuronal cytoplasm (and nuclei)
FTLD-UPS Ubiquitin NCI (NII) Neuronal cytoplasm (and nuclei)
FTLD-FUS FUS protein NCI (NII) Neuronal cytoplasm (and nuclei)
FTLD-ni None known NA NA
3R, 3 repeat tau; 4R, 4 repeat tau; FTLD, frontotemporal lobar degeneration (for various forms of FTLD see main text); GCI, glial cytoplasmic inclusions; LB, Lewy body; LN, Lewy neurite; NA,
not applicable; NCI, neuronal cytoplasmic inclusion; NFT, neurofibrillary tangle; NII, neuronal intanuclear inclusions; NT, neuropil thread.
expression of the amyloid precursor protein (APP) that is encoded structures can be seen on microscopic examination. However, it
on chromosome 21 leads to increased production of amyloid-β is not clear if the soluble oligomers /and/or the insoluble micro-
(Aβ) with consecutive formation of Aβ plaques. However, another scopically visible aggregates are the neurotoxic species. This has
possibility for increased production of amyloid-β in both sporadic important implications with respect to the interpretation of neu-
and hereditary (e.g. mutations in presenilin genes 1 and 2) neuropathological findings regarding disease mechanisms, as neu-
rodegenerative diseases and ageing might be increased activity ropathology detects insoluble aggregated proteins but not the
of β- and γ-secretase, which leads to the increased formation of soluble oligomers. It has indeed been suggested that the forma-
amyloid-β oligomers (for more details on amyloid processing see tion of insoluble aggregates represents a protective mechanism,
section Alzheimer’s disease). insofar as neurotoxic oligomers are ‘trapped’ in their aggregated
On the other hand, a decrease in protein degradation /and/or form and thereby lose their neurotoxic property. Alternatively,
elimination out of the brain may cause protein accumulation and the accumulation and aggregation of proteins might primarily
aggregation; cellular protein degradation and clearance is medi- be the result of synaptic /and/or cellular damage, rather than the
ated via both the ubiquitin-proteasome system (UPS; ubiquitin cause for cell death.
pathway) and the autophagy-lysosomal pathway (ALP). Failure of In most age-associated neurodegenerative diseases the amount
UPS and ALP might be caused by increasing amounts of proteins of pathology is crucial for the development of clinical symptoms. In
that overwhelm these systems /and/or malfunction of the systems particular, AD pathology is frequently present to a limited extent in
themselves, both probably resulting in protein misfolding and brains of nondemented older people. It is therefore still a matter of
aggregation. Similarly, reduced elimination out of the brain could discussion whether age-associated neurodegenerative diseases are
lead to accumulation of the respective protein; the CNS is devoid of a pure consequence of ageing rather than a distinct disease, or if
lymphatic vessels and solutes are partly transported out of the brain age is an important risk factor but would not inevitably lead to the
via drainage along the perivascular pathway, and impairment of the development of full-blown disease with overt clinical symptoms,
perivascular pathway has been suggested to result in Aβ accumula- even in individuals aged over 100 years.
tion (Weller et al., 2009). Neurodegeneration per se usually does not lead to fatal brain
It is generally assumed that misfolded proteins exert neuro- lesions that would cause the death of a patient, but naturally can
toxicity, but the underlying mechanisms are poorly understood. be considered as severe underlying disease that predisposes indi-
Misfolded proteins form soluble oligomers that aggregate into viduals to acquire potentially fatal diseases. Bronchopneumonia
insoluble structures, and by using appropriate histochemical most frequently represents the immediate cause of death in AD
techniques and immunohistochemical antibodies these insoluble (approximately 50%), while in patients with vascular dementia,
CHAPTER 6 neuropathology 89
cardiovascular disease has been shown to be the immediate cause In AD, neuronal loss has been described, in particular in the
of death in approximately 50% (Attems et al., 2005b). nucleus basalis of Meynert, the amygdala, olfactory bulb with
olfactory nucleus, locus ceruleus, and serotonergic raphe nuclei.
Alzheimer’s Disease Another important feature of AD is synaptic loss and there is strong
evidence that synapses play a major pathophysiological role in AD
AD is the most frequent neurodegenerative disease, accounting for (for review see Duyckaerts et al., 2009). Band-like spongiosis and
50–80% of dementias. Both prevalence and incidence increase with reactive astrocytosis may be present in AD. Despite the fact that
age; in 65- to 69-year-olds the prevalence is 1%, doubling with each the aforementioned lesions are of pathophysiological importance
subsequent 5-year increment. Above age 85 years prevalence rates for AD (neuronal and synaptic loss) or frequent findings (spongi-
range from 20% to 50%. Thereby, increasing age is the most impor- osis), they are not used for neuropathological diagnostic purposes,
tant risk factor for AD (see Chapters 31 and 33 for details and other which rely on the assessment of NFT/NT and Aβ plaques including
risk factors). neuritic plaques.
The neuropathological diagnosis of AD is based on the semiquan-
titative and topographical assessment of Aβ plaques, neurofibrillary Neurofibrillary tangles and neuropil threads
tangles/neuropil threads (NFT/NT), and neuritic plaques (NP). NFT and NT are aggregates of hyperphosphorylated microtubule-
However, additional neuropathological lesions such as cerebral amy- associated protein tau (MAP tau). MAP tau is particularly abun-
loid angiopathy (CAA) and granulovacuolar degeneration (GVD) are dant in axons and its physiological function is primarily to stabilize
typically present, but the assessment of those lesions is not required microtubules. Six isoforms of MAP tau exist in the human brain
for stating a neuropathological diagnosis, according to current diag- and, based on the number of C terminal microtubule-binding
nostic criteria. The loss of neurons (Gomez-Isla et al., 1996) and syn- repeat motifs, each 3 repeat (3R) and 4 repeat (4R) isoforms can
apses (Terry et al., 1991) in AD has been shown to be more directly be distinguished (for review see Mandelkow and Mandelkow,
related to the severity of the cognitive deficit, but assessing neuronal 2012). In a dephosphorylated state MAP tau binds to microtu-
and synaptic loss is technical and time consuming and it is currently bules, while phosphorylation leads to detachment from microtu-
not possible to evaluate those on a routine basis. bules. Under physiological conditions MAP tau is in a dynamic
equilibrium, on and off the microtubules, but under pathological
Neuropathology of AD conditions (neurodegeneration) MAP tau is hyperphosphorylated,
On gross examination, post-mortem brains of AD patients often resulting in the disintegration of microtubules and aggregation of
appear relatively normal without striking pathological changes, and insoluble filaments/fibrils composed of hyperphosphorylated MAP
reduction in brain weight may be minimal. However, if present, the tau (for review see Ballatore et al., 2007). NFT/NT are composed
characteristic atrophy in AD involves the entorhinal cortex, hip- of aggregates of hyperphosphorylated MAP tau, while so-called
pocampus, amygdala, and olfactory bulb, as well as the inferior pre-tangles represent nonaggregated hyper- or abnormally phos-
temporal, superior frontal, and middle frontal gyri (Halliday et al., phorylated MAP tau and are considered to be precursors of NFT/
2003) (Fig. 6.1). In all of these areas, tau pathology predominates NT. Importantly, the extent of NFT/NT in the neocortex has been
and atrophy in AD is not associated with the burden of Aβ plaques shown to correlate with cognitive deficits (for review see Nelson
(Josephs et al., 2008). Cerebral atrophy leads to enlargement of the et al., 2012).
ventricles (i.e. hydrocephalus internus e vacuo) and the cortical rib- Using appropriate immunohistochemical antibodies (e.g. AT8),
bon may be thin. The substantia nigra is well pigmented while the both NFT and NT can be visualized for histological assessment
locus ceruleus might be pale and the cerebellum is normal. (Fig. 6.2). Of note, in AD both 3R and 4R tau isoforms are present,
while some tauopathies are characterized by the exclusive presence
of either 3R or 4R tau (for details see section Frontotemporal lobar
degenerations). NFT are located in neuronal cell bodies and NT in
axons and dendrites, the latter appearing in immunohistochemi-
cally stained sections as immunopositive threads in the neuropil.
It has been suggested recently that the amount of NT rather than
NFT should be assessed for diagnostic purposes (Alafuzoff et al.,
2008). NFT and NT are predominantly found in the hippocam-
pus, the entorhinal cortex, and in layers III and VI of the isocortex.
However, NFT and NT are also present in the olfactory bulb and
subcortical nuclei, in particular the locus ceruleus, in early, preclin-
ical stages of AD (Attems et al., 2005c, 2012; Braak et al., 2011).
It is generally assumed that in AD, NFT and NT primarily mani-
fest in transentorhinal and entorhinal cortices and then gradually
spread towards the hippocampus and isocortex (Braak and Braak,
1991), and this pattern is the basis for diagnostic Braak stages (see
Fig. 6.1 Comparison between formalin-fixed brain slices of the left hemispheres section Neuropathological diagnostic criteria for AD). However,
(level of posterior hippocampus) of an aged nondemented individual (A) and an
AD patient (B). Note the marked atrophy (thinning of the gyri and deepening of
recently pretangle material (i.e. hyperphosphorylated but nonaggre-
the sulci) in B, in particular hippocampal atrophy (arrow in B) with widening of gated tau) was found in the locus ceruleus in the majority of individ-
the inferior horn of the second ventricle (asterisk in B). (Photographs by courtesy uals under 30 years of age who were devoid of cortical tau pathology,
of Simon Fraser and Arthur Oakley.) See also Plate 1. suggesting that tau pathology begins in the locus ceruleus rather
Fig. 6.2 In AD, high amounts of neurofibrillary tangles and neuropil threads are seen in the hippocampus (A), among other regions (see main text). Arrows in B
mark a neuronal cell body that comprises a neurofibrillary tangle, while ovals encircle some of the many neuropil threads in neuronal processes that are seen in this
photomicrograph. CA1, CA2, and CA4 hippocampal cornu ammonis (Ammon’s horn) sectors 1, 2, and 3, respectively; GR, granule cell layer of the dentate gyrus.
Immunohistochemistry with antibody against hyperphosphorylated tau AT8. Scale bars: A, 50 μm; B, 10 μm. (Modified from Montine et al., 2011.) See also Plate 2.
than the transentorhinal cortex (Braak and Del Tredici, 2011; Braak are usually large (up to several 100 μm) and show ill-defined bor-
et al., 2011), but this is controversial (Attems et al., 2012). ders; they are referred to as ‘lake-like’, ‘fleecy’, and ‘subpial band-like’.
Aβ plaques, on the other hand, are focal Aβ deposits; ‘mature’, ‘clas-
Amyloid-β plaques
sical’, and ‘neuritic’ plaques can be distinguished.
Aβ peptides (4 kDA) are generated by β- and γ-secretase cleavage It should be emphasized that neuritic plaques contain—in
of the transmembranous amyloid precursor protein (APP) and addition to Aβ aggregates—hyperphosphorylated tau in dys-
comprise peptides terminating at carboxyl terminus 40 (Aβ40) and trophic neurites (Fig. 6.4). Neuritic plaques are strongly associ-
42 (Aβ42) (Glenner and Wong, 1984). Cerebral Aβ depositions are ated with AD, while other forms of parenchymal Aβ deposits are
a hallmark of AD, but in the majority of clinicopathological correla- frequently seen to a considerable extent in post-mortem brains
tive studies, cerebral Aβ load does not correlate with clinical demen- of nondemented individuals (Jellinger and Attems, 2012) and,
tia (Nelson et al., 2012). Nevertheless, most therapeutic approaches as mentioned above, their density does not correlate with clini-
against AD target Aβ. This is mainly due to the so-called amyloid cal dementia (for review see Duyckaerts and Dickson, 2011).
cascade hypothesis that postulates that an increase in Aβ is the However, it has been suggested that Thal Aβ phases 4 and 5 are
trigger for a series of events that ultimately lead to the formation correlated with clinical dementia (see section Neuropathological
of NFT/NT (Hardy and Higgins, 1992; Hardy and Selkoe, 2002). diagnostic criteria for AD).
The observation that mutations in genes that lead to an increase in
Aβ (APP, presenilins) cause familial AD with tau pathology (NFT/
NT), while mutations that lead to tau pathology cause familial Additional lesions in AD
tauopathies without Aβ, argue in favour of the ‘amyloid cascade Although the neuropathological diagnosis of AD solely relies
hypothesis’. However, several observations suggest that the ‘amy- on the assessment of both Aβ and tau pathology, a variety of
loid hypothesis’ might not apply for sporadic, age-associated AD; in
early stages of AD, Aβ is mainly present in neocortical areas, while
tau pathology is seen in the transentorhinal cortex, and a large
autopsy study analyzing the prevalence of Aβ and tau pathology
as a function of age demonstrated that tau pathology precedes Aβ
(Duyckaerts and Hauw, 1997). Also, APP transgenic mice develop
abundant Aβ but no tau pathology (for review see Duyckaerts
et al., 2009). However, the frequent co-occurrence of Aβ and tau
strongly suggests a mutual interaction that awaits further elucida-
tion. Although both Aβ and tau have been extensively studied with
regard to their separate mode of toxicity (for review see Hardy,
2003), more recently light has been shed on their possible interac-
tions and synergistic effects in AD, linking Aβ and tau (Gotz et al.,
2010; Ittner and Gotz, 2010).
While Aβ40 constitutes the majority of cerebral Aβ (over 95%), Aβ42
is more aggregatable and hence believed to initiate the formation of
oligomers, fibrils, and plaques (Naslund et al., 1994; Younkin, 1995;
Masters and Beyreuther, 2011). Aβ aggregates extracellularly and, Fig. 6.3 Aβ depositions in the neocortex (A–C) show different morphology:
using appropriate antibodies (e.g. 4G8 antibody), various morpho- subpial band-like Aβ (arrows in A1), fleecy Aβ (arrowheads in B), and Aβ plaques
logical forms of Aβ aggregates can be detected (Fig. 6.3). The main (C). Immunohistochemistry with Aβ antibody 4G8. Scale bars: A, 500 μm; B,
parenchymal Aβ deposits are diffuse or focal. Diffuse Aβ deposits 20 μm; C, 50 μm. See also Plate 3.
Fig. 6.4 Neuritic plaques are a neuropathological hallmark lesion of AD and represent Aβ plaques that contain tau in distended neuronal processes (i.e. dystrophic
neurites). A–C are photomicrographs from adjacent histological sections. A, Gallyas silver stain visualizes both aggregated Aβ and tau and is therefore ideal to detect
neuritic plaques (ring in A1, neuritic plaque; arrow in A2, dystrophic neurite; arrowhead in A2, neurofibrillary tangle). On the other hand, the combined use of tau
(B, antibody AT8) and Aβ (C, antibody 4G8) immunohistochemistry on adjacent sections is also useful to detect neuritic plaques, as tau immunopositivity shows a
plaque-like pattern (ring in B) and Aβ immunostaining confirms the presence of Aβ plaques (C). Scale bars: A, B, and C, 200 μm. See also Plate 4.
additional neuropathological lesions are usually present. Indeed, Neuropathological diagnostic criteria for AD
the presence of multiple pathologies in brains of elderly demented
With the progression of the disease, the major neuropathological
(and nondemented) individuals is not an exceptional finding
lesions of AD progress stepwise, following a hierarchical pattern:
and this is addressed in detail in the final section of this chapter
on cerebral multimorbidity. However, there are some additional 1. Braak neurofibrillary stages describe the progression of NFT/NT
pathologies that are particularly frequent in AD; among those are from the transentorhinal and entorhinal areas (stages I and II) to
cerebral amyloid angiopathy (CAA), cerebrovascular lesions, hip- hippocampus (stage III), temporal cortex (stage IV) and finally
pocampal sclerosis, and granulovacuolar degeneration (GVD) (for other neocortical areas of which the occipital cortex (stages V and
review see Duyckaerts and Dickson, 2011). Although GVD may VI) is used for staging purposes (Braak and Braak, 1991; Braak
be present in diseases other than AD, it is regarded as a prima- et al., 2006).
rily AD-related phenomenon and therefore described in the next 2. Thal Aβ phases describe the progression of parenchymal Aβ
section, while the other pathologies are described in their own depositions; in phase 1 the isocortex is involved, the hippocam-
respective sections. pus and entorhinal cortex in phase 2, striatum and diencephalic
nuclei in phase 3, brainstem nuclei in phase 4, and finally the
Granulovacuolar degeneration cerebellum and additional brainstem nuclei in phase 5; phases 4
The term GVD describes vacuolar changes in the cytoplasm of and 5 are suggested to be correlated with clinical dementia (Thal
neurons of the medial temporal lobe, in particular in pyrami- et al., 2002b).
dal neurons of the hippocampus, often in association with NFT,
but they have also been described in other regions (e.g. neocor- Several criteria for the neuropathological diagnosis of AD are in use:
tex, amygdala). GVD has been reported to be more frequent in ◆ Age-adjusted criteria of the Consortium to Establish a Registry
AD compared to controls and may be present in tauopathies (e.g.
for Alzheimer’s Disease (CERAD) are based on the semiquantita-
Pick’s disease, progressive supranuclear palsy). It has been dem-
tive assessment of neuritic plaques (sparse, moderate, frequent)
onstrated recently that GVD first develops in neurons of the CA1/
in middle frontal gyrus, superior/middle temporal gyri, and infe-
CA2 hippocampal subfield and then expands in a predictable
rior parietal lobule (Mirra et al., 1991).
sequence into other brain regions, allowing for the distinction in
five different stages (Thal et al., 2011). Importantly, these stages are ◆ The National Institute of Aging Nancy and Ronald Reagan
related to AD pathology, pointing towards a role of GVD in AD Institute Criteria (NIA-RI Criteria) give a probability that a
pathogenesis. clinical dementia has been caused by AD and are based on both
CERAD criteria and Braak neurofibrillary stages: i.e. CERAD cause microbleeds (Greenberg et al., 2009) and in moderate to severe
neg. and Braak neg., no probability; CERAD A and Braak I/II, stages has been shown to be an independent risk factor for cognitive
low probability; CERAD B and Braak III/IV, medium probabil- impairment (Matthews et al., 2009).
ity; CERAD C and Braak V/VI, high probability (Hyman, 1998).
Histopathology of CAA
◆ Recently, new NIA–Alzheimer’s Association guidelines that com-
bine Thal Aβ phases, Braak neurofibrillary stages, and CERAD Aβ is initially deposited in the outer parts of the vessel wall, but
criteria have been proposed (Hyman et al., 2012; Montine et al., with increasing severity all layers of the vessel wall show Aβ deposi-
2012). According to these criteria, a neuropathological report tions accompanied by a loss of smooth muscle cells. In very severe
should state the Aβ plaque score (Thal et al., 2002b), Braak neu- stages of CAA the vascular architecture is disrupted, but endothe-
rofibrillary stage (Braak and Braak, 1991, Braak et al., 2006), and lial cells are usually preserved (Fig. 6.5).
CERAD neuritic plaque score (Mirra et al., 1993), which yield CAA may frequently show a patchy distribution and primarily
a combined ‘ABC score’. This ABC (Amyloid, Braak, CERAD) affects leptomeningeal and cortical vessels of neocortical regions
score should be stated regardless of clinical history, to reflect the (Thal et al., 2003). The occipital lobe has been reported to be the site
amount of ‘Alzheimer Disease Neuropathologic Change’. Table 6.2 that is both most frequently and severely affected by CAA, followed
illustrates how each A, B, and C score is transformed to state the by either frontal, temporal, or parietal lobes (Tomonaga, 1981;
level of AD neuropathological change on a four-tiered scale: ‘Not, Vinters and Gilbert, 1983; Pfeifer et al., 2002; Attems et al., 2005a).
Low, Intermediate, and High’. The NIA-AA guidelines for the neu- Aβ depositions in the walls of capillaries are referred to as capil-
ropathological assessment of AD await further validation. lary CAA (capCAA) and usually present as strong staining lining
capillary walls in sections stained with appropriate Aβ antibod-
ies. The presence of capCAA distinguishes two types of CAA:
Cerebral Amyloid Angiopathy CAA-type 1 is characterized by the presence of capCAA and may
CAA is defined as the deposition of a congophilic material (i.e. positive show additional Aβ depositions in noncapillary blood vessels,
staining with a Congo-red dye) in cerebral leptomeningeal and intra- whereas in CAA-type 2, Aβ depositions are restricted to leptome-
cortical arteries, arterioles, capillaries, and, rarely, veins. CAA is also ningeal and cortical arteries, arterioles, and, rarely, veins without
referred to as congophilic amyloid angiopathy (for review see Vinters, capillary involvement (Thal et al., 2002a). Of note, the frequency
1987; Attems et al., 2011a). CAA occurs in both hereditary or famil- of the APOEε4-allele in CAA-type 1 is more that four times higher
ial and sporadic forms. While the amyloid in sporadic, age-associated than in CAA-type 2 (Thal et al., 2002a). It has been shown recently
forms is predominantly composed of Aβ, the nature of amyloid depo- that the presence of capCAA identifies a subtype of sporadic AD
sitions in hereditary forms is determined by the respective underly- that is defined by characteristic neuropathological features and
ing mutation (e.g. A-Bri and cystatin C in BRI2 and cystatin C gene genotype-specific associations (Thal et al., 2010). Pericapillary Aβ
mutations, respectively). Sporadic CAA is frequently but not invari- deposits are predominately composed of Aβ42 and are clustered in
ably present in AD and is often observed in older individuals with- the glia limitans around capillaries and represent another type of
out AD (Chui et al., 1992; Greenberg, 1998; Jellinger, 2002; Attems capillary involvement (Attems et al., 2004, 2010).
and Jellinger, 2004). CAA is considered a risk factor for nontraumatic As of today, no standardized neuropathological consensus criteria
intracerebral lobar haemorrhage (ICH) in the elderly and is present in for the scoring of CAA have been established, but several methods
5–20% of all cases with ICH (Pezzini et al., 2009), but in large autopsy have been published and are usually used to describe the severity of
cohorts the prevalence of ICH in CAA cases was approximately 5%, CAA in post-mortem brains (Vonsattel et al., 1991; Olichney et al.,
being similar to cases without CAA (Attems et al., 2008). CAA may 1995; Thal et al., 2003; Attems et al., 2005a; Chalmers et al., 2009).
Table 6.2 ABC criteria for the diagnosis of Alzheimer’s disease (AD)-related pathology. The level of AD neuropathological change is determined
by assessing A, B, and C scores. A scores are related to Thal phase for Aβ-plaques (first column), B scores to neuritic Braak stages (bottom row), and
C scores to Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) stages (last column)
Level of AD neuropathologic change
Thal phase A B C CERAD
for Aβ-plaques 0 or 1 2 3
0 0 Not Not Not 0 Neg
1 or 2 1 Low Low Low 0 or 1 Neg or A
1 or 2 1 Low Intermediate Intermediate 2 or 3 A or B
3 2 Low Intermediate Intermediate any C Neg or A to C
4 or 5 3 Low Intermediate Intermediate 0 or 1 Neg or A
4 or 5 3 Low Intermediate High 2 or 3 B or C
Braak 0–II Braak III–IV Braak V–VI
(Modified from Montine et al., 2011.)
Fig. 6.5 Cerebral amyloid angiopathy (CAA) presents as Aβ depositions in the walls of meningeal (arrows in A) and cortical (arrowheads in A) arteries, while capillary
CAA refers to Aβ depositions in the walls of capillaries (arrowheads in B). CAA may be the cause for nontraumatic intracerebral haemorrhage (ICH). Photomicrograph
C shows an ICH (left upper part) in a patient with CAA; insets in C show photomicrographs of sections of the same block stained with anti-Aβ antibody 4G8,
demonstrating CAA in regions adjacent to the ICH, thereby suggesting that CAA might have been the cause for ICH in this case. A and B and insets in C, 4G8
immunohistochemistry; C, H/E. Scale bars: A and C, 500 μm; B, 50 μm. See also Plate 5.
Hippocampal Sclerosis protein, an inhibitor of phospholipase D2, and an active participant

in oxidative stress production, as well as being both neuroprotec-
Hippocampal sclerosis (HS) is defined as severe astrogliosis and tive and neurotoxic (for review see Spillantini, 2011). Under patho-
neuronal loss in the CA1 region of the hippocampus, which neu- logical conditions, phosphorylation of α-synuclein is enhanced (at
rons are particularly vulnerable to hypoxia, and in the subicu- Ser 129) and its aggregates are the main component of LB and LN
lum. HS occurs in up to 26% of older individuals, frequently as that may in addition contain ubiquitin and α-B-crystallin, among
an additional finding in AD, and is often accompanied by multi- others (for review see Jellinger, 2012c).
ple small infarcts in other brain regions/and/or leucencephalopa- LB occur in two types (Fig. 6.6): (1) classical LB are spherical
thy (Jellinger, 2007b). It has been shown that patients with HS are cytoplasmic inclusions 8–30 μm in diameter with a hyaline eosi-
older than those without HS and had more coronary artery dis- nophilic core, concentric lamellar bands, and a narrow pale-stained
ease, suggesting that related occult hypoxic ischaemic episodes halo that are predominantly seen in pigmented neurons of the sub-
may represent pathogenic factors (Attems and Jellinger, 2006). On stantia nigra, locus ceruleus, and dorsal motor nucleus of the vagus
the other hand, transactivation-responsive (TAR) DNA-binding nerve; and (2) cortical LB are eosinophilic, rounded, angular, or
protein 43 (TDP-43) pathology was detected in up to 70% of cases reniform structures without a halo, which are seen in iso- and allo-
with HS (Amador-Ortiz et al., 2007), and approximately 50% of cortical areas. Of note, while classical LB can be easily detected on
frontotemporal lobar degeneration-TDP (see section FTLD with routine H/E stained sections, cortical LB are not readily detected
TDP-43 pathology) cases show HS. In a recent autopsy study, and immunohistochemistry for α-synuclein should be used to
TDP-43 pathology was seen in 18% of cases with HS, and in 46% detect cortical LB and LN, the latter appearing as thin immunopo-
of cases with concomitantt AD pathology (Rauramaa et al., 2011). sitive threads in the neuropil (for review see Dickson et al., 2009;
However, a causal link between HS and TDP-43 has not been dem- Jellinger, 2011c).
onstrated and their frequent co-occurence might rather reflect a
mere coincidence of two relatively common pathologies (Davidson
Parkinson’s disease
et al., 2011). It has been suggested that HS may incorporate differ-
ent subtypes: HS with advanced age (HS-Aging), HS with seizures PD is one of the most frequent neurodegenerative movement dis-
(HS-SZ), with tauopathies (HS-tau), with non-tauopathy fronto- orders, with a mean age of onset between 55 and 65 years. The
temporal dementia (HS-FTD), and with cerebrovascular diseases prevalence in the general population is estimated at 0.3%, but this
(HS-CVD) (Nelson et al., 2011). increases with increasing age to 3% over 65 years. The four cardinal
clinical features are Tremor at rest, Rigidity, Akinesia (or bradyki-
nesia), and Postural instability (TRAP), often associated with gait
Lewy Body Diseases disturbances and falls (Jankovic, 2008). However, PD can present
Lewy body diseases (LBD) comprise Parkinson’s disease (PD), with a variety of additional clinical features and various clinical
dementia with Lewy bodies (DLB), and Parkinson’s disease demen- subtypes have been described and the neuropathology underly-
tia (PDD) (McKeith et al., 2005). The characteristic neuropatho- ing clinical variability in α-synucleinopathies (e.g., PD) has been
logical lesions of LBDs are Lewy bodies (LB) in neuronal somata reviewed recently (Halliday et al., 2011).
and Lewy neurites (LN) in cell processes. LB and LN are composed On macroscopic examination the most remarkable feature of
of α-synuclein aggregates. α-Synuclein is a 140-amino acid residue PD is pallor of the substantia nigra and often the locus ceruleus.
protein that is a member of a family of proteins that include β- and Histologically, abundant LB and LN are predominantly present in
γ-synucleins with unknown physiological function(s). α-Synuclein subcortical nuclei, e.g. substantia nigra, locus ceruleus, and dorsal
is a lipid binding protein (Goedert, 2001) located at presynaptic ter- motor nucleus of the vagal nerve. In particular, the substantia nigra
minals in proximity to synaptic vesicles, and it has been suggested pars compacta shows a severe depletion of pigmented/melanized
that its physiological function involves the modulation of synap- (45–66%) and of dopaminergic (60–88%) neurons, particularly in
tic transmission and neuronal plasticity (for review see Waxman an area projecting to the striatum (ventrolateral tier) (for review see
and Giasson, 2009). In addition, α-synuclein might be a chaperone Jellinger, 2011b, 2012b).
Fig. 6.6 Classical Lewy bodies in the substantia nigra (arrows in A and B) show a hyaline eosinophilic core and a narrow pale stained halo (B shows two Lewy bodies
in one pigmented neuron). Degenerating pigmented neurons lose melanin (pigment incontinence) that may be incorporated by macrophages (C). On H/E sections,
cortical Lewy bodies are rounded eosinophilic inclusions (D) in neuronal cell bodies that are readily detected with α-synuclein immunohistochemistry (E, arrowhead in
E1) that also stains Lewy neurites as thread- and dot-like structures in the neuropil (E). A–D, H/E; E, α-synuclein immunohistochemistry. Scale bars: A, 50 μm; B–D, 20 μm;
E, 200 μm. See also Plate 6.
Dementia with Lewy bodies and Parkinson’s the vagus nerve) become initially affected (stage 1) and pathol-
disease dementia ogy spreads gradually to the pons (locus ceruleus, stage 2),
midbrain (substantia nigra, stage 3), entorhinal cortex and
The distinction between DLB and PDD is primarily based on clini-
hippocampus (stage 4), and finally reaches the neocortex (stages
cal data; in PDD the onset of extrapyramidal symptoms should
5 and 6) (Braak et al., 2003). Using this classification, PD would
precede dementia by 1 year. Neuropathologically a significantly
reach stage 3, while DLB should at least show stage 4.
more severe Aβ plaque load in the striatum of DLB than in PDD
patients has been observed (Jellinger and Attems, 2006; Halliday 2. The Newcastle–McKeith criteria for LBD (McKeith et al., 2005)
et al., 2011). However, on neuropathological post-mortem exami- distinguish between brainstem predominant (PD), limbic (tran-
nation, clinical PDD might either present as DLB or as PD with an sitional; DLB), and diffuse neocortical (DLB) types. These criteria
additional dementing disease such as AD, and consequently those differ from Braak stages insofar as they do not strictly postulate a
latter cases are not strictly associated with DLB. stepwise progression of α-synuclein pathology from the medulla
DLB is the second most common age-associated neurodegen- oblongata to pons, midbrain, limbic areas, and neocortex, and
erative dementia, accounting for up to 26% of dementias in spe- cases might show severe neocortical α-synuclein pathology with
cialized referral centres (for review see Ince, 2011). The clinical only minimal involvement of brainstem regions. Thereby this
symptoms of DLB are dementia associated with core neuropsychi- staging system questions the validity of the system proposed by
atric features of fluctuating cognitive function, visual hallucina- Braak and colleagues (Braak et al., 2003).
tions, and in up to 75% of cases mild parkinsonism (McKeith et al., 3. Leverenz and colleagues (Leverenz et al., 2008) modified the
2005). For more detailed description of clinical findings in DLB, Newcastle–McKeith criteria by adding an amygdala predominant
see Chapter 35. type that may completely lack α-synuclein pathology in regions
Macroscopically, brains of DLB patients may appear normal or other than the amygdala. This amygdala predominant type fre-
show diffuse cerebral atrophy that is indistinguishable from AD. quently co-occurs with severe AD (e.g. in 24% of otherwise pure
The pigmentation of the substantia nigra varies from normal to AD cases (Uchikado et al., 2006)).
pale and the locus ceruleus is usually depigmented. Histologically,
DLB is characterized by abundant LB and LN in the neocortex,
in particular the cingulate gyrus and the limbic system including
Multiple System Atrophy
the entorhinal cortex and hippocampus. Similar to AD, additional Multiple system atrophy (MSA) is a sporadic neurodegenerative
pathologies are usually present in DLB and those are discussed in disease characterized clinically by parkinsonism, cerebellar ataxia,
the section Cerebral multimorbidity. autonomic dysfunction, and corticospinal signs. MSA primarily
involves striatonigral and olivopontocerebellar structures (for review
Neuropathological diagnostic criteria for see Jellinger and Lantos, 2010; Holton et al., 2011). On macroscopic
Lewy body disease examination considerable atrophy of the cerebellum and pontine base
is frequently observed and the atrophic putamen shows dark discol-
Several neuropathological staging systems for LBD are in use: oration. The substantia nigra and locus ceruleus are depigmented.
1. Braak stages for α-synuclein-related pathology postulate that The characteristic neuropathological lesios of MSA are oligodendro-
nuclei in the medulla oblongata (e.g. dorsal motor nucleus of glial α-synuclein cytoplasmic inclusions known as glial cytoplasmic
Fig. 6.7 In multiple system atrophy (MSA), aggregates of α-synuclein termed Papp-Lantos inclusions or glial cytoplasmic inclusions (arrows in B and C) are frequent in
the oligodendroglia. Photomicrographs are taken from sections of the external capsule. A and B, α-synuclein immunohistochemistry; C, H/E. Scale bars: A, 500 μm; B and
C, 20 μm. See also Plate 7.
inclusions (GCI or Papp-Lantos inclusions) that are associated with evenly distributed but in large clusters (Dickson, 2009; Munoz et al.,
myelin depletion. α-Synuclein also accumulates in oligodendroglial 2011). Ballooned neurons, similar to the ones seen in corticobasal
nuclei as well as in the cytoplasm and nuclei of neurons (Fig. 6.7). degeneration (CBD), may be present as well as tau immunoreactive
glial lesions that, interestingly, are predominately composed of 4R
tau (Zhukareva et al., 2002).
Frontotemporal Lobar Degenerations
(Including Tauopathies) Corticobasal degeneration (CBD)
The clinical symptoms in CBD are highly variable, the most char-
Frontotemporal lobar degenerations (FTLD), the third or fourth
acteristic being the corticobasal syndrome with asymmetrical
most frequent cause of dementias, usually show a frontotemporal
akinetic-rigid parkinsonism and cortical signs including apraxia,
pronounced cerebral atrophy. Microscopically they are charac-
dystonia, and myoclonus. However, some autopsy series indicate
terized by cellular inclusion bodies that are composed of tau or
that—although being characteristic—the corticobasal syndrome
TDP-43 or a protein encoded by the fused in sarcoma gene (FUS)
may not be the most common clinical presentation of CBD, as
and/or ubiquitin or neurofilament and/or a basophilic substance
frontal-type dementia has been reported to be more frequent. No
(basophilic inclusion bodies). According to the predominant pro-
epidemiological data are available for the prevalence and incidence
tein aggregates, the subtypes described in the following sections
of CBD, but its incidence has been estimated to be less than 1 per
can be distinguished, but there are cases with overlapping pathol-
100,000 people per year. CBD is a disease of middle to late age with
ogy (Mackenzie et al., 2011; Goedert et al., 2012; Halliday et al.,
no gender predisposition and age being the only known risk factor
2012; Seltman and Matthews, 2012). It should be mentioned that
(for a review see Dickson et al., 2011). Reflecting the broad spec-
FTLD-tau are frequently listed in the group of tauopathies that
trum of possible clinical presentations, the neuropathological find-
additionally include diseases that are not characterized by atrophy
ings in CBD are highly variable. However, the classical presentation
of frontal /and/or temporal lesions, such as parkinsonism-dementia
shows asymmetrical atrophy of cortical gyri that is most marked
complex of Guam and postencephalic parkinsonism.
in superior frontal and parietal parasagittal regions. Among other
macroscopic features, a red-brown discoloration may be seen in
FTLD-tau the globus pallidus and the substantia nigra shows loss of neu-
Pick’s disease (PiD) romelanin pigment, while the locus ceruleus appears normal. The
Pick’s disease, although a well-known eponym, is a rare cause of characteristic neuropathological finding of CBD is the accumula-
frontotemporal dementia (less than 5% of cases), with behavioural tion of 4R tau in the neuropil and in astrocytes in the cortex, white
abnormalities or with nonfluent progressive aphasia as the most matter, basal ganglia, thalamus, and brainstem. 4R tau may also be
common clinical presentation. Both sexes are equally affected, present in neurons, but the neuropathological hallmark lesion is
with an average age of onset at approximately 60 years and a dis- the astrocytic plaque, which is not seen in other diseases (Dickson,
ease duration of 8–10 years (for a review see Munoz et al., 2011). 2009; Kouri et al., 2011). The astrocytic plaque represents 4R tau
At post-mortem examination, the characteristic macroscopic fea- accumulation in distal segments of astrocytes, while the cell body is
ture is severe circumscribed atrophy of the frontal, temporal, and, free of accumulation. This morphology is reminiscent of a neuritic
sometimes, parietal lobes, while the motor and sensory cortices, plaque (ide nomen) but no Aβ accumulation is present in astrocytic
posterior two-thirds of the superior temporal gyrus, and occipital plaques. Astrocytic plaques predominately occur in the cortex, but
cortex are preserved. The brainstem and cerebellum show no mac- can also be present in other regions such as the caudate and puta-
roscopic abnormalities. Histologically the macroscopically atrophic men. Swollen cortical neurons that are termed ballooned neurons
areas as well as the deep grey matter and the brainstem, in particu- are another neuropathological hallmark of CBD.
lar monoaminergic nuclei, show neuronal loss and gliosis, which
also affects atrophic subcortical white matter. Pick bodies—the Progressive supranuclear palsy (PSP)
neuropathological hallmark lesion of PiD—are large argyrophilic, Symmetrical, akinetic-rigid parkinsonism with severe postural
neuronal, cytoplasmic, round inclusions predominately composed instability and supranuclear ophthalmoplegia are the typi-
of 3R tau. Pick bodies are frequent in hippocampal pyramidal neu- cal clinical symptoms of progressive supranuclear palsy (PSP)
rons of the CA1 sector, while CA2–4 and subiculum are less fre- (Steele–Richardson–Olszewski syndrome), but less frequent
quently involved. In other cortical areas, Pick bodies may occur less atypical presentations such as asymmetrical clinical syndromes
(e.g. corticobasal syndrome) may be seen (Steele et al., 1964; confined to allocortical regions, corresponding to neurofibrillary
Dickson et al., 2010). In the UK, the prevalence of PSP is esti- Braak stage III, and rarely mild isocortical involvement is seen.
mated at approximately 5 per 100,000, with the mean age at NFTD accounts for 5 to 7% of late-onset dementias and differs from
onset ranging from 66– 69 years and in 44–61% of cases affect- AD by later onset (80 years), shorter duration (5 years), less severe
ing females (Nath et al., 2001). The basal ganglia, subthalamic cognitive impairment, and the almost absence of APOEε4 geno-
nucleus, and substantia nigra are primarily affected, and usually type (Jellinger and Attems, 2007a). Recent studies showed absence
pathology in the cerebellar dentate nucleus is severe and associ- of soluble Aβ with the tau gene MAPT H1 haplotype, classifying it
ated with profound atrophy of the superior cerebellar peduncle as a specific tauopathy independent of amyloid (Santa-Maria et al.,
(Dickson, 2009). The characteristic neuronal lesion is the glo- 2012).
bose NFT that is composed of 4R tau and occurs in brainstem
nuclei and the nucleus basalis of Meynert, while in other regions FTLD with TDP-43 pathology
NFTs may be flame-shaped. Immunohistochemistry for 4R tau FTLD with TDP-43 pathology (FTLD-TDP) may show any of
also labels tuft-shaped astrocytes (tufted astrocytes) which are the three major clinical variants of the frontotemporal dementia
the most characteristic glial lesion in PSP (Yamada et al., 1992). syndrome, including behavioural variant, progressive nonfluent
In addition, 4R tau positive NT are seen in grey and white matter aphasia, and semantic dementia (Cairns et al., 2007b). The clini-
of both cortical and subcortical regions. The severity and distri- cal frontotemporal dementia syndrome accounts for 10–15% of all
bution of tau pathology varies in the different subtypes of PSP dementia cases, and some studies have shown that neuropathologi-
(PSP, parkinsonism/PSP-P, Richardson’s syndrome, and atypical cally FTLD-TDP is present in 50% of cases with clinical fronto-
forms) (Dickson et al., 2011). temporal dementia syndrome (Lipton et al., 2004; Neumann et al.,
Argyrophilic grain disease 2006; Cairns et al., 2007b; Davidson et al., 2007). The mean age of
onset of FTLD-TDP is approximately 60 years (33–89 years) with a
Argyrophilic grain disease (AGD) has a mean age of onset of
mean disease duration of 8 years (2–18 years). Neuropathologically
approximately 80 years, with males and females affected equally.
FTLD-TDPs are characterized by neuronal cytoplasmic inclusions
AGD is the most common sporadic tauopathy and is observed in
(NCI) that consist of phosphorylated TDP-43. TDP-43 is involved
up to 5% of late-onset dementia cases (Saito et al., 2002). However,
in regulating transcription and alternative splicing (Buratti and
AGD is a frequent finding in other neurodegenerative diseases and
Baralle, 2008), and physiologically TDP-43 is present in the
has been reported to be present in varying severity in up to 26% of
nucleus but under pathological conditions such as FTLD-TDP
AD cases (Fujino et al., 2005). Macroscopically the brain appears
is usually located in the cytoplasm (TDP-43 NCI in spinal motor
relatively normal with mild frontotemporal cortical atrophy, but
neurons are a characteristic feature of amyotrophic lateral scle-
severe atrophy of the ambient gyrus was described in late stages of
rosis). Macroscopically cerebral atrophy is variable, but frontal
the disease (Saito et al., 2002). In AGD, 4R tau aggregates occur in
and temporal lobes are most frequently affected, sometimes with
dendrites and present as comma- or grain-shaped structures in the
asymmetrical distribution. In addition to unspecific features of
neuropil of the medial temporal lobe, in particular in the hippoc-
neurodegeneration, TDP-43 positive NCI and dystrophic neurites
ampus Ammon’s horn sector CA1 and prosubiculum (for review
are usually seen in the frontotemporal neocortex and granule cell
see Tolnay and Braak, 2011). In addition, 4R tau is present in oli-
layer of the dentate gyrus (Cairns et al., 2007a) (Fig. 6.8). Neuronal
godendroglia, referred to as coiled bodies, that may be present in
intranuclear inclusions (NII) may also be present and, depending
other tauopathies but in AGD are a consistent finding in the white
on the amount and distribution of NCI and DN as well as on the
matter.
presence of NII, four neuropathological subtypes (types A, B, C, D)
Neurofibrillary tangle dominant dementia of FTLD-TDP have been described (Mackenzie et al., 2011).
Neurofibrillary tangle dominant dementia (NFTD) is characterized
by the presence of NFT/NT that contain both 3R and 4R tau, while FTLD-with ubiquitin positive inclusions (UPS)
Aβ pathology is in 75% of cases completely absent or present to a Ubiquitin positive inclusions are the characteristic lesion of
very limited degree. Importantly, in NFTD, tau pathology is mostly FTLD-UPS (ubiquitin-proteasome-system). It now appears
Fig. 6.8 TDP-43 positive neuronal cytoplasmic inclusions (NCI, arrow in A1 and B) and thread-like structures (arrowhead in A1) are the characteristic neuropathological
lesion for FTLD-TDP-43 and motor neuron disease (amyotrophic lateral sclerosis) but may be present as an unspecific feature in a variety of other diseases, including AD.
Of note, under physiological conditions TDP-43 is present in neuronal nuclei (ring in B). Immunohistochemistry with antibody against (nonphosphorylated) TDP-43.
Scale bars: A, 100 μm; B, 20 μm. See also Plate 8.
that most cases of sporadic FTLD-UPS have FUS immunoreac- infectious agens is the prion protein that essentially consists of PrPSc
tive pathology, but the term FTLD-UPS is still appropriate for a which presents as aggregates of an abnormally folded β-sheet-rich
familial form of FTLD, familial FTLD linked to chromosome 3 isoform of a normal cellular protein, of largely unknown function
(FTD-3). (possible role in myelin maintenance), termed PrPC (for review see
Aguzzi and Heikenwalder, 2006). Prion diseases occur in humans
FTLD-FUS and animals. Human prion diseases encompass sporadic (sporadic
FTLD-FUS are characterized by abnormal accumulation of the Creutzfeldt–Jakob disease (sCJD) and fatal sporadic insomnia),
FUS protein as the most prominent pathology, and include atypical inherited (familial CJD, Gerstmann–Sträussler–Scheinker syn-
FTLD with ubiquinated inclusions (aFTLD-U), neuronal interme- drome and fatal familial insomnia), and infectious forms (iatro-
diate filament inclusion disease (NIFID), and basophilic inclusion genic CJD, variant CJD (vCJD) and Kuru) (Parchi et al., 2011). Of
body disease (BIBD) (Mackenzie et al., 2010). A more detailed note, as vCJD predominantly occurred in the UK, the country with
description of FTLD-FUS is beyond the scope of this chapter (for the highest incidence of bovine spongiform encephalopathy (BSE,
comprehensive review see Cairns, 2011; Neumann and Mackenzie, a bovine prion disease), the exposure to prion proteins due to past
2011). consumption has been suggested to cause vCJD in humans (Will et
al., 1996). It is assumed that PrPSc is capable of inducing conforma-
FTLD with no inclusions tional conversion of PrPC into PrPSc, thereby leading to a spread of
The term FTLD-no inclusions (FTLD-ni) is used for FTLD cases aggregated PrPSc in the CNS.
that lack any histochemically and immunohistochemically detect- The most common form of prion disease in humans is sCJD,
able inclusions but otherwise show clinical features of dementia accounting for 85% of all CJD. No cause for sCJD has been identi-
and nonspecific neuropathological changes (e.g. atrophy, neuronal fied and the typical clinical picture is rapidly progressive dementia
loss). This type has previously been referred to as ‘dementia lack- with ataxia and myoclonus. The incidence of sCJD is between 1 and
ing distinctive histopathology’ (DLDH), but it is suggested that 2 cases per million per year and the mean age of onset is 65 years
FTLD-ni replaces the term DLDH, as the latter suggest that patho- (for a review see Budka et al., 2011). On macroscopic examination
logical lesions are completely absent (Mackenzie et al., 2009). the brain may appear normal but usually some degree of atrophy is
present. Histologically sCJD is characterized by a triad of spongi-
form change, neuronal loss, and astrocytic as well as microglial
Huntington’s Disease (HD) gliosis. Spongiform change presents as diffuse or focal small round
Huntington’s disease is an autosomal dominant hereditary disease. or oval vacuoles in the cerebral cortex (whole thickness or deep lay-
HD belongs to the family of trinucleotide repeat diseases that are ers), subcortical grey matter (almost constantly in the head of the
caused by expansions of pre-existing tandem repeat sequences with caudate nucleus), and cerebellar molecular layer. In addition, PrP
each of the diseases (e.g. myotonic dystrophy type 2, spinocerebel- immunohistochemistry should be used to confirm the diagnosis
lar ataxia) affecting different genes (for review see Clark, 2011). The (Budka et al., 2011). A consensus classification of human prion dis-
gene for HD (huntingtin or HTT) is located on the short arm of ease subtypes allows reliable identification of molecular subtypes
chromosome 4 (4p16.3) and the mutation consists of an expanded (Parchi et al., 2012).
repetition of the cytosine-adenine-guanine (CAG) trinucleotide
(The Huntington’s Disease Collaborative Research Group, 1993). Vascular Dementia (VaD)
HD equally affects men and woman, age at diagnosis is on aver-
Vascular dementia or vascular cognitive impairment (O’Brien et al.,
age 40 years, and the symptomatic prevalence ranges from 5–10
2003) can be defined as acquired cognitive impairment caused by
per 100,000 (Hedreen and Roos, 2011). The prodromal period
cerebrovascular disease. Due to the high variability of morphologi-
of HD, which may be present for many years before diagnosis, is
cal findings and multifactorial pathogenesis of VaD, no generally
characterized by motor abnormalities (subtle involuntary move-
accepted morphologic scheme for staging cerebrovascular lesions
ments and oculomotor dysfunction), while the typical symptoms
and no validated neuropathological criteria for VaD have been
of the disease include chorea and mental dysfunction, leading to
established to date (for review see Jellinger, 2007b, 2008). Therefore
dementia (Shoulson and Young, 2011). The brain in end-stage
many aspects of VaD, such as prevalence, morphology, and patho-
HD shows severe reduction in weight (900–1,000 g) mainly due
genesis, are a matter of ongoing discussions. VaD has, however,
to profound atrophy of the caudate nucleus and putamen, but the
been suggested to be the cause for dementia in approximately 10%
cerebral cortex, hippocampus, thalamus, and white matter are also
of cases. Clearly, cerebrovascular lesions can cause neuronal loss
affected (Shoulson and Young, 2011). Microscopically the basal
and—depending on the topographical localization—this process
ganglia show severe neuronal loss and gliosis, and on microscopic
may lead to dementia.
examination huntingtin or ubiquitin positive protein accumula-
The morphological substrates for VaD are extensive and only a
tions are frequently seen in cellular processes (neurites) and, to a
brief and incomplete description can be provided here (for compre-
lesser degree, as nuclear inclusion bodies (Hedreen and Roos, 2011;
hensive reviews see Jellinger, 2007b; Grinberg and Thal, 2010).
Shoulson and Young, 2011).
Two types of vessel disorders can potentially cause VaD:
1. Atherosclerosis is a very common vessel disorder in the older
Prion Diseases person, frequently affecting large to medium sized arteries of the
Prion diseases that are also known as transmissible spongiform entire cardiovascular system. In the cranium it often affects the
encephalopathies (TSEs) are caused by a potentially infectious circle of Willis, and an important extracranial site—with respect
agens and thereby differ from other neurodegenerative diseases. The to the development of cerebrovascular lesions—is the carotid
arteries, in particular at the level of the carotid bifurcation. special stains (e.g. Luxol Fast Blue), demyelination is indicated by
Atherosclerosis is initiated by thickening of the tunica intima (i.e. a pallor in staining intensity and arterioles in these areas may show
the innermost layer of a blood vessel), with subsequent accumula- concentric thickening of the vessel wall and perivascular spaces
tion of blood-derived lipids. This process is followed by splitting that are filled with macrophages (of note, small perivascular spaces
of the internal elastic lamina and finally cholesterol accumula- alone are not an indicator of small vessel disease but are an artifact
tion leads to the manifestation of atherosclerotic plaques that can caused by tissue processing; Fig. 6.9). Lacunes are cavities measur-
calcify. Atherosclerosis causes narrowing of the arteries’ lumina, ing 5–10 mm in diameter that are caused by either old (lacunar)
thereby reducing the blood supply for the supported region, but infarcts or less frequently haemorrhages. In addition to the white
this is frequently compensated by a shift in blood flow towards matter, they occur in the basal ganglia (deep grey matter lesions)
collateral arteries (e.g. vertebral arteries for internal carotid arter- and, if severe, are the morphologic substrate of the so-called status
ies). On the other hand, rupture of atherosclerotic plaques often lacunaris or état criblé.
leads to thrombosis that results in either occlusion of the lumen Ischaemic brain infarcts are caused by insufficient blood sup-
or thromboembolism. Depending on the size of the embolus, ply frequently due to thromboembolism. After 6–12 h the neurons
thromboembolism may cause lesions that range from ‘silent inf- show signs of acute ischaemic cell injury (e.g. eosinophilic cyto-
arcts’ to large cerebral infarcts with overt clinical symptoms. plasm and shrunken nucleus) and between 24 and 48 h neutrophils
2. Small vessel disease refers to pathological changes that affect appear that are replaced by macrophages after 48 h. Ten days after
small arteries and arterioles (Fig. 6.9). Small arteries may show the infarction, the area is liquefied, resulting in the formation of a
morphological changes that are similar to atherosclerotic changes cavity (third week) that becomes intersected by vascular and con-
(i.e. microatheroma) and are termed ‘small vessel arteriosclero- nective tissue strand and surrounded by gliosis (scarring, Fig. 6.10).
sis/atherosclerosis’. ‘Lipohyalinosis’ is used for asymmetric fibro- Haemorrhagic brain infarcts are characterized by secondary blood
sis/hyalinosis, while concentric hyaline thickening with lumen influx into the infarcted territory and microinfarcts refer to infarcts
stenosis is termed ‘arteriosclerosis’. The latter is particularly com- below 5 mm in diameter (for lacunar infarcts see above).
mon in the white matter and the main cause for white- and deep Cerebral haemorrhages are larger than 10 mm in diameter.
grey matter lesions, respectively. Causes for nontraumatic intracerebral haemorrhage include aneu-
rysms of cerebral arteries, coagulation disorders, atherosclerosis,
White matter lesions (leukoaraiosis) are increasingly detected
hypertension with small vessel disease, and CAA. Recently, cerebral
by modern neuroimaging methods (up to 60% in individuals over
microbleeds have gained wider attention as MRI detects small sig-
60 years of age), frequently clinically asymptomatic, and in most
nal voids, indicating perivascular accumulation of haemosiderin/
cases related to small vessel disease of the white matter. The associ-
haemosiderin-laden macrophages that are presumed to reflect old
ated neuropathological findings include demyelination, axonal loss,
cerebral microbleeds. Cerebral microbleeds are associated with
and lacunar infarcts (usually small), most frequently in the frontal,
CAA and, according to MRI studies, frequent in older people (e.g.
parietal, and occipital white matter (Schmidt et al., 2011). Using
Fig. 6.9 Small vessel disease is a frequent cause of white matter (WM) lesions. (A) Normal vessel and white matter; note the slight perivascular space that is regarded as an
artifact; (B) mildly enlarged perivascular spaces; (C) considerable enlargement of the perivascular space accompanied by moderate demyelination of the WM (pallor in myelin
staining); (D) severe WM lesions with enlarged perivascular spaces, WM pallor, and tissue loss; (E) normal artery; (F) artery with mild fibrosis; (G) artery with moderate fibrosis;
(H) severe arteriolar fibrosis with acellular concentric thickening of the vessel wall (usually associated with hypertension). A–D, Luxol Fast Blue histochemistry (myelin stain);
E–H, H/E. Scale bars: A and G, 50 μm; B, C, D, and H, 100 μm; E and F, 20 μm. (Photomicrographs by courtesy of Kirsty McAleese.) See also Plate 9.
Fig. 6.10 Old cystic infarct in the right frontal lobe in the territory of the middle cerebral artery (A; B, frontal plane). Scale bars: A, 50 mm; B, 20 mm. See also Plate 10.
23.5% in Vernooij et al., 2008). However, microbleeds are rarely A large autopsy study reported that 41.5% of clinically demented
seen at neuropathological post-mortem examination and their true patients (n = 1,700, mean age 84.3 ± 6.0 years) fulfilled neuropatho-
incidence is therefore controversial. logical criteria of pure AD, while AD with additional pathol-
It is beyond the scope of this chapter to provide a comprehensive ogy was seen in 43.2%; 23.2% showed additional cerebrovascular
review of the many possible ways to classify cerebrovascular pathol- pathology (e.g. infarcts, hippocampal sclerosis, lacunar state), 9.2%
ogy and several articles are suggested for further reading (O’Brien α-synuclein pathology, and 2.6% various other neuropathological
et al., 2003; Kalaria et al., 2004; Jellinger, 2007b, 2008; Gorelick lesions, while the remaining 15.3% did not show any AD pathology,
et al., 2011). As an example, a brief outline of the classification of including 10.7% of ‘pure’ vascular dementia and 5.5% other disor-
VaD according to the underlying disease is now provided. ders (Jellinger and Attems, 2007b; Jellinger, 2011d).
Another large autopsy study across nine centres of the BrainNet
1. Large vessel disease: Atherosclerosis of extra- or intracranial Europe Consortium included neuropathological data of 3,303
arteries may cause thromboembolism or hypoperfusion that cases (1,667, female; 1,636, male; mean age, 74.14 ± 12.07 years)
leads to single or multiple infarctions in the cerebral regions that that showed clinical dementia. Fifty three percent of cases showed
receive blood supply by the respective artery. Cognitive impair- mixed pathology, which was most frequently seen in LBD (PD,
ment might result from tissue loss with a volume over 50 ml 92%; LBD, 61%), followed by cases with cerebrovascular pathology
(Kalaria et al., 2004), or from smaller infarcts that are strategi- (65%), and AGD (61%), compared to AD (43%), PSP (22%), CBD
cally placed affecting functionally important brain areas and (21%), FTLD (9%), and CJD (2%). The most frequent additional
neuronal circuits (‘strategic’ infarct dementia syndrome). diagnoses in mixed cases was AD pathology in 89.6% (P < 0.01),
followed by cerebrovascular pathology (52.6%), synucleinopathy
2. Small vessel disease: Small vessel disease affects small arteries and
(50%), and AGD (11.4%) (Kovacs et al., 2008).
arterioles and includes arteriosclerosis/atherosclerosis, arteriolo-
The prevalence of mixed pathologies increases with age. In a con-
sclerosis, and lipohyalinosis (for review see Grinberg and Thal,
secutive autopsy series on 1,110 patients (64% female; mean age at
2010). Small vessel disease may cause ischaemic white matter
death 83.3 ± 5.6 years, range 60–103 years, 90% over age 70), the
damage and lacunar infarcts in the basal ganglia, both of which
prevalence of pure AD increased from 32.2% in the 7th decade to
may present a morphological correlate for VaD.
45.1% in the 9th decade, while it decreased in the 10th decade to
3. Hypoxic damage: Hypoxic damage caused by cerebral hypoper- 39.2%. By contrast, the prevalence of AD with minor cerebrovascu-
fusion may lead to ischaemic-anoxic damage and hippocampal lar lesions as well as mixed dementia (AD and vascular dementia)
sclerosis (see section Hippocampal sclerosis). both increased from 7.8% and 0% in the 7th decade to 32.9% and
7.5% in the 10th decade, respectively (Jellinger and Attems, 2010b).
As already outlined neurodegenerative diseases are neuropatho-
Cerebral multimorbidity logically characterized by the presence of discrete lesions in a spe-
It is becoming increasingly clear that the ageing brain is character- cific topographical pattern. However, data from many different
ized by the presence of multiple pathologies rather than the charac- autopsy studies point towards the presence of additional lesions in
teristic neuropathological lesions of one single neurodegenerative otherwise well-characterized cases, which fulfill the criteria for a
disease only. Of particular importance is the frequent presence of single distinct disorder (Fig. 6.11):
confounding processes in the aged brain that coexist with AD, as, ◆ In AD that is characterized by Aβ and tau pathology, Lewy
for example, cerebrovascular disease, LB pathology, AGD, TDP-43 bodies are seen in up to 43% (Uchikado et al., 2006) (AD with
pathology, and hippocampal sclerosis. Approximately two-thirds of amygdala LBs is considered a distinct and common form of
aged human brains contain nonAD pathology (Nagy et al., 1997; α-synucleinopathy (Uchikado et al., 2006)), TDP-43 inclusions
Nelson et al., 2007; Schneider et al., 2007a; Davidson et al., 2011), mainly restricted to the granule cell layer of the dentate gyrus and
which have, however, frequently been missed clinically and could entorhinal cortex in up to 43% (Duyckaerts et al., 2009; Davidson
not be identified without neuropathological examination using et al., 2011; Rauramaa et al., 2011; Robinson et al., 2011), and
modern biochemical and molecular-biological analyses (for review severe cerebrovascular lesions in up to 20% (Jellinger and Attems,
see Jellinger, 2013). 2007b) of cases, respectively.
Fig. 6.11 Multiple pathologies in brains of older demented patients. Full arrows point towards the characteristic neuropathology of the respective disease, while dotted
arrows point towards neuropathological lesions that are frequently seen in addition to the main pathological hallmark lesions. Approximate percentages are encircled.
AD, Alzheimer’s disease; LBD, Lewy body diseases; FTLD-TDP, frontotemporal lobar degeneration with TDP-43 pathology; VaD, vascular dementia. See also Plate 11.
◆ In LBD that is characterized by α-synuclein pathology, Aβ discussed (Lee et al., 2000; Zekry et al., 2003; Gold et al., 2007).
pathology was identified in 95%, considerable tau pathology AD pathology alone more frequently accounts for dementia than
(Braak stages V/VI) in 55%, and various degrees of cerebrov- both microscopic and macroscopic infarcts (Troncoso et al., 2008)
ascular pathology in 75% (Jellinger and Attems, 2008) of cases. and in advanced stages of AD concomitant small vascular lesions
Interestingly, Parkkinen and colleagues in investigating an do not significantly influence the overall state and progression of
autopsy cohort for the presence of α-synuclein pathology found cognitive decline; hence the severity and extent of AD pathology
that over 50% of cases with widespread α-synuclein did not show seems to overwhelm the effects of cerebrovascular disease (Lee
any clinical symptoms (Parkkinen et al., 2008). et al., 2000; Bennett et al., 2005; Jellinger, 2007b, 2008). However,
◆ Pure VaD without additional lesions is rare (e.g. 12.3% in Jellinger this does not preclude the possibility that cerebrovascular lesions
and Attems, 2010a), and frequently additional, often limited AD exert an influence on AD pathology itself and it has been suggested
pathology is present. However, since 50–85% of post-mortem that cerebrovascular lesions in AD lower the threshold for overt
brains from individuals (both demented and nondemented) who clinical dementia, i.e. less AD pathology is needed to cause clini-
die over 80 years of age show appreciable cerebrovascular lesions cal symptoms, if additional cerebrovascular pathology is present
(Petrovitch et al., 2005), a particular problem is the impact of (Jellinger and Attems, 2003).
cerebrovascular disease on cognition, especially in relation to AD Another example of the simultaneous presence of a distinct neu-
pathology (Bennett et al., 2005; Chui, 2006;, Chui et al., 2006; rodegenerative disease and age-associated change would be limited
Schneider et al., 2007a, 2007b). AD pathology (e.g. Braak stage III and Aβ-plaques) in DLB; here,
AD pathology could be regarded as age-associated change that
The high prevalence of multiple pathologies in brains of aged manifests independently of DLB. Similarly to the presumed impact
individuals could reflect the simultaneous presence of one single of cerebrovascular lesions in AD, α-synuclein pathology in DLB
distinct neurodegenerative disease and age-related changes that might be aggravated by age-associated AD pathology.
are not directly associated with the primary neurodegenerative In vitro studies demonstrated synergistic fibrillization of tau and
disease; e.g. the high prevalence of cerebrovascular lesions in AD α-syn (Giasson et al., 2003) and recent data from studies on trans-
brains of individuals aged 80+ probably mirrors the prevalence genic (tg) mice indeed suggest a mutual interaction of Aβ, tau, and
of age-associated vascular diseases in the aged per se, rather than α-synuclein pathologies; Clinton and colleagues (Clinton et al., 2010)
suggesting causal relationships between AD and cerebrovascular crossed 3xtg mice that develop both Aβ-plaques and tau pathology
lesions. The burdens of vascular and AD-type lesions are indeed (Oddo et al., 2003) with mice that express the A53T mutation in
considered to be independent of each other and are consistent with α-synuclein (M83-h) (Giasson et al., 2002). At 12 months of age
an additive (or even synergistic) effect of both types of lesions on these so-called DLB-AD mice had significantly higher detergent
cognitive impairment (Schneider et al., 2004; Jellinger and Attems, soluble and insoluble Aβ40, detergent insoluble Aβ42, and insoluble
2005; Jellinger, 2007a; Launer et al., 2008; Duyckaerts et al., 2009; tau levels than 3xtg mice, respectively. Moreover, detergent insolu-
Strozyk et al., 2010; Duyckaerts and Dickson, 2011). The thresh- ble α-synuclein as well as detergent soluble α-synuclein phosphor-
olds for vascular and degenerative lesions in distinguishing ‘pure’ ylated at serine 129 levels were significantly higher in DLB-AD mice
vascular dementia or AD from mixed cases have been critically compared to M83-h mice. Taken together, these results suggest that
Aβ, tau, and α-synuclein interact in vivo to promote the aggrega- Alafuzoff, I., et al . (2009a). Staging/typing of Lewy body related
tion and accumulation of each other (Clinton et al., 2010). alpha-synuclein pathology: a study of the BrainNet Europe Consortium.
Albeit directly comparable data on humans are not available, Acta Neuropathologica, 117, 635–52.
recent studies suggest similar interactions in human brains; e.g. in a Alafuzoff, I., et al. (2009b). Assessment of beta-amyloid deposits in human
brain: a study of the BrainNet Europe Consortium. Acta Neuropathologica,
study on PDD and PD cases, Compta and colleagues (Compta et al., 117, 309–20.
2011) found a significant correlation between Lewy body scores Amador-Ortiz, C., et al. (2007). TDP-43 immunoreactivity in hippocampal
and NFT Braak stages. On the other hand, the considerable overlap sclerosis and Alzheimer’s disease. Annals of Neurology, 61, 435–45.
between tau and α-synuclein pathologies could be explained by a Attems, J. and Jellinger, K.A. (2004). Only cerebral capillary amyloid
mechanistic linkage between tauopathies and synucleinopathies; it angiopathy correlates with Alzheimer pathology—a pilot study. Acta
has been suggested that reduced levels of proteasome 19S and 20S Neuropathologica, 107, 83–90.
subunits facilitate abnormal deposition of both tau and α-synuclein Attems, J. and Jellinger, K.A. (2006). Hippocampal sclerosis in Alzheimer
(Wills et al., 2010; Jellinger, 2011a, 2012a). disease and other dementias. Neurology, 66, 775.
Attems, J., Lintner, F., and Jellinger, K.A. (2004). Amyloid beta peptide
1–42 highly correlates with capillary cerebral amyloid angiopathy and
Future Directions in Neuropathological Alzheimer disease pathology. Acta Neuropathologica, 107, 283–91.
Attems, J., Jellinger, K.A., and Lintner, F. (2005a). Alzheimer’s disease
Assessment pathology influences severity and topographical distribution of cerebral
Currently, neuropathological assessment is based on semiquantita- amyloid angiopathy. Acta Neuropathologica, 110, 222–31.
tive scoring, usually on a four-tiered scale, e.g. 0, absent; 1, mild; Attems, J., et al. (2005b). Cause of death in demented and non-demented
2, moderate; 3, severe. Frequently, the criteria for scoring one of elderly inpatients; an autopsy study of 308 cases. Journal of Alzheimer’s
those categories have been standardized, e.g. (Alafuzoff et al., 2008, Disease, 8, 57–62.
Attems, J., Lintner, F., and Jellinger, K.A. (2005c). Olfactory involvement in
2009a, 2009b). While semiquantitative assessment allows for the
aging and Alzheimer’s disease: an autopsy study. Journal of Alzheimer’s
statement of routine diagnoses, it provides only a rough estimation Disease, 7, 149–57.
of the amount of pathology present. For example, we found that the Attems, J., Quass, M., and Jellinger, K.A. (2007). Tau and alpha-synuclein
area covered by immunopositivity for AT8 antibody (tau) in cases brainstem pathology in Alzheimer disease: relation with extrapyramidal
that were assigned the score ‘severe’ differs by nearly 100% between signs. Acta Neuropathologica, 113, 53–62.
cases. In the 90+ study, a prospective longitudinal population-based Attems, J., Lauda, F., and Jellinger, K.A. (2008). Unexpectedly low prevalence
study of ageing and dementia, the area covered by Aβ immunopo- of intracerebral hemorrhages in sporadic cerebral amyloid angiopathy:
sitivity in selected neocortical regions significantly correlated with an autopsy study. Journal of Neurology, 255, 70–6.
the presence of clinical dementia (Robinson et al., 2011); of note, Attems, J., et al. (2010). Capillary CAA and perivascular Abeta-deposition:
in most studies the extent of neocortical of Aβ immunopositiv- two distinct features of Alzheimer’s disease pathology. Journal of the
Neurological Sciences, 299, 155–62.
ity as assessed by semiquantitative scoring did not correlate with
Attems, J., et al. (2011). Review: sporadic cerebral amyloid angiopathy.
the presence of clinical dementia (for review see Duyckaerts et al., Neuropathology and Applied Neurobiology, 37(1), 75–93.
2009). It is likely that new clinicopathological phenotypes could be Attems, J., Thomas, A. and Jellinger, K. (2012). Correlations between cortical
identified by routinely assessing the amount of pathology in a more and subcortical tau pathology. Neuropathology and Applied Neurobiology,
quantitative way. Indeed, in quantitatively assessing tau pathology 38(6), 582–90.
in a large cohort (n = 889) of neuropathologically diagnosed AD Ballatore, C., Lee, V.M., and Trojanowski, J.Q. (2007). Tau-mediated
cases, Murray and colleagues (Murray et al., 2011) recently found neurodegeneration in Alzheimer’s disease and related disorders. Nature
hippocampal sparing and limbic predominant subtypes of AD, Reviews Neuroscience, 8, 663–72.
that differed in clinical presentation, age at onset, disease dura- Bennett, D. A., et al. (2005). Mild cognitive impairment is related to Alzheimer
disease pathology and cerebral infarctions. Neurology, 64, 834–41.
tion, and rate of cognitive decline from typical AD. These data are
Braak, H. and Braak, E. (1991). Neuropathological staging of Alzheimer-related
confirmed in another large series of autopsy-confirmed AD cases
changes. Acta Neuropathologica, 82, 239–59.
where typical AD accounted for 82.5%, hippocampal-sparing and
Braak, H. and Del Tredici, K. (2011). The pathological process underlying
limbic-predominant forms for around 9% each (Jellinger, 2012d). Alzheimer’s disease in individuals under thirty. Acta Neuropathologica,
Given that quantitative assessment of tau pathology alone in AD 121, 171–81.
cases points towards new clinicopathological phenotypes, we Braak, H., et al. (2003). Staging of brain pathology related to sporadic
believe that a more accurate, quantitative assessment of various Parkinson’s disease. Neurobiology of Aging, 24, 197–211.
neuropathological lesions is necessary to further elucidate possible Braak, H., et al. (2006). Staging of Alzheimer disease-associated neurofibrillary
mutual relationships between pathologies as well as their combined pathology using paraffin sections and immunocytochemistry. Acta
influence on the clinical picture. A deeper knowledge of underly- Neuropathologica, 112, 389–404.
ing pathologies in age-associated neurodegeneration and cerebrov- Braak, H., et al. (2011). Stages of the pathologic process in Alzheimer disease:
age categories from 1 to 100 years. Journal of Neuropathology and
ascular disorders is paramount to identify suitable targets for new
Experimental Neurology, 70, 960–9.
therapeutic approaches. Budka, H., et al. (2011). Sporadic Creutzfeldt-Jakob disease. In:
Dickson, D.W. and Weller, R.O. (eds) Neurodegeneration: the molecular
References pathology of dementia and movement disorders, 2nd edn. Chichester :
Aguzzi, A. and Heikenwalder, M. (2006). Pathogenesis of prion diseases: Wiley-Blackwell.
current status and future outlook. Nature Reviews Microbiology, 4, Buratti, E. and Baralle, F.E. (2008). Multiple roles of TDP-43 in gene
765–75. expression, splicing regulation, and human disease. Frontiers in
Alafuzoff, I., et al. (2008). Staging of neurofibrillary pathology in Alzheimer’s Bioscience, 13, 867–78.
disease: a study of the BrainNet Europe Consortium. Brain Pathology, Cairns, N.J. (2011). Neuronal intermediate filament inclusion disease. In:
18, 484–96. Dickson, D.W. and Weller, R.O. (eds) Neurodegeneration: the molecular
pathology of dementia and movement disorders, 2nd edition. Chichester: Giasson, B.I., et al. (2003). Initiation and synergistic fibrillization of tau and
Wiley-Blackwell. alpha-synuclein. Science, 300, 636–40.
Cairns, N.J., et al. (2007a). Neuropathologic diagnostic and nosologic criteria Glenner, G.G. and Wong, C.W. (1984). Alzheimer’s disease and Down’s
for frontotemporal lobar degeneration: consensus of the Consortium for syndrome: sharing of a unique cerebrovascular amyloid fibril protein.
Frontotemporal Lobar Degeneration. Acta Neuropathologica, 114, 5–22. Biochemistry and Biophysical Research Communications, 122, 1131–5.
Cairns, N.J., et al. (2007b). TDP-43 in familial and sporadic frontotemporal Goedert, M. (2001). Alpha-synuclein and neurodegenerative diseases. Nature
lobar degeneration with ubiquitin inclusions. American Journal of Review Neuroscience, 2, 492–501.
Pathology, 171, 227–40. Goedert, M., et al. (2012). Frontotemporal dementia: implications for
Chalmers, K., et al. (2009). Validation of international consensus criteria understanding Alzheimer disease. Cold Spring Harbor Perspectives in
for the assessment of cerebral amyloid angiopathy in post-mortem brain. Medicine, 2, a006254.
Proceedings of Sixth International Conference on Vascular Dementia, Gold, G., et al. (2007). Identification of Alzheimer and vascular lesion
Barcelona. thresholds for mixed dementia. Brain, 130, 2830–6.
Chui, H.C. (2006). Vascular cognitive impairment: today and tomorrow. Gomez-Isla, T., et al. (1996). Profound loss of layer II entorhinal cortex
Alzheimer’s and Dementia, 2, 185–94. neurons occurs in very mild Alzheimer’s disease. Journal of Neuroscience,
Chui, H.C., et al. (1992). Criteria for the diagnosis of ischemic vascular 16, 4491–500.
dementia proposed by the State of California Alzheimer’s Disease Gorelick, P.B., et al. (2011). Vascular contributions to cognitive impairment
Diagnostic and Treatment Centers. Neurology, 42, 473–80. and dementia: a statement for healthcare professionals from the American
Chui, H.C., et al. (2006). Cognitive impact of subcortical vascular and Heart Association/American Stroke Association. Stroke, 42, 2672–713.
Alzheimer’s disease pathology. Annals of Neurology, 60, 677–87. Gotz, J., et al. (2010). Dissecting toxicity of tau and beta-amyloid.
Clark, H.B. (2011). Introduction to trinucleotide repeat disorders. In: Neurodegenerative Diseases, 7, 10–12.
Dickson, D.W. and Weller, R.O. (eds) Neurodegeneration: the molecular Greenberg, S.M. (1998). Cerebral amyloid angiopathy: prospects for clinical
pathology of dementia and movement disorders, 2nd edition. Chichester: diagnosis and treatment. Neurology, 51, 690–4.
Wiley-Blackwell. Greenberg, S.M., et al. (2009). Cerebral microbleeds: a guide to detection and
Clinton, L.K., et al. (2010). Synergistic Interactions between Abeta, tau, and interpretation. Lancet Neurology, 8, 165–74.
alpha-synuclein: acceleration of neuropathology and cognitive decline. Grinberg, L.T. and Thal, D.R. (2010). Vascular pathology in the aged human
Journal of Neuroscience, 30, 7281–9. brain. Acta Neuropathologica, 119, 277–90.
Compta, Y., et al. (2011). Lewy- and Alzheimer-type pathologies in Parkinson’s Halliday, G.M., et al. (2003). Identifying severely atrophic cortical subregions
disease dementia: which is more important? Brain, 134, 1493–505. in Alzheimer’s disease. Neurobiology of Aging, 24, 797–806.
Davidson, Y., et al. (2007). Ubiquitinated pathological lesions in Halliday, G.M., Song Y.J., and, Harding, A.J. (2011). Striatal beta-amyloid in
frontotemporal lobar degeneration contain the TAR DNA-binding dementia with Lewy bodies but not Parkinson’s disease. Journal of Neural
protein, TDP-43. Acta Neuropathologica, 113, 521–33. Transmission, 118, 713–19.
Davidson, Y.S., et al. (2011). TDP-43 pathological changes in early onset Halliday, G.M., et al. (2011). Neuropathology underlying clinical variability
familial and sporadic Alzheimer’s disease, late onset Alzheimer’s disease in patients with synucleinopathies. Acta Neuropathologica , 122,
and Down’s Syndrome: association with age, hippocampal sclerosis and 187–204.
clinical phenotype. Acta Neuropathologica, 122, 703–13. Halliday, G.M., et al. (2012). Mechanisms of disease in frontotemporal
Dickson, D.W. (2009). Neuropathology of non-Alzheimer degenerative lobar degeneration: gain of function versus loss of function effects. Acta
disorders. International Journal of Clinical and Experimental Pathology, Neuropathologica, 124, 373–82.
3, 1–23. Hardy, J. (2003). The relationship between amyloid and tau. Journal of
Dickson, D.W. (2011). Introduction to neurodegeneration: the molecular Molecular Neuroscience, 20, 203–6.
pathology of dementia and movement disorders. In: Dickson, D.W. and Hardy, J.A. and Higgins, G.A. (1992). Alzheimer’s disease: the amyloid
Weller, R.O. (eds) Neurodeneration: the molecular pathology of dementia cascade hypothesis. Science, 256, 184–5.
and movement disorders, 2nd edition. Chichester: Wiley-Blackwell. Hardy, J. and Selkoe, D.J. (2002). The amyloid hypothesis of Alzheimer’s
Dickson, D.W., et al. (2009). Neuropathological assessment of Parkinson’s disease: progress and problems on the road to therapeutics. Science, 297,
disease: refining the diagnostic criteria. Lancet Neurology, 8, 1150–7. 353–6.
Dickson DW, et al. (2010). Neuropathology of variants of progressive Hedreen, J.C. and Roos, R.A.C. (2011). Huntington’s disease. In: Dickson,
supranuclear palsy. Current Opinions in Neurology, 23, 394–400. D.W., and Weller, R.O. (eds) Neurodegeneration: the molecular
Dickson, D.W., et al. (2011). Progressive supranuclear palsy and corticobasal pathology of dementia and movement disorders, 2nd edition. Chichester:
degeneration. In: Dickson, D.W. and Weller, R.O. (eds) Neurodegeneration: Wiley-Blackwell.
the molecular pathology of dementia and movement disorders, 2nd edition. Holton, J.L., Lees, A.J., and Revesz, T. (2011). Multiple system atrophy. In:
Chichester: Wiley-Blackwell. Dickson, D.W. and Weller, R.O. (eds) Neurodegeneration: the molecular
Duyckaerts, C. and Dickson, D.W. (2011). Neuropathology of Alzheimer’s pathology of dementia and movement disorders, 2nd edition. Chichester:
disease and its variants. In: Dickson, D.W. and Weller, R.O. (eds) Wiley-Blackwell.
Neurodegeneration: the molecular pathology of dementia and movement Huntington’s Disease Collaborative Research Group (1993). A novel
disorders, 2nd edition. Chichester: Wiley-Blackwell. gene containing a trinucleotide repeat that is expanded and unstable
Duyckaerts, C. and Hauw, J.J. (1997). Prevalence, incidence and duration of on Huntington’s disease chromosomes. The Huntington’s Disease
Braak’s stages in the general population: can we know? Neurobiology of Collaborative Research Group. Cell, 72, 971–83.
Aging, 18, 362–9; discussion 389–92. Hyman, B.T. (1998). New neuropathological criteria for Alzheimer disease.
Duyckaerts, C., Delatour, B., and Potier, M.C. (2009). Classification and basic Archives of Neurology, 55, 1174–6.
pathology of Alzheimer disease. Acta Neuropathologica, 118, 5–36. Hyman, B.T., et al. (2012). National Institute on Aging-Alzheimer’s
Fujino, Y., et al. (2005). Increased frequency of argyrophilic grain disease in Association guidelines for the neuropathologic assessment of Alzheimer’s
Alzheimer disease with 4R tau-specific immunohistochemistry. Journal disease. Alzheimer’s and Dementia, 8, 1–13.
of Neuropathology and Experimental Neurology, 64, 209–14. Ince, P. (2011). Dementia with Lewy bodies and Parkinson’s disease dementia.
Giasson, B.I., et al. (2002). Neuronal alpha-synucleinopathy with severe In: Dickson, D.W. and Weller, R.O. (eds) Neurodegeneration: the molecular
movement disorder in mice expressing A53T human alpha-synuclein. pathology of dementia and movement disorders, 2nd edition. Chichester:
Neuron, 34, 521–33. Wiley-Blackwell.
Ittner, L.M. and Gotz, J. (2010). Amyloid-beta and tau—a toxic pas de deux in Josephs, K.A., et al. (2008). Beta-amyloid burden is not associated with rates
Alzheimer’s disease. Nature Review Neuroscience, 12, 65–72. of brain atrophy. Annals of Neurology, 63, 204–12.
Jankovic, J. (2008). Parkinson’s disease: clinical features and diagnosis. Journal Josephs, K.A., et al. (2011). Neuropathological background of phenotypical
of Neurology, Neurosurgery and Psychiatry, 79, 368–76. variability in frontotemporal dementia. Acta Neuropathologica, 122,
Jellinger, K.A. (2002). Alzheimer disease and cerebrovascular pathology: an 137–53.
update. Journal of Neural Transmission, 109, 813–36. Kalaria, R.N., et al. (2004). Towards defining the neuropathological
Jellinger, K.A. (2007a). The enigma of mixed dementia. Alzheimer’s and substrates of vascular dementia. Journal of Neurological Science, 226,
Dementia: Journal of the Alzheimer’s Association, 3, 40–53. 75–80.
Jellinger, K.A. (2007b). The enigma of vascular cognitive disorder and Karbowski, M. and Neutzner, A. (2012). Neurodegeneration as a consequence
vascular dementia. Acta Neuropathologica, 113, 349–88. of failed mitochondrial maintenance. Acta Neuropathologica, 123(2),
Jellinger, K.A. (2008). The pathology of ‘vascular dementia’: a critical update. 157–71.
Journal of Alzheimer’s Disease, 14, 107–23. Kouri, N., et al. (2011). Corticobasal degeneration: a pathologically distinct
Jellinger, K.A. (2010). Basic mechanisms of neurodegeneration: a critical 4R tauopathy. Nature Reviews Neurology, 7, 263–72.
update. Journal of Cell Molecular Medicine, 14, 457–87. Kovacs, G.G., et al. (2008). Mixed brain pathologies in dementia: the BrainNet
Jellinger, K.A. (2011a). Interaction between alpha-synuclein and tau in Europe Consortium experience. Dementia and Geriatric Cognitive
Parkinson’s disease comment on Wills et al.: elevated tauopathy and Disorders, 26, 343–50.
alpha-synuclein pathology in postmortem Parkinson’s disease brains Launer, L.J., et al. (2008). AD brain pathology: vascular origins? Results from
with and without dementia. Experimental Neurology 225, 210–18. the HAAS autopsy study. Neurobioloy of Aging, 29, 1587–90.
Experimental Neurology, 227, 13–18. Lee, J.H., et al. (2000). Small concomitant vascular lesions do not influence
Jellinger, K.A. (2011b). Parkinson’s disease. In: Dickson, D.W. and Weller, rates of cognitive decline in patients with Alzheimer disease. Archives of
R.O. (eds) Neurodegeneration: the molecular patholgy of dementia and Neurology, 57, 1474–9.
movement disorders, 2nd edition. Chichester: Wiley-Blackwell. Leverenz, J.B., et al. (2008). Empiric refinement of the pathologic assessment
Jellinger, K.A. (2011d). Criteria for the neuropathological diagnosis of of Lewy-related pathology in the dementia patient. Brain Pathology, 18,
dementing disorders: routes out of the swamp? In: Braissant, O., et al. 220–4.
(eds) Recent Researches in Modern Medicine. Proceedings of the WSEAS Lipton, A.M., White, C.L., 3rd, and Bigio, E.H. (2004). Frontotemporal lobar
International Conference, 23–25 Feb, Cambridge. WSEAS Press. degeneration with motor neuron disease-type inclusions predominates
Jellinger, K.A. (2012a). Interaction between pathogenic proteins in in 76 cases of frontotemporal degeneration. Acta Neuropathologica
neurodegenerative disorders. Journal of Cellular Molecular Medicine, (Berlin), 108, 379–85.
16(6), 1168–83. Mackenzie, I.R., et al. (2009). Nomenclature for neuropathologic subtypes of
Jellinger, K.A. (2012b). Neuropathology of sporadic Parkinson’s disease: frontotemporal lobar degeneration: consensus recommendations. Acta
evaluation and changes of concepts. Movement Disorders , 27 (1), Neuropathologica, 117, 15–8.
8–30. Mackenzie, I.R., et al. (2010). Nomenclature and nosology for neuropathologic
Jellinger, K.A. (2012c). The role of α-synuclein in neurodegeneration—an subtypes of frontotemporal lobar degeneration: an update. Acta
update. Translational Neuroscience, 3, 75–122. Neuropathologica, 119, 1–4.
Jellinger, K.A. (2012d). Neuropathological subtypes of Alzheimer’s disease Mackenzie, I.R., et al. (2011). A harmonized classification system for
(Correspondence). Acta Neuropathologica, 123, 153–4. FTLD-TDP pathology. Acta Neuropathologica, 122, 111–13.
Jellinger, K.A. (2013). Challenges in the neuropathological diagnosis of Mandelkow, E.M. and Mandelkow, E. (2012). Biochemistry and cell biology
dementias. International Journal of Neuropathology, 1, 8–52. of tau protein in neurofibrillary degeneration. Cold Spring Harbor
Jellinger, K.A. and Attems, J. (2003). Incidence of cerebrovascular lesions in Perspectives in Medicine, 2, a006247.
Alzheimer’s disease: a postmortem study. Acta Neuropathologica, 105, Masters, C.L. and Beyreuther, K. (2011). Amyloid-β production. In:
14–7. Dickson, D.W. and Weller, R.O. (eds) Neurodegeneration: the molecular
Jellinger, K.A. and Attems, J. (2005). Prevalence and pathogenic role of pathology of dementia and movement disorders, 2nd edition. Chichester:
cerebrovascular lesions in Alzheimer disease. Journal of Neurological Wiley-Blackwell.
Science, 229–230, 37–41. Matthews, F.E., et al. (2009). Epidemiological pathology of dementia:
Jellinger, K.A. and Attems, J. (2006). Does striatal pathology distinguish attributable-risks at death in the Medical Research Council Cognitive
Parkinson disease with dementia and dementia with Lewy bodies? Acta Function and Ageing Study. PLoS Medicine, 6, e1000180.
Neuropathologica, 112, 253–60. McKeith, I.G., et al. (2005). Diagnosis and management of dementia with
Jellinger, K.A. and Attems, J. (2007a). Neurofibrillary tangle-predominant Lewy bodies: third report of the DLB Consortium. Neurology, 65,
dementia: comparison with classical Alzheimer disease. Acta 1863–72.
Neuropathologica, 113, 107–17. Mirra, S.S., et al. (1991). The Consortium to Establish a Registry for Alzheimer’s
Jellinger, K.A. and Attems, J. (2007b). Neuropathological evaluation of mixed Disease (CERAD). Part II. Standardization of the neuropathologic
dementia. Journal of the Neurological Sciences, 257, 80–7. assessment of Alzheimer’s disease. Neurology, 41, 479–86.
Jellinger, K.A. and Attems, J. (2008). Prevalence and impact of vascular and Mirra, S.S., Hart, M.N., and Terry, R.D. (1993). Making the diagnosis of
Alzheimer pathologies in Lewy body disease. Acta Neuropathologica, Alzheimer’s disease. A primer for practicing pathologists. Archives of
115, 427–36. Pathology and Laboratory Medicine, 117, 132–44.
Jellinger, K.A. and Attems, J. (2010a). Prevalence and pathology of vascular Montine, T.J., et al. (2012). National Institute on Aging-Alzheimer’s Association
dementia in the oldest-old. Journal of Alzheimer’s Disease, 1283–1298. guidelines for the neuropathologic assessment of Alzheimer’s disease: a
Jellinger, K.A. and Attems, J. (2010b). Prevalence of dementia disorders in the practical approach. Acta Neuropathologica, 123, 1–11.
oldest-old: an autopsy study. Acta Neuropathologica, 119, 421–33. Munoz, G.D., Morris, H.R., and Rossor, M. (2011). Pick’s disease. In:
Jellinger, K.A. and Attems, J. (2012). Neuropathology and general autopsy Dickson, D.W. and Weller, R.O. (eds) Neurodegeneration: the molecular
findings in nondemented aged subjects. Clinical Neuropathology, 31, pathology of dementia and movement disorders, 2nd edition. Chichester:
87–98. Wiley-Blackwell.
Jellinger, K.A. and Lantos, P.L. (2010). Papp-Lantos inclusions and Murray, M.E., et al. (2011). Neuropathologically defined subtypes of
the pathogenesis of multiple system atrophy: an update. Acta Alzheimer’s disease with distinct clinical characteristics: a retrospective
Neuropathologica, 119, 657–67. study. Lancet Neurology, 10, 785–96.
Nagy, Z., et al. (1997). The effects of additional pathology on the cognitive Sayre, L.M., Smith, M.A., and Perry, G. (2001). Chemistry and biochemistry
deficit in Alzheimer disease. Journal of Neuropathology and Experimental of oxidative stress in neurodegenerative disease. Current Medicinal
Neurology, 56, 165–70. Chemistry, 8, 721–738.
Naslund, J., et al. (1994). Relative abundance of Alzheimer A beta amyloid Schmidt, R., et al. (2011). Heterogeneity in age-related white matter changes.
peptide variants in Alzheimer disease and normal aging. Proceedings of Acta Neuropathologica, 122, 171–85.
the National Academy of Science USA, 91, 8378–82. Schneider, J.A., et al. (2004). Cerebral infarctions and the likelihood of
Nath, U., et al. (2001). The prevalence of progressive supranuclear palsy dementia from Alzheimer disease pathology. Neurology, 62, 1148–55.
(Steele-Richardson-Olszewski syndrome) in the UK. Brain, 124, Schneider, J.A., et al. (2007a). Mixed brain pathologies account for most
1438–49. dementia cases in community-dwelling older persons. Neurology, 69,
Nelson, P.T., et al. (2007). Clinicopathologic correlations in a large Alzheimer 2197–204.
disease center autopsy cohort: neuritic plaques and neurofibrillary tangles Schneider, J.A., et al. (2007b). Subcortical infarcts, Alzheimer’s disease
‘do count’ when staging disease severity. Journal of Neuropathology and pathology, and memory function in older persons. Annals of Neurology,
Experimental Neurology, 66, 1136–46. 62, 59–66.
Nelson, P.T., et al. (2011). Hippocampal sclerosis in advanced age: clinical Seltman, R.E. and Matthews, B.R. (2012). Frontotemporal lobar degeneration:
and pathological features. Brain, 134, 1506–18. epidemiology, pathology, diagnosis and management. CNS Drugs, 26,
Nelson, P.T., et al. (2012). Correlation of Alzheimer disease neuropathologic 841–70.
changes with cognitive status: a review of the literature. Journal of Shoulson, I. and Young, A.B. (2011). Milestones in Huntington disease.
Neuropathology and Experimental Neurology, 71, 362–81. Movement Disorders, 26, 1127–33.
Neumann, M. and Mackenzie, I.R.A. (2011). Frontotemporal lobar Spillantini, M.G. (2011). Introduction to α-sunucleinopathies. In: Dickson,
degeneration with FUS immunoreactive inclusions. In: Dickson, D.W. and D.W. and Weller, R.O. (eds) Neurodegeneration: the molecular pathology
Weller, R.O. (eds) Neurodegeneration: the molecular pathology of dementia of dementia and movement disorders, 2nd edition. Chichester :
and movement disorders, 2nd edition. Chichester: Wiley-Blackwell. Wiley-Blackwell.
Neumann, M., et al. (2006). Ubiquitinated TDP-43 in frontotemporal lobar Steele, J.C., Richardson, J.C., and Olszewski, J. (1964). Progressive
degeneration and amyotrophic lateral sclerosis. Science, 314, 130–3. supranuclear palsy. A heterogeneous degeneration involving the brain
O’Brien, J.T., et al. (2003). Vascular cognitive impairment. Lancet Neurology, stem, basal ganglia and cerebellum with vertical gaze and pseudobulbar
2, 89–98. palsy, nuchal dystonia and dementia. Archives of Neurology, 10,
Oddo, S., et al. (2003). Triple-transgenic model of Alzheimer’s disease with 333–59.
plaques and tangles: intracellular Abeta and synaptic dysfunction. Strozyk, D., et al. (2010). Contribution of vascular pathology to the clinical
Neuron, 39, 409–21. expression of dementia. Neurobiology of Aging, 31, 1710–20.
Olichney, J.M., et al. (1995). Cerebral infarction in Alzheimer’s disease is Terry, R.D., et al. (1991). Physical basis of cognitive alterations in Alzheimer’s
associated with severe amyloid angiopathy and hypertension. Archives disease: synapse loss is the major correlate of cognitive impairment.
of Neurology, 52, 702–8. Annals of Neurology, 30, 572–80.
Parchi, P., et al. (2011). Phenotypic variability of sporadic human prion disease Thal, D.R., et al. (2002a). Two types of sporadic cerebral amyloid angiopathy.
and its molecular basis: past, present, and future. Acta Neuropathologica, Journal of Neuropathological and Experimental Neurology, 61, 282–93.
121, 91–112. Thal, D.R., et al. (2002b). Phases of A beta-deposition in the human
Parchi, P., et al. (2012). Consensus classification of human prion disease brain and its relevance for the development of AD. Neurology, 58,
histotypes allows reliable identification of molecular subtypes: an 1791–800.
inter-rater study among surveillance centres in Europe and USA. Acta Thal, D.R., et al. (2003). Vascular pathology in Alzheimer disease: correlation
Neuropathologica, 124, 517–29. of cerebral amyloid angiopathy and arteriosclerosis/lipohyalinosis with
Parkkinen, L., Pirttila, T., and Alafuzoff, I. (2008). Applicability of current cognitive decline. Journal of Neuropathology and Experimental Neurology,
staging/categorization of alpha-synuclein pathology and their clinical 62, 1287–1301.
relevance. Acta Neuropathologica, 115, 399–407. Thal, D.R., et al. (2010). Capillary cerebral amyloid angiopathy identifies a
Perry, G., et al. (1998). Apoptosis and Alzheimer’s disease. Science, 282, distinct APOE epsilon4-associated subtype of sporadic Alzheimer’s
1268–9. disease. Acta Neuropathologica, 120, 169–83.
Petrovitch, H., et al. (2005). AD lesions and infarcts in demented and Thal, D.R., et al. (2011). Stages of granulovacuolar degeneration: their
non-demented Japanese-American men. Annals of Neurology, 57, relation to Alzheimer’s disease and chronic stress response. Acta
98–103. Neuropathologica, 122, 577–89.
Pezzini, A., et al. (2009). Cerebral amyloid angiopathy: a common cause of Tolnay, M. and Braak, H. (2011). Argyrophilic grain disease. In: Dickson,
cerebral hemorrhage. Current Medicinal Chemistry, 16, 2498–513. D.W. and Weller, R.O. (eds) Neurodegeneration: the molecular pathology
Pfeifer, L.A., et al. (2002). Cerebral amyloid angiopathy and cognitive of dementia and movement disorders, 2nd edition. Chichester :
function: the HAAS autopsy study. Neurology, 58, 1629–34. Wiley-Blackwell.
Rauramaa, T., et al. (2011). TAR-DNA binding protein-43 and alterations in Tomonaga, M. (1981). Cerebral amyloid angiopathy in the elderly. Journal of
the hippocampus. Journal of Neural Transmission, 118, 683–9. the American Geriatrics Society, 29, 151–7.
Robinson, J.L., et al. (2011). Neocortical and hippocampal amyloid-beta Troncoso, J.C., et al. (2008). Effect of infarcts on dementia in the Baltimore
and tau measures associate with dementia in the oldest-old. Brain, 134, longitudinal study of aging. Annals of Neurology, 64, 168–76.
3705–12. Uchikado, H., et al. (2006). Alzheimer disease with amygdala Lewy bodies:
Rohrer, J.D., et al. (2011a). The clinical and neuroanatomical phenotype of a distinct form of alpha-synucleinopathy. Journal of Neuropathology and
FUS associated frontotemporal lobar degeneration. Journal of Neurology, Experimental Neurology, 65, 685–97.
Neurosurgery and Psychiatry, 82, 1405–7. Vernooij, M.W., et al. (2008). Prevalence and risk factors of cerebral
Rohrer, J.D., et al. (2011b). Clinical and neuroanatomical signatures of tissue microbleeds: the Rotterdam Scan Study. Neurology, 70, 1208–14.
pathology in frontotemporal lobar degeneration. Brain, 134, 2565–81. Vinters, H.V. (1987). Cerebral amyloid angiopathy. A critical review. Stroke,
Saito, Y., (2002). Severe involvement of ambient gyrus in dementia with grains. 18, 311–24.
Journal of Neuropathology and Experimental Neurology, 61, 789–96. Vinters, H.V. and Gilbert, J.J. (1983). Cerebral amyloid angiopathy: incidence
Santa-Maria, I., et al. (2012). The MAPT H1 haplotype is associated with and complications in the aging brain. II. The distribution of amyloid
tangle-predominant dementia. Acta Neuropathologica, 124, 693–704. vascular changes. Stroke, 14, 924–8.
Vonsattel, J.P., et al. (1991). Cerebral amyloid angiopathy without and with Wyllie, A.H. (1997). Apoptosis: an overview. British Medical Bulletin, 53,
cerebral hemorrhages: a comparative histological study. Annals of 451–465.
Neurology, 30, 637–49. Yamada, T., McGeer, P.L., and McGeer, E.G. (1992). Appearance of paired
Waxman, E.A. and Giasson, B.I. (2009). Molecular mechanisms of nucleated, tau-positive glia in patients with progressive supranuclear
alpha-synuclein neurodegeneration. Biochimica Biophysica Acta, 1792, palsy brain tissue. Neuroscience Letters, 135, 99–102.
616–24. Younkin, S.G. (1995). Evidence that A beta 42 is the real culprit in Alzheimer’s
Weller, R.O., et al. (2009). Lymphatic drainage of the brain and the pathophysiology disease. Annals of Neurology, 37, 287–8.
of neurological disease. Acta Neuropathologica, 117, 1–14. Zekry, D., et al. (2003). The vascular lesions in vascular and mixed dementia:
Will, R.G., et al. (1996). A new variant of Creutzfeldt-Jakob disease in the the weight of functional neuroanatomy. Neurobiology of Aging, 24,
UK. Lancet, 347, 921–5. 213–9.
Wills, J., et al. (2010). Elevated tauopathy and alpha-synuclein pathology Zhukareva, V., et al. (2002). Sporadic Pick’s disease: a tauopathy characterized
in postmortem Parkinson’s disease brains with and without dementia. by a spectrum of pathological tau isoforms in gray and white matter.
Experimental Neurology, 225, 210–8. Annals of Neurology, 51, 730–9.
CHAPTER 7
Neurochemical
pathology of dementia
Margaret Ann Piggott
This chapter focuses on the neurodegenerative disorders Alzheimer’s forebrain neurons (Whitehouse et al., 1982) and cortical choline
disease (AD), Lewy body dementias (dementia with Lewy bodies acetyltransferase (ChAT, the enzyme that synthesizes ACh) and
(DLB) and Parkinson’s disease dementia (PDD)), frontotemporal acetylcholinesterase (AChE, which breaks down ACh) activities
dementia (FTD), and vascular dementia (VaD) which results from (Perry et al., 1978; Wilcock et al., 1982; Francis et al., 1999). These
cerebrovascular disease. These different conditions, which give rise studies predominantly involved autopsy tissue, which generally
to dementia syndromes, each have distinct neurochemical pathol- reflect end-stage processes. However, ChAT activity from a post-
ogies, with important implications for treatment. As increased mortem series including individuals without a diagnosis of demen-
age is the common risk factor generally associated with dement- tia in life but with significant Alzheimer-type pathology indicated
ing illnesses, neurochemical changes are set in the context of the reduced ChAT, interpreted as demonstrating cholinergic dysfunc-
changes that occur during ageing. A detailed understanding of the tion in the preclinical disease stage (Potter et al., 2011). Biopsy
neurotransmitter function in each condition can lead to rational investigations have also been undertaken (reviewed in Francis
drug design and treatment strategies appropriate for each group of et al., 1999). These biopsy studies of AD patients, on average 3.5
patients. Neurochemical pathology in transmitter systems underly- years after diagnosis, indicated that cholinergic neurotransmitter
ing clinical features of these disorders will be reviewed. pathology is likely to occur relatively early in the course of the dis-
ease, with cortical cholinergic presynaptic markers (ChAT activity,
Alzheimer’s Disease high affinity choline uptake, and ACh synthesis) being reduced to
about 50% of control values. In vivo mapping of cholinergic ter-
Alzheimer’s disease is the most common age-associated dementia, minals by SPECT, using [123I]iodobenzovesamicol (IBVM) binding
with prevalence rising from 4% at 65 to 20% after the age of 80 to the vesicular ACh transporter, demonstrated reduced choliner-
(Evans et al., 2004). Several neurotransmitter systems are affected gic innervation of the neocortex and hippocampus in AD (Kuhl et
in AD, undoubtedly combining their effects and resulting in a range al., 1996). Subsequently, Mazère et al. found reduced SPECT [123I]
of clinical symptoms. IBVM binding in early AD (MMSE scores higher than 21) in the
cingulate cortex and parahippocampal/amygdaloid region (which
Acetylcholine includes the area containing the nucleus basalis), suggesting that
The cholinergic system, arising from the basal forebrain (includ- cholinergic degeneration occurs in the early stage of AD (Mazère
ing septal nuclei, diagonal band, and nucleus basalis of Meynert, et al., 2008). High resolution MRI with cross-sectional mapping
which constitute the cholinergic projections to the neocortex), of the basal forebrain cholinergic nuclei area has shown gradually
innervates all areas of the cerebral cortex including hippocampus, reduced volume across the adult age range, with greater reductions
and the reticular nucleus of the thalamus. Cholinergic brainstem in posterior parts of the nucleus basalis in very mild AD, and more
neurons (from the pedunculopontine and laterodorsal tegmental pronounced reductions of the whole basal forebrain cholinergic
nuclei) innervate the thalamus and cerebellum. In contrast, the area in AD (Grothe et al., 2012). CSF ACh content has been noted
highly cholinergic putamen and caudate nucleus do not receive to be reduced to a third of nondemented controls (Jia et al., 2004).
such inputs, but have intrinsic cholinergic neurons (large striatal Importantly, in vivo (Kuhl et al., 1996; Jia et al., 2004; Mazère et al.,
interneurons). Activation of cholinergic receptors is procogni- 2008) and biopsy studies (Francis et al., 1999) have corroborated
tive, especially by improving attention (Voytko, 1996; Perry et al., the association between cholinergic marker loss and the degree of
1999). The importance of acetylcholine (ACh) to cognitive impair- cognitive impairment seen in autopsy investigations (Perry et al.,
ment in AD was recognized 30 years ago, with several early reports 1978; Wilcock et al., 1982; Potter et al., 2011). Others have argued
highlighting the significant reduction in both cholinergic basal that there is an initial stabilization or even up-regulation of cortical
ChAT activity during the early stages of AD (Davis et al., 1999; in Court et al., 2001a). There are two major subtypes of these recep-
DeKosky et al., 2002), but since ChAT activity is not the rate limit- tors: homomeric receptors that are composed of α7 subunits, and
ing step of ACh synthesis, such data do not preclude the likelihood heteromeric receptors. The latter have been demonstrated to con-
of cortical cholinergic synaptic dysfunction occurring at an early tain α4 and β2 subunits by immunoprecipitation in human brain
stage of the disease (Terry and Buccafusco, 2003). tissue, with populations of α2 subunits in the cortex and α6 and β3
Reduced thalamic and striatal ChAT activity in AD has also in the striatum (Gotti et al., 2006). Reductions in cortical hetero-
been observed in some, but not all, autopsy studies (Danielsson meric nAChRs have been observed universally in AD and this has
et al., 1988; Reinikainen et al., 1988; Xuereb et al., 1990), includ- been confirmed to be due to α4 and β2 losses (Gotti et al., 2006). A
ing reduced ChAT immunoreactivity (approximately 40%) in the decline in the α4β2 subtype in the striatum has also been reported
mediodorsal thalamic nucleus (Brandel et al., 1991). In vivo PET (Court et al., 2001a) and this loss is likely to be due to nAChRs
imaging of AChE with [(11)C]-methyl-4-piperidinyl propionate being lost from cortical inputs rather than nigrostriatal afferents
showed no reduction in thalamic cholinergic denervation in mild (Court et al., 2000; Pimlott et al., 2004). That α4β2 neuronal nic-
to moderate AD (Kotagal et al., 2012). Conversely, attenuated tha- otinic receptor loss occurs very early in AD (Perry et al., 2000;
lamic binding to the vesicular cholinergic transporter using IBVM O’Brien et al., 2007) is confirmed by in vivo PET imaging showing
has been reported in early onset AD (Kuhl et al., 1996). A reduc- reductions in both AD and mild cognitive impairment in patients
tion in the expression of ChAT mRNA has also been noted in the who later developed AD (Kendziorra et al., 2011), especially in
dorsal striatum in AD (Boissiere et al., 1997). These studies indicate brain regions affected by AD pathology (Sabri et al., 2008). This
that cholinergic dysfunction in AD may progress to extend beyond might give prognostic information about conversion from MCI to
basal forebrain projections. AD, although one study did not find a significant reduction in mild
AD (Mitsis et al., 2009).
Cholinergic receptors Cortical and hippocampal α7 receptor deficits have also been
Post-mortem studies have demonstrated changes in cholinergic reported by some but not by all investigators (Court et al., 2001a),
receptors in AD. Multiple muscarinic G-protein linked ACh receptors possibly depending on disease severity, but also complicated
(mAChRs) are expressed in human brain (M1–5), the most prevalent because, since α7 receptors are present on astrocytes and these
being M1 in cortical regions, and M2 (widely distributed). For M1 are up-regulated by gliosis in AD (Teaktong et al., 2003; Yu et al.,
mAChRs (predominantly postsynaptic), ligand binding studies indi- 2005), loss of the α7receptor subtype from neurons may be masked.
cate no loss in hippocampus (Mulugeta et al., 2003) or limbic cortex Alpha-7 receptors are also important as they likely interact with
or striatum (Piggott et al., 2003) or frontal or temporal cortex (Lai et β-amyloid (Parri and Dineley 2010; Parri et al., 2011), colocalizing
al., 2001). Modest changes have been reported, with reductions in with amyloid plaques (Yu et al., 2012), but not in all reports, e.g.
the hippocampus especially in late stage disease (Rodriguez-Puertas Small et al. (2007) found Aβ binds to membrane lipids but not the
et al., 1997). Conversely, slightly increased cortical density possibly α7 receptor (Small and Dubois 2007). Consequently, in areas of high
as a corollary of atrophy has been described (Nordberg et al., 1992; Aβ, α7 binding may be increased (Ikonomovic et al., 2009). Genetic
Overk et al., 2010). There is more unanimity in relation to disrup- variation in the CHNRA7 gene for the α7 receptor is associated
tion of M1 signalling in AD, with evidence of impaired coupling of with delusional symptoms in AD (Carson et al., 2008a) and influ-
cortical M1 receptors to G-proteins (Smith et al., 1987; Flynn et al., ences risk of AD (Carson et al., 2008b; Barabash et al., 2009). In the
1995b; Ladner and Lee, 1999; Warren et al., 2008). Coupling is more thalamus, which has high levels of nAChRs, significant deficits have
impaired with increasing severity of dementia (Tsang et al., 2006; not been observed except for a 25% reduction in α7 receptor bind-
Potter et al., 2011) and with severity of neuropathology (Perry et al., ing in the reticular nucleus (Court et al., 2001a), possibly reflecting
1998; Overk et al., 2010; Potter et al., 2011). Defective coupling can the presence of this receptor on basal forebrain cholinergic inputs.
be detected preclinically (Potter et al., 2011).
Reductions in cortical M2 mAChRs, considered to be predomi- Cholinergic treatment implications
nantly presynaptic on cholinergic terminals, have been reported in Cholinergic support through administration of cholinesterase
AD (Flynn et al., 1995a, 1995b; Lai et al., 2001; Mulugeta et al., 2003; inhibitors has efficacy in terms of attenuated cognitive deficits
Piggott et al., 2003), and such reductions are independent of ApoE and improved activities of daily living (Birks 2006; Tsuno, 2009).
allele type (Lai et al., 2006). M2 density appears unchanged in the Perhaps it is not surprising that the effects, although significant,
thalamus (Warren et al., 2007), while elevations have been observed are small, given the reduced coupling and/or loss of cholinergic
in the striatum (Piggott et al., 2003). Relative sparing of cortical M2 receptors in AD, as well as the fairly low ability of these drugs to
receptors in AD is associated with psychosis (Lai et al., 2001). There inhibit cholinesterase. Therapy with cholinesterase inhibitors also
are also indications that M2–G-protein coupling is impaired in AD, leads to some improvement in anxiety, depression, and apathy in
notably in areas most affected by pathology (Cowburn et al., 1996). AD (Feldman et al., 2001), implicating cholinergic mechanisms.
Assessments of M4 mAChRs show some inconsistency; increases in It may be that the cholinergic system is less about cognition than
frontal, temporal, and parietal cortices (Flynn et al., 1995a, 1995b) attention and arousal, and more about noncognitive behavioural
are reported, but others noted no change in M4 binding in fron- aspects (Francis et al., 2010). Treatment with cholinesterase inhibi-
tal cortex and striatum (Rodriguez-Puertas et al., 1997). Selective tors for longer or at higher doses could be beneficial for many
regional reductions in M4 binding in AD have been found in the patients (Sabbagh and Cummings 2011; Howard et al., 2012). New
mediodorsal nucleus of the thalamus (Warren et al., 2007) and in cholinesterase inhibitors, or hybrid types, may also offer advantages
hippocampus (Mulugeta et al., 2003). (Mehta et al., 2012).
Deficits in ionotropic nicotinic ACh receptors (nAChRs) are There seems to be a relationship between reduced cholinergic
widely reported in the hippocampus and neocortex in AD (reviewed transmission and the accumulation of AD-type pathology (Nordberg,
CHAPTER 7 neurochemical pathology of dementia 109
2006), with chronic cholinergic antagonism (in Parkinson’s disease) both NR1 in hippocampus (Ulas and Cotman, 1997; Sze et al., 2001;
being associated with an increased density of plaques and tangles Hynd et al., 2004c) and frontal cortex (Amada et al., 2005). NR2B
(Perry et al., 2003). Conversely, it has been demonstrated that long- and NR2A subunit types are reduced in brain areas susceptible to
term exposure to nicotine (tobacco use) is associated with reduced Alzheimer pathology (Sze et al., 2001; Hynd et al., 2004a). In terms
Aβ deposition (Hellstrom-Lindahl et al., 2004; Court et al., 2005). of the efficacy of NMDA receptor signals in AD it is worthy of note
In addition, treatment with an M1 agonist can reduce CSF amy- that the content of spermidine, a polyamine positive modulator of
loid β 1–42 peptide in patients with AD (Hock et al., 2003), and in these receptors, has been reported to be elevated in temporal cortex
model systems both amyloid and tau pathology (Caccamo et al., (Morrison and Kish, 1995). The activity of ornithine decarboxy-
2006). There is still a great deal of scope for addressing sympto- lase, a key enzyme involved in polyamine synthesis, has also been
matic and disease-modifying therapies through the cholinergic sys- found increased in temporal cortex in AD (Morrison et al., 1998).
tem, with better cholinesterase inhibitors (Mehta et al., 2012), via But since polyamine enhancement of MK801 binding (a measure
M1 receptors (Fisher, 2012), or nicotinic α7 receptors (Bencherif of NMDA channel opening) in areas susceptible to pathology has
and Lippiello, 2010; Pohanka, 2012). been shown to be attenuated in AD (Ragnarsson et al., 2002) and
with increasing age (Piggott et al., 1994), it is doubtful whether an
Glutamate increase in polyamine content would result in increased NMDA
While cholinergic mechanisms are implicated in attention and receptor activity.
arousal, glutamate systems likely underlie deficits of memory. Dewar et al. (1991) observed reductions of AMPA and kainite
Glutamate is the main excitatory neurotransmitter in the cortex receptor binding in the subiculum, and of AMPA but not kainate in
and hippocampus, and loss of glutamatergic neurons in the hip- CA1, in AD. AMPA receptor GluR1 and GluR2/3 subunit density
pocampus and entorhinal cortex occurs early in AD (reviews in was decreased in hippocampus in moderate to severe AD in areas
Schaeffer and Gattaz, 2008; Francis, 2009a; Francis et al., 2010). vulnerable to Alzheimer pathology (Armstrong and Ikonomovic,
Glutamatergic systems are complex, with different reuptake trans- 1996; Carter et al., 2004), while GluR1 RNA was reduced in dentate
porter types, multiple receptor subtypes (NMDA, AMPA, kainate, gyrus (Pellegrini-Giampietro et al., 1994). An increase or no change
and metabotropic-mGluR), subunits and subunit splice variants, in kainate binding has been reported in cortex (Cowburn et al.,
and receptor modulation by many common endogenous molecules 1989; Chalmers et al., 1990). This complex pattern of preservation
(e.g. Mg2+, Zn2+, and polyamines). One family of genes encodes of some subunits and loss of others was explored in the hippocam-
subunits (GluR1–4) that form AMPA-selective receptors, two pus (Armstrong et al., 2003), where, despite marked neuronal loss
families encode kainate-selective subunits (GluR5–7), and two and reductions in NR1, NR2B, and GluR2 subunits, there was pres-
encode NMDA-selective subunits (NMDAR1 andNMDAR2A–D). ervation among others, interpreted as compensatory up-regulation
Furthermore, there is intimate involvement of both neurons and of select subunits in surviving neurons and maintenance of some
glia in glutamate transmission. In addition to reduced overall balance between excitatory and inhibitory tone.
glutamate transmission due to neuron loss, the reuptake system There are three groups of metabotropic glutamate receptors:
is compromised (Procter et al., 1988; Francis, 2003; Chen et al., group I (mGluR1, mGluR5); group II (mGluR2, mGluR3); and
2011). Increased protein expression of enzymes involved in gluta- group III (mGluR4, mGluR6, mGluR7, mGluR8) (for review see
mate metabolism (including glutamine synthetase and glutamate Niswender and Conn, 2010). Metabotropic receptor binding was
dehydrogenase) has been reported in the prefrontal cortex in AD reduced in the subiculum and CA1 but not parahippocampal gyrus,
consistent with, and possibly a result of, increased local concentra- CA3, or dentate (Dewar et al., 1991). The presynaptic mGluR2
tion of extracellular glutamate (Burbaeva et al., 2005). It has been subtype is reduced in hippocampal areas in AD (Richards et al.,
hypothesized that impaired reuptake reduces clearance of glutamate 2010). The postsynaptic Group 1 metabotropic glutamate receptors
from the synapse, reducing NMDA receptor-mediated generation are reduced in the frontal cortex from early stages, correlate with
of long-term potentiation (LTP) and hence cognitive impairment pathology, and are associated with reduced mGluR/phospholipase
(Francis, 2003; Francis et al., 2010). Raised local glutamate concen- C (PLC) signalling and PLCβ1 isoform expression (Albasanz et
tration could also lead to excessive glutamate receptor activation al., 2005). The mGluR5 receptor affects the efficiency of glutamate
and calcium-mediated excitotoxic cell death (Greenamyre et al., neurotransmission (Niswender and Conn, 2010) and is reduced by
1988; Boksha, 2004; Hynd et al., 2004b). 50% in hippocampus (Minuzzi et al., 2009). Metabotropic mGluR5
The measurement of glutamate itself as a marker of presynap- play a role in mediating synaptic dysfunction caused by Aβ (Renner
tic glutamatergic transmission has been utilized, although it is not et al., 2010), with clustering at synapses an early event (Lacor et al.,
only synthesized in nerve terminals, but also integral to protein and 2011).
energy metabolism (Francis, 2003). Some studies indicate reduc-
tions in brain and CSF glutamate in AD, while other reports dem- Glutamate treatment implications
onstrate no change or an increase in glutamate (in the CSF) (Smith Given the likelihood of decreased glutamatergic transmission in
et al., 1985; Jimenez-Jimenez et al., 1998; D’Aniello et al., 2005). AD, it is at first glance surprising that the glutamatergic (NMDA
receptor) antagonist memantine is useful. But memantine has
Glutamate receptors shown modest benefits in cognition, function, and global and
Investigation of the glutamate system has shown evidence for recep- behavioural measures in moderate to severe AD (Herrmann et
tor alterations in AD (reviews in Schaeffer and Gattaz, 2008; Hu al., 2011). Possibly since in states of reduced membrane potential
et al., 2012). NMDA receptor binding is reduced in hippocampus the voltage-dependent Mg2+ blockage of NMDA receptors can be
(Penney et al., 1990; Ulas et al., 1992) and frontal cortex (Chalmers released, leading to excessive and neurotoxic entry of Ca2+ into
et al., 1990; Scheuer et al., 1996). Subunit density is also reduced, neurons, memantine’s effectiveness lies in acting as a substitute
Mg2+ block (Muir, 2006). Also, since memantine is an uncompeti- expression in CA1 (Iwakiri et al., 2005). Binding to the peripheral
tive, low-affinity, open-channel blocker, it is suggested that it may benzodiazepine receptor (incongruously as it occurs in the CNS),
be of benefit in AD by limiting glutamate transmission, as it were which is not coupled to GABAA, was increased in AD, coinciding
reducing signal to noise (Lipton, 2005; Francis, 2009a; Francis et with increased microglia (Gulyas et al., 2009).
al., 2010). Another suggestion for memantine’s mode of action is The GABA system and GABAA receptors are potential targets
that via NMDA inhibition it offsets the disinhibition of forebrain in AD, to address excitatory/inhibitory imbalance, reduce amy-
neuronal circuits resulting from reduced cholinergic, noradrener- loid pathology, and for treatment of behavioural and psychological
gic, serotonergic, and histaminergic neurotransmission (Schmitt, symptoms (Francis et al., 2010; Rissman and Mobley 2011).
2005). Promoting the activity of remaining glutamate neurons
while avoiding excitotoxicity are important treatment strategies; Serotonin
metabotropic receptors are also a focus for the development of Reductions in serotonin function in AD were noted as part of early
therapeutics (Gravius et al., 2010). investigations of neurochemical pathology in AD (Perry et al.,
1993). Cell loss and tangle pathology in the raphe nuclei results
GABA in reduced cortical serotonin in AD, with loss of serotonin inner-
As glutamate is the major excitatory neurotransmitter, so gam- vation and serotonin transporters (5-HTT) related to cognitive
ma-aminobutyric acid (GABA) is the inhibitory neurotransmitter. impairment and disease severity, particularly in early onset AD
While 90% of hippocampal neurons are glutamatergic, the remain- (Arai et al., 1992; Halliday et al., 1992). Changes in the serotonergic
ing 10% are mainly GABAergic, and whether through excessive system have been linked to neuropsychiatric symptoms in AD, such
glutamate transmission or glutamate neuron loss or GABA neu- as psychosis (Holmes et al., 1998; Nacmias et al., 2001), aggression
ron loss, or receptor changes in either system, the balance between (Sukonick et al., 2001; Lai et al., 2003b, 2011a; Garcia-Alloza et al.,
inhibitory and excitatory transmission is under threat in AD, poten- 2004; Zarros et al., 2005) and hyperphagic eating (Tsang et al.,
tially leading to behavioural disturbances (Garcia-Alloza et al., 2010).
2006) and seizures (for review see Rissman and Mobley, 2011). 5HT1A receptor immunoreactivity was reduced in hippocampus
GABA transmission is mediated by two main classes of receptors. (CA1 only) associated with neuron loss in AD Braak stage V/VI
GABAA are ligand-gated chloride ion channel receptors responsi- (Mizukami et al., 2011). Loss in the hippocampus as a whole, how-
ble for fast inhibitory transmission; GABAA receptors are modu- ever, correlated with depression (Lai et al., 2011b), and was asso-
lated by benzodiazepines, and comprise a pentameric assembly of ciated with aggression (Lai et al., 2003b). 5HT(1B/1D) reductions
many possible types of subunits (at least 20 genes code for different correlated with cognitive decline (Garcia-Alloza et al., 2004).
subunits). GABAB are G-protein-coupled pre- and postsynaptic 5HT2 receptors tend to be reduced, with faster decline in MMSE
receptors mediating slow inhibitory transmission. score correlating with 5HT2A loss (Lai et al., 2005) independently
In contrast to the marked deficits in cholinergic and glutamater- of cholinergic loss, or the presence or absence of BPSD. Widespread
gic systems, GABAergic neurons seem resistant to neurodegenera- reduction in 5HT2A receptors in early AD occurred in advance of
tion. Alterations in the GABA system are generally only reported other serotonergic losses (Marner et al., 2012) and was also seen in
to occur at later stages of the disorder as pathology and clinical amnestic MCI (Hasselbalch et al., 2008).
severity mount, and to be associated with selective receptor and 5HT3 receptors appear unaltered, as do 5HT4 receptors; the lat-
subunit changes. Reductions in cortical (frontal and temporal) lev- ter being positively coupled to adenylate cyclase and modulating
els of GABA have been observed (Lowe et al., 1988) in association the release of other transmitters (Lai et al., 2003a). Whether 5HT4
with depression (Garcia-Alloza et al., 2006). Conversely, increased coupling is compromised in AD (like muscarinic M1 receptors) has
plasma GABA levels were associated with depression and apathy in not been investigated. There are few reports about 5HT5, 5HT6, or
severe AD (Lanctot et al., 2007b). 5HT7; however, 5HT6 receptor loss is apparent in cortex, correlat-
Reduction in hippocampal benzodiazepine binding (GABAARs) ing with aggressive behaviour (Garcia-Alloza et al., 2004).
was correlated with neurofibrillary tangle numbers (Penney et al., These observations suggest that serotonergic disturbance is most
1990), while relatively increased GABAA/benzodiazepine bind- relevant to neuropsychiatric aspects of AD. However, clinical tri-
ing has been associated with depression in AD (Garcia-Alloza als of antidepressants have shown disappointing results (Banerjee
et al., 2006). No loss of GABAA/benzodiazepine sites was detected et al., 2011), and atypical antipsychotics with antagonism at 5HT
by SPECT imaging in mild cognitive impairment or in early AD receptors may be contraindicated (Francis et al., 2010). The 5HT6,
(Pappata et al., 2010). For GABAA subunits, α1 decreased with along with 5HT3 and 5HT4, receptors seem fruitful targets for ther-
tangle pathology, while β2/3, γ1/3, and γ2 were preserved (and for apeutic development (Buhot et al., 2000; Francis et al., 2010).
γ1/3 increased in very late stage disease) in hippocampus CA1–2 As with muscarinic receptors, there may be a relationship between
(Iwakiri et al., 2009). Reductions in GABAA α1 and α5 mRNA serotonin receptors and amyloid pathology, with activation of the
were observed in hippocampus in moderate cognitive impairment 5HT4 receptor stimulating the secretion of nonamyloidogenic
and probable AD compared to a nonimpaired group, and protein forms of the amyloid precursor protein in model systems (Robert et
expression of α1, β1, and β2 GABAARs was not changed, although al., 2001; Lezoualc’h and Robert, 2003).
α5 protein was reduced in CA1/2 in the most severe cases (Rissman
et al., 2003, 2007). Reduced α1, α2, α4, β2, and δ subunits occurred Noradrenaline
at more severe Braak stage (Luchetti et al., 2011). Early studies revealed deficits in the noradrenaline (NA) (or
At early stages of AD, no change or increased hippocampal norepinephrine (NE)) system in the brains of AD patients, with
GABABR1 receptor immunoreactivity (on pyramidal neurons) has reductions in NA being noted in the cingulate gyrus, substan-
been reported, while in advanced disease there is reduced neuronal tia innominata, putamen, hypothalamus, medial nucleus of the
thalamus, and globus pallidus (Arai et al., 1984). Reduced numbers Vascular dementia
of neurons in the noradrenergic locus ceruleus were also observed
in the majority of AD cases (Mann et al., 1980; Weinshenker, 2008). The term vascular dementia (VaD) is used to describe a heteroge-
There is evidence that surviving neurons may compensate by up- neous group of neuropathologies resulting from vascular disease
regulation of tyrosine hydroxylase mRNA and dendritic sprouting or injury. The pathological hallmarks can range from large focal
(Szot et al., 2006), consistent with studies showing increased levels lesions to a number of small vessel diseases, and VaD frequently
of the NA metabolite 3-methoxy-4hydroxyphenylglycol (MHPG) coexists with AD-type pathology. The site, extent, and combination
(Herrmann et al., 2004), indicative of increased NA turnover. The of lesions inevitably affect any resulting neurochemical deficit(s).
norepinephrine transporter is reduced in AD, correlating with Braak Hence it is not easy to summarize the relatively small-scale
stage (Gulyas et al., 2010). Receptors α1D and α2C are reduced in hip- studies of neurochemical changes in VaD that have been pub-
pocampus (Szot et al., 2006). It has been suggested that behavioural lished. Neurotransmitter deficits in VaD have been indicated by
and psychological symptoms of dementia, for example aggression post-mortem, CSF, and in vivo investigations (reviewed in Court
and agitation, may be linked to increased NA activity or adrenore- and Perry, 2003; Court et al., 2003).
ceptor hypersensitivity in AD. β-adrenergic receptor blockers may
have efficacy (Herrmann et al., 2004), and it has been observed
Acetylcholine
that beta-blocker treatment can result in lower incidence of AD Deficits in cortical and hippocampal cholinergic innervation in
(Khachaturian et al., 2006; Rosenberg et al., 2008; Yu et al., 2011). VaD (predominantly multi-infarct dementia, MID), assessed by
ChAT and high affinity choline uptake, have been observed in a
Dopamine number of studies (reviewed in Court and Perry, 2003; Court et al.,
Although dopamine receptors decline continuously across the 2003; Roman and Kalaria, 2006). However, one early investigation
lifespan and may play a role in the gradual decline of cognitive abil- of the number of neurons in the nucleus basalis of Meynert did not
ity, there is no particular loss of dopamine associated with features observe any reduction in MID (Mann et al., 1986), indicating that
of AD. There is no significant loss of substantia nigra neurons, and declines in cortical innervation in this condition are likely to be
no loss of striatal dopamine concentration or of receptors (Perry because of ischaemic disruption of cholinergic projections or neu-
et al., 1990c; Piggott et al., 1999), although there is some decline ronal function rather than extensive loss of cholinergic neurons, as
in dopamine transporter binding in nucleus accumbens with is the case in AD. A post-mortem study suggests that reduced corti-
increased age of onset in AD (Piggott et al., 1999). No satisfactory cal ChAT only occurs in VaD when in combination with AD-type
neurochemical substrate for the varied postural, gait, and move- pathology, and not in ‘pure VaD’ (Perry et al., 2005). This contrasts
ment changes that emerge in many patients with AD progression with the finding of reduced cortical ChAT in CADASIL, a genetic
has been established. However, rather than being dopamine-based, condition with severe vascular pathology (Keverne et al., 2007).
it is likely to be due to Alzheimer pathology and atrophy in basal With the wide variation in cholinergic change in subcortical ischae-
ganglia and brainstem nuclei, and/or extensive cholinergic derange- mic VaD (Nardone et al., 2008), the cholinergic status should be
ment. In spite of mainly preserved dopaminergic function, with the assessed to indicate if a cholinesterase inhibitor would be helpful.
cholinergic loss in AD the resulting imbalance between dopamine There is limited evidence in relation to changes in muscarinic
and ACh could be a substrate for psychosis (Reeves et al., 2009). In receptors in VaD, with little exploration of selected subtype
vivo D2 receptor binding potential in the right hippocampus was changes (M1–5) and none of functional coupling. Evidence from
reported significantly associated with verbal memory perform- M5-deficient mice indicates that this muscarinic subtype particu-
ance (Kemppainen et al., 2003); however, in vivo studies without larly mediates ACh dilation of cerebral blood vessels (Araya et al.,
post-mortem confirmation of diagnosis could consist of heteroge- 2006). A study of nAChRs showed no loss of cortical α4β2 receptor
neous cases, and variations in endogenous dopamine can result in binding or of α4 (and α7) subunit immunoreactivity (when smok-
different measures of D2 receptors in vivo. ing status, which affects receptor binding, was taken into account)
(Martin-Ruiz et al., 2000). This is in striking contrast to AD, DLB,
Neuropeptides and PD. The apparent intactness of at least nicotinic cholinergic
Deficits in neuropeptide Y (NPY) and somatostatin have been receptors in combination with reduced cholinergic innervation
reported, with reductions in CSF NPY correlated with disease dura- indicates that cholinergic medication may be efficacious in VaD,
tion (Edvinsson et al., 1993) and paralleling NPY deficits in plasma and this has some support (Erkinjuntti et al., 2004; Roman, 2005).
(Koide et al., 1995) and hippocampus (Martel et al., 1990), although In vivo imaging of α4β2 nicotinic receptors in VaD showed reduc-
no deficits in neocortex were observed (Allen et al., 1984). In addition in subcortical regions of dorsal thalamus and right caudate,
tion, no change in hippocampal NPY binding was observed (Martel with increases in cuneus cortex in the occipital lobe (Colloby et al.,
et al., 1990). NPY may be neuroprotective as it inhibits evoked gluta- 2011).
mate release, and thus NPY and NPY receptors could be therapeutic
targets (Silva et al., 2005). Reduced cortical/hippocampal somato- Biogenic amines
statin levels and immunoreactivity have also been shown to corre- Disruption of dopaminergic function is also suggested by altera-
late with disease severity (Allen et al., 1984; Dawbarn et al., 1986; tions in dopaminergic parameters in the CSF and striatum in VaD.
Chan-Palay 1987) and cognitive deficit (Dournaud et al., 1995). The The apparently elevated dopamine/homovanillic acid ratio in the
nucleus basalis of Meynert and substantia innominata are notable CSF, possibly suggestive of reduced brain dopamine metabolism,
because of observed elevations in NPY and galanin, presumably together with reduced striatal dopamine uptake markers, indicate
the result of relative neuronal sparing and/or compensation mecha- impaired striatal function (reviewed in Court et al., 2003). This, in
nisms (Allen et al., 1984; Chan-Palay 1988). common with reduced ChAT observed in the caudate and putamen
(the large interneurons of the striatum are cholinergic), is possibly as well as distinguishing features. These differences include fluc-
due to the particular vulnerability of the lenticulostriate vessels. tuations in attention, alertness, and cognitive performance, varying
Further evidence of nonspecific disruption of the striatum in VaD over minutes as well as days and weeks; persistent complex visual
is indicated by reduced levels of serotonin (5-HT) and its metabo- hallucinations; spontaneous parkinsonism; repeated falls; and sensi-
lite 5-hydroxy indole acetic acid (5-HIAA), although no decline tivity to neuroleptics. DLB and PDD are virtually indistinguishable
in 5-HT uptake sites was observed in this region (Gottfries et al., pathologically, but differ by clinical history in that PDD starts with
1994). 5HT1A and 5HT2A receptors measured post mortem were levodopa-responsive Parkinson’s disease (PD) at least 1 year before
increased in temporal cortex in multi-infarct vascular dementia but the onset of dementia; while in DLB, extrapyramidal symptoms
not in subcortical ischaemic vascular dementia or mixed AD/VaD (EPS) start later, and a proportion (about 20%) of patients have no
(Elliott et al., 2009). EPS. These clinical differences between AD and DLB/PDD reflect
Reduced 5-HT and/or 5-HIAA concentrations have been noted distinct neurochemical profiles, which have treatment implications,
in the CSF, hippocampus, and hypothalamus. In contrast, no reduc- such as response to cholinesterase inhibitors and levodopa, and the
tion in neocortical 5-HT uptake sites or receptors was observed, in importance of avoiding neuroleptic exposure in DLB/PDD.
addition to no diminution in the number of dorsal raphe seroton-
ergic neurons (Court et al., 2003). This is again suggestive of local Dopamine
attrition of neuronal fields rather than primary neuronal loss. No Reduced substantia nigra (SN) neuron density and low dopamine
change in the number of noradrenergic neurons in the locus ceru- concentration in the caudate nucleus were reported in DLB (Perry
leus has been noted. et al., 1990c; Marshall et al., 1994); subsequently, dopamine con-
centration and dopamine transporters were shown to be reduced
Neuropeptides in posterior striatum (Piggott et al., 1999) even in DLB cases with
Neuropeptide profiles in VaD have some overlap with AD. little movement disorder (Piggott and Perry 2010) (see Fig. 7.1).
Reductions in a number in CSF neuropeptide concentrations in The pattern of dopamine loss in DLB is less selectively focused on
VaD have been reported, including somatostatin, β-endorphin, the posterior putamen than in PD, and the caudate is more affected
and corticotrophin-releasing factor (Heilig et al., 1995). In contrast, (Piggott et al., 1999). This is a reflection of the pattern of SN neuron
some neuropeptides, including somatostatin and neurotensin, have loss with relative sparing of the medial SN in PD. SN and dopamine
been found to be increased in a number of brain regions, particu- transporter loss in DLB are more symmetrical between hemi-
larly the hypothalamus, possibly the result of reduced inhibitory spheres than in PD (Ransmayr et al., 2001; O’Brien et al., 2004;
serotonergic tone (Gottfries et al., 1994). Walker et al., 2004).
Since neurotransmitters modulate cerebral blood flow (reviewed Dopamine transporter loss affecting both caudate and putamen
in Court et al., 2002), it is possible that neurotransmitter changes rostrocaudally in DLB has been confirmed in vivo (Walker et al.,
in VaD not only contribute to behavioural and cognitive symptoms 2004; Colloby et al., 2005). Imaging dopamine transporters by
via direct effects on neuronal circuits, but also further compromise SPECT and PET is becoming established for differentiation of DLB
cerebral perfusion and blood–brain barrier control. Hence rational from AD (O’Brien et al., 2004; Sinha et al., 2012). Reduced nigros-
transmitter-based therapies may be of particular significance in VaD. triatal dopamine is the substrate for parkinsonism in DLB, as in
PD, but it may be that a moderate reduction in dopaminergic input
Dementia with Lewy Bodies and Parkinson’s results in more severe movement disorder in DLB than it would
in PD. This is due to the lack of compensatory changes, such as
disease dementia increased turnover of dopamine and up-regulation of D2 receptors,
Dementia with Lewy bodies (DLB) and Parkinson’s disease demen- which occur in PD but generally fail to occur in DLB (Piggott et al.,
tia (PDD) have clinicopathological features in common with AD, 1999), as well as the likelihood of other neurotransmitter system
Caudate Putamen
1.00 1.00
Control
125I PE2I binding fmol/mg
0.75 0.75 PD no dementia

PD+dementia
DLB+EPS later
0.50 0.50
DLB no EPS
AD
0.25 0.25
0.00 0.00
Fig. 7.1 Dopamine transporter density in caudae and putamen. AD, Alzheimer’s disease; C, control; DLB, dementia with Lewy bodies; +EPS, –EPS, with or without
extrapyramidal symptoms; PD, Parkinson’s disease; PDD, Parkinson’s disease dementia.
changes contributing to the movement disorder characteristics of Acetylcholine—presynaptic changes

DLB, as evidenced by the axial predominance and relative levo- Cholinergic losses are generally greater than in AD, in terms of
dopa resistance (Bonelli et al., 2004). Nigrothalamic dopamine is presynaptic cortical activities and in the striatum and the pro-
also likely to be reduced in DLB and PDD. The thalamus receives jection from the pedunculopontine nucleus to the thalamus. As
dopamine via nigrostriatal collaterals, which have been shown in AD, there is consistent involvement of the nucleus basalis of
to have depleted dopamine transporter immunoreactivity in the Meynert, but with Lewy body pathology and more extensive cell
1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-treated loss (Tiraboschi et al., 2000). Within the cerebral cortex, ChAT
monkey model of PD (Freeman et al., 2001). and AChE losses post mortem exceed those in AD (except in the
Postsynaptic dopaminergic measures hippocampus, where they are particularly pronounced in AD) and
are apparent early in the disease course (Perry et al., 1990a, 1990c;
Dopamine D2 receptor density is up-regulated in PD without
Tiraboschi et al., 2002). In PD with dementia, in vivo imaging of
dementia, in the striatum by more than 70% (Piggott et al., 1999), a
the vesicular ACh transporter showed extensive losses (Kuhl et al.,
compensatory change tending to ‘damp down’ overactivity of striat-
1996a). As in AD, cholinergic loss and ChAT reductions are cor-
opallidal neurons. D2 receptors in PD are also up-regulated twofold
related with cognitive decline in DLB (Perry et al., 1990d; Samuel
in all examined regions of the thalamus (Piggott et al., 2007a). In
et al., 1997; Tiraboschi et al., 2002) and in PD (Perry et al., 1985;
DLB there is a slight reduction (17%) in D2 receptor density in the
Mattila et al., 2001; Bohnen et al., 2006). ChAT deficits are greater
striatum (Piggott et al., 1999). D2 density was lowest in cases that
in some visual cortical areas in DLB cases with visual hallucinations
had severe neuroleptic sensitivity and was slightly higher in cases
compared to those without, for example in Brodmann area 36 of
that were tolerant of neuroleptic treatment (Piggott et al., 1998); it
the temporal cortex (Perry et al., 1990a; Ballard et al., 2000). It may
may be that cases with lower D2 density are more at risk of sudden
be that a propensity to visual hallucinations is increased with much
catastrophic blockade due to the D2 antagonist action of neurolep-
reduced ACh combined with a relatively active serotonergic system
tics. Failure to up-regulate D2 receptors in DLB could be due to
(Perry et al., 1990b).
intrinsic striatal pathology, with β-amyloid and synuclein pathol-
ogy reported (Duda et al., 2002; Jellinger and Attems 2006; Tsuboi
et al., 2007; Kalaitzakis et al., 2011), or perhaps due to deficits in Nicotinic receptor changes
other systems, for example in thalamo-striatal afferents, reduced Cortical binding to nAChRs containing α4 and β2 subunits is
excitatory glutamatergic inputs from the subthalamic nucleus, or reduced in DLB in common with AD (Perry et al., 1990d; Gotti
cholinergic system changes. et al., 2006; Colloby et al., 2010), but unlike AD, in DLB there is
In the thalamus there is a tendency for lower D2 binding in an apparent correlation between this nAChR deficit and cortical
all regions in DLB without EPS (DLB-EPS), slight but significant ChAT reduction (Reid et al., 2000). An imaging study has shown a
up-regulation only in ventrointermedius (motor function) and lat- correlation with cortical α4β2 and cognitive decline (Colloby et al.,
erodorsal (association) nuclei in DLB with EPS (DLB+EPS), and 2010). Somewhat surprisingly, heteromeric nAChR binding was
moderate, significant up-regulation in PDD cases in the ventroin- relatively preserved in the temporal cortex in cases of DLB with
termedius nucleus alone (Piggott et al., 2007a). D2 receptor bind- disturbed consciousness (Ballard et al., 2002b). Since patients with
ing in the thalamus in DLB or PDD did not vary with cognitive disturbances of consciousness are able, at least some of the time,
decline or visual hallucinations (Piggott et al., 2007a). Although to be more alert than at other times, the neurotransmitter systems
somewhat confounded by neuroleptic use, in DLB and PDD there must be capable of supporting the higher level of awareness. When
was higher thalamic D2 binding in cases with disturbances of con- cholinergic losses are extensive, a higher density of nicotinic recep-
sciousness, particularly in the GABAergic reticular nucleus (Piggott tors may enable small transmitter fluctuations to lead to variations
et al., 2007a). Probably D2 receptors located on reticular nucleus in consciousness and attention. There is also a link with hallucina-
GABAergic neurons will, being inhibitory, help maintain thalamic tions; while there are reductions in other cortical areas, there is rel-
and cortical activity, thus enabling fluctuations. In an environment atively higher α4β2 occipital binding in DLB patients with a recent
of reduced dopamine concentration, relatively higher D2 receptors history of hallucinating (O’Brien et al., 2008). Whether cortical
will amplify small transmitter changes, leading to variations in con- α7-containing receptors are reduced generally in DLB is equivocal
sciousness and attention. Similarly, nicotinic receptors were higher (Gotti et al., 2006), but it is perhaps most likely to occur in DLB
in some thalamic nuclei in cases with variations in consciousness cases with hallucinations (Court et al., 2001b).
(Pimlott et al., 2004) (see section Nicotinic receptor changes), In the striatum, nAChRs are more reduced in DLB and PD than
possibly suggesting that combined cholinergic and dopaminergic is the case in AD, notably α6, α4, β2, and β3-containing subtypes
therapy is required to treat variations in consciousness in DLB. (Gotti et al., 2006). Reduced nAChR binding in this region, which
In temporal cortex (Brodmann area 21) in DLB and PDD, is at least in part on dopaminergic terminals, is as severe in DLB as
D2 receptors are significantly reduced by over 40%, and are also PD (Gotti et al., 2006), perhaps indicating that loss of these recep-
reduced in cases with concomitant Alzheimer pathology, but are tors occurs at a relatively early stage of nigrostriatal degeneration
not reduced in pure AD (Piggott et al., 2007b). The reduced tem- (Perry et al., 1995). In contrast, α4β2-containing nAChRs visual-
poral cortical D2 density correlated with cognitive decline, but not ized with [125I]-5-IA85380 were not generally reduced in the tha-
with hallucinations or delusions, suggesting that neuroleptics may lamus in DLB (although there were reductions in PD), significant
have a deleterious effect on cognition in dementia with Lewy bod- deficits only being observed in cases without variations in levels
ies. Dopamine D1 and D3 receptors in the striatum were not altered of consciousness (Pimlott et al., 2004). Yet again, relative preserva-
in DLB in a post-mortem study mainly involving cases with little or tion of receptors (D2 in thalamus, α4β2 in temporal cortex and tha-
no parkinsonism (Piggott et al., 1999). lamus) is associated with fluctuations in consciousness. Reduced
α7 receptor binding has also been noted in the reticular nucleus of is, however, an increase in the numbers of 5-HT1A receptors in tem-
the thalamus in DLB (in common with AD), a region innervated by poral cortex in DLB and PDD with depression (Sharp et al., 2008),
cholinergic neurons from the basal forebrain (Court et al., 1999). and in PD 5-HT1A receptors numbers were increased in frontal and
Loss of nAChRs in DLB is likely to reflect reduced cholinergic temporal cortex compared to controls (although any relationship
innervation (cortex and thalamus), dopaminergic innervation to depression was not assessed in this study) (Chen et al., 1998).
(striatum), and also attenuation of pre- and postsynaptic receptors, Treatment of depression in DLB and PDD may be efficacious with
including those on glutamatergic, GABA-ergic, and serotonergic a 5-HT1A antagonist.
neurons (cortex, thalamus, and basal ganglia). Greater preservation of serotonergic function is related to more
behavioural and psychological symptoms in DLB. Probably the bal-
Muscarinic receptor changes ance between transmitter systems is important, as in the reported
Although presynaptic losses are extensive in DLB, postsynaptically relative preservation of 5HT markers in DLB, where in an envi-
there is less neuronal damage than in AD (Tiraboschi et al., 2000; ronment of a much reduced cholinergic system there is increased
O’Brien et al., 2001). Muscarinic M1 receptor modulation differs likelihood of visual hallucinations (Perry et al., 1990b).
between DLB and AD. While unchanged or slightly reduced in the
cortex in severe AD and with defective coupling, in DLB M1 recep- Glutamate
tors have been reported to be up-regulated in temporal and parietal Excitatory amino acid transmission occurs between several compo-
cortex (Perry et al., 1990d; Ballard et al., 2000), and higher in fron- nents of the basal ganglia circuit, which are likely to be affected in
tal cortex in PDD and DLB compared to AD (Warren et al., 2008). PDD and DLB, and excitotoxic mechanisms have been implicated
Such up-regulation in the temporal cortex in DLB was associated in the progress of neurodegenerative diseases. In PD, increased
with delusions (Ballard et al., 2000). In vivo imaging of M1/M4 output from the subthalamic nucleus is part of the neurochemical
with 123I QNB showed raised receptor density in occipital cortex pathology of the disorder, and glutamate antagonists have been used
(Colloby et al., 2006). Additionally, coupling to G-protein second therapeutically. However, investigations of glutamate markers have
messenger systems is found to be preserved in DLB compared to been few. No change was shown in glutamate transporter protein in
AD, in temporal cortex (Perry et al., 1998), and in frontal cortex cortex in two DLB cases (Scott et al., 2002), no change in NMDA
(Warren et al., 2008). In contrast to the cortex, striatal M1 receptor receptor immunoreactivity in entorhinal cortex and hippocampus
binding is reduced, in parallel with D2 receptors, in DLB (M1 and (Thorns et al., 1997), and no reduction in glutamate in CSF (Molina
D2 are distributed together mainly on the same population of pro- et al., 2005). GluR2/3 AMPA receptor immunoreactivity was, how-
jection neurons from striatum to external globus pallidus) (Piggott ever, decreased in entorhinal cortex and hippocampus (Thorns et
et al., 2003). This is possibly why cholinesterase inhibitor therapy al., 1997), and group I metabotropic mGluRs were reduced in DLB
tends not to provoke worsening of parkinsonism in DLB patients. In with concomitant Alzheimer pathology (Albasanz et al., 2005) in
DLB and PDD, M4 receptors are raised in cingulate with impaired parallel with increased Alzheimer pathology. Conversely, a 40%
consciousness (Teaktong et al., 2005), and in the insula cortex with increase in the levels of mGluR5 immunoreactivity in the fron-
the symptom of delusions, while M2 receptors tend to be higher in tal cortex, hippocampus, and putamen in DLB has been reported
relation to severity and duration of EPS in putamen and insula cor- (Price et al., 2010). Further studies are needed to determine the
tex. M2 binding was higher in cingulate cortex compared to con- extent of glutamate receptor changes in DLB/PDD.
trols, and higher M2 and M4 cingulate binding was associated with
visual hallucinations (Teaktong et al., 2005). In contrast, in thala- GABA
mus M4 receptors are reduced in DLB while M2 are unchanged While anxiety and insomnia are common complaints in DLB, which
(Warren et al., 2007). may respond to treatment with benzodiazepines, and GABAergic
systems are likely affected in DLB, there are few published reports
Serotonin of GABAergic changes in DLB. Selective dendritic derangement
Lewy body pathology and neuron loss have been reported in the of GABAergic medium spiny neurons in the caudate have been
raphe nucleus in DLB (Benarroch et al., 2007). Reduced serotonin reported, suggested to be linked to disrupted executive function
has been reported in striatum and cortex in DLB (Langlais et al., (Zaja-Milatovic et al., 2006), and similar disruption in the motor
1993; Perry et al., 1993; Ohara et al., 1998), but not in another study putamen in late stage PD has been found (Zaja-Milatovic et al.,
of the putamen (Piggott and Marshall, 1996). Serotonin transporter 2005). However, there was no change in benzodiazepine binding
binding is reduced by about 70% in temporal and parietal cortex (GABAA receptor) in the striatum in DLB, with a slight increase
in DLB, and 5-HT22A receptors are reduced by 50% in putamen in in the globus pallidus (Suzuki et al., 2002). No difference in GABA
PD (Piggott, unpublished), with 5-HT22A receptors also reduced in CSF concentration between controls and DLB has been found
temporal cortex in DLB and PDD (Cheng et al., 1991). Depression is (Molina et al., 2005).
a frequent symptom in PD and DLB, but there does not seem to be a
link between serotonin loss and depression (Francis, 2009b). There Noradrenaline
is no evidence that selective serotonin reuptake inhibitors are effec- Degeneration of the locus ceruleus and reduction in noradrenaline
tive in depression in PD (Ghazi-Noori et al., 2003), and untreated has been reported in PD, and suggested to be linked to symptoms
PD patients with depression showed no differences in CSF serot- of mood disorder and subtle cognitive change (Scatton et al., 1983;
onin metabolites compared to patients without depression (Kuhn Cash et al., 1987), and this loss is more extensive in PDD than PD
et al., 1996). DLB patients with a history of major depression actu- (Cash et al., 1987; Zweig et al, 1993) and DLB (where it is equiva-
ally had relatively higher serotonin transporter binding in parietal lent to AD) (Jellinger, 2000; Leverenz et al., 2001; Szot et al., 2006).
cortex than cases without depression (Ballard et al., 2002a). There Locus ceruleus neuron loss correlates with cognitive decline, with
evidence of compensation for the neuron loss in the remaining loss was profound in the caudate and putamen, but deficits were
locus ceruleus neurons in AD and DLB (Szot et al., 2006). In both also seen in the substantia nigra, nucleus basalis, locus ceruleus,
AD and DLB there was reduced α1D and α2C adrenergic receptor raphe nucleus, lateral thalamus, and occipital cortex (Nagaoka et
mRNA in hippocampus (Szot et al., 2006). Alpha-2 adrenergic al., 1995). Dopaminergic changes are related to aggression in FTD
receptor density was increased slightly in DLB in frontal cortex (Engelborghs et al., 2007).
(Leverenz et al., 2001).
Noradrenaline system changes in limbic areas may contribute to Serotonin
depression and to behavioural symptoms such as aggression and Serotonergic deficits are the most consistent neurochemical finding
pacing, to cognitive decline in cortical areas, and to movement in FTD, revealed in both autopsy and imaging studies. In the young
disorder in basal ganglia; but in the main these relationships are case of FTD reported by Nagaoka (Nagaoka et al., 1995), there was
still to be investigated. In DLB, noradrenaline is much reduced in a 50% reduction in serotonin in striatum, lateral thalamus, nucleus
putamen (Langlais et al., 1993), which could mitigate symptoms of basalis of Meynert, and hippocampus. In 12 FTD cases there was a
parkinsonism. 40% reduction in neuron number in the raphe (Yang and Schmitt,
2001). Serotonin reuptake site density and serotonin concentra-
Frontotemporal dementia tion have been found to be unchanged or increased (Sparks and
Markesbery 1991; Francis et al., 1993; Bowen et al., 2008), but in
Frontotemporal lobe dementia (FTD) groups together Pick’s dis- three sporadic cases, using cyanoimipramine rather than paroxetine
ease, primary progressive aphasia, and semantic dementia. FTD to visualize 5HT-T, densities were reduced by about 40% in puta-
involves atrophy and neuron loss in the frontal and temporal lobes men and in cingulate and insula cortex (Piggott, unpublished).
to greater or lesser extents, without amyloid plaques but with tau Postsynaptic serotonin receptor binding is consistently reduced
pathology. There is also variable degeneration, often asymmetric, in in FTD; a PET study in eight patients found 5HT2A receptor den-
the substantia nigra, hippocampus, thalamus, striatum, and amy- sity reduced in orbitofrontal, frontal medial, and cingulate cor-
gdala (Barnes et al., 2006). Studies of neurotransmitter changes in tex (Franceschi et al., 2005), and similarly a PET study of 5HT1A
FTD are relatively few, and involve a small number of cases. receptors found widespread cortical reductions (Lanctot et al.,
2007a). Of post-mortem studies, serotonin binding was reduced
Acetylcholine by 50% in hypothalamus, temporal, and frontal cortex (Sparks and
Cholinergic systems have generally been reported as normal in Markesbery, 1991), while postsynaptic 5HT1A receptors were low
FTD (Di Lazzaro et al., 2006), with sparing of the nucleus basalis in frontal cortex (Francis et al., 1993; Procter et al., 1999) and also
of Meynert (Foster et al., 1998; reviewed in Huey et al., 2006). An in temporal cortex (Bowen et al., 2008). 5HT2A receptor density
imaging study showed no change in AChE activity in FTD (Hirano was reduced in cortex and by more than half in frontal and tempo-
et al., 2010). No changes in ChATor M1 receptors were reported ral cortex (Bowen et al., 2008). 5HT2 binding was reduced by 30%
by Bowen et al. (2008). Some studies have found deficits, however, in caudate, putamen, and cingulate cortex in three cases (Piggott,
for example in muscarinic receptors in temporal cortex in Pick’s unpublished).
disease (Hansen et al., 1988; Weinberger et al., 1991) and in atypical FTD patients often have symptoms of stereotyped compulsive
Pick’s disease (Odawara et al., 2003). AChE inhibitor treatment in behaviour, impulsivity, and food cravings, which could be caused
FTD is generally not helpful (Huey et al., 2006), although there is by serotonin deficiency (Miller et al., 1995). Although large stud-
a report of benefit for behavioural symptoms but not for cognition ies of effective treatments have yet to be published, the behavioural
(Lampl et al., 2004). symptoms of FTD can improve with serotonin reuptake inhibitors,
according to some (Moretti et al., 2003; Ikeda et al., 2004; Mendez,
Dopamine 2009) but not all reports (Deakin et al., 2004). 5HT1A receptor
Basal ganglia involvement with substantia nigra cell loss occurs in antagonism might be a useful therapeutic approach (Bowen et al.,
many patients and is the likely cause of parkinsonian symptoms. 2008).
A PET study in 12 FTD patients found dopamine transporter loss
correlated with severity of motor items from the Unified Parkinson’s Glutamate
Disease Rating Scale (UPDRS) (Rinne et al., 2002). In this study, AMPA receptors are reduced in both temporal and frontal lobes
the average reduction in putamen was 82% of control, while in cau- in FTD (30%), with no reduction of kainate receptors and modest
date it was down by 86%. A SPECT DaTSCAN study reported defi- reduction of NMDA receptors (13%) (Procter et al., 1999; Bowen
cits of more than 60% in striatal dopamine transporters in seven et al., 2008). This was suggested to be consistent with dying back of
FTD patients, correlated with motor UPDRS score (Sedaghat et al., pyramidal neurons.
2007). These are similar to the reduction found in a post-mortem
study, with severe loss of striatal dopamine concentration and trans- Noradrenaline
porter binding in three sporadic FTD cases of older onset (52, 67, The locus ceruleus is generally spared in FTD (Yang and Schmitt,
and 80 years) (Piggott, unpublished). This group also had lowered 2001), or may have decreased neuron number in Pick’s disease
dopamine content in hippocampus and frontal cortex. The predi- (Arima and Akashi, 1990). CSF norepinephrine levels were posi-
lection cortical sites for atrophy in FTD, temporal and frontal lobes, tively correlated with aggression (Engelborghs et al., 2007). In the
receive strong dopaminergic inputs, and mesocortical dopamine single young case of FTD, noradrenaline concentration was less
insufficiency will likely contribute to the nonmotor features of than 20% of normal in the lateral thalamus and the nucleus basalis
depression and dementia (Knopman et al., 1990). In one young of Meynert, and slightly reduced in the locus ceruleus, amygdala,
FTD case (28 years) with behavioural changes, striatal dopamine and medial thalamus, although pathology in the locus ceruleus
appeared slight (Nagaoka et al, 1995). Idazoxan, an α2 noradrener- Bencherif, M. and Lippiello, P.M. (2010). Alpha7 neuronal nicotinic receptors:
gic receptor antagonist, may improve attention, verbal fluency, and the missing link to understanding Alzheimer’s etiopathology? Medical
planning (Coull et al., 1996). Hypotheses, 74, 281–85.
Birks, J. (2006). Cholinesterase inhibitors for Alzheimer’s disease. Cochrane
Neuropeptides Database of Systematic Reviews, 1, CD005593.
Bohnen, N.I., et al. (2006). Cognitive correlates of cortical cholinergic
Neuropeptides may be reduced in FTD, with low CSF lev- denervation in Parkinson’s disease and parkinsonian dementia. Journal
els of somatostatin, neuropeptide Y, antidiuretic hormone, and of Neurology, 253, 242–247.
corticotrophin-releasing factor reported (Edvinsson et al., 1993; Boissiere, F., et al. (1997). Choline acetyltransferase mRNA expression in
Minthon et al., 1997), with concentrations of somatostatin and neu- the striatal neurons of patients with Alzheimer’s disease. Neuroscience
ropeptide Y related to agitiation, irritability, and restlessness (Minthon Letters, 225, 169–72.
et al., 1997). Conversely, neuropeptide concentrations in hypothala- Boksha, I.S. (2004). Coupling between neuronal and glial cells via glutamate
mus were not reduced and not linked to aberrant eating (Piguet et al., metabolism in brain of healthy persons and patients with mental
disorders. Biochemistry (Moscow), 69, 705–19.
2011). Oxytocin may be a novel treatment (Jesso et al., 2011).
Bonelli, S.B., et al. (2004). L-dopa responsiveness in dementia with Lewy
bodies, Parkinson disease with and without dementia. Neurology, 63,
References 376–8.
Albasanz, J.L., et al. (2005). Impaired metabotropic glutamate receptor/ Bowen, D.M., et al. (2008). Imbalance of a serotonergic system in
phospholipase C signaling pathway in the cerebral cortex in Alzheimer’s frontotemporal dementia: implication for pharmacotherapy.
disease and dementia with Lewy bodies correlates with stage of Psychopharmacology (Berlin), 196, 603–10.
Alzheimer’s-disease-related changes. Neurobiology of Disease, 20, 685–93. Brandel, J.P., et al. (1991). Differential vulnerability of cholinergic projections
Allen, J.M., et al. (1984). Elevation of neuropeptide Y (NPY) in substantia to the mediodorsal nucleus of the thalamus in senile dementia of
innominata in Alzheimer’s type dementia. Journal of Neurological Alzheimer type and progressive supranuclear palsy. Neuroscience, 41,
Science, 640, 325–31. 25–31.
Amada, N., et al. (2005). Reduction of NR1 and phosphorylated Ca2+/ Buhot, M.C., Martin, S., and Segu, L. (2000). Role of serotonin in memory
calmodulin-dependent protein kinase II levels in Alzheimer’s disease. impairment. Annals of Medicine, 32, 210–21.
Neuroreport, 16, 1809–13. Burbaeva, G., et al. (2005). Glutamate metabolizing enzymes in prefrontal
Arai, H., et al. (1992). Neurotransmitter changes in early- and late-onset cortex of Alzheimer’s disease patients. Neurochemistry Research, 30,
Alzheimer-type dementia. Progress in Neuropsychopharmacology and 1443–51.
Biological Psychiatry, 16, 883–90. Caccamo, A., Oddo, S., and Billings, L.M. (2006). M1 receptors play a central
Arai, H., Kosaka, K., and Iizuka, R. (1984). Changes of biogenic amines role in modulating AD-like pathology in transgenic mice. Neuron, 49,
and their metabolites in postmortem brains from patients with 671–82.
Alzheimer-type dementia. Journal of Neurochemistry, 43, 388–93. Carson, R., et al. (2008a). Genetic variation in the alpha 7 nicotinic
Araya, R., et al. (2006). Loss of M5 muscarinic acetylcholine receptors leads acetylcholine receptor is associated with delusional symptoms in
to cerebrovascular and neuronal abnormalities and cognitive deficits in Alzheimer’s disease. Neuromolecular Medicine, 10, 377–84.
mice. Neurobiology of Disease, 24, 334–44. Carson, R., et al. (2008b). Alpha7 nicotinic acetylcholine receptor gene and
Arima, K. and Akashi, T. (1990). Involvement of the locus coeruleus in Pick’s reduced risk of Alzheimer’s disease Journal of Medicine and Genetics, 45,
disease with or without Pick body formation. Acta Neuropathologica, 79, 244–8.
629–33. Carter, T.L., et al. (2004). Differential preservation of AMPA receptor subunits
Armstrong, D.M. and Ikonomovic, M.D. (1996). AMPA-selective glutamate in the hippocampi of Alzheimer’s disease patients according to Braak
receptor subtype immunoreactivity in the hippocampal dentate gyrus of stage. Experimental Neurology, 187, 299–309.
patients with Alzheimer disease: Evidence for hippocampal plasticity. Cash, R., et al. (1987). Parkinson’s disease and dementia: norepinephrine and
Molecular and Chemical Neuropathology, 28, 59–64. dopamine in locus ceruleus. Neurology, 37, 42–6.
Armstrong, D.M., et al. (2003). Plasticity of glutamate and GABAA receptors Chalmers, D.T., et al. (1990). Differential alterations of cortical glutamatergic
in the hippocampus of patients with Alzheimer’s disease. Cellular binding sites in senile dementia of the Alzheimer type. Proceedings of the
Molecular Neurobiology, 23, 491–505. National Academy of Sciences USA, 87, 1352–6.
Ballard, C., et al. (2000). Delusions associated with elevated muscarinic M1 Chan-Palay, V. (1987). Somatostatin immunoreactive neurons in the human
receptor binding in dementia with Lewy bodies. Annals of Neurology, hippocampus and cortex shown by immunogold/silver intensification
48, 868–76. on vibratome sections: coexistence with neuropeptide Y neurons, and
Ballard, C., et al. (2002a). A positive association between 5HT re-uptake effects in Alzheimer-type dementia. Journal of Comparative Neurology,
binding sites and depression in dementia with Lewy bodies. Journal of 260, 201–23.
Affective Disorders, 69, 219–23. Chan-Palay, V. (1988). Galanin hyperinnervates surviving neurons of the
Ballard, C.G., et al. (2002b). Disturbances of consciousness in dementia with human basal nucleus of Meynert in dementias of Alzheimer’s and
Lewy bodies associated with alteration in nicotinic receptor binding in Parkinson’s disease: a hypothesis for the role of galanin in accentuating
the temporal cortex. Consciousness and Cognition, 11, 461–74. cholinergic dysfunction in dementia. Journal of Comparative Neurology,
Banerjee, S., et al. (2011). Sertraline or mirtazapine for depression in 273, 543–57.
dementia (HTA-SADD): a randomised, multicentre, double-blind, Chen, C.P., et al. (1998). Post-synaptic 5-HT1A and 5-HT2A receptors are
placebo-controlled trial. Lancet, 378, 403–11. increased in Parkinson’s disease neocortex. Annals of the New York
Barabash, A., et al. (2009). APOE, ACT and CHRNA7 genes in the conversion Academy of Science, 861, 288–9.
from amnestic mild cognitive impairment to Alzheimer’s disease. Chen, K.H., et al. (2011). Disturbed neurotransmitter transporter expression
Neurobiology of Aging, 30, 1254–64. in Alzheimer’s disease brain. Journal of Alzheimer’s Disease, 26, 755–66.
Barnes, J., et al. (2006). Measurements of the amygdala and hippocampus Cheng, A.V., et al. (1991). Cortical serotonin-S2 receptor binding in Lewy
in pathologically confirmed Alzheimer disease and frontotemporal lobar body dementia, Alzheimer’s and Parkinson’s diseases. Journal of the
degeneration. Archives of Neurology, 63, 1434–39. Neurological Sciences, 106, 50–5.
Benarroch, E.E., et al. (2007). Rostral raphe involvement in Lewy body dementia Colloby, S.J., et al. (2005). Progression of dopaminergic degeneration in
and multiple system atrophy. Acta Neuropathologica, 114, 213–20. dementia with Lewy bodies and Parkinson’s disease with and without
dementia assessed using 123I-FP-CIT SPECT. European Journal of Dewar, D., et al. (1991). Glutamate metabotropic and AMPA binding sites
Nuclear Medicine and Molecular Imaging, 32, 1176–85. are reduced in Alzheimer’s disease: an autoradiographic study of the
Colloby, S.J., et al. (2006). In vivo SPECT imaging of muscarinic acetylcholine hippocampus. Brain Research, 553, 58–64.
receptors using (R,R) 123I-QNB in dementia with Lewy bodies and Di Lazzaro, V., et al. (2006). In vivo cholinergic circuit evaluation in
Parkinson’s disease dementia. Neuroimage, 33, 423–9. frontotemporal and Alzheimer dementias. Neurology, 66, 1111–3.
Colloby, S.J., et al. (2010). Nicotinic 123I-5IA-85380 single photon emission Dournaud, P., et al. (1995). Differential correlation between neurochemical
computed tomography as a predictor of cognitive progression in deficits, neuropathology, and cognitive status in Alzheimer’s disease.
Alzheimer’s disease and dementia with Lewy bodies. American Journal Neurobiology of Aging, 16, 817–23.
of Geriatric Psychiatry, 18, 86–90. Duda, J.E., et al. (2002). Novel antibodies to synuclein show abundant striatal
Colloby, S.J., et al. (2011). Alterations in nicotinic alpha4beta2 receptor pathology in Lewy body diseases. Annals of Neurology, 52, 205–10.
binding in vascular dementia using 123I-5IA-85380 SPECT: comparison Edvinsson, L., et al. (1993). Neuropeptides in cerebrospinal fluid of patients
with regional cerebral blood flow. Neurobiology of Aging, 32, 293–301. with Alzheimer’s disease and dementia with frontotemporal lobe
Coull, J.T., Sahakian, B.J., and Hodges, J.R. (1996). The alpha(2) antagonist degeneration. Dementia, 4, 167–71.
idazoxan remediates certain attentional and executive dysfunction in Elliott, M.S., et al. (2009). Increased binding to 5-HT1A and 5-HT2A receptors
patients with dementia of frontal type. Psychopharmacology (Berlin), is associated with large vessel infarction and relative preservation of
123, 239–49. cognition. Brain, 132, 1858–65.
Court, J.A. and Perry, E.K. (2003). Neurotransmitter abnormalities in vascular Engelborghs, S., et al. (2007). The dopaminergic neurotransmitter system is
dementia. International Psychogeriatrics, 15, 81–7. associated with aggression and agitation in frontotemporal dementia.
Court, J., Perry, E., and Kalaria, R. (2002). Neurotransmitter control of Neurochemistry International, 52, 1052–60.
the cerebral vasculature and abnormalities in vascular dementia. In: Erkinjuntti, T., et al. (2004). Emerging therapies for vascular dementia and
Erkinjuntti, T. and Gauthier, S. (eds) Vascular cognitive impairment. vascular cognitive impairment. Stroke, 35, 1010–7.
Martin Dunitz, London, pp. 167–85.
Evans, J.G., Wilcock, G., and Birks, J. (2004). Evidence-based pharmacotherapy
Court, J., Perry, E., and Kalaria, R. (2003). Neurochemical changes in vascular of Alzheimer’s disease. International Journal of Neuropsychopharmacology,
dementia. In: Ames, D., et al. (eds) Cerebrovascular disease, cognitive 7, 351–69.
impairment and dementia. Martin Dunitz, London, pp. 133–51.
Feldman, H., et al. (2001). A 24-week, randomized, double-blind study of
Court, J., et al. (1999). Neuronal nicotinic receptors in dementia with Lewy
donepezil in moderate to severe Alzheimer’s disease. Neurology, 57, 613–20.
bodies and schizophrenia: α-bungarotoxin and nicotine binding in the
Fisher, A. (2012). Cholinergic modulation of amyloid precursor protein
thalamus. Journal of Neurochemistry, 73, 1590–7.
processing with emphasis on M1 muscarinic receptor: perspectives and
Court, J.A., et al. (2000). Nicotine binding in human striatum: elevation in
challenges in treatment of Alzheimer’s disease. Journal of Neurochemistry,
schizophrenia and reductions in dementia with Lewy bodies, Parkinson’s
120, 22–33.
disease and Alzheimer’s disease and in relation to neuroleptic medication.
Flynn, D.D., et al. (1995a). Differential alterations in muscarinic receptor
Neuroscience, 98, 79–87.
subtypes in Alzheimer’s disease: implications for cholinergic-based
Court, J., et al. (2001a). Nicotinic receptor abnormalities in Alzheimer’s
therapies. Life Sciences, 56, 869–76.
disease. Biological Psychiatry, 49, 175–84.
Flynn, D.D., et al. (1995b). Differential regulation of molecular subtypes of
Court, J.A., et al. (2001b). Visual hallucinations are associated with lower alpha
muscarinic receptors in Alzheimer’s disease. Journal of Neurochemistry,
bungarotoxin binding in dementia with Lewy bodies. Pharmacology,
64, 1888–91.
Biochemistry and Behavior, 70, 571–9.
Court, J.A., et al. (2005). Attenuation of Abeta deposition in the entorhinal Foster, N.L., et al. (1998). Clinical and neuropathological correlations with
cortex of normal elderly individuals associated with tobacco smoking. PET in FTDP-17 and sporadic frontal dementia. Neurobiology of Aging,
Neuropathology and Applied Neurobiology, 31, 522–35. 19, S296.
Cowburn, R.F., et al. (1989). Characterisation, density, and distribution of Franceschi, M., et al. (2005). Glucose metabolism and serotonin receptors in
kainate receptors in normal and Alzheimer’s diseased human brain. the frontotemporal lobe degeneration. Annals of Neurology, 57, 216–25.
Journal of Neurochemistry, 52, 140–7. Francis, P.T. (2003). Glutamatergic systems in Alzheimer’s disease.
Cowburn, R.F., et al. (1996). Acetylcholine muscarinic M2 receptor International Journal of Geriatric Psychiatry, 18, S15–21.
stimulated [35S]GTP gamma S binding shows regional selective changes Francis, P.T. (2009a). Altered glutamate neurotransmission and behaviour
in Alzheimer’s disease postmortem brain. Neurodegeneration, 5, 19–26. in dementia: evidence from studies of memantine. Current Molecular
D’Aniello, A., et al. (2005). Amino acids and transaminases activity in Pharmacology, 2, 77–82.
ventricular CSF and in brain of normal and Alzheimer patients. Francis, P.T. (2009b). Biochemical and pathological correlates of cognitive
Neuroscience Letters, 388, 49–53. and behavioural change in DLB/PDD. Journal of Neurology, 256, 280–5.
Danielsson, E., et al. (1988). VIP-sensitive adenylate cyclase, guanylate cyclase, Francis, P.T., Ramirez, M.J., and Lai, M.K. (2010). Neurochemical basis for
muscarinic receptors, choline acetyltransferase and acetylcholinesterase, symptomatic treatment of Alzheimer’s disease. Neuropharmacology, 59,
in brain tissue afflicted by Alzheimer’s disease/senile dementia of the 221–9.
Alzheimer type. Neurobiology of Aging, 9, 153–62. Francis, P.T., et al . (1993). Preliminary neurochemical findings in
Davis, K.L., et al. (1999). Cholinergic markers in elderly patients with early non-Alzheimer dementia due to lobar atrophy. Dementia, 4, 172–7.
signs of Alzheimer disease, Journal of the American Medical Association, Francis, P.T., et al. (1999). The cholinergic hypothesis of Alzheimer’s disease:
281, 1401–6. a review of progress. Journal of Neurology, Neurosurgery and Psychiatry,
Dawbarn, D., et al. (1986). Decreased somatostatin immunoreactivity but not 66, 137–47.
neuropeptide Y immunoreactivity in cerebral cortex in senile dementia Freeman, A., et al. (2001). Nigrostriatal collaterals to thalamus degenerate in
of Alzheimer type. Neuroscience Letters, 70, 154–9. parkinsonian animal models. Annals of Neurology, 50, 321–9.
Deakin, J.B., et al. (2004). Paroxetine does not improve symptoms and impairs Garcia-Alloza, M., et al. (2004). Differential involvement of 5-HT(1B/1D)
cognition in frontotemporal dementia: a double-blind randomized and 5-HT6 receptors in cognitive and non-cognitive symptoms in
controlled trial. Psychopharmacology (Berlin), 172, 400–8. Alzheimer’s disease. Neuropsychopharmacology, 29, 410–6.
DeKosky, S.T., et al. (2002). Upregulation of choline acetyltransferase activity Garcia-Alloza, M., et al. (2006). Involvement of the GABAergic system in
in hippocampus and frontal cortex of elderly subjects with mild cognitive depressive symptoms of Alzheimer’s disease, Neurobiology of Aging, 27,
impairment. Annals of Neurology, 51, 145–55. 1110–7.
Ghazi-Noori, S., et al. (2003). Therapies for depression in Parkinson’s disease. Hynd, M.R., Scott, H.L., and Dodd, P.R. (2004b). Glutamate-mediated
Cochrane Database of Systematic Reviews, 3, CD003465. excitotoxicity and neurodegeneration in Alzheimer’s disease.
Gottfries, C.G., et al. (1994). The neurochemistry of vascular dementia. Neurochemistry International, 45, 583–95.
Dementia, 5, 163–7. Hynd, M.R., Scott, H.L., and Dodd, P.R. (2004c). Selective loss of NMDA
Gotti, C., et al. (2006). Selective nicotinic acetylcholine receptor subunit receptor NR1 subunit isoforms in Alzheimer’s disease. Journal of
deficits identified in Alzheimer’s disease, Parkinson’s disease and Neurochemistry, 89, 240–7.
dementia with Lewy bodies by immunoprecipitation. Neurobiology of Ikeda, M., et al. (2004). Efficacy of fluvoxamine as a treatment for behavioral
Disease, 23, 481–489. symptoms in frontotemporal lobar degeneration patients. Dementia and
Gravius, A., et al. (2010). Metabotropic glutamate receptors as therapeutic Geriatric Cognitive Disorders, 17, 117–21.
targets for cognitive disorders. Current Topics in Medicinal Chemistry, Ikonomovic, M.D., et al. (2009). Cortical alpha7 nicotinic acetylcholine
10, 187–206. receptor and beta-amyloid levels in early Alzheimer disease. Archives of
Greenamyre, J.T., et al. (1988). Glutamate transmission and toxicity in Neurology, 66, 646–51.
Alzheimer’s disease. Progress in Neuropsychopharmacology and Biological Iwakiri, M., et al. (2005). Changes in hippocampal GABABR1 subunit
Psychiatry, 12, 421–30. expression in Alzheimer’s patients: association with Braak staging. Acta
Grothe, M., Heinsen, H., and Teipel, S.J. (2012). Atrophy of the cholinergic Neuropathologica (Berlin), 109, 467–74.
basal forebrain over the adult age range and in early stages of Alzheimer’s Iwakiri, M., et al. (2009). An immunohistochemical study of GABAA receptor
disease. Biological Psychiatry, 71, 805–13. gamma subunits in Alzheimer’s disease hippocampus: relationship to
Gulyas, B., et al. (2009). A comparative autoradiography study in post neurofibrillary tangle progression. Neuropathology, 29, 263–269.
mortem whole hemisphere human brain slices taken from Alzheimer Jellinger, K.A. (2000). Morphological substrates of mental dysfunction
patients and age-matched controls using two radiolabelled DAA1106 in Lewy body disease: an update. Journal of Neural Transmission
analogues with high affinity to the peripheral benzodiazepine receptor Supplementum, 59, 185–212.
(PBR) system. Neurochemistry International, 54, 28–36. Jellinger, K.A. and Attems, J. (2006). Does striatal pathology distinguish
Gulyas, B., et al. (2010). The norepinephrine transporter (NET) radioligand Parkinson disease with dementia and dementia with Lewy bodies? Acta
(S,S)-[18F]FMeNER-D2 shows significant decreases in NET density in Neuropathologica, 112, 253–60.
the human brain in Alzheimer’s disease: a post-mortem autoradiographic Jesso, S., et al. (2011). The effects of oxytocin on social cognition and
study. Neurochemistry International, 56, 789–98. behaviour in frontotemporal dementia. Brain, 134, 2493–501.
Halliday, G.M., et al. (1992). Brain stem serotonin-synthesizing neurons Jia, J.P., et al. (2004). Differential acetylcholine and choline concentrations in
in Alzheimer’s disease: a clinicopathological correlation. Acta the cerebrospinal fluid of patients with Alzheimer’s disease and vascular
Neuropathologica (Berlin), 84, 638–50. dementia. Chinical Medical Journal (England), 117, 1161–4.
Hansen, L.A., et al. (1988). Neocortical morphometry and cholinergic Jimenez-Jimenez, F.J., et al. (1998). Neurotransmitter amino acids in
neurochemistry in Pick’s disease. American Journal of Pathology, 131, cerebrospinal fluid of patients with Alzheimer’s disease. Journal of Neural
507–18. Transmission, 105, 269–77.
Hasselbalch, S.G., et al. (2008). Reduced 5-HT2A receptor binding in patients Kalaitzakis, M.E., et al. (2011). Striatal Abeta peptide deposition mirrors
with mild cognitive impairment. Neurobiology of Aging, 29, 1830–8. dementia and differentiates DLB and PDD from other parkinsonian
Heilig, M., et al. (1995). Cerebrospinal fluid neuropeptides in Alzheimer’s syndromes. Neurobiology of Disease, 41, 377–84.
disease and vascular dementia. Biology and Psychiatry, 38, 210–6. Kemppainen, N., et al. (2003). Hippocampal dopamine D2 receptors correlate
Hellstrom-Lindahl, E., et al. (2004). Reduced levels of Abeta 40 and Abeta 42 with memory functions in Alzheimer’s disease. European Journal of
in brains of smoking controls and Alzheimer’s patients. Neurobiological Neuroscience, 18, 149–154.
Disease, 15, 351–60. Kendziorra, K., et al. (2011). Decreased cerebral alpha4beta2* nicotinic
Herrmann, N., Lanctot, K.L., and Khan, L.R. (2004). The role of norepinephrine acetylcholine receptor availability in patients with mild cognitive
in the behavioral and psychological symptoms of dementia. Journal of impairment and Alzheimer’s disease assessed with positron emission
Neuropsychiatry and Clinical Neuroscience, 16, 261–76. tomography. European Journal of Nuclear Medicine and Molecular
Herrmann, N., Li, A., and Lanctot, K. (2011). Memantine in dementia: a Imaging, 38, 515–25.
review of the current evidence. Expert Opinions in Pharmacotherapy, 12, Keverne, J.S., et al. (2007). Cholinergic neuronal deficits in CADASIL. Stroke,
787–800. 38, 188–91.
Hirano, S., et al. (2010). Cholinergic imaging in corticobasal syndrome, Khachaturian, A.S., et al. (2006). Antihypertensive medication use and
progressive supranuclear palsy and frontotemporal dementia. Brain, incident Alzheimer disease: the Cache County Study. Archives of
133, 2058–2068. Neurology, 63, 686–92.
Hock, C., et al. (2003). Treatment with the selective muscarinic m1 agonist Knopman, D.S., et al. (1990). Dementia lacking distinctive histologic
talsaclidine decreases cerebrospinal fluid levels of A beta 42 in patients features: a common non-Alzheimer degenerative dementia. Neurology,
with Alzheimer’s disease. Amyloid, 10, 1–6. 40, 251–6.
Holmes, C., et al. (1998). 5-HT2A and 5-HT2C receptor polymorphisms and Koide, S., et al. (1995). Plasma neuropeptide Y is reduced in patients with
psychopathology in late onset Alzheimer’s disease. Human Molecular Alzheimer’s disease. Neuroscience Letters, 198, 149–51.
Genetics, 7, 1507–1509. Kotagal, V., et al. (2012). Thalamic cholinergic innervation is spared in
Howard, R., et al. (2012). Donepezil and memantine for moderate-to-severe Alzheimer disease compared to parkinsonian disorders. Neuroscience
Alzheimer’s disease. New England Journal of Medicine, 366, 893–903. Letters, 514, 169–72.
Hu, N.W., Ondrejcak, T., and Rowan, M.J. (2012). Glutamate receptors in Kuhl, D.E., et al. (1996). In vivo mapping of cholinergic terminals in normal
preclinical research on Alzheimer’s disease: update on recent advances. aging, Alzheimer’s disease, and Parkinson’s disease. Annals of Neurology,
Pharmacology and Biochemical Behaviour, 100, 855–62. 40, 399–410.
Huey, E.D., Putnam, K.T., and Grafman, J. (2006). A systematic review of Kuhn, W., et al. (1996). Depression in Parkinson’s disease: biogenic amines in
neurotransmitter deficits and treatments in frontotemporal dementia. CSF of ‘de novo’ patients. Journal of Neural Transmission, 103, 1441–5.
Neurology, 66, 17–22. Lacor, P., et al. (2011). Synapse disruption in ad brain: evidence for early
Hynd, M.R., Scott, H.L., and Dodd, P.R. (2004a). Differential expression of pathological clustering of the group I metabotropic glutamate receptor
N-methyl-D-aspartate receptor NR2 isoforms in Alzheimer’s disease. 5. Alzheimer’s and Dementia. Conference: Alzheimer’s Association
Journal of Neurochemistry, 90, 913–19. International Conference, 7(4 Suppl 1).
Ladner, C.J. and Lee, J.M. (1999). Reduced high-affinity agonist binding at Martin-Ruiz, C., et al. (2000). Nicotinic receptors in dementia of Alzheimer,
the M(1) muscarinic receptor in Alzheimer’s disease brain: differential Lewy body and vascular types. Acta Neurologica Scandinavica Supplement,
sensitivity to agonists and divalent cations. Experimental Neurology, 158, 176, 34–41.
451–8. Mattila, P.M., et al. (2001). Choline acetytransferase activity and striatal
Lai, M.K., et al. (2001). Psychosis of Alzheimer’s disease is associated with dopamine receptors in Parkinson’s disease in relation to cognitive
elevated muscarinic M2 binding in the cortex. Neurology, 57, 805–11. impairment. Acta Neuropathologica, 102, 160–6.
Lai, M.K., et al. (2003a). [3H]GR113808 binding to serotonin 5-HT(4) Mazère, J., et al. (2008). In vivo SPECT imaging of vesicular acetylcholine
receptors in the postmortem neocortex of Alzheimer disease: a transporter using [123I]-IBVM in early Alzheimer’s disease. NeuroImage,
clinicopathological study. Journal of Neural Transmission, 110 , 40, 280–288.
779–88. Mehta, M., Adem, A., and Sabbagh, M. (2012). New acetylcholinesterase
Lai, M.K., et al. (2003b). Reduced serotonin 5-HT1A receptor binding in inhibitors for Alzheimer’s disease. InternationalJournal of Alzheimer’s
the temporal cortex correlates with aggressive behavior in Alzheimer disease, 2012, 728983.
disease. Brain Research, 974, 82–7. Mendez, M.F. (2009). Frontotemporal dementia: therapeutic interventions.
Lai, M.K., et al. (2005). Loss of serotonin 5-HT2A receptors in the postmortem Dementia in Clinical Practice. Frontiers of Neurology and Neuroscience,
temporal cortex correlates with rate of cognitive decline in Alzheimer’s 24, 168–178.
disease. Psychopharmacology (Berlin), 179, 673–7. Miller, B.L., et al. (1995). Dietary changes, compulsions and sexual behavior
Lai, M.K., et al. (2006). Selective effects of the APOE epsilon4 allele on in frontotemporal degeneration. Dementia, 6, 195–9.
presynaptic cholinergic markers in the neocortex of Alzheimer’s disease. Minthon, L., Edvinsson, L., and Gustafson, L. (1997). Somatostatin and
Neurobiological Disease, 22, 555–61. neuropeptide Y in cerebrospinal fluid: correlations with severity of
Lai, M.K., et al. (2011a). Differential involvement of hippocampal serotonin1A disease and clinical signs in Alzheimer’s disease and frontotemporal
receptors and re-uptake sites in non-cognitive behaviors of Alzheimer’s dementia. Dementia and Geriatric Cognitive Disorders, 8, 232–9.
disease. Psychopharmacology (Berlin), 213, 431–9. Minuzzi, L., et al. (2009). Declines of hippocampal metabotropic glutamate
Lai, M.K.P., et al. (2011b). Differential involvement of hippocampal receptor in Alzheimer’s disease. Alzheimer’s and Dementia. Conference:
serotonin1A receptors and re-uptake sites in non-cognitive behaviors of Alzheimer’s Association International Conference on Alzheimer’s Disease,
Alzheimer’s disease. Psychopharmacology, 213, 431–439. 5(4 Suppl 1).
Lampl, Y., Sadeh, M., and Lorberboym, M. (2004). Efficacy of Mitsis, E.M., et al. (2009). 123I-5-IA-85380 SPECT imaging of nicotinic
acetylcholinesterase inhibitors in frontotemporal dementia. Annals of receptors in Alzheimer disease and mild cognitive impairment. Journal
Pharmacotherapy, 38, 1967–8. of Nuclear Medicine, 50, 1455–63.
Lanctot, K.L., et al. (2007a). Serotonin-1A receptors in frontotemporal Mizukami, K., et al. (2011). An immunohistochemical study of the serotonin
dementia compared with controls. Psychiatry Research, 156, 247–50. 1A receptor in the hippocampus of subjects with Alzheimer’s disease.
Lanctot, K.L., et al. (2007b). Behavioral correlates of GABAergic disruption Neuropathology, 31, 503–509.
in Alzheimer’s disease. International Psychogeriatrics, 19, 151–158. Molina, J.A., et al. (2005). Neurotransmitter amino acid in cerebrospinal
Langlais, P.J., et al. (1993). Neurotransmitters in basal ganglia and cortex fluid of patients with dementia with Lewy bodies. Journal of Neural
of Alzheimer’s disease with and without Lewy bodies. Neurology, 43, Transmission, 112, 557–63.
1927–34. Moretti, R., et al. (2003). Frontotemporal dementia: paroxetine as a possible
Leverenz, J.B., et al. (2001). Increased alpha 2-adrenergic receptor binding treatment of behavior symptoms. A randomized, controlled, open
in locus coeruleus projection areas in dementia with Lewy bodies. 14-month study. European Neurology, 49, 13–9.
Neurobiology and Aging, 22, 555–61. Morrison, L.D. and Kish, S.J. (1995). Brain polyamine levels are altered in
Lezoualc’h, F. and Robert, S.J. (2003). The serotonin 5-HT4 receptor and the Alzheimer’s disease. Neuroscience Letters, 197, 5–8.
amyloid precursor protein processing. Experimental Gerontology, 38, Morrison, L.D., Cao, X.C., and Kish, S.J. (1998). Ornithine decarboxylase in
159–166. human brain: influence of aging, regional distribution, and Alzheimer’s
Lipton, S.A. (2005). The molecular basis of memantine action in Alzheimer’s disease. Journal of Neurochemistry, 71, 288–94.
disease and other neurologic disorders: low-affinity, uncompetitive Muir, K.W. (2006). Glutamate-based therapeutic approaches: clinical trials
antagonism. Current Alzheimer’s Research, 2, 155–65. with NMDA antagonists. Current Opinions in Pharmacology, 6, 53–60.
Lowe, S.L., et al. (1988). Gamma-aminobutyric acid concentration in brain Mulugeta, E., et al. (2003). Loss of muscarinic M4 receptors in hippocampus
tissue at two stages of Alzheimer’s disease. Brain, 111, 785–99. of Alzheimer patients. Brain Research, 960, 259–62.
Luchetti, S., Huitinga, I. and Swaab, D.F. (2011). Neurosteroid and GABA-A Nacmias, B., et al. (2001). Psychosis, serotonin receptor polymorphism and
receptor alterations in Alzheimer’s disease, Parkinson’s disease and Alzheimer’s disease. Archives of Gerontology & Geriatrics, 33, 279–83.
multiple sclerosis. Neuroscience, 191, 6–21. Nagaoka, S., et al. (1995). A juvenile case of frontotemporal dementia:
Mann, D.M., Yates, P.O., and Marcyniuk, B. (1986). The nucleus basalis of neurochemical and neuropathological investigations. Progress in
Meynert in multi-infarct (vascular) dementia. Acta Neuropatholica Neuro-Psychopharmacology & Biological Psychiatry, 19, 1251–61.
(Berlin), 71, 332–7. Nardone, R., et al. (2008). Cholinergic dysfunction in subcortical ischaemic
Mann, D.M., et al. (1980). Changes in the monoamine containing neurones vascular dementia: A transcranial magnetic stimulation study. Journal of
of the human CNS in senile dementia. British Journal of Psychiatry, 136, Neural Transmission, 115, 737–43.
533–41. Niswender, C.M. and Conn, P.J. (2010). Metabotropic glutamate receptors:
Marner, L., et al. (2012). Loss of serotonin 2A receptors exceeds loss of physiology, pharmacology, and disease. Annual Review of Pharmacology
serotonergic projections in early Alzheimer’s disease: a combined and Toxicology, 50, 295–322.
[11C]DASB and [18F]altanserin-PET study. Neurobiology of Aging, 33, Nordberg, A. (2006). Mechanisms behind the neuroprotective actions of
479–87. cholinesterase inhibitors in Alzheimer disease. Alzheimer’s Disease and
Marshall, E.F., et al. (1994). Dopamine metabolism in post-mortem Associated Disorders, 20, S12–18.
caudate nucleus in neurodegenerative disorders. Neuroscience Research Nordberg, A., Alafuzoff, I., and Winblad, B. (1992). Nicotinic and muscarinic
Communication, 14, 17–25. subtypes in the human brain: changes with aging and dementia. Journal
Martel, J.C., et al. (1990). Neuropeptide Y receptor binding sites in human of Neuroscience Research, 31, 103–11.
brain. Possible alteration in Alzheimer’s disease. Brain Research, 519, O’Brien, J.T., et al. (2001). Progressive brain atrophy on serial MRI in dementia
228–35. with Lewy bodies, AD, and vascular dementia. Neurology, 56, 1386–8.
O’Brien, J.T., et al. (2004). Dopamine transporter loss visualized with FP-CIT Perry, E., et al. (2000). Nicotinic receptor subtypes in human brain ageing,
SPECT in the differential diagnosis of dementia with Lewy bodies. Alzheimer and Lewy body diseases. European Journal of Pharmacology,
Archives of Neurology, 61, 919–25. 393, 215–22.
O’Brien, J.T., et al. (2007). Alpha4beta2 nicotinic receptor status in Alzheimer’s Perry, E.K., et al. (2003). Increased Alzheimer pathology in Parkinson’s
disease using 123I-5IA-85380 single-photon-emission computed disease related to antimuscarinic drugs. Annals of Neurology, 54,
tomography. Journal of Neurology, Neurosurgery and Psychiatry, 78, 235–8.
356–62. Perry, E., et al. (2005). Absence of cholinergic deficits in ‘pure’ vascular
O’Brien, J.T., et al. (2008). Nicotinic alpha4beta2 receptor binding in dementia. Neurology, 64, 132–3.
dementia with Lewy bodies using 123I-5IA-85380 SPECT demonstrates Piggott, M.A. and Marshall, E.F. (1996). Neurochemical correlates of
a link between occipital changes and visual hallucinations. Neuroimage, pathological and iatrogenic extrapyramidal symptoms. In: Perry, R.H.,
40, 1056–63. McKeith, I.G., and Perry, E.K. (eds) Dementia with Lewy bodies: clinical,
Odawara, T., et al. (2003). Alterations of muscarinic acetylcholine receptors pathological, and treatment issues. Cambridge University, Cambridge, pp.
in atypical Pick’s disease without Pick bodies. Journal of Neurology, 449–67.
Neurosurgery and Psychiatry, 74, 965–7. Piggott, M. and Perry, E. (2010). Neurochemical pathology of Parkinson’s
Ohara, K., Kondo, N., and Ohara, K. (1998). Changes of monoamines in disease dementia. In: Emre, M. (ed.) Cognitive impairment and dementia
post-mortem brains from patients with diffuse lewy body disease. Progress in Parkinson’s disease. Oxford University, Oxford, pp. 153–70.
in Neuro-Psychopharmacology and Biology Psychiatry, 22, 311–17. Piggott, M.A., et al. (1994). Modulation of [3H]MK-801 binding by
Overk, C.R., et al. (2010). Cortical M1 receptor concentration increases polyamines in human development and aging. Neuroscientific Research
without a concomitant change in function in Alzheimer’s disease. Journal Communications, 15, 49–58.
of Chemistry and Neuroanatomy, 40, 63–70. Piggott, M.A., et al. (1998). Nigrostriatal dopaminergic activities in dementia
Pappata, S., et al. (2010). SPECT imaging of GABA(A)/benzodiazepine with Lewy bodies in relation to neuroleptic sensitivity: comparisons with
receptors and cerebral perfusion in mild cognitive impairment. European Parkinson’s disease. Biology and Psychiatry, 44, 765–74.
Journal of Nuclear Medicine & Molecular Imaging, 37, 1156–63. Piggott, M.A., et al. (1999). Striatal dopaminergic markers in dementia
Parri, R.H. and Dineley, T.K. (2010). Nicotinic acetylcholine receptor with Lewy bodies, Alzheimer’s and Parkinson’s diseases: rostrocaudal
interaction with beta-amyloid: molecular, cellular, and physiological distribution. Brain, 122, 1449–68.
consequences. Current Alzheimer’s Research, 7, 27–39. Piggott, M.A., et al. (2003). Muscarinic receptors in basal ganglia in dementia
Parri, H.R., Hernandez, C.M., and Dineley, K.T. (2011). Research update: with Lewy bodies, Parkinson’s disease and Alzheimer’s disease. Journal of
alpha7 nicotinic acetylcholine receptor mechanisms in Alzheimer’s Chemistry and Neuroanatomy, 25, 161–73.
disease. Biochemistry and Pharmacology, 82, 931–42. Piggott, M.A., et al. (2007a). Thalamic D2 receptors in dementia with
Pellegrini-Giampietro, D.E., Bennett, M.V., and Zukin, R.S. (1994). AMPA/ Lewy bodies, Parkinson’s disease, and Parkinson’s disease dementia.
kainate receptor gene expression in normal and Alzheimer’s disease International Journal of Neuropsychopharmacology, 10, 231–44.
hippocampus. Neuroscience, 61, 41–9. Piggott, M.A., et al. (2007b ). Selective loss of dopamine D2 receptors
Penney, J.B., et al. (1990). Excitatory amino acid binding sites in the in temporal cortex in dementia with Lewy Bodies, association with
hippocampal region of Alzheimer’s disease and other dementias. Journal cognitive decline. Synapse, 61, 903–11.
of Neurology, Neurosurgery and Psychiatry, 53, 314–20. Piguet, O., et al. (2011). Eating and hypothalamus changes in behavioral-variant
Perry, E.K., et al. (1978). Correlation of cholinergic abnormalities with frontotemporal dementia. Annals of Neurology, 69, 312–19.
senile plaques and mental test scores in senile dementia. British Medical Pimlott, S.L., et al. (2004). Nicotinic acetylcholine receptor distribution in
Journal, 2(6150), 1457–9. Alzheimer’s disease, dementia with Lewy bodies, Parkinson’s disease,
Perry, E., et al. (1998). Clinical neurochemistry: developments in dementia and vascular dementia: in vitro binding study using 5-[(125)i]-a-85380.
research based on brain bank material. Journal of Neural Transmission Neuropsychopharmacology, 29, 108–16.
(Budapest), 105, 915–33. Pohanka, M. (2012). Alpha7 nicotinic acetylcholine receptor is a target in
Perry, E.K., et al. (1985). Cholinergic correlates of cognitive impairment in pharmacology and toxicology. International Journal Molecular Sciences,
Parkinson’s disease: comparisons with Alzheimer’s disease. Journal of 13, 2219–38.
Neurology, Neurosurgery and Psychiatry, 48, 413–21. Potter, P.E., et al. (2011). Pre- and post-synaptic cortical cholinergic
Perry, E.K., et al. (1990a). Cerebral cholinergic activity is related to the deficits are proportional to amyloid plaque presence and density at
incidence of visual hallucinations in senile dementia of Lewy body type. preclinical stages of Alzheimer’s disease. Acta Neuropathologica, 122,
Dementia, 1, 2–4. 49–60.
Perry, E.K., et al. (1990b). Evidence of a monoaminergic-cholinergic Price, D.L., et al. (2010). Alterations in mGluR5 expression and signaling in
imbalance related to visual hallucinations in Lewy body dementia. Lewy body disease and in transgenic models of alpha-synucleinopathy—
Journal of Neurochemistry, 55, 1454–6. implications for excitotoxicity. PLoS One, 5(11), e14020.
Perry, E.K., et al. (1990c). Cholinergic and dopaminergic activities in Procter, A.W., et al. (1988). Evidence of glutamatergic denervation and
senile dementia of Lewy body type. Alzheimer’s Disease and Associated possible abnormal metabolism in Alzheimer’s disease. Journal of
Disorders, 4, 87–95. Neurochemistry, 50, 790–802.
Perry, E.K., et al. (1990d). Cholinergic nicotinic and muscarinic receptors Procter, A.W., Qurne, M., and Francis, P.T. (1999). Neurochemical features of
in dementia of Alzheimer, Parkinson and Lewy body types. Journal frontotemporal dementia. Dementia and Geriatric Cognitive Disorders,
of Neural Transmission. Parkinson’s Disease and Dementia Section, 2, 10 Suppl 1, 80–4.
149–58. Ragnarsson, L., et al. (2002). Spermine modulation of the glutamate(NMDA)
Perry, E.K., et al. (1993). Monoaminergic activities in Lewy body dementia: receptor is differentially responsive to conantokins in normal and
relation to hallucinosis and extrapyramidal features. Journal of Neural Alzheimer’s disease human cerebral cortex. Journal of Neurochemistry,
Transmission. Parkinson’s Disease and Dementia Section, 6, 167–77. 81, 765–79.
Perry, E.K., et al. (1995). Alteration in nicotine binding sites in Parkinson’s Ransmayr, G., et al. (2001). Striatal dopamine transporter function in
disease, Lewy body dementia and Alzheimer’s disease: possible index of dementia with Lewy bodies and Parkinson’s disease. European Journal of
early neuropathology. Neuroscience, 64, 385–95. Nuclear Medicine, 28, 1523–8.
Perry, E., et al. (1999). Acetylcholine in mind: a neurotransmitter correlate of Renner, M., et al. (2010). Deleterious effects of amyloid beta oligomers acting
consciousness? Trends in Neuroscience, 22, 273–80. as an extracellular scaffold for mGluR5. Neuron, 66, 739–54.
Reeves, S., et al. (2009). Increased striatal dopamine (D2/D3) receptor Sharp, S.I., et al. (2008). Cortical serotonin 1A receptor levels are associated
availability and delusions in Alzheimer disease. Neurology , 72, with depression in patients with dementia with Lewy bodies and
528–34. Parkinson’s disease dementia. Dementia and Geriatric Cognitive
Reid, R.T., et al. (2000). Nicotinic receptor losses in dementia with Lewy Disorders, 26, 330–8.
bodies: comparisons with Alzheimer’s disease. Neurobiology and Aging, Silva, A.P., et al. (2005). The putative neuroprotective role of neuropeptide Y
21, 741–6. in the central nervous system. Current Drug Targets and CNS Neurological
Reinikainen, K.J., et al. (1988). Cholinergic deficit in Alzheimer’s disease: Disorders, 4, 331–47.
a study based on CSF and autopsy data. Neurochemistry Research, 13, Sinha, N., Firbank, M., and O’Brien, J.T. (2012). Biomarkers in dementia with
135–46. Lewy bodies: A review. International Journal of Geriatric Psychiatry, 27,
Richards, G., et al. (2010). Altered distribution of mGlu2 receptors in 443–53.
beta-amyloid-affected brain regions of Alzheimer cases and aged Small, G. and Dubois, B. (2007). A review of compliance to treatment in
PS2APP mice. Brain Research, 1363, 180–90. Alzheimer’s disease: potential benefits of a transdermal patch. Current
Rinne, J.O., et al. (2002). Striatal dopamine transporter and extrapyramidal Medical Research Opinions, 23, 2705–13.
symptoms in frontotemporal dementia. Neurology, 58, 1489–93. Smith, C.C., et al. (1985). Putative amino acid transmitters in lumbar
Rissman, R.A. and Mobley, W.C. (2011). Implications for treatment: GABAA cerebrospinal fluid of patients with histologically verified Alzheimer’s
receptors in aging, Down syndrome and Alzheimer’s disease. Journal of dementia. Journal of Neurology, Neurosurgery and Psychiatry, 48,
Neurochemistry, 117, 613–22. 469–71.
Rissman, R.A., et al. (2003). Biochemical analysis of GABA(A) receptor Smith, C.J., et al. (1987). Guanine nucleotide modulation of muscarinic
subunits alpha 1, alpha 5, beta 1, beta 2 in the hippocampus of patients cholinergic receptor binding in postmortem human brain—a preliminary
with Alzheimer’s disease neuropathology. Neuroscience, 120, 695–704. study in Alzheimer’s disease. Neuroscience Letters, 82, 227–32.
Rissman, R.A., De Blas, A.L., and Armstrong, D.M. (2007). GABAA Sparks, D.L. and Markesbery, W.R. (1991). Altered serotonergic and
receptors in aging and Alzheimer’s disease. Journal of Neurochemistry, cholinergic synaptic markers in Pick’s disease. Archives of Neurology, 48,
103, 1285–92. 796–9.
Robert, S.J., et al. (2001). The human serotonin 5-HT4 receptor regulates Sukonick, D.L., et al. (2001). The 5-HTTPR*S/*L polymorphism and
secretion of non-amyloidogenic precursor protein. Journal of Biological aggressive behavior in Alzheimer disease. Archives of Neurology, 58,
Chemistry, 276, 44881–8. 1425–8.
Rodriguez-Puertas, R., et al. (1997). Autoradiographic distribution of M1, Suzuki, M., et al. (2002). Striatal monoaminergic terminals in Lewy body and
M2, M3, and M4 muscarinic receptor subtypes in Alzheimer’s disease. Alzheimer’s dementias. Annals of Neurology, 51, 767–71.
Synapse, 26, 341–50. Sze, C., et al. (2001). N-Methyl-D-aspartate receptor subunit proteins and
Roman, G.C. (2005). Rivastigmine for subcortical vascular dementia. Expert their phosphorylation status are altered selectively in Alzheimer’s disease.
Reviews in Neurotherapy, 5, 309–13. Journal of the Neurological Sciences, 182, 151–9.
Roman, G.C. and Kalaria, R.N. (2006). Vascular determinants of cholinergic Szot, P., et al. (2006). Compensatory changes in the noradrenergic nervous
deficits in Alzheimer disease and vascular dementia. Neurobiology of system in the locus ceruleus and hippocampus of postmortem subjects
Aging, 27, 1769–85. with Alzheimer’s disease and dementia with Lewy bodies. Journal of
Rosenberg, P.B., et al. (2008). Effects of cardiovascular medications on rate Neuroscience, 26, 467–78.
of functional decline in Alzheimer disease. American Journal of Geriatric Teaktong, T., et al. (2003). Alzheimer’s disease is associated with a selective
Psychiatry, 16, 883–92. increase in alpha7 nicotinic acetylcholine receptor immunoreactivity in
Sabbagh, M. and Cummings, J. (2011). Progressive cholinergic decline in astrocytes. Glia, 41, 207–11.
Alzheimer’s disease: consideration for treatment with donepezil 23 mg Teaktong, T., et al. (2005). Muscarinic M2 and M4 receptors in anterior
in patients with moderate to severe symptomatology. BMC Neurology, cingulate cortex: relation to neuropsychiatric symptoms in dementia
11, 21. with Lewy bodies. Behavioural Brain Research, 161, 299–305.
Sabri, O., et al. (2008). Acetylcholine receptors in dementia and mild cognitive Terry Jr, A.V. and Buccafusco, J.J. (2003). The cholinergic hypothesis of age
impairment. European Journal of Nuclear Medicine and Molecular and Alzheimer’s disease-related cognitive deficits: recent challenges and
Imaging, 35 Suppl 1, S30–45. their implications for novel drug development. Journal of Pharmacology
Samuel, W., et al. (1997). Dementia with Lewy bodies versus pure Alzheimer & Experimental Therapeutics, 306, 821–7.
disease: differences in cognition, neuropathology, cholinergic Thorns, V., et al. (1997). Alterations in glutamate receptor 2/3 subunits
dysfunction, and synapse density. Journal of Neuropathology and and amyloid precursor protein expression during the course of
Experimental Neurology, 56, 499–508. Alzheimer’s disease and Lewy body variant. Acta Neuropathologica,
Scatton, B., et al. (1983). Reduction of cortical dopamine, noradrenaline, 94, 539–48.
serotonin and their metabolites in Parkinson’s disease. Brain Research, Tiraboschi, P., et al (2000). Cholinergic dysfunction in diseases with Lewy
275, 321–8. bodies. Neurology, 54, 407–11.
Schaeffer, E.L. and Gattaz, W.F. (2008). Cholinergic and glutamatergic Tiraboschi, P., et al. (2002). Early and widespread cholinergic losses
alterations beginning at the early stages of Alzheimer disease: differentiate dementia with Lewy bodies from Alzheimer disease.
participation of the phospholipase A2 enzyme. Psychopharmacology Archives of General Psychiatry, 59, 946–51.
(Berlin), 198, 1–27. Tsang, S.W., et al. (2006). Impaired coupling of muscarinic M1 receptors to
Scheuer, K., et al. (1996). Cortical NMDA receptor properties and membrane G-proteins in the neocortex is associated with severity of dementia in
fluidity are altered in Alzheimer’s disease. Dementia, 7, 210–14. Alzheimer’s disease. Neurobiology and Aging, 27, 1216–23.
Schmitt, H.P. (2005). On the paradox of ion channel blockade and its benefits Tsang. S.W., et al. (2010). A serotoninergic basis for hyperphagic eating changes
in the treatment of Alzheimer disease. Medical Hypotheses, 65, 259–65. in Alzheimer’s disease. Journal of Neurological Science, 288, 151–5.
Scott, H.L., et al. (2002). Aberrant expression of the glutamate transporter Tsuboi, Y., Uchikado, H., and Dickson, D.W. (2007). Neuropathology of
excitatory amino acid transporter 1 (EAAT1) in Alzheimer’s disease. Parkinson’s disease dementia and dementia with Lewy bodies with
Journal of Neuroscience, 22, RC206. reference to striatal pathology. Parkinsonism and Relative Disorders, 13
Sedaghat, F., et al. (2007). Evaluation of dopaminergic function in Suppl 3, S221–4.
frontotemporal dementia using 123I-FP-CIT single photon emission Tsuno, N. (2009). Donepezil in the treatment of patients with Alzheimer’s
computed tomography. Neurodegenerative Dis, 4, 382–5. disease. Expert Reviews in Neurotherapy, 9, 591–8.
Ulas, J. and Cotman, C.W. (1997). Decreased expression of and histological abnormalities. Journal of Neurological Science, 57,
N-methyl-D-aspartate receptor 1 messenger RNA in select regions of 407–17.
Alzheimer brain. Neuroscience, 79, 973–82. Xuereb, J.H., et al. (1990). Parameters of cholinergic neurotransmission in
Ulas, J., et al. (1992). N-methyl-D-aspartate receptor complex in the the thalamus in Parkinson’s disease and Alzheimer’s disease. Journal of
hippocampus of elderly, normal individuals and those with Alzheimer’s Neurological Science, 99, 185–97.
disease. Neuroscience, 49, 45–61. Yang, Y. and Schmitt, H.P. (2001). Frontotemporal dementia: evidence
Voytko, M.L. (1996). Cognitive functions of the basal forebrain cholinergic for impairment of ascending serotoninergic but not noradrenergic
system in monkeys: memory or attention? Behavioural Brain Research, innervation. Immunocytochemical and quantitative study using a graph
75, 13–25. method. Acta Neuropathologica (Berlin), 101, 256–70.
Walker, Z., et al. (2004). Striatal dopamine transporter in dementia with Lewy Yu, J.T., et al. (2011). Roles of beta-adrenergic receptors in Alzheimer’s
bodies and Parkinson disease: a comparison. Neurology, 62, 1568–72. disease: implications for novel therapeutics. Brain Research Bulletin, 84,
Warren, N.M., et al. (2007). Muscarinic receptors in the thalamus in 111–17.
progressive supranuclear palsy and other neurodegenerative disorders. Yu, W.F., et al. (2005). High selective expression of alpha7 nicotinic receptors
Journal of Neuropathology and Experimental Neurology, 66, 399–404. on astrocytes in the brains of patients with sporadic Alzheimer’s disease
Warren, N.M., et al. (2008). Intact coupling of M1 receptors and preserved and patients carrying Swedish APP 670/671 mutation: a possible
M2 and M4 receptors in the cortex in progressive supranuclear palsy: association with neuritic plaques. Experimental Neurology, 192, 215–25.
contrast with other dementias. Journal of Chemistry and Neuroanatomy, Yu, W., et al. (2012). Upregulation of astrocytic alpha7 nicotinic receptors
35, 268–74. in Alzheimer’s disease brain-possible relevant to amyloid pathology.
Weinberger, D.R., et al. (1991). The distribution of cerebral muscarinic Molecular Neurodegeneration, International Conference on Molecular
acetylcholine receptors in vivo in patients with dementia. Neurodegeneration, Shanghai, p. 2011.
A controlled study with 123IQNB and single photon emission computed Zaja-Milatovic, S., et al. (2005). Dendritic degeneration in neostriatal medium
tomography. Archives of Neurology, 48, 169–76. spiny neurons in Parkinson disease. Neurology, 64, 545–7.
Weinshenker, D. (2008). Functional consequences of locus coeruleus Zaja-Milatovic, S., et al. (2006). Selective dendritic degeneration of medium
degeneration in Alzheimer’s disease. Current Alzheimer’s Research, 5, spiny neurons in dementia with Lewy bodies. Neurology, 66, 1591–3.
342–5. Zarros, A., Kalopita, K.S., and Tsakiris, S.T. (2005). Serotoninergic impairment
Whitehouse, P.J., et al. (1982). Alzheimer’s disease and senile dementia: loss and aggressive behavior in Alzheimer’s disease. Acta Neurobiologiae
of neurons in the basal forebrain. Science, 215, 1237–9. Experimentalis, 65, 277–86.
Wilcock, G.K., et al. (1982). Alzheimer’s disease. Correlation of cortical Zweig, R.M., et al. (1993). The locus ceruleus and dementia in Parkinson’s
choline acetyltransferase activity with the severity of dementia disease. Neurology, 43, 986–91.
CHAPTER 8
Molecular genetics and
biology of dementia
Denise Harold and Julie Williams
Dementia is a syndrome that can result from a variety of diseases Neuropathology of AD

and conditions, and is defined as a global decline in cognitive
Neuropathologically, the disease is characterized by extracellular
functioning, including difficulties with memory, language usage,
senile plaques (SPs) and intracellular NFTs. However, the mere
executive functioning, and activities of daily living. It is a major
presence of SPs and NFTs are not sufficient to indicate disease, for
health issue of increasing concern, both in the UK and world-
both are found in the brains of healthy, aged individuals. Rather it
wide. Alzheimer’s Research UK (2010) estimated the number of
is the density of these lesions, particularly in the limbic and associa-
patients with dementia in the UK to be over 820,000; the World
tion cortices, that signifies disease.
Alzheimer Report estimates the global figure to be 35.6 million
Amyloid plaques consist of extracellular deposits of β-amyloid
(Alzheimer’s Disease International, 2010). Disturbingly, this
(Aβ) peptides, proteolytic derivatives of the amyloid precursor
number is predicted to almost double every 20 years, such that,
protein (APP). When these plaques are associated with swollen,
barring successful therapeutic intervention, the expected number
distorted neuronal processes, they are called neuritic plaques. Two
of dementia sufferers in 2050 will be 115.4 million. However,
species of fibrillar Aβ, 40 and 42 amino acids long (Aβ40 and Aβ42,
there is hope for the future; recent advances in the molecular
respectively), are found in neuritic plaques. The slightly more hydro-
genetics of the dementias are pinpointing potential pathogenic
phobic form, Aβ42, is particularly prone to aggregation (Jarrett et al.,
mechanisms, and in time may provide therapeutic targets. In
1993). The neurites often contain paired helical filaments (indistin-
this chapter, we focus on the most common forms of demen-
guishable to those of NFTs), normal glial processes and abnormal
tia: Alzheimer’s disease, vascular dementia, and frontotemporal
organelles. Microglia are found within and adjacent to the central
dementia.
amyloid core and astrocytes are found at the plaque periphery. In
addition to amyloid fibrils, neuritic plaques include apolipoprotein
Alzheimer’s Disease E (APOE), clusterin (CLU), complement factors, α2-macroglobulin
Alzheimer’s disease (AD), a devastating, progressive neurode- (A2M), α1-antichymotrypsin, and low-density lipoprotein recep-
generative disorder, accounts for more than half of all cases of tor related protein (LRP), amongst other components.
dementia among people over 65 years of age (Alzheimer’s Disease Only a subset of all Aβ deposits in the AD brain is associated
International, 2009). The disease is characterized by cognitive with neuritic plaques. Many of the plaques found in limbic and
decline and behavioural symptoms, and mean survival after association cortices and almost all of those found in areas of the
symptom onset for AD is 10.3 years but may range from 2 to more brain not typically associated with AD do not have a fibrillar, com-
than 20 years (Mann et al., 1992). Diagnosis of AD was standard- pacted centre, nor are they associated with neuritic dystrophy or
ized in 1984 with criteria developed by the Work Group of the glial changes. These diffuse plaques are comprised primarily of
National Institute of Neurologic and Communicative Disorders Aβ42, with little or no Aβ40, and are believed to represent immature
and Stroke and the Alzheimer’s Disease and Associated Disorders lesions that are precursors to neuritic plaques. Lending support to
Association (NINCDS–ADRDA) (McKhann et al., 1984), and this theory is the fact that often only diffuse plaques are found in
these criteria were revised in 2011 to reflect knowledge accumu- limbic and association cortices of healthy, aged brains. In addition,
lated in the intervening 27 years (McKhann et al., 2011). Clinical individuals with Down’s syndrome often display diffuse deposits as
criteria allow a diagnosis of possible or probable AD, but a defini- early as their teenage years, but do not show neuritic plaques until
tive diagnosis requires examination of the brain post mortem and some two decades later, a time at which they first display abundant
demonstration of more neurofibrillary tangles (NFTs) and neu- NFTs in limbic and association cortices (Lemere et al., 1996).
ritic plaques, particularly in the cerebral cortex, than expected for NFTs are relatively insoluble, intracellular aggregates composed
the patient’s age. of paired helical filaments (PHFs) that occupy the cell body and
may extend into the dendrites but do not occur in the axon. PHFs of the disease; the subsequent identification of rare mutations in
consist of protofilaments arranged to form a tubule and con- three genes that cause early-onset AD has had a major impact on
tain abnormally phosphorylated microtubule-associated tau (τ) our understanding of the pathogenesis of the disease.
protein. Phosphate groups are attached and removed from τ in a
dynamic process that regulates the ability of the protein to facilitate The amyloid precursor protein (APP)
the assembly and stabilization of microtubules. In AD, the τ protein The first success came from analyses of chromosome 21. Middle-aged
accumulates an excess number of phosphates and becomes dys- individuals with Down’s syndrome (trisomy 21) commonly develop
functional, dissociating from the microtubule and resulting in the the clinical features of AD and invariably present with the neu-
destabilization and disrupted assembly of this important cytoskele- ropathological hallmarks of the disease at autopsy (Mann, 1988a,
tal component of the intracellular transport system (Lee et al., 1991; 1988b). This provided a good indicator for the presence of an AD
Clark et al., 1997). NFTs lead to the death of the nerve cells in which gene on chromosome 21. Indeed, the first positive AD genetic
they occur (although nerve cell death also occurs in regions with linkage was discovered between a locus on chromosome 21q and
few tangles). These lesions are not unique to AD and occur in other autosomal dominant EOAD (St George-Hyslop et al., 1987). In sub-
neurodegenerative disorders, including frontotemporal dementia, sequent analyses, some families with autosomal dominant EOAD
corticobasal degeneration, progressive supranuclear palsy, and failed to show linkage to chromosome 21 markers, which cast doubt
dementia pugilistica; as previously mentioned, a small number are on the initial positive linkage finding. As we now know, locus het-
also found in normal ageing. erogeneity exists within autosomal dominant EOAD and Goate
SPs and NFTs are regionally specific, occurring predominantly and colleagues (1991) went on to discover a point mutation in the
in the hippocampus, entorhinal cortex, and association areas of amyloid precursor protein (APP) gene in a family showing linkage
the neocortex. The cognitive phenotype of AD reflects the location on chromosome 21. Since then, a number of pathogenic mutations
of these lesions. Hippocampal dysfunction has been related to the have been identified in autosomal dominant EOAD families: there
memory impairment suffered by individuals with AD. Dysfunction are currently 33 APP mutations listed in the Alzheimer Disease and
of left posterior association cortex produces the fluent aphasia and Frontotemporal Dementia Mutation Database (<http://www.mol-
involvement of the right posterior association cortex leads to the gen.ua.ac.be/ADMutations/default.cfm>).
visuospatial dysfunction typical of patients with AD (Cummings,
2003). The presenilins (PSEN1 and PSEN2)
In addition to plaques and tangles, granulovacuolar degeneration As it was evident that APP mutations did not account for all cases
and amyloid angiopathy are evident in the brains of AD patients. of autosomal dominant EOAD, the search for further pathogenic
Granulovacuolar degeneration is highly selective for the pyramidal loci continued. In 1992, a locus on the long arm of chromosome
neurons of the hippocampus. Vacuoles are present in the cytoplasm 14 was detected by linkage analysis, and a few years later a gene,
of these cells, with each vacuole containing a single dense granule. named presenilin 1 (PSEN1), was identified and isolated by a posi-
Several vacuoles may occur within the same cell body. Amyloid tional cloning strategy (Van Broeckhoven et al., 1992; Sherrington
angiopathy refers to the deposition of fibrillar amyloid in small et al., 1995). Similarly, evidence for a locus on chromosome 1q was
arterioles, venules, and capillaries within the cerebral cortex. The reported in several kindreds known as the Volga-German families,
amyloid appears to principally be the Aβ40 species (Suzuki et al., a group of related kindreds of German–Russian origin with multi-
1994b). ple cases of autosomal dominant EOAD (Levy-Lahad et al., 1995b).
There is significant neuronal loss in AD. Degeneration of the Subsequently, the presenilin 2 gene (PSEN2) was identified based
basal forebrain cholinergic system is a classic feature of AD and on its homology to PSEN1 (Levy-Lahad et al., 1995a; Rogaev et al.,
has been correlated with the depth of dementia, at least in late 1995). (1The Alzheimer Disease and Frontotemporal Dementia
stages of the disease. In addition, neuronal loss or atrophy in the Mutation Database lists 185 PSEN1 pathogenic mutations and 13
locus ceruleus and raphe nuclei of the brainstem leads to deficits in PSEN2 pathogenic mutations. APP and PSEN1 mutations lead to
noradrenergic and serotonergic transmitters, respectively (Lantos AD with early onset (age 45–60) and rapid progression (death in
and Cairns, 2000). 6–8 years). Mutations in PSEN2 also produce autosomal dominant
AD, but with a more variable age at onset (Cummings, 2003).
Symptom onset in Alzheimer’s disease
AD has been traditionally divided into late-onset AD (LOAD) APP processing
with an age at onset greater than 60–65 years, and early-onset AD The amyloid precursor protein, APP, and related family members
(EOAD) with age at onset below this cut-off (the latter accounting amyloid-β (A4) precursor-like protein 1 and 2 (APLP1 and APLP2)
for less than 5% of all AD cases (Shastry and Giblin, 1999)). A small are homologous type I transmembrane proteins (Selkoe, 2001). APP
proportion of the already uncommon early-onset cases appear to is ubiquitously expressed and contains a large extracellular region,
segregate in an autosomal dominant manner. In a population-based a transmembrane domain, and a small cytoplasmic tail. Alternative
study in the city of Rouen, Campion and colleagues (1999) calcu- splicing of the APP gene on chromosome 21 results in at least eight
lated the prevalence of EOAD to be 41.2 per 100,000 individuals at isoforms of the protein (Bayer et al., 1999), but the predominant
risk and the prevalence of autosomal dominant EOAD to be 5.3 per isoforms are APP695, APP751, and APP770, containing 695, 751,
100,000 individuals at risk (12.9% of EOAD). and 770 amino acids, respectively. APP695, the isoform primarily
expressed in CNS neurons, differs from APP751 and APP770 in
Molecular genetics of EOAD that it lacks a Kunitz proteinase inhibitor (KPI) domain found in
Initial attempts to define the genetic architecture of AD began in the the other two isoforms. APP is synthesized in the endoplasmic
1980s and focused on the early-onset, autosomal dominant forms reticulum and undergoes several post-translational modifications
CHAPTER 8 molecular genetics and biology of dementia 125
including N-glycosylation, O-glycosylation, and tyrosyl-sulphation (Ehehalt et al., 2003). APP processing by BACE1 also occurs pref-
in the Golgi apparatus. This mature form of APP is transported to erentially within lipid rafts (Ehehalt et al., 2003). Like APP, BACE1
the cell surface via the secretory pathway. APP is also endocytosed can be internalized in endosomes, which may be the preferential
from the cell surface and processed in the endosomal–lysosomal site of BACE-1 activity due to its acidic pH (Vassar, 2001). Studies
pathway (Kang et al., 1987; Shioi et al., 1992, 1993; Suzuki et al., in BACE1 knockout mice provide strong evidence that BACE1 is
1994c; Thinakaran et al., 1995; Russo et al., 2001). the major β-secretase in brain. Knockout of the BACE1 gene com-
APP is processed via two proteolytic pathways involving the secre- pletely impairs the β-secretase cleavage of APP and abolishes the
tases α-, β-, and γ-secretase. In the ‘nonamyloidogenic’ pathway, generation of Aβ (Cai et al., 2001). These mice develop normally
APP is cleaved by α-secretase, which cuts within the Aβ domain and show no phenotypic alterations despite lacking the primary
and thus precludes Aβ peptide generation. Following the release β-secretase activity in brain (Luo et al., 2001). BACE1 can cleave
of the soluble extracellular α-N-terminal fragment (α-APPs), the APP either at Asp1 of the Aβ sequence or at Glu11 (the β and β’
C-terminal APP fragment (C83) undergoes γ-cleavage, leading to cleavage site, respectively) (Gouras et al., 1998). A homologous
the generation of the p3 peptide. Alternatively, in the amyloidog- enzyme to BACE1 has been identified. BACE2 is 51% identical
enic pathway APP is cleaved by β-secretase to generate soluble to BACE1, although it is apparently not significantly involved in
extracellular β-N-terminal fragment (β-APPs) and the C-terminal Aβ generation as it is only expressed at very low levels in the brain
APP fragment, C99. Subsequent cleavage of C99 by γ-secretase (Bennett et al., 2000). In vitro assays with peptide substrates dem-
results in the formation of the Aβ peptide. onstrate that BACE2 can cleave APP at the β-secretase site (Farzan
α-Secretase cleavage of APP occurs between residues Lys16 and et al., 2000; Hussain et al., 2000). However, it also cleaves between
Leu17 of the Aβ sequence and is the predominant pathway of APP Phe19-Phe20 and Phe20-Ala21 within the Aβ peptide, resulting in
metabolism in most cells (Nitsch et al., 1992). This occurs in both increased secretion of α-APPs and P3-like products and reduced
a constitutive manner and as a result of stimulation by regulatory production of Aβ species (Farzan et al., 2000).
factors (e.g. activation of protein kinase C by phorbol esters is able γ-Secretase is unusual in that it cleaves APP within the membrane
to up-regulate the α-cleavage pathway). Immunocytochemical bilayer, which was previously thought to be biochemically impos-
analysis has indicated that α-cleavage occurs within the endoplas- sible (Weihofen and Martoglio, 2003). It is, however, reminiscent of
mic reticulum and various compartments within the trans-Golgi the intramembrane proteolysis of Notch, which is cleaved during
apparatus (Nunan and Small, 2000). In addition, there is α-secretase development in a presenilin-dependent fashion, thereby releasing a
activity at the cell membrane (Parvathy et al., 1998, 1999). Three fragment (called the Notch intracellular domain (NICD)) that acti-
members of the ADAMs family of proteins have been demonstrated vates the transcription of genes involved in cell-fate determination
to have α-secretase activity. ADAMs are multidomain proteins and (Schroeter et al., 1998; De Strooper et al., 1999).
some members have been implicated in ectodomain shedding via Presenilin 1 and 2 (PSEN1 and PSEN2) are ubiquitously
a metalloprotease domain (Hooper et al., 1997). Experiments con- expressed proteins, comprising eight membrane-spanning regions
ducted on embryonic fibroblasts derived from mice with a knockout with cytoplasmic orientation for both the N- and C-termini (Huse
of ADAM17 (also known as the TNFα converting enzyme, TACE) and Doms, 2000; Mushegian, 2002). They share 76% homol-
show that phorbol esters-mediated release of α-APPs is completely ogy, with the area of highest variability being in the N-terminal
blocked by the loss of the protease (Buxbaum et al., 1998). However, region and the large cytoplasmic loop between transmembrane
constitutive α-secretase activity is not affected by disruption of domains 6 and 7 (Rogaev et al., 1995). Both presenilins undergo
the ADAM17 gene. Similarly, co-expression of ADAM9 with APP endoproteolytic cleavage within this large loop (Thinakaran et al.,
in COS cells leads to an increase in production of α-APPs upon 1996). The resulting N- and C-terminal fragments remain associ-
phorbol ester treatment of the cells, above that seen in the absence ated (Capell et al., 1998) and this association seems to be critical
of transfected ADAM9 (Koike et al., 1999). ADAM10 has been for presenilin function (Haass and Steiner, 2002). The presenilins
implicated in both the constitutive and protein kinase C-regulated were implicated in γ-secretase function when it was found that a
α-secretase pathways (Lammich et al., 1999; Skovronsky et al., PSEN1 knockout severely reduced Aβ generation (De Strooper
2000). A significant decrease of platelet ADAM10 levels has been et al., 1998). When the PSEN2 gene was eliminated as well, no Aβ
observed in AD patients, together with a similar decrease in α-APPs generation was observed at all (Herreman et al., 2000; Zhang et al.,
in both thrombin-activated platelets and CSF (Colciaghi et al., 2000). Moreover, experiments using several γ-secretase inhibi-
2002). Moreover, in an AD mouse model that overexpresses mutated tors have shown that they specifically bind to the presenilin N- or
human APP (V717I) in neurons, and therefore leads to Aβ deposi- C-terminal fragments and inhibit the γ-secretase activity (Esler et
tion and memory deficits, overexpression of ADAM10 increased al., 2000; Li et al., 2000). The presenilins contain two functionally
the secretion of α-APPs, reduced the formation of Aβ peptides, and important aspartate residues within transmembrane domains 6 and
alleviated cognitive deficits (Postina et al., 2004). 7 that appear to be necessary for normal γ-secretase activity. When
The enzyme responsible for β-secretase cleavage has been cloned either, or both, of these residues are mutated in PSEN1, Aβ and P3
and is called BACE1, for β-site APP cleaving enzyme (Hussain production are greatly reduced with a concomitant increase in the
et al., 1999; Sinha et al., 1999; Vassar et al., 1999; Yan et al., 1999; levels of APP C-terminal fragments (Wolfe et al., 1999; Herreman
Lin et al., 2000). BACE1 is a type 1 transmembrane protein, con- et al., 2000; Kimberly et al., 2002). Presenilins alone are not suffi-
taining two active site motifs with a conserved sequence, similar cient for γ-cleavage. Instead, γ-secretase activity is associated with
to aspartyl proteases. Like many proteases, BACE1 is synthesized a high molecular weight PSEN-containing complex (Capell et al.,
as a zymogen. After cleavage of a prodomain that maintains the 1998; Haass and Steiner, 2002; De Strooper, 2003). Biochemical
enzyme in a latent form, BACE1 is targeted through the secretory purification of this complex led to the identification of the type I
pathway to the plasma membrane and clusters within lipid rafts transmembrane glycoprotein nicastrin (Yu et al., 2000). A genetic
screen of Caenorhabditis elegans revealed two genes, aph-1 and et al., 2001). Dystrophic neurites in plaque vicinity, heavy astroglio-
pen-2, each encoding multipass transmembrane proteins that inter- sis, and microglia activation are also observed. Although neurode-
act strongly with the presenilin orthologue sel-12 and the nicastrin generation is not observed to a degree anywhere near that seen in
orthologue aph-2 (Francis et al., 2002). When all four components human AD, there is loss of synaptophysin staining in aged PDAPP
were expressed together in Saccharomyces cerevisiae, an organ- mice (Games et al., 1995), a modest loss of neurons in the APP23
ism that lacks any endogenous γ-secretase activity, fully active mice (Calhoun et al., 1998), and an even smaller, though statisti-
γ-secretase was reconstituted (Edbauer et al., 2003). cally significant, neuronal loss in the hippocampus of Tg2576 mice
(Takeuchi et al., 2000). It is perhaps notable that none of the APP
Mutations in APP, PSEN1, and PSEN2 or APP/PSEN mice shows evidence of NFTs, a fact often cited in
Alteration of APP processing appears to be the mechanism by which arguments against the amyloid cascade hypothesis.
autosomal dominant mutations in the APP and presenilin genes
cause AD. In APP, pathogenic mutations are found in exons 16 and Molecular genetics of LOAD
17, clustered around the β- and γ-secretase sites. The effects of these While much progress has been made in determining the causes of
mutations have been elucidated somewhat by cell culture studies. autosomal dominant EOAD, it must be remembered that this form
For example, a double mutation at amino acids 670 and 671 from of the disease is rare. The inheritance pattern of the more common
Lys–Met to Asp–Leu (the ‘Swedish’ mutation) that lies upstream late-onset form of the disease appears to be much more complex.
of the β-cleavage site results in a five- to eightfold increase in the There is considerable evidence that familial factors play an impor-
formation of both Aβ40 and Aβ42 (Citron et al., 1992). Mutations at tant role in the aetiology of LOAD. For example, it has long been
amino acid 717, such as the ‘London’ mutation (Val → Ile) and the known that there is a markedly increased cumulative risk of demen-
‘Indiana’ mutation (Val → Phe), lie close to the γ-secretase site and tia among first-degree relatives of individuals with AD. In a reanaly-
results in a more than twofold increase of the more insoluble Aβ42, sis of seven case-control studies, first-degree relatives of AD patients
which rapidly aggregates to form amyloid depositions (Suzuki had a 3.5-fold increase in risk for developing AD (van Duijn et al.,
et al., 1994a). Mutations in the presenilin genes are predominantly 1991). This relative risk decreased with increased age at onset of the
located in the highly conserved transmembrane domains and are proband, but even by age 80 years a statistically significant 2.6-fold
presumed to distort the precise conformation of the molecule increase in risk for first-degree relatives was observed. That a dis-
within the membrane. The effect of these mutations again appears ease shows familial aggregation, however, does not necessarily mean
to be the enhanced production of Aβ42 (Borchelt et al., 1996; Wolfe that it has a genetic component; the higher incidence of the disorder
et al., 1999). Therefore, in these families, the cause of AD appears to may simply be the result of shared family environment. Twin stud-
be a direct result of increased levels of Aβ, with Aβ42 in particular ies are commonly undertaken to distinguish between the effects of
associated with neurotoxicity. genetic and environmental factors, based on the differences in con-
cordance rates observed between monozygotic (MZ) and dizygotic
The amyloid cascade hypothesis (DZ) twins. A number of AD twin studies have been performed in
The discovery of the aforementioned mutations and their mech- the past. Perhaps the most comprehensive studies were based upon
anism of action led to the formulation of the amyloid cascade the well-established Swedish twin registry. The Swedish Study of
hypothesis, which postulates that elevated levels of Aβ result in Dementia in Twins reported data from twins who were reared apart,
the oligomerization, fibrillogenesis, and aggregation of the peptide. and a similar number of pairs who were reared together (Gatz et al.,
This is thought to initiate a cascade of injurious events, including 1997). The concordance rate for MZ twins for AD was 67%, com-
free radical production, glial activation, and direct neuronal dam- pared to 22% among DZ twins, resulting in a heritability estimate
age. Soluble Aβ oligomers are themselves a neurotoxic species prior of between 75% and 85%. Gatz and colleagues later extended this
to deposition in senile plaques (Walsh et al., 2002). They can lead to work, reporting analyses from the HARMONY study (Gatz et al.,
the generation of hydroxyl ions and oxidative injury of phospholi- 2006). They incorporated data from over 4000 twin pairs, aged over
pid membranes, and this loss of membrane integrity leads to cell 65, who had completed at least basic telephone cognitive screening.
death (Pappolla et al., 1998). Aβ may also exert toxicity through The heritability of AD was estimated to be between 58% and 79%.
influences on calcium channels, exaggerating calcium influx and Given the high heritability of LOAD, numerous genetic studies have
initiating cell death pathways. Although controversial, the hyper- been performed, aiming to identify susceptibility genes for the dis-
phosphorylation of τ is believed to be a downstream consequence ease, including linkage analyses, candidate gene studies, and, more
of increased Aβ levels and results in NFTs and cell death. The recently, genome-wide association studies.
aggregation of Aβ in plaques precipitates microglial activation and
the inflammatory response that surrounds the extracellular lesions. Candidate gene studies and apolipoprotein E (APOE)
These reactive changes may further compromise brain function. Successes in linkage analysis have not been restricted to autosomal
The combination of plaques, tangles, neuronal loss, and neuro- dominant EOAD families. Late-onset families have also been exam-
transmitter deficits results in AD. The generation of transgenic mice ined, leading to the identification of a linkage region for LOAD on
that overexpress human APP carrying autosomal dominant muta- chromosome 19 (Pericak-Vance et al., 1991). Subsequent studies
tions has lent support to the hypothesis that Aβ is central to AD. identified an association between variation at the apolipoprotein E
These mice produce an age-dependent increase in Aβ and develop (APOE) gene on chromosome 19q13.2 with LOAD (Strittmatter
deposits resembling senile plaques in the brain regions most heav- et al., 1993). There are three major isoforms of apoE—E2, E3,
ily affected in AD and demonstrate spatial memory deficits (Quon and E4—which differ from each other by one or two amino acids.
et al., 1991; Games et al., 1995; Hsiao et al., 1996; Sturchler-Pierrat These isoforms are coded for by alleles ε2, ε3, and ε4, respectively.
et al., 1997; Holcomb et al., 1998; Lewis et al., 2000, 2001; Chishti Numerous studies have consistently found an increased risk of
LOAD in carriers of the ε4 allele, whereas the ε2 allele appears a risk allele of frequency 0.25 in the population, conferring an odds
to have a protective effect. Warwick Daw and colleagues (2000) ratio of 1.3. This requirement for large sample sizes has been recog-
have estimated that APOE genotype can make a difference of up to nized and addressed by the more recent AD GWAS.
17 years in age at onset of AD, and presence of the ε4 allele is also In 2009, we undertook a large two-stage GWAS of AD as
associated with an earlier expression of clinical symptoms in carri- part of a collaborative consortium called GERAD (Genetic and
ers of specific APP (Sorbi et al., 1995) or PSEN1 mutations (Pastor Environmental Risk in Alzheimer’s Disease). In the first stage, 3941
et al., 2003). It should be stressed, however, that the ε4 allele is a AD cases and 7848 controls ascertained from Europe and the USA
risk factor rather than causative mutation for LOAD and is neither were genotyped at up to 529,205 autosomal SNPs; in the second
necessary nor sufficient to cause the disease. stage, a subset of markers was genotyped in an independent sam-
The relatively large effect size of the ε4 allele undoubtedly aided ple of 2023 cases and 2340 controls (Harold et al., 2009). Another
the early identification of APOE as a susceptibility locus in 1993. consortium, the European Alzheimer’s Disease Initiative (EADI),
However, little progress was made in the field of AD genetics in the also performed an independent GWAS, typing 537,029 autosomal
subsequent 15 years. This has not been due to lack of effort; accord- markers in 2032 AD cases and 5328 controls ascertained in France,
ing to the AlzGene database, over 1300 association studies in AD and following up a subset of SNPs in 3978 AD cases and 3297 con-
have been performed to date, examining over 650 genes (Bertram trols ascertained in Belgium, Finland, Italy, and Spain (Lambert
et al., 2007). These were predominantly candidate gene studies, et al., 2009). These studies were much larger, and therefore more
and although a number of significant associations with AD were powerful, than previous GWAS of AD (see Table 8.1). Both stud-
reported, most have failed to replicate in subsequent studies. There ies employed Illumina arrays, principally the Human610-Quad,
are problems inherent in the candidate gene approach: the candi- thus allowing direct comparability of the results. The best proxy for
date genes are typically selected either because of their position in an the APOE ε4 SNP (rs429358) on the Illumina Human610-Quad is
implicated region (e.g. a linkage region) or because something about rs2075650 (pairwise r2 = 0.48), and unsurprisingly this was the most
the function of the encoded protein suggests it may play a role in the significant SNP identified in both studies (GERAD P = 2 × 10–157;
disease. Linkage analyses have had limited success in the study of EADI P = 1 × 10–130). GERAD also reported genome-wide signifi-
complex diseases, and even a genuine region of linkage can be several cant evidence in stage 1, with support in an independent replication
megabases. Moreover, functional candidate gene studies suffer from sample, for two novel susceptibility loci: with rs11136000 in the CLU
the problem that understanding of the pathogenesis of a disease is gene (P = 8.5 × 10–10, OR = 0.86) and with rs3851179 and rs541458,
usually incomplete; thus, the number of plausible candidate genes two SNPs 5′ to the PICALM gene (P = 1.3 × 10–9, OR = 0.86 and
is large and the prior probability that a given variant is associated 8.3 × 10–10, OR = 0.86 respectively). Encouragingly, EADI also
with disease is low. Fortunately, thanks to our increased knowledge identified genome-wide significant association with the same allele
of human genetic variation provided by the Human Genome project of rs11136000 in CLU (P = 7.5 × 10–9, OR = 0.86) and found sup-
and the HapMap project, and improvements in high throughput gen- port for the PICALM locus (P = 0.03 and P = 3 × 10–3 for rs3851179
otyping, it has become possible to take a hypothesis-free approach to and rs541458 respectively). EADI identified genome-wide signifi-
association studies by genotyping markers throughout the genome, cant association in their combined sample with rs6656401 in the
regardless of their position or proposed functional nature. Since the CR1 gene (P = 3.7 × 10–9, OR = 1.21). This SNP is not present on
first genome-wide association studies (GWAS) in 2005, GWAS have the Illumina Human610-Quad (and was originally imputed in the
proved enormously successful in identifying susceptibility loci in EADI study), but GERAD detected association with a proxy SNP,
complex disease (Hindorff et al., 2012). rs3818361 (pairwise r2 = 0.83), also in the CR1 gene (GERAD P =
9.2 × 10−6, OR = 1.17; EADI P = 8.9 × 10−8, OR = 1.19).
Genome-wide association studies of Notably, the GERAD study identified a significant excess of
Alzheimer’s disease SNPs in its dataset that, while not meeting stringent criteria for
Since 2007, several GWAS of AD have been performed (see Table 8.1 genome-wide significance, still shows strong evidence for associa-
for studies including at least 150 individuals). The earlier studies tion with AD (P < 1 ×10–5), indicating that additional susceptibility
of AD were based on samples including up to 1100 AD cases; a genes remain to be identified. Included amongst these ‘sub-threshold’
reasonable approach at the time, given success stories like the iden- SNPs were variants 5′ to the BIN1 gene (e.g. rs744373, P = 3.2 × 10−6,
tification of CFH as a susceptibility gene for age-related macular OR = 1.17) and SNPs at the MS4A gene cluster (e.g. rs610932, P = 1.4
degeneration in a GWAS of 96 cases and 50 controls (Klein et al., × 10−6, OR = 0.87). The BIN1 locus received further support from a
2005), and the fact that at least one variant of large effect, APOE subsequent GWAS by the Cohorts for Heart and Aging Research in
ε4, is already known to exist in AD. Indeed, the APOE locus was Genomic Epidemiology (CHARGE) consortium, which performed
consistently identified, but although a number of additional prom- a three-stage analysis of new and previously published GWAS data
ising candidate genes were highlighted, there was little overlap in (Seshadri et al., 2010). In stage one, new data from four cohort stud-
the top results of these early GWAS (see Avramopoulos (2009) for a ies were included in a meta-analysis with previously reported results
detailed review). What these studies have told us is that additional from Reiman et al. (2007) and Carrasquillo et al. (2009), resulting
common susceptibility variants with effect sizes of the same mag- in a combined dataset of 3006 AD cases and 14,648 controls. The
nitude as the APOE ε4 allele are unlikely to exist. As with most most significant SNPs were then meta-analysed with data from
complex diseases, common risk variants for AD are expected to EADI (stage 2) and GERAD (stage 3). CHARGE identified asso-
be of small to moderate effect, with odds ratios less than 1.3. As ciation with the BIN1 SNP rs744373 in their stage 1 data (P = 4.9
such, more powerful samples are required. For example, a sample × 10–4, OR = 1.13); when combined with the GERAD and EADI
of 2000 cases and 5000 controls has just over 80% power to detect data, this SNP surpasses the threshold for genome-wide significance
genome-wide significance (typically accepted as P = 5 × 10–8), with (P = 1.59 × 10–11, OR = 1.15). It also replicated the association in
Table 8.1 Genome-wide association studies of Alzheimer’s disease

Study Year GWAS Stage 1 cases, Sample Novel genes (P ≤ 5 × 10–8)
controls
Grupe et al. * 2007 380, 396 UK, US NA
Coon et al. 2007 664, 422 US, Netherlands NA
Reiman et al. 2007 861, 550a US, Netherlands GAB2
Li et al. 2008 753, 736 Canada NA
Abraham et al. * 2008 1082, 1239b UK NA
Bertram et al. 2008 410 families US NA
Beecham et al. 2009 492, 498 US NA
Carrasquillo et al. 2009 844, 1255 US PCDH11X
Potkin et al. 2009 172, 209 US NA
Harold et al.† 2009 3941, 7848b,c,d Europe, US CLU, PICALM
Lambert et al. † 2009 2025, 5328 France CLU, CR1
Heinzen et al. 2010 331, 368e US NA
Seshadri et al. 2010 3006, 14648d,f Europe, US BIN1, EXOC3L2
Naj et al. 2010 931, 1104e US MTHFD1L
Lee et al. 2011 549, 544 Caribbean Hispanic NA
Wijsman et al. 2011 992 families US CUGBP2
Hu et al. 2011 1831, 1764g,h US, Canada NA
Hollingworth et al. ‡ 2011 6688, 13,685f,h,i,j Europe, US ABCA7, MS4A gene cluster, CD2AP, EPHA1, CD33
Naj et al. ‡ 2011 8309, 7366f,g,h,k,l US MS4A gene cluster, CD2AP, EPHA1, CD33
Antunez et al. 2011 3009, 3006f,g,h,l Spain, US, Canada NA
Logue et al. 2011 513, 496 African-American NA
Only studies with more than 150 cases have been included.
* Pooling study.
† ‡ Published in the same issue of Nature Genetics.
a Includes Coon et al. (2007) dataset.
b A subset of samples was also included in Grupe et al. (2007).
c A subset of samples was also included in Abraham et al. (2008).
d Includes Carrasquillo et al. (2009) dataset.
e Some overlap with Beecham et al. (2009).
f Includes/large overlap with Reiman et al. (2007) dataset.
g Includes Li et al. (2008) dataset.
h Includes Potkin et al. (2009) dataset.
i Includes Harold et al. (2009) dataset.
j Includes Lambert et al. (2009) dataset.
k Includes Naj et al. (2010) dataset.
l Includes Wijsman et al. (2011) dataset.
an independent sample from Spain (P = 0.02, OR = 1.17). Notably, the MS4A gene cluster (P = 1.2 × 10−16). The American Alzheimer’s
CHARGE replicated association with the CLU SNP rs11136000 and Disease Genetic Consortium (ADGC) also reported genome-wide
the PICALM SNP rs3851179 (stage 1 P = 5.0 × 10–4 and P = 1.2 × significant evidence at the MS4A gene cluster, further support for
10–5, respectively). Taken together, these studies provide compelling ABCA7 and suggestive evidence for association with SNPs at the
evidence that CLU, PICALM,CR1, and BIN1 are genuine suscepti- CD33, CD2AP, ARID5B, and EPHA1 loci. When combining data
bility genes for AD, and these findings have been replicated in sev- from ADGC and GERAD+, SNPs at CD33 (P = 1.6 × 10−9), CD2AP
eral independent datasets (Carrasquillo et al., 2010; Hollingworth (P = 8.6 × 10−9), and EPHA1 (P = 6.0 × 10−10) also exceed criteria
et al., 2011a; Hu et al., 2011; Naj et al., 2011). for genome-wide significant association with AD.
More recently, the GERAD consortium has reported findings
from an extended study (GERAD+), which included 19,870 AD Pathogenic mechanisms in LOAD
cases and 39,846 controls and identified genome-wide significant The elucidation of the pathogenic role of early-onset autosomal
evidence for association at the ABCA7 locus (P = 5.0 × 10−21) and dominant mutations in APP, PSEN1, and PSEN2 implicated
Aβ aggregation/clearance as a key process in AD, and it is interest- ABCA7, CD33, and EPHA1 all have putative functions in the
ing to note that the genetic risk factors identified in LOAD may well immune system. ABCA7 is known to modulate phagocytosis and
impact on this process. Apolipoprotein E and clusterin have both clearance of apoptotic cells. In response to stimulation with C1q
been shown to modify Aβ clearance at the blood–brain barrier (Bell or apoptotic cells, ABCA7 moves to the macrophage cell surface
et al., 2007). Levels of APOE protein appear to be inversely propor- and colocalizes with the low-density lipoprotein receptor-related
tional to APOE ε4 allele dosage, with protein levels reduced in ε4 protein 1 (LRP1). ABCA7 appears to facilitate the cell surface
homozygotes compared with heterozygotes. Conversely, clusterin localization of LRP1 and associated signalling via extracellular
levels are increased in proportion to APOE ε4 allele dose levels, sug- signal-related kinase (ERK), thereby promoting the phagoctyo-
gesting a compensatory induction of CLU in individuals with low sis of apoptotic cells (Jehle et al., 2006). CD33 is a member of the
APOE levels (Bertrand et al., 1995). Thus, CLU may modulate Aβ sialic-acid binding immunoglobulin-like lectins (Siglec) family. The
clearance from the brain in concert with APOE. Interestingly, com- CD33-related Siglecs are mainly expressed by mature cells of the
plement receptor 1, which is expressed widely on the extracellular innate immune system. Siglecs have the potential to mediate both
membrane of a number of blood cells but also in brain, has been cell–cell interactions and signalling functions in both the innate
shown to act as a receptor for Aβ, implicating CR1 in Aβ clearance and adaptive immune systems (Crocker et al., 2007). However,
from the brain and the circulatory system (Rogers et al., 2006). Also their precise functions and biologically relevant ligands are yet to
of note, ABCA7 has been shown to regulate APP processing and be determined. EPHA1 is a member of the ephrin receptor sub-
inhibit β-amyloid secretion in cultured cells overexpressing APP family. Ephrins and Eph receptors are membrane-bound proteins
(Chan et al., 2008). that have roles in cell and axon guidance (Martinez et al., 2005) as
Intriguingly, pathway analysis of GWAS data has implicated the well as in synaptic development and plasticity (Lai and Ip, 2009).
immune system and cholesterol metabolism in the aetiology of AD Functions in apoptosis and inflammation have also been proposed.
(Jones et al., 2010). The realization that the immune system, spe- These include Eph signalling as a physiological trigger for apoptosis
cifically the inflammatory response, plays a role in AD is not new (Depaepe et al., 2005) and the promotion of inflammation through
(Zotova et al., 2010); markers of inflammation have been shown to a decrease in expression of Eph receptors, including EphA1, on
associate with amyloid plaques (McGeer and McGeer, 2001) and leukocytes and endothelial cells, thus promoting adhesion of these
inflammatory processes have been proposed as pathogenic con- cells and thereby inflammation (Ivanov and Romanovsky, 2006).
tributors (Bates et al. 2009). However, the inflammatory response BIN1 and CD2AP may also have putative functions in immunity.
has generally been considered to be a secondary event. BIN1 knockout mosaic mice have been reported to show reduced
Both CLU and CR1 play significant roles in inflammation as well inflammation with aging (Chang et al., 2007). CD2AP is hypoth-
as in innate and adaptive immunity. CLU is a multifunctional mol- esized to function as a molecular scaffold for receptor patterning
ecule whose roles include modulation of the complement system; and cytoskeletal polarization events which are critical to the forma-
the protein is able to prevent the inflammatory response associated tion of an effective T cell–antigen-presenting cell junction (Dustin
with complement activation by inhibiting the membrane attack et al., 1998).
complex (MAC) (Jones and Jomary, 2002). As Aβ aggregation is an Cholesterol metabolism had also been implicated in AD prior
activator of the complement system, alterations in CLU structure to the recent GWAS findings. Epidemiological studies have shown
and/or expression may alter the rate of Aβ clearance via activation that high levels of cholesterol during middle age are correlated with
of the complement system. the subsequent development of dementia. Furthermore, statins,
CR1 is a polymorphic protein that has numerous different func- which lower cholesterol levels, may have a protective effect against
tions within the immune system. It is a member of the regulators of the onset of dementia (Duron and Hanon, 2008). It has also been
complement activation (RCA) family and acts as a negative regula- shown that dysregulation of cholesterol homeostasis can contribute
tor inhibiting both the classical and alternative pathways. On eryth- to neurodegeneration (Dietschy and Turley, 2001; Puglielli et al.,
rocytes CR1 is a vehicle for the clearance of C3b-coated immune 2003). APOE is the main cholesterol transporter within the brain
complexes, as it is involved in immune adherence and phagocy- (Beffert et al., 1998) and it has been shown that the ε4 isoform,
tosis (Birmingham et al., 2003). The longer forms of CR1 have an which is less stable than the neutral APOE ε3 isoform (Morrow
additional C3b binding site and thus clear C3b-coated immune et al., 2002), does not deliver cholesterol to neurons as effectively
complexes from the system at a faster rate than the shorter forms (Gong et al., 2002). CLU is the second major apolipoprotein within
of CR1, thus acting to dampen complement cascade activity. As the brain. It is a binding partner of APOE (May and Finch, 1992)
complement cascade activity is thought to heighten AD pathology, and parallels many of APOE’s properties within lipid metabolism
the longer forms of CR1 should confer relative protection against and trafficking. It is present in lipid particles (Stuart et al., 1992) and
the development of AD. However, recent evidence suggests that is involved in the transport of both cholesterol and phospholipids
the longer forms of CR1 are a risk factor for AD (Brouwers et al., (Calero et al., 1999). ABCA7, part of the ABC transporter super-
2011). The authors hypothesize that in this instance the role of family (Kaminski et al., 2000), also plays a role in lipid metabolism.
complement is actually neuroprotective, suggesting that the longer This family are transmembrane proteins that aid in the transport of
forms of CR1 excessively decrease C3b opsonization and clear- substrates across the cell membrane (Kim et al., 2008). ABCA7 is
ance of Aβ. Therefore, inhibition of the complement system by the expressed throughout the brain (Kim et al., 2005) and is the clos-
longer forms of CR1 actually confers disease risk. This theory is est homologue to ABCA1 (Kaminski et al., 2000). ABCA1 plays
supported by AD mouse models where inhibition of complement a critical role in peripheral lipid transport and regulates choles-
activation has been shown to correlate with enhanced Aβ plaque terol efflux from the plasma membrane to the apolipoproteins A–I.
deposition and loss of neuronal integrity (Wyss-Coray et al., 2002; ABCA7 itself is known to be involved in the release of cholesterol
Maier et al., 2008). and phopholipids from cells to lipoprotein particles (Abe-Dohmae
et al., 2004) and through this role may interact with both APOE lesions. VaD reflects a heterogeneous grouping of disorders (with
and CLU. subtypes including stroke-related dementia, subcortical VaD, mixed
Another mechanism implicated by the GWAS findings is AD and VaD), which presents a considerable obstacle to genetic
receptor-mediated endocytosis. Previous research has shown that investigation, as reflected in the limited number of such studies
one of the earliest changes detected in Alzheimer’s pathology is compared with AD. However, as with AD, advances have been
within the clathrin-mediated endocytic pathway (Cataldo et al., made in the understanding of Mendelian subtypes of the disorder.
1996). Two AD risk genes, PICALM and BIN1, have a direct role
in clathrin-mediated endocytosis (CME). PICALM is involved in Mendelian forms of vascular dementia
the initial assembly of the clathrin-coated vesicle, whereas BIN1 Cerebral arteriopathy with subcortical infarcts and
binds lipid membranes and detects and induces membrane cur- leucoencephalopathy
vature (Dawson et al., 2006). Two other AD susceptibility genes Cerebral autosomal dominant arteriopathy with subcortical inf-
have also been linked to endocytosis. Most members of the Siglec arcts and leucoencephalopathy (CADASIL) is the most common
family, which includes CD33, act as endocytic receptors, mediating form of hereditary recurrent stroke. It manifests as subcortical
endocytosis through a mechanism independent of clathrin (Tateno small vessel disease accompanied by recurrent transient ischaemic
et al., 2007). CD2AP is a scaffold adaptor protein (Dustin et al., attacks (TIAs), lacunar strokes, migraine, neuropsychiatric compli-
1998) that associates with cortactin, a protein also involved in the cations, and dementia (Tournier-Lasserve et al., 1993). CADASIL
regulation of receptor-mediated endocytosis (Lynch et al., 2003). can affect individuals of both sexes from their twenties upwards,
There are several hypotheses as to how defects in endocytosis and with a mean age at onset of first stroke of 46 years. Individuals suf-
intracellular trafficking could cause AD. APP is a cell-surface pro- fering from CADASIL display prominent signal abnormalities on
tein that is rapidly internalized via clathrin-mediated endocytosis brain MRI, including white matter abnormalities and small subcor-
(Nordstedt et al., 1993; Koo and Squazzo, 1994). The generation of tical infarcts, which can often be observed prior to the first stroke.
Aβ is initiated by β-secretase cleavage of APP and evidence sug- In 1993, genetic linkage analysis of two unrelated families mapped
gests that the amyloidogenic processing of APP by β-secretase takes the CADASIL locus to chromosome 19q12 (Tournier-Lasserve et
place in endosomes (Nordstedt et al., 1993; Vassar et al., 1999; Lah al., 1993). Subsequent studies refined the region of interest to a 2cM
and Levey, 2000). Therefore the production of Aβ is dependent on interval on chromosome 19p13.1 (Sabbadini et al., 1995; Ducros
the endocytosis (and recycling) of APP from the cell surface and its et al., 1996). Mutations in the NOTCH3 gene located within the
transit to the endosomes. Neurotransmitter release, crucial in neu- linkage region on chromosome 19 were later identified that seg-
ron function, is also reliant on CME (Smith et al., 2008). PICALM regated with the disease within families (Joutel et al., 1996, 1997).
is known to regulate the trafficking of VAMP2, a SNARE protein NOTCH3 is normally expressed in vascular smooth muscle cells
that has a prominent role in the fusion of synaptic vesicles to the and pericytes, including those of the cerebral vasculature (Joutel
presynaptic membrane in neurotransmitter release (Harel et al., et al., 2000; Prakash et al., 2002). To date, more than 180 patho-
2008). AD brains show a reduced number of synapses compared genic missense mutations, six deletions, an insertion, a frameshift,
to elderly controls, and stereological and biochemical analysis has and two duplications have been identified (Yamamoto et al., 2011),
shown that this reduction in synaptic density correlates better with and although it has been speculated that these mutations may affect
cognitive decline than with the accumulation of amyloid plaques receptor trafficking, processing, ligand binding, or signal trans-
(Masliah et al., 2001). There is also evidence that synapses within duction, the exact mechanisms underlying the pathogenesis of
the brains of those with AD may be dysfunctional even before they CADASIL are unclear.
visibly degenerate (Fitzjohn et al., 2001). An autosomal recessive disorder similar to CADASIL has also
Thus, the genetic risk factors for AD identified to date are not been described, and has been designated CARASIL (cerebral auto-
random, but instead point to defects in specific biological proc- somal recessive arteriopathy with subcortical infarcts and leucoen-
esses and pathways that contribute to the development of the dis- cephalopathy). The disease is characterized by nonhypertensive
ease. These findings have refined previous ideas and defined new leucoencephalopathy associated with alopecia and spondylosis,
putative disease mechanisms, providing new impetus for focused and, as in CADASIL, the recurrent strokes lead to progressive
studies aimed at understanding AD pathogenesis. The 10 identi- cognitive impairment, ultimately resulting in dementia in most
fied genes influencing LOAD account for approximately one-third individuals. CARASIL is very rare, with approximately 50 cases
of the genetic variance in AD and so a substantial proportion of described in Japan, two in China, and one in Spain (in a Caucasian
heritability remains unaccounted for. As such, the current genetic individual). Through genome-wide linkage analysis and fine map-
findings cannot be used for predictive purposes. However, fur- ping, Hara and colleagues (2009) identified homozygous mutations
ther research using more powerful genome-wide association and in the HTRA1 gene cosegregating with disease. The HTRA1 gene
next-generation sequencing approaches is likely to define more of on chromosome 10q25 encodes a serine protease that is known
the genetic architecture of AD. to influence different processes, including transforming growth
factor-β (TGF-β) signalling. Hara et al. demonstrated that mutated
forms of the HTRA1 protein exhibit decreased protease activity
Vascular Dementia and fail to repress TGF-β signalling; thus, elevated TGF-β signal-
Vascular dementia (VaD) is the second most common type of ling appears to result in CARASIL.
dementia, accounting for approximately 20% of all cases, with an
estimated 20,000 people a year developing VaD in the UK alone Hereditary cerebral amyloid angiopathy
(Heyman et al., 1998). The disease is characterized by a progressive Cerebral amyloid angiopathy (CAA) is a disorder characterized
deterioration in cognitive function, resulting from cerebrovascular by amyloid deposition in the walls of arteries, arterioles, and, less
frequently, veins of the cerebral cortex and the leptomeninges. CAA by Sudlow and colleagues (2006), incorporating data from 31 stud-
occurs most often as a sporadic condition in older people (and is ies including data from 5961 stroke sufferers and 17,965 healthy
highly prevalent in AD brains), but rare hereditary forms have controls, observed modest associations between the APOE ε4
been identified that are typically more severe and earlier in onset. allele and ischaemic stroke and subarachnoid haemorrhage, but
Hereditary cerebral haemorrhage with amyloidosis (HCHWA) not with intracerebral haemorrhage, which was associated with ε2
is characterized primarily by haemorrhagic strokes and demen- genotypes.
tia (Bornebroek et al., 1996; Maat-Schieman et al., 1996), which Others have sought to directly investigate the association between
are often accompanied by migraines and psychiatric problems. A APOE and VaD. Probably the best evidence has come from large
number of subtypes of HCHWA have been identified, with most population-based studies. For example, a study of the Framingham
resulting from mutations in the APP gene (Dutch, Italian, Flemish, cohort, involving over 1000 individuals aged between 70 and 100,
Piedmont, Arctic types), and one resulting from a mutation in the found that the APOE ε4 allele was associated with both dementia
CST3 gene (Icelandic type). As discussed previously, distinct muta- following stroke and multi-infarct dementia (Myers et al., 1996).
tions in APP have also been shown to result in EOAD, highlight- Likewise, Slooter and colleagues (1997) reported that VaD and
ing the overlap that exists between AD and VaD. CST3 encodes mixed VaD and AD were both increased among APOE ε4 carriers.
the cystatin C protein, which is a member of the type II family of Given the well-established and strong association between APOE
cysteine protease inhibitors. CST3-CAA results from a L68Q muta- and AD (Farrer et al., 1997), investigations of VaD are likely to
tion (Abrahamson et al., 1987; Palsdottir et al., 1988), and this leads be complicated by issues of diagnosis, as dementia sufferers often
to the deposition of an N-terminal degradation product of the exhibit mixed AD and vascular pathology (Blacker and Lovestone,
mutated cystatin C protein as vascular amyloid in the leptomenin- 2006). Indeed, the Gothenburg longitudinal study found that the
ges, cerebral cortex, basal ganglia, brainstem, and cerebellum. Two APOE ε4 allele was associated with an increased risk of AD and
mutations of the ITM2B gene (also known as BRI2) also result in mixed dementia, but not pure VaD (Skoog et al., 1998). This issue
forms of hereditary CAA, namely familial British dementia (FBD) can be addressed by studies performed in samples of dementia
and familial Danish dementia (FDD). ITM2B encodes a transmem- sufferers whose diagnosis has been confirmed at post mortem.
brane protein which is processed at the C-terminus to produce a Polvikoski and colleagues (2001) established the prevalence of neu-
secreted peptide that inhibits the deposition of Aβ. FBD results ropathologically defined AD and other dementias, together with
from a point mutation of the normal stop codon of the ITM2B gene APOE genotype, in 88% of the 85 years old and older residents of
(Vidal et al., 1999), whereas a decamer duplication between codons Vantaa, Finland. They noted a strong relationship between APOE
265 and 266 is associated with FDD (Vidal et al., 2000). Both muta- and both clinically diagnosed VaD and AD. However, neuropatho-
tions abolish the normal stop codon, resulting in an extended logical analyses showed that over half of those with clinically diag-
precursor protein. nosed VaD demonstrated AD pathology. This study suggests that
the association between APOE ε4 and VaD probably results from
Sporadic forms of vascular dementia an association with underlying and undiagnosed AD. Indeed, other
Sporadic and common forms of VaD are thought to result from a autopsy studies of VaD have found either no association with VaD
combination of genetic and environmental influences. A number and APOE or an association only in cases of mixed VaD and AD
of twin studies have sought to estimate the heritability of VaD. (Saunders et al., 1993; Betard et al., 1994).
In a study of Finnish twins using hospital discharge records, the Recently, a GWAS of VaD has been performed by Schrijvers and
probandwise concordance rates for VaD were 31% in MZ versus colleagues (2012). Study participants were from the Rotterdam
12.5% in DZ twins (Raiha et al., 1996), supporting a genetic com- Study, a large prospective population-based cohort in the
ponent. In contrast, the Norwegian register-based study of elderly Netherlands. They identified genome-wide significant association
twins, in which participants were interviewed and clinically exam- with a SNP close to the androgen receptor gene (AR) on the X
ined, found identical probandwise concordance rates of 29% in MZ chromosome in their discovery sample of 67 VaD cases and 5633
and DZ twins (Bergem et al., 1997). Despite being the smaller of controls (P = 1.3 × 10–8), which replicated in two independent sam-
the two studies, the study by Bergem and colleagues is probably ples (combined sample of 256 cases and 5839 controls; combined
the most methodologically sound and suggests that environmental P = 0.024). However, as the authors noted, the study included small
influences dominate in VaD. numbers of VaD cases, and additional replication will be necessary
Genetic studies of sporadic VaD have been limited and have before this can be considered an established association.
mostly taken the form of candidate gene studies. While a number of
associations have been reported (see Kim et al. (2011) for a review),
these are based on small sample sizes (typically less than 250 cases) Frontotemporal Dementia
and cannot be considered robust associations by current standards. Frontotemporal dementia (FTD) accounts for 5–10% of all cases
However, given its strong association with AD, the potential role of of dementia (and is the second most common early-onset demen-
APOE in VaD is perhaps worthy of mention. APOE has a complex tia) and results from degeneration of the frontal and temporal
relationship with cardiovascular disease and VaD. APOE genotype lobes, often in conjunction with the degeneration of subcortical
has been associated with a number of factors linked to VaD, includ- brain regions. This manifests clinically as progressive behavioural
ing cholesterol metabolism (Eichner et al., 2002), hypertension changes and frontal executive deficits and/or selective language
(Hirono et al., 2000), intracerebral haemorrhage (Rosand et al., difficulties. FTD is a pathologically heterogeneous disease, and
2000; Woo et al., 2002), ischaemic heart disease (Song et al., 2004), can be subdivided into three types based on the presence of pro-
and cerebral amyloid angiopathy (McCarron et al., 2000). APOE has tein inclusions in the brain: those with tau(τ)-positive inclusions;
been hypothesized to influence the risk of stroke. A meta-analysis those with ubiquitin-positive, τ-negative inclusions; and those
lacking distinctive histopathology (Mackenzie et al., 2006b). Most the maintenance of microtubule structure within the cell. If either
cases of FTD are sporadic; however, studies suggest that 25–50% of the three- or four- repeat isoforms fail to function, or if the ratio
of disease sufferers have a first-degree relative with FTD (Poorkaj of the two alters within the cell, microtubule formation and stabil-
et al., 2001; Rosso et al., 2003), with a substantial proportion of ity become compromised. Additionally, unused τ of either form
these patients reporting a family history of disease consistent with can be bundled into a tangle, which can impair cell function (Lee
autosomal dominant inheritance. Around 30% of familial FTD et al., 2001; Neary et al., 2005). MAPT mutations can be largely
cases are caused by mutations in the gene encoding τ (MAPT) and grouped according to their position in the gene, which defines
are characterized by τ-pathology (Morris et al., 2001; Rosso et al., their effect on MAPT mRNA and protein, which in turn influences
2003). However, a significant proportion of hereditary FTD does the type of resultant pathology. The majority of MAPT mutations
not result from mutations in MAPT (Tolnay and Probst, 2002), and occur within the C-terminal region of the gene and most within
further susceptibility loci have been identified. or adjacent to the microtubule-binding domains between exons 9
and 12, or close to exon 13. Intronic mutations, close to exon 10,
Frontotemporal dementia and parkinsonism linked to are also common and act to increase splicing of exon 10 (Hutton
chromosome 17 et al.; 1998, Poorkaj et al., 1998). Others have reported muta-
Frontotemporal dementia and parkinsonism linked to chromosome tions in exon 1 of the MAPT gene (Hayashi et al., 2002; Poorkaj
17 (FTDP-17) is an autosomal dominantly inherited disease, which et al., 2002). Intronic mutations and some mutations that affect
probably accounts for less than 5% of all dementias. FTDP-17 can splicing regulatory elements manifest at the level of mRNA splic-
largely be categorized into two major types: predominant-dementia ing, which leads to altered expressions of τ isoforms (Clark et al.,
and predominant-parkinsonism (van Swieten and Spillantini, 2007). 1998; Hutton et al., 1998; Spillantini et al., 1998b; van Swieten et
However, both dementia- and parkinsonism-predominant FTD al., 2007). Generally, MAPT mutations affect either τ microtubule
can occur in families carrying the same mutations (Yasuda et al., interactions and/or fibril formation or exon 10 splicing. However,
1999). The dementia phenotype is commonly characterized by cog- some mutations have multiple effects.
nitive difficulties, including problems with memory and language, The clinical presentation of FTDP-17 varies, to some extent,
and marked personality changes, including disinhibition and apa- according to the type and location of the MAPT mutation. For
thy (Neary et al., 2005). Those with the parkinsonism-predominant example, some mutations lead to very similar age of onset both
subtype usually experience gait impairment, rigidity, bradykinesia, within and between families, even among families from different
and resting tremor (Arima et al., 2000). In contrast to AD, those continents (Delisle et al., 1999; Arima et al., 2000; Kodama et al.,
with FTDP-17 have relatively intact episodic memory and have 2000). Some mutations are associated with disease onset in mid to
fewer difficulties with orientation. Rather, they present with deficits late adulthood (e.g. P301L (van Swieten et al., 1999)), whereas oth-
in verbal fluency, abstract thinking, attention, and executive func- ers can lead to an onset of clinical symptoms between 20 and 30
tion (see Hodges and Miller (2001) for a review of the neuropsy- years of age (e.g. P301S (Bugiani et al., 1999, Sperfeld et al., 1999),
chology of FTD). L315R (van Herpen et al., 2003), and G335S (Sperfeld et al., 1999)).
Spillantini and colleagues (1998a) first reported linkage to Disease onset after the age of 70 years is less common but is some-
chromosome 17q21–22 in families with FTD and parkinsonism. times observed in those with R5H (Hayashi et al., 2002) and 1260V
Further analysis of this region revealed mutations in MAPT as the (Grover et al., 2003) mutations.
cause of the symptoms of FTD (Clark et al., 1998; Goedert et al., Despite the robust association between MAPT mutations, τ
1999, Hutton et al., 1998; Poorkaj et al., 1998). To date, 44 MAPT pathology, and FTDP-17, mutations in MAPT do not appear to
disease mutations have been reported in 132 families (see <www. be a common cause of general dementia. Studies have generally
molgen.ua.ac.be/FTDMutations/>). FTDP-17 is now used as a col- reported no, or weak, association with AD (Roks et al., 1999; Myers
lective term for primarily autosomal dominant conditions linked et al., 2005; Mukherjee et al., 2007; Gerrish et al., 2012), Parkinson’s
to chromosome 17. Historically, several of these conditions have disease (Zhang et al., 2005; Zabetian et al., 2007), or sporadic FTD
been referred to as different nosological entities, e.g. Pick’s disease, (Verpillat et al., 2002; Bernardi et al., 2006). These findings are par-
familial subcorticol gliosis, and autosomal dominant dementia with ticularly interesting as a number of other dementias, including AD,
widespread NFTs (Foster et al., 1997; Spillantini et al., 1997, 1998a; are associated with extensive τ pathology (Lee et al., 2001).
Lee et al., 2001). FTDP-17 shows a primarily dominant mode of
inheritance, although recessive forms have also been described Non-MAPT frontotemporal dementia
(Rademakers et al., 2002; Nicholl et al., 2003). MAPT mutations A family history of similar neurodegenerative disease may be
are found in between 10% and 30% of FTD patients with a positive present in up to 50% of individuals with FTD; however, a substan-
family history of the disease and up to 70% of cases from families tial proportion of hereditary FTD does not result from mutations
exhibiting an autosomal dominant mode of disease transmission in MAPT (Tolnay and Probst, 2002). A number of loci and muta-
(Rizzu et al., 1999; Poorkaj et al., 2001; Bird et al., 2003). tions in genes other than MAPT have been shown to be associated
τ is involved in microtubule assembly and stabilization. In normal with FTD. The disease often appears with pathological ubiquitin
nerve cells there are six τ isoforms, which are produced by alternate inclusions, in the absence of τ pathology (FTD-TDP) (Forman
mRNA splicing of exons 9, 10, 11, and 12 (Goedert et al., 1989). et al., 2006). Two independent studies reported linkage to chromo-
They differ first in terms of the inclusion of certain N-terminal some 17q21–22, which was not attributable to known mutations
domains and, second, in the inclusion of three or four-repeat in the MAPT gene (Rosso et al., 2001; Rademakers et al., 2002).
domains within the C-terminal region of the molecule (Lee et al., Mackenzie et al. (2006a) later identified several mutations in the
2001). The repeat domains are important because they act as the progranulin gene (GRN), located 1.7 Mb centromeric of MAPT
microtubule-binding domains of τ and thus play a precise role in on chromosome 17q21.3, that segregated with FTD-TDP in eight
families. Cruts et al. (2006) have also found mutations in GRN in For example, nine genes influencing AD have been identified in
a Belgian family with autosomal dominantly inherited FTD-TDP. the past 3 years through GWAS, the first AD risk genes since 1993.
Furthermore, they reported that GRN mutations were over three These findings have refined previous ideas and defined new puta-
times more frequent than mutations in MAPT, emphasizing a major tive disease mechanisms, providing impetus for focused studies
role in FTD. Further autosomal dominant loci have been identified, aimed at understanding AD pathogenesis.
including CHMPB2 (Skibinski et al., 2005; Parkinson et al., 2006), In addition to providing new avenues for exploration, the recent
FUS (Kwiatkowski et al., 2009), and VCP (Watts et al., 2004;). genetic advances highlight the potential for gene discovery when
Even those with known mutations show wide variation in terms larger samples are analysed, using ever more sophisticated methods.
of age at disease onset and other clinical characteristics, which is To this end, the GERAD, EADI, ADGC, and CHARGE consor-
likely to be under genetic and environmental control. Wide clini- tia have begun the International Genomics of Alzheimer’s Project
cal and pathological differences between FTD sufferers have caused (IGAP). IGAP aims to combine GWAS data from 17,008 AD cases
some concerns about applying GWAS approaches, which assume and 37,646 controls and follow up the most interesting results in
aetiological homogeneity. Despite this, Van Deerlin and colleagues an independent sample of 8,572 AD cases and 11.312 controls. In
(2010) reported the first GWAS study of FTD in March 2010. They addition, we and others are seeking to identify new risk loci by
formed an international collaboration to compare a sample of 515 analysing refined disease phenotypes characterized by behavioural
individuals with FTD-TDP with 2509 control individuals and iden- symptoms that show evidence of heritability in AD, such as psycho-
tified association with several SNPs mapping to a single LD block on sis (Hollingworth et al., 2011b) or depression. GWAS data also offer
chromosome 7p21, which contains the gene TMEM106B (most sig- exciting opportunities to examine genetic loci contributing to multi-
nificant SNP rs1990622, P = 1.08 × 10–11, OR = 0.61). This has since ple phenotypes, e.g. we are currently comparing our AD GWAS data
replicated in independent samples comprising those with FTD-TDP with that of other neurological disorders such as Parkinson’s disease
(van der Zee and Van Broeckhoven, 2011) and those diagnosed clin- and ALS. We are also investigating whether there is a shared genetic
ically with FTD (van der Zee et al., 2011). The association is particu- basis for AD and cardiovascular disease, which may also be relevant
larly strong in GRN mutation carriers, and association between the for vascular dementia. It is clear that the coming years will be an
risk allele of rs1990622 and lower plasma granulin protein levels has exciting time for dementia genetics as we move closer and closer to
been also been identified (Finch et al., 2011; Cruchaga et al., 2011). completely unravelling the genetic architecture of these diseases.
A number of interesting subthreshold hits were reported in the
GWAS and have subsequently been the focus of a study by Rollinson References
and colleagues (2011). Although none of these loci replicated in a Abe-Dohmae, S., et al . (2004 ) . Human ABCA7 supports
clinical cohort of 470 patients, convincing evidence for association apolipoprotein-mediated release of cellular cholesterol and phospholipid
was found on chromosome 9 in a subgroup of 84 patients with FTD to generate high density lipoprotein. Journal of Biological Chemistry, 279,
and amyotrophic lateral sclerosis (ALS). In fact, studies dating back 604–11.
to 2006 have identified linkage to chromosome 9p21 in families Abrahamson, M., et al. (1987). Molecular cloning and sequence analysis
with FTD and/or ALS (Morita et al., 2006; Vance et al., 2006; Luty of cDNA coding for the precursor of the human cysteine proteinase
et al., 2008), although the causative mutation had proven difficult inhibitor cystatin C. FEBS Letters, 216, 229–33.
to pinpoint. However, using next-generation sequencing strategies, Alzheimer’s Disease International (2009). World Alzheimer Report 2009.
<http://www.alz.co.uk/research/files/World Alzheimer Report.pdf>.
two groups recently identified a GGGGCC hexanucleotide expan-
Alzheimer’s Disease International (2010). World Alzheimer Report 2010.
sion in an intron of the C90RF72 gene (DeJesus-Hernandez et al., <http://www.alz.co.uk/research/files/World Alzheimer Report2010.pdf>.
2011; Renton et al., 2011). Little is known about the function of the Alzheimer’s Research UK (2010). Dementia 2010. <http://www.dementia2010.
C90RF72 protein, but the hexanucleotide expansion leads to the loss org/reports/Dementia2010Full.pdf>.
of an alternatively spliced transcript and the formation of nuclear Arima, K., et al. (2000). Two brothers with frontotemporal dementia and
RNA foci, which may be toxic. Boeve and colleagues (2012) recently parkinsonism with an N279K mutation of the tau gene. Neurology, 54,
reported the results of screening cohorts of patients from the Mayo 1787–95.
Clinic with FTD and/or ALS. The frequency of the expansion among Avramopoulos, D. (2009). Genetics of Alzheimer’s disease: recent advances.
all screened cases was highest in those with FTD and ALS (21.6%), Genome Medicine, 1, 34.
and when restricting to cases with a positive family history, the Bayer, T. A., et al. (1999). It all sticks together—the APP-related family of
proteins and Alzheimer’s disease. Molecular Psychiatry, 4, 524–8.
frequency rose to 47.6%. The expansion was also detected in FTD
Beffert, U., et al. (1998). The neurobiology of apolipoproteins and their
only cases (14.7% of cases with a positive family history and 3.7% of receptors in the CNS and Alzheimer’s disease. Brain Research and Brain
sporadic cases) and in ALS only cases (24% of cases with a positive Research Review, 27, 119–42.
family history and 4.1% of sporadic cases). The C90RF72 expansion Bell, R. D., et al. (2007). Transport pathways for clearance of human
is therefore a major cause of FTD/ALS. Alzheimer’s amyloid beta-peptide and apolipoproteins E and J in the
mouse central nervous system. Journal of Cerebral Blood Flow and
Metabolism, 27, 909–18.
Conclusion Bennett, B. D., et al. (2000). A furin-like convertase mediates propeptide
cleavage of BACE, the Alzheimer’s beta—secretase. Journal of Biological
Identifying genes that contribute to disease risk is focusing
Chemistry, 275, 37712–7.
research on biological pathways affecting the development of dif-
Bergem, A. L., Engedal, K., and Kringlen, E. (1997). The role of heredity
ferent dementias. Until recently, most success had been achieved in late-onset alzheimer disease and vascular dementia. A twin study.
in identifying rare mutations causing Mendelian forms of demen- Archives of General Psychiatry, 54, 264–70.
tia. However, the advent of genome-wide association studies has Bernardi, L., et al. (2006). The effects of APOE and tau gene variability on risk
enabled the identification of a number of novel susceptibility loci. of frontotemporal dementia. Neurobiological Aging, 27, 702–9.
Bertram, L., et al. (2007). Systematic meta-analyses of Alzheimer disease genetic Citron, M., et al. (1992). Mutation of the beta-amyloid precursor protein in
association studies: the AlzGene database. Nature Genetics, 39, 17–23. familial Alzheimer’s disease increases beta-protein production. Nature,
Bertrand, P., et al. (1995). Association of apolipoprotein E genotype with brain 360, 672–4.
levels of apolipoprotein E and apolipoprotein J (clusterin) in Alzheimer Clark, C. M., Trojanowski, J. Q., and Lee, M.-Y. L. (1997). Neurofibrillary
disease. Brain Research and Molecular Brain Research, 33, 174–8. tangles in Alzheimer’s disease: clinical and pathological implications.
Betard, C., et al. (1994). Apo E allele frequencies in Alzheimer’s disease, Lewy In: Brioni, J. D. and Decker, M. W. (eds) Pharmacological treatment of
body dementia, Alzheimer’s disease with cerebrovascular disease and Alzheimer’s disease: molecular and neurobiological foundations. New
vascular dementia. Neuroreport, 5, 1893–6. York, Wiley-Liss.
Bird, T., et al. (2003). Epidemiology and genetics of frontotemporal dementia/ Clark, L. N., et al. (1998). Pathogenic implications of mutations in the tau
Pick’s disease. Annals of Neurology, 54 Suppl 5, S29–31. gene in pallido-ponto-nigral degeneration and related neurodegenerative
Birmingham, D. J., et al. (2003). A CR1 polymorphism associated with disorders linked to chromosome 17. Proceedings of the National Academy
constitutive erythrocyte CR1 levels affects binding to C4b but not C3b. of Science USA, 95, 13103–7.
Immunology, 108, 531–8. Colciaghi, F., et al. (2002). [alpha]-Secretase ADAM10 as well as [alpha]APPs
Blacker, D. and Lovestone, S. (2006). Genetics and dementia nosology. is reduced in platelets and CSF of Alzheimer disease patients. Molecular
Journal of Geriatric Psychiatry and Neurology, 19, 186–91. Medicine, 8, 67–74.
Boeve, B. F., et al. (2012). Characterization of frontotemporal dementia and/ Crocker, P. R., Paulson, J. C., and Varki, A. (2007). Siglecs and their roles in
or amyotrophic lateral sclerosis associated with the GGGGCC repeat the immune system. Nature Review of Immunology, 7, 255–66.
expansion in C9ORF72. Brain, 135, 765–83. Cruchaga, C., et al. (2011). Association of TMEM106B gene polymorphism
Borchelt, D. R., et al. (1996). Familial Alzheimer’s disease-linked presenilin with age at onset in granulin mutation carriers and plasma granulin
1 variants elevate Abeta1-42/1-40 ratio in vitro and in vivo. Neuron, 17, protein levels. Archives of Neurology, 68, 581–6.
1005–13. Cruts, M., et al. (2006). Null mutations in progranulin cause ubiquitin-positive
Bornebroek, M., et al. (1996). Hereditary cerebral hemorrhage with frontotemporal dementia linked to chromosome 17q21. Nature, 442,
amyloidosis-Dutch type (HCHWA-D): I—a review of clinical, radiologic 920–4.
and genetic aspects. Brain Pathology, 6, 111–4. Cummings, J. L. (2003). Alzheimer’s disease: from molecular biology to
Brouwers, N., et al. (2011). Alzheimer risk associated with a copy number neuropsychiatry. Seminars in Clinical Neuropsychiatry, 8, 31–6.
variation in the complement receptor 1 increasing C3b/C4b binding Dawson, J. C., Legg, J. A., and Machesky, L. M. (2006). Bar domain proteins:
sites. Molecular Psychiatry, 17, 223–33. a role in tubulation, scission and actin assembly in clathrin-mediated
Bugiani, O., et al. (1999). Frontotemporal dementia and corticobasal endocytosis. Trends in Cell Biology, 16, 493–8.
degeneration in a family with a P301S mutation in tau. Journal of Dejesus-Hernandez, M., et al. (2011). Expanded GGGGCC hexanucleotide
Neuropathology and Experimental Neurology, 58, 667–77. repeat in noncoding region of C9ORF72 causes chromosome 9p-linked
Buxbaum, J. D., et al. (1998). Evidence that tumor necrosis factor alpha FTD and ALS. Neuron, 72, 245–56.
converting enzyme is involved in regulated alpha-secretase cleavage of Delisle, M. B., et al. (1999). A mutation at codon 279 (N279K) in exon 10 of
the Alzheimer amyloid protein precursor. Journal of Biological Chemistry, the Tau gene causes a tauopathy with dementia and supranuclear palsy.
273, 27765–7. Acta Neuropathologica, 98, 62–77.
Cai, H., et al. (2001). BACE1 is the major beta-secretase for generation of Depaepe, V., et al. (2005). Ephrin signalling controls brain size by regulating
Abeta peptides by neurons. Nature and Neuroscience, 4, 233–4. apoptosis of neural progenitors. Nature, 435, 1244–50.
Calero, M., et al. (1999). Functional and structural properties of De Strooper, B. (2003). Aph-1, Pen-2, and Nicastrin with Presenilin generate
lipid-associated apolipoprotein J (clusterin). Biochemistry Journal, 344 an active gamma-Secretase complex. Neuron, 38, 9–12.
Pt 2, 375–83. De Strooper, B., et al. (1998). Deficiency of presenilin-1 inhibits the normal
Calhoun, M. E., et al. (1998). Neuron loss in APP transgenic mice. Nature, cleavage of amyloid precursor protein. Nature, 391, 387–90.
395, 755–6. De Strooper, B., et al. (1999). A presenilin-1-dependent gamma-secretase-like
Campion, D., et al. (1999). Early-onset autosomal dominant Alzheimer protease mediates release of Notch intracellular domain. Nature, 398,
disease: prevalence, genetic heterogeneity, and mutation spectrum. 518–22.
American Journal of Human Genetics, 65, 664–70. Dietschy, J. M. and Turley, S. D. (2001). Cholesterol metabolism in the brain.
Capell, A., et al. (1998). The proteolytic fragments of the Alzheimer’s Current Opinions in Lipidology, 12, 105–12.
disease-associated presenilin-1 form heterodimers and occur as a 100– Ducros, A., et al. (1996). Cerebral autosomal dominant arteriopathy with
150-kDa molecular mass complex. Journal of Biology and Chemistry, subcortical infarcts and leukoencephalopathy, genetic homogeneity, and
273, 3205–11. mapping of the locus within a 2-cM interval. American Journal of Human
Carrasquillo, M. M., et al. (2009). Genetic variation in PCDH11X is associated Genetics, 58, 171–81.
with susceptibility to late-onset Alzheimer’s disease. Nat ure Genetics, 41, Duron, E. and Hanon, O. (2008). Vascular risk factors, cognitive decline, and
192–8. dementia. Vascular Health Risk Management, 4, 363–81.
Carrasquillo, M.M., et al. (2010). Replication of CLU, CR1, and PICALM Dustin, M. L., et al. (1998). A novel adaptor protein orchestrates receptor
associations with Alzheimer disease. Archives of Neurology, 67, 961–4. patterning and cytoskeletal polarity in T-cell contacts. Cell, 94, 667–77.
Cataldo, A. M., et al. (1996). Properties of the endosomal-lysosomal system Edbauer, D., et al. (2003). Reconstitution of gamma-secretase activity. Nature
in the human central nervous system: disturbances mark most neurons Cell Biology, 5, 486–8.
in populations at risk to degenerate in Alzheimer’s disease. Journal of Ehehalt, R., et al. (2003). Amyloidogenic processing of the Alzheimer
Neuroscience, 16, 186–99. beta-amyloid precursor protein depends on lipid rafts. Journal of Cell
Chang, M. Y., et al. (2007). Bin1 ablation increases susceptibility to cancer Biology, 160, 113–23.
during aging, particularly lung cancer. Cancer Research, 67, 7605–12. Eichner, J. E., et al. (2002). Apolipoprotein E polymorphism and cardiovascular
Chan, S. L., et al. (2008). ATP-binding cassette transporter A7 regulates disease: a HuGE review. American Journal of Epidemiology, 155, 487–95.
processing of amyloid precursor protein in vitro. Journal of Esler, W. P., et al. (2000). Transition-state analogue inhibitors of
Neurochemistry, 106, 793–804. gamma-secretase bind directly to presenilin-1. Nature Cell Biology, 2,
Chishti, M. A., et al. (2001). Early-onset amyloid deposition and cognitive 428–34.
deficits in transgenic mice expressing a double mutant form of amyloid Farrer, L. A., et al. (1997). Effects of age, sex, and ethnicity on the association
precursor protein 695. Journal of Biological Chemistry, 276, 21562–70. between apolipoprotein E genotype and Alzheimer disease. A
meta-analysis. APOE and Alzheimer Disease Meta Analysis Consortium. Hindorff, L. A., et al. (2012). A catalog of published genome-wide association
Journal of the American Medical Association, 278, 1349–56. studies. <www.genome.gov/gwastudies> (accessed 05.03.2012).
Farzan, M., et al. (2000). BACE2, a beta—secretase homolog, cleaves at Hirono, N., et al. (2000). Effect of the apolipoprotein E epsilon 4 allele on
the beta site and within the amyloid-beta region of the amyloid-beta white matter hyperintensities in dementia. Stroke, 31, 1263–8.
precursor protein. Proceedings of the National Academy of Science USA, Hodges, J. R. and Miller, B. (2001). The neuropsychology of frontal variant
97, 9712–7. frontotemporal dementia and semantic dementia. Introduction to the
Finch, N., et al. (2011). TMEM106B regulates progranulin levels and the special topic papers: part II. Neurocase, 7, 113–21.
penetrance of FTLD in GRN mutation carriers. Neurology, 76, 467–74. Holcomb, L., et al. (1998). Accelerated Alzheimer-type phenotype in
Fitzjohn, S. M., et al. (2001). Age-related impairment of synaptic transmission transgenic mice carrying both mutant amyloid precursor protein and
but normal long-term potentiation in transgenic mice that overexpress presenilin 1 transgenes. Nature Medicine, 4, 97–100.
the human APP695SWE mutant form of amyloid precursor protein. Hollingworth, P., et al. (2011a). Common variants at ABCA7, MS4A6A/
Journal of Neuroscience, 21, 4691–8. MS4A4E, EPHA1, CD33 and CD2AP are associated with Alzheimer’s
Forman, M. S., et al. (2006). Frontotemporal dementia: clinicopathological disease. Nature Genetics, 43, 429–35.
correlations. Annals of Neurology, 59, 952–62. Hollingworth, P.J., et al. (2011b). Genome-wide association study of
Foster, N. L., et al. (1997). Frontotemporal dementia and parkinsonism linked Alzheimer’s disease with psychotic symptoms. Molecular Psychiatry,
to chromosome 17: a consensus conference. Conference Participants. 17(12), 1316–27.
Annals of Neurology, 41, 706–15. Hooper, N. M., Karran, E. H., and Turner, A. J. (1997). Membrane protein
Francis, R., et al. (2002). Aph-1 and pen-2 are required for Notch pathway secretases. Biochemistry Journal, 321 (Pt 2), 265–79.
signaling, gamma-secretase cleavage of betaAPP, and presenilin protein Hsiao, K., et al. (1996). Correlative memory deficits, Abeta elevation, and
accumulation. Development Cell, 3, 85–97. amyloid plaques in transgenic mice. Science, 274, 99–102.
Games, D., et al. (1995). Alzheimer-type neuropathology in transgenic mice Hu, X., et al. (2011). Meta-analysis for genome-wide association study
overexpressing V717F beta-amyloid precursor protein. Nature, 373, identifies multiple variants at the BIN1 locus associated with late-onset
523–7. Alzheimer’s disease. PLoS One, 6, e16616.
Gatz, M., et al. (1997). Heritability for Alzheimer’s disease: the study of Huse, J. T. and Doms, R. W. (2000). Closing in on the amyloid cascade:
dementia in Swedish twins. Journal of Gerontology A Biololgy Science recent insights into the cell biology of Alzheimer’s disease. Molecular
Medical Science, 52, M117–25. Neurobiology, 22, 81–98.
Gatz, M., et al. (2006). Role of genes and environments for explaining Hussain, I., et al. (1999). Identification of a novel aspartic protease (Asp 2) as
Alzheimer disease. Archives of General Psychiatry, 63, 168–74. beta-secretase. Molecular Cell Neuroscience, 14, 419–27.
Gerrish, A., et al. (2012). The role of variation at AbetaPP, PSEN1, PSEN2, Hussain, I., et al. (2000). ASP1 (BACE2) cleaves the amyloid precursor protein
and MAPT in late onset Alzheimer’s disease. Journal of Alzheimers at the beta-secretase site. Molecular Cell Neuroscience, 16, 609–19.
Disease, 28, 377–87. Hutton, M., et al. (1998). Association of missense and 5′-splice-site mutations
Goate, A., et al. (1991). Segregation of a missense mutation in the amyloid in tau with the inherited dementia FTDP-17. Nature, 393, 702–5.
precursor protein gene with familial Alzheimer’s disease. Nature, 349, Ivanov, A. I. and Romanovsky, A. A. (2006). Putative dual role of ephrin-Eph
704–6. receptor interactions in inflammation. IUBMB Life, 58, 389–94.
Goedert, M., et al. (1989). Multiple isoforms of human microtubule-associated Jarrett, J. T., Berger, E. P., and Lansbury, P. T., Jr (1993). The carboxy terminus
protein tau: sequences and localization in neurofibrillary tangles of of the beta amyloid protein is critical for the seeding of amyloid
Alzheimer’s disease. Neuron, 3, 519–26. formation: implications for the pathogenesis of Alzheimer’s disease.
Goedert, M., et al. (1999). Tau gene mutation in familial progressive Biochemistry, 32, 4693–7.
subcortical gliosis. Nature Medicine, 5, 454–7. Jehle, A. W., et al. (2006). ATP-binding cassette transporter A7 enhances
Gong, J. S., et al. (2002). Apolipoprotein E (ApoE) isoform-dependent phagocytosis of apoptotic cells and associated ERK signaling in
lipid release from astrocytes prepared from human ApoE3 and ApoE4 macrophages. Journal of Cell Biology, 174, 547–56.
knock-in mice. Journal of Biology and Chemistry, 277, 29919–26. Jones, L., et al. (2010). Genetic evidence implicates the immune system and
Gouras, G. K., et al. (1998). Generation and regulation of beta-amyloid cholesterol metabolism in the aetiology of Alzheimer’s disease. PLoS
peptide variants by neurons. Journal of Neurochemistry, 71, 1920–5. One, 5, e13950.
Grover, A., et al. (2003). A novel tau mutation in exon 9 (1260V) causes a Jones, S. E. and Jomary, C. (2002). Clusterin. International Journal of
four-repeat tauopathy. Experimental Neurology, 184, 131–40. Biochemistry and Cell Biology, 34, 427–31.
Haass, C. and Steiner, H. (2002). Alzheimer disease gamma-secretase: a Joutel, A., et al. (1996). Notch3 mutations in CADASIL, a hereditary adult-onset
complex story of GxGD-type presenilin proteases. Trends in Cell Biology, condition causing stroke and dementia. Nature, 383, 707–10.
12, 556–62. Joutel, A., et al. (1997). Strong clustering and stereotyped nature of Notch3
Hara, K., et al. (2009). Association of HTRA1 mutations and familial mutations in CADASIL patients. Lancet, 350, 1511–5.
ischemic cerebral small-vessel disease. New England Journal of Medicine, Joutel, A., et al. (2000). The ectodomain of the Notch3 receptor accumulates
360, 1729–39. within the cerebrovasculature of CADASIL patients. Journal of Clinical
Harel, A., et al. (2008). Evidence for CALM in directing VAMP2 trafficking. Investigations, 105, 597–605.
Traffic, 9, 417–29. Kaminski, W. E., et al. (2000). Identification of a novel human sterol-sensitive
Harold, D., et al. (2009). Genome-wide association study identifies variants at ATP-binding cassette transporter (ABCA7). Biochemistry and Biophysical
CLU and PICALM associated with Alzheimer’s disease. Nature Genetics, Research Community, 273, 532–8.
41, 1088–93. Kang, J., et al. (1987). The precursor of Alzheimer’s disease amyloid A4
Hayashi, S., et al. (2002). Late-onset frontotemporal dementia with a novel protein resembles a cell-surface receptor. Nature, 325, 733–6.
exon 1 (Arg5His) tau gene mutation. Annals of Neurology, 51, 525–30. Kim, W. S., Weickert, C. S., and Garner, B. (2008). Role of ATP-binding
Herreman, A., et al. (2000). Total inactivation of gamma-secretase activity cassette transporters in brain lipid transport and neurological disease.
in presenilin-deficient embryonic stem cells. Nature Cell Biology, 2, Journal of Neurochemistry, 104, 1145–66.
461–2. Kim, W. S., et al. (2005). Abca7 null mice retain normal macrophage
Heyman, A., et al. (1998). Cerebral infarcts in patients with autopsy-proven phosphatidylcholine and cholesterol efflux activity despite alterations
Alzheimer’s disease: CERAD, part XVIII. Consortium to Establish a in adipose mass and serum cholesterol levels. Journal of Biology and
Registry for Alzheimer’s Disease. Neurology, 51, 159–62. Chemistry, 280, 3989–95.
Kim, Y., et al. (2011). Genetic dissection of susceptibility to vascular dementia. Maat-Schieman, M. L., et al. (1996). Hereditary cerebral hemorrhage with
Psychiatry and Genetics, 21, 69–76. amyloidosis-Dutch type (HCHWA-D): II—A review of histopathological
Kimberly, W. T., et al. (2002). Complex N-linked glycosylated nicastrin aspects. Brain Pathology, 6, 115–20.
associates with active gamma-secretase and undergoes tight cellular Mackenzie, I. R., et al. (2006a). A family with tau-negative frontotemporal
regulation. Journal of Biology and Chemistry, 277, 35113–17. dementia and neuronal intranuclear inclusions linked to chromosome
Klein, R. J., et al. (2005). Complement factor H polymorphism in age-related 17. Brain, 129, 853–67.
macular degeneration. Science, 308, 385–9. Mackenzie, I. R., et al. (2006b). Dementia lacking distinctive histology
Kodama, K., et al. (2000). Familial frontotemporal dementia with a P301L tau (DLDH) revisited. Acta Neuropathologica, 112, 551–9.
mutation in Japan. Journal of Neurological Science, 176, 57–64. Maier, M., et al. (2008). Complement C3 deficiency leads to accelerated
Koike, H., et al. (1999). Membrane-anchored metalloprotease MDC9 has an amyloid beta plaque deposition and neurodegeneration and modulation
alpha-secretase activity responsible for processing the amyloid precursor of the microglia/macrophage phenotype in amyloid precursor protein
protein. Biochemistry Journal, 343 (Pt 2), 371–5. transgenic mice. Journal of Neuroscience, 28, 6333–41.
Koo, E. H. and Squazzo, S. L. (1994). Evidence that production and release of Mann, D. M. (1988a). Alzheimer’s disease and Down’s syndrome.
amyloid beta-protein involves the endocytic pathway. Journal of Biology Histopathology, 13, 125–37.
and Chemistry, 269, 17386–9. Mann, D. M. (1988b). The pathological association between Down syndrome
Kwiatkowski, T. J., Jr., et al. (2009). Mutations in the FUS/TLS gene on and Alzheimer disease. Mechanical Ageing Development, 43, 99–136.
chromosome 16 cause familial amyotrophic lateral sclerosis. Science, Mann, U. M., et al . (1992 ) . Heterogeneity in Alzheimer’s disease:
323, 1205–8. progression rate segregated by distinct neuropsychological and
Lah, J. J. and Levey, A. I. (2000). Endogenous presenilin-1 targets to endocytic cerebral metabolic profiles. Journal of Neurology and Neurosurgical
rather than biosynthetic compartments. Molecular Cell Neuroscience, 16, Psychiatry, 55, 956–9.
111–26. Martinez, A., et al. (2005). Disruption of ephrin-A/EphA binding
Lai, K. O. and Ip, N. Y. (2009). Synapse development and plasticity: roles of alters synaptogenesis and neural connectivity in the hippocampus.
ephrin/Eph receptor signaling. Current Opinions in Neurobiology, 19, Neuroscience, 135, 451–61.
275–83. Masliah, E., et al. (2001). Altered expression of synaptic proteins occurs early
Lambert, J. C., et al. (2009). Genome-wide association study identifies during progression of Alzheimer’s disease. Neurology, 56, 127–9.
variants at CLU and CR1 associated with Alzheimer’s disease. Nature May, P. C. and Finch, C. E. (1992). Sulfated glycoprotein 2: new relationships
Genetics, 41, 1094–9. of this multifunctional protein to neurodegeneration. Trends in
Lammich, S., et al. (1999). Constitutive and regulated alpha-secretase Neuroscience, 15, 391–6.
cleavage of Alzheimer’s amyloid precursor protein by a disintegrin McCarron, M. O., et al. (2000). Prospective study of apolipoprotein E
metalloprotease. Proceedings of the National Academy of Science USA, 96, genotype and functional outcome following ischemic stroke. Archives of
3922–7. Neurology, 57, 1480–4.
Lantos, P. and Cairns, N. (2000). The neuropathology of Alzheimer’s disease. McGeer, E. G. and McGeer, P. L. (2001). Innate immunity in Alzheimer’s
In: O’Brien, J., Arnes, D. and Burns, A. (eds) Dementia, 2nd edition. disease: a model for local inflammatory reactions. Molecular Interventions,
London, Arnold. 1, 22–9.
Lee, V. M., et al. (1991). A68: a major subunit of paired helical filaments and McKhann, G., et al. (1984). Clinical diagnosis of Alzheimer’s disease: report
derivatized forms of normal Tau. Science, 251, 675–8. of the NINCDS-ADRDA Work Group under the auspices of Department
Lee, V. M., Goedert, M., and Trojanowski, J. Q. (2001). Neurodegenerative of Health and Human Services Task Force on Alzheimer’s Disease.
tauopathies. Annual Review of Neuroscience, 24, 1121–59. Neurology, 34, 939–44.
Lemere, C. A., et al. (1996). Sequence of deposition of heterogeneous amyloid McKhann, G. M., et al. (2011). The diagnosis of dementia due to
beta-peptides and APO E in Down syndrome: implications for initial Alzheimer’s disease: recommendations from the National Institute on
events in amyloid plaque formation. Neurobiological Disease, 3, 16–32. Aging-Alzheimer’s Association workgroups on diagnostic guidelines for
Levy-Lahad, E., et al. (1995a). Candidate gene for the chromosome 1 familial Alzheimer’s disease. Alzheimers Dementia, 7, 263–9.
Alzheimer’s disease locus. Science, 269, 973–7. Morita, M., et al. (2006). A locus on chromosome 9p confers susceptibility to
Levy-Lahad, E., et al. (1995b). A familial Alzheimer’s disease locus on ALS and frontotemporal dementia. Neurology, 66, 839–44.
chromosome 1. Science, 269, 970–3. Morris, H. R., et al . (2001). The genetic and pathological classification
Lewis, J., et al. (2000). Neurofibrillary tangles, amyotrophy and progressive of familial frontotemporal dementia . Archives of Neurology, 58,
motor disturbance in mice expressing mutant (P301L) tau protein. 1813–16.
Nature Genetics, 25, 402–5. Morrow, J. A., et al. (2002). Apolipoprotein E4 forms a molten globule. A
Lewis, J., et al. (2001). Enhanced neurofibrillary degeneration in transgenic potential basis for its association with disease. Journal of Biology and
mice expressing mutant tau and APP. Science, 293, 1487–91. Chemistry, 277, 50380–5.
Li, Y. M., et al. (2000). Photoactivated gamma-secretase inhibitors directed to Mukherjee, O., et al. (2007). Haplotype-based association analysis of the
the active site covalently label presenilin 1. Nature, 405, 689–94. MAPT locus in late onset Alzheimer’s disease. BMC Genetics, 8, 3.
Lin, X., et al. (2000). Human aspartic protease memapsin 2 cleaves the Mushegian, A. (2002). Refining structural and functional predictions for
beta-secretase site of beta-amyloid precursor protein. Proceedings of the secretasome components by comparative sequence analysis. Proteins, 47,
National Academy of Science USA, 97, 1456–60. 69–74.
Luo, Y.R., et al. (2001). Mice deficient in BACE1, the Alzheimer’s beta-secretase, Myers, A. J., et al. (2005). The H1c haplotype at the MAPT locus is associated
have normal phenotype and abolished beta-amyloid generation. Nature with Alzheimer’s disease. Human Molecular Genetics, 14, 2399–404.
Neuroscience, 4, 231–2. Myers, R. H., et al. (1996). Apolipoprotein E epsilon4 association with
Luty, A. A., et al. (2008). Pedigree with frontotemporal lobar degeneration— dementia in a population-based study: the Framingham study. Neurology,
motor neuron disease and Tar DNA binding protein-43 positive 46, 673–7.
neuropathology: genetic linkage to chromosome 9. BMC Neurology, 8, 32. Naj, A. C., et al. (2011). Common variants at MS4A4/MS4A6E, CD2AP,
Lynch, D. K., et al. (2003). A Cortactin-CD2-associated protein (CD2AP) CD33 and EPHA1 are associated with late-onset Alzheimer’s disease.
complex provides a novel link between epidermal growth factor receptor Nature Genetics, 43, 436–41.
endocytosis and the actin cytoskeleton. Journal of Biology and Chemistry, Neary, D., Snowden, J., and Mann, D. (2005). Frontotemporal dementia.
278, 21805–13. Lancet Neurology, 4, 771–80.
Nicholl, D. J., et al. (2003). An English kindred with a novel recessive Rogers, J.M., et al. (2006). Peripheral clearance of amyloid beta peptide by
tauopathy and respiratory failure. Annals of Neurology, 54, 682–6. complement C3-dependent adherence to erythrocytes. Neurobiological
Nitsch, R. M., et al. (1992). Release of Alzheimer amyloid precursor derivatives Aging, 27, 1733–9.
stimulated by activation of muscarinic acetylcholine receptors. Science, Roks, G., et al. (1999). Mutation screening of the tau gene in patients with
258, 304–7. early-onset Alzheimer’s disease. Neuroscience Letters, 277, 137–9.
Nordstedt, C.P., et al. (1993). Identification of the Alzheimer beta/A4 amyloid Rollinson, S., et al. (2011). Frontotemporal lobar degeneration genome
precursor protein in clathrin-coated vesicles purified from PC12 cells. wide association study replication confirms a risk locus shared with
Journal of Biology and Chemistry, 268, 608–12. amyotrophic lateral sclerosis. Neurobiological Aging, 32, 758 e1–7.
Nunan, J. and Small, D. H. (2000). Regulation of APP cleavage by alpha-, Rosand, J., et al. (2000). Warfarin-associated hemorrhage and cerebral
beta- and gamma-secretases. FEBS Letters, 483, 6–10. amyloid angiopathy: a genetic and pathologic study. Neurology, 55,
Palsdottir, A., et al. (1988). Mutation in cystatin C gene causes hereditary 947–51.
brain haemorrhage. Lancet, 2, 603–4. Rosso, S. M., et al. (2001). Familial frontotemporal dementia with
Pappolla, M. A., et al. (1998). Evidence of oxidative stress and in vivo ubiquitin-positive inclusions is linked to chromosome 17q21-22. Brain,
neurotoxicity of beta-amyloid in a transgenic mouse model of Alzheimer’s 124, 1948–57.
disease: a chronic oxidative paradigm for testing antioxidant therapies in Rosso, S. M., et al. (2003). Frontotemporal dementia in The Netherlands:
vivo. American Journal of Pathology, 152, 871–7. patient characteristics and prevalence estimates from a population-based
Parkinson, N., et al. (2006). ALS phenotypes with mutations in CHMP2B study. Brain, 126, 2016–22.
(charged multivesicular body protein 2B). Neurology, 67, 1074–7. Russo, C., et al. (2001). Amino-terminal modification and tyrosine
Parvathy, S., et al. (1998). Alzheimer’s amyloid precursor protein alpha-secretase phosphorylation of [corrected] carboxy-terminal fragments of the
is inhibited by hydroxamic acid-based zinc metalloprotease inhibitors: amyloid precursor protein in Alzheimer’s disease and Down’s syndrome
similarities to the angiotensin converting enzyme secretase. Biochemistry, brain. Neurobiological Diseases, 8, 173–80.
37, 1680–5. Sabbadini, G., et al. (1995). Cerebral autosomal dominant arteriopathy with
Parvathy, S., et al. (1999). Cleavage of Alzheimer’s amyloid precursor protein subcortical infarcts and leucoencephalopathy (CADASIL). Clinical,
by alpha-secretase occurs at the surface of neuronal cells. Biochemistry, neuroimaging, pathological and genetic study of a large Italian family.
38, 9728–34. Brain, 118 (Pt 1), 207–15.
Pastor, P., et al. (2003). Apolipoprotein E epsilon4 modifies Alzheimer’s Saunders, A. M., et al. (1993). Association of apolipoprotein E allele epsilon
disease onset in an E280A PS1 kindred. Annals of Neurology, 54, 163–9. 4 with late-onset familial and sporadic Alzheimer’s disease. Neurology,
Pericak-Vance, M. A., et al. (1991). Linkage studies in familial Alzheimer 43, 1467–72.
disease: evidence for chromosome 19 linkage. American Journal of Schrijvers, E. M., et al. (2012). Genome-wide association study of vascular
Human Genetics, 48, 1034–50. dementia. Stroke, 43, 315–19.
Polvikoski, T., et al. (2001). Prevalence of Alzheimer’s disease in very elderly Schroder, R., et al. (2005). Mutant valosin-containing protein causes a novel
people: a prospective neuropathological study. Neurology, 56, 1690–6. type of frontotemporal dementia. Annals of Neurology, 57, 457–61.
Poorkaj, P., et al. (1998). Tau is a candidate gene for chromosome 17 Schroeter, E. H., Kisslinger, J. A., and Kopan, R. (1998). Notch-1 signalling
frontotemporal dementia. Annals of Neurology, 43, 815–25. requires ligand-induced proteolytic release of intracellular domain.
Poorkaj, P., et al. (2001). Frequency of tau gene mutations in familial and Nature, 393, 382–6.
sporadic cases of non-Alzheimer dementia. Archives of Neurology, 58, Selkoe, D. J. (2001). Alzheimer’s disease: genes, proteins, and therapy.
383–7. Physiololgical Review, 81, 741–66.
Poorkaj, P., et al. (2002). An R5L tau mutation in a subject with a progressive Seshadri, S., et al. (2010). Genome-wide analysis of genetic loci associated
supranuclear palsy phenotype. Annals of Neurology, 52, 511–6. with Alzheimer disease. Journal of the American Medical Association,
Postina, R., et al. (2004). A disintegrin-metalloproteinase prevents amyloid 303, 1832–40.
plaque formation and hippocampal defects in an Alzheimer disease Shastry, B. S. and Giblin, F. J. (1999). Genes and susceptible loci of Alzheimer’s
mouse model. Journal of Clinical Investigations, 113, 1456–64. disease. Brain Research Bulletin, 48, 121–7.
Prakash, N., et al. (2002). Mouse Notch 3 expression in the pre- and postnatal Sherrington, R., et al. (1995). Cloning of a gene bearing missense mutations
brain: relationship to the stroke and dementia syndrome CADASIL. Exp in early-onset familial Alzheimer’s disease. Nature, 375, 754–60.
Cell Res, 278, 31–44. Shioi, J., et al. (1992). Chondroitin sulfate proteoglycan form of the
Puglielli, L., Tanzi, R. E., and Kovacs, D. M. (2003). Alzheimer’s disease: the Alzheimer’s beta-amyloid precursor. Journal of Biology and Chemistry,
cholesterol connection. Nature Neuroscience, 6, 345–51. 267, 13819–22.
Quon, D., et al. (1991). Formation of beta-amyloid protein deposits in brains Shioi, J., et al. (1993). Chondroitin sulfate proteoglycan form of cellular and
of transgenic mice. Nature, 352, 239–41. cell-surface Alzheimer amyloid precursor. Neuroscience Letters, 154,
Rademakers, R., et al. (2002). Tau negative frontal lobe dementia at 17q21: 121–4.
significant finemapping of the candidate region to a 4.8 cM interval. Sinha, S., et al. (1999). Purification and cloning of amyloid precursor protein
Molecular Psychiatry, 7, 1064–74. beta-secretase from human brain. Nature, 402, 537–40.
Raiha, I., et al. (1996). Alzheimer’s disease in Finnish twins. Lancet, 347, Skibinski, G., et al. (2005). Mutations in the endosomal ESCRTIII-complex
573–8. subunit CHMP2B in frontotemporal dementia. Nature Genetics, 37,
Reiman, E. M., et al. (2007). GAB2 alleles modify Alzheimer’s risk in APOE 806–8.
epsilon4 carriers. Neuron, 54, 713–20. Skoog, I., et al. (1998). A population study of apoE genotype at the age of 85:
Renton, A. E.J., et al. (2011). A hexanucleotide repeat expansion in C9ORF72 relation to dementia, cerebrovascular disease, and mortality. Journal of
is the cause of chromosome 9p21-linked ALS-FTD. Neuron, 72, 257–68. Neurology and Neurosurgical Psychiatry, 64, 37–43.
Rizzu, P., et al . (1999 ) . High prevalence of mutations in the Skovronsky, D. M., et al. (2000). Protein kinase C-dependent alpha-secretase
microtubule-associated protein tau in a population study of competes with beta-secretase for cleavage of amyloid-beta precursor
frontotemporal dementia in the Netherlands. American Journal of protein in the trans-golgi network. Journal of Biology and Chemistry,
Human Genetics, 64, 414–21. 275, 2568–75.
Rogaev, E. I., et al. (1995). Familial Alzheimer’s disease in kindreds with Slooter, A. J., et al. (1997). Apolipoprotein E epsilon 4 and the risk of dementia
missense mutations in a gene on chromosome 1 related to the Alzheimer’s with stroke. A population-based investigation. Journal of the American
disease type 3 gene. Nature, 376, 775–8. Medical Association, 277, 818–21.
Smith, S. M., Renden, R., and Von Gersdorff, H. (2008). Synaptic vesicle Van Broeckhoven, C., et al. (1992). Mapping of a gene predisposing to
endocytosis: fast and slow modes of membrane retrieval. Trends in early-onset Alzheimer’s disease to chromosome 14q24.3. Nature Genetics,
Neuroscience, 31, 559–68. 2, 335–9.
Song, Y., Stampfer, M. J., and Liu, S. (2004). Meta-analysis: apolipoprotein Vance, C., et al. (2006). Familial amyotrophic lateral sclerosis with
E genotypes and risk for coronary heart disease. Annals of Internal frontotemporal dementia is linked to a locus on chromosome 9p13.2-
Medicine, 141, 137–47. 21.3. Brain, 129, 868–76.
Sorbi, S., et al. (1995). Epistatic effect of APP717 mutation and apolipoprotein Van Deerlin, V. M., et al. (2010). Common variants at 7p21 are associated
E genotype in familial Alzheimer’s disease. Annals of Neurology, 38, with frontotemporal lobar degeneration with TDP-43 inclusions. Nature
124–7. Genetics, 42, 234–9.
Sperfeld, A. D., et al. (1999). FTDP-17: an early-onset phenotype with Van Der Zee, J. and Van Broeckhoven, C. (2011). TMEM106B a novel risk
parkinsonism and epileptic seizures caused by a novel mutation. Annals factor for frontotemporal lobar degeneration. Journal of Molecular
of Neurology, 46, 708–15. Neuroscience, 45, 516–21.
Spillantini, M.G., et al. (1997). Familial multiple system tauopathy with Van Der Zee, J., et al. (2011). TMEM106B is associated with frontotemporal
presenile dementia: a disease with abundant neuronal and glial tau lobar degeneration in a clinically diagnosed patient cohort. Brain, 134,
filaments. Proceedings of the National Academy of Science USA, 94, 808–15.
4113–18. Van Duijn, C. M., et al. (1991). Familial aggregation of Alzheimer’s disease
Spillantini, M. G., Bird, T. D., and Ghetti, B. (1998a). Frontotemporal and related disorders: a collaborative re-analysis of case-control studies.
dementia and Parkinsonism linked to chromosome 17: a new group of EURODEM Risk Factors Research Group. International Journal of
tauopathies. Brain Pathology, 8, 387–402. Epidemiology, 20 Suppl 2, S13–20.
Spillantini, M. G., et al. (1998b). Mutation in the tau gene in familial multiple Van Herpen, E., et al. (2003). Variable phenotypic expression and extensive
system tauopathy with presenile dementia. Proceedings of the National tau pathology in two families with the novel tau mutation L315R. Annals
Academy of Science USA, 95, 7737–41. of Neurology, 54, 573–81.
St George-Hyslop, P. H., et al . (1987 ) . Th e genetic defect causing Van Swieten, J. and Spillantini, M. G. (2007). Hereditary frontotemporal
familial Alzheimer’s disease maps on chromosome 21. Science, 235, dementia caused by Tau gene mutations. Brain Pathology, 17, 63–73.
885–90. Van Swieten, J. C., et al. (1999). Phenotypic variation in hereditary
Strittmatter, W. J., et al. (1993). Apolipoprotein E: high-avidity binding to frontotemporal dementia with tau mutations. Annals of Neurology, 46,
beta-amyloid and increased frequency of type 4 allele in late-onset 617–26.
familial Alzheimer disease. Proceedings of the National Academy of Van Swieten, J.C., et al. (2007). The DeltaK280 mutation in MAP tau favors
Science USA, 90, 1977–81. exon 10 skipping in vivo. Journal of Neuropathology and Experimental
Stuart, W. D., et al. (1992). Structure and stability of apolipoprotein Neurology, 66, 17–25.
J-containing high-density lipoproteins. Biochemistry, 31, 8552–9. Vassar, R. (2001). The beta-secretase, BACE: a prime drug target for
Sturchler-Pierrat, C., et al. (1997). Two amyloid precursor protein transgenic Alzheimer’s disease. Journal of Molecular Neuroscience, 17, 157–70.
mouse models with Alzheimer disease-like pathology. Proceedings of the Vassar, R., et al. (1999). Beta-secretase cleavage of Alzheimer’s amyloid
National Academy of Science USA, 94, 13287–92. precursor protein by the transmembrane aspartic protease BACE.
Sudlow, C., et al. (2006). Does apolipoprotein E genotype influence the Science, 286, 735–41.
risk of ischemic stroke, intracerebral hemorrhage, or subarachnoid Verpillat, P., et al. (2002). Association between the extended tau haplotype
hemorrhage? Systematic review and meta-analyses of 31 studies among and frontotemporal dementia. Archives of Neurology, 59, 935–9.
5961 cases and 17,965 controls. Stroke, 37, 364–70. Vidal, R., et al. (1999). A stop-codon mutation in the BRI gene associated
Suzuki, N., et al. (1994a). An increased percentage of long amyloid beta with familial British dementia. Nature, 399, 776–81.
protein secreted by familial amyloid beta protein precursor (beta Vidal, R., et al. (2000). A decamer duplication in the 3′ region of the BRI
APP717) mutants. Science, 264, 1336–40. gene originates an amyloid peptide that is associated with dementia in
Suzuki, N., et al. (1994b). High tissue content of soluble beta 1-40 is linked a Danish kindred. Proceedings of the National Academy of Science USA,
to cerebral amyloid angiopathy. American Journal of Pathology, 145, 97, 4920–5.
452–60. Walsh, D. M., et al. (2002). Amyloid-beta oligomers: their production, toxicity
Suzuki, T., et al. (1994c). Cell cycle-dependent regulation of the and therapeutic inhibition. Biochemistry Society Transactions, 30, 552–7.
phosphorylation and metabolism of the Alzheimer amyloid precursor Warwick Daw, E., et al. (2000). The number of trait loci in late-onset
protein. EMBO Journal, 13, 1114–22. Alzheimer disease. American Journal of Human Genetics, 66, 196–204.
Takeuchi, A., et al. (2000). Age-related amyloid beta deposition in transgenic Watts, G. D., et al. (2004). Inclusion body myopathy associated with Paget
mice overexpressing both Alzheimer mutant presenilin 1 and amyloid disease of bone and frontotemporal dementia is caused by mutant
beta precursor protein Swedish mutant is not associated with global valosin-containing protein. Nature Genetics, 36, 377–81.
neuronal loss. American Journal of Pathology, 157, 331–9. Weihofen, A. and Martoglio, B. (2003). Intramembrane-cleaving proteases:
Tateno, H., et al. (2007). Distinct endocytic mechanisms of CD22 (Siglec-2) controlled liberation of proteins and bioactive peptides. Trends in Cell
and Siglec-F reflect roles in cell signaling and innate immunity. Molecular Biology, 13, 71–8.
Cell Biology, 27, 5699–710. Wolfe, M. S., et al. (1999). Two transmembrane aspartates in presenilin-1
Thinakaran, G., et al. (1996). Endoproteolysis of presenilin 1 and accumulation required for presenilin endoproteolysis and gamma-secretase activity.
of processed derivatives in vivo. Neuron, 17, 181–90. Nature, 398, 513–7.
Thinakaran, G., Slunt, H. H., and Sisodia, S. S. (1995). Novel regulation Woo, D., et al. (2002). Genetic and environmental risk factors for intracerebral
of chondroitin sulfate glycosaminoglycan modification of amyloid hemorrhage: preliminary results of a population-based study. Stroke, 33,
precursor protein and its homologue, APLP2. Journal of Biology and 1190–5.
Chemistry, 270, 16522–5. Wyss-Coray, T., et al. (2002). Prominent neurodegeneration and increased
Tolnay, M. and Probst, A. (2002). Frontotemporal lobar degeneration—tau as plaque formation in complement-inhibited Alzheimer’s mice. Proceedings
a pied piper? Neurogenetics, 4, 63–75. of the National Academy of Science USA, 99, 10837–42.
Tournier-Lasserve, E., et al. (1993). Cerebral autosomal dominant arteriopathy Yamamoto, Y., et al. (2011). Review: molecular genetics and pathology of
with subcortical infarcts and leukoencephalopathy maps to chromosome hereditary small vessel diseases of the brain. Neuropathology and Applied
19q12. Nature Genetics, 3, 256–9. Neurobiology, 37, 94–113.
Yan, R., et al. (1999). Membrane-anchored aspartyl protease with Alzheimer’s Zhang, J., et al. (2005). The tau gene haplotype h1 confers a susceptibility to
disease beta-secretase activity. Nature, 402, 533–7. Parkinson’s disease. European Neurology, 53, 15–21.
Yasuda, M., et al. (1999). A mutation in the microtubule-associated protein Zhang, Z., et al. (2000). Presenilins are required for gamma-secretase
tau in pallido-nigro-luysian degeneration. Neurology, 53, 864–8. cleavage of beta-APP and transmembrane cleavage of Notch-1. Nature
Yu, G., et al. (2000). Nicastrin modulates presenilin-mediated notch/glp-1 Cell Biology, 2, 463–5.
signal transduction and betaAPP processing. Nature, 407, 48–54. Zotova, E., et al. (2010). Inflammation in Alzheimer’s disease: relevance to
Zabetian, C. P., et al. (2007). Association analysis of MAPT H1 haplotype and pathogenesis and therapy. Alzheimers Research and Therapy, 2, 1.
subhaplotypes in Parkinson’s disease. Annals of Neurology, 62, 137–44.
CHAPTER 9
Psychiatric assessment
of older people
Alan Thomas
The fundamentals of psychiatric assessment for older people are that exposes the shallowness of what can be achieved in clinic, an
the same as for younger people (thorough history, detailed mental aspect that frequently strikes medical students when visiting with
state assessment, and a physical examination). However, the details consultants (Anderson and Aquilina, 2002). On arrival, the doctor
and balance of each of these components varies considerably from is already aware of the local environment, whether it is large houses
the assessment in younger adults due to the high prevalence of with spacious gardens or empty flats with boarded up windows, and
physical morbidity and the crucial importance of proper cogni- the quality of the patient’s own house before knocking on the door.
tive assessment. In the same way, whilst the same kind of follow-up On entering, he is rapidly able to determine the state of the entrance
assessments, e.g. social work assessment and investigations, e.g. area and main room and can observe any hazards, such as loose
neuroimaging, are needed, the assessments themselves are quite carpets or objects strewn on the floor, as well as the general level
different. of cleanliness. After politely seeking permission, a quick inspec-
tion of the kitchen, looking for out of date food, an empty fridge,
The Venue for the Initial Assessment or burned pans can be revealing. Interestingly, it was this aspect
of home-based assessments, the ability to understand the home
Assessment of the older person with a possible mental illness may
environment, that one study reported as the main reason general
take place in a range of settings, as considered below, and whilst
practitioners asked for such assessments (Hardy-Thompson et al.,
each has its strengths and weaknesses, the evidence about the clini-
1992). It is much easier for a key relative or carer to be present at the
cal and cost effectiveness of these different services is poor (Parker
patient’s home, and often several are present. As well as providing
et al., 2000). In practice, the place where the initial assessment is
corroboration for the history, they can assist the patient to remem-
conducted is constrained by the existing design of the local serv-
ber and clarify important elements of the history, and it may give
ice, the manner of the referral and the urgency of the clinical prob-
an insight into family dynamics and the kind of relationships the
lem. The assessing clinician needs to show flexibility to adapt his
patient enjoys. For those patients who have carers attending already,
assessment to the advantages and disadvantages of the different
there should be a written care plan available and daily notes from
locations.
the carer(s), giving useful additional information. An informative
practice is to ask patients to produce their medication for check-
Home-based Assessment ing, which, as well as confirming their current treatment, will often
Patients referred for old age psychiatry assessment should be seen reveal evidence of erratic compliance. A domiciliary assessment
initially at home. Such domiciliary assessment has several advan- also shows patients at their best, giving a more realistic assessment
tages over a hospital-based assessment for older people and is of their mood, cognition, and general behaviour. Although most
especially important for those with cognitive impairment. The of these advantages will be familiar to clinicians practising in the
patient is spared a time consuming, tiring, and perhaps expensive field, they are difficult to study, but one investigation reported a
journey to clinic. Clinics, especially psychiatric ones, are too often large reduction in nonattendance for home-based (1.7%) versus
unpleasant places to wait around and can be particularly distress- clinic-based (21.2%) assessments, leading to a more efficient use
ing to frail, older, and cognitively impaired people. Unsurprisingly of valuable medical time (Anderson and Aquilina, 2002). This also
therefore, when patients were asked, they overwhelmingly opted results in patients being followed-up more effectively (Benbow,
for home-based assessment over clinic-based assessment or assess- 1990) . Nonattendance at clinics leads to important delays in assess-
ment at a primary care clinic (Jones et al., 1987). For the clinician, a ment and on occasions assessment does not happen at all (Frankel
home-based assessment provides a depth and quality of information et al., 1989).
Assessment in Residential Homes The Psychiatric Assessment

When a referred patient is living in a residential care facility this is Aims of the assessment
usually because of the development of some form of behavioural A psychiatric assessment aims to achieve much more than a diagno-
disturbance in someone with a pre-existing illness, in most cases sis. A thorough assessment should enable the clinician to produce
dementia. They have often therefore been previously assessed and, well-reasoned differential diagnoses and have some confidence
providing their previous records are available, this enables the in the most likely main diagnosis. But the initial assessment also
psychiatrist to concentrate on the current issue(s). Where there aims to engage the patient and his or her family to facilitate further
has been no formal psychiatric assessment previously it can be assessment as necessary and to foster cooperation in all aspects of
very difficult to complete a full history, because informants with a future management. Thus the process of assessment should estab-
long-term knowledge of the patient are often not available. In some lish a good rapport with the patient and his/her family and car-
cases they may be available by telephone, but usually a pragmatic ers, and achieving such a positive relationship improves the quality
approach is necessary, making the best use of available sources of of information obtained. There is now a much greater tendency
information. As well as that in the referral letter, documents from amongst old age psychiatrists to disclose the diagnosis of demen-
social services recording the admission process to the home and tia, although other doctors in other specialities, e.g. neurology and
care records can be consulted. This can furnish valuable informa- geriatric medicine, remain reluctant to do so. This development has
tion about the previous history. Care staff are usually able to pro- probably been largely driven by the availability of licensed treat-
vide a reasonable account about the current problems, although it is ments for Alzheimer’s disease, but the NDS and NICE Dementia
prudent to enquire how well the carer-informant knows the patient Guidelines both encourage clinicians to discuss dementia diagnosis
and for how long. and its implications with patients and families because of the advan-
tages it brings, e.g. facilitating advanced care planning (NICE, 2006;
Outpatient Assessment Health, 2009). Disclosing and discussing diagnosis is an important
For most patients referred to old age psychiatry services an initial clinical skill and establishing a good rapport paves the way for deal-
assessment in their own home remains the choice for the reasons ing with this sensitive subject (Bamford et al., 2004).
outlined above. Whilst some people with functional illnesses, espe- As well as eliciting information to determine a diagnosis it is
cially those who are physically fit, may be adequately assessed in an important to identify other relevant problems that need dealing
outpatient clinic, in practice it is usually difficult to identify with with in their own right. The importance of noncognitive symp-
confidence such people from the referral letter alone. The main toms in dementia (commonly referred to as behavioural and psy-
group where outpatient clinic assessment can be advantageous is chological symptoms in dementia, BPSD) is now well recognized
in people with a possible early dementia, and over the last 20 years (Finkel and Burns, 2000), yet these usually play no role in the diag-
memory clinics focusing on such patients have become an increas- nosis that is based on cognitive symptoms, although new criteria
ingly prominent feature in old age psychiatry services (Lindesay include them as one element in diagnosis (McKhann et al., 2011).
et al., 2002). Such clinics often deal with younger, less cognitively Clarifying functional difficulties, such as with mobility or personal
impaired patients and for those who have a dementia they are ear- care, and the presence of other relevant medical illnesses is essen-
lier in the course of their illness (Luce et al., 2001). More recently, tial as well. Thus, for example, a summary of a patient assessment
memory clinics have expanded out of more academic settings and in should include, in addition to a diagnosis of dementia, reference
England, following the publication of the 2009 National Dementia to such issues as psychotic symptoms, mood disturbance, postural
Strategy (NDS) (Health, 2009), there has been an increase in spe- instability, difficulties using stairs, and bathing. The initial assess-
cialist memory assessment services. The NDS emphasized the ment should not, indeed cannot, aim to achieve a detailed under-
importance of early and accurate diagnosis as one of its key objec- standing of all these matters, but should identify the range of issues
tives. Memory services are discussed in detail in Chapter 24. that need further assessment. The initial assessment should there-
fore be holistic in aiming to achieve a clear overall understanding
of the patient’s complete set of needs.
Inpatient Assessment
A third to a quarter of new referrals to old age psychiatry serv- The initial assessment
ices come from wards in general hospitals, and assessment here Upon first meeting the patient, a good handshake, polite smile, and
is a very different experience from home-based assessments and clear explanation of who you are and why you have come begins
brings its own difficulties. The growing recognition that inpatient the process of forming a working relationship. The referral let-
assessments require a different approach has led to the increasing ter should have identified the key issues, but it is important at the
development in many places of specialist liaison old age psychiatry outset to ensure they are correct, and where the letter is unclear
services. The UK NICE Clinical Guidelines for Dementia state that to ask what the main problems are. It is common for patients to
every acute hospital should have a dedicated specialist liaison team, deny they have any problems, especially when they have cognitive
planned by acute, mental health, and social services, able to holisti- impairment, and not infrequently the patient is found to have had
cally assess and manage older people in the acute hospital setting no involvement in the referral process, which was initiated by con-
with dementia (NICE, 2006), and similarly one of the key objectives cerned family members. Tact and sensitivity are needed to explain
of the National Dementia Strategy is for a dedicated service with the concerns others have about the patient and to obtain patience
an identified lead clinician in the hospital (Health, 2009). Details and cooperation with the formal assessment itself. Seeking permis-
about such services and the special approach and skills needed for sion to speak with family or friends can be difficult in such circum-
inpatient old age psychiatry are discussed in Chapter 22. stances, but it is still important to try to do so. Explaining to the
CHAPTER 9 psychiatric assessment of older people 143
patient that understanding the whole picture is important and that the differential diagnosis of dementia. Thus changes in social inter-
it is standard practice to ask other people for information achieves action and eating behaviour are characteristic of FTD (Neary et al.,
agreement in most cases. 1998) (Chapter 36), and visual hallucinations and sleep changes
(especially REM sleep behaviour disorder; see Chapter 51) are
History diagnostic features of DLB (McKeith et al., 2005) (see Chapter 35).
A careful and detailed history remains the most important element
in the whole assessment; physical and mental state examinations Previous medical and psychiatric history
and special investigations only serve to clarify and confirm the his- Often, a referral from primary care will come with a computer
tory. It is therefore vital to establish a good rapport and give ade- printout of the patient’s previous illnesses and medication. Whilst
quate time to cover all the necessary aspects in the clinical history. such records are helpful as a starting point, they often contain
errors and omissions and it is prudent to confirm these illnesses,
Presenting complaint especially where there is apparent contradiction, e.g. a history of
One main problem may be given as the reason for referral, but typi- hypothyroidism but no prescription for thyroxine. Questions about
cally several interrelated issues are present. These need to be iden- previous psychiatric history should always be asked and any ill-
tified and clarified individually, and their relationship, temporal nesses in the months before and since the onset of the presenting
and otherwise, established. At the outset it is appropriate to check symptoms should be carefully assessed; cognitive decline and mood
that patients understand why they are being assessed and that the alterations after operations or major illnesses are not uncommon.
presenting problems are all those that need to be discussed. This If ‘vascular factors’ (stroke, transient ischaemic attacks, myocardial
may reveal important issues, such as that they resent having been infarction, angina pectoris, peripheral vascular disease, diabetes,
referred and are angry with their wife for having brought this about. hypertension, and falls) were not covered in the presenting com-
The development of each main symptom can then be covered, ask- plaint they should be asked about here, as evidence in support of or
ing about key aspects such as duration, pattern of onset, change against a diagnosis of VaD.
through time, and any known precipitants. The timing of the onset
of amnesia and cognitive decline is frequently difficult to ascertain, Current medication
but asking about this in relation to memorable dates can be use- Whilst a good referral usually lists the currently prescribed treat-
ful, e.g. was the person his usual self last birthday or at Christmas ments, it is prudent to check this information against the actual
or on holiday last summer? In old age psychiatry, people usually medication the patient is taking by asking the patient to produce
present with a history of insidious onset of amnesia, and whilst this his or her medication. As discussed earlier, this helps identify prob-
is characteristic of Alzheimer’s disease, dementia with Lewy bodies lems with compliance, but also enables one to check about non-
(DLB) also presents this way and it is a common pattern in vascular prescribed, alternative treatments. The use of substances such as
dementia (VaD), especially of the subcortical type. The classic VaD vitamins, ginseng, and gingko biloba is not uncommon in older
presentation, a sudden onset of dementia and/or a stepwise dete- people. Polite questioning about why these treatments, as well as
rioration, is unusual in old age psychiatry, probably because such the prescribed ones, are being taken can give useful insights into
patients usually present acutely to stroke services. current concerns and may raise issues not identified elsewhere.
A more rapid (but not sudden) onset of cognitive impairment, Checking the treatment actually being taken may also reveal an
over a few days or weeks, is consistent with a delirium or a depres- important recent change of medication; it is not uncommon for
sion and should lead to questions to identify these syndromes, medication to be altered in hospital and for the patient to be on dif-
e.g. for delirium the presence of an infection or fluctuating con- ferent treatments from the referral letter. Treatments with adverse
sciousness, and for depression of affective change and biological effects on cognition and behaviour may have been commenced, e.g.
symptoms. In someone with an insidious onset of amnesia then oxybutinin for urinary incontinence impairs cognition, and some-
questions to identify other cognitive symptoms may be appropri- times necessary treatments may have been stopped inadvertently,
ate, e.g. problems reading or writing, difficulties naming objects or e.g. antihypertensives.
recognizing objects or people, but such issues are usually identified
more clearly during cognitive testing. Of more importance during Family history
history taking is covering the range of noncognitive symptoms that Obtaining a family history of mental illness can be helpful because
occur in dementia. It is essential to ask about problems in everyday the major mental illnesses, including dementia, have a definite
function, in order to identify needs for which further assessment genetic contribution. In practice, however, it can be very difficult
and help can be given and to assess the severity of the dementia. to determine the validity of the information given for at least two
These are usually divided into basic activities of daily living (ability reasons. First, younger informants may not be able to corrobo-
to maintain personal care) and instrumental activities of daily living rate the information given by older patients, and where cognitive
(more complex everyday tasks) and the use of a structured assess- impairment is present this may not be accurate. Second, when
ment tool, e.g. the Bristol Activities of Daily Living (Bucks et al., informed that a parent had, for example, dementia it is not at all
1996), may help here. Behavioural and psychological disturbances certain this term corresponds to the clinical diagnosis a psychiatrist
are common in dementia, and questions about social interaction, would make. Thus the added diagnostic value of a family history of
mood, paranoia, hallucinations, wandering, aggression, sleep, and mental illness is unclear. Other family history about relationships
eating (both appetite and eating behaviour) are usually appropri- with parents and siblings may be helpful in understanding certain
ate. Identifying such symptoms is important because they are fre- behaviours or for ‘functional illnesses’, but again this information
quently troublesome and when present usually cause more distress can be difficult to verify and is probably of less importance than in
than cognitive symptoms. Specific symptoms are also important in younger adults.
Personal history causes problems, and one that is familiar to old age psychiatrists,
This naturally follows on from the family history and for early life is placement in residential care. Services and residential homes are
often raises the same problems of corroborating information. For aware of the difficulty a patient may have in adapting to this new
cognitively impaired patients it is usually helpful to move to the per- situation, but the spouse left behind has to adapt too and depres-
sonal history near the beginning of the interview because patients sion and alcohol abuse may result.
can talk happily about their earlier life and are not distressed by Clearly, in taking a personal history in old age psychiatry there
difficulties in remembering recent events. Taking a detailed per- is an immense amount of information that could be obtained.
sonal history serves two main functions in old age psychiatry: (1) Clinicians need to adapt their questions to the nature of the prob-
it enables the psychiatrist to assess the severity of amnesia without lem at hand, e.g. questions about relationships in earlier life are
the use of cognitive testing, which can be upsetting for patients who more important for people presenting with a depressive or anxiety
are anxious or in denial. Thus names, dates, and anecdotes from disorder than for those with probable dementia.
early in life may be easily remembered, but events later in work-
ing life and the names of children and especially grandchildren are
Personality
forgotten; the approximate age at which memory fragments gives At the initial assessment the clinician can only begin to understand
an indication of the severity of a dementia; and (2), as in general the personality of a new patient. There is limited value in asking
psychiatry, it helps understanding the patient as a person. either the patient or any informants directly about the patient’s
Obtaining a personal history in old age psychiatry follows personality; such questions provoke stereotyped answers lacking in
the same chronological order as with younger adults but there depth. If a detailed personal and social history has been obtained,
is clearly more ground to cover. Generally, events earlier in life then the questions about relationships with family, at school, and at
will be less important, but care should be taken to sensitively ask work, along with achievements and activities at school, work, and
about the quality of relationships and experiences at all stages of elsewhere, will have shed light on enduring personality traits that
the patient’s life. Unless issues emerge, questions about childhood may have changed with illness and are important in managing the
can be restricted to those to do with happiness and friendships at patient. A few extra questions to clarify matters as this history is
school and contentedness with family life at home. Older people taken may be all that is needed at this stage. For older people with
usually left full-time education at 14 or 15 and one should be care- long-term functional illnesses, the effects of their chronic illness
ful about inferring intelligence levels from such information or may now be indistinguishable from their personality, but developing
from patients’ own comments about their educational attainments. an awareness of personality traits is still important in understand-
The subsequent history of further training and their occupational ing their behaviour and relationships. For those with a more recent
record gives a surer indication of their ability. When eliciting this illness, especially a dementia, the premorbid personality moulds
information it is important to try to clarify the degree of autonomy the presentation of the illness. Someone who has lived independ-
and responsibility the patient enjoyed. Being told someone worked ently all his life, has strong opinions, and has always had things his
in a factory is of limited value: was this packing boxes, as a clerk, or own way is unlikely to settle quietly into a nursing home! A referral
as a production manager? When, as is frequently the case, someone from a ward or a residential home for ‘aggressive behaviour’ in such
held many jobs over the decades of their working life, it is necessary a person needs to be interpreted in this context of their personality,
to focus on their longest employments and any jobs that had special rather than lead to a prescription for antipsychotic medication.
importance. Comments made about a marriage or relationships
may raise this as a natural topic for discussion, but if not then it is Social history
important to elicit a marital history, again focusing on the nature The social history follows smoothly from the personal history,
of the relationship, enjoyed (or not), with the spouse and any chil- bringing it up to date. The pattern of the patient’s everyday life, his
dren and grandchildren. Where there have been several marriages or her activities, and relationships with family members and others
and sexual relationships, then matters may be very sensitive and it should be clarified and confirmed. The amount of care currently
may be appropriate to obtain only the broad outline about these. given to the patient, by family, friends, and formal carers, needs
Another area requiring tactful enquiry is the current sexual activity to be elicited and, in addition to information from the patient and
and degree of satisfaction with this and in the marital relationship informant(s), a written care plan may be available that can provide
in general. Some couples may find such questions inappropriate, evidence of day-to-day issues as well as of the programme of care.
but important information about the strength of the marriage may Other health professionals may be involved, e.g. a district nurse,
be revealed that impinge on the caregiving role. Bereavement and and the patient may attend day care or have a respite care pro-
other loss events are obviously much more frequently experienced gramme. Enquiring about the pattern of daily activity can bring out
by older people. Most cope very well, recognizing this as an ines- problems such as apathy or resistiveness, as well as providing fur-
capable part of growing old, but the loss of children, even when they ther evidence about the extent of current difficulties. Having estab-
themselves may have become old, is a severe blow for many and lished an understanding of the pattern of the patient’s living and
such losses often precipitate a search for help when other bereave- current support, the social history may conclude with more direct
ments have been borne well. Again, most cope well with serious questions about the use of alcohol and other substances. The abuse
illnesses, probably again because they are regarded as inevitable, of alcohol and benzodiazepines remains the major foci of concern,
but some disabling illnesses do cause depressive reactions. Some but increasingly old age psychiatrists will see people abusing other
people, especially men, find adapting to retirement difficult and substances and should be alert to this possibility.
this can create stress at home and consequent marital difficulty, and In recent years, the question of driving motor vehicles has
discreet questions about adapting to this major change are appro- become a prominent issue as an increasing number of older peo-
priate in those close to retirement. Another event that frequently ple, including women who used rarely to drive, own and drive their
own vehicles (Brown and Ott, 2004). If the patient is driving, the observing the patient at rest for tremor and bradykinesia, examining
clinician may ask others about any concerns they may have or inci- the arms for rigidity, and asking the patient to take a short walk to
dents that have occurred, and will need to consider whether to ask watch for gait changes and postural instability. At the same time the
the patient to stop driving if it appears he or she cannot safely drive examiner should look for focal neurological signs, especially those
any longer. Assessment clinically is difficult because, whilst several that may be due to stroke disease. Such an examination should not
neuropsychological measures, especially visuospatial and atten- replace a more detailed physical examination (see Chapter 11) and
tional tests, correlate well with driving performance (Adler et al., is necessarily limited at an initial assessment in someone’s home,
2005), such measures are not easily available and also do not pro- but it frequently adds important diagnostic information.
vide definitive evidence. Hence this review recommended a driv- Apathy is highly prevalent even in early dementia (Mega et al.,
ing assessment for all patients with mini-mental state examination 1996) and often mistaken for depression (Levy et al., 1998). The
(MMSE) scores of less than 24 or where concern exists (Adler et al., listless and disengaged presentation of someone with apathy can
2005). These issues are discussed in detail in Chapter 62. usually be distinguished from the withdrawn and retarded pic-
It is also important to ask some general questions about peo- ture in depression, although it can be difficult, especially in more
ple’s property and financial arrangements, as these will influence severely impaired patients, and both may be present. At interview
whether theya are at risk of exploitation and also the choices that someone with apathy may appear uninterested and switched off
they may have in the future. For example, ask whether they manage when the clinician is discussing the situation with an informant,
their finances themselves or whether they have help and whether but then warm up and engage well when directly addressed and
they own their house or rent it? Have they made any provisions show a reactive mood. On questioning why he has given up certain
for the future, such as making a Will or establishing a Lasting activities the apathetic patient will intimate he no longer has the
Power of Attorney? And are they in receipt of any benefits, such as drive to do them and in fact still enjoys visits from friends and fam-
Attendance Allowance? ily, whereas someone with depression will explain he does not enjoy
them any more, or at least not as he used to (anhedonia). Whilst
Mental state examination ‘frontal dementias’ may be accompanied by an apathetic picture,
It is important to remember that the mental state examination is a disinhibited presentation is also well recognized and much less
an assessment of the mental state of the patient at the time of the likely to be due to a bipolar illness than in younger adults. Subtler
interview, and so symptoms identified in the history, e.g. halluci- aspects of disinhibited behaviour, such as overfamiliarity, may be
nations, may not be manifest for recording as part of the patient’s difficult to distinguish from the normal range of social interaction,
current mental state. The mental state assessment begins on arrival but more overt behaviours, such as coarse joking, sometimes occur,
and continues throughout the interview and should be recorded in although even in such circumstances it is wise to consider whether
the standard way. this may be an aspect of the patient’s premorbid personality. In all
cases it is prudent to observe the reaction of relatives and to gently
Appearance and behaviour enquire about whether such comments or interactions represent a
Generally speaking, older people have maintained more formal change, before regarding them as pathological.
modes of dress and behaviour. Although this is changing, it means
the clinician may be informed that an apparently well-groomed Speech
man has slipped in his standards. It is a delicate matter to enquire Occasionally when dealing with older people the clinician will
directly about issues of dress and hygiene, but it is appropriate to encounter someone who is loud and garrulous, consistent with
gently ask where there is an apparent discrepancy between the FTD or mania, but quiet and impoverished speech is much more
patient’s state of grooming and that of the spouse or the surround- common, being a feature of the major degenerative dementias and
ings in the home. Another important element to consider in assess- depression. Dysphasia is another key feature of early dementing ill-
ing older people is the presence of sensory impairment. The severity nesses and one that does not occur in depression. Thus whilst pov-
of any deafness and blindness should be noted because of both the erty of speech does not in itself help to distinguish depression from
influence it has on the assessment process, especially on cognitive dementia, the presence or absence of dysphasia does. However,
testing, and the importance in ensuring that handicaps related to deafness is a frequent problem and at interview it can be very dif-
these are addressed during the management of the patient’s illness. ficult to distinguish whether an apparent failure to understand is
A brief assessment of general health, including changes in weight because of receptive dysphasia or deafness. In milder dementia,
and pallor, and the level of alertness are also important. subtle difficulties in speech structure occur, especially in finding
The clinician should be looking for the range of behavioural names, but these can be difficult to detect at interview because
changes manifest in functional illnesses, e.g. poor eye contact in patients can be adept at covering up these problems through the use
depression or suspiciousness in paranoid schizophrenia. In old age of circumlocutions. Whilst instruments exist for formally assessing
psychiatry, psychomotor changes can be especially prominent in dysphasia (see Chapter 10 on cognitive assessment) these are not
the affective disorders, with agitation a common feature in depres- well suited to regular clinical practice.
sion. When a patient exhibits apparent psychomotor retardation
it is important to consider whether this may be apathy or related Mood
to Parkinson’s disease (bradykinesia) rather than a depressive ill- Abnormalities of mood follow the same pattern as in disorders of
ness. The high prevalence of parkinsonism in older people and the younger adults. Lowering of mood in a depressive illness is often
diagnostic importance of this in dementia, as a hallmark feature of accompanied by a more prominent anxiety than in earlier life, with
DLB, means it is good practice to assess for parkinsonism in every other associated anxiety symptoms, and this can be misleading;
new patient. This can be done briefly in most settings by carefully an anxiety presentation in later life, especially in someone with
previous psychiatric history, should make one enquire carefully for leading to them engaging this stranger in conversation or becom-
a depression. The reluctance of many older people to express their ing angry at their presence. People seen on television are believed
feelings is better recognized and, although probably changing in the to be really present in the room with the patient in the ‘TV sign’,
young-old compared with the old-old, the clinician still needs to be and in the phantom boarder syndrome the patient believes extra
aware of this phenomenon. Elevation of mood in a manic illness people are living in the house (perhaps after failing to recognize
tends to be attenuated, like the rest of the illness, in later life and relatives). More sophisticated and complex delusions are unusual
also occurs in frontal dementias, and where present in dementia it in dementia but remain typical of schizophrenia in either its early
may be indicative of a more severe illness, especially when associ- or late onset forms.
ated with agitation. A feature of mood disturbance more specific to
old age psychiatry is the mood lability that occurs with cerebrovas- Perception
cular disease (Morris et al., 1993). Stroke disease, but also less obvi- Psychiatrists working with older people encounter patients with
ous cerebrovascular disease, is frequently accompanied by mood auditory hallucinations, and sometimes olfactory hallucinations,
abnormalities and, whether the prevailing change is an elevation or typical of schizophrenia, but more frequently they encounter
a depression of mood, it is often highly labile. Apparently spontane- prominent and persistent hallucinations in other sensory modali-
ous, but typically short-lived, episodes of weeping without any sus- ties. This is a well-recognized consequence of the ‘organic’ nature
tained lowering of mood should alert the clinician to the possibility of the illnesses that present to old age psychiatry. Visual and tac-
that this may be emotional lability related to stroke disease rather tile hallucinations are frequently seen in patients with delirium in
than a mood disorder as such (House et al., 1989). the general hospital but also in less overt delirium during home
assessments. Complex and enduring visual hallucinations, char-
Thought acteristically of people and animals, are a core feature of DLB
With increasing age, people are more likely to spend time review- (McKeith et al., 2005; Mosimann et al., 2006) (see Chapter 35),
ing their past life and thinking about its end, and this seems to be a but it is always important to consider whether they may be due to
normal phenomenon. If someone is depressed, they may well start a delirium. Often it can be difficult to be sure whether the visual
to feel that life is no longer worth living or that the future holds hallucination is truly occurring in the absence of a real stimulus or
little for them. They may even wish that their life would come to whether it is a misperception resulting from poor eyesight or some
end or that they might not wake up from their sleep one morn- other defect in visual processing. However, since such problems in
ing. Among people in their 90s it is not uncommon to hear them visual processing commonly occur in DLB, the presence of such
say that they feel they have lived too long. More active suicidal phenomena suggests such a diagnosis, even if frank hallucinations
thoughts, e.g. I wish that I was dead, or thinking about how one cannot be confirmed. Sometimes, persistent, complex visual hal-
might kill oneself, are less common and are more likely to occur in lucinations are the only phenomenological feature confirmed dur-
more severe depression. In the most extreme cases, a person will ing assessment; in particular, there may be no significant cognitive
be actively seeking to harm or kill him- or herself. In the interview, impairment or alteration in consciousness. Such patients are said
therefore, it is important to assess this range of thoughts in a sensi- to have Charles Bonnet syndrome and although many will proceed
tive but thorough manner. A sequence of questions starting with to decline cognitively and develop a dementia and other features of
‘How depressed do you get?’ is a useful approach. ‘Do you cry a DLB, this does not appear to happen to everyone presenting in this
lot?’ ‘Are there times when you feel hopeless or that life is not worth way. Most old age psychiatrists are familiar with a similar phenom-
living?’ ‘Are there ever times when you want your life to come to an enon of persistent auditory hallucinations in clear consciousness
end?’ ‘Do you ever think of harming yourself or doing away with and in the absence of cognitive decline (an auditory hallucinosis,
yourself?’ ‘Do you have any plans to kill yourself?’ It is useful to sometimes called auditory Charles Bonnet syndrome). Typically
note that in depressed older people, thoughts of wanting their life such patients experience musical hallucinations for hours on
to end are common but active suicidal ideation is much less so. end, hearing choirs singing hymns or bands playing old songs,
However, when it is present it is extremely serious and should be and show good insight, often giving detailed descriptions of their
responded to. hallucinations. Musical hallucinations are often association with
Abnormalities in the form of thought occur in schizophrenia, sensorineural deafness, which suggests that the phenomenon may
depression, and bipolar disorder as in younger people, but it is rare arise as a form of tinnitus that is organized into a more complex
to encounter formal thought disorder in late-onset schizophrenia. perception by the cerebral cortex. Neither the visual nor auditory
Incoherence of thought is, of course, common in dementing ill- variants of Charles Bonnet syndrome do well with antipsychotic
nesses, and can occur early on but is more common in advanced treatment (Batra et al., 1997).
disease.
Paranoid delusions, especially of theft, are extremely common Cognition
in dementia and can be difficult to distinguish from the effects Formal cognitive assessment is dealt with in detail in Chapter
of amnesia and from related confabulation (Burns et al., 1990). 10. However, the clinician should have been gleaning cogni-
Misidentification delusions are characteristic of dementia, and tive information throughout the assessment. A structured brief
whilst the Capgras syndrome (more accurately a symptom) is the instrument for assessing cognition should be used as part of
most well-known, there are several variants (Harwood et al., 1999). the cognitive assessment, and although many are available, the
All appear to be related to a failure to recognize or correctly iden- MMSE (Folstein et al., 1975) appears to remain the favourite of
tify an image. In the Capgras syndrome the patient believes a close most clinicians. Such instruments can provide an estimate of
relative has been substituted by an exact double. The ‘mirror sign’ the extent of cognitive impairment and through repeated use
occurs when patients fail to recognize their image in the mirror, over time enable the clinician to monitor illness progression or
response to antidementia treatment; patients ‘typically’ decline Physical examination

at 2–3 points per year on the MMSE (Salmon et al., 1990), but Even if assessing someone at home, a brief physical assessment is
in clinical practice immense variability between patients is appropriate. In those who are found to be physically unwell, per-
observed. Whilst dysphasia can be difficult to assess formally dur- haps delirious, this will need to include hydration status and cardi-
ing a standard clinical assessment it can frequently be observed ovascular, respiratory, and abdominal assessments prior to referral
during the process of history and mental state assessment. More to colleagues at the general hospital. In routine practice the need
obviously, inconsistencies and gaps in the patient’s account pro- is for a neurological examination, especially to look for features of
vide evidence of both the presence and the severity of amnesia. stroke disease and parkinsonism. Power, tone, coordination, basic
Other information on orientation, cognition, and perhaps spa- sensation, gait, and most cranial nerves can all be examined with-
tial or executive dysfunction may also be detected. Whilst this out the need for special equipment. Such an examination does not
information does not replace proper cognitive testing, it provides replace the need for a full assessment with related physical inves-
important additional evidence. tigations later but does provide valuable evidence to clarify the
potential causes of a dementia.
Insight
Insight is a complex, variable, and multidimensional phenom- The informant interview
enon (Howorth and Saper, 2003). In patients with ‘functional’
Obtaining a history from a family member, friend, or carer who
disorders the degree of insight varies in much the same way as in
knows the patient helps fill out the history and ensure that all prob-
younger adults. It is good practice to record a brief description of
lems have been identified and understood. In those with cognitive
the extent of insight, rather than a summary phrase; for example,
impairment it is essential because the patient is unable to provide
to say ‘she understands she has a depressive illness, of which her
reliable information. A decision about how much useful informa-
anxiety and tremor are manifestations, and is willing to take anti-
tion can be garnered from the patient can usually be made early in
depressant medication for this’ rather than simply ‘full insight’. In
the interview. Those with early dementia or other cognitive impair-
people with moderate to severe dementia there is usually very lit-
ment can typically give a reasonable account of their personal his-
tle insight, just an awareness perhaps that one is not right and an
tory until the recent past, and some information on major medical
acceptance that help is needed. However, patients may have more
or psychiatric illnesses as well, but for the more impaired an inform-
awareness of their illness than they are able to express verbally,
ant is needed for this as well and the interview with the patient
perhaps due to better preservation of implicit memory (Howorth
focuses on the mental state, especially the cognitive assessment.
and Saper, 2003). Such insight appears to be greater for the amne-
Whilst most informants are reliable historians who desire appro-
sia than for the impact of the cognitive decline on their function
priate treatment and help for the patient, this is not always the case
and perhaps least for the impact of their illness on other people.
and the assessing clinician should be alert to other possibilities.
In those with milder dementia there is typically more insight into
In some cases the informant is cognitively impaired themselves
the illness but the extent to which this is acknowledged varies
and on other occasions does not have sufficient knowledge of the
considerably. Often the clinician suspects the patient and family
patient. Occasionally conflict within the family is revealed by dif-
are colluding in denying the severity of problems because they are
ferent opinions being offered and an informant may not be a disin-
aware of the implications. There has been a trend recently towards
terested party but one with his or her own views about what should
disclosure of the diagnosis of dementia and whilst such openness
be done, e.g. a patient kept inappropriately at home so benefits and
is usually appropriate this may not always be the case, especially
allowances continue to be available.
when the family appears to prefer denial. When prescription of
antidementia treatment is indicated the issue of diagnosis is espe-
cially acute and difficult to avoid and sensitive handling of the Further Assessment
issue is needed. Following the initial assessment, the psychiatrist is usually in a posi-
tion to make a differential diagnosis and plan further assessments to
Assessment of capacity clarify or confirm this and to deal with other important issues that
Capacity and decision-making have become increasingly promi- have arisen. Inpatient admission and assessment is only necessary
nent features of old age psychiatry practice and capacity assessment in a small minority of cases, usually where there are severe behav-
and the relevant legislation is dealt with in detail in Chapter 63. ioural disturbances or significant immediate risks. Such admission
Key principles of the Mental Capacity Act 2005 (for England and provides for a thorough assessment but one that is limited by it
Wales) are that capacity is specific to the time of the assessment and not taking place in the home environment. More frequent is fur-
that it is functional in nature (capacity varies by the function being ther assessment through attendance at a day unit. This is a suitable
assessed). A formal assessment of capacity is not part of routine setting for coordinating a full physical assessment and associated
clinical assessment and indeed, for these reasons, strictly speak- physical investigations and one that allows nursing staff to provide
ing, it cannot be made unless it is requested because the clinician longer-term assessment of the patient’s interaction with other peo-
needs to know what the patient is being assessed for at any point in ple and to observe for significant psychopathology. In most cases,
time. When making an assessment the clinician needs to determine further assessment can be satisfactorily carried out at home, with
whether the four aspects of capacity (sufficient information has visits from other relevant professionals as required. For example, in
been conveyed; this information can be retained for long enough patients with borderline or mild cognitive impairments, suggesting
to make a decision; an ability to weigh matters up is present; a deci- they may have an early dementia, a more detailed neuropsychologi-
sion can be communicated to others) are fulfilled for each particu- cal assessment by a psychologist is helpful to clarify the pattern and
lar decision-making issue. extent of any deficits. Where difficulties in everyday functioning
have been identified or are likely to be present, it is appropriate to Benbow, S. (1990). The community clinic—its advantages and disadvantages.
refer to an occupational therapist for a detailed functional assess- International Journal of Geriatric Psychiatry, 5, 119–21.
ment. In such circumstances, and where financial issues arise and Brown, L. B. and Ott, B. R. (2004). Driving and dementia: a review of the
literature. Journal of Geriatric Psychiatry and Neurology, 17, 232–40.
care plans may need adjustment, the involvement of a social worker
Bucks, R. S., et al. (1996). Assessment of activities of daily living in dementia:
specializing in dealing with older people with mental illness is
development of the Bristol Activities of Daily Living Scale. Age and
also necessary. Community psychiatric nurses are able to monitor Ageing, 25, 113–20.
changes in mental state in response to treatment and to deal with Burns, A., Jacoby, R., and Levy, R. (1990). Psychiatric phenomena in
carer stress issues and related problems that may be present in their Alzheimer’s disease. I: Disorders of thought content. British Journal of
relationship with the patient. Psychiatry, 157, 72–6, 92–4.
Finkel, S. I. and Burns, A. (2000). Behavioural and psychological symptoms
of dementia (BPSD): a clinical and research update: introduction.
Principles of Management International Psychogeriatrics, 12, 9–12.
Management in old age psychiatry settings should develop seam- Folstein, M. F., Folstein, S. E., and McHugh, P. R. (1975). ‘Mini-mental state’.
lessly out of the assessment process. The initial clinical assessment A practical method for grading the cognitive state of patients for the
by the old age psychiatrist leads to the identification of the main clinician. Journal of Psychiatric Research, 12, 189–98.
needs of the patient and key carers. The psychiatrist will initiate, Frankel, S., Farrow, A., and West, R. (1989). Non-admission or non-invitation?
A case-control study of failed admissions. British Medical Journal, 299,
titrate, and monitor drug treatments targeted at key symptoms.
598–600.
When the initial assessment leads to a diagnosis of a functional ill-
Hardy-Thompson, C., Orrell, M. W., and Bergmann, K. (1992). Evaluating a
ness, the appropriate pharmacological treatment can be initiated psychogeriatric domiciliary visit service: views of general practitioners.
immediately and consideration given to the need for any additional British Medical Journal, 304, 421–2.
psychosocial intervention. These treatments are broadly the same Harwood, D. G., et al. (1999). Prevalence and correlates of Capgras syndrome
as for younger adults and their application to older people is disin Alzheimer’s disease. International Journal of Geriatric Psychiatry, 14,
cussed in later chapters (see Chapters 16–21). 415–20.
When dementia is diagnosed, cholinesterase inhibitors or Health, D. O. (2009). Living well with dementia: a national dementia strategy.
memantine may be commenced and, for both ‘functional’ and Department of Health, London.
dementia illnesses, the psychiatrist will engage other members of House, A., et al. (1989). Emotionalism after stroke. British Medical Journal,
298, 991–4.
the multidisciplinary team to conduct more detailed assessments
Howorth, P. and Saper, J. (2003). The dimensions of insight in people with
in specific areas. These assessments in turn clarify needs and lead dementia. Aging and Mental Health, 7, 113–22.
to focused interventions. The psychiatrist has a coordinating and Jones, S. J., Turner, R. J., and Grant, J. E. (1987). Assessing patients in their
consultative role in managing the overall strategy. In more straight- homes. Bulletin of the Royal College of Psychiatry, 11, 117–19.
forward cases the different individuals involved gradually withdraw Levy, M. L., et al. (1998). Apathy is not depression. Journal of Neuropsychiatry
as problems are solved, symptoms improve, and the patient settles and Clinical Neurosciences, 10, 314–9.
in to a stable existence again. In people with dementia, such a meet- Lindesay, J., et al. (2002). The second Leicester survey of memory clinics in
ing is usually followed by a period in which the agreed manage- the British Isles. International Journal of Geriatric Psychiatry, 17, 41–7.
ment plan is enacted and, if successful, all settles down again. In Luce, A., et al. (2001). How do memory clinics compare with traditional old
those who have severe functional illnesses, long-term follow-up by age psychiatry services? International Journal of Geriatric Psychiatry, 16,
837–45.
the psychiatrist and/or community nurse is usually the outcome,
McKeith, I. G., et al. (2005). Diagnosis and management of dementia with Lewy
rather than discharge from the service. In the most resistant and bodies: third report of the DLB Consortium. Neurology, 65, 1863–72.
severe illnesses, closer long-term follow-up may be needed by reg- McKhann, G. M., et al. (2011). The diagnosis of dementia due to
ular attendance at a day unit, e.g. to monitor psychosis in people Alzheimer’s disease: recommendations from the National Institute on
on clozapine or mood in those on combinations of antidepressants Aging-Alzheimer’s Association workgroups on diagnostic guidelines for
and mood stabilizers. Throughout the assessment and management Alzheimer’s disease. Alzheimer’s and Dementia, 7, 263–9.
process the psychiatrist has an essential leading role using his or her Mega, M. S., et al. (1996). The spectrum of behavioral changes in Alzheimer’s
expertise to make key decisions, supervise other team members, disease. Neurology, 46, 130–5.
and ensure the patient’s care is optimally and ethically delivered. Morris, P. L., Robinson, R. G., and Raphael, B. (1993). Emotional lability after
Management of dementia is discussed in more detail in Chapter 39, stroke. Australian and New Zealand Journal of Psychiatry, 27, 601–5.
Mosimann, U. P., et al. (2006). Characteristics of visual hallucinations in
and for other conditions in the specific relevant chapters.
Parkinson disease dementia and dementia with Lewy bodies. American
Journal of Geriatric Psychiatry, 14, 153–60.
References Neary, D., et al. (1998). Frontotemporal lobar degeneration: a consensus on
Adler, G., Rottunda, S, and Dysken, M. (2005). The older driver with clinical diagnostic criteria. Neurology, 51, 1546–54.
dementia: an updated literature review. Journal of Safety Research, 36, NICE (2006). Dementia: supporting people with dementia and their carers in
399–407. health and social care. Clinical Guidelines CG 42. NICE, London.
Anderson, D. N, and Aquilina, C. (2002). Domiciliary clinics I: effects on Parker, G., et al. (2000). Best place of care for older people after acute and
non-attendance. International Journal of Geriatric Psychiatry, 17, 941–4. during subacute illness: a systematic review. Journal of Health Services
Bamford, C., et al. (2004). Disclosing a diagnosis of dementia: a systematic and Research Policy, 5, 176–89.
review. International Journal of Geriatric Psychiatry, 19, 151–69. Salmon, D. P., et al. (1990). Longitudinal evaluation of dementia of
Batra, A., Bartels, M., and Wormstall, H. (1997). Therapeutic options in the Alzheimer type: a comparison of 3 standardized mental status
Charles Bonnet syndrome. Acta Psychiatrica Scandinavica, 96, 129–33. examinations. Neurology, 40, 1225–30.
CHAPTER 10
Clinical cognitive assessment
Christopher Kipps and John Hodges
Cognitive symptoms arise from the location of brain dysfunction from the patient is minimal. It allows a chance to assess both lan-
and are not linked directly to any particular pathology. In the early guage and cooperation without assistance or interruption from
stages of disease, symptoms may be nonspecific, and while certain the partner. Disparities between the two accounts are important
symptom clusters are commonly seen in particular disorders, atypi- as insight is often poor. A family history and risk factors, nota-
cal presentations are not infrequent. For example, in Alzheimer’s bly vascular, are particularly relevant, and should be specifically
disease, patients may present with a focal language syndrome enquired about. Considerable, and sometimes repeated, probing
instead of the more commonly appreciated autobiographical mem- may be needed, as a history of suicide or alcohol dependency in
ory disturbance, despite identical pathology. In our approach to the close relatives may disguise an unrecognized early onset demen-
cognitive assessment, we maintain a symptom-oriented approach. tia. Concomitant illness and medication use frequently underlie, or
This facilitates the localization of pathology and subsequent clinical complicate, cognitive complaints. We use a questionnaire filled out
diagnosis, which may then be supplemented by associated neuro- before the consultation, which saves time and helps draw attention
logical signs, imaging, or other investigations. to issues in the background history.
The purpose of the cognitive examination is to separate out those
patients in whom a firm clinical diagnosis can be made from those
in whom additional investigation is required. In this assessment, Orientation and Attention
the history and clinical examination are completely intertwined, Alertness and cooperation with the assessment should be noted,
and the ability to respond to conversational clues or provide details as these factors may impact on the subsequent findings. The level
of personal events is as much a consideration as any formal test- of alertness is an important clue to the presence of a delirium or
ing. Skilful examiners often weave their assessment into a relaxed the effects of medication. Delirium may be marked by both rest-
conversation with a patient, making it more enjoyable for both. lessness and distractibility, or the patient may be quiet, and drift
In neurodegenerative diseases in particular, poor memory and off to sleep easily during the consultation. If there is any concern
impaired insight make the perspective of an informant, who knows about the level of alertness of the patient, review of the medica-
the patient well, essential. tion list is often helpful. It may be misleading, and is frequently
A clear focus is needed from early in the consultation in order hopeless, to perform a detailed cognitive assessment on a patient
to direct the assessment to the areas of greatest relevance that need with diminished alertness. If that is the case, documentation of
specific and more detailed examination. A list of particularly disorientation and attention may be as much as can be achieved
criminating questions can be seen in Table 10.1. In the sections that initially.
follow, we divide the cognitive examination into a number of broad
domains: attention and orientation, memory, language, executive Orientation
function, apraxia, visuospatial ability, and behavioural.
Orientation is usually assessed to time, place, and person; it is not
particularly sensitive, and intact orientation does not exclude a
General significant memory disorder, particularly if there is concern about
We start by establishing a picture of premorbid functioning (e.g. memory from an informant.
education, employment, significant relationships). Learning a little Time orientation is the most helpful, and should include the
about the patient’s interests or hobbies allows one to tailor ques- time of day. Many normal people do not know the exact date, and
tions in the cognitive examination more precisely. The onset and being out by two days or less is considered normal when scoring
time course of the deterioration are as important as the cluster of this formally. Time intervals are often poorly monitored by patients
deficits, be they memory, language, visual function, behaviour, or with delirium, moderate to severe dementia, and in the amnesic
indeed psychiatric, and often the first noted deficit has diagnostic syndrome, and are easily tested by asking about the length of time
relevance. It is frequently very helpful to ask patients directly what spent in hospital.
they believe the reason is for their attendance in the clinic. Place should be confirmed, and asking what the name of the
We try to interview both the patient and informant independ- building is (e.g. outpatient clinic), rather than the name of the
ently, even when the amount of information likely to be obtained hospital, often produces a surprising lack of awareness of location.
Table 10.1 Discriminating features in common neurodegenerative diseases

Disease Key questions to ask: Clinical signs/investigations
Is there . . . ? Do they . . . ?
Alzheimer’s disease Repetition of the same question with rapid forgetting of the answer MRI: generalized atrophy, particularly medial
Get lost on familiar routes temporal and hippocampal regions
CSF biomarkers
Amyloid-PET imaging
Frontotemporal dementia (FTD) Loss of empathy MRI: frontal and/or right predominant temporal
—behavioural variant (bv) Eating changes lobe atrophy
Stereotypical behaviours Beware phenocopy bvFTD with normal imaging
Social inappropriateness
Disinhibition
Lack of insight
—semantic dementia Difficulty with naming objects MRI: anterior temporal lobe atrophy (usually left
An inability to understand words more than right)
—progressive nonfluent aphasia Mispronunciation of words with insertion of repeated or incorrect

syllables
—logopenic aphasia Slow speech and difficulty with naming objects
Longer sentences are not understood
Frontotemporal dementia with motor Rapid progression of a behavioural syndrome and/or language deficits Fasciculations (limb or tongue)
neuron disease (FTD-MND) —prominent delusions and/or hallucinations (transient) EMG
Progressive supranuclear palsy (PSP) A history of falls Axial rigidity
Profound apathy Vertical gaze palsy
Slowed movement and thinking Loss of postural reflexes
Lewy body disease Visual hallucinations Mild parkinsonism
Fluctuating cognitive ability Visuospatial dysfunction
Falls
REM sleep disorder
Corticobasal degeneration Difficulty using objects with one hand Limb apraxia
Limb having ‘a mind of its own’ Visuospatial dysfunction
Shock-like jerks Parkinsonism
May develop nonfluent language syndrome
Huntington’s disease Family history (autosomal dominant) Generalized chorea
Huntington’s gene
Vascular dementia Vascular risk factors Pyramidal signs
May have stepwise decline, but frequently do not Evidence of small and/or large vessel disease on
brain imaging
CJD Rapid cognitive decline Ataxia, tremor, myoclonus
Movement disorder MRI abnormalities on diffusion imaging
Neurosurgical procedure or graft
vCJD Rapid cognitive decline MRI—pulvinar sign
Sensory disturbance
Movement disorder
Since there are often visual and contextual cues present, this is less In the aphasic patient, earlier conversation should have revealed the
sensitive than orientation to time, and is particularly the case when true deficit, but a mistaken label of ‘confusion’ is frequently applied
assessing someone at home. because such patients either fail to comprehend the question or
Person orientation includes name, age, and date of birth. produce the wrong answer. Given a choice, they can usually pick
Disorientation to name is usually only seen in psychogenic amnesia. out their own name.
CHAPTER 10 clinical cognitive assessment 151
Attention often used. If care is not taken to ensure proper registration of the
Attention can be tested in a number of ways including serial 7s, digit items at the start of this test, the results may be confusing or mis-
span, spelling WORLD backwards, and recitation of the months of leading. Poor registration, usually a feature of poor attention or
the year in reverse order. Although serial 7s is commonly used, it executive dysfunction, may invalidate the results of recall or rec-
is frequently performed incorrectly by older people, as well as by ognition that test episodic memory. Free recall is harder than the
patients with impaired attention. Reverse-order months of the year recognition of an item from a list. Testing in the hearing impaired
is a highly overlearned sequence, and we prefer it as a measure of poses particular challenges but can be tested verbally by the use of
sustained attention. written instructions, in large print, after handing the patient his or
Digit span is a relatively pure test of attention and is dependent her spectacles.
on working memory, but it is not specific and can be impaired in Anterograde nonverbal memory can be assessed by asking a sub-
delirium, focal left frontal damage, aphasia, and moderate to severe ject to copy and later recall geometric shapes. Alternatively, it is
dementia. It should, however, be normal in the amnesic syndrome possible to hide several objects around the room at random, and
(e.g. Korsakoff ’s syndrome and medial temporal lobe damage). Start ask the patient to search for them several minutes later. This is
with three digits and ensure that they are spoken individually and an easy task, and inability to perform well is a convincing sign of
not clumped together in the way that one might recite a telephone memory impairment.
number (e.g. 3–7–2–5 and not 37–25, etc.). Normal digit span is Famous events, recent sporting results, or the names of recent
6 ± 1, depending on age and general intellectual ability. In older people prime ministers can all be used to test retrograde memory with-
or the intellectually impaired, 5 can be considered normal. Reverse out an informant. More remote autobiographical memory assess-
span is usually one less than forward span. In performing this test, it ment needs corroboration, and may be relatively preserved in early
is helpful to write out the numbers to be used before starting. Alzheimer’s disease. Autobiographical ‘lacunes’, where discrete
periods of time or events are forgotten, are a characteristic feature
of transient epileptic amnesia (TEA).
Memory Memory loss and learning impairment out of proportion to
Complaints about poor memory are the most frequent reason for other cognitive disturbance is known as the amnesic syndrome.
referral to a cognitive disorders clinic and often provide a good Generally, both anterograde and retrograde memory loss occur in
starting point for the consultation, despite not being very specific. parallel, such as in Alzheimer’s disease and head injury, but disso-
A useful framework for analysing memory complaints divides ciations occur. Disproportionately severe anterograde amnesia may
memory into several separate domains. Episodic memory (person- be seen when there is hippocampal damage, particularly in her-
ally experienced events) comprises anterograde (newly encountered pes simplex encephalitis, focal temporal lobe tumours, and infarc-
information) and retrograde (past events) components, and depends tion. Confabulation, for example in Korsakoff ’s syndrome, might
on the hippocampal-diencephalic system. A second important sys- be grandiose or delusional, but more often involves the misorder-
tem involves memory for word meaning and general knowledge ing and fusion of real memories that end up being retrieved out of
(semantic memory), the key neural substrate being the anterior context. A transient amnesic syndrome with marked anterograde,
temporal lobe. Working memory refers to the very limited capac- and variable retrograde, amnesia is seen in transient global amne-
ity that allows us to retain information for a few seconds, and uses sia (TGA), whilst ‘memory lacunes’, and repeated brief episodes of
the dorsolateral prefrontal cortex. The term ‘short-term’ memory is memory loss suggest TEA.
applied, confusingly, to a number of different memory problems, Simply asking both patient and informant to give an overall
but has no convincing anatomical or psychological correlate. memory rating (out of 10) is often helpful. It is seldom, if ever, that
truly amnestic patients will give themselves scores such as 0 or 1,
Episodic memory although their spouse might. The reverse is often true of those who
Anterograde memory loss is suggested by the following: forget primarily because of anxiety or depression.
◆ forgetting recent personal and family events (appointments, Working memory

social occasions) Lapses in concentration and attention (losing your train of thought,
◆ losing items around the home wandering into a room and forgetting the reason, forgetting a phone
◆ repetitive questioning number that has just been looked at) are common and increase
with age, depression, and anxiety. Such symptoms are much more
◆ inability to following and/or remember plots of movies, televi- evident to patients than to family members and, in isolation, are
sion programmes, details of current affairs usually not of great concern. It should be noted, however, that basal
◆ deterioration of message-taking skills ganglia and white matter diseases may present with predominantly
working memory deficits.
◆ increasing reliance on lists.
Retrograde memory loss is suggested by the following: Semantic memory
◆ memory of past events (jobs, past homes, major news items) Patients with semantic breakdown typically complain of ‘loss of
memory for words’. Vocabulary diminishes and patients substitute
◆ getting lost, with poor topographical sense (route-finding). words like ‘thing’. There is a parallel impairment in appreciating
Specific questions about the route taken to the hospital or recent the meaning of individual words which first involves infrequent or
events on the ward can be tested directly during conversation. unusual words. Word-finding difficulty is common in both anxi-
Recalling a name and address or the names of three items is also ety and ageing, but it is variable and not associated with impaired
comprehension. This is in stark contrast to the anomia (impaired semantic memory impairment seen in semantic dementia. Posterior
naming) in semantic dementia which is relentlessly progressive lesions, particularly of the angular gyrus, can produce quite pro-
and associated with both object agnosia (inability to recognize nounced anomia for visually recognized objects, and may be asso-
objects) and atrophy of the anterior temporal lobe, usually on the ciated with alexia.
left. Testing of language function is described in detail in the next
section. Comprehension
Difficulty with comprehension is often (incorrectly) assumed to be
Language a result of hearing impairment. Complaints of difficulty using the
The majority of language deficits are usually revealed within the telephone or withdrawal from group conversations may be more
first few minutes of listening to the patient speak, particularly subtle clues to its presence. It is useful to assess comprehension
where poor fluency, prosody, agrammatism, and articulation are in a graded manner, starting with simple and then more complex
involved. Evidence of word-finding impairments and parapha- instructions.
sic errors are also usually quickly apparent. Documenting several Use several common items (coin, key, pen) and ask the patient to
examples of these errors is often quite helpful to subsequent cli- point to each one in turn in order to assess single word comprehen-
nicians. Sometimes, however, a relatively fluent history may mask sion. There is a frequency effect, and if this test seems too easy, try
quite significant naming and single-word comprehension defi- harder items around the room. We also test comprehension by ask-
cits, and it is important to assess this routinely with infrequently ing patients to define the meaning of words such as hippopotamus,
encountered words and with directed questioning. caterpillar, encyclopedia, emerald, and perimeter, which is done at
the same time as assessing repetition (see below).
Naming Sentence comprehension can be tested with several common items
in order to devise syntactically complex commands. For example:
The degree of anomia is useful as an overall index of the sever-
touch the pen, and then the watch, followed by more difficult sen-
ity of a language deficit, and is a prominent feature in virtually all
tences such as touch the watch, after touching the keys and the pen.
post-stroke aphasic patients, in moderate stage Alzheimer’s disease,
Alternatively, ask ‘If the lion ate the tiger, who survived?’ Syntactic
as well as semantic dementia. Naming ability requires an integra-
ability is classically impaired with lesions of Broca’s area or the ante-
tion of visual, semantic, and phonological aspects of item knowl-
rior insular region and is commonly accompanied by phonological
edge. There is a marked frequency effect, and rather than using very
errors and poor repetition.
common items to test the patient, such as a pen or watch, it may be
more informative to ask about a winder, nib, cufflinks, or a stetho-
scope. Line drawings in the Addenbrooke’s Cognitive Examination Repetition
(ACE-R) (see section Cognitive assessment scales and Fig. 10.1) Use a series of words and sentences of increasing complexity.
are useful for assessing naming ability. Phonemic paraphasias (e.g. Repetition of polysyllabic words such as ‘hippopotamus’ or ‘cater-
‘electrickery’ for ‘electricity’) and semantic paraphasias (‘clock’ for pillar’ followed by enquiry as to their nature assesses phonological,
‘watch’, or ‘apple’ for ‘orange’) may also be seen, and reflect pathol- articulatory, and semantic processing simultaneously. Listen care-
ogy in Broca’s area and the posterior perisylvian region, respectively. fully for phonemic paraphasias during this task. Sentence repeti-
Broad superordinate responses, such as ‘animal’ may be given in tion can be tested with the well-known phrase ‘No ifs, ands or buts’
response to pictures of, for example, a camel, with the progressive which is somewhat surprisingly more difficult than repeating ‘The
Addenbrooke’s cognitive examination—ACE-R
Fig. 10.1 Line drawings used in Addenbrooke’s Cognitive Examination (revised version) to assess naming and comprehension. (© John Hodges, 2000.)
orchestra played and the audience applauded’. This can be further and abstract reasoning. Although executive function is generally
supplemented by using statements with several embedded clauses believed to be a (dorsolateral) frontal lobe function, this set of skills
such as ‘I only know that John is the one to help today’ from the is probably much more widely distributed in the brain than this.
Montreal Cognitive Assessment (MoCA). Inability to do this points Head injury is a common cause of impaired executive function,
to deficits in auditory working memory. which is also usually seen in Alzheimer’s disease, even in the early
stages. It is important not to forget that most of the frontal lobe is
Reading subcortical white matter, and consequently many of the leukodys-
Failure to comprehend is usually accompanied by an inability to trophies cause executive dysfunction and discrete frontal lobe signs.
read aloud, but the reverse is not necessarily true. Test this either Basal ganglia disorders also impair these skills, the prime example
by writing a simple command, such as ‘Close your eyes’, or by using being progressive supranuclear palsy (PSP).
a few phrases from a nearby newspaper. If a reading deficit is
detected, this should be characterized further. Letter and category fluency
Patients with so-called pure alexia exhibit the phenomenon of Letter and category verbal fluency are very useful tests, and poor
letter-by-letter reading, with frequent errors in letter identification. performance in both is common in executive dysfunction. Patients
Neglect dyslexia, seen in right hemisphere damage, is usually con- are asked to produce as many words as possible, starting with a par-
fined to the initial part of a word and can take the form of omissions ticular letter of the alphabet (F, A, and S are the commonly used
or substitutions (e.g. land for island, and fish for dish). Surface dys- letters). Proper names and the generation of exemplars from a sin-
lexics have difficulty in reading words with irregular spelling (e.g. gle stem (e.g. pot, pots, potter) are not allowed. Category fluency is
pint, soot, cellist, dough), which indicates a breakdown in the link- performed by, for example, asking for as many animals as possible
age of words to their underlying semantic meanings and is one of in 1 minute. Young adults can produce 20 animals, 15 animals is
the hallmarks of semantic dementia. Deep dyslexics are unable to low average, and less than 10 is definitely impaired. Letter fluency
read plausible nonwords (e.g. neg, glem, deak), and make semantic is usually more difficult (a score of 15 words per letter is normal),
errors (canary for parrot). and subjects with subcortical or frontal pathology score poorly on
both measures but worse on letter fluency. In contrast, patients with
Writing semantic deficits, such as semantic dementia and Alzheimer’s dis-
Writing is more vulnerable to disruption than reading and involves ease, have a more marked impairment for categories. Refinements,
coordination of both central (spelling) and more peripheral (letter such as categories of dogs and type of fruit, can be introduced to
formation) components. Central dysgraphias affect both written detect more subtle deficits.
and oral spelling. These syndromes are analogous to those seen in
the acquired dyslexias, and can be tested similarly.
Impulsivity, cognitive estimates, perseveration,
In general, intact oral spelling in the face of written spelling
impairments suggests a writing dyspraxia or neglect dysgraphia.
and proverbs
The former results in effortful, and often illegible, writing with fre- Impulsivity is thought to reflect failure of response inhibition and
quent errors in the shape or orientation of letters. Copying is also is seen in inferior frontal pathology. It can be assessed using the
abnormal. A mixed central and peripheral dysgraphia with spelling Go-No-Go task. The examiner instructs the patient to tap once in
errors that tend to be phonologically plausible is commonly seen response to a single tap and to withold a response for two taps. This
in corticobasal degeneration (CBD). Neglect dysgraphia results in test can be made more difficult by changing the initial rule after
misspelling of the initial part of words, and is frequently associated several trials (e.g. tap once when I tap twice, and not at all when I tap
with other nondominant parietal lobe deficits of visuospatial ability once). The ability to switch task and the inhibition of inappropri-
and perceptual function. ate, or perseverative, responses can also be assessed by asking the
patient to copy a short sequence of alternating squares and trian-
Acalculia gles, and then to continue across the page (Fig. 10.2). Perseveration
Acalculia refers to the inability to read, write, and comprehend in drawing one or other of the shapes may be seen in frontal lobe
numbers, and is not exactly the same as an inability to perform deficits, but the test is relatively insensitive. A tendency to clap more
arithmetical calculations (anarithmetria). Although simple calcu- than the same number of times as the examiner (usually three) also
lation is sufficient for most purposes, a full assessment of this skill suggests perseveration. Further clinical examples include palilalia
requires the patient to write numbers to dictation, copy numbers, and palalogia which are characterized by the repetition of sounds
and read them aloud. The left angular gyrus appears to be impor- and words respectively, whilst the repetition of whatever is heard is
tant for these numeracy skills. The patient should also be asked to known as echolalia.
perform oral arithmetic and written calculation and finally be tested The Cognitive Estimates Test may prompt bizarre or improb-
in ability to reason arithmetically (e.g. calculating change received able responses in patients with frontal or executive dysfunction.
when purchasing several items). The integration of several skills is
important here, including the retrieval of stored arithmetic facts
and the ability to manipulate numerical quantities arithmetically.
Executive and Frontal Lobe Function

Impairments in this domain typically involve errors of goal set- Fig. 10.2 A sequence of alternating shapes to test the ability to switch tasks and
ting, planning, judgement, initiation, flexibility, impulse control, suppress inappropriate perseverative responses.
Although it is a formal test, with defined scoring norms, it can be (orobuccal or limb), and to provide a description of impaired per-
performed at the bedside by asking, for example, the population formance, recording both spatial and temporal or sequencing errors
of a nearby city, of London, and of the UK, the height of the Post on several different types of task.
Office Tower, or the speed of a typical racehorse. Questions about A thorough assessment of apraxia should involve the following:
the similarity between two conceptually similar objects can be used ◆ Imitation of gestures, both meaningful (e.g. wave, salute,
to assess inferential reasoning which may be impaired in the same hitch-hiking sign) and meaningless (body and nonbody-oriented
way. Simple pairs such as ‘apples and oranges’ or ‘desk and chair’ are hand positions). Meaningful gestures should also be tested to
tested first, followed by more abstract pairs such as ‘love and hate’ command. See Fig. 10.3.
or a ‘poem and a statue’. Patients typically answer, quite concretely,
that two objects are ‘different’ or that they are ‘not similar’ and are ◆ Orobuccal movements (blow out a candle, stick out your tongue,
unable to form an abstract concept to link the pair. This often per- cough, lick your lips) and the use of imagined objects (comb your
sists despite encouragement to consider other ways in which the hair, brush your teeth, carve a loaf of bread) are assessed. A com-
items are alike. Testing of proverbs probably measures a similar mon error is to use a body part as a tool, such as a finger for a
skill, but it is highly dependent on premorbid educational ability toothbrush; if repeated after being corrected, this can be consid-
and cultural background. ered pathological. Actual use of the object generally elicits better
The three-step Luria test, a motor sequencing task, is thought to be performance than when it is mimed, and is typical of so-called
a left frontal lobe task, and is discussed more in the section Apraxia. ideomotor apraxia.
◆ A sequencing task such as the Luria three-step command (fist,
Behavioural assessment edge, palm) or the alternating hand movements test completes
Inappropriate behaviour is seldom, if ever, elicited from the history the assessment. This latter task is performed, after demonstra-
given by the patient, who may act quite normally during a clinical tion, with arms outstretched, and alternate opening and clos-
consultation. Direct questioning about conflict at work, with inter- ing of the fingers of one hand, while those of the other remain
personal relationships, or involvement with law enforcement agen- clenched in a fist.
cies may be helpful in determining the degree of insight; however, the ◆ Lower limb apraxia may be demonstrated by an inability to trace
corroboration of the history from an informant, interviewed alone, patterns with the feet on the ground in response to command.
is crucial. Spouses may mention embarrassing social behaviour, Relative preservation of some movements (e.g. bicycling move-
changes in food preference (in particular sweet foods), or inappro- ments with the legs whilst lying down) in the presence of gait
priate sexual behaviour. Ability to empathize and judge the emo- ataxia may also suggest an apraxic cause.
tional state of others is particularly disrupted in the frontotemporal
syndromes. The informant should also be questioned about irritabil- In general, apraxic signs are of limited localizing value, but the
ity, anxiety, and poor judgement. Apathy and poor motivation are left parietal and frontal lobes appear to be of greatest importance.
common features of Alzheimer’s disease and frontotemporal and Orobuccal apraxia is closely associated with lesions of the left infe-
subcortical dementias, but do not differentiate well between different rior frontal lobe and the insula, and commonly accompanies the
aetiologies. Impulsiveness, which is sometimes demonstrated clini- aphasia caused by lesions of Broca’s area. Progressive, isolated limb
cally by the Go-No-Go task described in the section Impulsivity, apraxia is virtually diagnostic of corticobasal degeneration.
cognitive estimates, perseveration, and proverbs, may be a marker
of impaired inhibition, an inferior frontal lobe function.
There are few clinical tests that reliably and objectively docu-
Visuospatial Ability
ment behavioural impairment. Of most use are a number of Deficits in the visuospatial domain are quite commonly associ-
behavioural inventories that list symptoms and their severity (e.g. ated with neurodegenerative diseases, particularly Alzheimer’s
the Neuropsychiatric Inventory (NPI), Cambridge Behavioural disease, but are often clinically silent and missed unless enquired
Inventory (CBI), and Frontal Behavioural Inventory (FBI)) and are about specifically (Fig. 10.4). Getting lost in familiar surroundings
either filled out by the carer prior to the consultation or scored in a (topographical disorientation), difficulties with dressing (dressing
structured clinical interview. apraxia), misreaching for objects, and the failure to identify famil-
The anatomic localization of many behavioural symptoms is iar faces are all markers of this type of impairment.
poorly understood, but there is an increasing awareness of the role Visual neglect may produce a failure to groom one half of the
of the right hemisphere, particularly the medial (anterior cingulate) body or eat what is placed on one side of a plate. Visual halluci-
and inferior (orbital) frontal and anterior temporal. nations invariably suggest an organic cause and are prominent in
dementia with Lewy bodies and acute confusional states. Formed
visual hallucinations may also be seen in the absence of cognitive
Apraxia impairment in the Charles Bonnet syndrome, and are often associ-
The inability to perform a movement with a body part despite intact ated with poor eyesight; insight is generally retained.
sensory and motor function is termed apraxia. Theoretically, the Information from the visual cortex is directed towards the tem-
concept can be divided into errors of action conception (knowledge poral or parietal cortex via one of two streams. The dorsal (‘where’)
of actions and of tool functions, e.g. the pupose of a screwdriver) stream links visual information with spatial position and orienta-
and action production (generation and control of movement). tion in the parietal lobe, whereas the ventral (‘what’) stream links
Although a number of categories, such as limb-kinetic, ideomo- this information to the store of semantic knowledge in the tempo-
tor, and ideational apraxia, exist, these labels are seldom useful in ral lobes. The frontal eye fields are important in directing attention
clinical practice. It is more helpful to describe the apraxia by region towards targets in the visual field.
Fig. 10.3 Examples of meaningful and meaningless hand gestures that may be used in testing for limb apraxia. It is important to ensure that there is no physical
limitation (e.g. severe arthritis, contractures, hemiplegic stroke) that prevents the patient from copying the gesture.
Fig. 10.4 Examples of poor visuospatial performance when

reproducing interlocking pentagons, the wire cube, or drawing
a clock face. This patient has dementia with Lewy bodies
(DLB), which frequently causes visuospatial impairment. Other
common causes include Alzheimer’s disease and corticobasal
degeneration (CBD).
Neglect cube, interlocking pentagons, or constructing a clock-face with

Neglect of personal and extrapersonal space is usually caused by numbers are good tests of constructional ability, and may also high-
lesions to the right hemisphere—usually the inferior parietal or pre- light neglect if it is present. Left-sided lesions tend to cause over-
frontal regions. Deficits can be uncovered by simultaneous bilateral simplification in copying, whereas right-sided lesions may result in
sensory or visual stimulation, or having the patient bisect lines of abnormal spatial relationships between the constituent parts of the
variable length. Letter and star cancellation tasks are similar, more figure. Dressing apraxia is easily tested by having the patient put on
formal tasks. Patients with object-centred neglect fail to copy one clothing that has been turned inside-out.
side of an object, and neglect dyslexics may not read the beginning Visual agnosias
of a line or word. Patients with anosognosia deny they are hemiple-
gic or even that the affected limb belongs to them. Visual object agnosias cause a failure of object recognition despite
adequate perception. Those with apperceptive visual agnosia have
normal basic visual functions, but fail on more complex tasks involv-
Dressing and constructional apraxia ing object identification and naming. However, they are able to name
Although deficits in dressing and constructional ability are termed objects to description, or by touch, indicating a preserved underlying
apraxias, they are best considered as visuospatial rather than motor semantic representation of the object. This phenomenon is described
impairments. Copying three-dimensional shapes such as a wire with bilateral occipitotemporal infarction. In cases of associative
visual agnosia, the deficit reflects a disruption of stored semantic by perceptual processing in subcortical structures and brainstem
knowledge, and involves all modalities accessing this information. nuclei.
This is always secondary to anterior temporal lobe pathology, typi-
cally semantic dementia and herpes encephalitis. To test for these Activities of daily living (ADLs)
syndromes, it is necessary to assess object naming and description, Recent research criteria for dementia include impaired ADLs in the
along with tactile naming, naming unseen objects to description, and definition of dementia. The ability to organize finances, use home
the ability to provide semantic information about unnamed items. appliances, to drive safely, and organize medication regimens are
higher order (instrumental) ADLs that are usually impaired ear-
Prosopagnosia lier in disease than more commonly assessed skills such as cooking,
Prosopagnosics cannot recognize familiar faces. Often other clues, walking, personal hygiene, and continence (basic ADLs). This is an
such as gait, voice, and distinctive clothing, are used to aid iden- area in which a reliable informant, who knows the patients well, is
tification. The deficit may not be entirely selective to faces, and essential.
often fine-grained identification within categories may also be
impaired (e.g. makes of car, types of flowers). Patients are generally Driving
able to characterize individual facial features, and if the underlying Driving is often a sensitive issue. Early cognitive impairment does
(semantic) knowledge associated with a particular person is not dis- not preclude driving but should prompt discussion of driving abil-
rupted, the ability to produce attributes of the face in question, if it ity. In general, spouses are fairly aware of changes in driving skill,
is named, remains intact. An occipitotemporal lesion underlies this and their concerns should not be dismissed lightly. Impairments
disability and is often associated with a field defect, achromatopsia, in visuospatial ability (e.g. copying the wire cube, pentagons,
or pure alexia. Whether it is necessary to have bilateral pathology drawing a clock face) are good markers of increased driving risk.
remains controversial. Where there is anterior right temporal lobe Often, cessation of driving can be negotiated, but in extreme cases,
involvement, as in the right temporal variant of semantic dementia, where poor insight conflicts with a sensible approach, keys can
person-based social knowledge is often profoundly affected. These be hidden, cars can be disabled, moved, or sold, and the licens-
patients also find it difficult to judge facial affect. In delusional misi- ing authority notified. An independent driving assessment may be
dentification syndromes such as the Capgras syndrome, patients are very advisable.
convinced that an impostor, who looks identical, has replaced a
close relative. It occurs in dementia and schizophrenia, and there is
a suggestion that the linkage of affective attributes to a face may be Cognitive Assessment Scales
disconnected from processing of its identification. Perhaps the most widely used cognitive rating scale is the
mini-mental state examination (MMSE), which, although useful,
Colour deficits is weighted significantly towards aspects of memory and atten-
Colour processing deficits such as achromatopsia (loss of ability tion. However, it provides relatively little by way of language test-
to discriminate colours) are often associated with pure alexia after ing, minimal assessment of visuospatial ability, and no testing
medial occipitotemporal damage, following left posterior cerebral of executive performance. It is scored out of 30, with a score of
artery infarction. Colour agnosia impairs tasks requiring retrieval 24 or less being regarded as abnormal. The patient’s educational
of colour information (e.g. ‘What colour is a banana?’), and colour background, age, and first language should be considered. It has
anomia (e.g. ‘What colour is this?’) refers to a specific disorder of the benefit of being fast to administer and is well recognized as a
colour naming despite intact perception and colour knowledge, screening instrument, but it is quite insensitive, particularly in the
probably caused by a disconnection of the language structures in context of frontal and subcortical pathology, as well as mild cogni-
the temporal lobe from the visual cortex. tive impairment.
The Addenbrooke’s Cognitive Examination (ACE) has been
Rare visual syndromes developed in an attempt to address the deficiencies of the MMSE.
A few rare syndromes are worthy of mention. Balint’s syndrome It was designed to be sensitive to the early stages of frontotemporal
consists of a triad of simultanagnosia (inability to attend to more dementia and Alzheimer’s disease and has been shown to be sensi-
than one item of a complex scene at a time), optic ataxia (inability tive to cognitive dysfunction in the Parkinson-plus syndromes. The
to guide reaching or pointing despite adequate vision), and occu- 100-point scale incorporates the items of the MMSE, but includes
lomotor apraxia (inability to voluntarily direct saccades to a visual more tests of executive function, visuospatial skill, and more com-
target). Fields may be full when challenged with gross stimuli, and plex language assessment. The revised version (ACE-R) provides
occulocephalic reflexes are intact. This syndrome results from bilat- subscores for five domains: orientation and attention, memory,
eral damage including the superior-parieto-occipital region, which verbal fluency, language, and visuospatial/perceptual functioning.
disrupts the dorsal (‘where’) visual processing stream linking vis- A cut-off of 88 provides high sensitivity but lower specificity, while
ual with parietal association areas. Possible causes include carbon a cut-off of 82 has very high specificity for dementia at the cost of
monoxide poisoning, watershed infarction, leucodystrophy, and the lower sensitivity.
posterior cortical variant of Alzheimer’s disease. Anton’s syndrome The Montreal Cognitive Assessment (MoCA) is a 30-point cogni-
is a visual agnosia, in which the patient denies any deficit and may tive screening test of orientation, memory, language, attention, and
attempt to negotiate the environment, invariably without success. executive function. This has good sensitivity for Mild Cognitive
In the curious phenomenon known as blindsight, visual stimuli can Impairment (MCI), Alzheimer’s disease, and other neurodegenera-
induce a response despite cortical blindness. It is probably mediated tive diseases.
The briefest of all of the screening examinations is the mental General Neurological Examination
test score (MTS), which is a 10-item scale assessing orientation,
memory (anterograde and retrograde), and attention. A score of A cognitive assessment is incomplete without a careful general neu-
6 or less is abnormal in older people, but as with the other cogni- rological examination. There are a number of clinical features that
tive rating scales, the profile of deficits is more instructive than the have particular importance. Although in the early stages of many
global score. It may help direct further, more detailed assessment, neurodegenerative diseases, clinical signs may be absent, this is not
but offers no advantages over the other scales other than speed of invariable. Table 10.2 highlights the important neurological find-
administration. ings associated with various cognitive disorders.
Formal Neuropsychological Assessment Conclusion

Unfortunately, neuropsychological services are not always availa- It is not possible to examine everything in the cognitive assess-
ble. In general, it is reasonable to reserve this facility for patients in ment and, as in most other areas of neurology, the history remains
whom we need more detailed assessment for diagnostic purposes, pre-eminent in guiding subsequent examination. The central role
or those in whom we wish to better characterize deficits for the pur- of a reliable informant, and the ability to immediately test hypoth-
poses of rehabilitation. Patients with clearcut deficits of moderate eses generated during the history-taking, distinguish this means of
severity are unlikely to need formal testing to reach a diagnosis. In neurological assessment.
contrast, neuropsychology has much to offer for patients in whom In some patients it is not possible to reach a firm diagnosis after
deficits are early and subtle, or who have marginal performance on a single cognitive assessment, even when in possession of a formal
cognitive screening scales. neuropsychological report. This is particularly true for the mild
Table 10.2 Associated neurological features in dementia

Neurological feature Common or early Less common or late
Extrapyramidal signs: Parkinson’s disease (PD), corticobasal degeneration (CBD), Alzheimer’s disease, dementia pugilistica, Wilson’s
-bradykinesia progressive supranuclear palsy (PSP), frontotemporal dementia disease (WD), neurodegeneration with brain
(FTD), dementia with Lewy bodies (DLB), vascular dementia (VaD) iron accumulation (NBIA), leucodystrophies,
-rigidity
dentato-rubro-pallido-luysian atrophy (DRPLA)
-tremor
Chorea Huntington’s disease (HD), DRPLA Autoimmune:
systemic lupus erythematosus (SLE)
Eyes:
-impaired saccades and/or gaze palsy PSP, HD, Niemann-Pick type C (NPC), Wernicke-Korsakoff (W-K)
-visual hallucinations DLB, delirium
-Kayser-Fleischer rings WD
Frontal signs: Subcortical dementia, leucodystrophy, FTD, PSP, WD, frontal
-utilization tumours, hydrocephalus
-grasp
-palmomental
-pout
Alien limb CBD Corpus callosum lesion
Dystonia CBD, WD, HD AD, Lesch-Nyhan
Myoclonus Prion disease, CBD, familial AD, encephalopathy (asterixis) MSA, DLB, post-anoxic, Whipples (facial
myorhythmia)
Ataxia Prion diseases, multiple sclerosis (MS), MSA, W-K, SCA, NBIA,
DRPLA, (WD), leucodystrophy
Orobuccal apraxia CBD, progressive nonfluent aphasia (PNFA)
Pyramidal signs MND, MS, MSA, SCA, leucodystrophy, prion disease, hydrocephalus FTD, familial AD
Peripheral neuropathy Leucodystrophy (especially metachromatic), deficiency states, NBIA, Neuroacanthocytosis
SCA syndromes
Muscle wasting, weakness, and MND-associated FTD
fasciculation
Anosmia Head injury, most neurodegenerative diseases (HD, PD, AD) Anterior cranial fossa tumour
stages of neurodegenerative diseases where symptoms may be non- Hodges, J.R. (1994). Cognitive assessment for clinicians. Oxford University
specific, and reflects the relative insensitivity of both clinical and Press, Oxford.
imaging assessment to early pathology. The time-honoured method Lezak, M.D. (2004). Neuropsychological assessment, 4th edition. Oxford
University Press, Oxford.
of longitudinal follow-up and repeated assessment in such cases is
Mathuranath, P.S., et al. (2000). A brief cognitive test battery to differentiate
invaluable and should not be forgotten.
Alzheimer’s disease and frontotemporal dementia. Neurology, 55, 1613–20.
Mioshi, E., et al. (2006). The Addenbrooke’s Cognitive Examination
References Revised (ACE-R): a brief cognitive test battery for dementia screening.
Bak, T.H., et al. (2005). Cognitive bedside assessment in atypical parkinsonian International Journal of Geriatric Psychiatry, 21, 1078–85.
syndromes. Journal of Neurology, Neurosurgery, and Psychiatry, 76, 420–2. Nasreddine, Z.S., et al. (2005). The Montreal Cognitive Assessment, MoCA:
Dubois, B., et al. (2000). The FAB: a Frontal Assessment Battery at bedside. a brief screening tool for mild cognitive impairment. Journal of the
Neurology, 55, 1621–6. American Geriatrics Society, 53, 695–9.
Heilman, K.M. and Valenstein, E. (2003). Clinical neuropsychology, 4th Rossor, M.N., et al. (2010). The diagnosis of young-onset dementia. Lancet
edition. Oxford University Press, Oxford. Neurology, 9, 793–806.
CHAPTER 11
Physical assessment
of older patients
Duncan Forsyth
Mental health problems in older age may be the presenting feature Older people are more prone to multiple pathologies, so that there
of other medical illness; they may present with somatization; or they may be several aetiologies for a given symptom, e.g. the combination
may exacerbate coexistent medical conditions. Thus, the patient of Parkinson’s disease, osteoarthritis, poor visual acuity from senile
may first be referred to a geriatrician rather than an old age psy- macular degeneration, new onset atrial fibrillation, and drug-induced
chiatrist to exclude ‘organic’ pathology. Equally, colleagues in old orthostatic hypotension may combine to explain postural instabil-
age psychiatry may need to involve the geriatrician to assist in the ity and falls. Thus, having found one possible cause for presenting
management of coexisting medical problems and to advise regard- symptoms, the clinician should consider whether others may also be
ing complex polypharmacy. All physicians should be aware of the contributing (this is in contrast to Ockham’s razor, which suggests
potential psychological impact of ill health and its treatments on one accepts the simplest explanation for an event unless there is rea-
their patients and carers. son to do otherwise); this will also require a judicious use of clinical
As a broad general rule, each individual tends to age at a rela- acumen in apportioning clinical relevance to every possible diagno-
tively constant rate; in other words, the ageing process is not a sud- sis and determining to what extent each should be investigated and
den occurrence, but its trajectory may be influenced by acute illness managed. Equally, the effects of several minor (individually unim-
and chronic disease processes. Frail older people may deteriorate pressive) interventions may combine to produce a major change, e.g.
rapidly and are more likely to develop secondary complications due the combination of quadriceps exercises to help stabilize an osteoar-
to their poor physiological reserve. The so-called nonspecific pres- thritic knee, analgesia for osteoarthritic pain, withdrawing diuretics
entation of disease typifies the geriatric giants with presentations to reduce postural hypotension, digoxin to slow atrial fibrillation,
of delirium, immobility (gone off feet), falls, incontinence, and and treatment of mild Parkinson’s may enable the physiotherapist to
increased dependency. This nonspecific presentation of disease in rehabilitate a previously immobile individual.
older life can often mean that symptoms of disease are mistaken for There is much contention in the literature regarding assessment
the process of ageing or worse still assigned as not coping (‘acopia’) scales, whether for disability, mobility, or cognition. It is not appro-
or ‘social problems’. Although typical diagnostic features of par- priate in the context of this chapter to discuss the pros and cons of all
ticular disease processes may be obscured in older age, making the scales; suffice it to say that clinicians should use scales that they are
diagnostic process more difficult, diagnostic features are seldom familiar with and understand, recognize their limitations, and that it
completely absent and clues will be found if care and attention are is the process of documenting change within these assessment scales
taken with a good history (including corroborative history from that is most important in determining whether a particular treat-
relatives and carers) as well as a thorough physical examination and ment modality is working or whether a disease process is progress-
appropriate investigations. For example, whilst truly silent myocar- ing. Any scales used need to be applicable to everyday clinical usage
dial infarction may occur and is more common with increasing age and may be different to those used in clinical trials, where more time
(Muller et al., 1990), most are not truly silent, as they come to light is available to complete more complex assessment scales. A prag-
as the result of a fall, evidence of left ventricular failure, or new matic approach is simply to be descriptive. For example, the person
onset atrial fibrillation. Atypical presentations of disease in older who can now walk using a Zimmer frame, with the assistance of
age have been overemphasized and, in the author’s view, generally one person and transfer with the help of one person, is improving
reflect inadequate assessment of the older person. if 1 week earlier he needed two people to assist with transfers and
could only stand with the help of two people. A smiling patient who
acknowledges your presence and initiates conversation is respond-
Learning point ing to the antibiotics for his severe pneumonia, if several days earlier
Patients admitted to the general hospital with a diagnosis of ‘acopia’ he could barely speak and showed no interest in his surroundings.
or as a ‘social admission’ generally have significant underlying On the other hand, the individual whose appetite and fluid intake is
pathology and are likely to benefit from comprehensive geriatric declining and who is becoming less engaged in his environment is
assessment. developing some illness and needs further attention.
90
Learning point 80
% Cognitive Function
70
Simple descriptive observation can be as useful as completing
60
assessment scales but needs to describe the change: 50
◆ Unhelpful descriptors include: ‘better’, ‘improving’, ‘confused’, 40
‘more confused’, ‘less confused’, ‘sitting out’. 30
20
◆ Helpful descriptors may be: ‘now transferring with the help of 10
one person’, ‘no longer eating or drinking’, ‘has started taking 0
1 2 3 4 5 6 7 8 9 10
interest again in his personal appearance’, ‘no longer wandering
or hallucinating’. Years
Normal cognition
Wherever possible and provided the patient is agreeable, it is Dementia
always useful to have a member of the family or a carer in attend- Delirium superimposed
ance when assessing the patient. The benefits are fourfold: addi- on dementia
tional information can be obtained from the third party; the family Fig. 11.1 Impact of illness on cognitive trajectory.
member/carer remains informed; carer strain may become evident;
and a care plan is better developed. It is beyond the scope of this chapter to cover the entire com-
plexity of geriatric medicine. It will be more appropriate to address
some broad principles and highlight specific areas that may com-
Learning point monly present as problems for the old age psychiatrist.
Involving family or carers in the assessment process:
◆ assists with information gathering and communication Observation
◆ provides an opportunity to educate the carer/family as well as the General observation of the patient before and during the actual
patient consultation can provide important clues, e.g. a resting tremor and
shuffling gait suggests a parkinsonian syndrome; failure to make
◆ may help identify carer strain
eye contact may alert to possible depression or a glass eye! Do not
◆ facilitates the care planning process miss any opportunity to assess gait: watch the patient walk in to
◆ will help emergency care staff determine if the person has the consultation room; lead you through his/her house; going to
changed and needs admission to hospital or can be allowed the toilet; and so on. Does the patient require assistance with walk-
home. ing, transfers, and so on? The following may give vital clues to falls
risk:
◆ ill-fitting footwear
The impact of illness on any older person can be devastating, both
physically and psychologically (see Psychological Manifestations ◆ postural instability and tendency to backwards lean on standing
of Disease). It is important to warn family members that rarely of postural hypotension
will individuals with any background of a chronic progressive dis- ◆ shuffling gait of extrapyramidal syndromes
ease return to their prior level of function upon recovery from any
acute illness. The extent of the difference in their level of function- ◆ hemiplegic gait
ing will be determined by the severity and duration of the acute ◆ antalgic gait of osteoarthritic hips
illness. In those with cognitive impairment, it is a strong clinical ◆ broad-based gait of postural instability due to cerebellar or pos-
impression that their cognitive abilities will not be the same after terior spinocerebellar tract pathology, middle ear disease, or
acute illness of any nature (shown schematically in Fig. 11.1). This peripheral neuropathy
means that care packages will need to be reviewed and amended,
where necessary, once individuals have recovered from their acute ◆ can they walk and talk at the same time?
illness. Early identification of these frail older people, many of ◆ do they take more than four steps to turn through 180 degrees?
whom will have some degree of cognitive impairment, with proac-
◆ slow and careful gait with visual impairment.
tive case management and early escalation of their care package
may help to avoid hospitalization and/or facilitate early supported Observing the face may reveal the blunted facial expression of
discharge. Parkinson’s or depression; pallor of anaemia or lack of sun-exposure
(start to think about vitamin D deficiency and osteomalacia);
vitiligo (which may mark other autoimmune conditions such as
Learning point diabetes, pernicious anaemia, Addison’s disease, and thyroid dys-
Keeping a register of vulnerable older people within primary care, function); orofacial tardive dyskinesia; ptosis (if symmetrical this
along with proactive case management, may help reduce the risk will be an ageing phenomenon, if unilateral this may be due to eye
of hospitalization and facilitate early supported discharge from surgery or neurological disease); facial palsy of new or old stroke;
hospital. or Bell’s palsy. Orange staining around the mouth/of teeth/on the
chin or beard in Parkinson’s patients identifies problems swallowing
CHAPTER 11 physical assessment of older patients 161
Stalevo tablets with associated drooling, and this should prompt a

swallowing assessment and review of their medication as well as Learning point
affording an explanation as to why their Parkinson’s symptoms are The absence of any change in behaviour after a fall, in people with
poorly controlled. dementia, can generally be taken to mean no damage was done.
Stooped posture may be the kyphosis of osteoporosis or due to In such circumstances their carer(s) need supporting to have the
Parkinson’s. A tentative handshake may identify rotator cuff injury confidence NOT to send the person to hospital.
or painful inflammatory arthritis.
Evidence of self-neglect or inability to attend to oneself due to ill
health may be obvious with dirty clothes, hands, and face; unshaven Blunted facial expression raises the possibility of a parkinsonian
and unkempt appearance; neglected nails (remember to check the syndrome, but in the absence of other extrapyramidal signs is more
feet and toenails). likely to be due to depression. However, the two do commonly
Body habitus will separate the well nourished from the poorly coexist. Also consider poor visual acuity and failure by the patient
nourished or cachexic due to underlying sinister pathology. to respond to visual cues.
Initial conversation will identify whether speech is appropriate Inability to engage patients in conversation requires a thorough
(delirium, psychosis), intelligible (e.g. dysarthria, dysphonia, or dys- assessment of hearing, e.g. look for wax, have they got a hearing aid
phasia suggesting neurological disease such as Parkinson’s, stroke, that they should be wearing (is it in working order—see Table 11.1)?
tumour, encephalopathy, or motor neuron disease), plausible and The prevalence of hearing impairment increases with advancing
accurate (delirium, dementia, delusional), hindered by shortness of age; over 80% of those aged over 80 years may benefit from a hear-
breath (severe respiratory or cardiac disease); and whether mood is ing aid. Whilst many older people with hearing impairment will
normal (depression, lability post-stroke). have sensorineural deafness and benefit from a hearing aid, initial
assessment must exclude blockage of the external auditory canal by
wax. Early referral, to the local audiology department, is recom-
Learning point mended for anyone with hearing impairment (not due to wax) for
The neuromuscular control of articulation is the same as for swal- assessment (Sharaf et al., 2006). Writing questions down or using
lowing, so a dysarthric patient is highly likely to be dysphagic as a microphone and headset communicator may overcome a hearing
well. impediment. If it does not, assuming that you are conversing in the
Being expert in the art of the ‘nontouch’ examination tech-

Table 11.1 Solving common problems with hearing aids
nique is a crucial part of the physical assessment of agitated,
paranoid, or aggressive individuals. This will help reduce the Hearing aid apparently not working
chance of personal injury and also of unexpectedly coming in to ◆ Check if aid is switched to the ‘T’ setting by accident.
contact with urine, blood, and faeces. For example, faecal soiling ◆ Check volume is at the correct level and not turned right down.
may alert to the possibility of constipation, with overflow diar- ◆ Check the battery is the right way round.
rhoea, and the distended suprapubic region to agitation second- ◆ Try putting in a new battery.
ary to coexistent urinary retention. Asking demented agitated
◆ For behind the ear (BTE) hearing aid, check that the tubing is not twisted,
patients why they keep hitting staff may also prove useful— squashed, or split
‘Because it hurts when they hold my arm’ was the response from
◆ Check whether there are droplets of condensation in the tubing. If there are,
one old man with an undiagnosed fractured head of humerus.
gently pull the soft tubing off the hooked part of the aid and blow down
Whilst people with dementia are more prone to fall, their behav- the tubing to remove the droplets; and/or
iour afterwards may be a consequence of the fall, not of their ◆ If it is a body-worn aid, the lead may need to be replaced.
pre-existing dementia. Check with their carers if there has been
◆ Check that the ear mould (and ear) is not blocked with wax.
a change in behaviour, look carefully for fractures, and consider
the possibility of subdural haemorrhage. Also remember that Whistling, squealing, sizzling, or squeaking
the fall may have been precipitated by intercurrent illness or This may be caused by ‘feedback’ when sound amplified by the hearing aid
change in medication. leaks out and is picked up by the hearing aid microphone. It may happen if:
◆ The ear mould is not put in properly—push it gently to check.
◆ There is excess wax in the ear—patient’s ears need to be checked.
Learning point ◆ The ear mould does not fit ear snugly enough—audiologist to review.
When faced with behaviourally disturbed patients, check if this is ◆ The volume is set too high—check the volume control.
usual for them. Where behaviour has changed, consider the follow- ◆ The ear mould has cracked, plastic hook or tubing in a BTE aid has become
ing common possibilities: loose or has split—the hearing aid centre should be contacted for help.
◆ pain, e.g. after a fall with or without fracture ◆ The tubing in a BTE hearing aid has become hard, causing the hearing aid
not to work well. If it splits, the aid may start to whistle. It can be replaced by
◆ constipation the patient or audiology staff.
◆ urinary retention (more common in men than women) Buzzing noises
◆ adverse drug reaction (has medication been changed recently?) This might be due to switching the hearing aid to the ‘T’ setting by accident.
Otherwise, buzzing generally means the hearing aid has developed a fault and
◆ infection. needs to be repaired.
patient’s first language, consider the possibilities of aphasia (usually

Box 11.1 Change in behaviour may herald onset of disease
due to cerebrovascular disease), depression, and frontotemporal
dementia. An 82-year-old man was recovering from severe community
acquired pneumonia with associated delirium. At week four his
delirium had resolved and he was able to mobilize with assist-
Learning point ance of one and a Zimmer frame. At week six, on the day before
Different tactics should be tried to facilitate listening for patients transfer to a community rehabilitation unit, the consultant noted
with hearing difficulties even if they have hearing aids. These that the patient was inattentive, drowsy, and engaged less in con-
include: versation than on his ward round 2 days earlier. Transfer to the
◆ attracting the person’s attention before speaking rehabilitation unit was deferred and investigations revealed a uri-
nary tract infection.
◆ facing the person when speaking and make sure the light is on
your face
◆ cutting down background noise whenever possible malignancy, Paget’s disease). Immobility may also result from
◆ speaking clearly and naturally, but not too quickly neurological weakness (stroke, peripheral neuropathy, motor neu-
ron disease); muscle damage (myopathy, hypokalaemia, diabetic
◆ the speaker should not shout as this may cause distortion amyotrophy, disuse) or reduced exercise tolerance (cardiac or res-
◆ rephrasing the words where necessary. Some words are more piratory disease, anaemia). Immobility may be the psychological
lip-readable than others consequence of falling (fear of further falls). Other causes include
◆ checking that instructions have been understood correctly painful bunions, onychogryphosis, foot ulceration, unsuccessful
orthopaedic procedures, and sedation.
◆ information could also be written down.
Drug History
Assessment of nutritional and hydrational status, including Any assessment is incomplete until all medications have been
any records of food and fluid intake, supplemented by records of checked. Do not rely upon patient recollection but ask to see the
weight, will help assess the general impact of illness on the indi- drugs and then go through each one as to how they are taken (dose,
vidual but does not necessarily give a clue as to the underlying diag- frequency, and adherence) and ask about possible side effects.
nosis. However, absence of weight loss in someone complaining of Around 20–25% of older people are not taking drugs that their gen-
weight loss and poor appetite, whether with or without other symp- eral practitioner (GP) expects them to be taking and about one-third
toms, suggests anxiety, depression, or attention-seeking behaviour. will be taking drugs that their GP does not know about; this lat-
Failure to gain weight as expected with appropriate medical treat- ter group will include complementary and alternative medicines
ment should also raise the possibility of missed or inadequately (CAM) as well as over the counter (OTC) medications. Up to 50%
treated comorbidities, coexistent depression, or poor psychologi- may be taking their medicines incorrectly (Royal Pharmaceutical
cal adjustment to their illness. Similarly, lack of interest in personal Society of Great Britain, 1997) and a similar percentage may be able
care (dirty and dishevelled or simply just not wanting to get dressed to stop their drugs (Walma et al., 1997).
or brush one’s hair) may reflect hypoactive delirium, subcortical or Concordance with medication regimes is greater if the patient
frontotemporal dementia, or depression. understands what the drug is for, when it should be taken, and what
In frail older individuals, multiple pathology may complicate the the potential benefits and disadvantages of taking the drug are. The
assessment of gait, e.g. poor visual acuity may make the individual ability to follow a complex medication schedule can be improved by
slow and cautious, as will fear of further falls; diabetic peripheral the use of a dosette box, pill-timer, SMS text messaging, or a carer
neuropathy may cause a shuffling gait due to altered propriocep- taking control of supervising the medication.
tion; and osteoarthritis of the lumbosacral spine and hips will cause
the individual to move in a stiff and rigid manner. Care should also
be taken to differentiate illusions in those with poor visual acuity Learning point
from true hallucinations, thereby saving the individual from an ◆ Do not assume that patients are taking their medication as per
unnecessary prescription of an antipsychotic. the instructions on the bottle.
Frail older people with long lengths of stay in hospital provide
◆ Do not assume that patients are taking the medication that their
an opportunity to teach nursing and medical staff observational
GP thinks they should be taking.
skills, so that they learn to recognize change in behaviour that alerts
to underlying illness (Box 11.1). Emphasizing these skills to care ◆ Do not assume that patients are only taking medication men-
home staff can help with the early recognition of illness behaviour tioned in any referral letter.
in care home residents with dementia, allowing early detection and ◆ The ability to accurately recall complex medication regimes pro-
treatment, and may avoid deterioration to the extent that hospitali- vides a useful screening assessment of cognition.
zation is required, with its attendant risks.
◆ Consider that any illness may be iatrogenic and assess medica-
Immobility tion accordingly.
This may be caused by pain or stiffness, in joints (arthritis or
infection); in muscles (trauma, polymyositis, polymyalgia rheu- The importance of medication risk in older age warranted a
matica, Parkinson’s disease), or bones (osteoporosis, osteomalacia, separate addendum to the National Service Framework for Older
People (NSFOP), published in England in 2001 (Department of which drugs are necessary and whether there is a liquid or patch
Health, 2001), and so this should be an integral part of the sin- formulation.
gle assessment process (SAP), set out in Standard 2 of the NSFOP.
Problems with medication are also often linked to stroke, falls, and Examination
mental health (Standards 5, 6, and 7 of the NSFOP). Risk factors for General aspects
iatrogenic drug problems include: When examining a patient of the opposite sex it is essential to have
◆ four or more drugs (remember OTCs!); a chaperone present. Examination technique will, by necessity, vary
according to how cooperative the patient is, and may not be com-
◆ specific drugs, e.g. warfarin, NSAIDs, diuretics, digoxin;
pleted in one go. Even the most expert clinician cannot get away
◆ recent discharge from hospital; entirely with the ‘hands-off ’ approach outlined in Observation).
◆ dementia; However, keen observation skills may focus the physical examina-
tion on the essential and the not so essential aspects. Holding an
◆ depression;
agitated or paranoid patient down is more likely to lead to violent
◆ poor vision/hearing/dexterity; behaviour than to facilitate a successful examination. As a general
◆ living alone or low levels of support. rule, if the patient cannot comprehend the reason for the examina-
tion (whether this is cardiac auscultation or digital rectal examina-
tion), you will be unlikely to succeed. If at first you don’t succeed,
Learning point try again later (Royal College of Physicians, 2006).
Reducing or omitting diuretics, angiotensin converting enzyme A calm reassuring manner, with appropriate introductions as to
inhibitors (ACE-I), and nonsteroidal anti-inflammatory drugs who you are, what you want to do, and why, will generally work
(NSAIDs) in prolonged hot weather can help reduce the incidence (but not always). Gentle hand holding is reassuring but can easily
of postural hypotension, electrolyte imbalance, renal impairment, be turned in to a restraining hand if the patient becomes aggressive.
and death in housebound or institutionalized frail older people. Likewise, careful positioning of one’s legs when facing the confused
and agitated patient appears nonthreatening but can also become a
defensive block if the patient kicks out. Body language is important
Which drugs are really necessary and why? to maintain a nonthreatening demeanour, but on occasions even this
Unfortunately, older people are largely excluded from clinical can be misinterpreted, e.g. with a deaf patient, as one leans in to speak
drug trials due to their high level of comorbidities, and so the clearly in the proffered ear this can be mistaken for an affectionate
evidence-base for most pharmacological treatments is weak or non- greeting and you are greeted with an unexpected peck on the cheek!
existent in older age. Older people are at increased risk of adverse Assessment of a confused individual must include assessment
effects of drugs, due to altered pharmacokinetics and pharmacody- of cognition and a screen for delirium. Assessment scales com-
namics, but may also have more to gain (absolute risk reduction) monly used by geriatricians include: Confusion Assessment
from preventive therapies, as more events occur in older age, e.g. Method (CAM) (Inouye et al., 1990), Folstein Mini-Mental State
the absolute benefits of thrombolysis in acute myocardial infarction Examination (MMSE) (Folstein et al., 1975), CLOX or clock draw-
increase with age. The complex interaction of age and disease on ing test (Royall et al., 1998), Informant Questionnaire on Cognitive
drug metabolism, e.g. drug metabolism declines in delirium (White Decline in the Elderly (IQCODE) (Jorm, 1994), and Geriatric
et al., 2005), may enable the judicious manipulation of side effects Depression Scale (GDS) (Yesavage and Brink, 1983). The invoca-
to advantage, e.g. in treating depression in Parkinson’s disease the tion of copyright with the MMSE may drive clinicians to use alter-
anticholinergic side effects of tricyclic antidepressants (TCAs) may native scales. However, those scales that incorporate the MMSE,
help reduce tremor. e.g. the Addenbrooke’s Cognitive Examination Revised (ACE-R)
In the later stages of dementia, or in an uncooperative patient, a (Mioshi et al., 2006), are also included in this copyright issue.
decision may be needed as to which medications are really neces-
sary. For example, in a bedfast, severely demented individual with The mouth and its contents
cardiovascular disease, is aspirin therapy still appropriate to reduce Angular stomatitis is usually due to dribbling and may be seen in
the possibility of a cardiovascular event? In an individual with Parkinson’s and in individuals with facial palsy or poorly fitting den-
severe dementia and inconsistent nutritional intake, for whom it tures; secondary infection may occur with candida. Herpes simplex and
has been agreed that artificial nutritional support is inappropriate, aphthous ulceration are often nonspecific markers of ill health. Oral
is continued bone prophylaxis with calcium and vitamin D sup- candidiasis may reflect poor oral hygiene, diabetes, recent treatment
plements or a bisphosphonate appropriate? Would substitution of with antibiotics, and can make swallowing very painful. Fasciculation
paracetamol for codeine-based analgesia reduce the necessity for of the tongue suggests motor neuron disease and requires further neu-
regular laxatives? If individuals cannot be relied upon to regu- rophysiological investigation, i.e. EMG and nerve conduction studies.
larly take their medication, even with supervision, is continued Mouth ulceration is often due to ill-fitting dentures, but the possibility
anticoagulation with warfarin wise or is there a safer alternative? of oral squamous cell carcinoma may need excluding.
A medication review should be undertaken at every opportunity Carious teeth may be the source of ‘unexplained’ fever and raised
and advice sought from a consultant geriatrician if the old age psy- inflammatory markers and can make the clinician feel very foolish
chiatrist is uncertain, e.g. is treatment with a statin still appropriate; for not having looked in the mouth.
would aspirin be safer than warfarin for stroke prevention? Dignity and nutrition are best maintained if dentures are well
If the patient is dysphagic or uncooperative (due to delir- fitting and worn regularly. Labelling dentures should reduce the
ium or dementia), this also provides an opportunity to consider chances of them being lost by staff or confused patients.
Skin Suspicious bruising includes: thumb marks (usually on limbs);

The dermis becomes thinner and more fragile with ageing due to bruising in the groins and inner thighs (most older people do not
changes in type-II collagen. Lack of elasticity with ageing renders fall astride an object); multiple bruising of multiple ages (but this
skin turgor useless in the assessment of hydrational status. person may just be a frequent faller); bruising with other inju-
Basal cell carcinoma is the commonest skin malignancy and ries suggestive of abuse, e.g. cigarette butt burns, scalds, friction
occurs most often on the face. It first appears as a pearly papule, burns. Bruising and a fearful or apprehensive patient should always
which slowly and inexorably enlarges. Metastatic spread is rare. make one consider the possibility of elder abuse (British Geriatrics
Squamous cell carcinoma presents as a reddened, indurated ulcer, Society, 2005). Any unexplained or suspicious injury in a vulner-
nodule, or plaque, which often arises in sun-exposed areas. It may able older person (more often than not they will be cognitively
metastasize. Malignant melanoma is an expanding pigmented impaired) should instigate a Safeguarding of Vulnerable Adults
lesion occurring anywhere on the body. Any pigmented skin lesion (SOVA) procedure.
that is changing in size, shape, or colour must be assessed by a der- A focal neurological deficit or fluctuating cognition in the pres-
matologist to exclude malignant transformation. ence of bruising should raise the possibility of subdural haematoma.
The risk of developing pressure sores is increased in those with Fluctuating cognition without evidence of trauma or focal neuro-
poor mobility, oedematous skin, poor nutrition, poor circulation logical deficit is almost always due to problems outside of the cra-
(peripheral vascular disease and cardiac failure), urinary and/or nium causing delirium (Royal College of Physicians, 2006).
faecal incontinence, or on long-term steroids. All individuals at risk
should have their pressure points examined (occiput, ears, shoul- Falls assessment and postural hypotension
ders, elbows, sacrum, spine, ankles, feet, and heels) and appropriate Measurement of resting (after 5 min recumbent or semirecumbent)
measures put in place to minimize that risk (NICE, 2005). and standing (after 2 min) blood pressure is mandatory in anyone
Leg ulceration may be due to arterial or venous disease or often who has fallen. Medication will need reviewing for drugs that might
a mixture of both. A long history of recurrent ulceration, usually of lower blood pressure either directly (vasodilators, volume depletors
the lower third of the shin above either malleolus, with surrounding (diuretics) or indirectly (centrally acting drugs affecting barorecep-
eczema and hyperpigmentation and palpable (unless obscured by tor reflexes, e.g. antiparkinsonian medication, opioid analgesia,
oedema) foot pulses is typical of venous ulceration. Arterial ulcera- and antidepressants). Other causes of volume depletion should be
tion is distinguished by absent foot pulses and well-circumscribed excluded, e.g. diarrhoea, vomiting, blood loss, and dehydration.
small punched-out ulcers often on the lateral aspect of the leg or Watching patients walk will identify gait abnormalities, dangerous
dorsum of the foot. Any ulceration will be painful; absence of pain footwear, postural instability, and whether they can concentrate on
usually implies coexistent peripheral neuropathy. other things whilst walking (walk and talk test). The shoulder-pull
test may be required to identify postural instability. A pull, from
behind, on the individual’s shoulders, whilst he or she stands with
Learning point feet slightly apart should not cause the individual to lose balance
Always be curious to see what lies beneath a dressing; this may hold or retropulse; if either occur then the patient is posturally unstable.
the clue as to why the individual is unwell, e.g. infected leg ulcers The author finds this to be a more useful test for postural instability
or gangrenous toes. than Romberg’s test. Individuals who use a walking aid are (usually)
by definition posturally unstable and so do not need this testing!
A corroborative history will help determine whether there was
Bruising and falls any loss of consciousness; this is unlikely if the individual can
The presence of bruising anywhere will usually be due to falls and recall the act of falling and coming to rest on the ground. However,
should prompt a look for other injury, sprain, or fracture. Pain many individuals, even those without cognitive impairment, do
from this injury may explain altered behaviour or worsened cogni- not remember falls that occurred within the last 3–12 months
tion. Common sites for fragility fractures are the hip, wrist, head of (Cummings et al., 1988). If loss of consciousness is suspected, then
humerus, and pelvis. Pelvic fracture should be suspected in anyone further tests are likely to be required to exclude a vascular (car-
who fails to mobilize after a fall with no other obvious reason not diac arrhythmia, acute coronary syndrome, pulmonary embolism,
to do so. Pain from a pelvic fracture may radiate to the low lum- stroke, or acute severe blood loss) or cerebral (epilepsy) cause.
bar spine and be misdiagnosed as osteoarthritis of the spine. Active Transient ischaemic attacks are focal neurological deficits resolving
flexion of the hip with extension of the knee (straight leg raising) within 24 h and do not cause loss of consciousness, but may cause
contracts muscles attached to all surfaces of the pelvic ring. This a fall!
simple test has a 95% positive predictive value and 90% negative Individuals with poor visual acuity are at increased risk of falling,
predictive value for pelvic fracture, and has greater diagnostic value as are those who have not adjusted to wearing bi- or varifocal lenses
than either downward pressure on the pubic bone or compression (especially when going down steps) (College of Optometrists and
of the iliac rings (Ham et al., 1996). British Geriatrics Society, 2011). Poorly fitting footwear and other
Individuals who are taking warfarin or long-term steroids or who environmental hazards should be identified and corrected.
have abnormal platelet function (e.g. due to bone marrow infiltra- The consequences of falls include: fracture (usually low velocity
tion or primary haematological malignancy) may bruise spontane- osteoporotic); subdural haemorrhage; tissue and muscle necrosis
ously or more extensively than would be expected for the severity from long periods of lying; acute tubular necrosis secondary to
of the injury. rhabdomyolysis; anaemia secondary to extensive bruising (beware
Bruising may result from elder abuse and should therefore always the individual on anticoagulants); hypothermia; fear of further falls
be accurately documented on body charts in the patient’s records. and resultant poor mobility. In institutionalized older people and
those who are housebound, the risk of falls may reduce with vita- Cooperative and uncooperative patients
min D supplementation due to improvements in muscle strength In individuals who demonstrate variable performance, physically
and balance. and/or cognitively, the differential diagnosis lies between:
◆ delirium
Learning point ◆ personality disorder
Falls assessment should include the following:
◆ dementia, particularly dementia with Lewy bodies
◆ Check lying and standing blood pressure.
◆ depression.
◆ Check temperature.
◆ Check for pain and document injuries (fractures and bruising). Differentiating these clinical conditions is generally easy but can
on occasions be difficult. For example, hypoactive delirium can be
◆ Review medication for drugs that might cause postural mistaken for subcortical dementia and depression, whilst agitated
hypotension. depression may be mistaken for symptoms of delirium. This may
◆ Review gait. lead to inappropriate prescribing and/or lack of appropriate thera-
◆ Review footwear. peutic interventions (e.g. rehabilitation).
Delirium is the most common complication of hospitalization
◆ Check balance, e.g. shoulder pull; walk and talk. among older adults, occurring in about 1/3–2/5 of elderly hospi-
◆ Neurological examination. talized patients. It is often not recognized, in up to two-thirds of
◆ Musculoskeletal examination. these cases. Delirium may be precipitated by any drug or illness in
susceptible individuals and often has more than one cause, which is
◆ Check vision and spectacles.
usually outside of the brain. Delirium is associated with an excess
◆ Consider other investigations to look for a cause, e.g. ambulatory mortality; excess morbidity (e.g. increased risk of hospital-acquired
ECG recording; EEG; FBC; vitamin D levels. infection, pressure sores, venous thromboembolic disease);
◆ Consider other investigations to look for complications, e.g. CK; increased lengths of stay; increased rates of institutionalization;
U&Es; X-rays. and higher readmission rates. For a full account of the prevention,
detection, and management of delirium the reader is referred to
◆ Review home for environmental hazards.
the Royal College of Physicians of London and British Geriatrics
◆ Assess cognition, e.g. dementia and/or delirium. Society National Guidelines (Royal College of Physicians, 2006)
◆ Consider elder abuse—‘Did they fall or were they pushed?’ and NICE Clinical Guideline 103 (NICE, 2010) (see also Chapter
40). Serial recording of cognitive function will help differentiate
(FBC, full blood count; CK, creatine kinase; U&Es, urea and electrolytes)
delirium from dementia and can be useful in determining whether
therapists should continue to try to engage the individual in reha-
bilitation. If delirium has been excluded then a trial of antidepres-
The neurological examination
sants may be warranted if depression cannot be excluded (in an
The neurological examination can be an ordeal for all concerned, and
effort to treat the treatable). Delirium symptoms may persist for
is an excellent test of cognition on behalf of the examiner and the
several months after resolution of the precipitating cause(s) and
examinee. Talk the patient through what you are going to do in sim-
may be a premonitory feature of dementia. Thus, individuals who
ple clear language—asking someone to relax all too frequently results
have had delirium should be followed up after discharge.
in entirely the opposite response! Use observational skills as much
The fluctuating course of delirium and its fluctuating recovery
as possible—muscle wasting may lead to difficulties rising from a
are often difficult for the patient’s family to understand—they see
chair or opening the door to the examination room. Muscle wasting
someone recovering, only to get worse and then recover and then
(sarcopenia) is common with ageing or may be secondary to disuse
deteriorate again. If medical and nursing staff do not recognize or
around arthritic joints, or reflect neurological disease (e.g. peripheral
understand delirium, their explanations to family members can
neuropathy, motor neuron disease, old polio). Reflexes may be dif-
seem confused and lead to hostility and complaints. A useful anal-
ficult to elicit or interpret; and power and tone may be difficult to
ogy is that of jet-lag: most people have experienced this form of
judge due to coexistent arthritis. Ankle jerks may be absent in around
delirium and can start to understand what is happening. The proc-
one in three normal older people. Pupils may be small and sluggish
ess of recovery is akin to setting your clock to local time, and so
to react to light, making fundoscopy difficult, even with mydriasis.
measures such as stopping delirium-inducing drugs or the resolu-
Fundoscopy may also be hindered by cataracts. Position sense is gen-
tion of infection do not immediately result in the body clock being
erally retained in older age, whilst vibration sense may be lost or not
reset or bring about an immediate resolution of confusion. As
understood. In general, testing sensation can be fraught with diffi-
dementia is the commonest risk factor for developing delirium it is
culty. Isolated upward gaze palsy is of dubious significance.
also worth advising relatives that the individual will rarely return to
his or her prior level of function.
Learning point
Progressive supranuclear palsy (PSP) may cause an isolated down-
ward gaze palsy not isolated upward gaze palsy. In PSP, upward gaze Learning point
palsy occurs in the context of other gaze paresis. These individuals ◆ Delirium is usually multifactorial, with the cause(s) being out-
are at increased risk of falling when they go down stairs. side of the brain.
◆ CT head scan is not necessary unless there is focal neurology or skills help to determine whether the individual is improving or
other reason to suspect intracranial pathology, e.g. falls and pos- deteriorating. If the latter occurs, a point will come at which they
sible subdural haematoma. are no longer able to mount resistance to more detailed examina-
◆ Delirium symptoms may persist for several months after resolution or investigation; hopefully this will not be beyond a point of
tion of the precipitating event. salvation, but often it is.
Sedating an uncooperative patient may be necessary to enable
◆ Family and carer(s) need information to help them understand essential investigations to be performed, e.g. head CT scan when
the cause(s) and natural history of delirium. behavioural disturbance could be due to a subdural haemorrhage
◆ Patients need ‘debriefing’ once they have recovered from delir- or intracranial tumour, and knowing this would alter management
ium so that they can understand what happened, as it will often plans. Sedation may also be necessary when patients are a risk to
have been a frightening experience for them and they may have themselves or others. However, in general, sedation is best avoided,
some recollection of events. as more often it makes things worse rather than better.
Investigations in an uncooperative patient Learning point

Even in cooperative patients one should always consider the utility of ◆ In general, only rendering patients comatose will prevent them
any investigation. Will the result influence this patient’s management wandering!
now or in the future? In uncooperative individuals, one may need to
practice empirical medicine and treat on the grounds of probabilities. ◆ Reality orientation for patients who are delirious may involve
Treatment may of course include masterly inactivity on the part of slipping in to their reality and welcoming them back to yours as
the doctor (so hard for many to practice!). Patients often do better delirium resolves!
when the doctor does nothing, as they are spared unnecessary and
inappropriate interventions and their attendant complications.
Specific Diseases
Learning point To diagnose a parkinsonian syndrome the individual must be
Things that help win over even the most belligerent patients bradykinetic and then have either rigidity and/or tremor. Tremor is
include: common in older age, occurring in 10% of over 65-year–olds, and
◆ a calm and nondistressing environment is not always due to Parkinson’s disease: equally, not all Parkinson’s
patients have tremor. Tremor in Parkinson’s is typically a rest
◆ friendly reassurance and consistent communication from staff tremor but may be present on action. Rigidity in older age is not
◆ offering adequate fluids and nutrition always classical cogwheel in nature. Idiopathic Parkinson’s disease
◆ waiting for the individual to settle down before performing com- can only be diagnosed if other parkinsonian syndromes have been
plex investigations excluded (Table 11.2). The diagnosis of idiopathic Parkinson’s dis-
ease in older age can be extremely difficult due to the impact of
◆ not arguing. other comorbid conditions and the fact that there are more causes
of parkinsonism in older age (Table 11.3).
Whilst waiting for a period of calm, attention to the basics of care Parkinson’s patients with motor fluctuations can cause particu-
(delirium prevention and management) will, for most individuals, lar difficulties, as they may be seen to undertake a task (e.g. walk-
be the most important aspects of their care. Time will also be well ing or feeding themselves) for which they require assistance later
spent in reviewing which drugs are really necessary, both now and the same day; this can lead to an incorrect assumption that they
for the future, if behavioural disturbance persists (see also Chapter are manipulative or awkward. Assessment by a Parkinson’s disease
40 for the assessment and management of delirium). For example, specialist should be undertaken to ensure optimal treatment for
annoying already confused and agitated individuals by insisting the Parkinson’s (pharmacological and nonpharmacological) and
that they take calcium and vitamin D supplements or aspirin is to establish that the motor symptoms of Parkinson’s are not being
unnecessary, whilst coaxing them to take their antiparkinsonian adversely affected by other extraneous events, e.g. physical illness,
medication may be time well spent. Parents quickly learn when psychological stress, or other mental health problems. Sudden
coaxing their baby or toddler to take food or medicine what will or changes in the natural trajectory of the disease will always be due
will not work and how to prioritize what is really necessary: such to the effect of extraneous factors on patients’ ability to compensate
skills are invaluable when dealing with a delirious frail older person for their symptoms and the disability, i.e. the effect of physical (e.g.
(but you do not need to be a parent to have them)! infection) or psychological (e.g. anxiety) stressors.
Faced with an uncooperative patient it will also be prudent
to reduce nursing observations (measurements of blood pres- Learning point
sure, pulse, temperature, etc.) to an absolute minimum. Indeed,
Parkinson’s disease is a slowly progressive neurodegenerative dis-
such patients provide an ideal opportunity for staff to refine their
ease which does not change suddenly. Sudden apparent change in
people-observation skills and to observe whether: the individual
Parkinson’s symptomatology will always be due to the effects of
has started to accept or is taking less fluid and food; is more or less
some other event (physical or psychological), e.g. anxiety, infection,
socially interactive; is wandering more or less; is starting to mobilize
constipation, heat wave, or change in any medication.
or has become less mobile; and so on. These simple observational
Table 11.2 Conditions to exclude before diagnosing idiopathic Table 11.3 Diagnostic uncertainty in Parkinson’s in older age
Causes of diagnostic uncertainty in determining the presence
Essential tremor of Parkinson’s
Drug-induced parkinsonism ◆ The presence of comorbidities (such as arthritis, depression, dementia,
phenothiazines muscle weakness, involuntary movements)

◆ Nonspecific presentation (falls, depression, slowing down, fatigue)
prochlorperazine
metoclopramide ◆ Atypical presentation (dysphagia, pain, dysarthria)
◆ Tremor is common in older people and may be atypical and difficult to

tetrabenazine
classify
Arteriosclerotic pseudoparkinsonism (vascular parkinsonism)
◆ Tremor-dominant Parkinson’s disease
Multisystem atrophy (MSA)
◆ Minor signs of extrapyramidal disturbance in older people associated with
Progressive supranuclear palsy (PSP)
cognitive impairment
Dementia with Lewy bodies
Causes of diagnostic uncertainty in determining the cause
Other causes of tremor
◆ There are more causes of parkinsonism in older people.
drugs: lithium, amiodarone, SSRIs
◆ Drug-induced parkinsonism increases in frequency with age.
anxiety
◆ Levodopa responsiveness is unreliable in older people as a marker for
hyperthyroidism Parkinson’s disease.
MPTP exposure ◆ Vascular parkinsonism and parkinsonism associated with dementia become
Old age! more prevalent with age and create diagnostic difficulty.
For most patients with Parkinson’s disease, it is crucial that they to consider alternatives to oral medication or advise whether the
receive their medication at the appropriate time, whether or not this patient is in fact dying and medication is no longer appropriate.
fits in with the timing of institutional drug rounds. It is imperative
that the hospital ward does not ‘run out’ of the Parkinson’s medica-
tion, and that if a patient’s medication is non-ward stock (which it Learning point
usually will be), the ward must obtain a supply before the next dose ◆ Controlled release levodopa is inactivated if crushed.
of medication is due (whatever day of the week or time of day it is!).
◆ Standard Sinemet can be crushed.
Failure to maintain normal dosing schedules runs the risk of:
◆ Standard Madopar can be converted to equivalent dose of dis-
◆ worsening Parkinson’s symptoms
persible formulation.
◆ decreased mobility
◆ Giving medication in yoghurt may aid swallowing.
◆ increased rigidity and pain
◆ increased tremor A reasonable aide-memoire for clinical pharmacology is that
◆ slowed cognition drugs that target a specific organ of the body will preferentially
‘poison’ that organ, i.e. drugs that act on the brain (e.g. Parkinson’s
◆ with attendant increased anxiety and risk of complications
drugs) will have cerebral side effects, typically postural hypoten-
◆ dysphagia with inability to maintain hydration and nutritional sion, delirium, and agitation, and these will be more likely to occur
intake in those who already have evidence of cognitive impairment.
◆ aspiration or hypostatic pneumonia The neuropsychiatry of Parkinson’s disease is complex and fasci-
nating. The cause of anxiety can often be very difficult to determine.
◆ constipation
This may be a response to motor fluctuations, particularly if they
◆ urinary sepsis are unpredictable or a symptom of depression. Parkinson’s symp-
◆ incontinence toms may seem much worse than the disease actually is when the
individual is depressed. Depression occurs in around two-thirds
◆ pressure sores of community-dwelling Parkinson’s patients (Meara et al., 1999).
◆ falls Depressive symptoms of poor attention, poor initiation, poor
◆ delirium construction, and increased perseveration all mimic Parkinson’s
dementia, so if in doubt, a therapeutic trial of antidepressants
◆ depression may be warranted, whilst watching out for problems of postural
◆ loss of faith in health professionals hypotension.
Cognitive deficits in Parkinson’s are commonly: executive func-
◆ complaints and litigation.
tion (most prominent and may be earliest); higher order atten-
If a Parkinson’s patient has problems taking oral medication, e.g. tion; memory; and spatial skills (visuomotor processing and visual
due to dysphagia, acute illness, or severe psychomotor retardation, attention). Executive dysfunction underlies several impairments,
then urgent input from the Parkinson’s specialist will be needed including memory dysfunction and problems with verbal fluency,
reasoning, spatial skills, and complex attention. Executive dys- that these give. These behaviours can be difficult to treat and are
function will limit the extent to which Parkinson’s patients may be yet another source of stress for carers (Evans et al., 2009; Spencer
deemed competent to consent to medical treatments and/or clini- et al., 2011).
cal trials (Dymek et al., 2001). Frontal executive dysfunction can be
found in around one-third of newly diagnosed cases of Parkinson’s
disease (Foltynie et al., 2004) and is a predictor for risk of develop- Learning point
ing subsequent Parkinson’s dementia. The prevalence of dementia Dementia in Parkinson’s:
is 40% in all Parkinson’s patients (Cummings, 1988) and corre- ◆ has a frequency six times that of age-matched controls
lates with age (0% less than 50 years; 69% more than 80 years) and
◆ is more common
duration of Parkinson’s (29% after 3 years and 78% after 8 years)
(Lieberman et al., 1979; Brown and Marsden, 1984; Aarsland et al., ◆ with age of onset of Parkinson’s more than 60 years
2003). Parkinson’s dementia renders individuals more vulnerable ◆ in later stages of Parkinson’s
to drug toxicity from their Parkinson’s medications and more likely
to need their medication supervised. Concomitant dementia repre- ◆ with more severe Parkinson’s
sents a significant clinical problem in Parkinson’s disease, and with ◆ if psychosis and confusion develop with levodopa.
it comes: (Levy et al., 2002)
◆ increased carer strain (Aarsland et al., 1999a)
◆ increased risk of institutionalization (Aarsland et al., 2000)
Stroke
◆ institutionalization associated with increased mortality (Louis et New onset stroke should be referred urgently to the local stroke
al., 1997). physician for assessment of type and size of stroke and for appro-
Around 30% of Parkinson’s patients develop hallucinations priate management to be instigated, as well as for exclusion of pos-
within the first 5 years after diagnosis (Fenelon et al., 2000). These sible brain tumour. The latter tends to have a more insidious onset
become permanent in over 80% (Graham et al., 1997) and about of neurological deficit rather than the abrupt onset of cerebrovascu-
10% go on to develop psychosis with delusions (Jenkins and Groh, lar disease. Sudden severe headache with neurological deficit may
1970; Aarsland et al. 1999b). Psychosis is the single most impor- represent subarachnoid haemorrhage and requires urgent assess-
tant precipitant for long-term institutional care in Parkinson’s ment and investigation.
(Aarsland et al., 2000; Goetz et al, 2001). Failing cognition or Transient neurological deficit (hemiplegia, dysphasia, hemi-
psychosis may make it necessary to sacrifice mobility in order to anopia) resolving in under 24 h is compatible with a diagnosis of
minimize psychotic symptoms and thereby preserve the sanity of transient ischaemic attack (TIA). The neurological deficit in TIAs is
carers and prevent the individual with Parkinson’s disease being focal in nature and, therefore, does not result in loss of conscious-
institutionalized. Close collaboration will be required with the local ness. Individuals may fall due to their hemiparesis, but there will
Parkinson’s specialist in these circumstances as there is a complex not be any evidence of loss of consciousness. Transient neurologi-
interplay between symptoms and treatment effects when dealing cal signs with loss of consciousness should raise the possibility of
with the neuropsychiatry of Parkinson’s. There is good evidence for epilepsy or cardiac arrhythmia.
the potential benefit of rivastigmine in both dementia with Lewy
bodies (DLB) (McKeith et al., 2000) and dementia in Parkinson’s
disease (PDD) (Emre et al., 2004) (see Chapter 35). Learning point
Impulsive and compulsive behaviours are more common in ◆ Once infarction has occurred, it is impossible to further damage
Parkinson’s patients treated with dopamine agonists, although these that area of the brain, though lesser degrees of brain injury may
are not always acknowledged by Parkinson’s patients and it may be repeated.
require specific questioning of their family or carer(s). These can
◆ A person who has recurrent episodes of similar neurological
include mild stereotyped behaviour at the expense of other activi-
deficit without cumulative long-term deficit may have:
ties (punding) and often reflects previous hobbies or careers, e.g.
rearranging clothing in drawers, sorting and resorting coin collec- ◆ recurrent epileptic seizures with Todd’s paresis or
tions. These behaviours are generally not distressing to the patient ◆ be failing to compensate for residual stroke deficit when
or those around them, although they can interfere with day-to-day stressed by additional illness.
functioning. Impulse control disorders tend to have more destruc-
tive consequences for the patient, e.g. hypersexuality and compul-
sive shopping/gambling. These behaviours are driven by a craving Cerebrovascular disease is an important major cause of dementia,
or tension with a temporary release by specific goal achievement with atrial fibrillation perhaps being the single most important risk
or social reward. The patient is often more distressed by the behav- factor for multi-infarct dementia. Assessment of cognition is hin-
iours and can try unsuccessfully to resist them. Both conditions are dered in individuals who are dysphasic. Individuals with hemiano-
associated with dopamine agonist use, particularly at higher doses. pia or visuospatial problems secondary to parietal lobe dysfunction
Other risk factors include male gender, younger age, and previous may appear confused due to their dyspraxia or problems in negoti-
impulsive behaviour patterns. The dopamine dysregulation syn- ating their way through their environment due to hemianopia. For
drome is addictive behaviour involving Parkinson patients seek- a comprehensive review of dyspraxias and dysphasias the reader is
ing increasing doses of dopaminergic drugs as they ‘enjoy’ the high referred to Hodges (1994).
Routine administration of nutritional supplements to stroke converting enzyme inhibitors (ACEIs); a review of hydrational sta-
patients does not improve overall outcome and should be reserved tus and, in particular, a review of the need to continue diuretics and
for those who are undernourished on admission or have deteriorat- ACEIs in very hot weather or when the individual is suffering from
ing nutritional status after their stroke (FOOD, 2005). Early enteral a diarrhoeal illness; consideration of prostatic hypertrophy and
feeding in those who are dysphagic after stroke reduces mortality obstructive uropathy in men (abdominal palpation will reveal the
but increases the number of severely disabled survivors. As dys- distended bladder and a rectal examination will reveal the enlarged
phagia post-stroke recovers within 2 weeks in over 80% of cases prostate).
and most patients show signs of recovery within a few days, there
is no need to rush into enteral feeding unless it is necessary for the
management of diabetes or Parkinson’s disease. Learning point
Thyroid disease Urinary retention is uncommon in women and should lead to a
search for pelvic or rectal tumour as well as a vaginal examination
Hypothyroidism should be considered in any depressed or apathetic
to exclude tumour and prolapse.
individual. Symptoms of cold intolerance, constipation, tiredness,
and slowing of mental and physical ability are indistinguishable
from symptoms of depression. Alopecia, bradycardia, slow relaxing
Electrolyte imbalance
reflexes, hoarse or gruff voice, swelling of the face, ataxia, and, more
rarely, pretibial myxoedema will help to confirm a clinical diagnosis Hyponatraemia, hyper-, and hypocalcaemia may all cause delirium
of hypothyroidism. Anxiety and hyperactive, even paranoid states or they may present with more subtle symptoms such as tired-
may be due to thyrotoxicosis. Clinical signs of lid retraction, lid ness, constipation, and reduced mobility. Hyponatraemia is most
lag, and exophthalmos along with warm peripheries, fine tremor, frequently secondary to diuretic drugs and selective serotonin
and a tachycardia that persists during sleep all help confirm a clini- receptor inhibitor (SSRI) antidepressants. The syndrome of inap-
cal diagnosis of thyrotoxicosis. Less common is apathetic hyper- propriate antidiuretic hormone secretion (SIADH) may occur with
thyroidism. Measurement of serum thyroid stimulating hormone infection; renal, hepatic, cardiac, and pituitary dysfunction; or by
(TSH) will help in the diagnosis of thyroid disorder. A normal TSH inappropriate production of ADH in carcinoma of the bronchus.
excludes hypo- and hyperthyroidism. However, an abnormal TSH, Hypercalcaemia is much more common than hypocalcaemia and
both high and low, may be due to nonthyroid disease, e.g. sever- should lead to a search for possible malignancy, measurement of
ity of illness (sick euthyroid) or the effects of drugs (amiodarone, parathormone to exclude hyperparathyroidism, as well as a review
steroids). In general, a high TSH with a low serum thyroxine (T4) of medications such as calcium and vitamin D supplements.
indicates hypothyroidism; other combinations are likely to require Hypocalcaemia may be due to malabsorption, vitamin D defi-
specialist interpretation. ciency, and hypoparathyroidism.
Adrenal disease Urinary tract infections

General malaise, weight loss, hypotension, and falls may be due to These are common and often present nonspecifically with confu-
hypoadrenalism, which may come on insidiously or may come to sion (delirium), anorexia, or fatigue. Asymptomatic bacteriuria is
the fore more precipitously when an individual is stressed by infec- frequent in older people and does not require treatment; if patients
tion. Hyperadrenalism, or Cushing’s syndrome, is uncommon but are unwell and no other cause for their illness has been identified
may present with depression or psychosis, both of which may also then this is not asymptomatic bacteriuria! Dipstick urinalysis that
occur with use of high-dose steroids in conditions such as asthma. is positive for nitrites and leucocytes has a positive predictive value
of over 90% for urinary tract infection. Negative dipstick testing of
Diabetes urine will obviate the need to send a urine sample to the laboratory
Management of a psychiatric patient with diabetes should be a col- unless there is unexplained fever, rising inflammatory markers, or
laboration between the psychiatrist and the diabetologist. Particular delirium. All too frequently, older people admitted to hospital will
attention needs to be given to diabetic patients who, by virtue of be treated for a presumed urinary tract infection when there is no
their psychiatric illness, are either not eating or not taking their evidence to support the presumption, and the fact that they get bet-
medication consistently. Under such circumstances, the potential ter is mistakenly assigned to the treatment with antibiotics when in
for either hypo- or hyperglycaemic complications is significant and fact they were likely to get better anyway. Blind treatment with anti-
may warrant inpatient treatment. biotics runs the risk of antibiotic-associated diarrhoea (Clostridium
Careful assessment for complications of diabetes is required, difficile diarrhoea) and so should not be encouraged.
with particular attention being paid to blood pressure, signs of car-
Anaemia
diac failure, assessment of renal function including urinalysis for
proteinuria and evidence of nephropathy, fundoscopy to assess for This often presents with nonspecific symptoms of malaise, poor
retinopathy, assessment for peripheral neuropathy, and evidence of mobility, apathy, possibly even self-neglect, falls, and confusion
neuropathic or arterial ulceration. (delirium). Significant anaemia may occur due to extensive bruis-
ing after a fall, especially in those on antiplatelet agents or warfarin.
Renal disease Hypochromic microcytic anaemia is most likely to be due to blood
Renal failure is rarely a cause of psychiatric symptoms except in a loss from the bowel. This requires a review of medication (is the
profoundly sick individual. Deterioration in renal function requires individual on aspirin, warfarin, steroids, or NSAIDs?) and a thor-
a review of medication that may adversely affect renal function, ough abdominal examination, which must include a rectal examina-
e.g. nonsteroidal anti-inflammatory drugs (NSAIDs), angiotensin tion. Other possible sources of bleeding may need to be considered,
e.g. postmenopausal bleeding. Macrocytic anaemia may be due to Table 11.4 Consequences of unrelieved pain
alcohol excess, hypothyroidism, folate, or B12 deficiency and it may
ACUTE
contribute to cognitive impairment. Normochromic normocytic
anaemia is often associated with chronic diseases such as rheuma- Increased risk of complications:
toid arthritis and chronic kidney disease. delirium
Leukonychia and a smooth shiny tongue may be seen in chronic deep vein thrombosis
iron deficiency anaemia. nausea and vomiting
Giant cell arteritis (temporal arteritis) respiratory infections
Giant cell arteritis (GCA) usually presents with headaches, visual Increased mortality
disturbance, and tender scalp, and sometimes causes pain around ACUTE AND CHRONIC
the shoulder girdle and pelvis (polymyalgia rheumatica). It may Behavioural changes
also present nonspecifically with malaise, weight loss, fever, or Depression
stroke. There may be low-grade anaemia and mild abnormalities of Psychosocial effects:
liver function, and there will almost always be a raised erythrocyte isolation
sedimentation rate (ESR) or C-reactive protein (CRP). Temporal impaired mobility
artery biopsy may be required to make the diagnosis. Urgent treat- disrupted sleep
ment with steroids is required to prevent blindness and, therefore,
changes in social roles and relationships
urgent referral should be made to a geriatrician.
Increased length of hospital stay
Specific Problems Increased risk of institutionalization
Decrease in successful rehabilitation
Assessing pain in cognitively impaired patients
Litigation
Despite the high prevalence of pain amongst older people and its
many consequences, pain is inadequately recognized and treated,
especially in those with severe cognitive impairment. Barriers
to assessment of pain include the inability of some older people,
especially those with severe dementia, to communicate their pain Most of the available standardized pain assessment tools are pri-
experience; and the misconception that pain is less severe in those marily forms of self-report. The verbalization of pain can be diffi-
with cognitive impairment. Use of other informants, direct obser- cult for patients who have cognitive impairment. Not only may they
vation of potential pain indicators, monitoring for changes in usual have problems localizing pain or describing its temporal relation-
activity and behaviour, and ruling out pain as a possible cause of ship, but also they may even have problems saying whether or not
behaviours through nondrug and analgesic trials should improve they have pain. As dementia progresses and verbal skills decline,
the assessment and management of pain in those with cognitive carers, nursing, and medical staff must increasingly rely on non-
impairment. verbal cues of physical and emotional pain (Table 11.5). Common
Unrelieved pain can have several adverse effects for the individ- behaviours associated with pain are shown in Table 11.6, but some
ual (Table 11.4) (Briggs, 2003), and these may lead to inappropri- patients demonstrate little or no specific behaviour associated with
ate management, e.g. disruptive behaviour may be inappropriately severe pain. Use of facial expressions or various behaviours may be
treated by sedation rather than by assessment of and alleviation of difficult in patients with Parkinson’s disease or facial palsy.
the pain. Poor pain management can also result in increased com- The altered affective response to pain, especially in people with
plaints and litigation by family members (advocates). Thus, it is Alzheimer’s disease, may reflect pathology in the medial pain sys-
important that we understand how those with dementia perceive tem, resulting in an inability to cognitively process the painful sen-
pain, that we recognize their pain and treat it. A thorough system- sation in the context of prior pain experience, attitudes, knowledge,
atic assessment is required in cognitively impaired older people to and beliefs (Scherder et al., 2005). Reactions to painful sensations
reveal covert pathology and to investigate and remedy symptoms may therefore differ from the typical response expected from a cog-
such as pain. The given history may not be accurate or may be nitively intact older person. For example, constipation can cause
absent, which can mislead the clinician. great distress in the cognitively impaired older person and may
The experience of pain is inherently subjective and will be modu- lead to aggressive or agitated behaviours. As there is no evidence
lated by a variety of factors, which include: mood state; perception that those with dementia experience less pain, we should assume
of control; expectations; social conditioning; cultural conditioning; that any condition that is painful to a cognitively intact person
and cognition. In cognitively impaired individuals, both the experi- would also be painful to those with advanced dementia who can-
ence and expression of pain may be altered. This poses difficulties not express themselves. For example, pain should be considered as
in assessing pain, as all pain rating scales require a reasonably high a possible explanation for a change in behaviour in an older person
level of cognitive and language ability. Selecting accurate and useful with advanced dementia. Although subtle changes in usual pat-
assessment instruments for use in those with cognitive impairment terns of behaviour or activity do not always mean that the patient
is a major problem and becomes more problematic as cognition is in pain, they should raise the suspicion and lead to a thorough
declines. Stolee et al. (2005) reviewed 30 instruments for assessing evaluation for possible pain-causing problems. Simply observ-
pain in cognitively impaired older people and found that for most, ing an individual at rest may not identify pain behaviours, which
reliability and validity data were basic or nonexistent, and that none may only occur during activities such as transferring, walking, and
proved to be both valid and reliable. repositioning.
Table 11.5 Nonverbal cues in the expression of pain Adequate pain assessment forms the basis for optimal pain con-
trol, and it has major implications for quality of life (QoL) and
Agitation or irritability
quality of care of older people. Unrelieved pain has been associated
Repetitive verbalization/shouting with altered immune function, impaired psychological function
Aggression (e.g. depression, anxiety, and fear), impaired physical function (e.g.
Fluctuating cognition impaired mobility and gait, delayed rehabilitation, falls), sleep dis-
Falls/withdrawal turbance, compromised cognitive function, and decreased sociali-
Decreasing functional ability zation (American Geriatrics Society, 2002). These may all result
Sweating in increased dependency as well as increased use of healthcare
Tachycardia/raised BP resources, with resultant increased costs. In those with severe cog-
nitive impairment it is all too easy to attribute these effects to their
dementia, rather than to unrecognized and untreated painful condi-
tions. For instance, demented patients with persistent pain are more
likely to be treated with benzodiazepines and antipsychotics than
Table 11.6 Common pain behaviours in older people with cognitive
impairment their nondemented counterparts (Balfour and O’Rourke, 2003).
An empirical trial of analgesia may be warranted if pain behav-
Facial expressions iours persist after other possible causes are ruled out or treated. The
Slight frown; sad, frightened face choice of an appropriate analgesic is challenging because it is diffi-
Grimacing, wrinkled forehead, closed or tightened eyes cult to determine the level of pain severity in persons with advanced
Any distorted expression dementia. Starting with paracetamol seems prudent, whilst titration
Rapid blinking
to stronger analgesics may be necessary before ruling out pain as the
aetiology for behaviour or activity changes. If analgesic use results
Verbalizations, vocalizations
in decreased pain-related behaviours, it seems reasonable to assume
Sighing, moaning, groaning
that pain was the likely cause and to continue pharmacological and/
Grunting, chanting, calling out or nonpharmacological interventions. The increased susceptibility
Noisy breathing of cognitively impaired older people to adverse drug effects clearly
Asking for help necessitates very careful monitoring of any analgesic trial.
Verbally abusive
Body movements
Learning point
Rigid, tense body posture, guarding
Fidgeting ◆ Pain is underreported and undertreated in cognitively impaired
Repetitive rubbing of an area (perhaps indicating where pain is located) older people.
Increased pacing, rocking ◆ Poorly treated pain is associated with increased disability, depres-
Restricted movement sion, behavioural problems (inappropriate prescription of neu-
Gait or mobility changes roleptics), and worsening cognitive function.
Changes in interpersonal interactions ◆ Most pain assessment scales rely upon verbal skills.
Aggressive, combative, resisting care ◆ Decline in verbal communication skills with worsening demen-
Decreased social interactions tia makes assessment increasingly problematic.
Socially inappropriate, disruptive
◆ A multifaceted approach using a combination of self-reported
Withdrawn
measures, family or carer input, and measures of functional
Changes in activity patterns or routines impairment/change, along with physiological or behavioural
Refusing food, appetite change measures should improve the accuracy of pain assessment and
Increase in rest periods improve the subsequent management of pain in this vulnerable
Sleep, rest pattern changes group of older people.
Sudden cessation of common routines
Increased wandering
Mental status changes
Crying or tears
Learning point
Increased confusion In a communicative individual with cognitive impairment the fol-
lowing tips will assist in the assessment of pain:
Irritability
Other restless or irritated behaviour ◆ Frame questions in the here and now.
Pulling at tubes ◆ Use concrete questions with yes/no responses.
(Adapted from American Geriatrics Society Panel on Persistent Pain in Older Persons, 2002). ◆ Repeat the question.
(Continued)
◆ Use validating questions. pressure would suggest right heart failure and, in the context of
◆ Ensure communication aids (spectacles, hearing aids) are worn acute shortness of breath, should lead to the consideration of pul-
and functioning. monary embolism, especially if there is new onset atrial fibrillation.
Individuals who are immobile may only have sacral oedema, espe-
◆ Give adequate time for the individual to respond to questions. cially if they have been bedfast.
Other possible causes of oedema include hypoalbuminaemia,
which may be due to malnutrition, malignancy, or nephrotic
Sleep disturbance syndrome.
Sleep deprivation undermines daily performance and will hinder Assessment of the jugular venous pressure in old age can be very
people’s ability to both compensate for any disability and give their difficult, due to cervical flexion. Bilateral basal crepitations are
optimum in rehabilitation. It is difficult to maintain sleep hygiene common and do not in themselves constitute a diagnosis of heart
in hospital, especially in dormitory wards, so it is no wonder that failure. If heart failure is a possibility then the patient needs to be
individuals’ performance upon returning home is often better than assessed by a geriatrician or a cardiologist.
anticipated from their assessments in hospital. Poor sleep may be
due to: Weight loss
Weight loss in older age must always raise the possibility of under-
◆ noisy environment
lying malignancy. Clues to a possible primary site may steer inves-
◆ insomnia tigations, e.g. weight loss in an 80-year-old lifelong smoker should
◆ pain raise the possibility of lung cancer. Respiratory symptoms such as
cough and haemoptysis may be further clues to a primary lung
◆ adverse effects of medication
cancer; whilst altered bowel habit may suggest bowel cancer; and
◆ depression or anxiety enlarged lymph nodes may reflect lymphoma or, if confined to a
◆ prostatism particular lymphatic drainage area, may point to the likely primary
site, e.g. axillary lymph nodes suggesting breast cancer. Physical
◆ detrusor instability examination is not complete without a thorough examination of
◆ delirium all lymphatic sites, breasts, thyroid, rectal, and pelvic areas. Other
investigations will be guided by a process of probabilities, but a
◆ uncompensated cardiac failure
minimum would include:
◆ chronic obstructive airways disease
◆ chest X-ray
◆ restless legs syndrome
◆ full blood count
◆ obstructive sleep apnoea (OSA)
◆ ESR
◆ rapid eye movement sleep behaviour disorder (REMSBD)
◆ serum immunoglobulins and urine for Bence-Jones protein (to
◆ caring role. look for myeloma)
Oedema ◆ serum calcium
The commonest cause of swollen feet is venostasis oedema. This ◆ liver and renal function tests.
typically improves with foot elevation, so it often lessens or dis- Weight loss may be a manifestation of individuals’ inability to
appears overnight only to return when the individual is upright. care for themselves due to either physical or psychological disease.
A previous history of varicose veins or veno-occlusive disease Weight loss may also be part of the later stages of a disease process,
may be helpful clues. Clinical appearance of dry, scaly skin (vari- e.g. Parkinson’s and Alzheimer’s diseases.
cose eczema), brown or purple pigmentation due to haemosiderin Monitoring of current weights and review of previous weights
deposition in the subcutaneous tissues, venous ulceration, or the will determine whether a complaint of weight loss is real or not. A
inverted champagne bottle appearance of lipodermatosclerosis all slow decline in weight over many years may simply be part of the
provide diagnostic pointers. Venous support stockings may help if natural ageing process, with loss of muscle and bone mass as well as
the older person can put them on and tolerate wearing them. Before natural shrinkage of internal organs.
recommending support stockings, care should be taken to exclude
significant coexistent peripheral vascular disease by measuring the Dysphagia
ankle-brachial pressure index (ABPI); an ABPI between 1.1 and Swallowing is a complex mechanism with no identified higher cor-
0.8 is normal. Leg elevation is generally all that is required, as well tical control centre. Dysphagia may occur due to stroke of any size
as reviewing the necessity for medication that might induce fluid and at any site. There may be local problems such as ill-fitting or
retention, e.g. NSAIDs or steroids. As the problem is not one of absent dentures; and oral pathology, such as a squamous cell car-
fluid overload, diuretics are rarely necessary and are highly likely cinoma, needs to be excluded. Neuromuscular disorders, such as
to cause side effects, e.g. electrolyte imbalance, dehydration, and Parkinson’s disease, motor neuron disease, and multiple sclero-
postural hypotension. sis, may also cause dysphagia. Intrinsic obstructive disease of the
Unilateral leg oedema may be due to deep vein thrombosis. If oesophagus, such as oesophageal cancer, or extrinsic compression
there is full leg veno-occlusive disease (iliofemoral thrombosis), of the oesophagus from other cancers or from an enlarged left
rectal and pelvic examinations are mandatory to look for possible atrium must also be considered. There are obviously psychologi-
pelvic tumours. Bilateral leg oedema with a raised jugular venous cal causes of dysphagia and one must also consider the possibility
of oesophageal candidiasis, especially if the individual is malnour- help encourage and increase oral intake. The management of dia-
ished or has recently been taking antibiotics. This list is by no means betes and Parkinson’s may cause particular problems if oral intake
exhaustive. is poor or inconsistent, and artificial hydration and nutritional
Management of dysphagia should be considered a medical support may be needed whilst a diagnosis is being sought and an
emergency—how many of us would care to go days (or even a day) appropriate treatment plan instigated; in such circumstances, early
without food? Urgent referral to a physician/geriatrician for inves- resort to nasogastric feeding may be the most appropriate course
tigation of nonpsychological dysphagia is, therefore, important. of action.
Investigations are likely to include chest X-ray, barium swallow, The British Medical Association (2007) and General Medical
and upper gastrointestinal endoscopy. If the individual has com- Council (2010) guidance provide an explicit framework for mak-
plete dysphagia for fluids and solids, artificial nutritional support ing difficult decisions necessary to provide optimum care for
(parenteral or via nasogastric tube) may be required whilst a diag- individuals who are unable to maintain their own hydration and
nosis is being sought. Dependent upon the cause of dysphagia, con- nutrition and are not competent to make decisions for themselves.
sideration will need to be given to the option of long-term artificial Anorexia, weight loss, and dysphagia are common in advanced
hydration and nutrition. Most commonly, this would be by percu- dementia and they may be precipitated or worsened by intercur-
taneous endoscopic gastrostomy (PEG) feeding tube. Drug therapy rent infection, environmental change, depression, pain, poor oral
of Parkinson’s disease, sadly, provides limited improvement for dys- hygiene, ill-fitting dentures, poor carer rapport, and lack of carer
phagic symptoms in Parkinson’s. support. In the absence of a reversible cause for declining oral
intake there is no evidence that artificial nutritional support using
a PEG tube improves the prognosis in advanced dementia; there is
Learning point no evidence that this intervention reduces aspiration risk, prolongs
Dysphagia is a medical emergency—consider how long you would survival, improves quality of life, improves nutritional or func-
be prepared to go without food and do not allow your patients to be tional status, or is well tolerated (Finucane et al., 1999; Mitchell
subjected to a longer period of starvation! et al., 2000, 2004). Where dietary intake is inadequate but death
is not imminent, a second opinion should be sought from a senior
clinician not involved in the individual’s care before the decision to
Managing hydration and nutrition withhold artificial feeding is finalized (General Medical Council,
Assessing hydrational status in older age can be very difficult due 2010).
to reduced skin turgor, wasting of facial muscles, and difficulties in
visualizing the jugular venous pressure. Postural hypotension or,
for those who cannot stand, a fall in blood pressure on changing Learning point
from lying to sitting in bed may be the only reliable clinical sign of
◆ Reduced oral intake is part of the natural dying process and needs
volume depletion.
to be differentiated from potentially treatable causes.
Undernutrition is a strong and independent predictor for mor-
bidity and mortality. Older people may have difficulty maintain- ◆ PEG tube feeding in advanced dementia improves neither quality
ing adequate fluid and food intake due to the effects of ill health, nor quantity of life.
reduced mobility, delirium, dysphagia (remember oral and ◆ Before making a decision to withhold artificial hydration and
oesophageal thrush after a course of antibiotics), depression, and nutrition from an individual who lacks capacity, a second opin-
dementia. These all need to be differentiated from the reduction ion should be obtained from an independent senior clinician.
in oral intake that is part of the natural process of dying. If the
individual is not dying then an appropriate management plan will
depend upon the underlying cause for the inability to maintain
adequate hydration and nutrition. Formal assessment of swal- Palliative Care
lowing by a Speech and Language Therapist with advice on how Most deaths occur in older age but sadly only one in four of us will
to enable swallowing, e.g. thickened fluids, and dietetic advice die at home. Thus, a key part of the assessment of any older person
regarding nutritional support for those who are undernourished should be the ability to recognize when death is imminent or when
or at risk of becoming undernourished is essential and should be treatments are futile, in order that individuals, their family, and
obtained as a matter of urgency. Without food and fluids, death other carers (health and social) can be prepared for death and not
will usually occur in less than 2 weeks. Without any food but with strive officiously to try to prevent the inevitable (Finucane, 1999).
fluids (even if this is only a small amount) individuals may sur- Death does not always give warning of its proximity, but one should
vive many weeks (this is similar to those on hunger strike), whilst always be on the lookout for it, especially in advanced or metastatic
with any fluid and miniscule amounts of food they may survive cancer; end-stage Parkinson’s or dementia; as cardiac, respiratory,
for months. renal, or hepatic failure progresses. Death may be heralded by:
Consideration needs to be given to the presentation of food (por- ◆ failure to respond as anticipated to supportive therapy, e.g. lack of
tion size, temperature, presentation), good oral hygiene, ensuring response to increasing doses of diuretics in cardiac failure
dentures are worn and well fitting, and the use of subcutaneous flu-
ids given overnight if daytime oral intake is inadequate. Overnight
◆ lack of response to potentially curative treatments, e.g. failure of
fluid supplementation has the advantage of allowing individuals to infection to respond to antibiotics
drink as much as they are able by day and does not restrict mobility ◆ increasing frequency of complications, e.g. rapid recurrence of
or rehabilitation with intravenous lines. Nutrition assistants can also aspiration pneumonia in dementia or Parkinson’s
◆ persistent coma after stroke the physical illness. Failure of an individual to achieve expected
◆ reducing fluid and nutritional intake in the absence of an obvious rehabilitation goals may be due to the development of depression, or
correctable cause. else that the patient does not share the same goals as the rehabilita-
tion staff, e.g. lack of engagement with therapists after a fall may be
The degree of warning that death is approaching may be very
because discharge from hospital means returning to a role of carer
short, although in retrospect one can often appreciate that the
for their demented spouse. In this example, a discussion with indi-
warning signs were there for some time before they were recog-
viduals will reveal an unmet need for home care for their spouse and
nized. There is an acknowledged need for palliative care in chronic
a need for a break from the caring role. Completing their rehabilita-
progressive neurological diseases such as dementia (Hughes et al.,
tion as an inpatient rather than using an early supported discharge
2005; National Council for Palliative Care, 2006).
scheme may benefit patients in the longer term, as the rehabilitation
Once it has been recognized that the individual is dying, appropri-
process offers a form of respite from their caring role.
ate palliative care can be instigated, with particular attention to pain
Common, medically unexplained symptoms that are due to psy-
control, anxiety and depression, nausea, breathlessness, delirium,
chological illness include fatigue, chest pain, dizziness, headache,
constipation, insomnia, cough, hiccoughs, and any other symptoms
back pain, and abdominal pain. A plethora of investigations will
that might occur (Ellershaw and Ward, 2003; Marie Curie Palliative
be normal, but individuals, and often their family too, will not be
Care Institute, 2009; Gold Standards Framework, 2013). Individuals
satisfied and seek consultations with multiple specialists. In older
who are conscious and clear in mind may wish to say ‘good-bye’ to
age, there is often an underlying concern that the individual has
their loved ones and put their affairs in order. Likewise, their family
cancer or other incurable disease; or this behaviour may emanate
and friends need to prepare themselves for the death and the griev-
from social isolation or a fear of dependency. Recognizing the indi-
ing process, and may also need to say ‘good-bye’. Health and social
vidual’s concerns can allow the clinician to tackle the psychological
care staff also need to be prepared for the person’s death and be con-
problem and prevent unnecessary and repetitive investigations.
fident of their role in the palliative care process.
A stroll through my ward provides me with a good barometer as
to how stressed the staff are. When staff are stressed, this often turns
Learning point out to be a transfer of anxieties from patients’ relatives. Troublesome
families, those that appear to demand inordinate amounts of time
◆ A subtle pointer to the fact that people may be coming towards from healthcare staff, are rarely the ‘families from hell’ that they
the end of their life is when the ‘acute physicians’ start to rec- are often portrayed to be. Of course, a small minority will never be
ognize them. This suggests frequent multiple admissions and satisfied no matter how good the care is. However, in general, the
a trajectory of chronic disease that may be reaching terminal family’s anxieties stem from poor communication; concern regard-
velocity. ing the nature, severity, and consequences of their loved one’s ill-
◆ Ask yourself ‘Would I be surprised if this person died in the next ness; failure to grasp the information given to them (either it has
6 months?’ If the answer is ‘no’, then consider the principles of the not been made simple enough or it was not detailed enough); lack
Gold Standards Framework. of continuity in carers; concern regarding discharge planning; or
correctly identifying deficiencies in care (inadequate staffing levels
or poor standards of care) (Table 11.7). Being aware of the ward
Psychological Manifestations of Disease ‘barometric pressure’ can allow timely input by a senior clinician
There are social and psychological aspects of all illnesses and it can at an early stage to deal with anxieties (family and staff ) and defuse
be difficult to tell if psychological problems have taken over from potential complaints.
Table 11.7 Common family concerns and likely explanation

Concern Explanation
Discharge will be too early ◆ They feel home or residential care is needed
◆ They were caring for their relative and have a holiday booked
Discharge home will be ◆ They do not understand what care is available
impossible ◆ They do not appreciate to what extent recovery is possible
◆ They are suffering from carer strain and need extra care
(this is often a family or friends who have managed without external help and now find the care needs are beyond their capabilities—
perhaps due to their own ill health or ageing)
Nobody seems to know ◆ Inconsistent staff
what is happening to my ◆ Overreliance on agency and locum staff
relative ◆ Failure to appreciate that medical staff work shifts and not understanding that out-of-hours staff are not ward based
◆ No firm diagnosis yet achieved (especially difficult for some to understand that medicine cannot always provide the answers)
I get conflicting reports as ◆ Lack of consistency in staff member liaising with the family
to how my relative is doing ◆ Inexperienced staff liaising with the family (nursing or medical)
◆ Staff not adequately explaining delirium
Conclusion References
In order that discharge planning may start on day one, frail older Aarsland, D., et al. (1999a). Mental symptoms in Parkinson’s disease are
people must have access to early specialist comprehensive geriatric important contributors to caregiver distress. International Journal of
assessment (CGA) (British Geriatrics Society, 2010), for it is difficult GeriaticPsychiatry, 14, 866–74.
to steer a course if you do not know what the problems (diagnoses) Aarsland, D., et al. (1999b). Prevalence and clinical correlates of psychotic
symptoms in Parkinson disease: a community-based study. Archives of
are. CGA is a multiprofessional endeavour and implies a team-based Neurology, 56, 595–601.
approach with clear understanding of community-based care and Aarsland, D., et al. (2000). Predictors of nursing home placement in
rehabilitation schemes. Admission assessment should include a Parkinson’s disease: a population-based, prospective study. Journal of the
structured description of the patient’s premorbid function and American Geriatrics Society, 48, 938–42.
mobility. This should be corroborated by discussion with relatives, Aarsland, D., et al. (2003). Prevalence and characteristics of dementia in
or professionals who know the patient, as acute illness, pain, frailty, Parkinson’s disease: an 8 year prospective study. Archives of Neurology,
and cognitive impairment, as well as fears for the future, may com- 60, 387–92.
promise the accuracy of the patient’s description. American Geriatrics Society Panel on Persistent Pain in Older Persons (2002).
Clinical practice guidelines: the management of persistent pain in older
The art within the science of physical assessment of older peo-
persons. Journal of the American Geriatrics Society, 50, S205–24.
ple is to engage eyes and ears with brain and only touch when
Balfour, J.E. and O’Rourke, N. (2003). Older adults with Alzheimer disease,
necessary! Excellent observational skills are paramount and no comorbid arthritis and prescription of psychotropic medications. Pain
clue must be missed or ignored (a good doctor should be a good Research and Management, 8, 198–204.
detective). Fathoming the contribution of each and every patho- Briggs, E. (2003). The nursing management of pain in older people. Nursing
logical process in those with multipathologies can be complicated Standard, 17, 47–53.
but immensely rewarding for clinician and patient. Apparently British Geriatrics Society (2005). Compendium document. Abuse of older
minor changes to each of several pathological processes can eas- people. <www.bgs.org.uk/Publications/Compendium/compend_4–10.
ily multiply to have a major impact on an individual’s physical htm> (accessed 20.12.2011).
British Geriatrics Society (2010). Best Practice Guide 3.5. Comprehensive
wellbeing and quality of life. Complex decisions may be required
assessment of the older frail patient. <http://www.bgs.org.uk/Publications/
as to the order of interventions most likely to provide best results Compendium/compend_3–5.htm> (accessed 20.12.2011).
(akin to those puzzles that require you to determine the least British Medical Association. (2007). Withholding and withdrawing
number of moves needed to rearrange the puzzle into a certain life-prolonging medical treatment: guidance for decision making, 3rd
order). edition. BMA, London.
The concept of atypical presentation of disease in older age has Brown, R.G., and Marsden, C.D. (1984). How common is dementia in
been overstated (Isaacs, 1992). Falls, immobility, incontinence, Parkinson’s disease? Lancet, ii, 1262–5.
delirium, and drug toxicity are typical presentations of pathol- College of Optometrists and British Geriatrics Society (2011). The importance
ogy in frail older people. They result directly from physiological of vision in preventing falls. <http://www.bgs.org.uk/pdf_cms/reference/
vision-in-falls.pdf> (accessed 01.2012).
changes of ageing that can render an older person more vulner-
Cummings, J.L. (1988). The dementias of Parkinson’s disease: prevalence,
able to stress and to changes in homeostatic processes, such as fluid characteristics, neurobiology, and comparison with dementia of the
balance (Rockwood et al., 2004). Thus, functions that require inte- Alzheimer type. European Neurology, 28 (Suppl 1), 15–23.
gration of higher order cortical process, such as staying upright, Cummings, S.R., Nevitt, M.C., and Kidd, S. (1988). Forgetting falls: the
maintaining balance, or walking, are more likely to fail, resulting in limited accuracy of recall of falls in the elderly. Journal of the American
falls, immobility, or delirium. These are not ‘social’ admissions or Geriatrics Society, 36, 613–16.
‘acopia’. These presentations of disease in older age demand rigor- Department of Health (2001). National Service Framework for Older People.
ous assessment and interventions. If individuals cannot cope, it is DH, London.
either because they have lost physical function, cognition, or confi- Dymek, M.P., et al. (2001). Competency to consent to medical treatment in
cognitively impaired patients with Parkinson’s disease. Neurology, 56,
dence, or because their social support network can no longer cope
17–24.
with their established care needs—either for physical or emotional/ Ellershaw, J. and Ward, C. (2003). Care of the dying patient: the last hours or
psychological reasons. The latter would be a social presentation and days of life. British Medical Journal, 326, 30–4.
would not require hospital admission but urgent social services Emre, M., et al. (2004). Rivastigmine for dementia associated with Parkinson’s
input or respite care. disease. New England Journal of Medicine, 351, 2509–18.
In patients with dementia, failure of language can provide par- Evans, A.H., et al. (2009). Impulsive and compulsive disorders in Parkinson’s
ticular problems in assessing pain, whilst failing cognition and lan- disease. Movement Disorders, 24, 1561–70
guage can hinder the neurological examination. Fenelon, G., et al. (2000). Hallucinations in Parkinson’s disease: prevalence,
Psychological problems for the patient can slow recovery and phenomenology and risk factors. Brain, 123, 733–45.
hinder discharge planning. They need to be recognized, discussed, Finucane, T.E. (1999). How gravely ill becomes dying: a key to end-of-life
care. Journal of the American Medical Association, 282, 1670–2.
and appropriately managed. So-called problem families are often
Finucane, T., Christmas, C., and Travis, K. (1999). Tube feeding inpatients
not the problem, and staff need to focus on the real issues, e.g. carer with advanced dementia: a review of the evidence. Journal of the
stress or poor communication, rather than becoming defensive or American Medical Association, 282, 1638–45.
antagonistic. Folstein, M.F., Folstein, S.E., and McHugh, P.R. (1975). ‘Mini-mental state’:
All healthcare professionals should remember that older peo- a practical method for grading the cognitive state of patients for the
ple do not choose to be ill or demented and that we are there to clinician. Journal of Psychiatric Research, 12, 189–98.
serve their needs, improve the quality of their lives, and, if possible, Foltynie, T., et al. (2004). The cognitive ability of an incident cohort of
increase the quantity. Parkinson’s patients in the UK. The CamPaIGN study. Brain, 127,
550–60.
FOOD Trial Collaboration. (2005). Effect of timing and method of enteral International Journal of Geriatric Psychiatry, 21, 1078–85. <http://egret.
tube feeding for dysphagic stroke patients (FOOD): a multi-centred psychol.cam.ac.uk/medicine/scales/MioshI_2006_ACE.pdf> (accessed
randomised controlled trial. Lancet, 365, 764–72. 29.11.2011).
General Medical Council (2010). Treatment and care towards the end of life: Mitchell, S., et al. (2000). A cross-sectional survey of tube-feeding decisions
good practice in decision making. http://www.gmc-uk.org/guidance/ in cognitively impaired older persons. Journal of the American Geriatrics
ethical_guidance/end_of_life_care.asp>(accessed 20.12.2011). Society, 48, 391–7.
Goetz, C.G., et al . (2001). Prospective longitudinal assessment of Mitchell, S., et al. (2004). Tube-feeding versus hand-feeding nursing home
hallucinations in Parkinson’s disease. Neurology, 57, 2078–82. residents with advanced dementia: a cost comparison. Journal of the
Gold Standards Framework (2013). <www.goldstandardsframework.org.uk> American Medical Directors Association, 5, S23–9.
(accessed 18.02.2013). Muller, R.T., et al. (1990). Painless myocardial infarction in the elderly.
Graham, J.M., Grunewald, R.A., and Sagar, H.J. (1997). Hallucinosis in American Heart Journal, 119, 202–4.
idiopathic Parkinson’s disease. Journal of Neurology Neurosurgery and National Council for Palliative Care (2006). Exploring palliative care for people
Psychiatry, 63, 434–40. with dementia. A discussion document. NCPC, London.
Ham, S.J., van Walsum, A.D., and Vierhout, P.A. (1996). Predictive value of NICE (2005). Clinical Guideline 29. Pressure ulcers: the management of
the hip flexion test for fractures of the pelvis. Injury, 27, 543–4. pressure ulcers in primary and secondary care. <http://www.nice.org.uk/
Hodges, J.R. (1994). Cognitive assessment for clinicians. Oxford University nicemedia/live/10972/57623/57623.pdf> (accessed 09.01.2012).
Press, Oxford. NICE (2010). Clinical Guideline 103. Delirium: diagnosis, prevention,
Hughes, J., Robinson, L., and Volicer, L. (2005). Specialist palliative care in and management . <http://www.nice.org.uk/nicemedia/
dementia. British Medical Journal, 330, 57–8. live/13060/49909/49909.pdf>(accessed 09.01.2012).
Inouye, S.K., et al. (1990). Clarifying confusion: the confusion assessment Rockwood, K., Mtinski, A.B., and MacKnight, C. (2004). Some mathematical
method. A new method for detection of delirium. Annals of Internal models of frailty and their clinical implications. Age and Ageing, 33, 430–2.
Medicine, 113, 941–8. Royal College of Physicians (2006). The prevention, diagnosis and management
Isaacs, B. (1992). The challenge of geriatric medicine. Oxford University Press, of delirium in older people: national guidelines. Royal College of Physicians/
Oxford. British Geriatrics Society, London. <http://www.rcplondon.ac.uk/sites/
Jenkins, R.B. and Groh, R.H. (1970). Mental symptoms in Parkinsonian default/files/concise-delirium-2006.pdf> (accessed 09.01.2012).
patients treated with L-dopa. Lancet, ii, 177–9. Royal Pharmaceutical Society of Great Britain (1997). From compliance to
Jorm, A.F. (1994). A short form of the Informant Questionnaire on Cognitive concordance: achieving shared goals in medicine. RPS, London.
Decline in the Elderly (IQCODE): development and cross-validation. Royall, D.R., Cordes, J.A., and Polk, M. (1998). CLOX: an executive clock
Psychological Medicine, 24, 145–53. drawing task. Journal of Neurology Neurosurgery and Psychiatry, 64,
Levy, G., et al. (2002). Combined effect of age and severity on the risk of 588–94.
dementia in Parkinson’s disease. Annals of Neurology, 51, 722–79. Scherder, E., et al. (2005). Recent developments in pain in dementia. British
Lieberman, A., et al. (1979). Dementia in Parkinson disease. Annals of Medical Journal, 330, 461–4.
Neurology, 6, 355–9. Sharaf, H., Fernihough, H., and Forsyth, D. (2006). Hearing aids: a clinician’s
Louis, E.D., et al. (1997). Mortality from Parkinson disease. Archives of guide. CME Journal Geriatric Medicine, 8 (2), 80–5.
Neurology, 54, 260–4. Spencer, A.H., et al. (2011). The prevalence and clinical characteristics of
Marie Curie Palliative Care Institute (2009). Liverpool Care Pathway for the punding in Parkinson’s disease. Movement Disorders, 26, 578–86.
Dying Patient. <http://www.liv.ac.uk/mcpcil/liverpool-care-pathway/ Stolee, P., et al. (2005). Instruments for the assessment of pain in older
documentation-lcp.htm#core> (accessed 12.2011). persons with cognitive impairment. Journal of the American Geriatrics
McKeith, I., et al. (2000). Efficacy of rivastigmine in dementia with Lewy Society, 53, 319–26.
bodies: a randomised, double-blind, placebo-controlled international Walma, E.P., et al. (1997). Withdrawal of long-term diuretic medication in
study. Lancet, 356, 2031–6. elderly patients: a double blind randomised trial. British Medical Journal,
Meara, R.J., Mitchelmore, E., and Hobson, J.P. (1999). Use of the GDS-15 as 315, 464–8.
a screening instrument for depressive symptomatology in patients with White, S., et al. (2005). Enzymes of drug metabolism. Age and Ageing, 34,
Parkinson’s disease and their carers in the community. Age and Ageing, 603–8.
28, 35–8. Yesavage, J.A. and Brink, T.L. (1983). Development and validation of a
Mioshi, E., et al. (2006). The Addenbrooke’s Cognitive Examination geriatric depression screening scale: a preliminary report. Journal of
Revised (ACE-R): a brief cognitive test battery for dementia screening. Psychiatric Research, 17, 37–49.
CHAPTER 12
Neuroimaging
Claire E. Sexton, Verena Heise,
and Klaus P. Ebmeier
With dramatic developments in imaging technology over the past matter hyperintensities using T2-weighted MRI or fluid attenuated
few decades, there has been a surge of neuroimaging studies, par- inversion recovery (FLAIR) images, white matter integrity using
ticularly in old age psychiatry. Age-related brain changes, together diffusion tensor imaging (DTI) (Fig. 12.2), and neuronal activation
with specific pathological abnormalities, have generated a host of patterns using functional MRI (fMRI).
case-control studies that illuminate the aetiology and pathophys-
iology of brain disease. However, little has as yet been translated T1-weighted MRI
into clinical practice. For most imaging protocols the necessary The physical principles of MRI are generally based upon the nuclear
large-scale naturalistic studies of, for example, memory clinic MR of protons. Briefly, protons possess magnetic spins that, in the
attenders with a validated diagnosis, which allow for the estimation absence of a magnetic field, are randomly oriented. An MRI scanner
of realistic sensitivity, specificity, and predictive values, are largely consists of a large magnet that exerts a powerful external magnetic
missing. To cover this division of neuroimaging into scientific clin- field (B0) in which spins align to create net magnetization in the
ical enquiry and clinical routine, the first section of this chapter direction of the magnetic field. During an MRI scan, application
aims to provide a brief overview of the theory behind neuroimag- of a radiofrequency pulse tilts the net magnetization away from B0.
ing techniques and outline commonly used analysis methods. The The spins then realign with B0, with the rate of recovery of mag-
second section will cover the current and future clinical applica- netization along the longitudinal axis, and the loss of magnetiza-
tions of neuroimaging. Particular focus will be given to the role tion in the transverse plane, defined by the time constants T1 and
of neuroimaging in making a diagnosis, predicting outcome, and T2, respectively. The precessing magnetization can be detected with
understanding illness mechanisms. a radiofrequency-tuned coil, with the signal dependent upon the
proton density, the T1 and T2 relaxation constants, and the rela-
Theory tive timing of external radiofrequency pulses and field gradients.
Since proton density, T1, and T2 are tissue-specific properties, it is
Computed X-ray tomography possible to vary the scan repeat time (repetition time, TR) and the
Computed tomography (CT) is a structural imaging technique time between transmitting a radiofrequency pulse and measuring
that is based upon the degree of attenuation of X-rays in neural a signal (echo time, TE) in an MRI pulse sequence in order to be
tissue, i.e. their electron density. Attenuation is greater in hyper- sensitive to proton density, T1, or T2 and create the relevant image
dense tissues, such as bone consisting of calcium [20Ca], phosphate contrast. Spatial position can be encoded by imposing a gradient
[15P, 8O], and carbonate [6C, 8O] minerals, which appear white in a on the field strength of the underlying magnetic field, as there is a
CT image, compared with hypodense tissues, such as cerebrospinal direct relationship between field strength B0 and the resonance radi-
fluid (CSF) [1H, 8O], which appear dark. Compared with magnetic ofrequency, so that frequency can be directly translated back into
resonance imaging (MRI), soft tissue contrast is low, and dense spatial position.
bony structure can obscure underlying soft tissue signals (e.g. pos- T1-weighted MRI provides particularly good contrast between
terior fossa). The further development of spiral CT scanners has grey matter and white matter and is thus the preferred modality
provided significantly improved resolution and speed of scanning, to isolate and examine grey matter. In T1-weighted images, grey
resulting in a reduced radiation dose (Fig 12.1). matter appears dark grey, white matter light grey, and CSF black.
The most common analysis techniques use regions-of-interest
Magnetic resonance imaging (ROI) and voxel-based morphometry (VBM). ROI analysis of grey
MRI is a noninvasive technique that can provide high resolution matter volumes has advanced from tracing brain regions by hand
in vivo images of the brain. Different MRI sequences allow various to fully automated segmentation methods (Babalola et al., 2009).
aspects of the brain’s structure and function to be investigated. Grey Furthermore, shape analysis techniques can provide greater detail of
matter is most commonly studied using T1-weighted MRI, white the location of structural changes in subcortical structures (Styner
Fig. 12.1 Computed X-ray tomography (CT).
Fig. 12.2 Structural MRI. Examples of (a) T1-weighted

axial image, (b) T2-weighted image, and (c) a fractional
anisotropy (FA) map derived from DTI data in the same
plane.
et al., 2003; Patenaude et al., 2011). However, whilst automated seg- from CSF so that it appears black on the MR image, which increases
mentation methods overcome some of the limitations associated the contrast between CSF and WMH and improves recognition of
with time consuming and expert-dependent manual techniques, periventricular WMH. FLAIR images are now often used alongside,
differences occurring outside of the ROIs go unobserved. or instead of, T2-weighted images in the assessment of WMH.
VBM can provide an assessment of grey matter across the whole Functional magnetic resonance imaging
brain. VBM typically involves: segmentation of T1-weighted images
fMRI is an MRI technique that provides an indirect measure of neu-
into grey matter, white matter, and CSF; registration of images into
ral activity by measuring the haemodynamic response to neuronal
a standard space to make them comparable between different sub-
activity. Increased neuronal activity is associated with an increase
jects; optional intensity modulation of images to compensate for
in blood flow that exceeds the increase in oxygen consumption,
the amount of expansion or contraction of each image; and spatial
resulting in an increase in the ratio of oxy- to deoxyhaemoglobin.
smoothing of images (Ashburner and Friston, 2000; Good et al.,
As haemoglobin is diamagnetic when oxygenated, but paramag-
2001). A voxel-by-voxel whole-brain statistical comparison of grey
netic when deoxygenated, neuronal activity leads to a decrease in
matter can then be made between subject, between diagnostic
magnetic susceptibility and an increase in the blood oxygen lev-
groups, and along dimensions, such as age or cognitive function.
el-dependent (BOLD) signal (Matthews and Jezzard, 2004).
VBM is limited by the quality of interparticipant image registration
The majority of fMRI studies to date have compared the BOLD
and the amount of spatial smoothing of data (Bookstein, 2001).
signal between a baseline and an activation condition. Particularly
T2-weighted MRI/FLAIR relevant to psychiatric research are fMRI experiments that assess
T2-weighted images are obtained by manipulating differences in activation during cognitive tasks or exposure to emotional stimuli,
T2-relaxation times. In T2-weighted images, grey matter appears such as happy, sad, angry, fearful, and neutral faces. Stimuli can be
light grey, white matter dark grey, and CSF white. Periventricular presented in a variety of ways, with block, event-related, mixed, and
and deep white matter hyperintensities (WMH) are often assessed self-driven experimental designs commonly employed (Amaro and
using visual rating scales on T2-weighted images. However, it can Barker, 2006). Analysis of task-based fMRI typically involves cor-
be difficult to differentiate between periventricular WMH and CSF relating the time course of BOLD signal change in each voxel with a
in T2-weighted images. By using a FLAIR sequence it is possible to model time course based on the expected neural response.
null the signal from any particular tissue so that it appears black on A recent development in fMRI has been the focus on activity in
the MR image. When studying WMH it is useful to null the signal the resting brain. In the resting brain, the BOLD signal exhibits
CHAPTER 12 neuroimaging 179
low-frequency spontaneous fluctuations (0.01–0.1 Hz) that are Both are not routinely available for regional cerebral blood flow
temporally correlated and correspond to resting-state networks scanning and require further development, before their theoretical
(RSNs). Commonly identified networks include: visual, auditory, advantage of not using ionizing radiation can be fully exploited.
sensory-motor, default-mode, executive control, and frontal-parietal
Diffusion tensor imaging
RSNs. The amplitude of the BOLD signal at rest is comparable with
that displayed in task-related experiments (Damoiseaux et al., DTI is an MRI technique that can identify the main orientation and
2006). Furthermore, there is close correspondence between the determine integrity of white matter tracts in vivo by measuring the dif-
major brain networks identified at rest and those identified during fusion of water in neural tissue. If there is unrestricted mobility of water
task-related fMRI experiments (Smith et al., 2009). molecules in all directions, diffusion is described as isotropic (‘turns
The two most popular methods for analysis of functional con- equally in all directions’). In contrast, if there is restricted mobility of
nectivity within RSNs are single seed region analysis and inde- water molecules in any direction, diffusion is described as anisotropic.
pendent components analysis (ICA). Single seed region analyses Within white matter, it is thought that the parallel arrangement of
assess the temporal correlation between the BOLD time course of axonal membranes plays the primary role in restricting perpendicular
a predefined seed region and that of either a priori defined target water diffusion and generating anisotropy, while myelination acts to
regions or the whole brain on a voxel-wise basis. Single seed region enhance the degree of anisotropy (Beaulieu, 2002).
analysis can only isolate a single network based upon the a priori The three dimensions of diffusion can be modelled as an ellip-
defined seed region. For example, a seed in posterior cingulate cor- soid whose shape is characterized by the three eigenvalues (λ1, λ2,
tex/precuneus is often used to investigate the default mode network λ3). If diffusion is isotropic, diffusion in all three main axes is equal
(DMN), a network that is active when subjects are resting in the (λ1 = λ2 = λ3) and the diffusion ellipsoid becomes a sphere. If dif-
scanner and not engaged in a task. fusion is anisotropic (λ1 > λ2, λ3), the major eigenvector reflects
ICA, in contrast, decomposes the whole dataset into a set of sta- the direction of maximum diffusivity, which, in turn, is assumed
tistically independent spatial component maps and associated time to reflect the orientation of fibre tracts (Fig. 12.3). DTI can there-
courses without a prespecified model. Voxel values reflect the cor- fore be used to study the orientation and integrity of white matter
relation between the voxel’s time series and the mean time series tracts by estimating the degree of anisotropy (fractional anisotropy,
of that particular RSN (Beckmann and Smith, 2004; Beckmann FA), the overall displacement of molecules (mean diffusivity, MD;
et al., 2005). ICA can detect and separate several RSNs at once. trace; apparent diffusion coefficient), axial diffusivity (DA, diffusiv-
As with other data reduction component analyses, the spatial pat- ity parallel to the white matter tract), and radial diffusivity (DR,
tern of RSNs generated using ICA varies according to the number diffusivity perpendicular to the white matter tract) (Fig. 12.4). Such
of components allowed. Moreover, if ICA is performed for each measurements can be extracted locally in previously defined ROI
participant separately, it can be difficult to identify the same net- analysis or by tractography, or they can be measured globally, using
work consistently across participants. An alternative approach is VBA or tract-based spatial statistics (TBSS).
to perform ICA across all participants, and then, based upon the ROI analysis involves placement of an ROI of a fixed size and
group-level components identified, generate individualized RSNs shape over a predefined white matter region. Such an approach
involving both spatial and temporal regression (Filippini et al., typically only studies a small portion of a white matter tract and is
2009; Zuo et al., 2010). RSNs identified in this fashion are remark- insensitive to differences occurring outside of, or overlapping, the
ably robust across thousands of resting scans (Smith et al., 2009). ROI. Tractography algorithms, moving from voxel to voxel via the
Blood flow magnetic resonance imaging

(arterial spin labelling) (a) λ1 (b)
λ1
The most commonly used MR techniques to image perfusion, that
is, the rate at which nutrient-supplying blood passes through neu-
ral tissue, are dynamic susceptibility contrast MRI (DSC-MRI) and
arterial spin labelling (ASL). λ2 λ3
DSC-MRI is a relatively quick technique that uses gadolinium
λ2
as an exogenous contrast agent in order to provide several haemo- λ3
dynamic measures. For example, the signal-time curve generated
from the passage of the gadolinium bolus through tissue can be Fig. 12.3 Diffusion ellipsoids. In panel (a) isotropic diffusion is modelled as a
used to calculate measures of cerebral blood volume, mean transit sphere. In panel (b) anisotropic diffusion is modelled as an ellipsoid.
time, and cerebral blood flow. However, there are relative contrain-
dications for gadolinium injection, including severe renal failure
(stage 4 or 5; GFR < 30 ml/min per 1.73 m2), where there is a dose Measure Formula
ependent risk of nephrogenic systemic fibrosis. 1 [(λ1 λ 2 )2 (λ 2 λ 3 )2 (λ 3 − λ1 )2 ]
Fractional anisotropy
In contrast, ASL uses magnetically labelled water molecules in 2 (λ12 + λ 22 + λ 23 )
the blood as an endogenous contrast agent to provide a measure λ1 + λ 2 + λ 3
of cerebral blood flow. Briefly, ASL involves acquisition of a tag Mean diffusivity
3
image, in which inflowing arterial blood water is labelled by mag- Axial diffusivit
i y λ1
netic inversion, and a control image, in which inflowing blood is λ2λ3
not labelled. The difference between control and tag images is then Radial diffusivity
2
calculated, which is proportional to cerebral blood flow. Fig. 12.4 Formulae for DTI measures.
major direction of diffusion, can automatically reconstruct entire

white matter tracts, but the possibility of type II (false negative)
errors remain (Mori and van Zijl, 2002). NAA
VBA involves spatial registration of each scan to a standard space
and performing a voxel-by-voxel whole-brain statistical comparison of Creatine
white matter between groups. While VBA can provide a global assess- Choline
Amplitude
ment of white matter, it is limited by the quality of between-participant
image registration (Smith et al., 2006). TBSS projects all participants’
FA data onto an average FA tract skeleton before applying voxelwise
statistics. TBSS minimizes the effects of misalignment and is more
robust and sensitive than VBA (Smith et al., 2007). However, as the
TBSS skeleton consists of peak FA values, assumed to be at the centre
of the tract, TBSS may be insensitive to differences that are not uni-
form throughout the width of the tract. 4 3 2
Frequency (ppm)
Magnetic resonance spectroscopy Fig. 12.5 Proton MRS spectrum. Choline (cell membrane marker); Creatine
Magnetic resonance spectroscopy (MRS) is an MRI technique used (energy metabolism); NAA, n-acetyl aspartate (neuronal marker).
to quantify levels of metabolites in localized brain regions. It is (Images supplied courtesy of Dr. Charlotte Stagg, Junior Research Fellow St
based upon the idea that atomic nuclei in different molecular envi- Edmund Hall, and Oxford Centre for Functional MRI of the Brain.)
ronments are subject to slightly different magnetic fields (B0) and
therefore precess at different resonance frequencies. Therefore, the
amplitude (power) of the MRS signal at specific frequencies can be ‘direction of view’ of the camera is determined by collimators, which
used to measure the concentration of signal-generating molecules admit photons only from a specific direction (see Fig. 12.6a).
within a tissue. Data from MRS are not illustrated as an image but This implies that much of the radiation is absorbed within the
as a frequency spectrum. Typically, measurements are obtained in collimator’s walls, thereby reducing the sensitivity of SPECT (com-
single voxels or across a slice of the brain, but it is also possible to pare positron emission tomography (PET) below). Common nuclei
employ MRS across the whole brain. used for labelling are iodine-123 and technetium-99m. Both gen-
MRS can be used to measure precession frequencies of a variety erate γ-radiation, with a half-life of 13.2 h (123I) and 6 h (99mTc),
of atomic nuclei including lithium (7Li), carbon (13C), fluorine (19F), respectively. Particularly 123I, therefore, allows for synthesis of the
and sodium (23Na). However, the most frequently employed meth- tracer in a commercial laboratory and transport to the scanner.
ods within old age psychiatry research are based on proton MRS While 99mTc is widely generated and available locally, its use is only
(1H-MRS) and phosphorous MRS (31P-MRS). Metabolites studied practicable if the labelling can be done without special skills and
using 1H-MRS include n-acetyl aspartate (NAA), a marker of neu- radio-chemical equipment, quickly, and to a high standard, as, for
ronal integrity; choline (Cho) (which includes free choline, phos- example, with 99mTc-HMPAO, where both components (99mTc and
phocholine, and glycerophosphocholine), a marker of membrane HMPAO) are simply mixed in a vial.
integrity and myelination; myoinositol (MI), a marker of glial integ- A variety of biologically relevant molecules have been
rity; and the neurotransmitters glutamate and glutamine (Glx). employed in brain imaging with SPECT, from lipophilic com-
Metabolites studied using 31P-MRS include phosphomonoester pounds that are metabolized and trapped in brain cells on first
(PME) and phosphodiester (PDE), markers of the synthesis and pass after intravenous injection and mark cerebral perfusion
breakdown of membrane phospholipids, respectively; α-, β-, and (99mTc-exametazime = 99mTc-hexamethylpropyleneamine oxime =
γ-nucleoside triphosphate (NTP), markers of ATP; phosphocreat- 99mTc-HMPAO; 99mTc-ethylcysteinate dimer = 99mTc-ECD;
ine (PCr), which acts as a buffer of ATP; and inorganic phosphate n-isopropyl-123I-P- iodoamphetamine = 123I-IMP) to radiolabelled
(Pi), which can be used to estimate intracellular pH. receptor ligands, forexample dopamine-, serotonin-, muscarinic-,
As there are several experimental factors that can influence the nicotinic-, and transporter binding sites (Table 12.1). Of practical
signal amplitude, for example the MRI scanner system and the clinical importance today are some of the SPECT perfusion markers
brain location being evaluated, metabolite concentration is nor- (99mTc-HMPAO and 99mTc-ECD) mentioned in this section, and
mally measured with respect to a reference signal of known con- the dopamine transporter ligand 123I-labeled N-(3-fluoropropyl)-
centration, in most cases water or creatine (Cre), a marker of total 2β-carbomethoxy-3β-(4-iodophenyl) nortropane (FP-CIT; see
cellular creatine stores. Making a diagnosis for its clinical application).
SPECT perfusion ligands are injected and taken up into brain cells
Emission computed tomography
on first pass. This means that once the tracer is injected, the patient
Single photon emission computed tomography can be imaged at any time afterwards, as long as the isotope is still
SPECT exploits the fact that high energy radiation (γ-radiation), active. An additional advantage is that the tracer remains in a pattern
like X-rays, penetrates human tissue, so that the three-dimensional representing a snapshot of brain perfusion within 2 min of injection,
distribution of a molecule labelled with a γ-emitter can be recon- so that any later interventions, sedation, or even anaesthetic do not
structed from the activity maps recorded from outside the head. change the image and can be used to optimize the quality of the scan.
Γ-radiation is measured with γ-cameras that translate γ-photons, The size of the γ-emitter atoms used (isotopic mass of 123I = 122.9;
via scintillation counters and crystals, into electrical signals, which 99mTc = 98.9) means that any radiolabelled molecules are likely to
can then be amplified and analysed in a quantitative fashion. The have a pharmacology different from that of the mother compound,
(a) Gamma Camera (b) PET Camera

Gamma
emitter
Gamma
photons Positron
emitter Coincidence
180°
Collimator
Crystal 2 Annihilation
photons Detector
Fig. 12.6 Methods of determining site of

radiation in SPECT (a) and PET (b).
Table 12.1 Biologically relevant molecules used in SPECT and PET substituted for the equivalent nonradioactive isotope (or halogen).
studies The short half-life implies that any synthesis of labelled tracer has
to be done in-house, which increases the price of such ligands to
Tracer Isotope-imaging Clinical use Measure a level that can only be afforded in specialist centres. In addition
mode
to this, the short half-life of positron emitters such as 11C and 15O
EDT 99mTc-SPECT Yes Perfusion requires the tracer injection to be done in the scanner, which makes
HMPAO 99mTc- SPECT Yes Perfusion the procedure potentially difficult for certain patient groups.
123I-SPECT
An exception to this are 18F-fluoro-deoxyglucose and 18F-labelled
5-I-A-85380 No Nicotinic receptor amyloid ligands (Herholz and Ebmeier, 2011), which can be
FP-CIT 123I-SPECT Yes Dopamine transporter imported from within a few hundred kilometres and are injected
Iodobenzamide 123I-SPECT No Dopamine D2 receptor well before the imaging session. For this reason, we will limit our
123I-SPECT
discussion to such potentially clinically useful substances.
Iodo-QNB No Muscarinic receptor
The central concept to understand with PET is that positrons
123I-SPECT
Iomazenil No Benzodiazepine emitted from the radioisotopes are annihilated within seconds
receptor and millimetres of their origin, and generate two photons of
FDG 18F-PET Yes Glucose uptake defined energy that travel in opposite directions. An instrument
(aerobic + anaerobic) that can detect such coincident photons is able to locate their
Glucose 11C-PET No Glucose uptake
source on the line between the activated detectors (Fig. 12.6b).
(aerobic) Any photon that is not matched by a coincident partner or is
15O-PET
of a lower energy than the predicted one is likely to be due to
Oxygen No Oxygen uptake scatter and can be ignored. Thus PET is generally more sensitive
Water 15O-PET Yes Blood flow than SPECT, where much of the potentially available informa-
MP4A 11C-PET No AChE activity tion is absorbed by the collimators (see Single Photon Emission
11C-PET
Tomography). In common with SPECT is the exposure to radio-
Nicotine No Nicotinic receptor
active isotopes, which may be trivial in the single case, but adds
11C-PET
Pi(ttsburgh No Insoluble fibrillary to the overall population exposure to ionizing radiation. For this
compound) B amyloid β (plaques) reason, BOLD fMRI and arterial spin labelling MRI have taken
Florbetaben 18F-PET Not yet – over virtually all experimental research applications of 15O-water
18F-PET PET for blood flow, and to an extent 18F-FDG PET, for the
Florbetapir Yes –
measurement of regional cerebral metabolism. An exception is
Flutemetamol 18F-PET Not yet – 18F-FDG PET in dementia research, where the logistics of scan-
FDDNP 18F-PET No NFTs and Aβ plaques ning (injection at rest 30–45 min before the scan), its availability
18F-PET in oncology centres, its quantitative nature, and the age group of
Altanserin No 5-HT2A receptor
the patients involved (where radiation risk is minimized) make it
WAY-100635 18F-PET No 5-HT1A receptor the preferred method of imaging current brain function (Herholz
et al., 2002).
Amyloid plaques are one of the hallmarks of Alzheimer pathol-
ogy. It is, therefore, not surprising that efforts have been made
so any pharmacological and toxicological investigations have to be
to label amyloid in vivo (Herholz and Ebmeier, 2011). As amy-
repeated independently, which slows down ligand development.
loid deposition is found in healthy older people’s brains, amy-
Positron emission tomography loid scanning cannot be diagnostic for Alzheimer’s disease (AD).
In contrast to this, convenient positron emitters are available Current efforts focus on the predictive validity of increased
in the form of 11C (radioactive half-life approximately 20 min), amyloid binding in mild cognitive impairment (MCI), as some
15O (approximately 2 min), and 18F (approximately 110 min), all people in this diagnostic category have normal and others have
abnormal amyloid scans. A potential shortcoming of the ligands

in development is nonspecific white matter binding, but this may
100
be overcome in practice (Herholz and Ebmeier, 2011). If future
Spatial resolution [mm]

therapeutic agents were to remove amyloid deposits, an in vivo
amyloid ligand would provide a way to monitor treatment suc- EEG/ fMRI PET
10 MEG
cess directly.
Electromagnetic fields
1
Electroencephalography
Electroencephalography (EEG) is a noninvasive functional tech-
nique that records electrical fields along the scalp using electrodes
0.001 0.01 0.1 1 10 100
placed in a cap over the head. The recorded electric potentials
Temporal resolution [s]
are generated by populations of thousands of neurons that have
a similar orientation and fire synchronously. The pyramidal Fig. 12.7 Spatial and temporal resolution of different neuroimaging techniques.
neurons of the cortex have these properties and therefore their
activity is thought to underlie EEG signals. While the temporal Several techniques can be used to reconstruct the source of neu-
resolution is much higher compared with fMRI—signals can be ronal activity, e.g. equivalent current dipole modelling and beam-
recorded only milliseconds after the presentation of a stimulus— former approaches. All of these techniques use specific constraints
the spatial resolution is greatly reduced (Fig. 12.7). One reason is to locate neuronal activity because theoretically there are an infinite
that the strength of the electric potential that is measured by EEG number of solutions to explain the MEG/EEG signal. This is known
falls off with the square of the distance, which impairs the detec- as the inverse problem.
tion of subcortical neuronal activity. Additionally, the skull and
scalp cause distortions of the signal which makes exact localiza-
tion difficult. Clinical Applications
Magnetoencephalography What is involved for the patient?
In contrast to EEG, magnetoencephalography (MEG) is used to A structural MRI scan with sequences sensitive to pathological
measure magnetic fields that are induced by the electric activity changes typically lasts less than 30 min, but in research settings,
of neurons. These magnetic fields are measured using supercon- multimodal MRI acquisition can take up to 1 h. A major limita-
ducting quantum interference devices (SQUIDs) as sensors that are tion to scan length is participant comfort and compliance, which
arranged in a helmet placed over the head. also has an effect on image quality (e.g. movement artefacts). There
One advantage of this method is that the magnetic field is less are a number of contraindications to MRI, including pacemakers,
influenced by changes of conductivity between brain, skull, and certain metallic implants, and severe claustrophobia. CT has sev-
scalp, which improves localization of neuronal activity. A disad- eral advantages over MRI as it is cheap, widely available, and often
vantage of MEG is that it can only record signals from neuronal suitable for patients with contraindications to MRI. However, most
pathways that run in parallel to the skull. Therefore, the main con- research is now conducted using MRI techniques as it provides
tribution to MEG signals comes from neurons located within corti- higher resolution images, greater soft tissue contrast, and does not
cal sulci. In contrast, EEG measures activity particularly in the gyri involve ionizing radiation. While MRI typically requires insertion
and is, therefore, more sensitive to sources located deeper within into a narrow scanner gantry, the more open design of CT, but also
the brain. of emission tomography scanners, makes them more acceptable for
There are several ways of analysing EEG and MEG data. One might some patients.
be interested in the electrophysiological response with respect to the PET and SPECT scans are used, according to their availability, in
onset of a specific stimulus. This response is called event-related various centres, with a trend for PET to become more widely availa-
potential (ERP) in EEG and event-related field (ERF) in MEG and ble, as PET-CT scanners are acquired by many services for oncologi-
occurs phase-locked to stimulus onset. The average of many syn- cal examinations, although their availability for psychiatric patients
chronized EEG traces removes noise and emphasizes any regular is likely to remain limited. Both are likely to fulfil the requirements
responses to the stimulus. Similarly, EEG/MEG can be phase-locked of diagnosis and differential diagnosis and can be used interchange-
to the subject’s response, if the main interest is in processes leading ably (Ebmeier, 2010), except in cases where appropriate ligands are
up to a motor response. It is alternatively possible to study changes not available in a modality (e.g. dopamine transporter ligands are
in the power of oscillations in electric activity in specific frequency not commercially available for PET, and amyloid ligands for SPECT
bands. More synchronized neuronal firing is thought to underlie are not on the horizon). Both scan modalities are relatively user
increased power of oscillations. Typical frequency bands of interest friendly, with an open gantry and little noise. Their main limitation
are delta (less than 4 Hz), theta (4–7 Hz), alpha (8–13 Hz), beta (14– is the radiation dose administered parenterally, both in terms of
30 Hz), and gamma (30–100 Hz). The power of oscillations can be the necessity of an injection and the (usually very small) risk to the
measured with respect to a stimulus onset (induced activity), which patient from ionizing radiation.
is called event-related synchronization (ERS) or desynchronization EEG and ERPs are no longer commonly used in clinical psychi-
(ERD). It can also be measured independently of a stimulus (spon- atric practice, except in the differential diagnosis of epilepsy and
taneous activity) when subjects are resting during the measurement multiple sclerosis, although they have several advantages compared
and not engaged in a specific task. with MRI. EEG equipment is user friendly, cheaper, and more
mobile, and recordings do not require a strong magnetic field. Potential diagnostic use
Patients merely have to sit still, or engage in simple tasks, usually The MRI technique that is arguably closest to translation to the
attentional auditory or visual tasks. Nevertheless, analysis of EEG clinic is T1-weighted MRI in the diagnosis of AD (Dubois et al.,
data is not straightforward, requires special neurophysiological 2010; McKhann et al., 2011). A key feature of AD is early, localized
skills, and is strongly affected by current mental state and by medi- atrophy of the medial temporal lobe, including the hippocampus
cation, which may be one explanation why it is not generally used and entorhinal cortex, which is followed by generalized neocor-
in the clinic. tical volume loss. T1-weighted MRI is an ideal tool to examine
MEG is not generally available and is predominantly used for such patterns of atrophy in vivo, with volumetric measures found
research, where it is well tolerated even by very young and old and to correlate well with both neuropathological disease progression
cognitively impaired patients. (Gosche et al., 2002; Jack et al., 2002; Vemuri et al., 2008) and the
degree of cognitive impairment (Hua et al., 2008; Vemuri et al.,
Making a diagnosis 2009). Moreover, T1-weighted MRI studies of AD have recently
Current clinical practice progressed from reporting case-control group differences to auto-
MRI is used to aid in the diagnosis of some dementias (see matic classification of individual scans. A multisite study involving
Table 12.2). As indicated in Table 12.2, CT and MRI are commonly 98 AD and 109 control participants obtained a specificity rate of
used to exclude ‘organic’ causes of dementia and other psychiatric 95% (controls correctly identified as controls, i.e. correct negative
syndromes. CT is used in the diagnosis of stroke (haemorrhagic rate) and a sensitivity rate of 71% (patients correctly identified as
or old ischaemic), large tumours, intra- and extracranial bleed- patients, i.e. correct positive rate) (Vemuri et al., 2009). Smaller
ing, atrophy, (communicating) hydrocephalus, and trauma. MRI is studies have also reported success in discriminating between AD
preferred for the investigation of conditions only detectable with and other dementias (Kloppel et al., 2008).
higher resolution and greater soft tissue contrast, such as temporal There is an extensive literature on regional cerebral blood flow
lobe epilepsy (medial temporal gliosis), multiple sclerosis, certain and regional glucose metabolism studies in dementia (Ebmeier,
tumours (e.g. pituitary adenomas), and certain vascular changes. 2010; Herholz, 2011), so that recent proposals for diagnostic cri-
While perfusion SPECT remains a tool for the occasional patient teria for Alzheimer’s dementia suggest 18F-FDG-PET as a topo-
with an unclear differential diagnosis (National Institute for Health graphical diagnostic marker (Dubois et al., 2010; McKhann et al.,
and Clinical Excellence, 2007), SPECT of the dopamine transporter 2011). In addition, the use of amyloid tracer uptake has been sug-
(FP-CIT; DatScan) has been quickly established as a reliable marker gested as a pathophysiological marker of AD, although there are
of Lewy body disease, be it Parkinson’s disease or Lewy body a number of caveats extending from availability of 11C-labelled
dementia (McKeith et al., 2007). compounds (Pittsburgh compound B, PiB) to the relative lack of
Dopamine transporter (DAT) density is highest in striatum, i.e. specificity of the available ligands, that show abnormalities already
caudate and putamen. With age-related nigrostriatal degenera- in MCI and healthy volunteers (Herholz and Ebmeier, 2011). There
tion, DAT binding also gradually decreases with age. In Parkinson’s are, however a number of 18F-labelled tracers under development
disease and Lewy body dementia, however, there is usually a pro- (flutemetamol, florbetapir, and florbetaben) that will at least make
nounced reduction in FP-CIT signal, often with greater reduction amyloid imaging available away from radiochemistry laboratories
in activity in the putamen. Because antipsychotic drugs bind to (Herholz and Ebmeier, 2011).
the postsynaptic D2 receptors and not DAT, drug-induced parkin- Other neuroimaging markers of AD have not yet been tested in
sonism shows a normal uptake pattern for FP-CIT. Vascular par- large-scale classification studies, but may have potential to reach
kinsonism may show a pronounced lateral asymmetry of striatal that stage. For example, DTI studies of AD have typically detected a
binding, but idiopathic Parkinson’s disease also often presents with pattern of decreased FA and increased MD values in AD compared
asymmetrical FP-CIT uptake. with controls. A meta-analysis of 33 case-control studies of AD
Table 12.2 MRI used in the diagnostic criteria of some dementias

Disease Diagnostic criteria MRI marker Implementation
Alzheimer’s dementia NINDS-ADRDA (McKhann et al., 1984) ‘Normal CT’ Exclusion
NIA-AA (McKhann et al., 2011) Medial temporal atrophy Most validated structural
topographical marker
Creutzfeldt-Jacob disease Collie et al., 2001; Tschampa et al., 2005 Cortical diffusion changes; pulvinar sign Diagnostic
Dementia with Lewy bodies McKeith 2006 Preserved medial temporal lobes, relative to AD Supportive
Frontotemporal degeneration Neary et al., 1998 Focal frontal or temporal atrophy Supportive
Multiple system atrophy Gilman et al., 2008 Atrophy of putamen, middle cerebellar Additional feature
peduncle, pons, and/or cerebellum
Vascular dementia NINDS-AIREN (Roman et al., 1993) Strategic infarction—extensive white matter Mandatory
changes
(a) (b)
Fig. 12.8 DatScan SPECTs (dopamine transporter) of (a) Lewy body dementia (note apparent greater activity across the brain, due to normalization procedure) and
(b) Alzheimer’s dementia. (Images supplied courtesy of Prof John T. O’Brien and Dr Sean Colloby, Institute for Ageing and Health, Newcastle University.) See also Plate 12.
Fig. 12.9 T1-weighted MRI scans to illustrate Scheltens scale medial temporal atrophy (MTA) scores 0–4. (Images supplied courtesy of Prof Philip Scheltens,
Neuroscience Campus, VrijeUniversiteit Amsterdam.)
found that AD was associated with reduced FA and elevated MD spectra (Jeong, 2004). This could be used for diagnostic pur-
values in widespread regions, including within frontal and tempo- poses. For example, a specificity rate of 83% and sensitivity of
ral white matter, the posterior cingulum, corpus callosum, superior 87% was achieved comparing global amplitudes of power in the
longitudinal fasciculus, and uncinate fasciculus (Sexton et al., 2011). alpha and theta frequency range between 38 AD patients and 24
Furthermore, several studies have reported a significant association healthy controls (Huang et al., 2000). Differences in amplitude
between decreased MMSE in AD and reduced FA (Bozzali et al., and latency of several ERP components have also been reported
2002; Duan et al., 2006; Ukmar et al., 2008; Mielke et al., 2009) or when comparing AD patients with healthy controls (Jackson and
increased MD (Bozzali et al., 2002; Duan et al., 2006), indicating Snyder, 2008).
that DTI metrics may be sensitive markers of disease progression. To date there are only a few MEG studies in AD. Two studies have
Although EEG has been shown to be useful in AD research for reported differences between AD patients and healthy controls
the last 30 years, there has not been a proper translation into the that might be useful for clinical practice: in a study of resting-state
clinic. One reason might be that there are no large-scale stud- MEG, 22 AD patients showed reduced mean frequency of power
ies investigating the clinical use of EEG to discriminate between spectra compared with 21 healthy controls (Fernandez et al.,
AD patients and healthy controls. This is surprising given the 2006a). Further analysis of the same subject group showed that
good results of resting-state EEG studies. AD is characterized relative power in the 16- to 28-Hz frequency band can differentiate
by slowing of cortical rhythms. Early stages of AD are associ- between AD patients and healthy controls with 80% specificity and
ated with increased theta band activity and decreased beta band 81% sensitivity (Fernandez et al., 2006b).
activity followed by a decrease in alpha activity. Additionally, In addition to the growing number of case-control studies of AD,
dementia severity correlates with the mean frequency of power neuroimaging research is used to investigate changes brought about
(a) (b)
Abnormal areas FDG
NC AD MCI
AD
Fig. 12.10 18F-FDG and 11C-PIB scans in healthy controls and patients with MCI and Alzheimer’s dementia. (a) Amyloid scan with florbetapir in a normal control (NC)
and an AD patient. Note tracer uptake in white matter of the healthy volunteer brain. (b) FDG scan in a patient with MCI and 2 years later, when he had progressed
to AD. The left column highlights the abnormal areas in black (association cortex). (Images supplied courtesy of Prof Karl Herholz, Wolfson Molecular Imaging Centre,
University of Manchester). See also Plate 13.
by other types of dementia, for example dementia with Lewy bod- whilst prediction of first episodes or relapses would provide an
ies (DLB), vascular dementia (VaD), and frontotemporal demen- opportunity for preventative measures.
tia (FTD). Several MRI, DTI, and MRS studies report differences
between dementia types (Ebmeier et al., 2011). Although there is an Predicting outcome
abundance of EEG studies in AD patients, only a few studies inves- Neuroimaging techniques are not currently used in the clinic for
tigate different dementia types using EEG or MEG. Some of them predicting outcome, but significant research has been performed
report differences in resting-state EEG (Jeong, 2004). Nevertheless, in this field. In particular, many studies have been performed that
larger studies are required to show if neuroimaging markers can have the long-term aim of predicting conversion to AD from at-risk
help facilitate diagnosis when the clinical presentation of a patient groups, including MCI and apolipoprotein E (ApoE) ε4 carriers.
is ambiguous. MCI is a descriptive diagnosis that identifies patients referred
Neuroimaging is not yet used routinely for the assessment of with memory impairment beyond that expected for age and educa-
late life depression (LLD), nor have large-scale classification stud- tion, who do not display functional deficits, and so do not qualify
ies been performed. Many research studies have used neuroim- for a diagnosis of dementia. A recent meta-analysis, however, calcu-
aging to identify differences between patient and control groups. lated the annual conversion rate from MCI to AD to be only 5–10%
There are several reasons that have contributed to difficulties in (Mitchell and Shiri-Feshki, 2009). ApoE is a lipid-transporting
translating their results into clinically useful diagnostic markers protein that exists in three different isoforms: ApoE2, 3, and 4; the
of LLD. First, there is frequently great heterogeneity in the anat- human APOE gene has three allelic variants (ε2, ε3, and ε4). The
omy of results reported. For example, while grey matter volume ε4 allele is associated with higher risk of developing early-onset
reductions in LLD have been most frequently reported within (Okuizumi et al., 1994) and late-onset (Corder et al., 1993) AD.
frontal lobe and hippocampus, changes have also been observed Nevertheless, it is neither necessary nor sufficient for the develop-
in the parietal lobe, amygdala, putamen, and thalamus, particu- ment of AD. That is why investigating the effect of the APOEε4-allele
larly in late-onset illness (Sexton et al., 2012b). Second, there is on brain structure and function might help to understand more
often a lack of specificity of results reported for LLD. For exam- about AD development and could help identify people at specific
ple, reduced hippocampal volume has also been reported in many risk before onset of the disease.
other patient groups, including MCI, AD, FTD, and VaD (van de Several T1-weighted MRI studies have examined conversion
Pol et al., 2006; Scher et al., 2011). Third, there is the question of from MCI to AD. Ferreira et al. (2011) performed a meta-analysis
what value a neuroimaging marker would have above and beyond of six longitudinal VBM studies of MCI, in which 142 of 429 partic-
the clinical assessment already used. An ideal marker would not ipants converted to AD. Conversion was associated with decreased
only reliably discriminate between patient groups, but also predict grey matter in the left hippocampus and parahippocampal gyrus.
outcome. For example, prediction of treatment response would Specificity and sensitivity rates in predicting conversion for indi-
allow the most likely effective treatment to be administered first, viduals with MCI have also been calculated by some studies. For
example, Devanand et al. (2008) used a model combing hippocam- For example, several neuroimaging studies have shown effects of
pal and entorhinal volumes with functional and cognitive impair- homocysteine levels on the brain, and among others plasma homo-
ment measures, to predict conversion from MCI to AD with 85% cysteine levels depend on nutrition and vitamin supply. High plasma
sensitivity and 90% sensitivity. homocysteine levels increase the risk for AD (Seshadri et al., 2002),
Reduced white matter integrity has also been detected in peo- which might be due to increased cortical and hippocampal atrophy
ple at higher risk of developing AD, including those with MCI and (den Heijer et al., 2003). Additionally, a recent study has reported
ApoE ε4 carriers. A meta-analysis of 21 case-control studies of that administration of B vitamin supplements can slow the rate of
MCI found that differences between MCI and controls paralleled brain atrophy in MCI subjects (Smith et al., 2010). Therefore, neu-
the differences between AD and controls, but fewer regions reached roimaging markers can be used to identify lifestyle factors affect-
statistical significance (Sexton et al., 2011). As with AD, several ing AD, which may be important to delay disease onset and study
studies have reported a significant association between decreased preventative measures.
MMSE and reduced FA in MCI (Ukmar et al., 2008; Bai et al., 2009; Illuminating the question as to how APOE genotype may affect
Mielke et al., 2009).With regard to APOE ε4, a TBSS study identi- susceptibility to AD, many studies have shown functional and
fied a widespread pattern of elevated MD in APOEε4 carriers aged structural differences between APOEε4 carriers and noncarriers. A
50–78 years compared with age-matched controls, and reduced FA network that is active when the brain is ‘at rest’, i.e. awake but not
in carriers aged 20–35, again compared with age-matched controls doing a specific task, has attracted special attention. This so-called
(Heise et al., 2011). default-mode network (DMN) is of interest because amyloid plaque
Many studies have shown the predictive value of EEG measures deposition seems to occur in regions that show high DMN activity
for progression of healthy subjects to MCI and AD (Jeong, 2004). (Buckner et al., 2005) and DMN activity differs between APOEε4
A 7-year follow-up study of 44 healthy participants with subjec- carriers and noncarriers (Filippini et al., 2009). An association
tive memory complaints showed that EEG at baseline can predict between neuronal activity and the processing of amyloid precursor
conversion to MCI with 95% sensitivity and 94.1% specificity and protein has also been shown in vitro (Kamenetz et al., 2003), which
conversion to dementia with 96.3% sensitivity and 94.1% specifi- suggests the hypothesis that genetically based differences in neuro-
city (Prichep et al., 2006). These high values were achieved with nal activity may play a role in AD development.
‘eyes-closed’ resting data. Subjects also had to discontinue any cen- In LLD, neuroimaging studies have played a key role in the devel-
trally acting medication at least 2 weeks before the baseline scan, opment and assessment of the vascular depression hypothesis. The
which probably limits the potential clinical use. The authors further vascular depression hypothesis proposes that cardiovascular factors
did a post-hoc logistic regression analysis based on nine quantita- may ‘predispose, precipitate, or perpetuate’ late-onset depression
tive (Q)EEG baseline variables, and ‘variables with the highest sig- (LOD) (typically age of onset is more than 60 years) (Alexopoulos
nificance in the prediction equation’ were mean frequency of the et al., 1997). Specifically, vascular factors are hypothesized to
total spectrum, mean frequency in the delta band, absolute power cause white matter damage within frontal-subcortical circuits
across all frequency bands on the right hemisphere, and absolute (Alexopoulos et al., 1997; Alexopoulos, 2002). In support of this
power in the theta frequency band. hypothesis, LOD has often been found to be associated with more
Several studies have used EEG also to investigate the effect of frequent and severe WMH compared with early-onset depression
APOE genotype on brain function. In AD patients and in healthy (EOD) (Herrmann et al., 2008). More frequent and severe WMH
subjects, differences between APOEε4 carriers and noncarriers have also been linked to poor antidepressant treatment response
were reported. This affects task-related ERP amplitudes and laten- in LLD (Hickie et al., 1995, 1997; Simpson et al., 1998). Expanding
cies (Espeseth et al., 2009) as well as resting-state EEG measure- on this research, Sheline et al. (2010) have reported that WMH
ments (Babiloni et al., 2006). severity predicted Montgomery-Åsberg Depression Rating Scores
(MADRS) over a 12-week course of treatment. Although some
Understanding illness mechanisms studies of WMH have employed automated methods, the majority
Neuroimaging can contribute in different ways to the question how of studies have used rating scales that can suffer from intra- and
and why diseases develop. Functional and structural measures can inter-rater reliability and a lack of anatomical specificity, and may
be used to evaluate the effects of drug treatment, investigate the also be insensitive to subtle changes in WM integrity. DTI studies
effects of lifestyle factors on the brain and disease risk, and pro- of LLD overcome many of these limitations and have also provided
vide insights into mechanisms that underlie genetic susceptibility support for the vascular hypothesis of depression. For example,
to diseases. Sexton et al. (2012a) found that later age at onset was significantly
With respect to AD, examples of these approaches are studies associated with reduced FA of widespread tracts, in particular the
looking at the effect of cholinesterase inhibitor (ChEI) treatment anterior thalamic radiation and superior longitudinal fasciculus.
on the brain (Venneri, 2007). In healthy subjects and AD patients, Similarly, Framingham Stroke scores predicted greater disruption
ChEI administration modulates brain function, and in AD patients of white matter connectivity (Allan et al., 2011).
it might slow down hippocampal atrophy. However, longitudinal
studies of large patient populations are required to understand the Understanding resilience: neuronal scaffolding
long-term effects of ChEI administration. Given that, overall, only Ageing is characterized by decreasing grey matter density, reducing
around 50% of patients show a positive response to medication, it white matter integrity, and (somewhat paradoxically) increasing
is also important to investigate if structural or functional imaging neuronal activity. fMRI studies report increased bilateral activa-
measures can be used to predict the response to drug treatment. tion of frontal and other association cortices in older as opposed to
Although there are genetic risk factors for AD development, younger subjects, which is associated with better cognitive perform-
there are important effects of lifestyle, such as diet, on AD risk. ance (Cabeza et al., 1997, 2002; Grady et al., 1999; Reuter-Lorenz
et al., 1999; Rypma et al., 2001; Reuter-Lorenz, 2002; Fera et al., will be of benefit to our patients. We may, nevertheless, be the last
2005). Hippocampal underactivity in well-functioning older sub- generation of clinicians who can claim with quiet satisfaction that
jects is combined with increased bilateral frontal activation com- our attentive interview and expert examination are as good as any
pared with younger volunteers (Morcom et al., 2003; Cabeza et al., brain imaging currently available to make a diagnosis and decide
2004; Gutchess et al., 2005; Persson et al., 2006). This increase in on the best management for our patients.
activity has been interpreted as a compensatory process, i.e. the
brain has to work harder to achieve the same level of performance. References
A number of authors have converged on the view that it may delay Alexopoulos, G.S. (2002). Frontostriatal and limbic dysfunction in late-life
performance failure and clinical decompensation even in the face of depression. American Journal of Geriatric Psychiatry, 10(6), 687–95.
progressive atrophy of critical brain structures (e.g. hippocampus); Alexopoulos, G.S., et al. (1997). ‘Vascular depression’ hypothesis. Archives of
General Psychiatry, 54(10), 915–22.
and it is accompanied by subjective awareness of cognitive prob-
Allan, C.L., et al. (2011). Does the Framingham Stroke Risk Profile
lems or increased effort during task performance (Buckner, 2004; predict white-matter changes in late-life depression? International
van Norden et al., 2008). The concept of neuronal scaffolding is a Psychogeriatrics, 17, 1–8.
formulation of this general idea that successful ageing is associated Amaro, E., Jr. and Barker, G.J. (2006). Study design in fMRI: basic principles.
with increased recruitment of additional neuronal resources in the Brain and Cognition, 60(3), 220–32.
face of deteriorating brain structure (Reuter-Lorenz and Cappell, Ashburner, J. and Friston, K.J. (2000). Voxel-based morphometry—the
2008; Han et al., 2009; Park and Reuter-Lorenz, 2009). Scaffolding methods. Neuroimage, 11(6 Pt 1), 805–21.
is a process that results in changes in brain function through Babalola, K.O., et al. (2009). An evaluation of four automatic methods of
strengthening of existing connections, formation of new connec- segmenting the subcortical structures in the brain. Neuroimage, 47(4),
tions, and disuse of connections that have become weak or faulty 1435–47.
Babiloni, C., et al. (2006). Apolipoprotein E and alpha brain rhythms in
(Park and Reuter-Lorenz, 2009). Consistent with this, a posterior–
mild cognitive impairment: a multicentric electroencephalogram study.
anterior shift of brain activity can be observed with increasing age, Annals of Neurology, 59(2), 323–34.
suggesting that frontal lobe networks are preferentially activated in Bai, F., et al. (2009). Abnormal integrity of association fiber tracts in amnestic mild
a compensatory manner and so may provide the necessary second- cognitive impairment. Journal of Neurological Science, 278(1–2), 102–6.
ary scaffold (Davis et al., 2008). Therefore, damage to the primary Beaulieu, C. (2002). The basis of anisotropic water diffusion in the nervous
structure involved in a neurocognitive task (for example, the hip- system—a technical review. NMR Biomedicine, 15(7–8), 435–55.
pocampus in memory) must exceed a certain threshold to reveal Beckmann, C.F. and Smith, S.M. (2004). Probabilistic independent component
the presence of objective memory impairment. However, a parallel analysis for functional magnetic resonance imaging. IEEE Transactions
process affecting the compensating networks (typically generalized in Medical Imaging, 23(2), 137–52.
vascular pathological change) may worsen the impact of the pri- Beckmann, C.F., et al. (2005). Investigations into resting-state connectivity
using independent component analysis. Philosophical Transactions of the
mary pathology or even be necessary to reveal a classical clinical
Royal Society of London Series B Biological Sciences, 360(1457), 1001–13.
condition (Buckner, 2004). This dual mechanism may explain the Bookstein, F.L. (2001). ‘Voxel-based morphometry’ should not be used with
biphasic clinical course of initial compensated memory impair- imperfectly registered images. Neuroimage, 14(6), 1454–62.
ment with insight, followed by dementia and loss of function in AD Bozzali, M., et al. (2002). White matter damage in Alzheimer’s disease
(Buckner, 2004). While hippocampal lesions occur early and are assessed in vivo using diffusion tensor magnetic resonance imaging.
the locus of maximum plaque and tangle formation, older people Journal of Neurology, Neurosurgery and Psychiatry, 72(6), 742–6.
in general and AD patients in particular show a deterioration in the Buckner, R.L. (2004). Memory and executive function in aging and AD:
frontal executive network that is characterized by reduction in fluid multiple factors that cause decline and reserve factors that compensate.
intelligence, i.e. abstract thinking and reasoning. Structurally, there Neuron, 44(1), 195–208.
Buckner, R.L., et al . (2005). Molecular, structural, and functional
may be white matter hyperintensities, reductions in white matter
characterization of Alzheimer’s disease: evidence for a relationship
integrity on DTI, and frontal atrophy. This putative mechanism between default activity, amyloid, and memory. Journal of Neuroscience,
may be generally applicable because both whole-brain and hippoc- 25(34), 7709–17.
ampal atrophy trajectories have been related to the onset of cogni- Cabeza, R., et al. (1997). Age-related differences in neural activity during
tive impairment and dementia: in one study hippocampal atrophy memory encoding and retrieval: a positron emission tomography study.
preceded dementia by 3 years and whole-brain atrophy by 1 ear Journal of Neuroscience, 17(1), 391–400.
(Ridha et al., 2006). Consistent with this, hippocampal atrophy Cabeza, R., et al. (2002). Aging gracefully: compensatory brain activity in
rates have been shown to predict MCI, while global atrophy rates high-performing older adults. Neuroimage, 17(3), 1394–402.
predicted dementia onset (Henneman et al., 2009). Cabeza, R., et al. (2004). Task-independent and task-specific age effects on
brain activity during working memory, visual attention and episodic
retrieval. Cerebal Cortex, 14(4), 364–75.
Conclusion Collie, D.A., et al. (2001). MRI of Creutzfeldt-Jakob disease: imaging features
It has taken half a century to develop methods that allow us to and recommended MRI protocol. Clinical Radiology, 56(9), 726–39.
examine brain structure and function in vivo. These secrets have Corder, E.H., et al. (1993). Gene dose of apolipoprotein E type 4 allele and the
risk of Alzheimer’s disease in late onset families. Science, 261(5123), 921–3.
been well guarded by the compact packaging and the complex
Damoiseaux, J.S., et al. (2006). Consistent resting-state networks across
structure of our brains. While giant strides are being made in
healthy subjects. Proceedings of the National Academy of Science USA,
unravelling aspects of normal and abnormal brain function, analys- 103(37), 13848–53.
ing this most complex of all systems will take some time to achieve. Davis, S.W., et al. (2008). Que pasa? The posterior-anterior shift in aging.
Obvious limits are given by ethics and by the absence of credible Cerebral Cortex, 18(5), 1201–9.
(animal or computing) models for most human mental disorders. Den Heijer, T., et al. (2003). Homocysteine and brain atrophy on MRI of
It will take all our ingenuity to develop insights and methods that non-demented elderly. Brain, 126(Pt 1), 170–5.
Devanand, D.P., et al. (2008). Combining early markers strongly predicts Hua, X., et al. (2008). Tensor-based morphometry as a neuroimaging
conversion from mild cognitive impairment to Alzheimer’s disease. biomarker for Alzheimer’s disease: an MRI study of 676 AD, MCI, and
Biology and Psychiatry, 64(10), 871–9. normal subjects. Neuroimage, 43(3), 458–69.
Duan, J.H., et al. (2006). White matter damage of patients with Alzheimer’s Huang, C., et al. (2000). Discrimination of Alzheimer’s disease and mild
disease correlated with the decreased cognitive function. Surgical and cognitive impairment by equivalent EEG sources: a cross-sectional and
Radiologic Anatomy, 28(2), 150–6. longitudinal study. Clinical Neurophysiology, 111(11), 1961–7.
Dubois, B., et al. (2010). Revising the definition of Alzheimer’s disease: a new Jack, C.R., Jr., et al. (2002). Antemortem MRI findings correlate with
lexicon. Lancet Neurology, 9(11), 1118–27. hippocampal neuropathology in typical aging and dementia. Neurology,
Ebmeier, K.P. (2010). Is there still a place for perfusion SPECT in the diagnosis 58(5), 750–7.
of dementia? Open Nuclear Medicine Journal, 2, 40–5. Jackson, C.E. and Snyder, P.J. (2008). Electroencephalography and
Ebmeier, K.P., et al. (2011). Other magnetic resonance imaging techniques. event-related potentials as biomarkers of mild cognitive impairment and
International Psychogeriatrics, 23 Suppl 2, S50–7. mild Alzheimer’s disease. Alzheimer’s Dementia, 4 1 Suppl 1, S137–43.
Espeseth, T., Rootwelt, H., and Reinvang, I. (2009). Apolipoprotein E Jeong, J. (2004). EEG dynamics in patients with Alzheimer’s disease. Clinical
modulates auditory event-related potentials in healthy aging. Neuroscience Neurophysiology, 115(7), 1490–505.
Letters, 459(2), 91–5. Kamenetz, F., et al. (2003). APP processing and synaptic function. Neuron,
Fera, F., et al. (2005). Neural mechanisms underlying probabilistic category 37(6), 925–37.
learning in normal aging. Journal of Neuroscience, 25(49), 11340–8. Kloppel, S., et al. (2008). Automatic classification of MR scans in Alzheimer’s
Fernandez, A., et al. (2006a). MEG spectral profile in Alzheimer’s disease and disease. Brain, 131(Pt 3), 681–9.
mild cognitive impairment. Clinical Neurophysiology, 117(2), 306–14. Matthews, P.M. and Jezzard, P. (2004). Functional magnetic resonance
Fernandez, A., et al. (2006b). Quantitative magnetoencephalography of imaging. Journal of Neurology, Neurosurgery and Psychiatry, 75(1), 6–12.
spontaneous brain activity in Alzheimer disease: an exhaustive frequency McKeith, I.G. (2006). Consensus guidelines for the clinical and pathologic
analysis. Alzheimer’s Disease and Associated Disorders, 20(3), 153–9. diagnosis of dementia with Lewy bodies (DLB): report of the consortium
Ferreira, L.K., et al. (2011). Neurostructural predictors of Alzheimer’s on DLB international workshop. Journal of Alzheimer’s Disease, 9 3 Suppl,
disease: a meta-analysis of VBM studies. Neurobiology of Aging, 32(10), 417–23.
1733–41. McKeith, I., et al. (2007). Sensitivity and specificity of dopamine transporter
Filippini, N., et al. (2009). Distinct patterns of brain activity in young carriers imaging with 123i-FP-CIT SPECT in dementia with Lewy bodies: a
of the ApoE-epsilon4 allele. Proceedings of the National Academy of phase III, multicentre study. Lancet Neurology, 6(4), 305–13.
Science USA, 106(17), 7209–14. McKhann, G., et al. (1984). Clinical diagnosis of Alzheimer’s disease: report
Gilman, S., et al. (2008). Second consensus statement on the diagnosis of of the NINCDS-ADRDA work group under the auspices of Department
multiple system atrophy. Neurology, 71(9), 670–6. of Health and Human Services Task Force on Alzheimer’s Disease.
Good, C.D., et al. (2001). A voxel-based morphometric study of ageing in 465 Neurology, 34(7), 939–44.
normal adult human brains. Neuroimage, 14(1 Pt 1), 21–36. McKhann, GM., et al. (2011). The diagnosis of dementia due to
Gosche, K.M., et al. (2002). Hippocampal volume as an index of Alzheimer Alzheimer’s disease: recommendations from the National Institute on
neuropathology: findings from the Nun study. Neurology, 58(10), Aging-Alzheimer’s Association workgroups on diagnostic guidelines for
1476–82. Alzheimer’s disease. Alzheimer’s Dementia, 7(3), 263–9.
Grady, C.L., et al. (1999). The effects of age on the neural correlates of episodic Mielke, M.M., et al. (2009). Regionally-specific diffusion tensor imaging in
encoding. Cerebral Cortex, 9(8), 805–14. mild cognitive impairment and Alzheimer’s disease. Neuroimage, 46(1),
Gutchess, A.H., et al. (2005). Aging and the neural correlates of successful 47–55.
picture encoding: frontal activations compensate for decreased Mitchell, A.J. and Shiri-Feshki, M. (2009). Rate of progression of mild
medial-temporal activity. Journal of Cognitive Neuroscience, 17(1), cognitive impairment to dementia—meta-analysis of 41 robust inception
84–96. cohort studies. Acta Psychiatrica Scandinavica, 119(4), 252–65.
Han, S.D., Bangen, K.J., and Bondi, M.W. (2009). Functional magnetic Morcom, A.M., et al. (2003). Age effects on the neural correlates of successful
resonance imaging of compensatory neural recruitment in aging and memory encoding. Brain, 126(Pt 1), 213–29.
risk for Alzheimer’s disease: review and recommendations. Dementia Mori, S. and Van Zijl, P.C. (2002). Fiber tracking: principles and strategies—a
and Geriatric Cognitive Disorders, 27(1), 1–10. technical review. NMR Biomedicine, 15(7–8), 468–80.
Heise, V., et al. (2011). The ApoE var epsilon 4 allele modulates brain white National Institute for Health and Clinical Excellence (2007). Dementia: a
matter integrity in healthy adults. Molecular Psychiatry, 16(9), 908–16. NICE-SCIE guideline on supporting people with dementia and their carers
Henneman, W.J., et al. (2009). Hippocampal atrophy rates in Alzheimer in health and social care. NICE, Leicester.
disease: added value over whole brain volume measures. Neurology, Neary, D., et al. (1998). Frontotemporal lobar degeneration: a consensus on
72(11), 999–1007. clinical diagnostic criteria. Neurology, 51(6), 1546–54.
Herholz, K. (2011). Perfusion SPECT and FDG-PET. International Okuizumi, K., et al. (1994). ApoE-epsilon 4 and early-onset Alzheimer’s.
Psychogeriatrics, 23 Suppl 2, S25–31. Nature Genetics, 7(1), 10–1.
Herholz, K. and Ebmeier, K. (2011). Clinical amyloid imaging in Alzheimer’s Park, D.C. and Reuter-Lorenz, P. (2009). The adaptive brain: aging and
disease. Lancet Neurology, 10(7), 667–70. neurocognitive scaffolding. Annual Review of Psychology, 60, 173–96.
Herholz, K., et al. (2002). Direct comparison of spatially normalized PET Patenaude, B., et al. (2011). A Bayesian model of shape and appearance for
and SPECT scans in Alzheimer’s disease. Journal of Nuclear Medicine, subcortical brain segmentation. Neuroimage, 56(3), 907–22.
43(1), 21–6. Persson, J., et al. (2006). Structure-function correlates of cognitive decline in
Herrmann, L.L., Le Masurier, M., and Ebmeier, K.P. (2008). White matter aging. Cerebral Cortex, 16(7), 907–15.
hyperintensities in late life depression: a systematic review. Journal of Prichep, L.S., et al. (2006). Prediction of longitudinal cognitive decline in
Neurology, Neurosurgery and Psychiatry, 79(6), 619–24. normal elderly with subjective complaints using electrophysiological
Hickie, I., et al. (1995). Subcortical hyperintensities on magnetic resonance imaging. Neurobiology of Aging, 27(3), 471–81.
imaging: clinical correlates and prognostic significance in patients with Reuter-Lorenz, P. (2002). New visions of the aging mind and brain. Trends in
severe depression. Biological Psychiatry, 37(3), 151–60. Cognitive Science, 6(9), 394.
Hickie, I., et al. (1997). Subcortical hyperintensities on magnetic resonance Reuter-Lorenz, P.A. and Cappell, K.A. (2008). Neurocognitive aging and the
imaging in patients with severe depression—a longitudinal evaluation. compensation hypothesis. Current Directions in Psychological Science,
Biological Psychiatry, 42(5), 367–74. 17(3), 177–82.
Reuter-Lorenz, P.A., Stanczak, L., and Miller, A.C. (1999). Neural recruitment Smith, S.M., et al. (2006). Tract-based spatial statistics: voxelwise analysis of
and cognitive aging: two hemispheres are better than one, especially as multi-subject diffusion data. Neuroimage, 31(4), 1487–505.
you age. Psychological Science, 10(6), 494–500. Smith, S.M., et al. (2007). Acquisition and voxelwise analysis of multi-subject
Ridha, B.H., et al. (2006). Tracking atrophy progression in familial Alzheimer’s diffusion data with tract-based spatial statistics. Nature Protocols, 2(3),
disease: a serial MRI study. Lancet Neurology, 5(10), 828–34. 499–503.
Roman, G.C., et al. (1993). Vascular dementia: diagnostic criteria for Smith, S.M., et al. (2009). Correspondence of the brain’s functional
research studies. Report of the NINDS-AIREN international workshop. architecture during activation and rest. Proceedings of the National
Neurology, 43(2), 250–60. Academy of Science USA, 106(31), 13040–5.
Rypma, B., et al. (2001). Age differences in prefrontal cortical activity in Styner, M., et al. (2003). Statistical shape analysis of neuroanatomical
working memory. Psychology and Aging, 16(3), 371–84. structures based on medial models. Medical Image Analysis, 7(3),
Scher, A.I., et al. (2011). Hippocampal morphometry in population-based 207–20.
incident Alzheimer’s disease and vascular dementia: the HAAS. Journal Tschampa, H.J., et al. (2005). MRI in the diagnosis of sporadic
of Neurology, Neurosurgery and Psychiatry, 82(4), 373–6. Creutzfeldt-Jakob disease: a study on inter-observer agreement. Brain,
Seshadri, S., et al. (2002). Plasma homocysteine as a risk factor for dementia and 128(Pt 9), 2026–33.
Alzheimer’s disease. New England Journal of Medicine, 346(7), 476–83. Ukmar, M., et al. (2008). Evaluation of white matter damage in patients with
Sexton, C.E., et al. (2011). A meta-analysis of diffusion tensor imaging in Alzheimer’s disease and in patients with mild cognitive impairment
mild cognitive impairment and Alzheimer’s disease. Neurobiology of by using diffusion tensor imaging. Radiology and Medicine, 113(6),
Aging, 32(12), 2322 e5–18. 915–22.
Sexton, C.E., et al. (2012a). Magnetic resonance imaging in late-life Van De Pol, L.A., et al. (2006). Hippocampal atrophy on MRI in
depression: multi-modal examination of network disruption. Archives of frontotemporal lobar degeneration and Alzheimer’s disease. Journal of
General Psychiatry, 69(7), 680–9. Neurology, Neurosurgery and Psychiatry, 77(4), 439–42.
Sexton, C.E., Mackay, C.E., and Ebmeier, K.P. (2012b). A systematic review Van Norden, A.G., et al . (2008). Subjective cognitive failures and
and meta-analysis of magnetic resonance imaging studies in late-life hippocampal volume in elderly with white matter lesions. Neurology,
depression. American Journal of Geriatric Psychiatry, 29 February 71(15), 1152–9.
(Epub). Vemuri, P., et al. (2008). Alzheimer’s disease diagnosis in individual subjects
Sheline, Y.I., et al. (2010). Support for the vascular depression hypothesis using structural mr images: Validation studies. Neuroimage, 39(3),
in late-life depression: results of a 2-site, prospective, antidepressant 1186–97.
treatment trial. Archives of General Psychiatry, 67(3), 277–85. Vemuri, P., et al. (2009). MRI and CSF biomarkers in normal, MCI, and AD
Simpson, S., et al. (1998). Is subcortical disease associated with a poor subjects: diagnostic discrimination and cognitive correlations. Neurology,
response to antidepressants? Neurological, neuropsychological and 73(4), 287–93.
neuroradiological findings in late-life depression. Psychology and Venneri, A. (2007). Imaging treatment effects in Alzheimer’s disease. Magnetic
Medicine, 28(5), 1015–26. Resonance Imaging, 25(6), 953–68.
Smith, A.D., et al. (2010). Homocysteine-lowering by B vitamins slows Zuo, X.N., et al. (2010). Reliable intrinsic connectivity networks: test-retest
the rate of accelerated brain atrophy in mild cognitive impairment: a evaluation using ICA and dual regression approach. Neuroimage, 49(3),
randomized controlled trial. PLoS One, 5(9), e12244. 2163–77.
CHAPTER 13
Psychopharmacology
in older people
Craig W. Ritchie
Using drugs in older people presents the clinician with some Polypharmacy
unique challenges. Older people are a distinct but heterogeneous It is clear that older patients who are at greater risk as a group of
population when considering the use of psychotropic medication. more numerous different physical comorbidities are likely to be on
Gerontological factors that may alter the individual’s pharmacoki- a high number of concomitant medications, sometimes inappro-
netics and the pharmacodynamics of the drug interact with clinical priately. Highlighting this, in a recent US study of 786 people over
or geriatric factors such as concurrent medical illness, compliance, the age of 65 receiving home care, the average number of medi-
and polypharmacy to generate a different (and at times unpredict- cations received was 8.0 (SD = 3.7), and the review noted that in
able) reaction to a drug in an older person compared to a younger almost one-third there was possible inappropriate use of at least
one. This chapter will discuss psychopharmacology that is relevant one drug and in 10% there would have been dangerous drug inter-
specifically to older people. The pharmacodynamic effects of drugs actions (Cannon et al., 2006). These data were entirely consistent
(i.e. their mode of action) are the same as they are in younger popu- with a pan-European study of 2707 people (again living at home)
lations and will not be dealt with here, an excellent review of this which showed that 20% of patients were being prescribed at least
area being found elsewhere (Shiloh et al., 2002). Moreover, other one drug inappropriately and this was more likely to occur in peo-
chapters in this book discuss the treatments of specific mental ill- ple receiving more drugs, those living in poorer socioeconomic
nesses and will consider pharmacological interventions from an conditions, suffering from depression, or being prescribed anxi-
efficacy, effectiveness, and tolerability perspective. Instead, this olytics (Fialova et al., 2005). Polypharmacy increases the likelihood
chapter will focus on issues pertinent to prescribing psychotropics of drug–drug interactions, which will be discussed in more detail
in older people and how age-related (gerontological) and geriatric in the Metabolism section of this chapter. Predicting what a drug–
issues affect this. drug interaction will do can be problematic when only two drugs
Despite the particular issues in prescribing psychotropics in older are involved, but when interactions are between (the average) eight
people, (excepting dementia drugs) there are very few well-designed drugs, then how these all interact would be virtually impossible to
clinical trials of drugs commonly used in older people to educate predict with any certainty. Moreover, where observable side effects
clinicians with regard to their pharmacokinetics, efficacy, optimal develop, these can be managed, but it is likely that drug–drug inter-
dosing regimes, safety, and interactions (Ritchie 2005; Marriott actions may manifest as very low-grade side effects that do not
et al., 2006; Mottram et al., 2006; Rapoport et al., 2007). Instead, old reach the awareness of the prescribing clinician but still have an
age psychiatrists and other clinicians have to extrapolate findings impact on a patient’s quality of life and wellbeing. It has also been
from younger populations that have been subject to higher quality calculated that with two drugs, the risk of a drug–drug interaction
research or prescribe from ‘first principles’ of physiology and phar- is 5.6% but rises to 100% if eight drugs are coprescribed (Sloan,
macology, supporting this with clinical experience. 1983).
Clinical Geriatrics and Psychopharmacology Compliance

Clinical geriatricians (and psychogeriatricians) manage diseases or Several terms are often now used to describe a patient’s taking of
illness states in older people. The use of drugs in this context is medication: adherence, concordance, and compliance. Each has a
educated by several relevant factors associated not with ageing itself slightly different meaning that articulates the relationship between
but with the presence of illness in older people and the way they are prescriber and patient, though all are reasonably synonymous as a
managed by their prescribing clinician. These are: description of how much of prescribed medication is taken. In the
following, terms are used as they were by the authors cited. As well
◆ polypharmacy
as the risks involved in polypharmacy, poor prescribing practice
◆ compliance may lead to uncertainty with regard to compliance or concordance
◆ clinical comorbidities between patient and the prescribing clinician (Bergman-Evans,
2006). From a large study in 11 European countries, Cooper et Gerontological Changes and
al. noted that 12.5% of 3881 older adults were not fully adher-
ent to their prescribed medication (Cooper et al., 2005). Factors Psychopharmacology
that raised the risk of nonadherence included problem drinking, Gerontological changes are part of normal ageing (i.e. in the
cognitive impairment, and an absence of regular clinical review. absence of any definable disease pathology) that may affect the use
Interventions to improve compliance in older patients though have of psychotropic medication.
been reported, with the most ‘striking’ improvement being noted Gerontological changes can affect the pharmacodynamic effects
with strategies incorporating telephone reminders to patients (van of drugs as the target organ may become more susceptible to both
Eijken et al., 2003). Poor concordance or compliance is highly rel- the development of side effects and manifestation of pharmacologi-
evant clinically as it makes it hard for a clinician to discern the cal efficacy. They can also affect the pharmacokinetics of the drug.
tolerability or efficacy of a drug on the background of uncertainty
regarding compliance. Changing pharmacodynamics
In psychiatry, the most relevant target organ is the brain. Ageing is
Clinical comorbidities associated with varying degrees of neuronal loss even without the
The compliance and polypharmacy issues discussed already could development of a neurodegenerative disorder (Thal et al., 2004),
be considered as nonbiological or external factors that create chal- with some, in particular cholinergic, neurons being more suscep-
lenges in using drugs in older patients. There are, however, very tible than others (McKinney and Jacksonville, 2005). In this con-
important biological or internal factors as well that are worthy of text, a dose of drug given to an older person will lead to a greater
consideration when using psychotropic medications in this popu- proportion of receptors being occupied than would be observed if
lation. Drug–patient interactions are relevant when the patient in the same dose were given to a younger person. The effect of this
question exhibits one or several clinical comorbidities. would be to create both a greater efficacy of the drug as well as an
As a consequence of physical illness, older patients may be par- increased likelihood of predictable side effects, e.g. tardive dyski-
ticularly vulnerable to the unwanted but well-recognized adverse nesia (Jeste, 2000).
effects of psychotropic medications. Examples of this include the Even taken in isolation of the pharmacokinetic effects of ageing
increased risk of tardive dyskinesia with increasing age associated to be articulated in the next section, these pharmacodynamic effects
with conventional neuroleptics (Jeste, 2000), increased risk of delir- would argue that lower doses are required in older patients for sim-
ium with drugs that have anticholinergic effects (Tune, 2001), and ilar efficacy and tolerability as is seen in younger populations.
postural hypotension (Mukai and Lipshiz, 2002) with psychotropic
drugs that inhibit α1 adrenergic and H1 histamine receptors. Older Unpredictable pharmacokinetics
people are more vulnerable to these effects because of, for exam- This section will articulate gerontological changes that may take
ple, an increased likelihood of substantia nigra or cholinergic path- place which affect the pharmacokinetics of drugs in older people,
way degeneration or autonomic instability, respectively. They are i.e. what the body does to the drug. The manner in which the body
also less able to tolerate the consequences of such interactions. For interacts with the drug can be summarized by the acronym ADME
example, an older person who suffers orthostatic hypotension may (Absorption (of drug), Distribution (of drug), Metabolism (of drug)
well fall and fracture a hip, whereas a younger person with a simi- and Elimination (of drug)). Polypharmacy, more prevalent in older
lar drop in blood pressure would simply feel a little light-headed, people, may cause drug–drug interactions which take place usually
and stronger muscle tone, reaction times, and visuospatial aware- through either effects on metabolism (cytochrome P450 system) or
ness would prevent the fall and if these factors did not, the femur distribution of drug (e.g. competition for plasma protein binding).
is less likely to fracture than the possibly osteopenic bone of the This problem is particularly relevant in older people and it will be
older person. considered here in some detail.
There may be other characteristics of individuals who are receiv-
ing the drug that make them even more vulnerable to side effects. Absorption
By way of example, there have been concerns raised regarding Absorption of a drug is defined as the process by which the drug
an increased risk of stroke and the use of atypical antipsychotics is moved from the site of administration into the plasma of the
in the management of behavioural problems in dementia (Smith systemic circulation. By definition then, drugs given intravenously
et al., 2004). If this risk is genuine, and this has been challenged (IV) are not absorbed.
(Schneider et al., 2006), then it would appear to be restricted to The vast majority of psychotropic drugs are given orally, with the
only those with pre-existing cerebrovascular damage (Shah and exception of intramuscular long-acting depots of either traditional
Suh, 2005). This latter observation would explain why this puta- or atypical antipsychotics. Intramuscular injections of acute-acting
tive safety concern has not been demonstrated in patients with psychotropics are also available for rapid tranquilization. There is
schizophrenia (Ritchie, 2005). This example illustrates the point little evidence to suggest that absorption of drugs following intra-
that medical morbidity can render the patient more vulnerable to muscular injections is any different in older people than it is in
side effects that, in younger patients and in healthy older people, younger people.
would not be observed. The only other means of administration used for psychotropic
In conclusion, because of the clinical or geriatric effects listed, drugs are sublingual, and this applies to preparations of Olanzapine
i.e. polypharmacy, uncertainty about compliance, comorbid physi- (Zyprexa Velotabs) and Risperidone (Quicklets), and IV. Sublingual
cal illness, and an increased likelihood of developing symptomatic administration of a drug is used to try to achieve more rapid absorp-
side effects, much initial thought and ongoing additional vigilance tion (Markowitz, 2006) and decreased first pass metabolism by the
needs to be exercised in prescribing in older people. liver. There is little evidence to suggest that sublingual absorption of
CHAPTER 13 psychopharmacology in older people 193
drugs is affected by ageing (Kharasch et al., 2004). The IV prepara- through disease, the use of most psychotropics is affected as first
tions that conceivably may be used in older people are of benzodi- pass metabolism decreases notably. In patients with hepatic disease
azepines, but as IV administration by definition avoids absorption, a great deal of caution must be used in prescribing psychotrop-
any age-related differences with IV drugs are only related to distri- ics (British National Formulary, 2006). However, for drugs (like
bution, metabolism, elimination, and pharmacodynamic changes. diazepam) with a lower extraction ratio, the age-related reduction
Absorption of drugs given orally is through the gut wall predom- in liver size does not affect the drug’s pharmacokinetics, though
inantly in the stomach or the small intestine. A drug’s absorption a reduction in enzymatic metabolic degradation associated with
is affected by its state of ionization and this is dependent upon pH. ageing (Herrlinger and Klotz, 2001; Kinirons and O’Mahony, 2004)
Psychotropic drugs are almost exclusively absorbed in the small will reduce the amount of drug metabolized in the liver and hence
intestine as at low gastric pH. These drugs are highly ionized which increase the drug’s bioavailability. The effects of ageing on hepatic
has the effect of inhibiting the drug’s movement across lipid cell drug metabolism are revisited in the section on Metabolism.
membranes. Occasionally drugs excreted in the bile are reabsorbed
in the distal small bowel so-called enterohepatic circulation. Distribution
Therefore, there are three main factors that could influence the To achieve their effect, all psychotropic drugs must cross the blood–
absorption of orally administered, psychotropic drugs into the sys- brain barrier; to do this these drugs must be highly lipophilic. This
temic circulation and these are bowel transit time, hepatic extrac- means that they will also be able to accumulate in body fat and
tion from the portal circulation, and intraluminal gastrointestinal have, what is termed, a high volume of distribution. Distribution of
pH. Gastric emptying (Gainsborough et al., 1993) and small bowel drugs is also affected by the amount of binding to plasma proteins
motility are not affected by normal ageing (Madsen and Graff, 2004); that a drug exhibits.
however, some drugs given in older people will affect gastric motil- Volume of distribution is the extent of distribution relative to the
ity. Drugs with significant anticholinergic effects are commonly amount of drug in the plasma. It does not refer to any actual vol-
prescribed in older people and will tend to cause constipation or ume but rather defines the relationship between plasma drug con-
reduced gut motility through inhibition of the bowel’s stimulation centrations and the amount of drug in the body. Accordingly, the
by the vagal nerve (Ness et al., 2006); accordingly, coadministered volume of distribution (for drugs with very low plasma concentra-
drugs will remain available for absorption for longer and hence may tions and high distribution to, e.g. fat), rather confusingly, can be
achieve higher plasma concentrations. However, as the steady state greater than the ‘total volume’ of the body. The following formula
of a drug is determined predominantly by metabolism rather than helps explain this:
absorption, delayed bowel transit time, in practice, has little effect
Volume of distribution (VD) = Amount of drug in body (mass)/
on steady state of drugs given chronically. More importantly, drugs
plasma drug concentration
that increase gastrointestinal motility may decrease the amount
of drug absorbed, leading to subtherapeutic levels of psychotrop- A proportional increase in fat mass as people age (Kyle et al.,
ics; cholinergic, adrenergic, and serotonergic drugs may all lead to 2001) due mainly to a reduction in muscle mass (Ruiz-Torres et al.,
increased gastric motility and this should be borne in mind when 1995) may thus create a larger bioavailable store of parent drug in
prescribing drugs with these properties or, for similar reasons, in lipid compared to younger, leaner patients. This may increase the
patients with diarrhoea of any other aetiology. drug’s half-life, as has been observed with diazepam. The effect of
The other main factor that affects absorption of drug going into this is to increase the length of time to reach steady state (if dosing
the systemic circulation is how much drug is extracted by the intervals are reduced) but with little effect on its eventual steady
liver, also known as first pass metabolism. There is ongoing debate state plasma concentration, providing metabolic pathways are not
about hepatic changes with ageing and their effect on drug absorp- compromised. If single doses are given, e.g. for tranquilization, then
tion (Cusack, 2004). In normal ageing, liver weight decreases the higher volume of distribution in older people will prolong the
(Schmucker, 1998) and hepatocytes decrease in both number and duration of action of the drug. If dosing intervals are not adjusted,
metabolic function, not through a reduction in enzymatic activity because most psychotropic drugs use metabolic pathways that (at
(Schmucker et al., 1990; Shimada et al., 1994) but possibly through therapeutic doses) are not saturated, there is unlikely to be accu-
a reduction in oxygen diffusion to hepatocytes following defenes- mulation of drug beyond the normally expected steady state. This
tration, thickening of the hepatic sinusoid endothelium, and infre- is because the main influence on steady state is neither absorption
quent development of the basal lamina (Le Couteur et al., 1991, nor distribution but metabolism. In conclusion, alterations to the
2001). This effect, known as pseudocapillarization, reduces oxygen volumes of distribution of drugs given chronically to older people
supply to hepatocytes which is necessary for energy-dependent will have little effect on steady state.
phase 1 enzymatic metabolism (Le Couteur et al., 1998, 1999, Distribution of drug is also affected by binding to plasma pro-
2001; Cogger et al., 2003; McLean et al. 2003). Therefore decreased teins. Basic drugs bind mainly to albumin and acidic ones to α1-
hepatic extraction (by reduced liver mass) is augmented by a reduc- glycoprotein. Most psychotropic drugs are basic. An alteration of
tion in metabolic enzyme function (by reduced hepatocyte oxygen- plasma protein binding in drugs that are normally, extensively pro-
ation). Accordingly, more active drug is passed into the systemic tein bound (for example fluoxetine), may alter the free drug con-
circulation. The net effect of decreased liver mass and function with centration. Protein binding can therefore be affected by a reduction
ageing is probably only relevant for drugs with a high extraction in plasma albumin or competition between drugs for protein bind-
ratio like clomethiazole, where there would be a reduction in the ing, as is the case with the coadministration of fluoxetine and war-
amount of drug removed from the portal circulation for metabo- farin. From this latter example, when prescribing SSRIs in patients
lism—hence more parent drug passes into the systemic circulation. on warfarin, the INR must be checked and adjustments of the dose
Beyond normal ageing in patients with impaired hepatic function of warfarin made accordingly.
Plasma albumin levels decrease slightly in normal ageing (Wallace able to induce Cyto P450 isoenzymes, though recent research has
and Verbeek, 1987) but more dramatically during acute illness, liver demonstrated that the ability of drugs to induce the activity of Cyto
disease, postoperatively, and where malnutrition is present. In these P450 isoenzymes does not diminish with healthy ageing (Cusack,
circumstances to a greater or lesser degree, while total drug levels 2004). Hence, enzyme induction described in younger populations
may remain constant, the amount of free drug could increase. remains relevant in older people. Moreover, phase 2 conjugation
reactions are little affected by age (Durnas, 1990).
Metabolism
It is also recognized that the activity of certain metabolic path-
The metabolism of psychotropic drugs involves two types of proc- ways can show variation at a population level, i.e. people can be
ess, referred to as phase 1 and phase 2 metabolism. The aim of grouped into effective (or fast) and ineffective (or slow) acetyla-
metabolism (or biotransformation) of drugs is to create molecules tors. Genetic polymorphisms that predict this grouping can be dis-
that are less lipid soluble (or more lipophobic). When in this state, cerned through reasonably straightforward genomic analysis. This
they do not easily diffuse from the distal tubule in the kidney back technology, if applied judiciously, may help clinicians predict who
to the peritubular capillaries, after having passed to there in the is more or less likely to have low (or higher) plasma levels of drug
glomerular filtrate from plasma through Bowman’s capsule into following the same dose based on their genotype. This observation
the renal tubule. The metabolism of drugs uses enzymes located on comes within the broader and emerging field of pharmacogenetics
the endoplasmic reticulum of hepatic cells, though some metabolic Prediction through genetic analysis of a patient’s ability to metab-
enzymes are also located in luminal cells of the small intestine. olize drug, prior to commencing treatment, would help avoid toler-
Phase 1 metabolism involves oxidation, reduction, or hydrolysis ability issues, which may aid compliance. This technology in theory
and may result in molecules that retain some pharmacological offers a great deal (particularly in managing depression (Steimer
activity, e.g. fluoxetine and its metabolite norfluoxetine. Sometimes et al., 2001)), but its use has been very limited in practice (Nebert
products of phase 1 metabolism are sufficiently lipophobic to be et al., 2003). The use of this technology, however, in older people
excreted in urine, but often they require a further metabolic step to may be of particular benefit given the high levels of polypharmacy
ensure excretion. and greater susceptibility to the toxic effect of drugs.
Phase 2 metabolism almost invariably leads to the creation of Table 13.1 lists commonly used psychotropic medication and
pharmacologically inactive compounds that are highly lipophobic. which isoenzyme is involved in its metabolic biotransformation. A
Phase 2 reactions involve the addition of a small, highly polar mol- drugs effect to either inhibit or induce the isoenzymes activity is
ecule to the parent compound or drug; this process is known as also presented.
conjugation. Conjugation may involve the addition of glucuronic
acid, acetic acid, sulphuric acid, or an amino acid. These processes, Excretion
like phase 1 reactions, take place predominantly in the liver’s hepa- Drugs are excreted either in faeces or in urine. Psychotropic drugs
tocytes on smooth endoplasmic reticulum, but other tissues may are mainly excreted (eliminated) through the kidneys. As all psy-
also be involved in phase 2 reactions, e.g. lung and kidney. chotropic drugs are lipid soluble (lipophilic), they need to be metab-
Phase 1 processes are energy dependent, requiring the presence olized through phase 1 and phase 2 reactions to make them more
of both nicotinamide adenine dinucleotide phosphate (NADPH) water soluble, as it is in this state that they are less able to move pas-
and the haem-containing protein cytochrome P450 (CytoP450). sively across lipid cell membranes. Free drug (whether lipophilic
Various different types of cytochrome P450 exist known as isoen- or hydrophilic) in plasma is able to pass from the plasma into the
zymes, some of which are listed in Table 13.1. These enzymes are renal tubules from Bowman’s capsule in the glomerular filtrate.
also sometimes referred to as microsomal enzymes, a microsome Large plasma proteins do not pass into the glomerular filtrate under
being a small vesicle that is derived from the endoplasmic retic- physiological conditions, so protein-bound drug cannot pass into
ulum; it is these microsomes that actually contain the cells’ cyto- the renal tubule. As well as passive movement through glomerular
chrome P450 enzymes. filtration into the renal tubule (which usually accounts for approxi-
As noted previously, energy deficiency in ageing hepatocytes fol- mately 20% of free drug), the remaining 80% of drug retained in the
lowing pseudocapillarization and subsequent impairment in oxy- peritubular capillaries may be subject to active secretion into the
gen diffusion to the cell can lead to a reduction in the activity of renal tubule, though this does not happen to an appreciable extent
microsomal enzymes. Reduced hepatic blood flow and reduced with psychotropic drugs. Once in the renal tubule, lipid-soluble
microsomal function could lead to a reduction in the metabo- drugs will diffuse passively back into the peritubular capillaries and
lism of active drugs in older people. This, in theory, could lead to hence back into the systemic circulation. Through metabolic proc-
increases in steady state concentrations of drugs in healthy older ess, lipid-soluble drugs are made more water soluble or hydrophilic
people. However, as stated previously, this situation is much more and thus do not leave the renal tubule and hence are eliminated in
likely to occur in the presence of hepatic failure or when drugs are the urine.
competing for microsomal enzyme activity. Gerontological effects on renal clearance of drugs have an impor-
Drug–drug interactions often take place as a result of the drugs’ tant effect on prescribing in older people. Kidneys change with
impact on phase 1 reactions (Table 13.1). normal ageing; they shrink, the intrarenal vascular intima thick-
Three scenarios can occur. First, two (or more) drugs can com- ens, and glomeruli become sclerotic, with both processes impairing
pete to be a substrate for a single phase 1 isoenzyme; second, one diffusion. There is also infiltration of chronic inflammatory cells
(at least) drug may inhibit a specific or group of Cyto P450 isoen- and fibrosis in the renal stroma (Muhlberg and Platt, 1999; Cusack,
zymes; and third, drugs may induce the activity and/or production 2004). A reduction in the number of glomeruli is observed as well
of Cyto P450 isoenzymes. As the last process is an adaptive or phys- as impairment of function of those surviving (Hoang et al., 2003).
iological response, it has been questioned whether older people are Some drugs that are hydrophilic (e.g. antibiotics) will be more
Table 13.1 Commonly used drugs in psychogeriatics and impact on CytoP450 system
Drug Is a substrate of … Inhibitor of … Stimulator of …
Venlafaxine CYP2C; CYP2D6; CYP3A4 CYP2C; CYP2D6; CYP3A4
Mirtazapine CYP2D6; CYP1A2; CYP3A4 None
Duloxetine CYP2D6, CYP1A2
Fluoxetine CYP1A2; CYP2C; CYP2D6; CYP3A4 CYP2C; CYP2D6; CYP3A4
Paroxetine CYP1A2; CYP2D6; CYP3A4 CYP2D6
Citalopram CYP2C; CYP3A4 None
Fluvoxamine CYP1A2; CYP2C; CYP3A4 CYP1A2; CYP2C; CYP3A4
St John’s wort CYP2D6, CYP3A4 ‘CYP450’a
Nefazodone CYP1A2; CYP2D6; CYP3A4 CYP3A4
Haloperidol CYP1A2; CYP2D6 CYP2D6
Clozapine CYP1A2; CYP2D6 None
Olanzapine CYP1A2; CYP2D6 None
Risperidone CYP2D6 None
Sulpiride None None
Quetiapine CYP3A4 None
Sertindole CYP2D6; CYP3A4 None
Aripiprazole CYP3A4; CYP2D6 None
Diazepam CYP2C; CYP3A4 None
Clomethiazole CYP2E1
Temazepam
Lorazepam
Zopiclone
Zolpidem
Carbamazepine CYP2D6; CYP3A4 CYP2D6
Sodium valproate CYP2D6
Donepezil CYP2D6; CYP3A4 Not reported Not known
Galanthamine CYP2D6; CYP3A4 Not reported Not reported
Rivastigmine Very little CytoP450 metabolism None None
Memantine Very little CytoP450 metabolism None None
Ginkgo biloba Not reported CYP3A4b
a Obach, 2000
b Yale et al., 2005
This table can be supplemented from an excellent US website that details the pharmacokinetics and pharmacodynamics of most psychotropics—<http://www.rxlist.com>.
affected by renal impairment than those that are hydrophobic (e.g. a measure of the patient’s renal function. This is usually described
psychotropics) (Meyers and Wilkinson, 1989). The latter group are by the creatinine clearance, which is a surrogate of the glomerular
rendered inactive by biotransformation to their hydrophilic prod- filtration rate (GFR). To measure this from serum creatinine levels,
uct that remains in the systemic circulation. Hence dose reduction the Cockcroft-Gault formulae is used:
of psychotropic medication in the context of normal ageing is not
usually necessary for renal reasons and even in mild renal failure ( age)
g weight ..23
23 ( .85 iif female)
Creatinine clearance (ml/min) =
little adjustment of these drugs is necessary (see Table 13.2) (British Creat [micromol/L]
National Formulary, 2011). This is not the case with more severe
renal impairment. In women, this result is multiplied by 0.85 (Cockcroft and Gault,
In order to decide whether there is a need to adjust psycho- 1976). A web-based calculator can be used to help with this cal-
tropic drug dosage on the basis of renal impairment one requires culation: <http://nephron.com/cgi-bin/CGSI.cgi>. However, this
Table 13.2 Commonly prescribed psychotropic drugs in older people and modification necessary in presence of renal impairment. Advice from
BNF, September 2011. Absence of advice does not necessarily equate to an absence of concern but rather indicates that no information is available.
Also, adjustments listed are for adults and further adjustments may be necessary or advised in older people per se
Drug (degree of impairment) Comment
Anxiolytics and hypnotics (severe) Start with small doses due to increased cerebral sensitivity.
Amisulpiride Halve dose if eGFR between 30 and 60 ml/min per 1.73 m2; one-third dose if eGFR between 10 and 30 ml/min per 1.73 m2;
with no information available, if eGFR < 10 ml/min per 1.73 m2.
Antipsychotics (severe) Start with small doses; increased cerebral sensitivity (see specific advice on amisulpiride, clozapine, olanzapine, and
quetiapine).
Chloral hydrate (severe) Avoid.
Clozapine (severe) Avoid.
Duloxetine Avoid if eGFR < 30 ml/min per 1.73 m2.
Fluvoxamine (moderate) Start with smaller doses.
Galantamine (severe) Avoid if eGFR < 30 ml/min per 1.73 m2.
Lithium Avoid if possible or reduce dose and monitor plasma concentrations carefully.
Memantine Reduce dose to 10 mg/day if eGFR 30–49 ml/min per 1.73 m2. If well tolerated after at least 7 days, dose can be increased in
steps to 20 mg daily. If eGFR is between 5 and 29 ml/min per 1.73 m2 then reduce dose to 10 mg. Should be avoided if eGFR
< 5 ml/min per 1.73 m2.
Mirtazapine Clearance reduced by 30% if eGFR < 40 ml/min per 1.73 m2; clearance reduced by 50% if eGFR < 10 ml/min per 1.73 m2.
Olanzapine Consider starting dose of 5 mg.
Paroxetine Reduce dose if eGFR < 30 ml/min per 1.73 m2.
Quetiapine Manufacturer advises initial dose of 25 mg daily, increased in daily steps of 25–50 mg.
Valproate (mild to moderate) Reduce dose.
Venlafaxine Use half normal dose if eGFR 10–30 ml/min per 1.73 m2; immediate release tablets may be given once daily.
Gerontological factors Gerontological factors

increasing ‘Bioavailability’ decreasing ‘Bioavailability’
• Absorption: •Absorption:
• Reduced hepatic extraction, • nil

smaller fraction of drug being •Distribution
subject to first pass
metabolism • nil
• Distribution •Metabolism
• Increased volume of • Induced microsomal enzymes

distribution resulting from following drug–drug
higher proportion of body fat interactions
•Reduced albumin for protein •Excretion

binding • nil
• Metabolism
•Reduced microsomal enzyme
activity through reduced
hepatocyte oxygenation
• Excretion
•Reduced kidney mass
•Impaired function of nephron
Fig. 13.1 Gerontological factors and their influence on bioavailability of drugs.
method to ascertain renal function has been criticised as under- Fialova, D., et al. (2005). Potentially inappropriate medication use among
estimating renal function in healthy, older patients (Fliser et al., elderly home care patients in Europe. Journal of the American Medical
1999). Association, 293(11), 1348–58.
Fliser, D., et al. (1999). Renal handling of drugs in the healthy elderly.
Despite these caveats, thresholds do exist that are used to quan-
Creatinine clearance underestimates renal function and pharmacokinetics
tify renal failure into mild (20–50 ml/min); moderate (10–20 ml/ remain virtually unchanged. European Journal of Clinical Pharmacology,
min), and severe (< 10 ml/min). These may be helpful in assisting 55, 205–11.
with dose selection in older patients and should therefore be used Gainsborough, N., et al. (1993). The association of age with gastric emptying.
routinely. Age and Ageing, 22, 37–40.
Herrlinger, C. and Klotz, U. (2001). Drug metabolism and drug interactions
in the elderly. Best Practice Research and Clinical Gastroenterology, 15,
Conclusion 897–918.
Pharmacodynamics has not been discussed in as much detail in Hoang, K., et al. (2003). Determinants of glomerular hypofiltration in aging
this chapter as pharmacokinetics because the mode of action of humans. Kidney International, 64, 1417–24.
each psychotropic drug is similar between young and older people. Jeste, D.V. (2000). Tardive dyskinesia in older patients. Journal of Clinical
Alterations to dosing and administration of drugs in older people Psychiatry, 61 Suppl 4, 27–32.
are mainly driven by pharmacokinetic differences and the increased Kharasch, E.D., Hoffer, C., and Whittington, D. (2004). Influence of age on
the pharmacokinetics and pharmacodynamics of oral transmucosal
frailty of older people. This frailty will lead to side effects of drugs
fentanyl citrate. Anesthesiology, 101, 738–43.
having a greater effect on the patient’s wellbeing. Polypharmacy,
Kinirons, M.T. and O’Mahony, M.S. (2004). Drug metabolism and ageing.
mediated by multiple medical problems, is also a problem and, British Journal of Clinical Pharmacology, 57, 540–4.
though not unique to older people, is certainly more commonly Kyle, U.G., et al. (2001). Fat-free mass percentiles in 5225 healthy subjects
observed in the latter. Moreover, this group is less likely to tolerate aged 15 to 98 years. Nutrition, 17(7–8), 675.
the polypharmacy that they are often exposed to. Le Couteur, D.G. and McLean, A.J. (1998). The aging liver. Drug clearance
As a result of both the geriatric and gerontological changes artic- and an oxygen diffusion barrier hypothesis. Clinical Pharmacokinetics,
ulated above, clinicians should obviously exercise caution in pre- 34, 359–73.
scribing psychotropics in older people. On balance, gerontological Le Couteur, D.G., et al. (1999). Oxidative injury reproduces age-related
factors tend to increase the bioavailability of drugs in older people impairment of oxygen-dependent drug metabolism. Pharmacology
Toxicology, 85, 230–2.
(Fig. 13.1), so we will not go far wrong if we simply adhere to the
Le Couteur, D.G., et al. (2001). Pseudocappilarization and associated energy
familiar adage: limitation in the aged rat liver. Hepatology, 33, 537–43.
‘Start low and go slow.’ Madsen, J.L. and Graff, J. (2004). Effects of ageing on gastrointestinal motor
function. Bowel Transit Time, 33(2), 154–9.
Finally, as can be seen from the review of this area, much work
Markowitz, J.S., et al. (2006). Pharmacokinetics of olanzapine after single-
needs to be done to test whether what is considered from first prin- dose oral administration of standard tablet versus normal and sublingual
ciples actually translates into being clinically relevant. The only administration of an orally disintegrating tablet in normal volunteers.
way to achieve this is to undertake well-designed clinical trials; Journal of Clinical Pharmacology, 46(2), 164–71.
accordingly, old age psychiatrists should be advocating on behalf Marriott, R.G., Neil, W., and Waddingham, S. (2006). Antipsychotic
of patients to research funders, both public and commercial, that medication for elderly people with schizophrenia. Cochrane Database of
our patients should be subject to the benefits (enjoyed by younger Systematatic Reviews, 25(1), CD005580.
populations) of a better evidence base. This will only come from McKinney, M. and Jacksonville, M.C. (2005). Brain cholinergic vulnerability:
better clinical trials research in this population. relevance to behaviour and disease. Biochemistry and Pharmacology,
70(8), 1115–24.
McLean, A.J., et al. (2003). Age related pseudocapillarization of the human
References liver. Journal of Pathology, 200, 112–17.
Bergman-Evans, B. (2006). Evidence-based guideline, improving medication Meyers, B.R. and Wilkinson, P. (1989). Clinical pharmacokinetics of
management for older clients. Journal of Gerontology Nurses, 32(7), antibacterial drugs in the elderly. Implications for selection and dosage.
6–14. Clinical Pharmacokinetics, 17, 385–95.
British National Formulary (2011). British National Formulary. BMJ Mottram, P., Wilson, K., and Strobl, J. (2006). Antidepressants for depressed
Publishing Group, London. elderly. Cochrane Database of Systematic Reviews, 25(1), CD003491.
Cannon, K.T., Choi, M.M., and Zuniga, M.A. (2006). Potentially Muhlberg, W. and Platt, D. (1999). Age-dependent changes of the kidneys:
inappropriate medication use in elderly patients receiving home pharmacological implications. Gerontology, 45, 243–53.
health care: a retrospective data analysis. American Journal of Geriatric Mukai, S. and Lipsitz, L.A. (2002). Orthostatic hypotension. Clinical Geriatric
Pharmacotherapy, 4(2), 134–43. Medicine, 18(2), 253–68.
Cockcroft, D.W. and Gault, M.H. (1976). Prediction of creatinine clearance Nebert, D.W., Jorge-Nebert, L., and Vessell, E.S. (2003). Pharmacogenomics
from serum creatinine. Nephron, 16, 31–41. and ‘individualised drug therapy’: high expectations and disappointing
Cogger, V.C., et al. (2003). Hepatic sinusoidal pseudocapillarisation with aging achievements. American Journal of Pharmacogenomics, 3(6), 361–70.
in the non-human primate. Experimental Gerontology, 38, 1101–07. Ness, J., et al. (2006). Anticholinergic medications in community-dwelling
Cooper, C., et al. (2005). The AdHOC study of older adults’ adherence to older veterans: prevalence of anticholinergic symptoms, symptom
medication in 11 countries. American Journal of Geriatric Psychiatry, burden, and adverse drug events. American Journal of Geriatrics and
13(12), 1067–76. Pharmacotherapy, 4(1), 42–51.
Cusack, B.J. (2004). Pharmacokinetics in older persons. American Journal of Obach, R.S. (2000). Inhibition of human cytochrome P450 enzymes by
Geriatric Pharmacotherapy, 2(4), 274–302. constituents of St. John’s Wort, an herbal preparation used in the
Durnas, C., Loi, M., and Cusack, B.J. (1990). Hepatic drug metabolism and treatment of depression. Journal of Pharmacology and Experimental
aging. Pharmacokinetics, 19, 359–89. Therapy, 294(1), 88–95.
Rapoport, M., Polson, C., and Ritchie, C.W. (2007). Pharmacological Japanese and 30 Caucasians. Journal of Pharmacology and Experimental
treatment of psychosis and depression in neurological disease in older Therapy, 270, 414–23.
adults. In: Tyrer, P. and Silk, K. (eds) Effective treatments in psychiatry. Sloan, R.W. (1998). Drug interactions. American Family Physician, 27,
Cambridge University Press, Cambridge. 229–38.
Ritchie, C.W. (2005). The use of antipsychotic medication for schizophrenia Smith, D.A. and Beier, M.T. (2004). Association between risperidone
occurring in late life. In: Hassett, A., Ames, D., and Chiu, E. (eds) treatment and cerebrovascular adverse events: examining the evidence
Psychosis in the elderly. Taylor and Francis, Abingdon. and postulating hypotheses for an underlying mechanism. Journal of
Ruiz-Torres, A., et al. (1995). Are anthropometric changes in healthy adults American Medical Directors Association, 5(2), 129–32.
caused by modifications in dietary habits or by aging? Gerontology, 41, Steimer, W., et al. (2001). Pharmacogenetics: a new diagnostic tool in the
243–51. management of antidepressive drug therapy. Clinica Chimica Acta,
Schmucker, D.L. (1998). Aging and the liver: an update. Journal of Gerontology 308(1–2), 33–41.
Series A Biological Science and Medical Science, 53, B315–20. Thal, D.R., Del Tredici, K., and Braak, H. (2004). Neurodegeneration
Schmucker, D.L., et al. (1990). Effects of age and gender on in vitro properties of in normal brain aging and disease. Scientific Aging and Knowledge
human liver monooxygenases. Clinical Pharmacology Therapy, 48, 365–74. Environment, 23, pe26.
Schneider, L.S., et al. (2006). Effectiveness of atypical antipsychotic drugs Tune, L.E. (2001). Anticholinergic effects of medication in elderly patients.
in patients with Alzheimer’s disease. New England Journal of Medicine, Journal of Clinical Psychiatry, 62 Suppl 21, 11–14.
355(15), 1525–38. van Eijken, M., et al. (2003). Interventions to improve medication compliance
Shah, A. and Suh, G.H. (2005). A case for judicious use of risperidone and in older patients living in the community: a systematic review of the
olanzapine in behavioural and psychological symptoms of dementia literature. Drugs and Aging, 20(3), 229–40.
(BPSD). International Psychogeriatrics, 17(1), 12–22. Wallace, S.M. and Verbeek, R.K. (1987). Plasma protein binding of drugs in
Shiloh, R., Nutt, D., and Weizman, A. (2002). Atlas of psychiatric the elderly. Clinical Pharmacokinetics, 12, 41–72.
pharmacotherapy. Martin Dunitz, London. Yale, S.H. and Glurich, I. (2005). Analysis of the inhibitory potential of
Shimada, T., et al. (1994). Interindividual variations in human liver Ginkgo biloba, Echinacea purpurea, and Serenoa repens on the metabolic
cytochrome P-450 enzymes involved in the oxidation of drugs, activity of cytochrome P450 3A4, 2D6, and 2C9. Journal of Alternative
carcinogens and toxic chemicals: studies with liver microsomes of 30 Complementary Medicine, 11(3), 433–9.
CHAPTER 14
Brain stimulation therapies
Daniel W. O’Connor and Chris Plakiotis
Electroconvulsive therapy, transcranial magnetic stimulation, and include an age-related intolerance of antidepressant and antipsy-
deep brain stimulation all entail the delivery of electrical energy, chotic medications coupled with an increase with age in depres-
either episodically or continuously, to the brain with the object of sive psychomotor retardation and psychosis, both of which predict
relieving mental or neurological symptoms. This chapter discusses a good response to ECT. It might also reflect a view by psychiatrists
their indications, application, effectiveness, and safety with respect that severely depressed old people benefit more from medically ori-
to depression and other psychiatric conditions. Because the lat- ented therapies. It is conceivable, therefore, that treatment patterns
ter two treatments are relatively novel, the bulk of this chapter is will change yet again now that medications are safer and the role of
devoted to electroconvulsive therapy. psychotherapy is expanding.
Does ECT work?

Electroconvulsive Therapy Depression
Electroconvulsive therapy (ECT) is an effective treatment of severe A series of randomized controlled trials in the 1970s and 1980s estab-
depression, mania, catatonia, and some cases of schizophrenia. It is lished that real ECT was superior to sham ‘treatments’ that entailed
important, therefore, that old age psychiatrists have a good under- an anaesthetic but no electrical stimulation. In a meta-analysis of
standing of its uses, benefits, and risks. Most researchers have failed these studies, depressed patients given real ECT progressed faster
to distinguish between older and younger patients in their reports and scored 9.7 fewer points on the 64-point Hamilton Depression
and so much of the evidence presented in this chapter is generic in Rating Scale (HDRS) 2–4 weeks later (UK ECT Review Group,
nature and readers must apply the information as best they can to 2003).
their own practice. ECT also proved superior to a range of older and newer antide-
ECT’s mechanisms of action are still poorly understood. It induces pressants over periods of 3–12 weeks in all but two of the13 studies
the release of prolactin, oxytocin, and other hormones related to summarized by the UK ECT Review Group (2003). This difference
the hypothalamic-pituitary axis; it boosts cerebral blood flow and in response translated to 5.2 fewer points on the HDRS. In one of
metabolism, particularly in frontal and anterior temporal areas; the largest trials, 66% of patients given ECT had no or few symp-
and it stimulates neurogenesis in animal models (McCormick et al., toms after 4 weeks compared with 42% of those on imipramine
2007; Bolwig, 2011). Contrary to expectation, ECT has no proven, 200 mg daily (Medical Research Council, 1965).
consistent links with changes in the production, reuptake, and ECT can certainly work quickly and effectively when adminis-
metabolism of the neurotransmitters serotonin and noradrenaline, tered by skilled clinicians to carefully selected patients. In a large,
which explains perhaps its benefits in patients resistant to pharma- prospective North American study of suprathreshold bilateral ECT,
cotherapy (Grover et al., 2005). 75% of patients had remitted after seven sessions as shown by scores
of 10 or less on the HDRS (Husain et al., 2004). The same might not
Patterns of use apply, though, when patients with complex, chronic conditions are
ECT declined in use in most countries from the 1970s onwards, due treated in mainstream facilities using suboptimal methods. As an
to the advent of effective antidepressant and antipsychotic medica- example, in a survey of hospitals in New York state where treatment
tions and to its depiction in popular media as a tool to control aber- practices varied widely, HDRS scores fell below 10 points in only
rant behaviour. Few countries or regions publish complete datasets, 47% of all cases. To make matters worse, 40% of accrued improve-
but, judging from the number of research publications, ECT and ment was lost within 10 days and relapse rates ranged from 46% to
other brain stimulation therapies are in the ascendancy. 79% across sites (Prudic et al., 2004).
ECT is prescribed more to older people than younger ones in There have been remarkably few prospective controlled studies
Australia (Plakiotis et al., 2012), Denmark (Munk-Olsen et al., of ECT in older people. When O’Leary et al. (1994) extracted data
2006), the US (Olfson et al., 1998), and probably most other coun- from a larger study regarding its 35 aged subjects, their scores on
tries. In Victoria, Australia, where data from all public and private the HDRS fell 31.7 points on average after six real ECTs versus 10.3
hospitals are collated systematically, applications rose steeply with with sham ones. Bilateral ECT worked a little better and faster than
age until 85 years (Plakiotis et al., 2012) (Fig. 14.1). Likely reasons unilateral treatment. Similarly, Flint and Rifat (1998) found in an
100
90
80
Rate per 100,000 persons

70
60
50
40 Female
Male
30
20
0
25− 35− 45− 55− 65− 75− >84
Age
Fig. 14.1 ECT application rates per 100,000 for age- and sex-specific populations in 2007, Victoria, Australia (Reproduced with permission from Plakiotis, C., George, K.
and O’Connor, D.W. (2012). Has ECT utilisation remained stable over time? A decade of ECT service provision in Victoria, Australia. Australian and New Zealand Journal of
Psychiatry. © Wiley 2012).
open, nonrandom trial that 88% of aged patients with depressive rates, however defined, of up to 80% in cases of depression comorbid
psychoses responded positively to ECT within 6 weeks compared with Alzheimer’s disease, vascular dementia, Lewy body dementia,
with 25% of those treated with adequate doses of nortriptyline and and frontotemporal dementia (Price and McAllister, 1989; Rao and
perphenazine. HDRS scores fell to 10 or less 3 weeks sooner with Lyketsos, 2000; Rasmussen et al., 2003). These reports take the form
ECT than medication. of small, retrospective case series that are subject to bias: positive
Most other reports describe series of depressed psychogeriatric reports are more likely to be published than negative ones. To coun-
in-patients. In the four largest series, outcomes were mostly positive. ter this, Nelson and Rosenberg (1991) compared 21 consecutive
Benbow (1987) discharged 52% of patients as ‘well’ and another 28% patients with dementia and comorbid major depression with 84 sim-
as ‘improved’, while Godber et al. (1987) found that 74% of patients ilarly aged nondemented ones. The dementia group derived almost
made a ‘full recovery’ or were ‘much improved’. In a report by Tew et al. as much benefit from ECT. While confusion post-ECT was more
(1999), 41% of their ‘oldest-old’ patients made a complete recovery, as common and correlated with the degree of dementia, mini-mental
did 39% of those described by Brodaty et al. (2000). state examination (MMSE) scores actually improved on average by
1.6 points. Confusion complicated ECT in half the cases described
Response predictors
by Rao and Lyketsos (2000) but it lasted only a few days and treat-
Older patients responded better to ECT than younger ones in some ments continued uneventfully. Few reports provide follow-up data,
studies (Wilkinson et al., 1993; O’Connor et al., 2001) but not in though in one, relapse rates were high (Rasmussen et al., 2003). With
others (Hickie et al., 1996; Birkenhäger et al., 2010). Outcome respect to mania in dementia, McDonald and Thompson (2001)
correlates better with depressive psychosis and melancholia, espe- described two patients who responded well to ECT in the short term
cially as reflected by objective signs of psychomotor agitation and but required ongoing treatment to hold symptoms at bay.
retardation, than with age per se. Depressed patients with psy- ECT is not a standard treatment of the disturbed behaviours like
chotic symptoms scored 5.9 points less on average on the HDRS screaming and aggression that arise in mid- to late-stage demen-
than nonpsychotic patients after treatment by Mulsant et al. (1991) tia, except where patients have comorbid depression and fail to
and 8% fewer points after treatment by Petrides et al. (2001). The respond to antidepressant medication. Notwithstanding this, there
combination of psychotic symptoms with observable psychomotor are detailed reports of patients who were not obviously depressed
change has even stronger predictive capacity. Deluded-retarded but benefited greatly from ECT (Carlyle et al., 1991; Holmberg
cases in one placebo-controlled trial lost 34.7 points on the HDRS et al., 1996; Grant and Mohan, 2001; Bang et al., 2008; Sutor and
in contrast to 16.2 in the retarded, nondeluded group and 10.7 in Rasmussen, 2008). Their disturbed behaviours, which posed seri-
the nonretarded, nondeluded group (Buchan et al., 1992). ous risk to themselves and others, resolved quickly, but about half
Certain factors reduce, but do not preclude, the likelihood of required maintenance treatments.
responding to ECT. These include chronicity and resistance to ECT can also be delivered safely and effectively to patients
antidepressant medications (Dombrovski et al., 2005); subcortical with mental disorders secondary to stroke (Currier et al., 1992),
hyperintensities (Steffens et al.,, 2001); medial temporal lobe atrophy Huntington’s disease (Ranen et al., 1994), head injury (Kant et al.,
(Oudega et al., 2011); and personality disorder (Black et al., 1988). 1999), and intellectual disability (van Waarde et al., 2001), though
Dementia and other organic mental disorders relapse rates tend to be high.
Two-thirds of the British old age psychiatrists surveyed by Benbow
(1991) agreed that ECT was ‘often or sometimes’ appropriate when Schizophrenia, mania, and catatonia
depression overlaid on dementia failed to respond to antidepressant All major clinical guidelines accept acute schizophrenia, schizoaf-
medication. Published reports are favourable with improvement fective disorder, mania, and catatonia as possible indications for
CHAPTER 14 brain stimulation therapies 201
ECT (American Psychiatric Association, 2001; Royal College of carbonate. Over the ensuing 24 weeks, 84% of those on no continu-
Psychiatrists, 2005; Royal Australian and New Zealand College of ing medication had relapsed, compared with 60% in the nortriptyl-
Psychiatrists, 2007). ine group and 39% of those on combination therapy. Relapse was
The use of ECT in acute schizophrenia plummeted once antip- more common in patients who had failed to respond to previous
sychotic medications became available in the 1950s. ECT works a pharmacotherapy. In one naturalistic outcome study, relapse rates
little faster than antipsychotics and is just as effective in reducing were higher in cases of psychotic depression (31% versus 22%),
psychotic symptoms (Abraham and Kulhara, 1987), but medica- comorbid psychiatric disorder (43% versus 27%), and personality
tions are simpler to administer on a long-term basis and are pre- disorder (23% versus 14%). Relapse occurred most often within the
ferred by most patients. Despite recent advances, there remain first 12 weeks of follow-up (Prudic et al., 2004).
patients who do not respond to medications, cannot tolerate them, It is now standard practice to introduce an antidepressant medi-
or require urgent intervention to prevent self-injury. Naturalistic cation, either alone or together with an antipsychotic or mood sta-
studies suggest, but do not prove, that half or more of those who bilizer, during the ECT course or soon afterwards (Haskett and Loo,
failed to benefit from standard medications improved substantially 2010). Where this is known to be ineffective or unsafe, continuing
once ECT was added. ECT responders had fewer negative symp- or maintenance ECT is an option. Continuation therapy (C-ECT)
toms and had been unwell for shorter periods than nonresponders follows directly from an acute course of ECT with the object of
(Chanpattana and Andrade, 2006). Old age need not be a barrier to preventing relapse. Treatment intervals are typically stretched
recovery. Kramer (1999) described five women aged 58–74 years from weekly to monthly depending on progress. Maintenance
whose florid schizophrenia or schizoaffective disorders improved ECT (M-ECT) refers somewhat arbitrarily to treatments extend-
dramatically with ECT. Against this, all of six aged patients with ing beyond 6 months, and occasionally for years, where the risk of
late-onset schizophrenia reported by Figiel et al. (1992) became relapse is known to be high (Fink et al., 1996). Intervals between
confused after ECT and none showed symptomatic improvement. treatments are typically extended until breakthrough symptoms
In cases of mania, ECT performs as well or better than standard provide clues to the optimal schedule. More recently, Lisanby et al.
pharmacotherapy (Loo et al., 2011; Versiani et al., 2011). There is (2008) devised a treatment algorithm based on patients’ depression
little information regarding older patients specifically, but clinical rating scores to bring some rigor to this process.
experience suggests that those who cannot tolerate antipsychotic or In a retrospective chart review of 58 aged patients with severe,
mood stabilizing medications, or are so manic that they look to be medication-resistant depression or psychosis, hospital readmis-
delirious, respond best to ECT. Only small numbers of treatments sions fell by 53% in number and 79% in duration in the 2 years after
may be required, even in life-threatening cases of ‘delirious mania’ M-ECT started compared with the previous 2 years (O’Connor
(Karmacharya et al., 2008). et al., 2010a). Within the actual M-ECT treatment period, which
Catatonia is now uncommon, but it arises occasionally in schizo- varied widely, admissions fell by 90% in number and 97% in dura-
phrenia, depression, mania, and organic conditions including brain tion compared with the same time beforehand. Similar findings
tumour, head trauma, stroke, encephalitis, and metabolic distur- emerged in earlier reviews of adult patients (Petrides et al., 2011).
bance with any or all of the following signs: mutism, negativism, M-ECT has also been used to limit relapse in bipolar disorder and
echolalia, echopraxia, muscle rigidity, posturing, and waxy flexibil- schizophrenia (Vanelle et al., 1994; Chanpattana et al., 1999).
ity. Its severest form, lethal catatonia, is associated with extreme The National Institute for Health and Clinical Excellence (2003),
agitation, stupor, fever, sweating, and autonomic instability, leading in a review of ECT, concluded that maintenance therapy could not
to dehydration, exhaustion, and death if left untreated. It must be be recommended on the grounds that its longer-term benefits and
distinguished from neuroleptic malignant syndrome and malig- risks had yet to be established. While case reports are subject to
nant hyperthermia by means of a reliable medication history. ECT publication bias and lack the authority of controlled trials, pains-
proved effective in 60% of cases after an average of 15 treatments taking clinical histories are still persuasive. M-ECT looks to be safe
(van Waarde et al., 2010a), though maintenance therapy may be and there is no evidence of progressive cognitive impairment (Rami
required to prevent relapse (Wilkins et al., 2008). et al., 2004), even in a patient who received 430 treatments over an
Other conditions 8-year period (Barnes et al., 1997).
Since M-ECT is usually delivered on an outpatient basis, patients
ECT can improve tremor, rigidity, bradykinesia, and gait in
must be cooperative, medically stable, and conversant with practi-
Parkinson’s disease for periods of several weeks but is rarely used
cal requirements. Processes must also be in place to conduct physi-
for this indication (Kennedy et al., 2003). It is of greater benefit
cal and mental state checks at predetermined intervals, to gauge the
in neuroleptic malignant syndrome, a condition that overlaps in
intervals between treatments, and to decide when treatment ends
form with catatonia and is precipitated by antipsychotic medica-
(O’Connor et al., 2010b).
tion. Patients who fail to respond to supportive therapy, benzodi-
azepines, and dopamine agonists, or whose lives are in danger, may Treatment
benefit from a brief course of ECT. Care is required, however, since
Anaesthesia
patients are often medically unstable and require constant monitor-
ing (Troller and Sachdev, 1999). The ideal anaesthetic induction agent ensures rapid, painless uncon-
sciousness, an adequate seizure, and speedy recovery with minimal
Relapse and maintenance therapy post-ECT confusion. Methohexital has a rapid onset, short dura-
Relapse is common in the period following ECT. In a study by tion of action and no anticonvulsant properties. By contrast, propo-
Sackeim et al. (2001), patients whose major depression remitted fol is sometimes painful on injection and raises seizure thresholds,
successfully with ECT were randomly assigned to receive pla- though not to a significant degree in usual doses. Its major advan-
cebo, nortriptyline, or a combination of nortriptyline and lithium tage is that pulse rate, blood pressure, and heart rhythm are a little
more stable (Hooten and Rasmussen, 2008). Other agents include Stimulus dosing
alfentanil, remifentanil, etomidate, ketamine, and sevoflurane. In former times, the same electrical dose was given to every patient.
Differences between them with respect to ECT practice are unclear There is no place for this now: treatments are individualized to
(Hooten and Rasmussen, 2008). Little available information refers maximize efficacy and reduce side effects. There is debate, how-
specifically to older people. ever, about how best to achieve this. One approach, called stimulus
Muscle relaxants like succinylcholine virtually eliminate the risk dose titration, entails giving one or more stimulations in the first
of fractures during ECT, though two cases of vertebral compression session to ascertain threshold. One widely used protocol recom-
fractures were reported by Mulsant et al. (1991) in patients known mends that, for women receiving unilateral ECT, electrical charge
to be osteoporotic. Doses of relaxant can be increased if desired is delivered at 32 mC, 48 mC, and then 80 mC until a convulsion is
in these situations. There is no uniform view on the routine use provoked. Once threshold is established, a therapeutic stimulus is
of anticholinergics (usually atropine or glycopyrrolate) just prior delivered at between three times and six times this level, depend-
to treatment to reduce salivation and bradycardia. Most centres ing on local practice. A ‘multiplier’ of three is recommended by the
now use them selectively, given the increased risks of tachycardia Royal Australian and New Zealand College of Psychiatrists (2007).
due to unopposed sympathetic activity, myocardial ischaemia, and Other authorities are less specific (Royal College of Psychiatrists,
post-ECT confusion (Mondimore et al., 1983). 1995; American Psychiatric Association, 2001). For bilateral ECT,
Seizure threshold the ‘multiplier’ is typically one and a half times threshold.
Tonic-clonic seizures are critical to ECT’s success (Cronholm and In experienced hands, stimulus titration takes just a few min-
Ottoson, 1960) and the electrical charge must therefore exceed sei- utes to complete. Inexpert clinicians are less sure-handed, result-
zure threshold to ensure that a convulsion ensues. Thresholds can ing in longer anaesthesia and slower recovery. There is also a risk
vary tenfold or more between patients but lie mostly between 20 that repeated subconvulsive stimulations will precipitate cardiac
and 100 millicoulombs (mC)(Sackeim et al., 1987). Increasing age, arrhythmias. This certainly happened in the 1940s and 1950s but is
male gender, and bilateral electrode placement all predict higher rarely seen now (McCall et al., 1994). There is no evidence that sub-
initial thresholds (Sackeim et al., 1987; van Waarde and van der convulsive stimulation causes more cognitive impairment (Prudic
Mast, 2010b). Men prescribed bilateral ECT will therefore have et al., 1994).
higher mean thresholds than women prescribed unilateral ECT. Age-based dosing strategies are simpler, faster, and possibly safer
Electrical charge is not critical in itself. What matters more is that for very old, frail patients. According to one formula for unilateral
energy levels exceed threshold by a defined amount. Bilateral ECT is ECT, doses for patients aged 60, 70, and 80 years are about 300 mC,
effective at, or just above, seizure threshold. Unilateral ECT is not, as 350 mC, and 400 mC respectively. Doses for bilateral ECT are half
shown in a critical study by Sackeim et al. (2000) in which ‘low dose’ these. While not as individualized as stimulation titration, it works
unilateral treatment (50% above threshold) was much less effective than well enough in most instances. According to Tiller and Ingram
‘high dose’ unilateral treatment (500% above threshold). ‘Moderate (2006), age-based dosing led to only 2% of their women patients
dose’ ECT (150% above threshold) occupied an intermediate position. and 7% of men receiving subtherapeutic doses. Against this, doses
‘High dose’ unilateral ECT worked almost as well as bilateral ECT (50% would have been too high, at seven or more times threshold, for
above threshold) in lifting depressive symptoms (Figure 14.2). 30% of women and 8% of men. In addition, if treatment is not pro-
gressing well, it is harder to take sensible corrective action if thresh-
old is not known. It is advisable therefore to establish threshold,
35 unless there are good reasons not to do so.
Low dose unilateral
Moderate dose unilateral Seizure monitoring
30 High dose unilateral Threshold levels tend to rise as treatment progresses, more for bilat-
High dose bilateral eral than unilateral ECT, and electrical charge must be increased to
keep pace if this happens (Sackeim et al., 1987). It is not practicable
25
to recheck patients’ thresholds week after week. Instead, clinicians
make judgements based on EEG monitoring of seizure duration
and morphology. This is an imprecise science. ECT machines now
HDRS score
20
produce measures of seizure duration, amplitude, interhemispheric
synchronicity, and postseizure electrical suppression, but none of
15 this information can be interpreted exactly.
Greater seizure intensity, regularity and postseizure suppression
10 correlate with better outcomes (Plakiotis and O’Connor, 2011), but
what is cause and what is effect is not clear: patients who respond
well to ECT may possess this sort of profile from the outset (Nobler
5 et al., 1993; Perera et al., 2004). A good ‘electrical picture’ might
therefore predict recovery, not cause it. To complicate matters,
older people typically have shorter, weaker, and less synchronous
0
Baseline After 6th After last 1 week later
seizures, irrespective of clinical outcome, and their thresholds rise
treatment treatment faster as treatment progresses (Nobler et al., 1993). In the absence
Fig. 14.2 Hamilton Depression Rating Scale scores for patients treated with high of clear evidence, it seems reasonable to increase electrical charge
and low energy, unilateral and bilateral ECT (Sackeim et al., 2000). by 50 mC at a step if seizures become obviously shorter and weaker.
Clinicians will naturally differ to some extent in how they interpret this, Dolenc and Rasmussen (2005) cite reports in which lithium
technical parameters (Little et al., 2002). and ECT were combined safely. Though the risk of serious mishap
Treatment variants is slight, lithium should be stopped before starting ECT.
Efforts continue to find ways to boost treatment efficacy while
minimizing adverse effects. One alternative to the standard bilat-
Safety
eral (actually bitemporal) approach is bifrontal ECT in which the ECT is generally very safe. In the US, there were no deaths related
electrodes are placed 3–5 cm above the outer angle of both orbits. to the 73,440 treatments administered to Veterans Affairs patients
In a recent trial, bitemporal ECT worked a little faster, but the two in 1999–2010, placing the risk of ECT at the bottom of the range
placements were otherwise indistinguishable (Kellner et al., 2010). for procedures entailing general anaesthesia (Abrams, 1997; Watts
In another approach, the standard electrical square wave stimu- et al., 2011). This is not because treatment is reserved for physically
lus is shortened from a pulse width of 0.5–2 milliseconds (ms) to fit patients. The reverse is often the case. ECT is given quite com-
an ultrabrief pulse of 0.3 ms, with the object of depolarizing neu- monly to patients with serious medical comorbidities compounded
rons and stimulating seizures, at lower electrical doses. Unilateral by dehydration, inanition, and exhaustion secondary to depression
ultrabrief ECT is certainly effective and is associated with sig- or psychosis. It might actually be safer than psychiatric medications
nificantly fewer cognitive sequelae than standard unilateral ECT in such circumstances. In a chart review by Zielinski et al. (1993),
(Sackeim et al., 2008), but response is slower and more treatments 11 of 21 patients with cardiac disease were forced to discontinue
are required (Loo et al., 2008). Counterintuitively, bilateral ultrabrief treatment with tricyclic antidepressants because of cardiovascular
stimuli are strikingly less effective than unilateral ultrabrief stimuli complications compared with only two of 40 similar patients given
(Sackeim et al., 2008). ECT. Similarly, rates of cardiovascular and gastrointestinal adverse
Some patients have very brief or low-amplitude seizures, and poor events were more common in 39 very old medically treated depres-
clinical outcomes, even at maximal electrical stimulation. Efforts to sives compared with 39 carefully matched patients treated with
lengthen and optimize seizures have included hyperoxygenation, ECT. Their rates of hypertension, myocardial infarction, and heart
pretreatment with oral or intravenous xanthine alkaloids (caffeine, failure were virtually identical (Manly et al., 2000). There are risks,
theophylline, aminophylline), and the addition of short-acting opi- however, as the following sections make clear.
oids (alfentanil, remifentanil) to the induction agent. None of these
Cardiovascular effects
strategies has been subject to rigorous study and xanthine alkaloids
can lead to prolonged seizures, tachycardia, and cardiac arrhyth- ECT exerts its effect on the heart principally by direct neuronal
mias (Loo et al., 2010). transmission of impulses from the hypothalamus to the heart via
With respect to frequency, ECT is traditionally administered parasympathetic and sympathetic tracts. During and immediately
three times a week, but twice weekly treatments are just as effec- after electrical stimulation, an intense parasympathetic surge flows
tive (Charlson et al., 2012), require no more treatments in total, through the vagal nerve to the heart, resulting in a transient sinus
and cause less cognitive impairment (Shapira et al., 1998). Barring bradycardia, with periods of asystole lasting for 2 s on average but
life-threatening cases, there is no benefit in administering ECT occasionally for 5 s or more (Burd and Kettl, 1998). A sympathetic
more often. surge then follows, leading to an average rise in blood pressure of
55 mmHg, and in pulse rate of 37 beats per minute, together with
Concomitant medications clinically benign premature atrial and ventricular contractions.
Benzodiazepines raise seizure thresholds and should therefore be Rare complications at this time include acute myocardial infarction
avoided, reduced, or stopped if possible (Pettinati et al., 1990), but and ventricular fibrillation (Burd and Kettl, 1998). Pulse rates and
modest doses (e.g. lorazepam up to 3 mg daily) have little impact in blood pressure both rise more during bilateral than unilateral ECT,
reality (Boylan et al., 2000) and are permitted in most ECT research and more with higher than lower energy levels (Gangadhar et al.,
trials to lessen anxiety and agitation. Mood stabilizers that are also 2000). Cardiac function usually returns to normal within 15 min,
anticonvulsants (e.g. carbamazepine and valproate) should also be but ST changes and bursts of bigeminy or trigeminy are a little more
stopped if possible. common than usual in the following 24 h (Huuhka et al., 2003).
Medications with anticholinergic properties have been linked
with worsened cognition post-ECT, especially when taken in com- Frequency and severity of adverse events
bination. Mondimore et al. (1983) found that eight of their 12 Aged patients are at higher risk of adverse events. In a report by
patients with high serum anticholinergic levels lost two or more Burke et al. (1987), 15% of their patients aged 60 years and over
points on the MMSE after a single ECT compared with only one experienced a cardiorespiratory complication (mostly changes
of eight patients with lower anticholinergic levels. Tricyclic antide- in blood pressure) compared with 3% of younger patients, and
pressants, some antipsychotics, and even lithium carbonate all have 15% sustained a fall versus none in the younger group. Rates of
discernable anticholinergic activity, as do some analgesics, antihis- post-ECT confusion were 18% and 13%, respectively. Altogether,
tamines, and antispasmodics (Chew et al., 2008). 35% of older patients had a complication, but most were transient
Lithium might be hazardous too when combined with ECT, and settled spontaneously. Side effects also rose with age in a chart
given occasional reports of severe confusion, prolonged seizures, review by Alexopoulos et al. (1984). Cardiovascular complications
and catatonia, even with ‘normal’ blood levels (Sartorius et al., arose in 19% of their patients aged 65 years and over versus 1% of
2005). By way of confirmation, rates of confusion were higher at younger patients, and 14% had neurological complications (mostly
22% in 27 patients given lithium and ECT concurrently, compared confusion) compared with 12%. Apart from an 87-year-old man
with 12% in those whose lithium was stopped just prior to ECT, who died of a myocardial infarction 48 h after his second treatment,
and 6% in those with no exposure (Penney et al., 1990). Against most patients went on to complete their course.
Octogenarians were found by Cattan et al. (1990) to have minutes but extends occasionally for hours and rarely for several
adverse events roughly twice as often as ‘younger old’ patients. days. Risk factors for emergent confusion include bilateral elec-
Cardiovascular complications and falls were both more frequent in trode placement (Sackeim et al., 1993); high electrical dose relative
very old patients than younger ones at 36% versus 12%, and 36% to threshold (Sackeim et al., 1993); a sinusoidal waveform (Daniel
versus 14%, respectively. Confusion was the commonest prob- and Crovitz, 1986); advanced age (Tomac et al., 1997); pre-existing
lem, arising in 59% and 45% of the two groups. Outcomes were dementia (Rao and Lyketsos, 2000); and concomitant anticholiner-
more benign in the patients aged 75 years and over described by gic medications (Mondimore et al., 1983).
Gormley et al. (1998), of whom 7% suffered prolonged confusion Most studies show some reduction in the acquisition and reten-
and 4% became manic. All problems resolved within 2 weeks. tion of new verbal and nonverbal material (anterograde memory)
Similarly, 32% of the patients aged 85 years and over of Tomac et al. over the course of a treatment cycle. Improvement starts within a
(1997) experienced prolonged confusion and 10% fell. One of the few days of treatment completion, more quickly for information
34 patients sustained a fracture between treatments and six died acquisition than retention, and continues for up to 6 months. Recall
of causes unrelated to ECT within 45 days after the last treatment, of past events (retrograde memory) is affected too, especially for
attesting to their very high rates of physical comorbidity. personal or autobiographical memories, and for recent events more
Other uncommon side effects include dental injury, fractures than distant ones (Ingram et al., 2008). Autobiographical memory
secondary to falls (de Carle and Kohn, 2001), and vertebral com- is difficult to measure in a standardized way and so the nature,
pression fractures due to poorly modified seizures (Mulsant et al., extent, and duration of lapses is unclear. Knowledge of events in the
1991). Very rare sequelae include stroke (Bruce et al.,, 2006 and sta- days or weeks prior to treatment may never be recovered, however,
tus epilepticus (Srzich and Turbott, 2000). Note that neurological extending in some cases to a more than a year beforehand (Ingram
asymmetries are common after ECT but generally resolve within et al., 2008).
20 min (Kriss et al., 1978). Risk factors for anterograde and retrograde amnesia include
bilateral electrode placement (Sackeim et al., 2000), high electri-
Prevention and management
cal dose relative to threshold (Sackeim et al.,, 2000), advanced age
Contraindications to ECT include recent myocardial infarction and (Zervas et al., 1993), and limited education (Legendre et al., 2003).
stroke, severe valvular heart disease, clinically significant arrhyth- Impaired cognition at baseline and prolonged disorientation after
mias, unstable angina, uncompensated cardiac failure, and cardiac treatment also increase the risk of retrograde amnesia (Sobin et al.,
and arterial aneurysms, but these rules are not absolute. Patients 1995). Other possible factors include substance abuse, cardiac dis-
who refuse to eat or drink or take essential medications might ease, and neurological disorders (Sackeim, 2000).
respond better to ECT than other treatments after a detailed evalu- Remarkably few studies have focused on old people, despite
ation of their mental and physical health (American Psychiatric their higher rates of premorbid cognitive impairment, and most
Association, 2001; Royal College of Psychiatrists, 2005). ECT has were small and of poor quality. Post-treatment confusion is cer-
been applied safely and effectively in patients with recent myocar- tainly more common and cognition typically declines as treatment
dial infarction (Magid et al., 2005), cardiomyopathy (Adabag et al., progresses, more for bilateral than unilateral ECT (O’Connor et al.,
2008), aortic aneurysm (Mueller et al., 2009), pulmonary embolism 2010c), but recovery is usually rapid (Russ et al., 1990; Rubin et al.,
(Suzuki et al., 2008), and cerebral aneurysms and angiomas (Kang 1993). Published reports may not be fully representative of clinical
and Passmore, 2004). experience, however, since many severely depressed patients cannot
Examples of the strategies required to minimize risk in medi- participate in cognitive testing or consent to research. The patients
cally compromised patients are described by Tess and Smetana who are most vulnerable to cognitive side effects might therefore be
(2009). These include pretreatment with atropine or glycopyrrolate excluded from study (Gardner and O’Connor, 2008).
to prevent bradycardia (though atropine may worsen post-ECT In the longer term, some cases of dementia must be expected
confusion); pretreatment with betablockers to limit hyperten- in older age groups because of the established links between
sion and tachycardia (though unopposed parasympathetic activ- depression, on the one hand, and incipient cerebrovascular and
ity can worsen bradycardia); reductions in anticoagulant therapy Alzheimer’s diseases, on the other. While rates look high at 14%
for patients with an intracranial mass or aneurysm; and additional in the patients aged 65+ years who were followed for 3 years by
muscle relaxation for those with fractures or marked osteoporosis. Godber et al. (1987) and 36% in those aged 75+ years followed for 5
A careful physical examination, ECG, measures of renal function, years by Brodaty et al. (2000), there is no reason to believe that ECT
and detailed discussions with medical and anaesthetic colleagues plays a causative role.
are vital. Other laboratory and imaging results add little extra Clinical guidelines highlight the need for cognitive assessment
information (Lafferty et al., 2001). Implanted cardiac pacemakers before, during, and after an ECT course but leave the choice of tests
present no special risk (Dolenc et al., 2004). to the clinician (American Psychiatric Association, 2001; Royal
College of Psychiatrists, 2005). Porter et al. (2008) proposed using
Cognition
the MMSE as a baseline measure and checking reorientation in the
Objective changes recovery room after every treatment. Later, they recommend a brief
No evidence has emerged from numerous anatomical and imaging test of verbal learning after every third treatment together with an
studies that ECT causes altered brain structure, neuronal death, or abbreviated autobiographical memory questionnaire after every
cerebral atrophy (Devanand et al., 1994; Ende et al., 2000). There sixth treatment. All these tests can be administered by a trained
is no doubt, however, that ECT can result in objective cogni- ECT nurse coordinator.
tive disruption. Patients are sometimes confused and disoriented Cognitive impairment is prevented or minimized by admin-
on wakening from anaesthesia. This typically lasts just for a few istering unilateral ECT in the first instance, titrating stimulus
intensity, prescribing only two treatments each week, and limiting Clinicians will naturally advocate for ECT when its use is war-
the number of treatments in a course, subject to clinical progress ranted. There is a fine line, however, between persuasion and
(Prudic, 2008). coercion. Patients’ apprehension typically settles with appropri-
ate reassurance and factual responses to questions, but doctors
Subjective reports
and patients sometimes perceive this process differently (McCall,
While complaints of poor memory subside as depression remits 2006). In a review of 17 reports by Rose et al. (2003), only half
(McCall et al., 1995; Coleman et al., 1996), between 29% and the respondents believed they had been given an adequate expla-
88% of patients describe some persistent forgetfulness in coming nation of ECT and most had limited knowledge of actual proce-
weeks and months (Brodaty et al., 2001; Rose et al., 2003; Philpot dures. Patients sometimes recounted giving ‘consent’ while heavily
et al., 2004). Gaps in memory of the 6 months before treatment sedated or threatened by legal compulsion. These recollections
and the 2 months afterwards are common and mostly well toler- might sometimes be inaccurate, but small numbers of patients are
ated (Freeman et al., 1980). Of more concern to patients are ‘holes’ left traumatized as a result. An account by Johnstone (1999) of the
in past memories and difficulties with remembering faces, names, perceptions of shame and despair by 20 aggrieved ECT recipients
and lists (Freeman et al., 1980). Risk factors for subjective deficits is required reading.
include bilateral electrode placement and high electrical charge Some of the patients most likely to benefit from ECT are unable
(Squire and Slater, 1983; Coleman et al., 1996). to consent by virtue of marked depression or psychosis. Pointers to
Mood plays a role too. Patients whose depression persists despite lack of capacity include: an inability or refusal to speak; a rejection
treatment tend to report greater memory difficulties and to per- of general nursing and medical care; rapid forgetting of basic infor-
ceive ECT more negatively (Freeman and Kendall, 1980; Coleman mation about the nature and consequences of ECT; and marked
et al., 1996). This is not surprising: mood correlates strongly with ambivalence. Treatment can be applied in most jurisdictions on an
self-rated memory, even in community populations (O’Connor involuntary basis with review by a court or tribunal. It is important
et al., 1990). to document the reasons for mandating ECT; the pros and cons of
Self-perceived cognitive changes, when assessed using standard alternative treatments; patients’ mental state and capacity to make
questionnaires, are only weakly associated with scores on objec- informed decisions; and family viewpoints. Except in emergencies,
tive tests, perhaps because the questionnaires fail to tap patients’ second opinions are encouraged. The limited research conducted to
personal memories. In fact, responses to a single, general question date suggests that nonconsenting patients often rate their outcome
about memory changes correlate quite well with objective perform- as satisfactory (Wheeldon et al., 1999). Patients who accept ECT
ance, suggesting that complaints have real validity (Brakemeier passively but cannot contribute to the decision-making process are
et al., 2011). best treated involuntarily to ensure legal oversight and protection
Previously, psychiatrists tended to blame complaints on patients’ (Law-Min and Stephens, 2006).
ongoing psychopathology, their loss of self-efficacy, or excessive ECT typically entails a series of treatments over a period of weeks.
concern with normal cognitive changes (Prudic et al., 2000). Others Mental capacity will often improve as depression or psychosis
were more circumspect. Freeman et al. (1980) noted that ECT abate, with the result that noncompetent patients become compe-
‘complainers’ were only fractionally more depressed on mood rat- tent and can participate in the decision to continue with treatment.
ing scales than ‘noncomplainers’. Following from this, and from the Conversely, a patient made confused by ECT may lose competence
observation that the temporal gradient in complaints fitted better for a period of time. Ethical practice requires a constant monitoring
with ECT than depression as causative agent, Sackeim (2000) con- of patients’ mental states, attitudes to ECT, and decisional capacity.
cluded that ‘attributing these subjective deficits to ongoing psycho- Repeated explanations of the reasons for administering ECT and
pathology or natural disease progression would seem disingenuous its likely benefits and side effects may be required if patients fail to
and defensive’. An account by one patient is especially compelling. recall initial discussions.
Donahue (2000) wrote that ECT (initially unilateral and then bilat-
eral) saved her from suicide. She felt grateful to be alive and would
have ECT again if necessary, but whole chunks of her life had been Attitudes to ECT
lost—a price she accepted but could not deny. Patients’ views
Fear of ECT is widespread. In a survey of 56 Australian hospital
Ethical issues visitors, none of whom had received ECT, many agreed that it
For patients with the capacity to consent, ECT should be adminis- ablated memory and rendered patients zombies (Kerr et al., 1982).
tered only with their agreement. Consent must be voluntary, based Psychiatrists attribute this fear to misinformation spread by popu-
on an adequate explanation of the nature of ECT, its practical impli- lar movies and certain lobby groups, but, in reality, a dread of elec-
cations and side effects, tailored to individual circumstances. Patients tricity is instilled in children from an early age.
bothered by cardiac symptoms, for example, need more detailed Malcolm (1989) detailed the concerns expressed by 100 patients
information about cardiovascular sequelae. Verbal discussion is best prior to ECT. Their fears included brain damage, memory loss,
supplemented by an informational brochure that patients and fami- pain, being seen by strangers while unconscious, having a heart
lies can digest at leisure, modelled perhaps on the exemplary docu- attack, and developing epilepsy. Their knowledge of ECT was
ment prepared by the American Psychiatric Association (2001). rudimentary: only 16% knew that a convulsion was induced and
Moderately depressed, cognitively intact older patients absorb and many expected a single treatment. Older people were more stoical:
retain this sort of information quite adequately (Lapid et al., 2004). 52% denied fear compared with 30% of younger ones. When the
A visit to the ECT treatment suite to explain its procedures and same patients were asked later what aspects most perturbed them,
equipment can allay patients’ and relatives’ concerns. they nominated waiting for treatment, seeing equipment laid out,
hearing staff talk in the background, breathing into an anaesthetic these lapses detracted from patients’ recoveries was unclear, but the
mask, and waking up afterwards. authors concluded that ‘the wide variability in how ECT is con-
In a retrospective survey, only 6% of 70 British patients rated ducted undoubtedly raises public health concerns’.
ECT as much worse than a visit to the dentist (Benbow and In former times, training deficiencies were commonplace. Of 160
Crentsil, 2004). By contrast, 7% of Australian patients said that they junior doctors in England and Wales, 63% had watched an instruc-
dreaded it and another 12% found it frightening (Kerr et al., 1982). tional video but only 53% had been supervised by an experienced
Occasionally patients develop an overwhelming horror of ECT, psychiatrist when giving their first treatment and only 4% were
usually after lengthy or repeated courses (Fox, 1993). routinely supervised thereafter (Duffett and Lelliott, 1998). Similar
Not all patients are negatively disposed. A psychiatrist who findings emerged in New South Wales, Australia, where 20% of
received ECT himself described a lifting of mood after the first doctors were not supervised in their first session by another medi-
treatment. Side effects—nausea, stiffness of the jaw, topographic cal practitioner (Halliday and Johnson, 1995).
disorientation, and mild memory loss—were a small price to pay
Clinical guidelines
in his view (A Practising Psychiatrist, 1965). Another patient, while
grateful for treatment, described the stigma and inconvenience of Three national professional bodies have since published detailed
maintenance ECT (Hensley, 2008). accounts of ECT’s indications and contraindications; proper
Patients’ views are coloured in part by their mental status. assessment and treatment procedures; legal and ethical issues;
Depression brings with it fears of therapeutic nihilism. As these and training and supervision requirements (American Psychiatric
fears remit with treatment, opinions of ECT may become more Association, 2001; Royal College of Psychiatrists, 2005; Royal
positive (Chakrabarti et al., 2010). Favourable views post-ECT Australian and New Zealand College of Psychiatrists, 2007). The
correlate with severity of mental disorder and disability pre-ECT, American Psychiatric Association stipulates that trainee psychia-
fewer treatment side effects, and a perception of good care while in trists should complete at least 10 treatments under direct supervi-
hospital, but not with legal status (Wheeldon et al., 1999; Philpot sion and then attend regular ECT review meetings. ‘Privileging’ to
et al., 2004; Rosenquist et al., 2006). This suggests that the sickest administer ECT independently requires objective evidence of safe,
patients, who stand to derive the greatest benefit from ECT, come proficient practice that conforms to local policy. Further education
to perceive it most favourably. might include attendance at an ECT course, a structured clinical
In an intriguing meta-analysis of 16 published reports, positive practicum, supervised reading, and participation in quality assur-
views of ECT emerged most commonly from studies conducted ance activities to identify gaps in performance and take corrective
shortly after ECT by psychiatrists using brief checklists. Studies action.
conducted by independent investigators some time after admission A detailed checklist prepared by the Royal College of Psychiatrists’
using open-ended questions were much less encouraging (Rose Research Unit (2006) of basic, standard, and ideal policies and
et al., 2003). This suggests that structured assessments conducted procedures covering all aspects of physical facilities, staff training,
by doctors in medical environments may fail to tap patients’ true patient assessment and treatment, consent, and follow-up provides
perceptions. a template for self-appraisal in any ECT service and is highly rec-
Proposed strategies to improve acceptance of ECT include edu- ommended. A suitable curriculum for psychiatry trainees has been
cational videos, use of ECT-experienced volunteers, minimizing proposed by Dolenc and Philbrick (2007), covering all important
time in the waiting room, and reducing exposure to technical par- areas of theory and clinical application.
aphernalia (Westreich et al., 1995; Koopowitz et al., 2003; Parvin
et al., 2004). The video deployed in one study made no discern- Transcranial Magnetic Stimulation
able difference to patients’ knowledge, perhaps because they were
too depressed to benefit, but it was greatly appreciated by families, Repetitive transcranial magnetic stimulation (TMS) involves using
whose support is often critical (Westreich et al., 1995). a pulsed magnetic field to induce local electric current in the cere-
bral cortex, targeting neural circuits implicated in mood regulation.
Families’ views The brief magnetic pulses are comparable to those used in mag-
Little research has been conducted of family members’ attitudes netic resonance imaging and are administered in rapid succession
to ECT, both before and after treatment. This is a deficiency, since to achieve changes in neurotransmission that persist beyond the
an informed, supportive family can do much to support patients stimulus period (Kim et al., 2009; George and Aston-Jones, 2010).
through their illness and treatment. In the meantime, two compel- During the procedure, patients sit in a specially designed chair with
ling accounts have been published of relatives’ battles with sceptical a coil positioned over their scalp. Several brief pulses are admin-
psychiatrists to secure ECT for a father and son respectively whose istered to determine the minimum amount of power required to
mental disorders had failed to respond to medications and psycho- produce hand twitching (the ‘motor threshold’), thereby allowing
therapy. Their efforts were rewarded eventually by excellent clinical individualized treatment dosing. The coil is then positioned over the
outcomes (D’Agostino, 1975; A Grateful Parent, 2005). left prefrontal cortex for therapeutic stimulation. Treatments typi-
cally take 20–40 min on 5 days per week over 4–6 weeks depending
Practice standards and training on clinical progress. As no sedation is required, normal activities
Variations in practice can be resumed immediately afterwards (George and Aston-Jones,
In New York state, Prudic et al. (2001) were perturbed to find that 2010).
17% of 59 mental health facilities administered the same electrical TMS is contraindicated in patients with nonremovable metal
charge to all patients; 8% of facilities failed to monitor seizure dura- implants in their heads, or within 30 cm of the coil, as heating or
tion; and 20% neglected to check patients’ cognitive status. Whether movement of these objects due to the magnetic field may result in
serious injury or death. Examples of prohibited devices include interface held close to the generator allows noninvasive setting and
aneurysm clips or coils, cardiac pacemakers or cardioverter defi- adjustment of stimulus parameters including temporary and per-
brillators, and metallic eye implants. TMS is also contraindicated in manent stimulus deactivation (Holtzheimer and Mayberg, 2011).
patients with a history of seizures or with potentially epileptogenic The mechanisms of action of DBS are more complex than sim-
frontal cortical ischaemia. Subcortical ischaemia is not a contrain- ply producing a ‘reversible lesion’ (as an alternative to ablative neu-
dication. Its safety in patients with dementia is not established rosurgery) in stimulated grey matter. Neuronal effects of DBS are
(Jorge and Robinson, 2011). both inhibitory and excitatory in nature, depending on the location
TMS has proven effectiveness as an antidepressant. In one of two and mix of cell bodies and white matter fibres in the stimulated
large, multicentre trials, O’Reardon et al. (2007) compared active field, as well as the stimulus parameters themselves (Holtzheimer
versus sham TMS in patients with medication-resistant major and Mayberg, 2011). At a systems level, DBS is presumed to act on
depression. Active TMS administered daily for a 4- to 6-week period cortico-striatal-thalamic-cortical (CSTC) neural circuits implicated
to the left prefrontal cortex was twice as likely as sham TMS to in psychiatric conditions such as obsessive-compulsive disorder
induce remission at 6 weeks. Only 4.5% of patients withdrew from and depression (Ward et al., 2010).
treatment because of adverse events. In the second trial, George Studies of DBS for psychiatric disorders are limited in size, but
et al. (2010) reported that remission was significantly more likely the results look promising. Lakhan and Callaway (2010) recently
(odds ratio, 4.2) following at least 3 weeks of active high-intensity analysed 16 trials reporting on the efficacy of DBS in treating
TMS, in contrast to sham TMS, delivered to the left prefrontal cor- obsessive-compulsive disorder, treatment-resistant depression,
tex. A significant interaction was found between clinical benefit and or both. The largest of these for obsessive-compulsive disorder,
the degree of antidepressant treatment resistance. However, overall examining subthalamic nucleus stimulation in 16 patients, showed
6-week remission rates with active TMS were modest in both tri- significantly lower mean symptom scores after active compared
als: 17.4% in the former and 14.1% in the latter, using the 24-item to sham treatment (19 versus 28, using the Yale-Brown Obsessive
HDRS. Age did not significantly predict treatment response in Compulsive Scale) and significantly higher mean global functioning
either study. scores (56 versus 43, using the Global Assessment of Functioning
In a trial of TMS for the treatment of vascular depression in older Scale) (Mallet et al., 2008). The largest study of DBS for depres-
adults, active treatment proved superior to sham (Jorge et al., 2008), sion, involving 20 patients who received subcallosal cingulate gyrus
with higher doses resulting in better 3-week response (39.4%) and stimulation, showed 60% response and 35% remission rates, using
remission (27.3%) rates, using the 17-item HDRS. Response rates the 17-item HDRS (Lozano et al., 2008). Other disorders of inter-
were lower, though, with older age and smaller frontal grey mat- est include addiction and Tourette’s syndrome (Holtzheimer and
ter volumes, possibly because of ischaemic damage to white matter Mayberg, 2011).
pathways connecting the left dorsolateral prefrontal and anterior DBS is invasive and adverse effects include seizure induction,
cingulate cortices. Responders and nonresponders did not dif- haemorrhage, superficial infection, and anaesthetic complications.
fer significantly in baseline neuropsychological performance and Stimulation itself may generate anxiety and fear. Unlike lesional
TMS itself did not give rise to cognitive deficits in this vulnerable psychosurgery, however, DBS is potentially reversible, producing
population. minimal tissue destruction and being easily adjusted or switched
Despite its cognitive-sparing properties, the time and expense off. Clarification of optimal stimulus targets, treatment profiles,
associated with TMS limits its utility as a first-line psychiatric treat- patient selection criteria, and long-term outcomes are future
ment and its role relative to ECT requires elucidation. Randomized research directions (Fitzgerald, 2011).
trials favour ECT over TMS in treating depression, showing sig-
nificantly higher response or remission rates for ECT (58.8% versus Conclusion
38.0%) (Rasmussen, 2008) and significantly lower Beck Depression
Inventory (BDI) suicide subscale scores (0.5 versus 1.2) (Keshtkar ◆ ECT is an effective, accepted treatment of depression. It
et al., 2011). works faster, and may be safer than pharmacotherapy for frail,
Suggested areas for future research include optimization of older patients but relapse rates are high once treatment stops.
stimulation frequency, intensity, and scheduling; stimulation of Continuing pharmacotherapy is mandatory. ECT works best in
cortical sites other than the left prefrontal cortex; combining treat- patients with psychomotor changes and psychotic symptoms.
ment with pharmacotherapy or psychotherapy; identification of the ◆ There is a place for ECT in medication-refractory cases of mania,
patients likely to benefit most; and the role of TMS in other condi- acute schizophrenia, and catatonia.
tions including anxiety disorders and psychosis (Kim et al., 2009;
◆ Electrical stimulation must always exceed seizure threshold, more
Jorge and Robinson, 2011).
for unilateral than bilateral ECT. Stimulus titration ensures that
energy charges are individualized. Alternative strategies include
Deep Brain Stimulation age-based formulae. Fixed dose, high energy treatments are not
acceptable.
Deep brain stimulation (DBS) has generated interest as a psy-
chiatric treatment modality following its success in treating ◆ Very old patients have higher rates of cardiovascular compli-
medication-resistant Parkinson’s disease. It involves implanting one cations, confusion, and falls. Twice-weekly ECT is adequate.
or more electrodes into specific brain regions using imaging-guided Benzodiazepines, anticonvulsants, and medications with anti-
stereotactic neurosurgery. The electrodes are then connected via cholinergic properties should be stopped if possible.
subcutaneous extension wires to an implanted pulse generator con- ◆ Cognition is often improved by ECT. Complaints of memory loss
taining a battery and stimulation computer. A handheld computer abate with time, but gaps in memory may persist.
◆ Capacity to consent to ECT can change as treatment progresses Bruce, B.B., Henry, M.E., and Greer, D.M. (2006). Ischemic stroke after
and explanations may need to be repeated. Educational brochures electroconvulsive therapy. Journal of ECT, 22, 150–2.
and videos are encouraged. Buchan, H., et al. (1992). Who benefits from electroconvulsive therapy?
Combined results of the Leicester and Northwick Park Trials. British
◆ ECT is perceived negatively by many patients and families and Journal of Psychiatry, 160, 355–59.
great efforts must be made to address their concerns accurately Burd, J. and Kettl, P. (1998). Incidence of asystole in electroconvulsive therapy
and to reduce the risk of poor outcomes. in elderly patients. American Journal of Psychiatry, 6, 203–11.
Burke, W.J., et al. (1987). The safety of ECT in geriatric psychiatry. Journal of
◆ Such a complex treatment should never be administered by
the American Geriatrics Society, 35, 516–21.
untrained staff. Carlyle, W., Killick, L., and Ancill, R. (1991). ECT: an effective treatment
◆ Few published reports focus just on frail, aged patients. Further in the screaming demented patient. Journal of the American Geriatrics
research is required to ensure that treatment of this vulnerable Society, 39, 637.
group is made as safe and effective as possible. Cattan, R.A., et al. (1990). Electroconvulsive therapy in octogenarians.
◆ TMS is better tolerated, but less effective, than ECT as a treat- Chakrabarti, S., Grover, S., and Rajagopal, R. (2010). Electroconvulsive therapy:
ment of major depression. a review of knowledge, experience and attitudes of patients concerning
the treatment. World Journal of Biological Psychiatry, 11, 525–37.
◆ DBS requires major surgery and its role is limited to the treatment
Chanpattana, W. and Andrade, C. (2006). ECT for treatment-resistant
of very severe, chronic, and disabling psychiatric disorders. schizophrenia: a response from the Far East to the UK NICE report.
Journal of ECT, 22, 4–12.
References Chanpattana, W., et al. (1999). Continuation ECT in treatment-resistant
Abraham, K.R. and Kulhara, P. (1987). The efficacy of electroconvulsive schizophrenia: a controlled study. Journal of ECT, 15, 178–92.
therapy in the treatment of schizophrenia: a comparative study. British Charlson, F., et al. (2012). ECT efficacy and treatment course: a systematic
Journal of Psychiatry, 151, 152–5. review and meta-analysis of twice vs thrice weekly schedules. Journal of
Abrams, R. (1997). The mortality rate with ECT. Convulsive Therapy, 13, ECT, 138, 1–8.
125–7. Chew, M.L., et al. (2008). Anticholinergic activity of 107 medications
Adabag, A.S., Kim, H.G., and Al Aloul, B. (2008). Arrhythmias associated commonly used by older adults. Journal of the American Geriatrics
with electroconvulsive therapy in hypertrophic cardiomyopathy. Pacing Society, 56, 1333–41.
and Clinical Electrophysiology, 31, 253–6. Coleman, E.A., et al. (1996). Subjective memory complaints prior to and
A Grateful Parent. (2005). A family member’s experience with ECT. Journal following electroconvulsive therapy. Biological Psychiatry, 39, 346–56.
of ECT, 21, 199. Cronholm, B. and Ottoson, J. (1960). Experimental studies of the therapeutic
Alexopoulos, G.S., et al. (1984). Medical problems of geriatric psychiatric action of electroconvulsive therapy in endogenous depression. Acta
patients and younger controls during electroconvulsive therapy. Journal Psychiatrica et Neurologica Scandinavica, 35, 69–102.
of the American Geriatrics Society, 32, 651–4. Currier, M.B., Murray, G.B., and Welch, C.C. (1992). Electroconvulsive therapy
American Psychiatric Association (2001). The practice of ECT: for post-stroke depressed geriatric patients. Journal of Neuropsychiatry
recommendations for treatment, training and privileging, 2nd edition. and Clinical Neurosciences, 4, 140–4.
American Psychiatric Association, Washington, DC. D’Agostino, A.M. (1975). Depression: schism in contemporary psychiatry.
A Practising Psychiatrist (1965). The experience of electro-convulsive American Journal of Psychiatry, 132, 629–32.
therapy. British Journal of Psychiatry, 111, 365–7. Daniel, W.F. and Crovitz, H.F. (1986). Disorientation during electroconvulsive
Bang, J.B., et al. (2008). ECT treatment for two cases of dementia-related therapy. Annals of the New York Academy of Sciences, 462, 293–306.
pathological yelling. Journal of ECT, 20, 379–80. De Carle, A.J. and Kohn, R. (2001). Risk factors for falling in a psychogeriatric
Barnes, R.C., et al. (1997). Maintenance electroconvulsive therapy and unit. International Journal of Geriatric Psychiatry, 16, 762–7.
cognitive function. British Journal of Psychiatry, 170, 285–7. Devanand, D.P., et al. (1994). Does ECT alter brain structure? American
Benbow, S.M. (1987). The use of electroconvulsive therapy in old age Journal of Psychiatry, 151, 957–70.
psychiatry. International Journal of Geriatric Psychiatry, 2, 25–30. Dolenc, T.J. and Philbrick, K.L. (2007). Achieving competency in
Benbow, S.M. (1991). Old age psychiatrists’ views on the use of ECT. electroconvulsive therapy: a model curriculum. Academic Psychiatry, 31,
International Journal of Geriatric Psychiatry, 6, 317–22. 65–7.
Benbow, S.M. and Crentsil, J. (2004). Subjective experience of electroconvulsive Dolenc, T.J. and Rasmussen, K.G. (2005). The safety of electroconvulsive
therapy. Psychiatric Bulletin, 28, 289–91. therapy and lithium in combination: a case series and review of the
Birkenhäger, T.K., et al. (2010). Influence of age on the efficacy of literature. Journal of ECT, 21, 165–70.
electroconvulsive therapy in major depression: a retrospective study. Dolenc, T.J., et al. (2004). Electroconvulsive therapy in patients with cardiac
Journal of Affective Disorders, 126, 257–61. pacemakers and implantable cardioverter defibrillators. Pacing and
Black, D.W., et al. (1988). The importance of Axis II in patients with major Clinical Electrophysiology, 27, 1257–63.
depression: a controlled study. Journal of Affective Disorders, 14, 115–22. Dombrovski, A.Y., et al. (2005). Predictors of remission after electroconvulsive
Bolwig, T. (2011). How does electroconvulsive therapy work? Canadian therapy in unipolar major depression. Journal of Clinical Psychiatry, 66,
Journal of Psychiatry, 56, 13–18. 1043–9.
Boylan, L.S., et al. (2000). Determinants of seizure threshold in ECT: Donahue, A.B. (2000). Electroconvulsive therapy and memory loss: a
benzodiazepine use, anesthetic dosage, and other factors. Journal of ECT, personal journey. Journal of ECT, 16, 133–43.
16, 3–18. Duffett, R. and Lelliott, P. (1998). Junior doctors’ training in the theory and
Brakemeier, E.L., et al. (2011). Self-evaluation of the cognitive effects of the practice of electroconvulsive therapy. Journal of ECT, 14, 127–30.
electroconvulsive therapy. Journal of ECT, 27, 59–66. Ende, G., et al. (2000). The hippocampus in patients treated with
Brodaty, H., et al. (2000). A prospective follow-up study of ECT outcome in electroconvulsive therapy. Archives of General Psychiatry, 57, 937–43.
older depressed patients. Journal of Affective Disorders, 60, 101–11. Figiel, G.S., et al. (1992). The treatment of late onset psychoses with
Brodaty, H., et al. (2001). ‘Side effects’ of ECT are mainly depressive phenomena electroconvulsive therapy. International Journal of Geriatric Psychiatry,
and are independent of age. Journal of Affective Disorders, 66, 237–45. 7, 183–9.
Fink, M, et al. (1996). Ambulatory electroconvulsive therapy: report of a task Jorge, R.E. and Robinson, R.G. (2011). Treatment of late-life depression: a
force of the Association for Convulsive Therapy. Convulsive Therapy, 12, role of non-invasive brain stimulation techniques. International Review
42–55. of Psychiatry, 23, 437–44.
Fitzgerald, P.B. (2011). The emerging use of brain stimulation treatments for Jorge, R.E., et al. (2008). Treatment of vascular depression using repetitive
psychiatric disorders. Australian and New Zealand Journal of Psychiatry, transcranial magnetic stimulation. Archives of General Psychiatry, 65,
45, 923–38. 268–76.
Flint, A.J. and Rifat, S.L. (1998). The treatment of psychotic depression in Kang, N. and Passmore, M.J. (2004). Successful ECT in a patient with an
later life: a comparison of pharmacotherapy and ECT. International orbital cavernous hemangioma. Journal of ECT, 20, 267–71.
Journal of Geriatric Psychiatry, 13, 23–8. Kant, R., Coffey, C.E., and Bogyi, A.M. (1999). Safety and efficacy of ECT in
Fox, H.A. (1993). Patients’ fear of and objection to electroconvulsive therapy. patients with head injury: a case series. Journal of Neuropsychiatry and
Hospital and Community Psychiatry, 44, 357–60. Clinical Neurosciences, 11, 32–7.
Freeman, C.P.L. and Kendall, R.E. (1980). ECTI: patients’ experiences and Karmacharya, R., England, M.L., and Öngur, D. (2008). Delirious mania:
attitudes. British Journal of Psychiatry, 137, 8–16. clinical features and treatment response. Journal of Affective Disorders,
Freeman, C.P.L., Weeks, D., and Kendall, R.E. (1980). ECT: II: patients who 109, 312–6.
complain. British Journal of Psychiatry, 137, 17–25. Kellner, C.H., et al. (2010). Bifrontal, bitemporal and right unilateral electrode
Gangadhar, B.N., et al. (2000). Cardiovascular response during ECT: a placement in ECT: randomised trial. British Journal of Psychiatry, 196,
cross-over study across stimulus conditions. Journal of ECT, 16, 177–82. 226–34.
Gardner, B.K. and O’Connor, D.W. (2008). A review of the cognitive effects of Kennedy, R., Mittal, D., and O’Jile, J. (2003). Electroconvulsive therapy in
electroconvulsive therapy in older adults. Journal of ECT, 24, 68–80. movement disorders: an update. Journal of Neuropsychiatry and Clinical
George, M.S. and Aston-Jones, G. (2010). Noninvasive techniques for probing Neuroscience, 15, 407–21.
neurocircuitry and treating illness: vagus nerve stimulation (VNS), Kerr, R.A., et al. (1982). ECT: misconceptions and attitudes. Australian and
transcranial magnetic stimulation (TMS) and transcranial direct current New Zealand Journal of Psychiatry, 16, 43–9.
stimulation (tDCS). Neuropsychopharmacology, 35, 301–16. Keshtkar, M., Ghanizadeh, A., and Firoozabadi, A. (2011). Repetitive
George, M.S., et al. (2010). Daily left prefrontal transcranial magnetic transcranial magnetic stimulation versus electroconvulsive therapy for
stimulation therapy for major depressive disorder: a sham-controlled the treatment of major depressive disorder, a randomized controlled
randomized trial. Archives of General Psychiatry, 67, 507–16. clinical trial. Journal of ECT, 27, 310–14.
Godber, C., et al. (1987). Depression in old age: prognosis after ECT. Kim, D.R., Pesiridou, A., and O’Reardon, J.P. (2009). Transcranial magnetic
International Journal of Geriatric Psychiatry, 2, 19–24. stimulation in the treatment of psychiatric disorders. Current Psychiatry
Gormley, N., et al. (1998). The safety and efficacy of electroconvulsive therapy Reports, 11, 447–52.
in patients over age 75. International Journal of Geriatric Psychiatry, 13, Koopowitz, L.R., et al. (2003). The subjective experience of patients who
871–4. received electroconvulsive therapy. Australian and New Zealand Journal
Grant, J.E. and Mohan, S.N. (2001). Treatment of agitation and aggression in of Psychiatry, 37, 49–54.
four demented patients using ECT. Journal of ECT, 17, 205–9. Kramer, B.A. (1999). ECT in elderly patients with schizophrenia. American
Grover, S., Mattoo, S.K., and Gupta, N. (2005). Theories on mechanism of Journal of Geriatric Psychiatry, 7, 171–4.
action of electroconvulsive therapy. German Journal of Psychiatry, 8, Kriss, A., et al. (1978). Neurological asymmetries immediately after
70–84. unilateral ECT. Journal of Neurology, Neurosurgery and Psychiatry, 41,
Halliday, G. and Johnson, G. (1995). Training to administer electroconvulsive 1135–44.
therapy: a survey of attitudes and experiences. Australian and New Lafferty, J.E., et al. (2001). Laboratory screening prior to ECT. Journal of ECT,
Zealand Journal of Psychiatry, 29, 133–8. 17, 158–65.
Haskett, R.F. and Loo, C. (2010). Adjunctive psychotropic medications Lakhan, S.E. and Callaway, E. (2010). Deep brain stimulation for
during electroconvulsive therapy in the treatment of depression, mania obsessive-compulsive disorder and treatment-resistant depression:
and schizophrenia. Journal of ECT, 26, 196–201. systematic review. BMC Research Notes, 3, 60.
Hensley, M.A. (2008). The meaning of electroconvulsive therapy: a patient’s Lapid, M.I., et al. (2004). Decisional capacity of depressed elderly to consent to
perspective. Journal of ECT, 24, 112–13. electroconvulsive therapy. Journal of Geriatric Psychiatry and Neurology,
Hickie, I., et al. (1996). Prediction of ECT response: validation of a refined 17, 42–46.
sign-based (CORE) system for defining melancholia. British Journal of Law-Min, R. and Stephens, J.P. (2006). Capacity, compliance and
Psychiatry, 169, 68–74. electroconvulsive therapy (ECT): the practice of ECT among consultant
Holmberg, S.K., Tariot, P.N., and Challapalli, R. (1996). Efficacy of ECT psychiatrists. Psychiatric Bulletin, 30, 13–5.
for agitation in dementia: a case report. American Journal of Geriatric Legendre, S.A., et al. (2003). The influence of cognitive reserve on memory
Psychiatry, 4, 330–4. following electroconvulsive therapy. Journal of Neuropsychiatry and
Holtzheimer, P.E. and Mayberg, H.S. (2011). Deep brain stimulation for Clinical Neurosciences, 15, 333–9.
psychiatric disorders. Annual Review of Neuroscience, 34, 289–307. Lisanby, S.H., et al. (2008). Toward individualized post- electroconvulsive
Hooten, W.M. and Rasmussen, K.G. (2008). Effects of general anesthetic therapy care: piloting the symptom-titrated, algorithm-based longitudinal
agents in adults receiving electroconvulsive therapy: a systematic review. ECT (STABLE) intervention. Journal of ECT, 24, 179–82.
Journal of ECT, 24, 208–23. Little, J.D., et al. (2002). Australian and US responses to electroconvulsive
Husain, M.M., et al. (2004). Speed of response and remission in major therapy dosage selection. Australian and New Zealand Journal of
depressive disorder with acute electroconvulsive therapy. Journal of Psychiatry, 36, 629–32.
Clinical Psychiatry, 65, 485–91. Loo, C.K., et al. (2008). A comparison of RUL ultrabrief pulse (0.3 ms) ECT and
Huuhka, M.J., et al. (2003). Cardiac arrhythmias induced by ECT in elderly standard RUL ECT. International Journal of Neuropsychopharmacology,
psychiatric patients: experience with 48-hour Holter monitoring. Journal 11, 883–90.
of ECT, 19, 22–5. Loo, C., Simpson, B., and MacPherson, R. (2010). Augmentation strategies in
Ingram, A., Saling, M.M., and Schweitzer, I. (2008). Cognitive side effects of electroconvulsive therapy. Journal of ECT, 26, 202–7.
brief pulse electroconvulsive therapy: a review. Journal of ECT, 24, 3–9. Loo, C., et al. (2011). Physical treatments for bipolar disorder: a review
Johnstone, L. (1999). Adverse psychological effects of ECT. Journal of Mental of electroconvulsive therapy, stereotactic surgery and other brain
Health, 8, 69–85. stimulation techniques. Journal of Affective Disorders, 132, 1–13.
Lozano, A.M., et al. (2008). Subcallosal cingulate gyrus deep brain stimulation Oudega, M.L., et al. (2011). White matter hyperintensities, medial lobe
for treatment-resistant depression. Biological Psychiatry, 64, 461–7. atrophy, cortical atrophy, and response to electroconvulsive therapy in
Magid, M., et al. (2005). Use of electroconvulsive therapy in a patient 10 days severely depressed elderly patients. Journal of Clinical Psychiatry, 72,
after myocardial infarction. Journal of ECT, 21, 182–5. 104–12.
Malcolm, K. (1989). Patients’ perceptions and knowledge of electroconvulsive Parvin, M.H., Swartz, C., and LaMontagne, B. (2004). Patient education by
therapy. Psychiatric Bulletin, 13, 161–5. electroconvulsive therapy-experienced volunteer. Journal of ECT, 20,
Mallet, L., et al. (2008). Subthalamic nucleus stimulation in severe 127–9.
obsessive-compulsive disorder. New England Journal of Medicine, 359, Penney, J.F., et al. (1990). Concurrent and close temporal administration of
2121–34. lithium and ECT. Convulsive Therapy, 6, 139–45.
Manly, D.T., Oakley, S.P., and Bloch, R.M. (2000). Electroconvulsive therapy Perera, T.D., et al. (2004). Seizure expression during electroconvulsive
in old-old patients. American Journal of Geriatric Psychiatry, 8, 232–6. therapy: relationships with clinical outcome and cognitive side effects.
McCall, W.V. (2006). Refusal versus reluctance. Journal of ECT, 22, 89–90. Neuropsychopharmacology, 29, 813–25.
McCall, W.V., Reid, S., and Ford, M. (1994). Electrocardiographic and Petrides, G., et al. (2001). ECT remission rates in psychotic versus
cardiovascular effects of subconvulsive stimulation during titrated right nonpsychotic depressed patients: a report from CORE. Journal of ECT,
unilateral ECT. Convulsive Therapy, 10, 25–33. 17, 244–53.
McCall, W.V., et alC. (1995). Comparison of the efficacy of titrated, Petrides, G., Tobias, K.G., and Kellner, C.H. (2011). Continuation and
moderate-dose and fixed, high-dose right unilateral ECT in elderly maintenance electroconvulsive therapy for mood disorders: review of
patients. American Journal of Geriatric Psychiatry, 3, 317–24. the literature. Neuropsychobiology, 64, 129–40.
McCormick, L.M., et al. (2007). Metabolic correlates of antidepressant and Pettinati, H.M., et al. (1990). Evidence for less improvement in patients taking
antipsychotic response in patients with psychotic depression undergoing benzodiazepines during unilateral ECT. American Journal of Psychiatry,
electroconvulsive therapy. Journal of ECT, 23, 265–73. 147, 1029–35.
McDonald, W.M. and Thompson, T.R. (2001). Treatment of mania in Philpot, M., et al. (2004). Eliciting users’ views of ECT in two mental health
dementia with electroconvulsive therapy. Psychopharmacology Bulletin, trusts with a user-designed questionnaire. Journal of Mental Health, 13,
35, 72–82. 403–13.
Medical Research Council (1965). Clinical trial of the treatment of depressive Plakiotis, C. and O’Connor, D.W. (2011). Ictal electroencephalographic
illness. British Medical Journal, 1, 881–6. measures in electroconvulsive therapy practice. Current Psychiatry
Mondimore, F.M., et al L. (1983). Post-ECT confusional states associated with Reviews, 7, 265–80.
elevated serum anticholinergic levels. American Journal of Psychiatry, Plakiotis, C., George, K., and O’Connor, D.W. (2012). Has electroconvulsive
140, 930–1. therapy utilisation remained stable over time? A decade of ECT service
Mueller, P.S., et al. (2009). Safety of electroconvulsive therapy in patients with provision in Victoria, Australia. Australian and New Zealand Journal of
unrepaired abdominal aortic aneurysm. Journal of ECT, 25, 165–8. Psychiatry, 46(4), 522–31.
Mulsant, B.H., et al. (1991). A prospective naturalistic study of Porter, R.J., Douglas, K., and Knight, R.G. (2008). Monitoring of cognitive
electroconvulsive therapy in late-life depression. Journal of Geriatric effects during a course of electroconvulsive therapy: recommendations
Psychiatry and Neurology, 4, 3–13. for clinical practice. Journal of ECT, 24, 25–34.
Munk-Olsen, T., et al. (2006). Electroconvulsive therapy: predictors and Price, T.R.P. and McAllister, T.W. (1989). Safety and efficacy of ECT in
trends in utilization from 1976 to 2000. Journal of ECT, 22, 127–32. depressed patients with dementia: a review of clinical experience.
National Institute for Health and Clinical Excellence. (2003). Guidance on the Convulsive Therapy, 5, 61–74.
use of electroconvulsive therapy. NICE, London. Prudic, J. (2008). Strategies to minimise cognitive side effects with ECT:
Nelson, J.P. and Rosenberg, D.R. (1991). ECT treatment of demented elderly aspects of ECT technique. Journal of ECT, 24, 46–51.
patients with major depression: a retrospective study of efficacy and Prudic, J., et al. (1994). Acute cognitive effects of subconvulsive electrical
safety. Convulsive Therapy, 7, 157–65. stimulation. Convulsive Therapy, 10, 4–24.
Nobler, M.S., et al. (1993). EEG manifestations during ECT: effects of electrode Prudic, J., Peyser, S., and Sackeim, H.A. (2000). Subjective memory complaints:
placement and stimulus intensity. Biological Psychiatry, 34, 321–30. a review of patient self-assessment of memory after electroconvulsive
O’Connor, D.W., Pollitt, P.A., and Roth, M. (1990). Coexisting depression therapy. Journal of ECT, 16, 121–32.
and dementia in a community survey of the elderly. International Prudic, J., Olfson, M., and Sackeim, H.A. (2001). Electro-convulsive therapy
Psychogeriatrics, 2, 45–53. practices in the community. Psychological Medicine, 31, 929–34.
O’Connor, D.W., et al. (2010a). The effectiveness of continuation-maintenance Prudic, J., et al. (2004). Effectiveness of electroconvulsive therapy in
ECT in reducing depressed older patients’ hospital re-admissions. Journal community settings. Biological Psychiatry, 55, 301–12.
of Affective Disorders, 120, 62–6. Rami, L., et al. (2004). Cognitive status of psychiatric patients under
O’Connor, D.W., et al. (2010b). Continuation-maintenance electroconvulsive maintenance electroconvulsive therapy: a one-year longitudinal study.
therapy: quality standards in 3 Australian aged psychiatry services. Journal of Neuropsychiatry and Clinical Neurosciences, 16, 465–71.
Journal of ECT, 26, 95–7. Ranen, N.G., Peyser, C.E., and Folstein, S.E. (1994). ECT as a treatment
O’Connor, D.W., et al. (2010c). Cognition in elderly patients receiving of depression in Huntington’s disease. Journal of Neuropsychiatry and
unilateral and bilateral electroconvulsive therapy: a prospective, Clinical Neurosciences, 6, 154–9.
naturalistic comparison. Journal of Affective Disorders, 124, 235–40. Rao, V. and Lyketsos, C.G. (2000). The benefits and risks of ECT for patients
O’Connor, M.K., et al. (2001). The influence of age on the response of major with primary dementia who also suffer from depression. International
depression to electroconvulsive therapy. American Journal of Geriatric Journal of Geriatric Psychiatry, 15, 729–35.
Psychiatry, 9, 382–90. Rasmussen, K.G. (2008). Electroconvulsive therapy versus transcranial
O’Leary, D., et al. (1994). The effectiveness of real versus simulated magnetic stimulation for major depression: a review with
electroconvulsive therapy in depressed elderly patients. International recommendations for future research. Acta Neuropsychiatrica, 20,
Journal of Geriatric Psychiatry, 9, 567–71. 291–4.
Olfson, M., et al. (1998). Use of ECT for the inpatient treatment of recurrent Rasmussen, K.G., et al. (2003). Electroconvulsive therapy for patients with
major depression. American Journal of Psychiatry, 155, 22–9. major depression and probable Lewy body dementia. Journal of ECT,
O’Reardon, J.P., et al. (2007). Efficacy and safety of transcranial magnetic 19, 103–9.
stimulation in the acute treatment of major depression: a multisite Rose, D., et al. (2003). Patients’ perspectives on electroconvulsive therapy:
randomized controlled trial. Biological Psychiatry, 62, 1208–16. systematic review. British Medical Journal, 326, 1363.
Rosenquist, P.B., et al V. (2006). What predicts patients’ expressed likelihood Suzuki, K., et al. (2008). Safety of electroconvulsive therapy in psychiatric
of choosing electroconvulsive therapy as a future treatment option. patients shortly after the occurrence of pulmonary embolism. Journal of
Journal of ECT, 22, 33–7. ECT, 24, 286–8.
Royal Australian and New Zealand College of Psychiatrists (2007). Tess, A.V. and Smetana, G.W. (2009). Medical evaluation of patients
Electroconvulsive therapy: clinical memorandum 12. Royal Australian and undergoing electroconvulsive therapy. New England Journal of Medicine,
New Zealand College of Psychiatrists, Melbourne. 360, 1437–44.
Royal College of Psychiatrists (2005). The ECT handbook, 2nd edition, ed. Tew, J.D., et al. (1999). Acute efficacy of ECT in the treatment of major
A.I.F. Scott. Royal College of Psychiatrists, London. depression in the old-old. American Journal of Psychiatry, 156, 1865–70.
Royal College of Psychiatrists’ Research Unit (2006). The ECT Accreditation Tiller, J.W.G. and Ingram, N. (2006). Seizure threshold determination for
Service (ECTAS): standards for the administration of ECT. Royal College electroconvulsive therapy: stimulus dose titration versus age-based
of Psychiatrists, London. estimations. Australian and New Zealand Journal of Psychiatry, 40,
Rubin, E.H., et al. (1993). The nature and time course of cognitive side effects 188–92.
during electroconvulsive therapy in the elderly. Journal of Geriatric Tomac, T.A., et al. (1997). Safety and efficacy of electroconvulsive therapy in
Psychiatry and Neurology, 6, 78–83. patients over age 85. American Journal of Geriatric Psychiatry, 5, 126–30.
Russ, M.J., et al. (1990). Cognitive effects of ECT in the elderly: preliminary Troller, J.N. and Sachdev, P.S. (1999). Electroconvulsive treatment of
findings. International Journal of Geriatric Psychiatry, 5, 115–18. neuroleptic malignant syndrome: a review and report of cases. Australian
Sackeim, H.A. (2000). Memory and ECT: from polarisation to reconciliation. and New Zealand Journal of Psychiatry, 33, 650–9.
Journal of ECT, 16, 87–96. UK ECT Review Group. (2003). Efficacy and safety of electroconvulsive
Sackeim, H., et al. (1987). Seizure threshold in electroconvulsive therapy: therapy in depressive disorders: a systematic review and meta-analysis.
effects of sex, age, electrode placement, and number of treatments. Lancet, 361, 799–808.
Archives of General Psychiatry, 44, 355–60. Van Waarde, J.A., Stolker, J.J., and van der Mast, R.C. (2001). ECT in mental
Sackeim, H.A., et al. (1993). Effects of stimulus intensity and electrode retardation: a review. Journal of ECT, 17, 236–43.
placement on the efficacy and cognitive effects of electroconvulsive Van Waarde, J.A., et al. (2010a). Electroconvulsive therapy for catatonia:
therapy. New England Journal of Medicine, 328, 839–46. treatment characteristics and outcomes in 27 patients. Journal of ECT,
Sackeim, H.A., et al. (2000). A prospective, randomized, double-blind 26, 248–52.
comparison of bilateral and right unilateral electroconvulsive therapy Van Waarde, J.A. and van der Mast, R.C. (2010b). Seizure thresholds in elderly
at different stimulus intensities. Archives of General Psychiatry, 57, patients treated with electroconvulsive therapy for major depressive
425–34. disorder: a review. Current Psychiatry Reviews, 6, 184–90.
Sackeim, H.A., et al. (2001). Continuation pharmacotherapy in the prevention Vanelle, J.-M., et al. (1994). Maintenance ECT in intractable manic-depressive
of relapse following electroconvulsive therapy: a randomized controlled disorders. Convulsive Therapy, 10, 195–205.
trial. JAMA, 285, 1299–307. Versiani, M., Cheniaux, E., and Landeira-Fernandez, J. (2011). Efficacy and
Sackeim, H.A., et al. (2008). Effects of the pulse width and electrode placement safety of electroconvulsive therapy in the treatment of bipolar disorder: a
on the efficacy and cognitive effects of electroconvulsive therapy. Brain systematic review. Journal of ECT, 27, 153–64.
Stimulation, 1, 71–83. Ward, H.E., Hwynn, N., and Okun, M.S. (2010). Update on deep brain
Sartorius, A., Wolf, J., and Henn, F.A. (2005). Lithium and ECT—concurrent stimulation for neuropsychiatric disorders. Neurobiology of Disease, 38,
use still demands attention: three case reports. World Journal of Biological 346–53.
Psychiatry, 6, 121–4. Watts, B.V., et al. (2011). An examination of mortality and other adverse
Shapira, B., et al. (1998). Cost and benefit in the choice of ECT schedule: events related to electroconvulsive therapy using a national adverse event
twice versus three times weekly ECT. British Journal of Psychiatry, 172, report system. Journal of ECT, 27, 105–8.
44–8. Westreich, L., et al. (1995). Patient knowledge about electroconvulsive
Sobin, C., et al. (1995). Predictors of retrograde amnesia following ECT. therapy: effect of an informational video. Convulsive Therapy, 11, 32–7.
American Journal of Psychiatry, 152, 995–1001. Wheeldon, T.J., et al. (1999). The views and outcomes of consenting and
Squire, L.R. and Slater, P.C. (1983). Electroconvulsive therapy and complaints non-consenting patients receiving ECT. Psychological Medicine, 29, 221–3.
of memory dysfunction: a prospective three-year follow-up study. British Wilkins, K.M., Ostroff, R., and Tampi, R.R. (2008). Efficacy of electroconvulsive
Journal of Psychiatry, 142, 1–8. therapy in the treatment of nondepressed elderly patients: a review of the
Srzich, A. and Turbott, J. (2000). Nonconvulsive generalised status epilepticus literature. Journal of Geriatric Psychiatry and Neurology, 21, 3–11.
following electroconvulsive therapy. Australian and New Zealand Journal Wilkinson, A.M., Anderson, D.N., and Peters, S. (1993). Age and the effects
of Psychiatry, 34, 334–6. of ECT. International Journal of Geriatric Psychiatry, 8, 401–6.
Steffens, D.C., et al. (2001). Severity of subcortical gray matter hyperintensity Zervas, I.M., et al. (1993). Age-dependent effects of electroconvulsive therapy
predicts ECT response in geriatric depression. Journal of ECT, 17, on memory. Convulsive Therapy, 9, 39–42.
45–9. Zielinski, R.J., et al. (1993). Cardiovascular complications of ECT in
Sutor, B. and Rasmussen, K.G. (2008). Electroconvulsive therapy for agitation depressed patients with cardiac disease. American Journal of Psychiatry,
in Alzheimer’s disease: a case series. Journal of ECT, 24, 239–41. 150, 904–9.
CHAPTER 15
Person- and
relationship-centred
dementia care: past,
present, and future
John Keady and Mike Nolan
It has been recognized for some time that supporting people with Nolan et al., 2002a, 2002b, 2006). It begins with a brief considera-
dementia and their family carers represents the most significant tion of the nature of dementia and its history and prevalence before
challenge facing health and social care systems throughout the world charting the rise of person-centred and relational models. This is
over the next 50 years (Jennings, 1999). However, debates as to the followed by a more detailed consideration of the conceptual and
best model and care pathway to adopt continue to evolve, with vari- empirical bases for the ‘Senses Framework’, followed by a case study
ous approaches to the support of people with dementia and their that illustrates its practical application. The chapter concludes with
family carers being advocated. For instance, during the last 30 years a glance towards the future and proposes that relationship-centred
or so, services and professional attitudes have moved from a largely care could extend its triadic structure by adding a fourth domain,
deficit-based model focusing on the person-with-dementia through the informal, community-based social networks of people with
to a much more personalized, individualized, and self-deterministic dementia and their families as they live their lives ‘outside their
approach with an emphasis on the person-with-dementia (Kitwood, front door’; what we have termed ‘The Neighbourhood Space’. As
1997: 7), a momentum that, in England, has culminated in the pub- Keady et al. (2012) have recently argued, accounting for, respecting,
lication of the National Dementia Declaration (<www.dementiaac- and authenticating the relationships that people with dementia and
tion.org.uk>, accessed 14.02.2012) with its central aim to ‘change their families have with neighbours, the community at large, shops,
the experience of living with dementia in England for good’ (p. 2). friends, and where ‘life is lived everyday’ are important new dimen-
Recent times have also seen the emergence of relational and dynamic sions of the social world that have received relatively little attention
models that transcend the essentially individualistic approach of in the literature or in policy formation.
person-centred care (Post, 2001) and take a triadic perspective with
a focus on the relationships and interactions between people with Dementia: Demography and Definitions
dementia, their family carers, and formal service systems (Nolan
et al., 2002a, 2006). Such trends reflect both advances in thinking The NHS Confederation (2010: 5) included the following defini-
about the nature of the ‘dementia experience’ and the importance tion of dementia in a recent report on improving hospital care for
now accorded to the active engagement of users, carers, and profes- people with dementia:
sionals. This approach recognizes that each of these partners is an Dementia is a syndrome (a group of related symptoms) that is associ-
‘expert’ in their own right and that a complete understanding will ated with an ongoing decline of the brain and its abilities. These include
only emerge when these differing, but complementary, perspectives thinking, language, memory, understanding and judgement; the con-
are brought together (Nolan et al., 2007). sequences are that people will be less able to care for themselves.
This chapter presents an historical overview of these changes The most commonly occurring dementia in both older and
and then focuses on a particular approach to working with people younger people is Alzheimer’s disease (Alzheimer’s Society, 2007),
with dementia, family carers, and paid carers based on the concept although there are many different causes of dementia which vary in
of relationship-centred care (Tresolini and the Pew Fetzer Task their presentation and progression. The greatest risk for the acquisi-
Force, 1994), as captured in the ‘Senses Framework’ (Nolan, 1997; tion of dementia is increasing age, with one in five people aged over
80 having a form of dementia (Alzheimer’s Society, 2007). The NHS After Auguste Deter’s death in the asylum on 8 April 1906, it was
Confederation (2010) report also suggests that at present up to the staining techniques applied to her brain by Dr Alzheimer and
70% of acute hospital beds in the UK are occupied by older people his colleagues that identified the amyloid plaques and neurofibril-
(p. 6) and the Royal College of Psychiatrists (2005) previously high- lary tangles that, to this day, remain confirmatory biological mark-
lighted that a fifth of all hospital patients will have a dementia. ers for presence of the disease and gave rise to the description of
Worldwide, 36 million people live with dementia, with these the ‘clinical symptoms’ of ‘presenile dementia’ (Maurer and Maurer,
numbers projected to double every 20 years to 66 million by 2030 2003). The search to further understand the biological function-
and 115 million by 2050 (Alzheimer’s Disease International, 2009). ing of the brain, and the death/dysfunction of the responsible
Of perhaps more concern, the Alzheimer’s Disease International neurotransmitters, largely marked Alzheimer’s disease as ‘medical
(2011) report on the benefits of early diagnosis and intervention territory’ for much of the twentieth century with its diagnostic clas-
suggests that as many as 28 million of the world’s 36 million people sification, and that of ‘dementia’, written into international mental
with dementia have yet to receive a diagnosis, and therefore ‘do not health disease classifications (World Health Organization, 1992).
have access to treatment, information, and care’ (p. 6). In the UK, Arguably, the direct alignment between ‘Alzheimer’s disease’ and
it is estimated that there are currently around 700,000 people with ‘mental health/illness’ promoted a deficit-based model and further
dementia which is representative of one person in every 88—or exposed those living with the condition to the stigma of mental ill-
1.1%—of the entire UK population. The total number of people ness, as well as to the stigma of ageing. Indeed, closer inspection
with dementia in the UK is forecast to rise to over 900,000 by 2021 of the quotation earlier reveals the ‘malignant positioning’ (Sabat,
and over 1.7 million by 2051, an increase of 38% over the next 15 2001, 2002) by Dr Alzheimer of Auguste Deter as ‘a patient’ (rather
years and 154% over the next 45 years (Alzheimer’s Society, 2007). than as ‘a person’) and an inmate of the confining and austere envi-
Both the National Dementia Strategy for England (Department of ronment of the Frankfurt am Main insane asylum. In such a setting,
Health, 2009a) and the National Dementia Declaration (<www. opportunities for individual expression and self-determination
dementiaaction.org.uk>, accessed 14.02.2012) state that the finan- would have been limited, to say the least.
cial cost of dementia in the UK is around £20 billion a year, and Similarly, the subsequent focus on the ‘staining technique’ on her
rising, with two-thirds of people with dementia currently living deceased brain, rather than attempting to find meaning in Auguste
at home and one-third resident in a care home. Moreover, esti- Deter’s profoundly self-aware words of ‘I have lost myself ’, effectively
mates currently suggest that there are over 15,000 younger peo- closed the door on what is now central to dementia care practice,
ple (i.e. under the age of 65 years) living with dementia in the UK namely the use of life story work (Haight et al., 2003; Bruce and
(Alzheimer’s Society, 2007) with patchy access to specialist services Schweitzer, 2008; Kellett et al., 2010; McKeown et al., 2010). Auguste
and support. This is regrettable, as a narrative review by Roach et al. Deter’s life story did not emerge until 100 years after her death when
(2009) on the experience of living with young-onset dementia high- Page and Fletcher (2006) pieced together the fragments of her life
lighted the importance of dedicated service provision and found that they were able to recover from various sources. In the story that
that age, employability, current family composition, and presen- emerged, Mrs Deter was not simply ‘a patient’ whose brain gave life
tation of symptoms related to dementia combine to significantly to Alzheimer’s disease, but until the time of her death at the age of
raise the stress for individuals and their families, especially when 56, she remained a married woman who had had a daughter called
teenage children are present in the home. Such an intergenerational Thekla. As Page and Fletcher (2006) reveal, Auguste Deter married
impact of the lived experience of dementia is a timely reminder Karl, a railway clerk, when she was 23 years of age and it was at
that ‘dementia’ extends beyond the traditional ‘dyad’, however that this time in the young couple’s life that they moved to Frankfurt to
relationship is defined, and reaches out to other members of the continue their life from within the protestant religion. It was also
family (Egset and Myklebust, 2011), and, as we will suggest, the reported that after his wife’s admission to the asylum in 1901, Karl
wider community. Deter struggled to pay the fees for his wife’s care and he was seen to
More overt recognition that dementia is not confined to later life is visit her ‘frequently’. Rather movingly, Mrs Deter is described in the
also a timely reminder of the history of the condition and the work of admission notes as being a ‘tall woman with long brown hair, brown
the German psychiatrist and neuropathologist Dr Alois Alzheimer, eyes and elegantly long fingers’ (Page and Fletcher, 2006: 578) and,
from whom the disease takes its name. In November 1901, at the age as far as it can be ascertained from the available writings, she was a
of 51, Auguste Deter was admitted to the Frankfurt am Main insane woman who had a preserved sense of identity—in the eyes of her
asylum at the behest of her husband. The family doctor’s admission husband and daughter—as a wife and as a mother.
note from 1901 survives and it shows that Mrs Deter’s condition We would like to think that Auguste Deter had a sense of belong-
required ‘treatment’ from the local asylum as she was experiencing: ing and significance for those who loved her, but defining people
weakening of the memory; persecution mania; sleeplessness; rest- with dementia from within their own relationships, life story, and
lessness; and an inability to perform any physical or mental work preserved attributes was not the dominant narrative of the twenti-
(Maurer and Maurer, 2003). Dr Alois Alzheimer himself performed eth century. This discourse was not to change until society began
various cognitive, psychological, and dexterity tests at the time of to question the morality and ethical standpoint of institutional and
Mrs Deter’s admission, which included an appraisal of her reading asylum-based care, and influential academics, practitioners, and
and writing performance. For example, Alzheimer recorded Mrs policy-makers began to challenge the long-held assumption that
Deter’s performance in the following way (Maurer et al., 1997): living with dementia equated simply to ‘coping with a living death’
… when she has to write Mrs Auguste Deter, she writes Mrs and we (Woods, 1989) in which ‘the self ’ had long since vanished (Cohen
must repeat the other words because she forgets them. The patient and Eisdorfer, 1986). As we will now explore, person-centred care
is not able to progress in writing and repeats, ‘I have lost myself ’. was central to the emergence of a new, more holistic vision of
(p. 1548) dementia and how to respond to it.
CHAPTER 15 person- and relationship-centred dementia care: past, present, and future 215
Person-Centred Care to reconceptualize the experience of dementia. After attempting to

agree a meaning to Rose’s actions and behaviours with colleagues
As noted in the previous section, during significant parts of the at the Bradford Dementia Group (see Kitwood, 1990a for a fuller
twentieth century dementia had a chequered history, with the discussion), they constructed a multidimensional theory that iden-
provision of care being largely institutionally based and task ori- tified a range of social and subjective factors that they believed
entated (Keady et al., 2009). Indeed, at the time, the metanarrative shaped Rose’s experiences. Whilst the emerging theory focused
of dementia was largely constructed around challenge, hopeless- predominantly on the more ‘advanced’ stages of dementia in resi-
ness, and the loss of self (for a review see Nolan, 2003), with little, dential care, mirroring Rose’s personal circumstances, it neverthe-
if any, effort made to provide supportive, evidence-based psycho- less placed the person with dementia at its heart. It was from these
social interventions to enable people living with the condition to observations that Kitwood (1988) reconceptualized the experience
reach (their) optimum quality of life. This therapeutic nihilism was of living with dementia along the following lines:
not to change until the introduction of reality orientation in the
early 1980s (Holden and Woods, 1982) and the construction of a SD = P + B+ H + NI + SP
more sympathetic, and empathetic, policy architecture that began
In this equation SD refers to senile dementia which is viewed as
to value people with dementia. Two reports published 1 year apart
the product of the complex interactions between the remaining five
proved particularly influential in challenging the prevailing pub-
elements of the equation:
lic and professional attitudes at the time, namely Organic Mental
Impairment in the Elderly (Royal College of Physicians, 1981) and P = Personality, which includes coping styles and defences
The Rising Tide (Health Advisory Service, 1982). For the first time, against anxiety;
these reports both signposted the importance and human worth
B = Biography, and responses to the vicissitudes of later life;
of people with dementia and highlighted the need for well-staffed
resources in both community and residential settings. In particu- H = Health status, including the acuity of the senses;
lar, The Rising Tide (Health Advisory Service, 1982) addressed the
NI = Neurological impairment, separated into its location, type,
needs of both people with dementia and their carers and set out
and intensity;
the key components of a comprehensive service that was to include
‘support, advice and relief at times of special difficulty to families’ SP = Social psychology which constitutes the fabric of everyday life.
(p. 17), with primary health and social services seen as the ‘essential
Kitwood (1988) suggested that the equation accounted for most
ingredients’ in providing a successful comprehensive service.
of the phenomenon associated with the range of dementias and
This commitment to improving the lives of people with dementia
explained the unique course of each person’s dementia by combin-
and carers was reinforced by the King’s Fund Centre (1984) in a
ing ‘structural’ and ‘conjunctural’ means of explanation. This theory
project paper that detailed the principles of good service practice.
reflects Kitwood’s (1990b,, 1997) critique of the past failings of care
This was achieved by outlining five ‘key principles’ that provided
environments and approaches to people with dementia which,
philosophical beliefs about personal empowerment for people
for Kitwood (1990b), created a ‘malignant social psychology’ that
with dementia. The five principles (King’s Fund Centre, 1984: 7–8,
inhibited the full expression and selfhood of people with dementia.
slightly abridged) called for an acknowledgment that:
Crucial to the emerging theory was the acceptance of the construct
1. People with dementia have the same human value as anyone else of ‘personhood’ and the recognition that a ‘malignant social psy-
irrespective of their degree of disability or dependence. chology’ had been developed that ‘bore down powerfully’ on those
2. People with dementia have the same varied human needs as any- with dementia (Kitwood and Bredin, 1992a). Kitwood (1990b)
one else. initially outlined 10 components that illustrated the elements of
this ‘malignant social psychology’ but later extended these to 17
3. People with dementia have the same rights as other citizens. by the time of the publication of the now seminal text Dementia
4. Every person with dementia is an individual. Reconsidered: The Person Comes First (Kitwood, 1997: 46–7).
5. People with dementia have the right to forms of support that do Examples of the factors that Kitwood believed created a ‘malignant
not exploit family and friends. social psychology’ include:
Whilst this was a most helpful start, the case for putting the ‘self ’ ◆ Infantilization: implying that a dementia sufferer has the mental-
back into the individual experience of living with dementia was ity or capability of a baby or young child
more eloquently articulated by Dr (later Professor) Tom Kitwood ◆ Stigmatization: turning a dementia sufferer into an alien, a dis-
and his colleagues at the Bradford Dementia Group. They devel- eased object, an outcast, especially through verbal labels
oped an elegant social theory of personhood in dementia that was,
in time, to influence UK public policy and have a global impact. ◆ Outpacing: the delivery of information or instruction at a rate far
True to the philosophy of a person-centred approach, this model beyond what can be processed
did not emerge from the consideration of data collected from sig- ◆ Objectification: treating a person like a lump of dead matter; to
nificant numbers of individuals but, as with Alzheimer’s work, was be measured, pushed around, drained, filled, and so on
built around a single case history, or psychobiography as it was orig-
inally named, undertaken by Kitwood himself (Kitwood, 1990a). ◆ Ignoring: carrying on (in conversation or action) in the presence
This psychobiography was conducted by Kitwood with a person of a person as if they were not there.
called ‘Rose’ whose struggle to assert her personality through the Kitwood and colleagues cogently argued that ‘the dementia’ is not
mask of her confusion triggered Kitwood’s thoughts on the need the main problem, rather it is ‘our’ (individual, carer, professional,
society) inability to accommodate ‘their’ view of the world. Kitwood by family carers (Department of Health, 2009a). Studies on the
and Bredin (1992a) suggested that this ‘them’ and ‘us’ divide creates experiences of family carers for people with dementia have a long
a dialectic tension that is reinforced over the years by the devalued and distinguished history, from early pioneering work conducted
status of someone who is labelled as ‘demented’ and who is conse- by Zarit and colleagues (1980, 1986) in the USA which continues to
quently seen to be cognitively incompetent (see also Sabat et al., the present day. However, much of this early work was undertaken
2011). Underpinning these observations is the belief that if the ele- in North America and focused almost exclusively on the ‘burden’
ments of a ‘malignant social psychology’ can be identified, appraised, of family care and the negative consequences that caring has on the
and overcome, then care (both for family and professional carers) health and wellbeing of family carers, a perspective that continues
can be improved and that the person with dementia will achieve to dominate the landscape of dementia care and practice.
a greater sense of personal ‘wellbeing’. This focus on wellbeing is As Kitwood was developing his new approach to dementia care,
reflected in Kitwood’s (1997) definition of personhood: a parallel, but separate, stream of work was being undertaken at
It is a standing or status that is bestowed upon one human being, by Bangor University, North Wales, that sought to provide a more
others, in the context of relationship and social being. It implies recog- rounded and holistic view of family care. The original focus of this
nition, respect and trust. Both the according of personhood, and the work was on carers of older people and parent carers of people with
failure to do so, have consequences that are empirically testable. (p. 8) learning disabilities, rather than carers of people with dementia per
To promote the practice of person-centred care the observational se. It clearly demonstrated that a view of caring that was predicated
method of Dementia Care Mapping (Kitwood, 1990b; Kitwood and on burden alone was inadequate and that, in reality, caring com-
Bredin, 1992b) was devised with the goal of not only enhancing care prises a complex mix of burdens and rewards and that it is the bal-
in formal settings, but also capturing it in a measurable way. Kitwood ance between the two that is the most important factor (Nolan and
and Bredin (1992a) argued that a positive change to the social envi- Grant, 1989; Nolan et al., 1990; Grant and Nolan, 1993). Several
ronment could cause a reversal of the accepted ‘decline’ trajectory in studies showed that carers’ subjective interpretations of their
dementia (i.e. from mild to moderate to severe stages) and named situation, and the nature and quality of their relationship with the
this ‘rementia’. Such a paradigm shift had implications for the caring person they were supporting, were most influential in determining
professions, as people with dementia were now seen to exert a sense their levels of stress rather than the objective circumstances of care,
of agency, an agency that could only be realized through significant such as the amount of direct practical and instrumental help they
changes in professional attitudes, practices, and cultures. provided (Grant et al., 1990; Nolan et al, 1990, 1994).
Following Professor Kitwood’s untimely death at the end of the This called attention to the dynamic nature of the relationship
1990s, the mantle of developing person-centred care was taken up between carers and cared-for persons that was consistent with the
by Professor Dawn Brooker who, at the time, was also working (at the time) recently published and now seminal work of Kahana
out of the Bradford Dementia Group in the UK. In a subsequent and Young (1990). These authors were ahead of their time in criti-
influential paper, Brooker (2004) developed the VIPS model of cizing the largely unidirectional and unidimensional models of care
person-centred care: that were predominant at the time, in which the cared for person
was seen only as a source of stress and burden on the carer, resulting
◆ Value of all human lives in universally negative consequences for the latter. Instead, Kahana
◆ Individualized approach recognizing uniqueness and Young (1990) called for the development of more dynamic and
relational models that acknowledged and explored the possibilities
◆ seeing the world from the Perspective of the service user
of both negative and positive consequences for both the carer and
◆ Social environment that promotes wellbeing. the cared for person. Such ‘dyadic’ models were, they argued, essen-
The VIPS model provides a helpful memory aide highlighting the tial if more appropriate and sensitive support was to be developed.
importance and value of people with dementia as well as reflecting In looking to the future, these authors suggested that the dyadic
a more holistic appreciation of the relationships that infuse daily model itself needed to expand to account for triadic interactions
life. that included the perspectives of service providers as well.
Whilst Kitwood’s work was promoting a new vision of the experi- This work appeared as a book chapter (a source of expertise and
ence of dementia from the perspective of people with dementia, a inspiration that is sadly all too often overlooked today with the
parallel stream of work was exploring the carer’s perspective. present hegemony of the ‘journal’), coinciding with the work being
undertaken at Bangor University, and stimulated further exploration
into the relational dynamics of caring. In a seemingly serendipitous
Dementia: Focusing on Family Carers manner at virtually the same time, John Rolland (1988, 1994), an
There can be no doubting the enormous impact that Kitwood’s work American psychologist, published his model of working with peo-
has had on the care of people with dementia, with the previously ple with disabilities, their family carers, and formal services, which
largely nihilistic approach being replaced by a much more positive he called the ‘therapeutic quadrangle’. Rolland’s interest was in the
and dynamic model centred on the personhood of the individual. experience of living with and supporting people with long-term
However, as noted above, Kitwood’s initial focus was largely on care conditions, which he argued are far more heterogeneous in their
provided in an institutional setting, and whilst the role of staff in presentation and effects than acute conditions. He postulated that in
creating a positive (or malignant) social psychology was acknowl- order to develop truly responsive interventions there was a need to
edged, the contribution of family carers was largely overlooked in look at the interactions between the person with a long-term condi-
the construction of person-centred care. This left a considerable gap tion, the family, and the professional system, all within the context
in our understanding, for then, as now, the majority of people with of living with a particular form of chronic condition, which vary
dementia live in the community, supported, as needed, primarily in several important dimensions, such as onset, course, incapacity,
Family career needed in the future. However, whilst this is a cognitive and specula-
tive activity, it can still have a profound effect on people’s lives, for
example in the case of an only daughter who may not move overseas
in anticipation that her ageing mother might need care in the future.
Preventive care
Professional Older person This typically comprises caring at a distance and involves subtle
efforts to ‘keep an eye’ on ageing parents, for example to see that
they are taking medication or that their diet remains good. This is
more overt and purposeful than anticipatory care, but is still never-
theless largely hidden from the older person.
Illness/disability Supervisory care

Fig. 15.1 The therapeutic quadrangle. (From Nolan et al. (1994) p. 24.) This type of care is now becoming increasingly obvious and com-
prises more direct action, such as shopping with or for older peo-
ple, directly ensuring they take medications, checking to see that
and outcome. Thus, he argued, a condition such as stroke, which their fridge is adequately stocked, and taking action if it isn’t.
usually has a sudden and often unexpected onset and causes vary-
ing degrees of incapacity but which subsequently follows a relatively Instrumental care
stable course (in the absence of another episode), makes very dif- This is the type of activity most usually defined as caring and
ferent demands than a condition such as dementia, with its usu- involves providing direct hands-on assistance usually with a range
ally insidious onset, highly variable and unpredictable course, and of activities of daily living or instrumental activities of daily living.
slowly increasing levels of incapacity. As such, it requires very differ- It is the primary type of ‘caring’ activity that is recognized as such
ing forms of support for all concerned in the process. by formal services, and often the only one with which a carer is
The work of Kahana and Young (1990) and Rolland (1988, 1994), likely to be offered help.
coinciding as it did with the newly emerging focus on the dynamics
of caring relationship in the UK, exerted a considerable influence Protective care
on the direction of future studies, with Nolan et al. (1994) adopt- This again comprises a subtle range of activities which daughters
ing the ‘therapeutic quadrangle’ as an organizing framework for (in the theoretical sample that predominantly comprised the study)
much of their on-going work (see Fig. 15.1). Subsequently, Keady undertook to ‘protect’ their mothers from knowledge of their
(1994a, 1994b; Nolan et al., 1995, 1996; Keady, 1996, 1997, 1999) increasing frailty and cognitive abilities.
took the lead in a number of studies with people with dementia
and their carers, and it is to this body of work that we now turn our Bowers’ (1987) data indicated that it is protective care that daughters
attention. see as the most stressful and the most important, and instrumental
care that they see as the least stressful and important. However, par-
Developing New Understandings of ‘Caring’ adoxically, services focused almost exclusively on the latter and did
and the Dementia Experience not usually even acknowledge the former. This was often a source
of conflict between family and formal carers.
Around the time that Rolland (1988, 1994) and Kahana and Young Whilst much taken with this model, at the time we argued that,
(1990) published their work, some exciting new insights in to based on our data collected over a number of years, anticipatory care
meaning of caring in dementia from the perspective of family car- was not confined to the early stages of caring (as Bowers contended)
ers were emerging. For example, Motenko (1989) highlighted the but occurred throughout the caring trajectory (Nolan et al., 1995,
potential for carers of people with dementia to experience both 1996). What changed was the nature of what was being anticipated.
gratifications and frustrations, reinforcing the earlier conclusions Moreover, we also suggested that whilst well-meaning, protective
of Hirschfield (1981, 1983) on the importance of mutuality in care could be potentially paternalistic and that the best caring rela-
caring relationships, or the extent to which carers find gratifica- tionships were characterized by reciprocal care in which both par-
tion and meaning in their role. Concurrently, Barbara Bowers, ties made a valued contribution. Another important element to the
an American nurse, was exploring the meanings that caregiving caring dynamic was reconstructive care in which carers helped the
daughters of people with dementia give to their experiences and person they were supporting to reconstruct a positive view of them-
developed a much more sophisticated and nuanced typology of selves. It was how such reconstructive and reciprocal care developed
caring (Bowers, 1987). This comprised of a number of types of over time that was the primary focus of Keady’s on-going work.
care, many of which were deliberately kept ‘invisible’ from the Building on a temporal model of the dementia caring experience
person with dementia and some of which occurred a long time as it unfolds over time (Wilson, 1989a, 1989b), Keady (Keady and
before actual hands-on instrumental care was needed. These are Nolan, 1994a, 1994b; Keady, 1996, 1997, 1999; Nolan et al., 1996;
now described. Grant et al., 2003) developed a six-stage model of the caring trajec-
tory that comprised the following phases:
Anticipatory care
This typically occurs long before any physical care is needed and
◆ Building on the past
involves carers anticipating what they would do if ever care was ◆ Recognizing the need
◆ Taking it on diagnosed and progressed over time, it was the carer who increas-
◆ Working through it ingly did most of the ‘work’, and Keady termed their efforts to keep
the person with dementia engaged as ‘maintaining involvement’
◆ Reaching the end (Keady, 1997).
◆ A new beginning. Since that time, several studies have further explored the dynam-
ics of caring relationships in dementia (see for example Brodaty
‘Building on the past’ acknowledges that caring relationships do et al., 2003; Selwood et al., 2007), and our understanding was mov-
not appear fully formed ‘out of the blue’ but rather are usually pre- ing closer to addressing Kahana and Young’s (1990) prophetic call
ceded by a long relationship and that the quality of this relationship for the development of much more intricate models of the ‘give and
is very influential on the subsequent caring dynamic. ‘Recognizing take’ of caring relationships as they unfold over time. Recent studies
the need’ is a period, either extended (as in dementia) or short (as in have also seen the emergence of ‘couplehood’ in dementia, captur-
stroke) when carers become aware that the nature of their relation- ing in more nuanced ways dementia as a shared experience.
ship with their loved ones is changing and that differences in roles
and responsibilities are becoming apparent. ‘Taking it on’ should be From Personhood to Couplehood
a time when informed decisions about whether or not to ‘become a
Probably the most extensive of these studies was that undertaken by
carer’ are made, based on as full an understanding of the potential
Hellström and colleagues in Sweden (Hellström et al., 2005a, 2005b,
consequences as possible. However, this rarely happens in practice
2007). This study involved following 20 couples (one of whom had
and carers often ‘take on’ their role uninformed and lacking the skills
dementia) over a 5-year period and, wherever possible, interview-
and knowledge that they need to ‘do caring well’ (Schumacher et al.,
ing both spouses several times. In total, over 150 interviews were
1998). ‘Working through it’ is often described as the ‘long haul’ of
completed. This study has, to the best of our knowledge, collected
caring and is the period during which, if they get any support at all,
the most extensive data of this type yet available and demonstrated
carers are most likely to be in receipt of formal services. Sadly, they
several important points. First, that it is possible to obtain rich
are rarely supported when they ‘reach the end’, through either the
and meaningful data from people with dementia over an extended
death of the person with dementia or placement in a care home. At
period of time, with 12 of the original 20 people with dementia still
such times, carers often receive no support at all. Of course for many
being interviewed 5 years later. Second, that for most of the spouses
carers this is not the end of their support for the person with demen-
the idea of ‘couplehood’ was far more meaningful than was the idea
tia as the family will often continue to play an important part in the
of ‘personhood’ in that both parties described themselves prima-
lives of the person with dementia. It is here where ‘A new begin-
rily in relation to each other rather than as separate individuals.
ning’ enters the life of (ex) family carers and different opportunities
Third, in the majority of relationships both spouses adopted a wide
open up, which may, or may not, include continuity and exposure
range of subtle tactics to ensure that they both experienced the best
to dementia, such as taking on a voluntary role with the Alzheimer’s
quality of life possible. Fourth, that until the very late stages, the
Society or taking part in one of the many dementia café initiatives
people with dementia were very active participants in ‘maintain-
that are in existence across the UK. However, as far as services are
ing involvement’ and often developed deliberate strategies to ‘hand
concerned, their role as a ‘carer’ has ended.
over’ responsibilities to their partner when they appreciated that
Keady originally developed this model following in-depth inter-
their own capacity was becoming limited.
views with carers of people with dementia (Keady and Nolan, 1994a,
In describing these interactions Hellström et al. (2007) identi-
1994b; Keady, 1999) and it was subsequently elaborated upon with
fied a wide range of processes, some of which reflected well-estab-
data from carers of physically frail older people and parent car-
lished patterns of relating, others of which had been developed and
ers of people with learning disability (Nolan et al., 1996). Whilst
evolved in response to the dementia experience. Taken together,
these studies tended to fit into the existing paradigm of treating
these processes created a ‘nurturative relational context’ (Hellström
people with dementia and carers separately, towards the end of his
et al., 2005a, 2005b) in which spouses sought to both ‘sustain cou-
interviews Keady was able to involve some people with dementia.
plehood’ and ‘maintain involvement’. In order to do so they focused
Consequently, a much more subtle picture began to emerge (Keady,
their attention primarily on the present rather than looking too
1999; Keady and Nolan, 2003) which allowed the ‘dynamics’ of
far into the future, but at the same time they often ‘took risks’ so
dementia to be better understood. Keady was able to describe a
that the person with dementia was not unduly constrained in their
delicate and often longstanding dialectic that took place between
actions. However, most of their focus was on the dynamics of their
people with (as yet undiagnosed) dementia and their carers in
relationship and involved:
which the former began to notice that something was wrong and
made efforts to hide this and the latter sooner or later ‘suspected’ ◆ ‘Talking things through’, so that each partner was aware of the
that something was wrong and made efforts to try to find out what thoughts and feelings of the other
(Keady and Nolan, 1995a, 1995b; Keady and Gilliard, 1999). This ◆ ‘Being appreciative and affectionate’, and complimenting each
involved ‘boxing and coxing’ in which both parties ‘worked’ hard other on their appearance and the efforts they made to keep their
but often towards different ends and sometimes to the detriment relationship ‘special’
of their relationship. In such circumstances some carers and their
partner tended to ‘work apart’ over time (Keady and Nolan, 2003). ◆ ‘Making the best of things’ by sharing enjoyable activities, seeing
On the other hand, when couples developed an open stance to the the positive things in life, and living for the moment
changing behaviour of one partner and were more proactive in dis- ◆ ‘Keeping the peace’, which involved being aware of, and seeking
cussing potential difficulties and addressing them positively, then to avoid, those activities that triggered a negative response in
they ‘worked together’ far more effectively. As the dementia was their partner.
As noted, this subtle range of activities were actively practised by care’. They described the importance of relationships in the fol-
both spouses, often over a prolonged period of time, and studies lowing way:
such as this attest not only to the importance of relationships in
understanding the dementia experience, but also to the continued relationships are critical to the care provided by nearly all practition-
ers and a sense of satisfaction and positive outcomes for patients and
agency of the person with dementia.
practitioners. Although relationships are a prerequisite to effective
Although it had taken over a decade, studies such as these were care and teaching, there has been little formal acknowledgement of
now beginning to realize the vision of Kahana and Young (1990). their importance and few formal efforts to help students and practi-
The insights provided by such knowledge have a number of impli- tioners learn to develop effective relationships in health care.
cations for the care and support given to people with dementia (Tresolini and Pew-Fetzer Task Force, 1994: 11)
and their family carers that we will go on to consider shortly.
However, before doing so, there is a need for an approach that In promoting a more holistic vision of healthcare, the Task Force
more fully realizes Rolland’s (1988, 1994) aspiration of a ‘thera- focused on several areas including:
peutic quadrangle’. ◆ the social, economic, environmental, cultural, and political con-
texts of care
Squaring the Quadrangle: ◆ the subjective experience of illness

The ‘Senses Framework’ and ◆ the reality that relationships develop between practitioners,
patients, families, and the wider community over time.
Relationship-Centred Care
They suggested that it was the interaction of these factors that should
Following Kitwood’s (1997) pioneering work, ‘person-centred care’
lie at the heart of a healthcare system based on relationship-centred
became a watch-word for good quality individualized care for peo-
care and that relationships form the ‘foundation’ of any therapeu-
ple with dementia and his ideas have been further developed by
tic or healing activity. Whilst they went on to outline the basic ele-
the more recent additions of Brooker (2004). However, the notion
ments of such a system, they also recognized that the concept of
of person-centred care has also become the policy ‘mantra’ of the
relationship-centred care was still emerging, and that further work
last decade and the ‘personalization’ agenda still dominates the lat-
was needed to put the concept into practice in ways that ensured an
est policy pronouncements (Social Care Institute for Excellence,
appropriate balance between the needs of everyone involved in health-
2010). Unfortunately, unlike the contribution of Kitwood and,
care relationships. The ‘Senses Framework’ (Nolan, 1997; Nolan et al.,
separately, Brooker, whose work has had an undoubted impact,
2001, 2002a, 2002b, 2006) provides one way of achieving this.
the reality of policy rarely matches the rhetoric and there remains
The ‘Senses Framework’ was developed in parallel to, but without
considerable confusion about what ‘person-centred’ care actually
awareness of, the emergence of relationship-centred care described
means (McCormack, 2004). Increasingly, at least in policy terms,
previously. Its roots lie in the early work of Nolan, Grant, and Keady
it has become associated with concepts such as independence and
(1996, 1998) on the caring relationship, and the influences exerted
autonomy, with a focus on ‘individuals’ and their circumstances
by the thinking of Kahana and Young (1990) and Rolland (1988,
rather than the nexus of relationships parting, which we all partici-
1994) have already been acknowledged. Further stimulus was pro-
pate in. This is a manifestation of the move towards the application
vided by a presentation given by Mulrooney at the World Congress
of consumerist principles to health care (McCormack, 2004) and
of Gerontology in Adelaide in 1997. Mulrooney (1997) promoted a
the emergence of concepts such as successful ageing, which lionize
vision of person- and relationship-centred care that was not based
independence and autonomy (Scheidt et al., 1999). This represents
on notions of autonomy and individuality but rather was predicated
a retrograde step and overlooks the complexity of the vast major-
on three core principles:
ity of real-life situations, especially in conditions such as dementia.
Ethicists such as Evans (1999) and MacDonald (2002) have become ◆ respecting personhood
increasing critical of the ‘politics of independence’, arguing that ◆ valuing interdependence
such a stance tends to negate the shared responsibility that we all
have to provide adequate support for the most frail and vulnerable ◆ investing in caregiving as a choice.
members of society. This model, whilst recognizing that each person is unique and
The tensions apparent in many modern-day health and welfare has individual needs and intrinsic worth, sees people as being
systems, with their goal of promoting independence, were acknowl- primarily interdependent, a value base that recognizes the reci-
edged a number of years ago by a major ‘task force’ established in procity that is inherent in the best of relationships and promotes
America to consider how to create a healthcare system fit to meet a balance between dependence and independence (Mulrooney,
the future needs of American society (Tresolini and Pew Fetzer 1997). However, in situations where substantial care and support
Task Force, 1994). Following a period of ‘intense national debate’, are required, personhood can only be respected and interdepend-
it was recognized that the current system did not meet the needs of ence valued if the person providing the care does so willingly. This
a population that was increasingly diverse ethnically and also suf- may not always be the case for family carers. Such considerations
fered primarily from long-term conditions. Such conditions pose also apply to those in paid caring roles, whether professional or
the greatest future challenge to healthcare, but the system in the not. Indeed, this belief was central to Kitwood’s (1997) vision of
US, possibly even more so than in the UK, is geared to address person-centred care and he argued that if paid carers are not valued
acute care needs. The task force concluded that there was a need and accorded status, then they will bring a feeling of being devalued
to develop an entirely new model of healthcare, predicated on a into their work, which will inevitably be reflected in the care they
different philosophy, and they termed this ‘relationship-centred provide for people with dementia.
An invitation to deliver a keynote address on the future role of accorded it mean that there is little to be gained by way of a sense
nurses in the delivery of health and social care for older people of significance. Small wonder then that standards of care have been
(Nolan, 1997) led to the emergence of the ‘Senses Framework’. The historically poor, as captured graphically in the titles of seminal
paper focused in particular on addressing the needs of frail older works such as The Last Refuge (Townsend, 1962) and Sans every-
people who require ongoing support. It argued that the evolution thing (Robb, 1967). Despite considerable improvements, scandals
of modern-day healthcare has put ‘cure’ centre stage and that car- still arise with uncomfortable regularity.
ing has become an essentially devalued activity. As a consequence, The success of Kitwood’s work and the enthusiasm with which it
those working in acute care have a clear sense of therapeutic direc- has been adopted is probably due in no small measure to the fact
tion and an explicit goal, cure, to aspire to. Moreover, such a goal is that it helped to create the ‘Senses’ for staff working with people
highly valued and the ‘excitement’ of working in an acute, hi-tech with dementia. After years of therapeutic nihilism they suddenly
environment is constantly reinforced in the media and exalted by had a sense of purpose (to realize person-centred care) and, with
the public. In the absence of cure, the fledging discipline of geriatric the advent of Dementia Care, Mapping a way of demonstrating that
medicine substituted rehabilitation, based on achieving a certain they had achieved this. As the popularity of this approach spread it
degree of functional independence, as its therapeutic aim (Wilkin helped to create a sense of belonging to an increasingly large group
and Hughes, 1986). However, those working in longer-term care of people delivering person-centred care and, as the value accorded
environments have always lacked a sense of therapeutic direction, to this grew, so too did the sense of significance and importance
hence the nihilism described by Kitwood (1997). The efforts of attached to this area of practice.
those working in long-term care have been consistently denigrated Whilst the ‘Senses’ were initially largely conceptual, lacked a clear
and accorded little or no value and status, as reflected in the aca- empirical basis, and had been developed with a long-term setting
demic literature where it has been termed ‘aimless residual care’ in mind, a major study in 1999 (Davies et al., 1999) demonstrated
(Evers, 1981), or at best ‘good geriatric care’ (Reed and Bond, 1991), their value in an acute hospital context. Following the publication
whose aim is to keep older people safely ‘warehoused’. of the Not Because They Are Old report (Health Advisory Service
Nolan (1997) argued that given the prevailing demography, with 2000, 1998), which once more highlighted the lamentable stand-
those aged 85+ being the most rapidly increasing section of the ards of care received by older people in acute care settings, Davies
population, health and social care had to create a positive environ- and colleagues were commissioned by Help the Aged and The
ment for long-term care in order to ensure that the needs of frail Order of St Johns Trust to identify acute hospitals in which older
older people were adequately met and that those providing care people and their families had received good care and to try to dis-
could experience a feeing of job satisfaction. He suggested that in cern what it was that differentiated them from settings in which
a positive long-term care environment older people should experi- older people received poor care. Following a number of detailed
ence six ‘Senses’. These were a Sense of: case studies, Davies et al. (1999) concluded that it was the leader-
◆ Security: to feel safe physically, psychologically, and existentially ship of the ward sister (manager) that was the primary factor. In
care environments in which older people and their families praised
◆ Belonging: to maintain important relationships and to feel part of care, the sister created a ‘positive culture’, which had at its heart
a valued group or community three essential features:
◆ Continuity: to be able to create links between the past, the present, ◆ ‘Valuing fundamental practice’ so that essential elements of
and the future, to experience consistent care delivered by known care such as help with personal hygiene, feeding, and toilet-
people ing were given a high priority. Sister herself often took part in
◆ Purpose: to be able to engage in valued activities, to have some- such activities and provided a clear role model. Such work was
thing to ensure the meaningful passage of time seen as ‘significant’ and provided a strong sense of purpose and
achievement.
◆ Achievement: to be able to achieve valued goals, to feel that your
efforts are valued ◆ ‘Fostering a stable environment’ where there was little staff turn-
over, creating a strong sense of belonging and continuity. At the
◆ Significance: to feel that you, who you are, and what you do in
same time, staff felt safe to question practice and innovate where
some way ‘matter’ to others who are important to you.
this was seen as necessary.
However, he also argued that if staff were to create such an envi-
◆ ‘Establishing clear and equitable therapeutic goals’ so that staff
ronment for older people, then they too have to experience the
knew what was expected of them. This involved negotiating the
‘Senses’ for themselves. So, for example, staff need to feel secure
goals of care with the active involvement of older people and
in their terms of employment and safe to raise any concerns that
their families.
they might have about standards of care. They have to feel that they
belong not only to a staff team or group but also to a wider com- This study demonstrated the value of the Senses in an acute set-
munity of practitioners, something that is hard to achieve in many ting and identified a wide range of factors that created the Senses
long-term care environments where staff turnover is often very not only for staff and older patients but also for family carers
high, compromising senses of belonging and continuity. The lack (Davies et al., 1999).
of therapeutic direction in long-term care has already been alluded During this and later studies, an environment in which the
to and this all but negates a sense of purpose and achievement for Senses were met for all the major stakeholders, and not just one
staff. The low status of work with frail older people (and indeed group such as older people or staff, was termed an ‘enriched envi-
those whose needs fall outside the acute sector generally, such ronment’ (Nolan et al., 2002b, 2006; Brown et al., 2008), and this is
as people with learning disabilities) and the lack of importance captured in the matrix shown in Fig. 15.2.
Stakeholder Older person Staff Family carers Students

Senses
Security
Belonging
Continuity
Purpose
Achievement
Singnificance
Fig. 15.2 The components of an enriched environment. (From Nolan et al. (2006) p. 125.)
As will be apparent, this matrix also includes students, as it has The project team included John Keady and those mentioned in the
been demonstrated that if students experience an enriched learn- Acknowledgements.
ing environment during their clinical placements, then they are far The Senses Framework was developed by Nolan and colleagues
more likely to choose to go to work with older people than if they over several years (as described in Squaring the Quadrangle: The
experience an impoverished environment where the Senses are not ‘Senses Framework’ and Relationship-Centred Care) and covers
created (Nolan et al., 2002b, 2006; Brown et al., 2008). Importantly, the Senses of achievement, belonging, continuity, purpose, security,
the Senses have been developed, refined, and shaped with the active and significance. The case study is taken from the ‘for profit’ care
involvement of staff, older people, family carers, and students who home. Ethical approval to conduct the study was granted on 2 June
have taken part in several studies spanning a number of years (see, 2009 (IRAS reference number 09/H1302/43).
e.g., Davies et al., 1999; Nolan et al., 2001, 2002b, 2006; Faulkner
et al., 2006; Brown et al., 2008). Care home setting
Now that we have considered the evolution of relationship-centred This purpose-built care home was set in a residential suburb of the
care and the Senses, we present a case study that describes their northwest of England. The care home had five ‘houses’ that allowed
recent application in a care home setting. specialist care and attention, including to residents with dementia,
with a total capacity of 150 residents. Four houses were registered
Relationship-Centred Care and the Senses for nursing care and one house was registered for dementia nursing
care. The care home had single occupancy rooms and offered sev-
Framework: Case Study—The Senses eral different facilities on-site, such as hairdressing. The care home
in Practice (SiPs) Care Home Practice was awarded a Care Quality Commission rating of 2 (good) in the
Development Project February 2010 inspection. Carers were given the opportunity prior
to the training programme to share some of their thoughts about
Background the care home and care provision and some of these are reproduced
This case study is taken from a BUPA-funded study that below:
involved two care homes in the northwest of England and
◆ Staff always make me feel very welcome, they chat to me during
took place between June 2009 and September 2010 (Keady et
the visit. My husband seems much happier than he was in the last
al., 2011). One was a ‘for profit’ care home whilst the other
home he was in.
was a ‘not-for-profit’ care home. The overall aim of the pilot
practice development project was to develop a staff education ◆ It would be nice if there were more trips out.
programme that reflected the importance of relationships and ◆ Recently the carpet and soft furnishings have been renewed. This
provided participating staff with an opportunity to identify how has made the visiting much more pleasant.
they might enhance the care of people with dementia using the
Senses Framework. Nolan et al. (2008) suggest that educational Sessions were delivered by two of the project team: John Keady
interventions need to be embedded into an organizational cul- (Professor of Older People’s Mental Health Nursing) and Caroline
ture that provides opportunities for staff to engage in problem Swarbrick (CS) (Research Associate with a PhD background in
solving and promotes relationship-centred care. The objectives dementia care).
of the SiPs project were to:
◆ Develop and deliver a training programme that encouraged
The SiPs training programme
staff to develop dementia care practice using the principles of The SiPs training programme comprised eight sessions of 1 hour
relationship-centred care and the Senses Framework to build duration. The sessions were facilitated predominantly in a desig-
upon commitment and confidence nated training room on the first floor of the main building. The
sessions were jointly delivered by JK and CS and followed a similar
◆ Evaluate the impact of the training programme on the experi-
cycle, with each session focusing on a different Sense and struc-
ences of staff, residents, and families within the home
tured as shown in Table 15.1.
◆ Evaluate how relationship-centred care contributes to the provi- The day and timing of the session were negotiated with each care
sion of individualized care. home to ensure maximum staff attendance and minimum disruption
Table 15.1 The Senses in Practice (SiPs) training programme

Session Title Activity
1 Knowing why we care: understanding the values and skills Completion of attitudes and values questionnaire
we bring to caring
2 Creating a sense of continuity The use of life story work
3 Creating a sense of significance in everyday practice The use of memory chests
4 Creating a sense of belonging in everyday practice Understanding what home means to people
5 Creating a sense of purpose in everyday practice Understanding how we each define purpose in our everyday lives
6 Creating a sense of achievement in everyday practice Understanding what makes people happy in their lives
7 Creating a sense of security in everyday practice Feeling safe with those around you
8 Putting it all together Discussion of record keeping to reflect how staff implement the Senses Framework for
people with dementia
to the home. The training programme comprised a series of interac- ◆ significant fundraising activities to support the work of the
tive workshops where staff were encouraged to consider how impor- home
tant each of the Senses was for them as individuals in their own lives ◆ the reinstatement of uniforms for staff (at the request of staff and
using activities and guided discussions. Staff were encouraged to families)
draw on their personal and professional experiences and were sup-
ported in further discussions focusing on how their knowledge and ◆ a regular education programme for staff
understanding might improve the experience of routine care for ◆ the creation of a ‘skills profile’ for qualified staff
people with dementia. Each staff group developed their own defini-
◆ the introduction of PAT dog ‘Missie’
tions of the Senses Framework to support their use in practice (see
Table 15.2). In the care home, attendance at the workshops fluctu- ◆ the entire redesign of the garden project with the input of local
ated, although it always had a ‘core group’ of attendees. university students
Through this approach, staff attending the training sessions devel- ◆ significant building modifications.
oped a Creating the Senses for… booklet which could be completed by (Adapted from Davies et al., 2007)
residents, families, and staff. The purpose of the booklet was to offer
insight into the resident’s life, such as hobbies and holidays enjoyed. Similar longer-term projects using the Senses Framework in the
The aim was that the booklet would be located in the resident’s community have resulted in the introduction of innovative respite
room, giving the resident a sense of ownership and empowerment. care schemes that have transformed the experiences of people with
Moreover, during the sessions, the value of sensory boxes and mem- dementia, families, and staff (Ryan et al., 2008) and in providing a
ory boxes was discussed, and staff showed a keen interest in pursuing practice lens to view the meaning of interactions for younger people
these ideas. Posters were displayed around the care home inviting with dementia (Davies-Quarrell et al., 2010). However, these inter-
staff, family members, and visitors to donate sensory items, particu- ventions do not represent ‘quick fix’ solutions and have involved
larly those with a reminiscence theme. As one of the care home staff considerable efforts that have fundamentally changed the culture
shared on the final evaluation form: ‘I look at the person [now], not of dementia care, much as was Kitwood’s (1997) original goal. We
the illness.’ It is a fitting testimony to the power of the Senses and life would like to conclude this chapter by looking to the future and
story work and their potential to transform lives and attitudes. It also suggesting two additional areas where we believe there is a need for
provides an insight into how care home staff can develop an individ- culture change and where the Senses may have a role to play. These
ualized, activities role that is drawn from the biography of the person are the diagnostic process and addressing the needs of people with
with dementia, to complement their role in providing physical care. dementia and their families at a societal level.
The above case study describes a pilot study, but other more
intensive interventions have demonstrated the potential for the
Senses Framework to significantly improve outcomes for people Conclusion
with dementia, family carers, and staff in care home facilities. For Support for people with dementia and their families has undergone
example, a 3-year study by Davies et al. (2007) actively engaged a sea change over recent years. Where once there was little inter-
staff and families in enriching the environment in a care home for est and limited, if any, policy initiatives, it is now almost impos-
people with dementia, and resulted in highly significant changes sible to keep abreast of the range of strategies, policy updates, and
including: evidence-based practice guidelines that have emerged to inform
◆ the introduction of an activities programme everyday work and commissioning protocols. However, one area
where there is a centralized focus in the literature is on the impor-
◆ establishing a relatives support group tance and value of an early diagnosis of dementia. Recently, the
◆ designing and producing a welcome booklet for new residents World Alzheimer Report 2011 on early diagnosis and intervention
and relatives suggested that the benefits of early diagnosis included relief from
Table 15.2 The Senses Framework and meanings: an overview developed by care home staff
Sense For residents For families For staff
Purpose Sense of identity Recognizing the family’s importance of caring Having a goal
Individual care for the resident Being needed
Being recognized by residents
Achievement Accomplishment Involving families in the care of the resident Developing a life story book with one of the
Being content residents
Creating memory boxes with residents and their
families
Security Familiarity with the environment Having a key worker to talk to Feeling wanted
Familiar voices Knowing the resident is ‘safe in the hands of Being recognized by residents and their families
Having keys (such as door keys) staff ’ Team working (especially between day and night
Memories Having the opportunity to be involved with staff )
the activities Being flexible
Routines
Getting to know the staff and who they are Knowing people
Stories of ‘home’ or the place the resident
grew up Staff understanding the resident’s biography Working on the same unit and developing
Directions to bedroom Staff recognition that not all families want to relationships
visit and there are many different reasons for (‘smiling faces’)
Photographs and belongings in the
such ‘Christmas families’
bedroom The joy of caring
Significance Being treated as an individual Involving families in helping to develop life Accomplishments being acknowledged
story books and memory boxes
Belonging Feeling of a community Being included in the running of the home ‘Home’ and what ‘makes a home’
Familiar things, such as photographs and Being comfortable Being part of the care home culture
ornaments Feeling integral to the care home Knowing staff and residents
Being comfortable
Continuity Routine Being involved in the residents’ care Working on one unit
Things to do Being kept up-to-date Learning about residents’ life stories
Developing memory boxes and life story
books
Continuity in terms of transition between
home in the community and care home
Ensuring the wellbeing of residents
stress due to a better understanding of symptoms, risk reduction, dementia remains a cause for concern. For example, the same
maximizing decision-making autonomy, and receiving the diag- report by Alzheimer’s Disease International (2011) indicates that
nosis as a human right (Alzheimer’s Disease International, 2011: only 20–50% of people with dementia living in high-income coun-
27). Moreover, to reinforce these perceived benefits, the report tries have a diagnosis, with that percentage falling significantly in
recommended that ‘every country should have a national demen- middle- or low-income countries (p. 4). These findings stand in
tia strategy’ (Alzheimer’s Disease International, 2011: 7), with contrast to the ‘Value of Knowing’ telephone survey (n = 2,678) that
each strategy promoting early diagnosis and intervention through examined public perceptions and awareness of Alzheimer’s disease
awareness raising, training of the health and social care workforce, for adults aged 18 and over in five countries (the US, Germany,
and health system strengthening. In England, objective 2 of the France, Spain, and Poland (<http://www.alzheimer-europe.org/
National Dementia Strategy (Department of Health, 2009a) pro- Research/Value-of-knowing>, accessed 14.02.2012), where one of
vides a national illustration of this global ambition, with its stated the main findings was that more than 8 in 10 adults—from 85% in
aim being to deliver a ‘good-quality early diagnosis and interven- Poland to 95% in Spain—said that if they were exhibiting confusion
tion for all’ (p. 11) whilst linking this outcome to a commissioning and memory loss, they would go to a doctor to determine if the
framework and service delivery pathway that views the diagnosis of cause of the symptoms was Alzheimer’s disease.
dementia as a specialist activity best conducted through a memory These seemingly contrasting findings frame an important ques-
clinic (Department of Health, 2009b). tion, namely why, despite apparently positive public attitudes
Whilst it could be argued that the current push towards early towards a diagnosis and increased policy and service attention in
diagnosis and intervention is both laudable and necessary, the this area, does the under-reporting and underdetection of demen-
limited numbers of older (and younger) adults actually coming tia exist? In response, a number of practical, sociological, and
forward for a memory assessment to consider the existence of a organizational explanations have been put forward, including: poor
attitudes of general practitioners and primary care towards making the Neighbourhood Space now adds a further relational dynamic
a diagnosis (Ahmad et al., 2010; Thomas, 2010); the existence of that reframes the focus to include the relationships that people with
stigma that surround dementia and its diagnosis (MacRae, 1999; dementia and their families have when living their lives ‘outside the
Hamilton, 2008; Alzheimer’s Society, 2010); an overall lack of pro- front door’. The Neighbourhood Space complements existing rela-
fessional knowledge about dementia (Turner et al., 2004; Wilcock tionships with the family carer and care staff and sets a challenge for
et al., 2009; Ahmad et al., 2010); a threatening diagnostic process local civic amenities and town and community transport planning
that probes for loss and not abilities (Keady and Gilliard, 2002); authorities, for example, to take into account the cognitive disabili-
and limited public health information about dementia and its ini- ties that arise from living with a dementia and the impact that this
tial onset and presentation, especially as it impacts upon the neigh- has on judgement, decision-making, and recall, so that at the very
bourhood and informal social networks (National Institute for least people with dementia experience a sense of security in their
Health and Clinical Excellence/Social Care Institute for Excellence, environment.
2006; Georges et al., 2008; Ward et al., 2011; Keady et al., 2012). However, to truly enrich the Neighbourhood Space we need to
All of the above suggests that, in terms of the Senses, there often address wider questions such as: How can the Senses of belong-
currently exists an ‘impoverished environment’ with respect to ing and continuity for people with dementia and their families be
people’s first contact with the ‘professional system’ and their sub- sustained? How can we ensure that they have access to meaningful
sequent experience of the diagnostic process. So, a key question activities that provide a Sense of purpose and achievement? And
becomes: ‘How can we create an enriched environment for people perhaps most importantly of all: How can we ensure that people
who suspect they have dementia and their families?’ We need to with dementia and their families have a Sense of significance and
begin to explore how we can create a sense of security when people genuinely feel that they ‘matter’ to the society in which they live?
first present with their fears about their ‘memory’ so that their con- There is of course a significant part that professional and paid car-
cerns are not dismissed as ‘just old age’. We also need to address the ers can play in addressing such questions, but the societal role is one
stigma associated with ‘dementia’ (Alzheimer’s Society, 2010), and that is lamentably underdeveloped. In our view, the present policies
hopefully high-profile ‘cases’ such as Sir Terry Pratchett will help of personalization and the ‘politics’ of independence do little to fos-
to create a more enriched and open debate about the future chal- ter a wider debate, and neither does the largely rhetorical call for
lenges of helping people to live well with dementia (Department of a ‘Big Society’. Rather, what we need to do is to extend the notion
Health, 2009a, 2010). We need to make the diagnostic procedure of a relationship-centred approach to care beyond the ‘therapeutic
less potentially alienating and fragmented so that senses of belong- quadrangle’ to enable debate to occur at a fundamental level about
ing and continuity are established from the outset. the values that society believes should underpin support and help
In our view, there is also much that could be done to ‘enrich’ the for those who need it. Such is the challenge for the future if the aim
community and neighbourhood environment in which people with of the National Dementia Declaration (<www.dementiaaction.org.
dementia and their families primarily live (National Institute for uk>, accessed 14.02.2012) ‘to change the experience of living with
Health and Clinical Excellence/Social Care Institute for Excellence, dementia in England for good’ (p. 2) is to be realized.
2006; Georges et al., 2008; Ward et al., 2011; Keady et al., 2012),
as current strategies are vague and fail to adequately account for
the everyday lives of people with dementia that take place in public Acknowledgements
places, such as in streets, shops, and civic amenities, e.g. the local Thanks to Dr Caroline Swarbrick and Dr Christine Brown Wilson,
leisure centre, cinema, and public library, and the responsibility that School of Nursing, Midwifery, and Social Work, University of
society has (and not just professional services or carers) to enrich Manchester, for their essential work and roles on the SiPs pilot
the life world of people with dementia. practice development project and for consenting to this case study
Evidence of the importance of everyday access to community being shared in the chapter. Special thanks are also extended to Dr
facilities was evident in an exploratory report by ‘Innovations in Clive Bowman at BUPA and BUPA Giving for their funding of the
Dementia’ that consulted with people with dementia both in com- SiPs project, and to all the staff who participated in the sessions in
munity group and individual settings. This report suggested that the two care homes in the northwest of England.
people with dementia make use of the physical environment,
such as local facilities, support services, social networks, and local References
groups, and that they want to keep in touch with their local com-
Ahmad, S., et al. (2010). GPs’ attitudes, awareness, and practice regarding
munities through membership of local groups, using local facilities, early diagnosis of dementia. British Journal of General Practice, 60, 578,
walking, and using support services (see <http://www.innovation- e360–5.
sindementia.org.uk/>,accessed 14.02.2012). Arguably, it is at this Alzheimer’s Disease International (2009). World Alzheimer report 2009.
personal–public space interface that relationship-centred care, Alzheimer’s Disease International, London.
expressed through the Senses Framework, can be further devel- Alzheimer’s Disease International (2011). World Alzheimer report 2011:
oped to include a new dimension for people with dementia and the benefits of early diagnosis and intervention. Alzheimer’s Disease
their families, one that we have named ‘The Neighbourhood Space’. International, London.
In charting new directions that would improve the quality of lives Alzheimer’s Society (2007). Dementia UK: a report into the prevalence and
cost of dementia, prepared by the Personal Social Services Research Unit
for people with dementia and their families, we need to consider
(PSSRU) at the London School of Economics and the Institute of Psychiatry
how the Neighbourhood Space can be enriched so that people with at Kings College London for the Alzheimer’s Society. Alzheimer’s Society,
dementia feel safe to more fully participate and ‘maintain their London.
involvement’ (Keady, 1997, 1999) with their community. With Alzheimer’s Society (2010). My name is not dementia: people with dementia
subtle additions to the original ‘reach’ of the Senses Framework, discuss quality of life. Alzheimer’s Society, London.
Bowers, B.J. (1987). Inter-generational caregiving: adult caregivers and their Hellström, I., Nolan, M.R., and Lundh, U. (2007). Sustaining ‘couplehood’:
ageing parents, Advances in Nursing Science, 9, 20–31. spouses’ strategies for living positively with dementia. Dementia: The
Brodaty, H., Green, A., and Koschera, A. (2003). Meta-analysis of psychosocial International Journal of Social Research and Practice, 6, 383–409.
interventions for caregivers of people with dementia. Journal of the Hirschfield, M.J. (1981). Families living and coping with the cognitively
American Geriatric Society, 51, 1–8. impaired. In: Copp, L.A. (ed.) Care of the ageing: recent advances in
Brooker, D. (2004). What is person centred care in dementia? Reviews in nursing, pp. 159–67. Churchill Livingstone, Edinburgh.
Clinical Gerontology, 13, 215–22. Hirschfield, M.J. (1983). Home care versus institutionalization: family
Brown, J., Nolan, M.R., and Davies, S. (2008). Bringing caring and competence caregiving and senile brain disease. International Journal of Nursing
into focus in gerontological nursing: a longitudinal, multi-method study. Studies, 20, 23–32.
International Journal of Nursing Studies, 45, 654–67. Holden, U.P. and Woods, R.T. (1982). Reality orientation: psychological
Bruce, E. and Schweitzer, P. (2008). Working with life history. In: Downs, M. approaches to the ‘confused’ elderly. Churchill Livingstone, Edinburgh.
and Bowers, B. (eds) Excellence in dementia care: research into practice, Jennings, B. (1999). A life greater than the sum of its parts: ethics, dementia
pp. 168–86. Open University Press, Maidenhead. and quality of life. Journal of Mental Health and Ageing, 5, 95–111.
Cohen, D. and Eisdorfer, C. (1986). The loss of self. WW Norton, New York. Kahana, E. and Young, R. (1990). Clarifying the caregiving paradigm:
Davies, S., et al. (1999). Good practice in the hospital care of older people. Help challenges for the future. In: Biegel, D.E. and Blum, A. (eds) Ageing and
the Aged, London. caregiving: theory, research and policy, pp. 76–97. Sage, California
Davies, S., et al. (2007). Changing the culture within care homes for older Keady, J. (1996). The experience of dementia: a review of the literature and
people. In: Nolan, M. R., et al. (eds) User participation in health and social implications for nursing practice. Journal of Clinical Nursing, 5, 275–88.
care research: voices, values and evaluation, pp. 50–68. Open University Keady, J. (1997). Maintaining involvement: a meta concept to describe the
Press, Maidenhead. dynamics of dementia. In: Marshall, M. (ed.) The state of the art in
Davies-Quarrell, V., the ACE Club, and Keady, J. (2010). The ACE approach: dementia care, pp. 25–31. Centre for Policy on Ageing, London.
promoting well-being and peer support for younger people with Keady, J. (1999). The dynamics of dementia: a modified grounded theory study.
dementia. Journal of Mental Health Training, Education and Practice, 5, PhD Thesis. University of Wales, Bangor.
41–50. Keady, J. and Gilliard, J. (1999). The early experience of Alzheimer’s disease:
Department of Health (2009a). Living well with dementia: a national dementia implications for partnership and practice. In: Adams, T. and Clarke, C.
strategy. Department of Health, London. (eds.) Dementia care: developing partnerships in practice, pp. 227–56.
Department of Health (2009b). Joint commissioning framework: national Baillière Tindall, London.
dementia strategy. Department of Health, London. Keady, J. and Gilliard, J. (2002). Testing times: the experience of
Department of Health (2010). Quality outcomes for people with dementia: neuropsychological assessment for people with suspected Alzheimer’s
building on the work of the national dementia strategy. Department of disease. In: Harris, P. B. (ed.) The person with Alzheimer’s disease:
Health, London. pathways to understanding the experience, pp. 3–28. Johns Hopkins
Egset, A.S. and Myklebust, J.O. (2011). Dementia in the family: two Norwegian University Press, Baltimore.
case studies. Dementia: The International Journal of Social Research and Keady, J. and Nolan, M.R. (1994a). Working with dementia sufferers and their
Practice, 10, 571–85. carers in the community: exploring the nursing role. International Nursing
Evans, M. (1999). Ethics: reconciling conflicting values in health policy. School Conference, University of Ulster, Coleraine, Northern Ireland, 30 August
for Health, London. 1994.
Evers, H.K. (1981). Multi-disciplinary teams in geriatric wards: myth or Keady, J. and Nolan, M.R. (1994b). Younger onset dementia: developing a
reality. Journal of Advanced Nursing, 6, 205–14. longitudinal model as the basis for a research agenda and as a guide to
Faulkner, M., et al. (2006). Development of the combined assessment of interventions with sufferers and carers. Journal of Advanced Nursing, 19,
residential environments (CARE) profiles. Journal of Advanced Nursing, 659–69.
55, 664–77. Keady, J. and Nolan, M. (1995a). IMMEL: assessing coping responses in the
Georges, J., et al. (2008). Alzheimer’s disease in real life—the dementia carer’s early stages of dementia. British Journal of Nursing, 4, 309–14.
survey. International Journal of Geriatric Psychiatry 23, 546–55. Keady, J. and Nolan, M. (1995b). IMMEL 2: working to augment coping
Grant, G., Nolan, M.R., and Ellis, N.C. (1990). A reappraisal of the Malaise responses in early dementia. British Journal of Nursing, 4, 377–80.
Inventory. Social Psychiatry and Psychiatric Epidemiology, 25, 170–8. Keady, J. and Nolan, M.R. (2003). The dynamics of dementia: working
Grant, G. and Nolan, M.R. (1993). Informal carers: sources and concomitants together, working separately, or working alone? In: Nolan, M. R., et al.
of satisfaction. Health and Social Care, 1, 147–59. (eds) Partnerships in family care: understanding the caregiving career, pp.
Grant, G., Nolan, M., and Keady, J. (2003). Supporting families over the 15–32. Open University Press, Maidenhead.
life-cycle: mapping temporality. Journal of Intellectual Disability Research, Keady, J., Page, S., and Hope, K. (2009). The person with dementia. In:
47(4/5), 342–51. Norman, I.J. and Ryrie, I. (eds) The art and science of mental health
Haight, B.K., et al. (2003). Life review: treating the dyadic family unit with nursing: a textbook of principles and practice, 2nd edition, pp. 550–75.
dementia. Clinical Psychology and Psychotherapy, 10, 165–74. Open University Press, Maidenhead.
Hamilton, H.E. (2008). Narrative as snapshot: glimpses into the past in Keady, J., Brown Wilson, C., and Swarbrick, C. (2011). Developing the senses
Alzheimer’s discourse. Narrative Inquiry, 18, 53–82. in practice: a pilot practice development project involving two care homes
in the north west of England. Final report to BUPA Giving, Leeds.
Health Advisory Service (1982). The rising tide. National Health Service,
Sutton. Keady, J., et al. (2012). Neighbourhoods and dementia in the health and
social care context: a realist review of the literature and implications for
Health Advisory Service 2000 (1998). ‘Not because they are old’: an
UK policy development. Reviews in Clinical Gerontology, 22, 150–63.
independent inquiry into the care of older people on acute wards in general
hospitals. Health Advisory Service 2000, London. Kellett, U., et al. (2010). Life stories and biography: a means of connecting
family and staff to people with dementia. Journal of Clinical Nursing, 19,
Hellström, I., Nolan, M.R., and Lundh, U. (2005a). ‘We do things together’:
1707–15.
a case study of ‘couplehood’ in dementia. Dementia: The International
Journal of Social Research and Practice, 4, 7–22. King’s Fund Centre (1984). Living well into old age: applying principles of good
practice to services for people with dementia. Report Number 63. King’s
Hellström, I., Nolan, M.R., and Lundh, U. (2005b). Awareness context theory
Fund Publishing Office, London.
and the dynamics of dementia: improving understanding using emergent
fit. Dementia: The International Journal of Social Research and Practice, 4, Kitwood, T. (1988). The technical, the personal and the framing of dementia.
269–95. Social Behaviour, 3, 161–80.
Kitwood, T. (1990a). Understanding senile dementia: a psychobiographical Nolan, M.R., et al. (2006). The Senses Framework: improving care for older
approach. Free Associations, 1, 60–76. people through a relationship-centred approach. Getting Research into
Kitwood, T. (1990b). The dialectics of dementia: with particular reference to Practice (GRiP) Report No 2, University of Sheffield.
Alzheimer’s disease. Ageing and Society, 10, 177–96. Nolan, M.R., et al. (eds) (2007). User participation in health and social
Kitwood, T. (1997). Dementia reconsidered: the person comes first. Open care research: voices, values and evaluation. Open University Press,
University, Buckingham. Maidenhead.
Kitwood, T. and Bredin, M. (1992a). Towards a theory of dementia care: Nolan, M.R., et al. (2008). The role of education and training in achieving
personhood and well-being. Ageing and Society, 12, 269–87. change in care homes: a literature review. Journal of Research in Nursing,
Kitwood, T. and Bredin, K. (1992b). A new approach to the evaluation of 13, 411–33.
dementia care. Journal of Advances in Health and Nursing Care, 1, Nolan, P. (2003). Voices from the past: the historical alignment of dementia
41–60. care to nursing. In: Keady, J., Clarke, C.L., and Adams, T. (eds) Community
MacDonald, C. (2002) Nursing autonomy as relational. Nursing Ethics, 9, mental health nursing and dementia care: practice perspectives, pp. 3–16.
94–102. Open University Press, Maidenhead.
MacRae, H. (1999). Managing courtesy stigma: the case of Alzheimer’s Page, S. and Fletcher, T. (2006). Auguste D: one hundred years on. Dementia:
disease. Sociology of Health and Illness, 21, 54–70. The International Journal of Social Research and Practice, 5, 571–83.
Maurer, K. and Maurer, U. (2003). Alzheimer: the life of a physician and career Post, S.G. (2001). Comments on research in the social sciences pertaining to
of a disease. Columbia University Press, New York. Alzheimer’s disease: a more humble approach. Aging and Mental Health,
Maurer, K., Volk, S., and Gerbaldo, H. (1997). Auguste D and Alzheimer’s 5 (Suppl 1), S17–19.
disease. Lancet, 349, 1546–9. Reed, J. and Bond, S. (1991). Nurses assessment of elderly patients in hospital.
McCormack, B. (2004). Person-centredness in gerontological nursing: an International Journal of Nursing Studies, 28, 55–64.
overview of the literature. International Journal of Older People Nursing, Roach, P., et al. (2009). Subjective experiences of younger people with
in association with Journal of Clinical Nursing, 13(3A), 33–8. dementia and their families: a review and implications for UK research,
McKeown, J., et al. (2010). The use of life story work with people with policy and practice attention. Reviews in Clinical Gerontology, 18,
dementia to enhance person-centred care. International Journal of Older 165–74.
People Nursing, 5, 148–58. Robb, B. (1967). Sans everything: a case to answer. Nelson, London.
Motenko, A.K. (1989). The frustrations, gratifications and well-being of Rolland, J.S. (1988). A conceptual model of chronic and life threatening
dementia caregivers. Gerontologist, 29, 166–72. illness and its impact on families. In: Chilman, C.S., Nunnally, E.W., and
Mulrooney, C.P. (1997). Competencies needed by formal caregivers to enhance Cox, F.M. (eds) Chronic illness and disabilities, pp. 17–68. Sage, Beverly
elders’ quality of life: the unitility of the ‘person- and relationship-centred Hills.
caregiving (PRCC) trait’. Proceedings 16th Congress of the International Rolland, J.S. (1994). Families, illness and disability: an integrative treatment
Association of Gerontology, Adelaide. model. Basic Books, New York.
National Institute for Health and Clinical Excellence/Social Care Institute for Royal College of Physicians (1981). Organic mental impairment in the elderly:
Excellence (2006). Dementia: supporting people with dementia and their a report of the Royal College of Physicians by the College Committee on
carers in health and social care. NICE clinical practice guideline 42. NICE, Geriatrics. Royal College of Physicians, London.
London. Royal College of Psychiatrists (2005). Who cares wins—improving the
Nolan, M. (1997). Health and social care: what the future holds for nursing. outcome for older people admitted to the general hospital: guidelines for
Plenary presentation given at the 3rd RCN Older Persons European the development of liaison mental health services for older people. Report
Conference and Exhibition, Harrogate, 5 November 1997. of a Working Group for the Faculty of Old Age Psychiatry. Royal College
Nolan, M.R. and Grant, G. (1989). Addressing the needs of family carers: of Psychiatrists, London.
a neglected area of nursing practice. Journal of Advanced Nursing, 14, Ryan, T., et al. (2008). Using the Senses Framework to achieve
950–61. relationship-centred dementia care services: a case example. Dementia:
Nolan, M.R., Grant, G., and Ellis, N.C. (1990). Stress is in the eye of the The International Journal of Social Research and Practice, 7, 71–93.
beholder: reconceptualizing the measurement of carer burden. Journal Sabat, S.R. (2001). The experience of Alzheimer’s disease: life through a tangled
of Advanced Nursing, 15, 544–55. veil. Blackwell, Oxford.
Nolan, M.R., et al. (1994). A framework for assessing the needs of family carers: Sabat, S.R. (2002). Surviving manifestations of selfhood. Dementia: The
a multi-disciplinary guide. BASE Publications, Stoke–on–Trent. International Journal of Social Research and Practice, 1, 25–36.
Nolan, M.R., Keady, J., and Grant, G. (1995). Developing a typology of family Sabat, S.R., et al. (2011). The ‘demented other’ or simply ‘a person’? Extending
care: implications for nurse and other service providers. Journal of the philosophical discourse of Ursula Naue and Thilo Kroll through an
Advanced Nursing, 21, 256–65. appreciation of the situated self. Nursing Philosophy, 12, 282–92.
Nolan, M.R., Grant, G., and Keady, J. (1996). Understanding family care: a Scheidt, R.J., Humphreys, D.R., and Yorgason, J.B. (1999). Successful ageing:
multidimensional model of caring and coping. Open University Press, what’s not to like? Journal of Applied Gerontology, 18, 277–82.
Buckingham. Schumacher, K.L., et al. (1998). Family caregiving skill: development of the
Nolan, M.R., Grant, G., and Keady, J. (1998). Assessing carer’s needs: a concept, image. Journal of Nursing Scholarship, 30, 63–70.
practitioner’s guide. Pavilion Publications, Hove. Selwood, A., et al. (2007). Systematic review of the effect of psychological
Nolan, M.R., Davies, S., and Grant, G. (eds) (2001). Working with older people interventions on family caregivers of people with dementia. Journal of
and their families: key issues in policy and practice. Open University Press, Affective Disorders, 101, 75–89.
Buckingham. Social Care Institute for Excellence (2010). Personalisation: a rough guide.
Nolan, M.R., et al. (2002a). Towards a more inclusive vision of dementia Social Care Institute for Excellence, London.
care practice and research. Dementia: The International Journal of Social The NHS Confederation (2010). Acute awareness: improving hospital care for
Research and Practice, 1, 193–211. people with dementia. The NHS Confederation, London.
Nolan, M.R., et al. (2002b). Longitudinal study of the effectiveness of Thomas, H. (2010). Attitudes of primary care team to diagnosing dementia.
educational preparation to meet the needs of older people and carers: Nursing Older People, 22, 23–7.
The AGEIN (Advancing Gerontological Education in Nursing) Project. Townsend, P. (1962). The last refuge: a survey of residential institutions
English National Board for Nursing, Midwifery and Health Visiting, and homes for the aged in England and Wales. Routledge, Kegan Paul,
London. London.
Tresolini, C.P. and the Pew-Fetzer Task Force (1994). Health professions Wilson, H.S. (1989a). Family caregivers: the experience of Alzheimer’s
education and relationships-centred care: a report of the Pew-Fetzer Task disease. Applied Nursing Research, 2, 40–5.
Force on Advancing Psychosocial Education. Pew Health Professions Wilson, H.S. (1989b). Family caregiving for a relative with Alzheimer’s
Commission, San Francisco. dementia: coping with negative choices. Nursing Research, 38, 94–8.
Turner, S., et al. (2004). General practitioners’ knowledge, confidence and Woods, R.T. (1989). Alzheimer’s disease: coping with a living death. Souvenir
attitudes in the diagnosis and management of dementia. Age and Ageing, Press, London.
33, 461–7. World Health Organization (1992). The ICD-10 classification of mental and
Ward, R., Campbell, S., and Keady, J. (2011). Friends for life: a local evaluation behavioural disorders. Clinical descriptions and diagnostic guidelines.
of the one year (2010–2011) Peer Support Project for People with Dementia Division of Mental Health, World Health Organization, Geneva.
in Salford. University of Manchester, Manchester. Zarit, S.H., Reever, K E., and Bach-Peterson, J. (1980). Relatives of the
Wilcock, J., et al. (2009). Concordance with clinical practice guidelines for impaired elderly: correlates of feelings of burden. Gerontologist, 20,
dementia in general practice. Aging and Mental Health, 13, 155–61. 649–55.
Wilkin, D. and Hughes, B. (1986). The elderly and the health services. In: Zarit, S.H., Todd, P A., and Zarit, J.M. (1986). Subjective burden of
Phillipson, C. and Walker, A. (eds) Ageing and social policy: a critical husbands and wives as caregivers: a longitudinal study. Gerontologist,
assessment, pp. 163–83. Aldershot, Gower. 26, 260–6.
CHAPTER 16
Psychological treatments
Philip Wilkinson
The last few years have seen an increase in the range of psychologi- a depressed dementia caregiver might be to establish a new social
cal interventions used in the treatment of mental disorders in older network; although the endpoint is the same, the two therapies will
people and an increase in the number of clinical trials, particularly employ different techniques to reach it. Both interpersonal therapy
of cognitive behaviour therapy. However, given the anticipated and psychodynamic therapy have a focus on interpersonal relation-
needs of an expanding population of older people, there is still a ships, but in the former this is made explicit through the use of an
significant paucity of research. This chapter reviews the diverse interpersonal inventory, and in the latter it might be more implicit
range of psychological treatments that have been used to help older through exploration of the therapeutic relationship.
people, the factors that influence their suitability for older people’s Different psychotherapeutic approaches may be used alongside
needs, and the availability of treatments in mental health services. each other. For example, individual therapy with a dementia car-
egiver might be combined with family meetings (Mittelman et al.,
The Range of Psychological Treatments 2003). Integrative models of psychotherapy, in which methods
from different therapies are blended together, are also used in the
Used with Older Adults treatment of older people’s personality problems. These include
Table 16.1 lists some of the therapies currently used with older cognitive analytic therapy (Hepple, 2002) and dialectical behaviour
people. Treatments can be divided into those devised specifically therapy (Lynch et al., 2007).
for older adults (usually to help patients with dementia) and those
developed with younger adults and later used with older people Indications for Using Psychological
(such as treatments for depression). Therapies in the first category
include reminiscence therapy, validation therapy, and reality orien- Treatments with Older Adults
tation. These were some of the first treatment approaches to be for- Psychological treatments might be offered to older patients or their
malized and evaluated, starting in the 1960s. Reminiscence therapy caregivers for a number of reasons (Table 16.2); often they will be
uses group discussions, photographs, recordings, and objects to used alongside biological and social interventions. An understand-
trigger personal memories and promote wellbeing. Many of the tri- ing of psychological treatment models can also help clinicians to
als used in their evaluation have significant shortcomings by today’s construct useful formulations of clinical problems even if formal
methodological standards. A Cochrane review (Woods et al., 2005) psychotherapy is not employed.
of reminiscence therapy for people with dementia did not identify A significant development in recent years has been the inclusion
any rigorous trials or economic analyses in this field. A large mul- of simple psychological interventions, such as problem-solving
ti-centre trial published more recently of reminiscence groups for therapy, in multicomponent collaborative care interventions for
people with dementia and their family caregivers did not provide depressed older adults (Hunkeler et al., 2006). Trials of cogni-
support for the effectiveness or cost-effectiveness of reminiscence tive behaviour therapy and interpersonal therapy are reviewed in
groups and revealed raised anxiety and stress amongst caregivers Chapters 17 and 18.
(Woods et al. 2012).
The other treatments described in this chapter are the more Making Psychological Treatments Available
prominent therapies from the second category. They are psychody-
namic psychotherapy, family therapy, cognitive behaviour therapy, to Older People
and interpersonal psychotherapy. In recent years the latter two Enabling older people to access psychological treatments has
therapies have begun to occupy an important place alongside phar- always presented particular challenges; addressing these challenges
macological treatments in the management of common psychiatric is crucial in improving services. While many patients may be suf-
disorders of old age. ficiently fit and mobile to attend clinics independently, others may
While categorization of psychological treatments is useful, it is require provision of special transport or home visits by therapists.
also liable to create false distinctions between therapies whose aims As patient transport is costly, providing treatment on a group basis,
and methods overlap. For example, in both cognitive behaviour where feasible, may improve efficiency as well as providing second-
therapy and interpersonal therapy, an important step in treating ary social benefits for patients. Other ways to improve patients’
Table 16.1 Some of the psychological treatments used with older memory may also help to compensate for any deficits in sponta-
adults neous recall. Sensory deficits can often be circumvented by using
technical aids or by audio recording therapy manuals for visually
Cognitive behaviour therapy
impaired patients. Adaptations of cognitive behaviour therapy and
Problem solving therapy interpersonal therapy are discussed in Chapters 17 and 18.
Dialectical behaviour therapy
Interpersonal therapy
Research into Psychological Treatments
Family (systemic) therapy
Reminiscence therapy
with Older People
Validation therapy Opponents of psychological treatment research argue that tri-
Psychoanalytic and psychodynamic psychotherapy als are more likely to be flawed than trials of medication, making
Cognitive analytic therapy their conclusions less credible. A particular problem is the difficulty
standardizing a psychological intervention and the performance of
Cognitive stimulation therapy
therapists in a clinical trial (see Parry 2000 for a fuller discussion
of these issues). However, if research into psychological treatments
is not conducted, then older people stand to lose out as agencies
commissioning healthcare seek to prioritize empirically supported
Table 16.2 Indications for psychological treatments with older adults
interventions. It is also important to establish whether potentially
Patient preference, as an alternative treatment to medication costly treatments deliver benefits above and beyond those that are
e.g. in the treatment of an anxiety disorder known to be derived from nonspecific social interventions (Pitkala
Augmenting the effect of psychotropic medication et al., 2011).
e.g. in the treatment of a depressive disorder Patients’ interests are best served when treatment decisions are
informed by empirically based findings rather than the therapist’s
To avoid the use of potentially harmful medication
experience and preference alone (Mansfield and Addis, 2001a).
e.g. in managing behavioural symptoms of dementia
The following considerations will help clinicians to apply the find-
To foster adherence to medication ings of research studies to their patients (Straus and McAlister,
e.g. in the treatment of depression 2004).
To help distressed caregivers
e.g. treating a dementia caregiver experiencing depressive and anxiety The similarity of the patient to those in the study
symptoms In order for a randomized controlled trial to produce valid results,
To alleviate psychological problems related to ageing as many participants as possible should complete the intervention
e.g. to help achieve contentment and acceptance of ageing or to resolve and be monitored for the whole of the planned follow-up period.
disputes within a family brought on by the illness of an older family To ensure this is the case, psychological treatment trials with older
member adults may exclude people with significant physical illness or cogni-
As part of a collaborative care intervention in the treatment of depressive tive impairment and those living in institutions who cannot reach
illness treatment centres; these factors, however, could be important in
To provide the clinician and patient with a psychological formulation of a determining how patients respond to treatment. Biological vari-
patient’s problems ables in older people may also influence the generalizability of find-
ings of research trials performed with younger adults. For example,
as late-life depression is associated with vascular, inflammatory,
access to treatment include bibliotherapy (Smith et al., 1997), as and immune changes (Alexopoulos, 2005) it cannot be assumed
well as telephone and internet-based systems (Rollman, 2010). It that interventions shown to be efficacious with younger people will
is also important to consider who will be available to support the be as beneficial with older adults.
use of a psychological intervention with an older person. Where
patients with dementia living in care homes are concerned, inter- The feasibility of the treatment in the clinical setting
ventions may be better directed at paid carers rather than directly at Key considerations are the availability of trained therapists to pro-
patients themselves. Cognitive therapy with a family caregiver takes vide the intervention and the ease with which patients can access
this a step further by combining care skills training with attention these therapists. The shorter and simpler an intervention is, the
to the carer’s own emotions and reactions. more likely patients are to engage and benefit from it. Use of a
treatment manual can help in the delivery of a psychological treat-
Taking account of sensory and cognitive changes ment and helps to ensure that the practice of a therapy is faithful
Deficits in verbal reasoning, speed of responses, and sensory func- to that in a research study. This is particularly important when the
tion may result in difficulty understanding the complex verbal con- terminology applied to treatments is misleading. For example, the
tent of some psychological treatments. Therefore, evaluation for term ‘reminiscence therapy’ can apply to a range of therapeutic
psychological treatment should include at least a brief assessment approaches with older people that involve any sort of life review
to look for possible deficits (Gallagher-Thompson and Thompson, and that are used both in depression and dementia (Woods, 2004).
2010). Frequently asking patients to summarize their understand- Another benefit of therapy manuals is that they can form the basis
ing of therapy sessions helps the therapist to check the understand- of dissemination and training in a new intervention (Mansfield and
ing of therapy session content; use of cued recall and recognition Addis, 2001b).
CHAPTER 16 psychological treatments 231
The likely benefits of the treatment younger people by improving access to specialist psychological serv-
The clinician needs to be aware of the clinical outcomes that have ices. The Dementia Guideline published in 2006 by the UK National
been measured in a research trial. Often these are scores on symp- Institute for Health and Clinical Excellence also promotes the use of
tom rating scales that might be difficult to apply directly to clinical psychological treatment to help caregivers (NICE, 2006).
populations. More pragmatic trial outcomes include relapse rates Little is known of older people’s understanding of and attitudes
of a disorder, admission rates to a care home, or prescribing rates to psychological treatments. There is some evidence that many
of potentially harmful antipsychotic medication in dementia. Some depressed older people are positively inclined towards learning
therapies, such as psychodynamic therapy, may have aims that are behavioural strategies to help them to manage their depression, and
not related directly to symptom severity, such as developmental that while some might regard psychological treatment with scep-
goals; this makes it more difficult finding measurable outcomes to ticism, many can see the benefits of talking to a psychotherapist
capture change. (Lawrence et al., 2006). In patients of all ages attending primary care,
In treatment studies of psychological therapies the negative psychotherapy is more likely than antidepressant medication to be
effects of treatment are less often measured than they are in tri- perceived as solving underlying problems and is generally preferred,
als of drug treatments, although it has been shown that a propor- although this is not so strong with older patients (Van Schaik et al.,
tion of patients receiving psychological therapy for depression may 2006). This may reflect lack of understanding of psychotherapies by
deteriorate during treatment (Ogles et al., 1995). Psychological older patients and suggests a need for patient education.
treatments are, by their nature, expensive to provide, so cost effec- In an effort to improve access to psychological treatments in the
tiveness is also an important outcome to measure (Bosmans et al., UK, a system of graded provision (stepped care) has been introduced
2007; Holman et al., 2011). under the Improving Access to Psychological Therapies programme
The current evidence on implementing psychological interven- (<www.iapt.nhs.uk>). This has seen the expansion of low-intensity
tions in dementia care is largely derived from intensive, short-term provision in primary care for milder disorders. Referral rates by
studies. There remains a challenge to carry out trials that are prag- general practitioners of older adults has been lower than expected,
matic and applicable to real-world settings in which benefits for which may reflect patchy availability of therapists skilled in work-
caregivers are adequately assessed (Ministerial Advisory Group on ing with older people, under-recognition of mental health prob-
Dementia Research, 2011). lems in older people, or practical problems such as lack of transport
(James, 2010).
It is often debated whether psychological treatments for older
The patient’s values
people should be provided in generic old age mental health serv-
The clinician should be in a position to give clear information ices or in separate psychological treatment services for adults of all
about the nature, duration, and likelihood of benefits of treatment. ages. The second model is difficult to support, however, as work
Patients who are naturally more autonomous may choose a psycho- with older people requires additional skills such as cognitive assess-
logical intervention in preference to a drug treatment, while those ment, an understanding of physical illness, and an ability to liaise
who prefer treatments requiring little effort may make the opposite with hospitals and care homes. It makes sense that the development
choice. and evaluation of treatments continue to take place within the con-
text of integrated old age psychiatry services.
Psychological Treatment Services
References
for Older Adults Alexopoulos, G. (2005). Depression in the elderly. Lancet, 365, 1961–70.
A survey of UK National Health Service psychotherapy depart- Bosmans, J.E., et al. (2007). Cost-effectiveness of interpersonal psychotherapy
ments showed that formal psychotherapy provision for older people for elderly primary care patients with major depression. International
was rare and that referral rates of older people to specialist psy- Journal of Technology and Assessment in Healthcare, 23, 480–7.
chotherapy services were low compared with younger adults. This Collins, E., Katona, C., and Orrell, M.W. (1997). Management of depression in
applied especially to the older old population (Murphy, 2000). It is the elderly by general practitioners: referral for psychological treatments.
more difficult to determine the extent of provision within generic British Journal of Clinical Psychology, 36, 445–8.
Department of Health (2005). Securing better mental health for older adults.
psychiatric services for older people. A study of general medical
<www.dh.gov.uk> (accessed 12.12.2011).
practitioners’ attitudes to the referral of older people with depres-
Gallagher-Thompson, D. and Thompson, L. (2010). Treating late-life
sion for psychological treatments revealed that 93% would consider depression: a cognitive-behavioral approach. therapist guide. Oxford
referring older people for psychological treatment, but only 44% University Press, New York.
had ever done so (Collins et al., 1997). In the US, uptake of treat- Hepple, J. (2002). Cognitive analytic therapy. In: Hepple, J., Pearce, J., and
ments by older people has also been shown to be poor (Wei et al., Wilkinson, P. (eds) Psychological therapies with older people, pp. 128–60.
2005). Brunner-Routledge, Hove.
Although provision and referral rates appear to be poor, the Holman, A., et al. (2011). Cost-effectiveness of cognitive behaviour therapy
importance of providing psychological treatments for older peo- versus talking and usual care for depressed older people in primary care.
BMC Health Service Research, 11, 33.
ple is recognized by policy-makers. The UK Department of
Hunkeler, E.M., et al. (2006). Long term outcomes from the IMPACT
Health’s document Securing Better Mental Health for Older Adults randomised trial for depressed elderly patients in primary care. British
(Department of Health, 2005) reviewed progress with the National Medical Journal, 332, 259–62.
Service Framework for Older People published in 2001. It reported James, I.A. (2010). Cognitive behavioural therapy with older people.
that there was still progress to be made in ensuring that older peo- Interventions for those with and without dementia. Jessica Kingsley,
ple benefited from the developments that had become available to London.
Lawrence, V., et al. (2006). Coping with depression in later life: a qualitative Pitkala, K.H., et al. (2011). Effects of socially stimulating group intervention
study of help-seeking in three ethnic groups. Psychological Medicine 36, on lonely, older people’s cognition: a randomized, controlled trial.
1375–83. American Journal of Geriatric Psychiatry, 19, 654–63.
Lynch, T.R., et al. (2007). Treatment of older adults with co-morbid Rollman, B. (2010). Commentary. Evidence Based Mental Health, 13, 20.
personality disorder and depression: a dialectical behavior therapy Smith, N.M., et al. (1997). Three-year follow-up of bibliotherapy for
approach. International Journal of Geriatric Psychiatry, 22, 131–43. depression. Journal of Consulting and Clinical Psychology, 2, 324–7.
Mansfield, A.K. and Addis, M.E. (2001a). Manual-based treatment. Part 2: Straus, S.E. and McAlister, F. (2004). Applying the results of trials and systematic
the advantages of manual-based practice in psychotherapy. Evidence reviews to our individual patients. Evidence Based Mental Health, 4, 6–7.
Based Mental Health, 4, 100–1. The Ministerial Group on Dementia Research (2011). Headline report to the
Mansfield, A.K. and Addis, M.E. (2001b). Manual-based psychotherapies in Department of Health. <www.dh.gov.uk/en/Publicationsandstatistics/
clinical practice. Part 1: assets, liabilities, and obstacles to dissemination. Publications/PublicationsPolicyAndGuidance/DH_127750> (accessed
Evidence Based Mental Health, 4, 68–9. 10.12.2011).
Mittelman, M.S., Epstein, C., and Pierzchala, A. (2003). Counseling the Van Schaik, A., et al. (2006). Interpersonal psychotherapy for elderly patients
Alzheimer’s caregiver. AMA Press, New York. in primary care. American Journal of Geriatric Psychiatry, 14, 777–86.
Murphy, S. (2000). Provision of psychotherapy services for older people. Wei, W., et al. (2005). Use of psychotherapy for depression in older adults.
Psychiatric Bulletin, 24, 184–7. American Journal of Psychiatry, 162, 711–17.
National Institute for Health and Clinical Excellence (2006). Dementia: Woods, B. (2004). Review: reminiscence and life review are effective therapies
supporting people with dementia and their carers. Clinical guideline. for depression in the elderly. Evidence Based Mental Health, 7, 81.
<www.nice.org.uk> (accessed 12.12.2011). Woods, B., et al. (2005). Reminiscence therapy for dementia. Cochrane Library,
Ogles, B.M., Lambert, M.J., and Sawyer, J.D. (1995). Clinical significance doi: 10.1002/14651858.CD001120.pub2. <www.thecochranelibrary.
of the National Institute of Mental Health Treatment of Depression com> (accessed 28.12.2011).
Collaborative Research Program data. Journal of Consulting and Clinical Woods, R., et al. (2012). REMCARE: reminiscence groups for people
Psychology, 63, 321–6. with dementia and their family caregivers—effectiveness and cost-
Parry, G. (2000) Evidence based psychotherapy: special case or special effectiveness pragmatic multicentre randomised trial. Health Technology
pleading? Evidence Based Mental Health, 4, 35–6. Assessment, 16(48).
CHAPTER 17
Cognitive behaviour therapy
Philip Wilkinson
Cognitive behaviour therapy (CBT) began to emerge as a distinct emphasis on the disorder-specific formulation than on individual
psychological treatment in the 1970s. With its roots in behaviour formulations.
therapy and cognitive psychology, CBT combines attention to Cognitive behaviour therapy involves an active exchange of
behavioural patterns with a focus on thought processes. Developed ideas between patient and therapist, with the therapist able to draw
initially as an intervention for depressive disorder, it subsequently upon a range of techniques to help the patient to recognize streams
became a mainstay in the treatment of anxiety disorders. New appli- of thought and to modify the negative meanings that are given to
cations for CBT have since been developed such that, for younger situations. Negative thoughts may be tackled directly through diary
adults at least, it is now a core mental health intervention. monitoring or through the use of behavioural experiments. These are
The central concept of the cognitive behavioural model is that emo- planned activities undertaken by patients in order to disconfirm exist-
tional states, both positive and negative, are determined by the way in ing unhelpful thoughts and provide evidence for new, more adaptive
which the person processes information about the self and the world thoughts. Many of these exercises are carried out between therapy
around. Negative automatic thoughts fuel unpleasant emotional sessions by the patient and backed up with written notes and records.
states and particular patterns of behaviour which, in turn, maintain Behavioural experiments also provide the therapist and patient with
distorted thinking patterns (Beck et al., 1979). Unhelpful negative further information about the causes of symptoms and the nega-
thinking patterns may be triggered by events in the outside world, tive effects of behaviours that patients may have been using to try to
but underlying them are belief systems, or assumptions, that are ameliorate their distress (Bennett-Levy et al., 2004). For instance, in
often quite specific to the individual. Older people may hold personal attempting to conceal their anxiety, socially anxious patients might
beliefs that were helpful to them in their younger years but that give employ behaviours that actually draw attention to them.
rise to problems in the face of negative life events or impoverished The competent delivery of CBT requires the therapist to work col-
social circumstances in later life (James et al., 1999). For instance, an laboratively with the patient in an organized approach. Treatment is
older lady who holds the belief ‘my place is in the family’ may lead supported by therapist supervision and use of therapist competence
a contented life for many years, only to become depressed when her measures (Improving Access to Psychological Therapies, 2007).
adult children and grandchildren move away from her area, leaving Measurement scales are used to assess the severity of the patient’s
her on her own. Beliefs held by older people may also be influenced symptoms as well as changes specific to CBT, such as the patient’s
by society and generation. Typically these beliefs begin in childhood thinking style and core beliefs. For a description of scales suitable
and are reinforced over the years; they include negative attitudes to for use with older people, see James et al. (2010).
ageing such as ‘growing older is growing weaker’ or ‘old people must
not be a burden to their families’, as well as ideas about illness such as Assessment of Patients for CBT and
‘depression is shameful’ (Laidlaw et al., 2004).
In CBT, information about the patient’s thought patterns and Adaptations for Working with Older People
underlying beliefs is used to derive a case formulation. This is Based on their considerable clinical experience and numerous stud-
combined with hypotheses about the triggers to the episode, ies of CBT with older people, Gallagher-Thompson and Thompson
which, in older people, often include retirement, ill health, and (2010) suggest a number of prerequisites for determining suitability
relationship strains (Thompson, 1996). The formulation, therefore, of CBT. These include the patient being able to adequately process
helps the therapist and patient to reach a shared understanding of information and enjoying sufficient physical health to attend treat-
why and how an episode of depression or anxiety has come about, ment sessions. In patients with memory problems, they also recom-
helps them in devising therapeutic interventions, and keeps the mend assessment of the specific role of psychological factors such
therapy on track. This formulation-driven approach to treatment as anxiety about the memory loss, and suggest the use of behav-
is combined with an empirically based treatment protocol specific ioural experiments to tackle this.
to the disorder in question. Formulations can take many forms, Specific strategies are used to help the older patient successfully
according to the needs of the patient and the aims of the interven- engage in treatment; these include involvement of family members
tion (James, 2010). CBT is usually delivered to the patient as an in treatment sessions to share the formulation and to understand
individual; when it is used as a group treatment, there is a greater the role of self-defeating behaviours (Gallagher-Thompson and
Thompson, 2010). Top-up telephone conversations with the thera- Table 17.1 Major cognitive behaviour therapy strategies used in the
pist between treatment sessions may also be useful (Mohlman, treatment of depression
2004).
Cognitive strategies Distraction techniques
CBT in the Treatment of Depression Counting thoughts
Behavioural strategies Monitoring activities, pleasure, and mastery
Trials of CBT with depressed older people
Scheduling activities
A large body of evidence now supports the efficacy of CBT as a
Graded task assignment
treatment for depression with adults of working age (NICE, 2010).
Individual CBT is as effective as antidepressants in reducing symp- Cognitive-behavioural Identifying negative automatic thoughts
toms of depression and produces more enduring benefit than anti- strategies Questioning negative automatic thoughts
depressant treatment alone; it also appears to be better tolerated Behavioural experiments
than antidepressants. Adding CBT to antidepressant treatment can Preventative strategies Identifying assumptions
also improve outcome in more severe depression and possibly in
Challenging assumptions
chronic depression.
Use of setbacks
There is a plethora of published clinical trials of CBT with
older adults. Most of these, however, have methodological short- Preparing for the future
comings such as small sample sizes, reliance on relatively young (Taken from Fennell, M.J.V. (1989) with author’s permission.)
media-recruited participants, absence of suitable control groups,
and unspecified treatment techniques (Mackin and Areán, 2005).
Comparison groups in these trials include other psychotherapies in the effects of retreating to bed for a long period with the effects of
individual and group format, placebo, waiting list, and antidepres- engagement in an activity, whilst recording feelings, and sense of
sant medication. A Cochrane Systematic Review of psychological mastery and pleasure (Fennell et al.., 2004). This demonstrates to
treatments for depressed older adults included five trials comparing the patient that increasing activity lifts mood.
a range of cognitive behavioural approaches with waiting list con-
trols in a total of 153 participants (Wilson et al., 2008). The cognitive Behavioural activation
behavioural therapies were found to be significantly more effective Behavioural activation can be used as a component of CBT or as a
than waiting list controls in reducing depression severity. However, stand-alone intervention in depression. It aims to overcome the lack
in view of the heterogeneity in the studies and low number of par- of response-contingent positive reinforcement in depressed patients’
ticipants, the authors recommend caution in generalizing these lives. Negative life events are seen as decreasing the likelihood of adap-
findings to clinical populations. In the NICE Depression Guideline tive behaviours for vulnerable individuals, which leads to a downward
(2010), only two acute-phase randomized controlled trials with spiral of low mood and a decrease in motivation (Martell et al., 2010).
older adults met the reviewers’ quality standards (Thompson, 2001; Depressed individuals are more likely to be sensitive to negative stim-
Laidlaw et al., 2008). It was concluded that CBT may be efficacious, uli, and behavioural activation aims to increase the likelihood that
but, again, caution was advised in interpreting the results due to beneficial patterns of behaviour will be positively reinforced.
the small number of participants. A further randomized controlled
trial (Serfaty et al., 2009) compared CBT with two conditions: treat- Problem-solving therapy
ment as usual and treatment as usual with talking control in 204 Problem-solving techniques may be used as a component of CBT or
people aged 65 or over with a Geriatric Mental State diagnosis of on their own in the form of problem-solving therapy (PST). PST is an
depression. Based on improvement in Beck Depression Inventory intervention of six to eight sessions that draws on cognitive behav-
score, CBT was of greater benefit than treatment as usual and talk- ioural principles. Behavioural aspects include tackling the avoid-
ing control over a period of 10 months. Group CBT has also been ance and reduction in pleasurable activities that occur in depression;
evaluated in the prevention of depression recurrence (Wilkinson cognitive aspects include addressing the negative perceptions that
et al., 2009), but at present there is insufficient evidence from which may interfere with finding practical solutions to problems. Therapy
to draw any clear conclusion on its efficacy (NICE, 2010). There follows a standard set of stages: definition of problems; establishing
remains a need for large, high-quality randomized trials of CBT to realistic goals; generating, choosing, and implementing solutions;
replicate current findings with participants who are typical of those and evaluating outcomes (Mynors-Wallis, 2001).
encountered in routine clinical practice.
Content of CBT in the treatment of depression Case example The following case example demonstrates the use
of formulation, activity monitoring, and questioning of negative
A typical course of individual CBT for depression lasts up to 16 ses-
automatic thoughts in the treatment of a depressed older patient.
sions and involves a range of therapeutic strategies (Table 17.1).
Two years after the death of her husband, Mrs Lloyd had moved
Behavioural strategies are used early on in treatment with the aim
to a new city to live near her son and his family. She had always
of making some immediate positive impact on mood and helping
been active and high-achieving and hoped to be useful to them.
the patient to engage with therapy. Activity monitoring using a sim-
However, her vision was failing due to macular disease which
ple diary can show patients how inactivity worsens low mood and
she found difficult to tolerate as it prevented her from doing the
the effect of negative thinking on underestimating achievement
things for the family that she would have wished to do. She soon
and enjoyment. A useful behavioural experiment to demonstrate
became depressed and frustrated.
the effect of inactivity to a tired, depressed patient is to compare
CHAPTER 17 cognitive behaviour therapy 235
In working on a case formulation with her therapist (Fig. 17.1), a tendency then to reflect on the day with self-critical and other
Mrs Lloyd soon recognized her tendency to try to do a lot of negative thoughts.
things at once or to fuss over her son and his family in an effort to Mrs Lloyd undertook the following behavioural experiment.
appear useful; she also tended to discount her son’s positive state- Having predicted that her son and his wife would be disap-
ments about her. She had become less motivated with a tendency pointed in her if she chose to entertain an old friend rather than
to stay in bed till late and to neglect her own care; she had also offer to look after her grandchildren one day, she put it to the
become anxious before social occasions such as church meetings. test. She observed and tackled her thoughts before and during
The formulation process also included discussion of her earlier life her friend’s visit and tried to limit herself to asking her son only
experiences. She recognized the influence of her father as she was once if they could manage without her that day. It turned out
growing up: he had always insisted that she strive to achieve the that she had an enjoyable day with her friend and discovered
highest standards and to help others at all times. Having drawn up that this had helped her to feel more independent of her son
this formulation, Mrs Lloyd said that, for the first time, events in and to provide some useful topics of conversation when she next
her life were beginning to ‘connect up’. saw him.
Therapy involved identifying the activities that Mrs Lloyd could As Mrs Lloyd’s mood improved she was able to challenge some
still enjoy despite her impaired vision, such as simple gardening of the assumptions that she had expressed, such as ‘unless I suc-
and listening to music in the evening. She recorded her negative ceed at this now, my whole life has been wasted’. She described
thoughts in a diary which enabled her to see that ‘I’ve got into herself as thinking straight by the end of the therapy, with a more
the habit of seeing the threat in every situation’ (Table 17.2). realistic and compassionate view of herself. She drew up a list of
Sometimes she noticed her mood was quite low when she sat strategies to remind her how to recognize depression in the future
down to listen to music; thought records revealed that she had and how to manage setbacks should they occur.
Early experience
Critical father
Pursuit of high standards at home and in family
Unhelpful beliefs
Unless I succeed, my whole life has been wasted
To be worthwhile, I must be useful to other people
Critical incidents
Move to be near son
Failing vision
Negative automatic thoughts
I’m no use to my family now

I don’t fit in here
Life shouldn’t have turned out like this
Physical Moods Behavioural

Trying too hard to help son
Tired
Depressed Looking for reassurance
No energy
Anxious Avoiding social situations
Not sleeping
Giving up and staying in bed
Fig. 17.1 Cognitive behavioural case formulation of Mrs Lloyd, taking into account both her current symptoms and her underlying vulnerability to depression.
Table 17.2 Thought diary of Mrs Lloyd

Situation Emotion (severity) Negative automatic Evidence that it’s Evidence that it’s Alternative Emotion
thoughts true not true thoughts
Waking up at 6 a.m. Despondent (90%) It’s another awful I feel so depressed Yes, but this is just If I take one thing Despondent (20%)
day the depression at a time and
speaking follow my activity
I shan’t be able to I tire so easily, If I stay in bed I’m schedule then I’ll
achieve anything I should rest only going to feel manage
worse
About to visit son and his Sad (80%) I’m no use to my I am older and This is an There are still Sad (10%)
family family now don’t have the all-or-nothing enough things
strength that I did thought. When about me that my
I wrote a list, I family value
identified many
ways in which they
value me
Angry (30%) Life shouldn’t have This wasn’t what No-one can If I take life as it Angry (0%)
turned out like this I was expecting in predict the future. comes I can make
my retirement It’s just down to the best of it
chance. This isn’t
going to help me
to get what I want
Mindfulness-based cognitive therapy Table 17.3 Structure of a mindfulness-based cognitive therapy course
Mindfulness-based cognitive therapy (MBCT) combines teaching
Session one Understanding automatic pilot thinking
on the role of negative automatic thoughts with meditation prac-
tices. It was developed with the aim of preventing the recurrence of Learning to move attention from thoughts to body
depression, following on from the successful treatment of stress in Session two Dealing with barriers
patients with chronic pain and physical illness (Kabat-Zinn, 1990; Learning to cope with the chatter of the mind and
Baer, 2003). Practising mindfulness entails becoming aware that controlling reactions to everyday events
the mind has a will of its own and that mood and actions are often Session three Mindfulness of the breath
governed by a stream of thoughts about past and future events.
Using awareness of breathing to become more focused
Bringing the patient’s attention to an awareness of the present pro- on the present moment
vides a better sense of control over events (Elliston, 2001). The aim,
therefore, is to respond mindfully to adverse situations rather than Session four Staying present
react automatically to streams of negative thoughts according to Education about depression as a package of symptoms
old, ingrained patterns of behaviour. and mindfulness as an alternative perspective
The principle behind MBCT for depression is straightforward: Session five Allowing/letting be
being able mentally to step aside from a stream of thought should Developing an attitude of acceptance because pushing
prevent engagement with the negative automatic thoughts that away unpleasant experience makes it worse
maintain depressed mood. This contrasts with traditional CBT in Session six Thoughts are not facts
which patients learn to engage with and re-evaluate their negative
Finding ways of reducing the degree of identification
thoughts. This ability to switch attention away from futile thought with thoughts and choosing whether to engage
processes is believed to be particularly helpful to sufferers of recur- with them
rent depression in whom minor dips in mood can rapidly lead to
Session seven How can I best take care of myself?
depression because of their tendency to ruminative thinking (Segal
et al., 2002). Identifying signs of relapse and developing an
action plan
MBCT is usually delivered in class format; Table 17.3
shows the structure of a typical course. Participants learn medi- Session eight Using what you have learned to deal with future moods
tation techniques, breathing exercises, and movement exercises Building in regular practice by linking to important goals
such as yoga, and there is an expectation of daily practice of and activities
meditation. The NICE Depression Guideline (NICE, 2010) states (Based on Segal et al. 2002.)
that there is some evidence with younger adults suggesting a
clinically significant difference favouring MBCT on reducing the As MBCT helps patients to cease trying to solve insoluble
likelihood of relapse of depression after 60 weeks of follow-up, problems, it has obvious appeal as a treatment for older adults
particularly in people who have had more than two episodes of facing recurrent depression, chronic physical illness, and other
depression. stressors. Other potential benefits for older people include the
support of class membership, the destigmatizing value of sharing et al., 2004). Patients with generalized anxiety disorder have a
experiences, the provision of education about depression, and a number of fears about their everyday circumstances and engage
method that may be easier to grasp and practise than standard in repetitive worry in an attempt to solve problems; they may then
CBT (Smith, 2007). Mindfulness meditation has also been used to worry about how their worrying is affecting them, leading to a
help older people to tolerate and manage chronic pain (Morone vicious circle of anxiety (Wells, 1997).
et al., 2008).
CBT in the Treatment of Patients with
CBT in the prevention of depression Physical Illness
Older people at risk of developing major depression include those
with chronic illness, functional impairment, and low-level symp- Trials of CBT in older adults with physical health
toms of anxiety and depression. An example of an intervention problems
aimed at preventing depression in older people with physical illness A meta-analysis of ransomized controlled trials of CBT and
(selective prevention) is the use of a problem-solving therapy for problem-solving therapy for depression in people with a range of
people with age-related macular degeneration (Rovner et al., 2007). physical illnesses (Beltman et al., 2010) showed benefits of CBT
Interventions to prevent progression of subsyndromal depression with similar effect sizes in the different physical conditions. CBT
and anxiety to major depression (indicated prevention) include for chronic sleep problems in later life combines education about
use of CBT in a stepped care approach in primary care (Van’t sleep hygiene and normal sleeping patterns in old age with specific
Veer-Tazelaar et al., 2009). cognitive strategies to address dysfunctional attitudes about sleep
and the impact of not sleeping. Montgomery and Dennis (2009)
reviewed the data from six trials of CBT for sleep problems in
CBT in the Treatment of Anxiety older people and concluded there was a small effect of CBT, best
Trials of CBT with anxious older people demonstrated for sleep maintenance insomnia. CBT may also help
older people whose sleep problems are secondary to medical ill-
The UK National Institute for Health and Clinical Excellence con-
ness (Rybarczyk et al., 2005).
cluded that CBT is an efficacious treatment for generalized anxi-
ety disorder and panic disorder in younger adults (NICE, 2011). Content of CBT in the treatment of patients with
Research into treatments for anxiety disorders in older people,
physical illness
however, lags behind that in younger adults. Most trials with
older people address efficacy of CBT in the treatment of general- CBT can be used to help older people with physical illness who
ized anxiety disorder, some involving relaxation training as well develop excess disability as a result of anxiety or depression. The
as cognitive restructuring techniques (Ayers et al., 2007; Stanley aim of treatment is to help the patient to recognize the effects of
et al., 2009). In a large trial in primary care (Zijlstra et al., 2009), thinking errors and unhelpful behaviours, as illustrated in Fig. 17.2
CBT was used to treat the fear of falling and associated avoidance and the following case example.
behaviours in anxious older people. There were benefits for up to
14 months in fear of falling and perceived control over falls, but
not in activity avoidance. Other trials have studied the interaction Case example Mrs Woods was an 81-year-old lady living in sup-
between CBT and medication in anxious older adults (Gorenstein ported accommodation. She had severe chronic obstructive pul-
et al., 2005; Schuurmans et al., 2006). However, there remains a monary disease which confined her to her flat. She experienced
need for large, adequately powered trials of CBT in the treatment episodes of severe breathlessness and dizziness when she had not
of late-life anxiety disorders with participants who are repre- seen anyone for some time, accompanied by thoughts such as
sentative of patients seen in routine clinical practice with longer ‘I’m getting a chest infection and no-one will be here to help me’.
follow-up periods. Her chest physician confirmed that her chest disease was not the
cause of these episodes. She monitored her breathing excessively
Content of CBT in the treatment of anxiety from the moment she woke, which in turn made her more anx-
While there is overlap between anxiety and depression in terms ious and produced hyperventilation and breathlessness. She also
of symptoms and cognitions (negative predictions and a nega- used her beta-agonist inhaler excessively which made her shaky
tive bias in appraising current experiences), in anxiety the nega- and light-headed.
tive appraisals tend to be more selective and specific to certain During CBT, over a period of a few weeks, Mrs Woods moni-
situations (Beck et al, 1985). The experience of anxiety may also tored her chest symptoms and her mood. She began to recog-
tend to increase the probability of the feared event occurring. nize the symptoms of anxiety when they occurred and learned
For example, a socially anxious patient may hesitate in social to differentiate them from symptoms of a chest infection, the
situations, thereby undermining social performance. There are anxiety having a rapid onset and being accompanied by fear, and
a number of disorder-specific models of anxiety disorders. For the infection coming on more slowly and being accompanied by
instance, according to the cognitive model of panic disorder, general malaise and a raised temperature. She recorded and chal-
sufferers misinterpret physical sensations of anxiety as signs of lenged the negative thoughts she experienced and experimented
imminent physical catastrophe (Clark, 1986). In health anxi- with not resorting to extra doses of her inhaled medication. With
ety, patients are afraid of having an undiagnosed serious illness; her agreement, the warden of the establishment where she lived
they monitor and overinterpret physical symptoms and engage was also included in one treatment session so she would know
in safety-seeking behaviours that exacerbate their anxiety (Silver how to help Mrs Woods.
Premorbid Limits imposed Limits imposed by fears,

functional by pathology and thinking errors, and self-
abilities physical impairment limiting actions Fig. 17.2 Zone of therapeutic work in CBT with patients
Increasing experiencing physical illness and excess disability.
loss of p. 37, James, I. A. (2010), Cognitive Behavioural Therapy with
function Older People, Jessica Kingsley Publishers, London and
Philadelphia. (Reproduced with permission from James, Ian
Zone in which CBT can be used to reduce Andrew. Cognitive Behavioural Therapy with Older People, 2010.
levels of disability and handicap © Jessica Kingsley, 2010)
CBT with Dementia Caregivers Table 17.4 Outline of caregiver CBT intervention
Trials of CBT with caregivers Assessment session Depression scores

Identifying and managing depression in caregivers is important Education booklet on dementia
as it is associated with increased breakdown of care and unhelpful Sessions 1, 2, and 3 Education sessions Assessment of caregiver’s model
caregiver behaviours (Cooper et al., 2010). NICE (2006) evaluated of illness and provision of
data from two systematic reviews and 25 trials. It was concluded alternative model if indicated
that there is evidence for the effectiveness of CBT on symptoms Written information on
of depression and anxiety in caregivers who have clinically sig- dementia, caregiver stress, and
nificant symptoms. The review also concluded that therapy availability of local services
for the carer can improve the mental state of the person being Session 4 Stress assessment Explanation of the effects
cared for and delay the move to institutional care. Reframing of of stress and monitoring of
negative automatic thoughts appears to be an active ingredient symptoms
of therapy, at least in reducing caregiver depression and anxiety Sessions 5 and 6 Relaxation training Progressive relaxation and body
(Vernooij-Dassen et al., 2011). There are also recent descriptions scan taught and practised as
of MBCT used with dementia caregivers (Franco et al., 2010; homework
Oken et al., 2010). Sessions 7 and 8 Managing Explanation of role of negative
One of the efficacious CBT interventions is that evaluated by psychological automatic thoughts in stress
Marriott et al. (2000) in comparison with two control conditions, responses to stress Thought records kept and
one an in-depth family interview (also the initial component of challenging of negative
the CBT intervention) and the other usual care only. Patients with automatic thoughts begins
Alzheimer’s dementia and their caregivers were recruited from
Sessions 9 and 10 Managing Tackle problems such as isolation
National Health Service community psychiatry services. To be behavioural and self-sacrificing behaviour
included in the trial, caregivers needed to achieve caseness levels of responses to stress Encourage meeting of own needs
symptoms on the General Health Questionnaire. Compared to the
control conditions, the intervention reduced burden and brought Session 11 Coping skills Identify caregiver’s problematic
about a greater reduction in number of cases on the General Health assessment coping behaviours and thought
patterns
Questionnaire, with a numbers needed to treat of two, at 3 months
follow-up. Sessions 12 and 13 Coping skills: Teach effective coping skills
patient behaviour including use of thought
Content of CBT with dementia caregivers management challenging
The structure of Marriott’s intervention, described in the sec- Sessions 14 and 15 Coping with Identify feelings of loss and,
tion Trials of CBT with caregivers, is summarized in Table 17.4. feelings of loss where appropriate, use thought
The aims of the intervention are: first, to help caregivers appreci- challenging to help caregiver to
ate that stressful interactions can lead to disturbed behaviours draw upon remaining positive
aspects of relationship
in the dementia sufferer and increased distress in the caregiver;
second, to help reduce stressful interactions using behaviour (Taken with permission from therapy manual prepared by A. Marriott and C. Donaldson
management techniques and calmer ways of responding; third, (unpublished) and used in trial by Marriott et al. (2000).)
to reduce stress by the use of relaxation, challenging of negative
automatic thoughts, and the development of the caregiver’s social shame in these situations. Other authors also stress the impor-
networks; and fourth, to help the caregiver cope with feelings of tance of sharing a formulation of the dementia sufferer’s symp-
loss. Coping skills training (sessions 12 and 13) might include, toms with the wider family so as to validate the main caregiver’s
for instance, helping the caregiver to deal with his embarrass- role (Mittelman et al., 2003).
ment when friends visit and his wife, with dementia, becomes Psychological treatment may also be of benefit to the caregiver
agitated and restless. First, the therapist would explain that peo- after the person with dementia has been admitted to a care home.
ple with dementia are susceptible to stress and noisy environ- The following account is written by a caregiver who received CBT
ments and would suggest trying to reduce this source of stress; to help him to manage symptoms associated with his wife’s anxi-
second, the therapist would introduce questioning of negative ety and the guilt he experienced when she moved into a nursing
automatic thoughts to reduce the caregiver’s embarrassment and home.
Case example My wife, Elizabeth, was diagnosed with Depression Associated with Personality
Alzheimer’s disease 4 years ago. We had been married for over Disorder
50 years. After the initial diagnosis, I tried to take care of her at
home, but this was very stressful indeed for both of us and after The finding that certain personality traits in older people are associ-
3 years it became clear that her condition had deteriorated to such ated with recurrent depressive disorder has led to the use of a modi-
an extent that she would have to be cared for professionally, in a fied form of CBT, dialectical behaviour therapy (DBT) (Cheavens
care home. The actual decision to send her into care was a very and Lynch, 2008). DBT assumes a biological predisposition to
painful one for me and my family. After such a long life together, I increased emotional sensitivity and impaired emotional regulation.
had very strong guilt feelings and much stress. I was worried that It combines individual therapy to tackle self-defeating behaviours,
she would not be looked after as well as at home and I became such as repeated disengagement from treatment, with group train-
very fearful about visiting her. I was also concerned about my ing in emotional regulation and managing interpersonal relation-
own future and the emptiness of my life alone at home. ships. Story-based (dialectical) strategies are also used, alongside
I found it impossible to explain to family and friends just how I traditional cognitive techniques.
was feeling. Instead of looking forward to the visits to my wife in
the care home, I dreaded the thought of going there. I had many Depression and Anxiety Associated with
worries about these visits. I was worried that Elizabeth would ask Memory Loss
me such questions as ‘Where am I?’ and ‘Why can’t I go home?’ I
was afraid that she would ask me to take her home, and I did not Psychological interventions for people with memory problems have
know how to deal with such a situation. It was at this point that I two aims: to delay the progression of the memory deficit and to
began to visit a cognitive therapist. From the beginning, the thera- reduce the symptoms of anxiety and depression associated with it
pist seemed to understand my concerns. Sometimes I broke down by increasing the sufferer’s sense of control. Therapies with the first
in sessions, something I could not let myself do with my family aim include memory training and cognitive stimulation treatments
and friends, and it was very helpful to have the therapist’s under- targeted at improving memory and executive skills; therapies with
standing and sympathy. Gradually I was able to express the emo- the second aim are CBT-based.
tions that previously I had hardly been able to understand myself. People experiencing memory problems may develop unhelp-
The therapist told me at once that the feelings and problems I ful beliefs about their problems that cause symptoms, such as ‘in
was experiencing were quite normal reactions for somebody in order to engage in social activities I must remember names and
my position and was able to offer some explanations of the symp- faces’ or ‘I’m the only person who forgets things’ (Kipling et al.,
toms and characteristics of Alzheimer’s disease. For example, he 1999). Patients with mild dementia may be able to use cognitive
told me that, while my wife’s behaviour might sometimes appear techniques to challenge negative automatic thoughts and to cope
normal, it was very probable that she had no real understanding with challenging situations. For example, they might be assisted to
of what she was saying and would certainly have no memory of challenge the self-critical and hopeless thoughts that occur when
any question that she might have asked me. This was an impor- they experience episodes of memory loss. Those with more severe
tant assurance to me, because I realized that I need have no fear cognitive impairment who are unable to use these techniques may
that I would upset her if I did not reply directly to difficult ques- benefit from behavioural activation and pleasant events scheduling
tions. The therapist persuaded me to keep this fact in my mind to overcome depressive withdrawal (Teri and Gallagher-Thompson,
at each visit, and I found that it helped me very much indeed to 1991). Despite the development of these techniques, however, NICE
overcome my worry about seeing my wife. (2006) found no adequately designed trials of interventions aimed
I sometimes wondered whether the method the therapist was directly at dementia sufferers.
using to confront my situation was helpful. He persuaded me to
analyse my thoughts and feelings before each visit to Elizabeth CBT for Psychotic Symptoms
and to write them down after I had returned. I was taught to
question my own feelings and fears, and to rationalize the results In recent years, there have been significant advances in the under-
by filling in a form which he gave me. At first I found this disci- standing and psychological treatment of psychotic symptoms, and,
pline very hard, but I came to realize how valuable it was, espe- for younger adults, CBT is now recommended as an intervention
cially when I was in a mood of despair in which feeling sorry for in schizophrenia. Psychological models of psychosis propose that
myself was uppermost in my mind. The habit of forcing myself adverse environmental conditions interact with a vulnerability to
to write down my thoughts and emotions helped very much to psychosis to trigger emotional changes and to disrupt attention,
address them and to put them into perspective. I learned to see perception, and judgement (Garety, 2001). Anomalous thoughts
the positive side of the situation; although my wife was in care, are then attributed to external sources, giving rise to delusions and
I was able to remember how difficult life had been for her when hallucinations.
she was still at home and I began to notice the simple things that Cognitive behavioural interventions for psychosis focus directly
we could enjoy doing together in the care home. on delusional beliefs, beliefs about hallucinations, and other asso-
I benefited very greatly from the therapy. My feelings of guilt ciated problems such as social anxiety (Close and Schuller, 2004).
about sending Elizabeth away from home are now gone, as With older adults, these approaches have been applied, but not
I realize that what I had to do was in her best interests. The formally evaluated, in the treatment of distressing visual halluci-
visits do not distress me any more and I enjoy spending time nations (Collerton and Dudley, 2004) and psychotic depression
with her. (Wilkinson and Schuller, 2003). Granholm et al. (2004) describe
the use of thought diaries to help patients reattribute distressing
auditory hallucinations to their illness and to avoid responding <www.dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/@dh/@en/

directly to them. They also suggest a number of adaptations for documents/digitalasset/dh_078535.pdf> (accessed 12.12.2011).
working with older patients with psychosis, including repeated James, I.A. (2010). Cognitive behavioural therapy with older people.
interventions for those with and without dementia. Jessica Kingsley,
presentations of material, use of acronyms and mnemonics, and
London.
role play (Granholm et al., 2004).
James, I.A., Kendell, K., and Reichelt, F.K. (1999). Conceptualizations of
depression in older people: the interaction of positive and negative
References beliefs. Behavioural and Cognitive Psychotherapy, 27, 285–90.
Ayers, C., et al. (2007). Evidence-based psychological treatments for late-life Kabat-Zinn, J. (1990). Full catastrophe living. Piatkus, New York.
anxiety. Psychology and Aging, 22, 8–17. Kipling, T., Bailey, M., and Charlesworth, G. (1999). The feasibility of a
Baer, R. (2003). Mindfulness training as a clinical intervention: a conceptual cognitive behavioural therapy group for men with mild/moderate
and empirical review. Clinical Psychology Science and Practice, 10, cognitive impairment. Behavioural and Cognitive Psychotherapy, 27,
125–43. 189–93.
Beck, A.T, et al. (1979). Cognitive therapy of depression. Guilford, New York. Laidlaw, K., Thompson, L.W., and Gallagher-Thompson, D. (2004).
Beck, A.T., Emery, G., and Greenberg, R.L. (1985). Anxiety disorders and Comprehensive conceptualization of cognitive behaviour therapy for late
phobias: a cognitive perspective. Basic Books, New York. life depression. Behavioural and Cognitive Psychotherapy, 32, 389–99.
Beltman, M., Oude Voshaar, R., and Speckens, A. (2010). Cognitive-behavioural Laidlaw, K., et al. (2008). A randomised controlled trial of cognitive behaviour
therapy for depression in people with a somatic disease: meta-analysis of therapy versus treatment as usual in the treatment of mild to moderate
randomised controlled trials. British Journal of Psychiatry, 197, 11–19. late life depression. International Journal of Geriatric Psychiatry, 23,
Bennett-Levy, J., et al. (2004). Behavioural experiments: historical and 843–50.
conceptual underpinnings. In: Bennett-Levy, J., et al. (eds) Oxford Mackin, R.S. and Aréan, P.A. (2005). Evidence-based psychotherapeutic
guide to behavioural experiments in cognitive therapy, pp. 1–20. Oxford interventions for geriatric depression. Psychiatric Clinics of North
University Press, Oxford. America, 28, 805–20
Cheavens, J.S. and Lynch, T.R. (2008). Dialectical behavior therapy for Marriott, A., et al. (2000). Effectiveness of cognitive-behavioural family
personality disorders in older adults. In: Gallagher-Thompson, D., intervention in reducing the burden of care in carers of patients with
Steffen, A.M., and Thompson, L.W. (eds) Handbook of behavioral and Alzheimer’s disease. British Journal of Psychiatry, 176, 557–62.
cognitive therapies with older adults. Springer, New York. Martell, C., Dimidjian, S., and Hermann-Dunn, R. (2010). Behavioral
Clark, D.M. (1986). A cognitive model of panic. Behaviour Research and activation for depression. A clinician’s guide. Guilford Press, New York.
Therapy, 24, 461–70. Mittelman, M.S., Epstein, C., and Pierzchala, A. (2003). Counseling the
Close, H. and Schuller, S. (2004). Psychotic symptoms. In: Bennett-Levy, J. Alzheimer’s caregiver. AMA Press, New York.
et al. (eds) Oxford guide to behavioural experiments in cognitive therapy, Mohlman, J. (2004). Psychosocial treatment of late-life generalized anxiety
pp. 245–63. Oxford University Press, Oxford. disorder: current status and future directions. Clinical Psychology Review,
Collerton, D. and Dudley, R. (2004). A cognitive behavioural framework 24, 149–69.
for the treatment of distressing visual hallucinations in older people. Montgomery, P. and Dennis, J. (2009). Cognitive behavioural interventions
Behavioural and Cognitive Psychotherapy, 32, 443–55. for sleep problems in adults aged 60+. Cochrane Database of Systematic
Cooper, C., et al. (2010). Family carers’ distress and abusive behaviour: Reviews, 10.1002/14651858.CD003161. <www.thecochranelibrary.com>
longitudinal study. British Journal of Psychiatry, 196, 480–5. (accessed 12.12.2011).
Elliston, P. (2001). Mindfulness in medicine and everyday life. British Medical Morone, N.E., Greco, C.M., and Weiner, D.K. (2008). Mindfulness meditation
Journal, 323(2), 7322. for the treatment of chronic low back pain in older adults: a randomized
Fennell, M., Bennett-Levy, J., and Westbrook, D. (2004). Depression. In: controlled pilot study. Pain, 134, 310–19.
Bennett-Levy, J., et al. (eds) Oxford guide to behavioural experiments in Mynors-Wallis, L. (2001). Problem-solving treatment in general psychiatric
cognitive therapy, pp. 205–22. Oxford University Press, Oxford. practice. Advances in Psychiatric Treatment, 7, 417–25.
Fennell, M.J.V. (1989). Depression. In: Hawton, K., et al. (eds) Cognitive NICE (2006). Dementia: the NICE–SCIE guideline on supporting people with
behaviour therapy for psychiatric problems, pp. 169–234. Oxford dementia and their carers in health and social care. National Practice
University Press, Oxford. Guideline 42. <www.nice.org.uk> (accessed 12.12.2011).
Franco, C., Del Mar Sola, M., and Justo, E. (2010). Reducción del malestar NICE (2010). The treatment and management of depression in adults (updated
psicológico y del la sobrecarga en familiares cuidadores de enfermos de version). National Clinical Practice Guideline 90. <www.nice.org.uk>
Alzheimer mediante la aplicación du un programa de entrenamiento (accessed 12.12.2011).
en mindfulness (conciencia plena). Revista Española de Geriatría y NICE (2011). Generalised anxiety disorder and panic disorder (with or
Gerontología, 45, 252–8. without agoraphobia) in adults. Management in primary, secondary and
Gallagher-Thompson, D. and Thompson, L. (2010). Treating late-life community care. Clinical Guideline 113. <www.nice.org.uk> (accessed
depression: a cognitive-behavioral approach. Therapist guide. Oxford 12.12.2011).
University Press, New York. Oken, B. S., et al. (2010). Pilot controlled trial of mindfulness meditation
Garety, P., et al. (2001). A cognitive model of the positive symptoms of and education for dementia caregivers. Journal of Alternative and
psychosis. Psychological Medicine, 31, 189–95. Complementary Medicine, 16, 1031–8.
Gorenstein, E.E., et al. (2005). Cognitive-behavioral therapy for management Rovner, B., et al. (2007). Preventing depression in age-related macular
of anxiety and medication taper in older adults. American Journal of degeneration. Archives of General Psychiatry, 64, 886–92.
Psychiatry, 13, 901–9. Rybarczyk, B., et al. (2005). A placebo-controlled test of cogntive-behavioral
Granholm, E.S., et al. (2004). Group cognitive-behavioral social skills training therapy for comorbid insomnia in older adults. Journal of Consulting and
for older outpatients with chronic schizophrenia. Journal of Cognitive Clinical Psychology, 73, 1164–74.
Psychotherapy: An International Quarterly, 18, 265–79. Schuurmans, J., et al. (2006). A randomized, controlled trial of the
Improving Access to Psychological Therapies (2007). The competences effectiveness of cognitive-behavioral therapy and sertraline versus a
required to deliver effective cognitive and behavioural therapy for people waitlist control group for anxiety disorders in older adults. American
with depression and with anxiety disorders. Department of Health, Journal of Psychiatry, 14, 255–63.
Segal, Z.V., Williams, J.M.G., and Teasdale, J.D. (2002). Mindfulness-based Vernooij-Dassen, M., et al. (2011). Cognitive reframing for carers of people
cognitive therapy for depression: a new approach to preventing relapse. with dementia. Cochrane Library. <http://onlinelibrary.wiley.com/
Guilford, New York. doi/10.1002/14651858.CD005318.pub2/full> (accessed 30.12.2011).
Serfaty, M.A., et al. (2009). Clinical effectiveness of individual cognitive Wells, A. (1997). Cognitive therapy of anxiety disorders. A practice manual
behavioral therapy for depressed older people in primary care: a and conceptual guide. Wiley, Chichester.
randomized controlled trial. Archives of General Psychiatry, 66, Wilkinson, P. and Schuller, S. (2003). Late-onset auditory hallucinations
1332–40. treated with cognitive behaviour therapy. International Journal of
Silver, A., et al. (2004). Health anxiety. In: Bennett-Levy, J., et al. (eds) Oxford Geriatric Psychiatry, 18, 537–9.
guide to behavioural experiments in cognitive therapy, pp. 81–98. Oxford Wilkinson, P., et al. (2009). A pilot randomised controlled trial of a brief
University Press, Oxford. cognitive behavioural group intervention to reduce recurrence rates
Smith, A., Graham, L., and Senthinathan, S. (2007). Mindfulness-based in late life depression. International Journal of Geriatric Psychiatry, 24,
cognitive therapy for recurring depression in older people: a qualitative 68–75.
study. Aging and Mental Health, 11, 346–57. Wilson, K., Mottram, P.G., and Vassilas, C. (2008). Psychotherapeutic
Stanley, M.A., et al. (2009). Cognitive behavior therapy for generalized anxiety treatments for older depressed people. Cochrane Database of Systematic
disorder among older adults in primary care: a randomized clinical trial. Reviews, 1, CD004853. doi: 10.1002/14651858.CD004853.pub2. <www.
Journal of the American Medical Association, 301, 1460–7. thecochranelibrary.com/view/0/index.html> (accessed 12.12.2011).
Teri, L. and Gallagher-Thompson, D. (1991). Cognitive-behavioral Zijlstra, G.A.R., et al. (2009). Effects of a multicomponent cognitive
interventions for treatment of depression in Alzheimer’s patients. The behavioral group intervention on fear of falling and activity avoidance
Gerontologist, 31(3), 413–16. in community-dwelling older adults: results of a randomized controlled
Thompson, L.W. (1996). Cognitive-behavioral therapy and treatment for trial. Journal of the American Geriatrics Society, 57, 2020–8.
late-life depression. Journal of Clinical Psychiatry, 57, 29–37.
Thompson, L.W., et al. (2001). Comparison of desipramine and cognitive/ Further reading
behavioral therapy in the treatment of elderly outpatients with Gallagher-Thompson, D. and Thompson, L. (2010). Treating late-life
mild-to-moderate depression. American Journal of Geriatric Psychiatry, depression: a cognitive-behavioral approach. Therapist guide. Oxford
9, 225–40. University Press, New York.
Van’t Veer-Tazelaar, P.J., et al. (2009). Stepped-care prevention of anxiety and James, I.A. (2010). Cognitive behavioural therapy with older people.
depression in late life: a randomized controlled trial. Archives of General Interventions for those with and without dementia. Jessica Kingsley,
Psychiatry, 66, 297–304. London.
CHAPTER 18
Interpersonal psychotherapy
Philip Wilkinson
Interpersonal psychotherapy (IPT) is based on the straightforward treatment completers, IPT was also superior in improving social
observation that depression has a negative impact on the sufferer’s functioning.
relationships and that, in some instances, the origins of the depres- IPT was used as part of a care management intervention with
sion appear to lie within those relationships. The model does not depressed older adults in primary care that was shown to reduce
ignore the biological underpinnings of depression or the role of severity of depression and suicidal ideation when compared with
physical treatments, but it places the emphasis of intervention in usual care (Bruce et al., 2004). IPT has also been compared with
the interpersonal domain. Having initially been developed as a depression care management in the treatment of depressed older
pragmatic and structured form of therapy for depression treatment adults showing a partial response to the antidepressant escitalopram.
trials, IPT has since been applied in a range of settings for a number After 16 weeks’ treatment, IPT was not superior to the care manage-
of applications. ment intervention at inducing remission (Reynolds et al., 2010).
IPT is informed by attachment, communication, and social the- Two significant trials of IPT with older people have been in
ories. Early experiences and unmet attachment needs are seen as the maintenance phase of treatment for depression. These trials
important in the aetiology of depression, with episodes of illness (Maintenance Therapies for Late Life Depression; MTLLD I and
being triggered by social pressures and disruption of interpersonal MTLLD II) examined the effect of IPT with or without antidepres-
networks. Early practitioners of IPT were, therefore, quick to see sant treatment in participants who had recovered from an episode
the relevance of the model to late-life depression (Sholomskas et al., of depression while receiving a combination of IPT and antide-
1983). Interpersonal problems are classified into four categories: pressant. In MTLLD I (Reynolds et al., 1999), participants were
complicated bereavements, role transitions, interpersonal disputes, relatively young outpatients (60 years or over) who had suffered at
and interpersonal deficits. A role transition may be the loss of a least two episodes of depression. In maintenance treatment they
role, such as retirement from work, or the acquisition of a new role, received nortriptyline alone, drug placebo alone, IPT alone, or a
such as becoming a carer for a spouse with dementia. A role dis- combination of both treatments. Follow-up was at monthly inter-
pute may occur, for instance, when an older person dealing with ill vals for 3 years. The combination produced the lowest recurrence
health resists accepting the practical and emotional support offered rate on the Hamilton Rating Scale for Depression and was supe-
by an adult son or daughter. rior to drug placebo, IPT alone, and possibly to nortriptyline alone,
although this comparison did not achieve statistical significance.
IPT in the Treatment of Depression Participants aged 70 and over had the highest rate of recurrence
in the first year, and a secondary analysis suggested that the great-
Trials of IPT in depression est benefit of combined pharmacotherapy and psychotherapy in
There is good evidence for the efficacy of IPT as an acute treat- reducing recurrence was in this group.
ment of depression in younger adults. The UK National Institute In the MTLLD II study (Reynolds et al., 2006), participants were
for Health and Clinical Excellence (NICE) recommends IPT for the aged 70 years or over and had more medical illness than those
treatment of moderate to severe depression in combination with in MTLLDI. Some also had cognitive impairment and may have
antidepressant medication and, on its own, for the treatment of been developing dementia. Half of the participants were in their
persistent subthreshold depressive symptoms or mild to moderate first episode of depression. They were randomly assigned to par-
depression. Treatment duration should typically be in the range of oxetine alone, placebo alone, monthly IPT alone, or a combination
16–20 weekly sessions (NICE, 2010). of paroxetine and IPT. Follow-up was for 2 years. Randomization
IPT has been evaluated as a treatment for major depression in was stratified according to number of episodes of depression, use of
older adults in primary care (van Schaik et al., 2006). Ten ses- augmented pharmacotherapy in the acute phase, and degree of cog-
sions of IPT (without concurrent antidepressant treatment) were nitive impairment. In contrast to MTLLD I, combined treatment
found to be superior to general practitioners’ care at 6 months’ did not produce lower recurrence rates than antidepressant alone
follow-up in reducing the percentage of patients with a diagnosis and IPT was no more efficacious than drug placebo. The authors
of depression but not in inducing remission according to cut-off suggest that the cases of depression in the MTLLD II participants
on the Montgomery-Åsberg Depression Rating Scale. Among were more resistant to treatment due to cognitive impairment.
Content of IPT in the treatment of depression problems, and telephone top-up sessions may be beneficial (Miller,
The focus of IPT is current relationships in the outside world rather 2008).
than internalized past relationships or the relationship with the Further illustration of how each focus may be addressed is given
therapist. The aims of treatment are to improve these relationships in the following sections.
where possible or, if more appropriate, to modify the patient’s reli-
Grief and loss
ance on unhelpful relationships. Assessment focuses on the patient’s
styles of attachment and communication, beginning with a detailed Grief and loss are the reactions to the death of a significant person,
inventory of relationships. Relationships reviewed include not only or people, that give rise to depression. The therapist’s role is to facil-
family and social relationships but also those with other key peo- itate the grieving process by clarifying the nature of the relationship
ple, such as professionals. Education on the nature of depression is with the deceased and by helping the patient look for alternative
given, emphasizing that depression is an illness. In the initial stages supports and interests. With older adults, the grief reaction may
of therapy, patients are encouraged to adopt the sick role and to not be for someone who has actually died but for a spouse who has
accept the limitations imposed on them by their depression. become disabled or has dementia. In this case, the work may be
IPT follows a set structure, the plan and time frame being made to grieve for the loss of the healthy partner and to make necessary
explicit from the start (Fig. 18.1). Once the initial assessment ses- changes in the current social world.
sions are completed, one or two therapeutic foci are selected. The
foci chosen are likely to be those most associated with hopelessness
Role disputes
and suicidal risk. Foci most often selected with older people are role Role disputes are interpersonal struggles typically with the part-
transition and role dispute (Miller and Reynolds, 2002). ner or other family member. They may become apparent when
In the middle sessions of therapy the therapist uses a range older people suffer disabling physical illnesses. These disputes both
of techniques to promote change in the patient’s interpersonal contribute to depression and are made worse by the social with-
behaviour and to help solve problems related to the chosen foci drawal associated with depression. The therapist’s role is to help the
(Weissman et al., 2000). Then, in the concluding sessions, the ther- patient to understand his or her role in the dispute, link it with the
apist and patient review progress together by looking again at the depression, and then to take a problem-solving approach to chang-
interpersonal inventory and problem areas, before plans are made ing aspects of the relationship. It may be that the problem can be
for tackling problems that may occur in the future. There may then overcome if communications are improved, or in some cases it may
be maintenance sessions for ongoing work on existing interper- be that the relationship cannot be repaired. With older adults, it is
sonal problems or work on new problems. often preferable to try to bolster a relationship rather than risk a
Few adaptations are necessary when using IPT with older peo- significant loss.
ple without cognitive impairment. Hearing acuity and educational
status should be taken into account and sessions may need to be Role transition
shortened for medically ill patients. It may also help to involve the Role transition means a shift in an interpersonal role such as
spouse or any other significant person in evaluating interpersonal retirement or increased dependency on an adult son or daughter
after the onset of chronic illness. The therapist helps the patient to
understand the link between the role transition and the depres-
Assessment Appropriateness for IPT
sion and to look for positive as well as negative aspects of the
Initial sessions Sessions 1−2 transition in order for the patient to develop a greater sense of
Interpersonal formulation mastery in the new role. During IPT with older adults, an ini-
Diagnosis tial focus of role transition often resolves, only to reveal more
Therapeutic contract
Interpersonal Inventory long-standing role disputes, often with the spouse (Miller and
Reynolds, 2002).
Middle sessions Sessions 3−12
Interpersonal work Interpersonal deficits
Grief and loss Specific techniques
Interpersonal disputes Interpersonal deficits (or interpersonal sensitivity) is the least vali-
Role transitions dated of the four problem areas in IPT and so is usually avoided as
Interpersonal sensitivity the therapeutic focus if possible. It implies a long-term difficulty in
forming interpersonal relationships and may be accompanied by
Conclusion of Sessions 13−14
a poor interpersonal network. This category will include patients
acute treatment Separation reponses with personality disorder but could also embrace those with
Review of progress chronic low-grade mood disorder. As with other problem areas in
Contingency planning IPT, treatment does not focus on the early origins of the deficits but
on ways of reducing social isolation, which may include identifying
Maintenance Sessions 15+ and reversing unhelpful patterns of relating, or improving assert-
Maintenance contract iveness skills.
Prevention of relapse
Therapeutic Techniques in IPT

Fig. 18.1 The structure of interpersonal therapy.
(Reproduced by permission of Hodder Education from Stuart, S. and Robertson, Within the theoretical model of IPT, various therapeutic techniques
M. (2003). Interpersonal Psychotherapy. A Clinician’s Guide. London: Arnold.) can be employed to tackle the identified problems. These include
CHAPTER 18 interpersonal psychotherapy 245
education on the nature of depression, facilitation of expression Maintenance IPT

of mood, problem solving, and cognitive strategies aimed at the
patient’s attitudes to roles and relationships. Sometimes directive When IPT is used as a maintenance therapy in depression, as in
and practical interventions may also be useful, such as help with the MTLLD studies, therapy sessions may be less frequent and
transport or finances. spaced over weeks or months. The therapist aims to identify factors
involved in the recurrence of depression including social deficits
and long-standing patterns of interpersonal behaviour (Frank et al.,
Case example The following case example demonstrates the 1993). Therapy is likely to involve a greater number of foci than
use of IPT with an older adult (taken from Miller and Reynolds, acute phase treatment and to include a plan for identifying the early
2002, with permission of Brunner-Routledge, publisher). warning signs of future episodes of depression. When patients have
Mrs Jackson, a 65-year-old, white, married female, presented been depressed for long periods, they need help in adapting to being
in her fourth episode of major depression, never having had any a well person again, which in itself constitutes a role transition.
previous experience in psychotherapy. She presented with sev-
eral psychosocial stressors, most prominently recent retirement. IPT in Depression with Cognitive
She was extremely anxious and guarded at the onset of therapy,
and reported an almost complete remission of depressive symp- Impairment
toms in the first week. Within several weeks, however, her symp- The MTLLD II study, in which patients with both depressive dis-
toms returned to their original severity. Mrs Jackson was quite order and cognitive impairment did not benefit from maintenance
anxious and had a difficult time engaging actively in therapy. IPT, has led to a number of modifications of IPT for this group
After an initial cautious start, the educational component of (Miller and Reynolds, 2007). Dealing with role transition may need
IPT appeared to pay off, and she began to engage more actively. a modified approach to take account of the impact of the cognitive
Gradually, she began talking about her difficulties adjusting to impairment (Table 18.1). Family members may be invited to joint
retirement. These difficulties included time management, learn- sessions to foster better communication and to help to prevent car-
ing to manage money, and setting boundaries on her availability egivers from misconstruing as deliberate acts the problem behav-
for baby-sitting her grandchildren. The first five to eight sessions iours that result from the cognitive impairment. Family members
focused on these role transition issues. themselves may need support with the role transition involved in
Once Mrs Jackson began to feel somewhat better and as her trust becoming caregivers.
in her therapist deepened, she began to reveal more deep-seated
resentments and conflicts toward her husband. She requested Group IPT
that the focus shift away from her problems with retirement onto
her conflicts with him. She stated that she was now willing to IPT can also be provided in group format. Before group sessions
examine her own feelings and behaviours and work on ways to begin, the patient receives two individual sessions for assessment
try to make life better with her husband. She grappled with many and preparation for the group. An advantage of the group format
long-standing conflicts with him. Each situation that she brought is that it provides an ‘interpersonal laboratory’ in which interper-
to light manifested an underlying imbalance of power and con- sonal functioning can be observed and new ways of relating to
trol. Mrs Jackson described her husband as a benign dictator, other people tried out (Wilfley et al., 2000). Psychoeducational
but a dictator nonetheless. In exploring these issues in IPT, she aspects of treatment are strengthened and participants can join one
concluded that she must try to speak up more, and to be clearer another in applying therapeutic techniques such as decision analy-
about her needs. Initially, this created more conflict as well as sis (Scocco et al., 2002), while the forum can provide practical and
intense internal dissonance. She eventually began to recognize empathic interactions between group members (Roberston, 2004).
her own responsibility in allowing her husband to control even
minor decisions, and she recognized it was not easy for her to
assert herself. As her IPT therapist encouraged her to try alter- Table 18.1 Comparing the traditional IPT approach to role transition
with the approach taken in IPT-ci (IPT for cognitive impairment)
native strategies, she was both surprised and delighted to find
that her husband was more willing to share in decision-making Traditional IPT IPT-ci
than she thought possible. With practice, she eventually became
more comfortable in this newly acquired role. Through her active Help the patient to Help the patient to accept the lost role
accept the lost role
participation in IPT, Mrs Jackson made healthy adjustments to
retirement, and reduced the role disputes that chronically char- Help the patient to Remind the patient of abilities that remain
acterized her marriage. The depression that had resulted from explore positive aspects intact that could be further developed or
feeling hopelessly stuck was resolved. of the new role enhanced to help to compensate for lost
In this case the patient was experiencing a role dispute with her abilities
husband. If irresolvable, such a situation can undermine self-esteem Help the patient to Help the patient to foster new attachments
and result in depression. In this case the patient was willing to try acquire new skills to commensurate with his or her current abilities
to renegotiate the role by first examining her own contribution to meet the challenges of and, when necessary, help the patient to accept
the imbalance of power. The therapist used a range of techniques the new role increased dependency on others
with Mrs Jackson including detailed clarification of the problem (Taken with kind permission of Springer Science and Business Media B.V. from: Miller, M.D.
and communication analysis to identify communication failures (2008). Using interpersonal therapy (IPT) with older adults today and tomorrow: a review
of the literature and new developments. Current Psychiatry Reports, 10, 16–22; table 2,
and to devise more effective ways of communicating. p 19.)
It may be difficult, however, to maintain an adequate focus on each Cyranowski, J.M., et al. (2005). Interpersonal psychotherapy for depression
participant’s chosen problem area, so occasional individual sessions with panic spectrum symptoms: a pilot study. Depression and Anxiety,
may be required. 21, 140–2.
Frank, E., et al. (1993). Interpersonal psychotherapy in the treatment of
late-life depression. In: Klerman, G.L. and Weissman, M.M. (eds)
IPT in the Treatment of Grief New applications of interpersonal psychotherapy. American Psychiatric
Publishing, Washington, DC.
Although there is no consensus about when grief becomes a clinical
Lenze, E.J., et al. (2003). Good treatment outcomes in late-life depression
problem requiring treatment, it is generally accepted that features with comorbid anxiety. Journal of Affective Disorders, 77, 247–54.
warranting intervention include delayed onset of grief, persist- Miller, M.D. (2008). Using interpersonal therapy (IPT) with older adults
ent symptoms of depression, preoccupation with thoughts of the today and tomorrow: a review of the literature and new developments.
deceased, intrusive thoughts related to the death, and persistent Current Psychiatry Reports, 10, 16–22.
avoidance of reminders of the loss. Severe grief reactions might be Miller, M.D. and Reynolds, C.F. (2002). Interpersonal psychotherapy. In:
precipitated by the loss of apparently unimportant figures because Hepple, J., Pearce, J., and Wilkinson, P. (eds) Psychological therapies with
of unexpressed grief relating to losses much earlier in life (Frank older people, pp. 103–27. Brunner-Routledge, Hove.
et al., 1993). Miller, M.D. and Reynolds, C.F. (2007). Expanding the usefulness
Drawing upon attachment and communication theories, IPT of interpersonal psychotherapy (IPT) for depressed elders with
co-morbid cognitive impairment. International Journal of Geriatric
aims to help the grieving patient to relinquish the attachment to Psychiatry, 22, 101–5.
the lost attachment figure, to communicate the loss to others, and NICE (2010). The treatment and management of depression in adults (updated
to develop new attachments (Stuart and Robertson, 2003). Grief version). National Clinical Practice Guideline 90. <www.nice.org.uk>
is sometimes chosen as a focus in IPT in situations where the (accessed 12.12.2011).
object of attachment has not actually died but has become una- Reynolds, C.F., et al. (1999). Nortriptyline and interpersonal psychotherapy
vailable through separation or illness. This situation might apply, as maintenance therapies for recurrent major depression: a randomized
for instance, to older people caring for a spouse after a stroke or controlled trial in patients older than 59 years. Journal of the American
impaired by dementia. In this situation, the aim of IPT is to help the Medical Association, 281, 39–45.
patient maximize the remaining positive aspects of the relationship, Reynolds, C.F., et al. (2006). Maintenance treatment of major depression in
old age. New England Journal of Medicine, 354, 1130–8.
grieve for what has been lost, and to give support in making the
Reynolds, C.F., et al. (2010). Treating depression to remission in older adults:
most of the social network. a controlled evaluation of combined escitalopram with interpersonal
psychotherapy versus escitalopram with depression care management.
IPT in the Treatment of Anxiety International Journal of Geriatric Psychiatry, 25, 1134–41.
Roberston, M., et al. (2004). Group-based interpersonal psychotherapy
Depressed patients with comorbid anxiety have often been thought for posttraumatic stress disorder: theoretical and clinical aspects.
to be less likely than nonanxious patients to respond to IPT, International Journal of Group Psychotherapy, 52(4), 145–75.
although a secondary analysis of data from the MTLLD II study Scocco, P., De Leo, D., and Frank, E. (2002). Is interpersonal psychotherapy
demonstrated that, generally, comorbid anxiety did not affect treat- in group format a therapeutic option in late-life depression? Clinical
ment outcomes (Lenze et al., 2003). When anxiety is present along- Psychology and Psychotherapy, 9, 68–75.
side depression, some authors suggest certain modifications to Sholomskas, A.J., et al. (1983). Short-term interpersonal therapy (IPT) with
IPT such as providing an explanation of anxiety and its impact on the depressed elderly: case reports and discussion. American Journal of
Psychotherapy, 37, 552–66.
relationships and the incorporation of cognitive behavioural strate-
Stuart, S. and Roberston, M. (2003). Interpersonal psychotherapy. a clinician’s
gies (Cyranowski, 2005). In the last few years, IPT has also been guide. Arnold, London.
used in the management of primary anxiety disorders, including Van Schaik, A., et al. (2006). Interpersonal psychotherapy for elderly patients
post-traumatic stress disorder and social phobia. Intuitively, IPT in primary care. American Journal of Geriatric Psychiatry, 14, 777–86.
should help to overcome some of the social effects of anxiety, such Weissman, M.M., et al. (2000). Comprehensive guide to interpersonal
as avoidance and interpersonal sensitivity, and help older adults psychotherapy. Basic Books, New York.
when anxiety disorders are triggered by retirement or traumatic Wilfley, D.E., et al. (2000). Interpersonal psychotherapy for groups. Basic
events (Robertson et al., 2004). Books, New York.
References Further reading

Bruce, M.L., et al. (2004). Reducing suicidal ideation and depressive symptoms Hinrichsen, G.A. and Clougherty, K.F. (2006). Interpersonal psychotherapy
in depressed older primary care patients. A randomized controlled trial. for depressed older adults. American Psychological Association,
Journal of the American Medical Association, 291, 1081–91. Washington, DC.
CHAPTER 19
Psychodynamic
psychotherapy
Jane Garner
Man can change and go on changing as long as he lives.

(Karen Horney, 1950)
A psychodynamic approach emphasizes the uniqueness of now been extended by later analysts to include the whole lifespan.
each patient. It is based on a psychoanalytic approach which According to Freud, our mental health results from the interac-
has its roots in Freud’s theories and human concepts of devel- tion between early childhood experience and innate aspects of our
opment, relationships, and experience. Psychoanalysis is both a internal world. We are unaware of most mental activities—we put
technique of investigation of mental processes and a theory of ambivalent or hostile feelings, guilt, and anxiety into this uncon-
treatment, although explanation and cure do not inevitably go scious area of emotional life. It is a complex area of wishes and
hand in hand. It is an everyday experience in clinical life that hostility: not only are some of these feelings unconscious, but also
treatment plans do not necessarily work out as we had hoped or they are actively withheld from consciousness. The conscious part
expected. operates on the ‘reality principle’ and the dynamically unconscious
For various reasons, such as negative social attitudes towards part is motivated by basic instinctual urges, the ‘pleasure princi-
old age and a tendency to regard older people’s problems as ple’. Reality orientation is turned off in sleep and a wish-fulfilling
inevitable and irreversible, older people are less likely to be mechanism pervades our dreams, not always obvious or mani-
offered psychological treatments than their younger counter- fest but latent and requiring understanding. Our internal world is
parts. These negative stereotypes are sometimes held by those peopled by important figures (objects) from the past that we have
employed to provide health and social services (Garner and incorporated into our psyche. They are from the past but interact
Ardern, 1998). Age is seen as inevitably associated with men- with the present and may be in conflict with our current needs or
tal deterioration, and so psychological treatments are ruled out. wishes. Our symptoms and personality difficulties have meaning
It is of course true that changes occur with ageing, e.g. slowing that is hidden from us. The relationship with a professional sensi-
at cognitive tasks requiring rote learning, speed, and energy. tive to these ideas may be not only therapeutic but also diagnostic,
However, in other areas, where knowledge and experience of through feelings that are communicated to the therapist.
the world can be used, older people do better. There is some There is no reason why these assumptions should not apply what-
decrease in the capacity of working memory; however, despite ever the person’s age. They are certainly relevant to the mental state of
that, emotion-related information is recalled more often by older older people. However, Freud (1905) expressed a different opinion:
than by younger adults (Carstensen and Turk-Charles, 1994). The age of the patients has this much importance in determining their
In many respects older people are well placed to benefit from fitness for psychoanalytic treatment, that, on the one hand, near or
psychotherapeutic approaches, and this is also borne out by the above the age of 50 the elasticity of the mental processes, on which
evidence that exists. the treatment depends is as a rule lacking—old people are no longer
This chapter presents the assumptions on which dynamic psy- educable—and on the other hand, the mass of material to be dealt with
chotherapy is based and questions the neglect of older people in would prolong the duration of the treatment indefinitely.
this area of clinical thinking. It notes the similarities and differences Hildebrand (1982) pointed out the irony that Freud was 49 when
between this type of work with older and younger patients. It is he wrote this and launched his exciting new metapsychology into
suggested that a psychotherapeutic approach to care can promote the world. Freud, who saw the feminine as an absence of a penis
more interest in psychological aspects of older patients’ lives, expe- and age as the absence of youth, was no doubt firmly embedded in
riences, and relationships. the culture of his time (Woodward, 1991) and its attitudes. Actually
he bore his own old age with stoicism, dominated as it was by a
Theoretical Context painful carcinoma of the jaw.
Freud was a neurologist who anticipated that eventually there
Freud and others would be reconciliation of neurology and psychology. There is recent
Psychoanalytic theory is a theory of development, originally set out scientific progress in this borderland between disciplines; there are
by Freud, whose theory ended in early adulthood. This work has neurobiological foundations to psychotherapy. For example, the
right orbital prefrontal cortex has been suggested (Schore, 1997) static. There remains a life-long fluctuation between a predomi-
as the interface between biology and psychoanalysis. Psychological nantly selfish and self-serving attitude and one of generosity and
treatments change functional imaging (mainly frontal), with the concern. Under intensified anxiety, one’s attitude may slip back
degree of change related to degree of clinical change. Psychotherapy in these ever-shifting mental states. An older adult may be well
also has a significant effect on serotonin metabolism (Viinamäki placed to get a balance and be more predominantly in the con-
et al., 1998), increasing the number of serotonin receptors (Karlsson, cerned, depressive position, not just through therapy but as a con-
2011). Attachment styles have physical sequelae and neural corre- sequence of experience and dealings with the world. Our ability to
lates (Ciechanowski et al., 2001; Buchheim et al., 2011). Freud’s view be alone with ourselves, and to cope with the loneliness of old age,
of the organization of mind seems to be valid in the light of modern will depend on whether inside we feel predominantly persecutory,
neuroscience, although the language used is completely different critical, and aggressive, or loving, supportive, and forgiving (Hess,
(Solms, 2004). Although the idea of unconscious mental life was ini- 2004).
tially rejected as illogical, neurobiologists are now more accepting. There are numerous instances of how the work of other theo-
For example, patients with bilateral occipital lobe damage are blind rists, who may not themselves have specifically addressed issues
in their conscious life and will say ‘I don’t know’ when asked to guess of ageing, has been extended by those working with older people.
where a light is, but nevertheless guess accurately. Being conscious For example, O’Connor (1993) and Evans (2004) have used the self
of seeing is not necessary for visual processing. psychology of Heinz Kohut to elucidate the vulnerabilities of age.
After Freud, however, many analysts have not adhered to the Waddell (2007) used ideas from Bion to understand dementia and
notion of inflexibility and ineducability of older people. (This of Lipinska (2009) did the same with techniques from Rogers. Miesen
course accords with the personal experience of those of us who are (1993) employed Bowlby’s attachment theory, originally devel-
over 50!) Karl Abraham (1919) wrote that the age of the neurosis oped considering small children in hospital, to understand some
was more important than the age of the patient. For him, the ‘older behaviour problems in older people with dementia. Terry (2006)
patient’ was in his or her 30s or 40s, but nevertheless this idea of drew on the work of Melanie Klein to understand ‘terrors’ of old
chronicity remains useful. age and anticipation of death, and also the way carers are made
Carl Jung was the first adult developmentalist. He saw personality to feel. Hess (2004) uses the same to emphasize the importance
development and an unfolding programme of universal archetypes of loneliness in later life and the narcissistic tyranny some people
continuing throughout life. ‘The afternoon of life is just as full of wield in old age. Psychoanalytic training and research has long
meaning as the morning; only its meaning and purpose are differ- used mother and infant observation for training psychotherapists,
ent’ (Jung, 1953). Especially after his split with Freud, Jung went on observing an infant weekly from birth to 2 years old (Bick, 1964).
to think more carefully about the second half of life and its possibil- This technique can be extended to older people in geriatric wards,
ities (Bacelle, 2004). The discipline he founded, ‘analytical psychol- and residential and nursing homes to enhance understanding of the
ogy’, was to be an extension of, not an alternative to, psychoanalysis emotional states of institutionalized populations (McKenzie-Smith,
(Bacelle, 2004). Jung (1931) drew an analogy between life and 1992; Davenhill et al., 2007). Psychodynamic observation opens the
the daily course of the sun. In the first half, the rays are extended possibility of thoughtful intervention, and the integration of psy-
outwards—the young person’s life is mounting and unfolding with chotherapeutic theories and techniques expands the boundaries of
aims of nature, propagation of children, and entrenchment in the traditional psychodynamic approaches.
world. In the afternoon, the second half of life, the rays are drawn
back. For the older person, this is for self-illumination and to give Erikson and others
serious attention to the self. He stressed that this attention was not Freud’s ‘structural’ model of the mind focused primarily on the
to turn the older person into a niggardly hypochondriac, but for internal world. The so-called Neo-Freudians restated psychoana-
a focus on culture and spirituality. The psychological shift is thus lytic theory in sociological terms, putting greater focus on the exter-
from developing and asserting the personality to assessing the nal world with a more interpersonal (as opposed to intrapersonal)
meaning and quality of life. The earlier task of conforming and of model. From this tradition, Erik Erikson, a professor of human
being a collective person is superseded in the afternoon by the task development at Harvard University, supplemented and extended
of ‘individuation’ of understanding and identifying who you are, Freud’s ideas beyond early adulthood. Development was seen to
considering the past, present, and future—the future where you can continue beyond libidinal development throughout the life cycle,
be the individual that you are, not alienated from your self, and not and included cultural as well as intrapsychic factors. He described
rushing around working and doing. Individuation is the goal of life ‘Eight Ages of Man’, with specific polarities to be negotiated at each
and therapy. (Erikson, 1959, 1966) (Table 19.1).
Jung (1933) also described the worldwide archetype of ‘Senex’, The life cycle is charted through psychoanalysis and social sci-
the old man who has seen enough to be content, representing wise ence. Although there is a sequence of stages, this model also reflects
acceptance but also a wish to go on learning more. Senex needs gradual development. Each stage indicates the conflict to be nego-
Puer (the inner child) in order to move on, and the young person tiated, the phase-specific task required in order to have a reason-
has a notion of Senex within. Puer and Senex are potential arche- able balance: balance, not achievement or failure. For each conflict,
types for us all. There is a need to synthesize these and other dispa- there are potential negative outcomes, which are a dynamic coun-
rate aspects of the self. terpart to successful resolution and may be stimulated by difficul-
Melanie Klein is not seen as an adult developmentalist. She ties and stresses. The solution to each of these developmental crises
described (Klein, 1946) the shift from the paranoid schizoid is carried forward to the next and each is dependent upon the solu-
position to the depressive position in which the infant is able to tion to earlier ones, throughout life. The child is not only father
recognize good and bad in the same parental figures. This is not to the man but also grandfather to the old man. Erikson writes of
CHAPTER 19 psychodynamic psychotherapy 249
Table 19.1 The eight ages of man other older people, misanthropic, and contemptuous. In old age, the
despair may be manifest as anger, jealousy, hatred, and bitterness: ‘If
Age Conflict Resolution Culmination in that is it, that is all, then it is nothing’. Ego integrity involves a prefer-
old age
ence for order and meaning, a sense of world order and spirituality.
Infancy Basic trust Hope Appreciation of The older person will not only be involved in the integrity–
(0–1) versus interdependence despair contention but also be concerned in all the previous life
basic mistrust and relatedness stage issues. Perhaps the most potent of the life stages for the older
Early childhood Autonomy Will Acceptance of the
person fearing dependency is the first one, when the infant negoti-
versus cycle of life, from ates basic trust versus mistrust. Will mother and food be reliably
(1–3)
shame and integration to available for the baby, not in terms of quantity but of quality? In
doubt disintegration later life, this becomes a question of whether to trust others or not,
of whether they will be sufficient for one’s needs. Fear and terror of
Play age Initiative versus Purpose Humour, empathy, dependency in old age can be a reflection of earlier needs not hav-
(3–6) guilt resilience
ing been met. If basic care needs and dependency come to the fore
School age Industry versus Competence Humility, acceptance during this time of life, they may have profound effects. Someone
(6–12) inferiority of life course and lacking inner security as a consequence of inadequate earlier expe-
unfulfilled hopes riences may present with anxiety, panic, and depression, which may
Adolescence Identity versus Fidelity Sense of complexity be accompanied by fear that their carers may hate them or be dis-
(12–19) role confusion of life gusted by them (Martindale, 1989). Patients with problems nego-
tiating dependency are common in clinical practice—both those
Early adulthood Intimacy versus Love Sense of complexity
who could apparently do more for themselves but complain that
(20–25) isolation of relationships,
nothing is done for them, and also those who, despite clear evi-
value of tenderness
and loving freely dence to the contrary, claim they can manage perfectly well without
‘interference’. In contrast, if the initial experience was good enough,
Adulthood Generativity Care Caring for others, an older person may accept with reasonable equanimity the pros-
(25–64) versus empathy, and
pect of being dependent in the future.
Stagnation concern
Other analysts have taken up the theme of psychological devel-
Old age Ego integrity Wisdom Existential opment during adulthood. For example, Colarusso and Nemiroff
(65–death) versus identity; integrity (1981) presented a series of hypotheses on adult development
Despair strong enough (Table 19.2). The sixth hypothesis refers to how bodily changes in
to bear physical adulthood can influence psychological development. The body–
disintegration
mind relationship is emphasized in work with children but rela-
(Erikson 1959, 1966.) tively ignored in therapy with adults. Both patient and therapist will
be aware of changes in physical function and appearance due to
‘crisis’, but he considers the conflict at each stage to be a ‘decisive ageing. The altered representation of body image due to involution
encounter’ with the environment. They are particular moments in or physical illness induces shifts in self-representation and feelings
psychosocial growth, turning points at tasks that may be healthily of identity. The patient may avoid examining biological realities
surmounted and mastered or reacted to in a self-alienating way to and the mind–body relationship because of the narcissistic injury
portend regression and retardation. involved in ageing; the same may be true for the therapist, but it is
The first five stages are the familiar developmental phases important not to ignore this dimension.
described by Freud. The final three stages chart our course in adult- In a later work, Nemiroff and Colarusso (1985) discussed ‘devel-
hood beyond libidinal development. Intimacy versus isolation is opmental resonance’. This proposes that, as we all have the same life
young adulthood, eager for partnership and intimacy. Avoiding it
may presage isolation. Generativity versus stagnation is the concern
for guiding the next generation, including non-genital creativity. Table 19.2 Seven hypotheses for adult development
The counterpart is interpersonal impoverishment and stagnation.
1. The nature of developmental processes is the same in the adult and
It is surprising that Erikson has been criticized for ageism on the
the child.
basis that he leaves the existential question to the end. Actually, he
2. Development during the adult years is an ongoing dynamic process.
stated that each phase is related to all others, every strength and
weakness has its precursors or derivatives in each stage. Ego integrity 3. Adult development is concerned with the continuing evolution and use of
psychic structure rather than its formation.
versus despair is the final stage of development. There is no clear
definition of ego integrity, but one who possesses it accepts that his 4. The fundamental issues of childhood continue in altered form as central
aspects of adult life.
life was his own responsibility. He accepts his family of origin, his
position in world affairs, and his personal life as it has been, as the 5. Development processes are influenced by the adult past, not only by
childhood.
only one he could have had in the context within which he lived.
Without integrity, he will be beset by despair that life is too short and 6. Bodily changes influence development both in childhood and adult life.
that death is too near, leaving no time for an alternative course to be 7. A central theme is the growing awareness of the finiteness of time and
taken. Rather than valuing ‘having been’, he will fear and regret the inevitability of death, including one’s own.
proximity of ‘not being’. He may be disgusted by his older self and (Colarusso and Nemiroff, 1981.)
Table 19.3 Developmental tasks and difficulties in later life Creativity may continue undiminished into old age, and there
are numerous well-known examples, e.g. Hokusai, Picasso, Goya,
◆ Fear of diminution/loss of sexual potency
Matisse, and Verdi. Others may discover abilities previously
◆ Threat of redundancy in work roles; being replaced by younger people unrealized. However, Jung (1933: 127) tells us: ‘We must not for-
◆ The need to reconsider and possibly remake the marital relationship after get that only a very few people are artists in life; that the art of
children have left life is the most distinguished and rarest of all the arts’. Ageing
◆ Awareness of one’s own ageing, illness, and possible dependence itself for those with sufficient internal resources can be a crea-
◆ Awareness that what one can achieve now is limited tive process. Thus negative experiences can be a possible source
◆ The feeling of having failed as a parent (paradoxically exacerbated in the of growth to strengthen the individual (Salzberger-Wittenberg,
childless) 1970) and creativity can emerge from loss (Storr, 1988). Even
◆ Loss of a partner and of intimacy though dementia is scarcely something to be desired, it has the
◆ The fact of one’s own death in terms of narcissistic loss and pain potential to reveal previously hidden feelings and to reconcile
relationships.
(Hildebrand, 1982.)
Retirement from paid work is a significant milestone: ‘laboro
ergo sum’. For some, work may be a denial of ageing and death; and
development framework, even when the therapist is much younger identity, self-esteem, and a sense of social inclusion may be tied up
than the patient, both can have an understanding of all phases of the with working life. However, for others, casting off the shackles of
patient’s life. The therapist can draw on his or her own experience work comes as great release and they are able to use the increased
from earlier stages and from the seeds of phases to come. Nonetheless, time available to realize new skills and put energy into making and
counter transference issues are affected by the fact of the therapist keeping relationships.
being younger and not having direct personal experience of old age One of the universal activities of later life is life review. As we age,
themselves. The therapist will be the object of transferential feelings we begin to appreciate within ourselves the processes of develop-
from all periods of the patient’s life cycle, so it is important that this ment and changes that have occurred over the life cycle. We under-
is covered in training for work with older patients (Pinquart and stand that the present is meaningful because of its connections with
Sörensen, 2001). the past. Sometimes, a midlife awareness of mortality may prompt
Another useful model is that of Hildebrand (1982), who super- an appraisal of life, or else retirement may encourage a period of
vised and facilitated a ‘revolutionary’ workshop at the Tavistock self-reflection. For Schmid (1990), lifespan review is a powerful
Clinic for clinicians seeing older patients. He drew on the work of psychodynamic issue that dominates old age. This is not simply
Erikson and of Pearl King (1974, 1980) and delineated particular for reminiscence or the retrieval of facts, but life review involves
tasks and difficulties which need to be negotiated for a content later evaluation and resolution. It aims to put life into perspective for
life (Table 19.3). Throughout life, disturbances in earlier stages can self-discovery, to achieve a sense of completeness, of connected-
produce psychopathological change later. Issues previously dormant ness and coherence over the lifespan. The process is both intra- and
may re-emerge with force in later years. Sandler (1978) wrote of interpersonal (Molinari, 1999). One reviews the past in order to
‘the senescent adolescent’ for whom previous conflicts needed to be understand the present. This is a normal developmental task in
reworked. Some apparently ‘younger’ behaviours may not be due to a later life: in part, the propensity for life review is brought about by
re-emergence of a previous problem but represent a current conflict an awareness of the finiteness of life and approaching death. All
about getting older being unbearable. For example, episodes of sex- therapy involves a degree of life review, and this is especially so with
ual promiscuity may be an attempt to prove that youth and potency older patients.
are still there. In working with older people, we should maintain a It is easy to perceive of old age as a time of loss and decline, but
balanced awareness between developmental issues specific to later a more constructive alternative view is the maturity model (Knight,
life and what has re-emerged from earlier times (Terry, 1997). 1992), which emphasizes positive aspects rather than charting defi-
cits. According to this model, healthy people feel more comfortable
Positive and Negative Aspects of Ageing with themselves as they age. They have learned from observations
in life, they have experience and knowledge of human interactions,
Positive aspects of ageing, development, life review, and they may have expertise in different areas, including employ-
and maturity ment, social, and family contexts. Freud described being mature as
A model of ageing based on impairment ignores the many aspects being able to love and to work. For Klein, as mentioned, it was an
of later life that are significantly more positive and may even bring increased capacity to live in the depressive position; and for Bion
advantages (Garner, 2009). Older people are not a homogeneous (1970), maturity was being able to keep developing. Waddell (1998)
mass, as might be thought from reading newspapers and policy wrote:
documents. Indeed, there is more diversity, more individuality in The difference between maturity and immaturity hinges not on the
older rather than younger citizens. They have anything up to nine fact of chronological years but on a person’s capacity to bear intense
decades of personal experience and the patterns of their lives have emotional states; on the extent to which it is possible to think about
been highly varied, whether for better or worse. Psychological and reflect on psychic pain as a consequence of having found and sus-
changes are not merely as a result of a decline in cognition, but tained a relationship with external and internal figures who are able
many skills are stable across the life-course. Over the years, we so to do.
have refined the strengths in our character and developed skills for Perhaps a state of maturity confers a sense of inner calm, free
coping with a variety of challenges. We have increased capacity for from basic drives competing for behavioural expression. For exam-
delayed gratification and for procrastinating less. ple, maturity could be seen as the capacity to recognize and enjoy
sexuality and sexual attraction without necessarily needing to act the prevailing zeitgeist and by seeing the most impaired people in
it out. Maturity is about having gained experience, not pining after our wards and clinics. It is easy to adopt an impairment-disease
lost youth or compulsive attempts to stay young. Danielle Quinodoz approach, the idea that old age is a loss, the deficit of youth, health,
(2009) acknowledges the negativity and pejorative tone associated cognition, etc., and therefore to focus on pharmacological inter-
with all expressions indicating old age and reflects that her own ventions for both physical and mental health problems. Davenhill
book, Growing Old, has an awful title. It is possible to shift the (2007) writes of the unconscious factors involved in the silencing of
emphasis by adding a preposition: Growing into Old Age (Garner, debate regarding long-term care for people in later life, and about
2004a) suggests a more productive, developmental, active, not pas- the dissolution of local authority and NHS provision in this area so
sive process into maturity. Jung (1933: 118) writes: that care has been shifted into the fragmented business sector. The
Society does not value these feats of the psyche very highly; its prizes ruling market hegemony where human need is seen as despicable
are always given for achievement and not for personality—the latter dependence and where ‘competition and survival ennoble the soul’
being rewarded for the most part posthumously. (Bell, 1996) only serves to increase the vulnerability of patients
who are doubly stigmatized as a consequence of being mentally
Prejudice as a barrier to treatment ill and old (Arden et al., 1998). Improving Access to Psychological
The misery of a child is interesting to a mother, the misery of a young Therapies (IAPT, 2008) is an initiative from the Department of
man is interesting to a young woman, the misery of an old man is Health to be warmly welcomed, but so far only a small minority of
interesting to nobody. patients referred are over 65 years. IAPT has so far really failed the
Victor Hugo, Les Misérables (1862)
needs of older people.
To Dorian Gray, in Oscar Wilde’s story, the idea of getting old was Many older people feel a discrepancy between their own experi-
such an anathema that he split off his ageing self and projected it ence of their life and self as basically continuous and the loss of social
onto the hated and feared portrait kept out of sight. The images identity they suffer in old age. This diminishes older people in the
we have of old age, carried in our conscious and unconscious lives, view of society and also in their own eyes, especially if they internal-
are for the most part negative (Garner and Ardern, 1998). They are ize these negative attitudes. An older person is part of society, so he
reflected culturally in the depiction of later life in art and litera- is aware of our fears and stereotyping, and to some extent he shares
ture. For example, in the Allegory of the Four Seasons by the Italian them. He does not wish to join a club ‘full of old people’ or be admit-
painter Bartolomeo Manfredi (circa 1610), Autumn is a virile, fer- ted to a ‘geriatric’ ward. Maintaining integrity or sense of self despite
tile male who has produce of sweet and luscious fruits; he is kiss- the negative images of ageing projected onto the old is a major task
ing Spring and embracing Summer who are both beautiful women. in old age (Evans, 1998). The situation may change as people from
Winter is an old man, excluded and shivering outside the group. more recent birth cohorts with a greater sense of entitlement reach
Another example is the psychodynamic literature commenting on old age and assert themselves more actively (Patrick, 2006).
the tale of Little Red Riding Hood (Zipes, 1993), where nobody This model of ageing parallels Freud’s emotional-depletion view.
seems concerned about the fate of the older woman. Although both The loss-deficit idea characterizes old age as loss of job, status, and
the young girl and her grandmother were devoured by the wolf, all finances, decreased social engagement and loneliness, poor health,
thought and sympathy is with the girl (Garner and Bacelle, 2004). disability, dependence, loss of cognitive skills, bereavement, and
Zoja (1983), a Jungian analyst with a background in sociology, death. These aspects are important in later life, but they do not
has commented on our productivistic, positivistic, entrepreneurial apply to everyone or only apply partially. These malignant percep-
society where what is most valued is young, active, and mascu- tions are mirrored in the way older people see themselves, so few
line, and how we view being old merely as the absence of youth. seek psychotherapy; and also by professionals, so few people are
Although this is a feature of postmodern western society, it may offered the treatment.
well have preceded that time. We live in a youth-centred culture The deficit approach, which views ageing as a narcissistic wound
and define old age in the negative. Certainly, we praise old people that cannot be healed, is unlikely to encourage therapists to take
not for ageing well but for seeming younger than they are (even on older patients, and it overemphasizes the difference between
if that is due to the skill of the plastic surgeon). Some of the tra- work with younger and older adults. Psychotherapists prefer to
ditional roles of older people have been taken over: for example, see younger rather than middle-aged and then middle-aged rather
wisdom by professionals, and storytelling by the media. Memories than older patients (Zivian et al., 1992). This is partly because of
in some cultures are now stored in computers. And, for all that, we limited experience and knowledge of psychotherapy in later life
can compile a long list of thoughtful, skilled, creative people who in and also because of erroneous perceptions of prognosis (therapeu-
later years have continued to contribute actively in all sorts of ways, tic nihilism). Therapists express personal anxiety about growing
but it is probably not completely helpful to look at later life in an old, a reluctance to be associated with low-status patients, and con-
idealized way. To erect an iconography of age against which most cern at limited training opportunities. Although older people have
people would fail by comparison may only serve to compound feel- problems common to all age groups, outcome is enhanced if the
ings of inadequacy. therapist not only is generally well qualified but also has special-
There is a risk that we simply think of older people as a single ized training in working with older adults (Pinquart and Sörensen,
group rather than celebrating their individuality. Terms like the 2001). The therapists have to deal with their own socially deter-
‘rising tide’ or the ‘demographic time bomb’ promote a perception mined ageism as well as personal feelings; perhaps they have unre-
of old age as simply a burden on the younger taxpayer. This ageist solved issues (which may or may not have been recognized) with
atmosphere inevitably influences policymakers and those who pro- parents or grandparents. In the therapy, they will have to face ideas
vide services. Even old age psychiatrists, who in many ways have of illness, disability, dependency, death—not only the patient’s but
championed the cause of older patients, are probably influenced by also their own.
A further prejudice in this work is that associated with this par- enhanced sense of how any individual life is part of a larger histori-
ticular type of therapy—a notion that psychodynamic concepts and cal and cultural circumstance.
treatments lack empirical support and scientific evidence. Perhaps
dynamic therapists have not always been their own best champi- Assessment
ons, preferring the successes of single case studies rather than ran- Suitability for treatment at any age depends on psychological mind-
domized controlled trials which often do not assess the range of edness, motivation, ego strength, and the ability to enter and sus-
phenomena that can change in psychotherapy, from alleviation of tain an intensive therapeutic relationship. In public health services,
symptoms to the development of new personal capacities. the problems presented are likely to be multiple and complex, so
the therapist should engage a flexible, thoughtful attitude to refer-
rals and avoid having too many exclusion criteria.
Themes and Special Considerations in For a brief approach, patient and therapist need to adhere to a
Psychotherapeutic Work with Older suitable focus for therapy and accept it will end at the agreed time.
Patients In all cases, there is a need to manage the expectations of change
alongside realistic goals. Younger patients may be able to exercise
Loss and mourning what they have learned about themselves by adopting a new course,
The reasons for older people coming to psychotherapy will be many lifestyle, or behaviour. Due to the constraints of the external world
and varied, but certain general themes, especially loss and unre- of the older patient, change may be more manifest in their inner
solved conflicts from the life stages discussed previously, will fea- world—bringing about a new relationship with internal objects so
ture strongly. Loss is inevitably a common theme, although not the that they can achieve a good quality of life, whatever their circum-
whole picture. We lose many things over the years—actual things; stances. Expectations of regaining youth or achieving happiness are
parents and siblings, spouse, and friends, employment, financial not realistic, but an acceptance of, and equanimity about, what is
status, strength and health, independence, home, and neighbours, still possible certainly can be achieved.
etc. The ability to face loss without feeling a sense of catastrophe Patients may feel undeserving of therapy because of their age, or
is founded in early childhood capacities to bear psychic reality. they may belong to a generation that lacks understanding of psy-
Therapy will need to recognize what is being experienced in rela- chotherapy, so time may need to be spent at the beginning engaging
tion to later life and what is being revived from an earlier time. them and discussing the rationale for treatment. If there is some
An ostensibly small loss may revive feelings from previous bereave- doubt on either side of the therapeutic relationship, it may be useful
ments. Losses are specific and particular for the person involved and to use a few sessions for assessment, seeing how the patient reacts to
will also include the loss of potential opportunities or attainments this unusual private space, the therapeutic relationship, and explo-
(‘I will now never be able to fulfil my dream of . . .’). Mourning these ration of the inner world.
losses and acknowledging the loss of the younger self can give a Several positive factors may enhance the ability of older people
sense of freedom for investment in the time that is to come. to engage in exploratory reflection. These may include, for example,
In order to move through the phases of life, we need to give up the idea of a last chance for understanding; calming of instinctual
and mourn previous ones. Mourning is a normal part of life, of life impulses may reduce the fearsome effect of internal objects; and the
review, and of therapy. If crises and losses, common in old age, can immediacy of some losses makes it harder to deny anxieties (King,
initiate a successful mourning process, then they do not have to 1974). Grotjahn (1955) pointed out that resistance to unpleas-
be psychologically incapacitating (Pollack, 1978). Pollack (1982) ant insight is lessened in old age as the demands of reality finally
writes of mourning liberation being the focus of work in later life; become acceptable.
liberation as it allows the past to appropriately become the past so The apparently simple act of taking a psychiatric history begins
that there may be a libidinal investment in the present and future. to help people shape a narrative of their life. Winnicott (1971)
Mourning is for the self, parts of which are lost and changed, for wrote about therapeutic consultation and, although his work was
others, and for unfulfilled hopes and aspirations. Suppressed with children, similar principles apply in work with adults. The aim
mourning will lead to psychological problems, such as depression, is to promote understanding of their successes and regrets in life,
anger, or hate. Appropriate sadness is a healthy way to review losses not what happened to them passively but the active role they played
in life and can lead to further development and creativity. in their own history. Telling the tale to an empathic listener may
Paradoxically, as one acknowledges loss, internal objects can encourage a reluctant patient to accept therapy.
be reclaimed. The natural losses occurring in treatment—timed Storytelling is important; a natural means to explain actions,
end of the session, breaks—can be used in therapy to aid mourn- thoughts, and feelings in narrative form. The past needs to be sur-
ing and adaptation. To be successful, psychotherapy involves loss veyed receptively, reconsidering past experiences and their meaning.
(Wolff, 1977): giving up maladaptive, erroneous, infantile wishes Giving new meaning to the old story may be important, especially
and behaviour. Moving through life, too, things need to be given if the tale reflects maladaptive perceptions. An assessment aims to
up. Successful ageing is the ability to mourn for the self, opening create the opportunity to reflect on the process of life review, as a
up possibilities and freedoms for the years to come. The losses and creative, reflective, and reframing activity, jointly undertaken. The
traumata of old age may spur someone with sufficient internal therapist will look at the possibilities for integrating unresolved
resources to further positive development. Settlage (1996) described developmental and relational aspects of experience.
therapeutic work with a centenarian where he felt development and In therapy it must be possible for any or all stories to be voiced,
creativity had been stimulated by the changes and losses of old age. not only those that show the patient in a socially acceptable light.
Therapy may lead to an expanded understanding and acceptance of Reflection on a life is about striving for new insights, themes, and
self and of death to come (Beadleson and Lara, 1988), as well as an meanings that link memories across the lifespan, giving a sense of
self that is continuous over time. The person may come to under- probably will not need to return. Managing endings is a significant
stand he/she is an ‘expert’ on the time in which he/she has lived. and difficult part of all therapy, especially if it is time limited, as is
Re-examining the personal narrative over time may give greater probable in a public heath service. The ending and the loss it repre-
capacity to accept the present and cope with potential future sents should be discussed from the beginning of therapy and linked
difficulties. with previous and anticipated losses. The patient may grieve and/
or be angry. Managing that anger will be helpful so that the positive
Practical considerations aspect of the therapy can still serve as an internal resource in years
Themes in psychological work with older people show some dif- to come.
ferences from work with younger adults and some modifications The prospect of impending death is a complex issue. Some peo-
to the therapeutic process may be necessary (Garner and Evans, ple start to acknowledge their own mortality in mid life (Jacques,
2010), although generalizations about older people should be 1965; Knight, 1986) so, for them, negotiating death is perhaps a
avoided, since there is great variation between individuals. It should less difficult task in later years. For many others though, death is
be borne in mind that all of us have basic psychological needs for a persecutory or depressive anxiety (Segal, 1958; Carvalho, 2008).
respect, security, and self-determination, at whatever age. People For Turner (1992), death is a longitudinal issue—if one fears death
should be treated with equality, that is, fairly, equitably, and impar- when young, one also fears it when old. Putting the problem on
tially; which is not to say they should all be treated identically. the ‘old’ could be seen as a projection by younger people of their
The importance of life review has already been mentioned. Butler fears and prejudices. Older people who have not been able to nego-
(1963) developed the life review technique based on the concepts tiate the integrity–despair stage will have fears of ‘not being’, and
from Erikson’s scheme. The idea of ego integrity, assessing one’s be unable to value ‘having been’. Bion (1962) writes of ‘the name-
experiences and circumstances, has a natural association with life less dread’: death or even worse—a personal feeling, an existential
review. Reviewing the past is not a direct, literal retrieval of memo- terror of cosmic loneliness in which there is no spouse, no parent,
ries, or a photographic construction of what happened. What we see no-one. External relationships counteract this, as do affective mem-
of our past is filtered through subsequent experience, personality, ories of relationships with good objects. The person may have con-
our current situation, our own theories of ageing, and the transfer- structed his or her life so as not to be exposed to these terrors, but
ence relationship with the therapist. In therapy, the therapist facili- with age and threat of dependency this defensive manoeuvre may
tates and enhances the personal review, to make it more conscious begin to disintegrate. Fears of falling to pieces, or of not being held,
and deliberate (Knight, 1992). The therapist will need to respond return. The therapist’s task is to understand the revival of infantile
not only to the weaknesses and difficulties but also to the strengths terrors without infantilizing the older adult (Terry, 2006). Shame
of the older adult who has lived and coped over many decades of and humiliation accompany the fears of loneliness, desperation for
history and problems. an intimate relationship, but inability to achieve it. Klein (1963)
In working with older patients, the therapeutic emphasis on inte- describes the dreaded loneliness in which one feels left alone with
riority needs to expand to include reality, both biological and social. the bad parts of the self, the unintegrated parts of the self. The anxi-
For many people, advancing years bring multiple and chronic health eties of age are not so different from earlier, universal anxieties—
problems. Handicaps may accumulate, engendering pain and dis- fear of loss, abandonment, and loss of autonomy—but the realities
ability that are unlikely to be alleviated by the exploratory approach of later life may enhance a person’s fears.
of psychotherapy. One of the skills in working with this patient Dementia is not an inevitable part of old age, but it does affect
group is to understand and acknowledge this biological reality, 25% of the over 80s and is a significant cause of illness and dis-
while incorporating it into the therapy along with its symbolism ability. People with a diagnosis of dementia can also benefit from
and meaning, the transference, and the intrapsychic world. Also, a psychodynamic approach, although they are neglected in that
the physical reality of someone’s health may need to be accommo- respect even more than other old people. Not everything that hap-
dated in the therapy setting. For example, is there a reliable lift or pens to a person with dementia is organically determined. The
should a room be available on the ground floor? If the patient needs illness trajectory is not solely determined by neurophysiology.
help to rise from the chair it would be insensitive not to offer it, but For example, emotional memory is a midbrain function (Van der
at the same time the therapist should understand that the patient Kolk and McFarlane, 1996) rather than cortical and so is retained
may find the helping hand either intrusive or comforting. The room for much longer. Ability to make relationships persists. And even
needs to be a neutral space protected from outside distraction, with though verbal skills and conscious memories may diminish, the
the patient free to bring any thoughts, feelings, and fantasies, and emphasis of psychoanalytic understanding on the unconscious and
the therapist must be open to whatever is brought. on nonverbal communication makes it a particularly useful para-
Time in all aspects is an issue in work with older patients. Some digm to address these difficulties (Sinason, 1992; Garner, 2004b;
patients work well and quickly, recognizing that time is now short. Duffy, 2006; Evans, 2008). Working in this way may make it easier
for patients to accept other aspects of care that they were resisting.
Depend upon it sir, when a man knows he is to be hanged in a fort- Patients may find some alleviation of terror, a sense of containment
night, it concentrates his mind wonderfully.
and continuity for a more peaceful end.
(Boswell, Life of Johnson, 1777)
Families may support and help facilitate the therapy, e.g. they
Other patients, perhaps linking discharge with death, need may bring the patient to the session; however, it could also be
to know that the therapist, or the institution where they work, asked whether this is truly helpful or if patients would feel they had
will be available to them forever (until their death) (King, 1980; more psychological space or more autonomy if they came to the
Martindale, 1989). It may be better to terminate gradually or let appointment by public or hospital transport. On the other hand,
patients know they may come back—with that reassurance, they families may disapprove and be resistant, they may resent or envy
the relationship developed between the patient and therapist. With the patient. Good supervision to explore these counter transferential
ageing of the older generation, power relations within the family feelings is essential in psychotherapy, and probably in any work with
will change and dynamics shift. The once powerful dictator may older patients.
fall, as described so eloquently by Shakespeare in King Lear (Hess, Interpretation of the transference leads to insight, which in turn
1987). Families often express the view that they have the right to leads to improved interpersonal function over time. This aspect of
know what is being discussed in the sessions, a form of infantiliza- psychoanalytic theory has recently been supported by evidence from
tion that reflects society’s ageism. Nonetheless, the rules of confi- a randomized clinical trial (Johansson et al., 2010). Patients who
dentiality apply whatever the age of the patient. If resources allow, it found transference interpretations particularly useful were those
may be appropriate to allocate a different member of staff to answer with personality disorder pathologies and more severe and chronic
the family’s concerns. difficulties in establishing stable and fulfilling relationships.
The relationship between the patient and therapist is also a real
Transference and counter transference one and the most significant agent in therapy. The therapeutic alli-
Understanding the concepts of transference and counter transfer- ance is the sine qua non of effective therapy; it is the collaborative
ence is helpful in any interaction with any patient. It is the cor- working relationship that keeps the treatment going, whatever the
nerstone of psychodynamic work—how these are manifest is the emotional ups and downs of the transference relationship. The
guide to the patient’s unconscious. Memories are not constituted of alliance forms from initial encounters and is then facilitated by
a series of facts: it is not necessary to reconstruct a lifetime of his- reciprocal experiences. Previous poor attachment experiences can
tory, and what is not recalled will be available and experienced in be modified. One aim of the therapeutic alliance is to provide the
the relationship with the therapist. patient with a different, improved attachment experience.
The type of transference has more to do with relationship experi-
ences across the decades rather than chronological age. It is more Effectiveness
varied with an older patient and can span generations. The thera- Although, as mentioned earlier, there have not been many
pist may represent parent or grandparent, child or grandchild, col- high-quality trials involving psychodynamic psychotherapy, the sit-
league, contemporary, partner, or lover. This will change throughout uation has improved in recent years. Shedler (2010) has compared
the work and is a rich source of information about the patient. An effect sizes from meta-analyses of recent treatment of outcome stud-
idealized transference is likely to cover underlying hostility—per- ies (different therapies and antidepressant medication). Available
haps in these circumstances an envy of youth or holidays, or an evidence indicates that effect sizes for dynamic psychotherapy are
imagined wonderful sexual life. as large as those reported for other treatments. In addition, when
Working with patients who will inevitably be older, possibly longer-term follow-up is included, effect sizes are larger, suggesting
much older, maybe with physical problems or a dementing illness, that the dynamic therapy sets in motion psychological processes
evokes particular feelings in staff, as they may see the patient as a that lead to ongoing change, even after the ending of therapy.
grandparent, parent, or themselves in old age. Unhelpful sentiments There is also good evidence that older people have outcomes from
of pity may emerge or else sadism stimulated by an unequal power different kinds of psychotherapy—including psychodynamic—
relationship. Conversely, it is possible to idealize the older patient that are at least equal to those for younger patients (Knight, 1996;
as the wished-for parent or grandparent and to fear failure before Woods and Roth, 1996). Indirect measures such as staff turnover
this older figure. A dependent patient may be feared and rejected, and instance of abuse may be useful outcome measures to assess
as the patient’s feelings of helplessness projected onto the therapist quality of service delivery and the benefit of using the skills of a
may leave the therapist feeling impotent, paralysed, or infused with psychoanalytic psychotherapist (Garner, 2002).
infantile wishes to dominate and control the ‘parent’. The therapist
may be frightened of imagined escalating demands and depend- Using psychodynamic ideas in the old age
ency of the patient—the paradox of being controlled by the help- psychiatry service
less (Martindale, 1989; Terry, 2006). It is anxiety provoking for a All recent reports on services emphasize the need for accessibil-
therapist of any age to be confronting difficult issues of death and ity and equity across patient groups (e.g. Age Concern, 2007).
dependency; it will evoke fears about one’s own ageing and eventual Although there is good evidence for the effectiveness of psychoana-
death and perhaps a wish to inappropriately discharge the patient lytic work with older adults, resource and prejudicial constraints
who has projected unwelcome dependency. suggest that few patients will be taken on for individual or group
Professionals may have difficulty working with this group of work. However, the principles and understanding derived from the
patients, feeling that rather than being able to be, they feel compelled theory and practice of this type of therapy may be used to inform the
to do, to act, as a doctor, nurse, social worker, etc., and then they may service as a whole. Caring for dependent older people is demand-
feel guilt at not doing. Issues related to sexuality may be pressing for ing and stressful. A psychodynamic perspective may be used as an
the patient and ignored by the therapist who finds it easier to think adjunct to biological, social, and other psychological approaches,
of old age as lacking sexuality, a view that may derive from the thera- providing opportunities for containment and reflection, under-
pist’s own oedipal reactions to parents but that is also prevalent in standing the fears of both the patients and the staff who care for
society. The therapist may find an eroticized transference from this them (Wesby, 2004; Garner, 2008). Our healthcare institutions
supposedly sexless older person difficult to acknowledge. Counter appear to be constructed as a form of social defence—to contain
transference is a tool in therapy, not a problem, but it needs to be experiences of doubt, uncertainty, anxiety, and guilt (Menzies-Lyth,
recognized, accepted, and understood. It is only problematic when 1987). Caregiving tasks need to be turned into opportunities to
the therapist reacts unthinkingly and unknowingly in response to interact psychologically, with benefit not only to patients but also
to staff, who are likely then to experience less dissatisfaction and Bick, E. (1964). Notes on infant observation in psychoanalytical training.
‘burnout’. Staff feel better about their work if relationships are more International Journal of Psychoanalysis, 45, 558–66.
meaningful. Bion, W.R. (1962). The psychoanalytic study of thinking. International
Journal of Psychoanalysis, 43, 306–10.
The counter transferential feelings discussed with reference to the
Bion, W. (1970). Attention and interpretation. Tavistock, London.
therapist will also affect other staff, who need training and supervi-
Buchheim, A., et al. (2011). Neural correlates of emotion, cognition, and
sion incorporating these ideas. Strong feelings can be engendered attachment in borderline personality disorder and its clinical implications.
by projections from psychotic, borderline, and dependent patients. In: Levy, R.A., Ablon, J.S., and Kächele, H. (eds) Psychodynamic
Staff also need to tolerate the sadness and regret some patients feel psychotherapy research: practice based evidence and evidence based
at the end of their lives and the hostility and guilt of some family practice, pp. 239–56. Humana Press, New York.
members. Butler, R.N. (1963). The life review: an interpretation of reminiscence in the
Psychotherapeutic skills relevant to old age psychiatry include the aged. Psychiatry, 26, 65–76.
ability to listen and empathize, showing openness to the patient’s Carstensen, L.L. and Turk-Charles, S. (1994). The salience of emotion across
emotions, being able to make sense of the patient’s experience, using the adult life span. Psychology and Aging, 9, 259–64.
Carvalho, R. (2008). The final challenge: ageing, dying, individuation. Journal
one’s personal emotional response as a source of understanding,
of Analytical Psychology, 53, 1–18.
and being able to communicate such understanding to the patient.
Ciechanowski, P.S., et al. (2001). The patient provider relationship: attachment
Staff should be able to contain anxiety and despair, rather than theory and adherence to treatment in diabetes. American Journal of
being compelled to act, while using realistic judgement to decide Psychiatry, 158, 29–35.
when action is necessary. We need to be able to identify idealization Colarusso, G.A. and Nemiroff, R.A. (1981). Adult development: a new
and a distorted perception of the staff/patient relationship, and this dimension in psychodynamic theory and practice. Plenum, New York.
can enable us to bear hostility and criticism without retaliation. If Davenhill, R. (2007) No truce with the furies: issues of containment in the
a service is able to use these ideas, it is likely there will also be less provision of care for people with dementia and those who care for them.
absenteeism, fewer complaints, and fewer serious incidents. In: Davenhill, R. (ed.) Looking into later life: a psychoanalytic approach to
depression and dementia in old age, pp. 201–21. Karnac, London.
Davenhill, R., Balfour, A., and Rustin, M. (2007). Psychodynamic observation
Conclusion and old age. In: Davenhill, R. (ed.) Looking into later life: a psychoanalytic
Psychoanalytic theory provides a framework in which to under- approach to depression and dementia in old age, pp. 129–44. Karnac,
stand human situations, patients’ symptoms and diagnoses, the London.
reactions of their families, and the fears and anxieties of staff. It Duffy, M. (2006) Psychotherapeutic interventions for older persons with
emphasizes the uniqueness of individuals. Unfortunately, nega- dementing disorders. In: Brody, C.M. and Semel, V.G. (eds) Strategies for
therapy with the elderly, pp. 15–30. Springer, New York.
tive attitudes to old age and therapeutic possibilities in later life
Erikson, E. (1959). Identity and the life cycle. Psychological Issues Monograph
decrease the likelihood of patients’ psychological needs being ade-
1. International Universities Press, New York.
quately met. For those who are able to access psychotherapeutic Erikson, E. (1966). Eight ages of man. International Journal of Psychiatry, 2,
help the outcome is comparable to, and sometimes better than, that 281–307.
for younger patients. Evans, S. (1998). Beyond the mirror: a group analytic exploration of late life
Development continues from birth to death through differ- and depression. Ageing and Mental Health, 2, 92–3.
ent phases. Everyone is engaged in a life review, which becomes Evans, S. (2004). The old self: Kohut, Winnicott and others. In: Evans, S.
more intense as we age. We all consider our lives, our relation- and Garner, J. (eds) Talking over the years: a handbook of dynamic
ships, our time in history, and this may be aided by a therapist psychotherapy with older adults, pp. 57–69. Brunner Routledge, Hove.
sensitive to these ideas. The therapist needs to be open to reflect- Evans, S. (2008). Beyond forgetfulness’: how psychoanalytic ideas can help us
to understand the experience of patients with dementia. Psychoanalytic
ing on his or her own thoughts and prejudices about ageing.
Psychotherapy, 22, 155–76.
Psychodynamic ideas can usefully inform the old age psychiatry
Freud, S. (1905/1953). On psychotherapy. In: Strachey, J. (ed. and trans.) The
service to the benefit of patients, their families, and the staff who standard edition of the complete psychological works of Sigmund Freud.
work there. Vol. 6, pp. 249–63. Hogarth, London.
Garner, J. (2002). Psychodynamic work and older adults. Advances in
References Psychiatric Treatment, 8, 128–37.
Abraham, K. (1919/1988). The applicability of psychoanalytic treatment to Garner, J. (2004a). Growing into old age: Erikson and others. In: Evans, S.
patients at an advanced age. In: Bryan, A. and Strachey, A. (eds) Selected and Garner, J. (eds) Talking over the years: a handbook of dynamic
papers on psychoanalysis. Karnac, London. psychotherapy with older adults, pp. 71–86. Brunner Routledge, Hove.
Age Concern (2007). Improving services and support for older people with Garner, J. (2004b). Dementia. In: Evans, S. and Garner, J. (eds) Talking
mental health problems. Second report from the UK Inquiry into Mental over the years: a handbook of dynamic psychotherapy with older adults,
Health and Well-Being in Later Life. Age Concern England, London. pp. 215–30. Brunner Routledge, Hove.
Ardern, M., Garner, J., and Porter, R. (1998). Curious bedfellows: Garner, J. (2008). WOT! No psychotherapist. APP Newsletter. Psychoanalytic
psychoanalytic understanding and old age psychiatry. Psychoanalytic Psychotherapy, 40, 7–8.
Psychotherapy, 12, 47–56. Garner, J. (2009). Considerably better than the alternative: positive aspects of
Bacelle, L. (2004). On becoming an old man: Jung and others. In: Evans, S. getting older. Quality in Ageing, 10, 5–8.
and Garner, J. (eds) Talking over the years: a handbook of dynamic Garner, J. and Ardern, M. (1998). Reflections on old age. Ageing and Mental
psychotherapy with older adults, pp. 29–41. Routledge, Hove. Health. 2, 92–3.
Beadleson, B.M. and Lara, L.L. (1988). Reminiscence: nursing actions for the Garner, J. and Bacelle, L. (2004). Sexuality. In: Evans, S. and Garner, J. (eds)
acutely ill geriatric patient. Issues in Mental Health Nursing, 9, 83–94. Talking over the years: a handbook of dynamic psychotherapy with older
Bell, D. (1996). Primitive mind of state. Psychoanalytic Psychotherapy, adults, pp. 247–63. Brunner Routledge, Hove.
10, 45–7.
Garner, J. and Evans, S. (2010). Psychodynamic approaches to the challenges Molinari, V. (1999). Using reminiscence and life review as natural therapeutic
of ageing. In: Pachana, N.A., Laidlaw, K., and Knight, B.G. (eds) Casebook strategies in group therapy. In: Duffy, M. (ed.) Handbook of counseling
of clinical geropsychology: international perspectives on practice, pp. 55–72. and psychotherapy with older adults, pp. 154–65. Wiley, New York.
Oxford University Press, Oxford. Nemiroff, R.A. and Colarusso, G.A. (1985). The race against time: psychotherapy
Grotjahn, M. (1955). Analytic psychotherapy with the elderly. Psychoanalytic and psychoanalysis in the second half of life. Plenum, New York.
Review, 42, 419–27. O’Connor, D. (1993). The impact of dementia: a self psychological perspective
Hess, N. (1987). King Lear and some anxieties of old age. British Journal of Journal of Gerontological Social Work, 20, 113–28.
Medical Psychology, 60, 209–15. Patrick, E. (2006). Older, wiser and unlikely to present. Therapy Today, 17
Hess, N. (2004). Loneliness in old age: Klein and others. In: Evans, S. (3), 4–8.
and Garner, J. (eds) Talking over the years: a handbook of dynamic Pinquart, M. and Sörenson, S. (2001). How effective are psychotherapeutic
psychotherapy with older adults, pp. 19–27.. Brunner Routledge, Hove and other psychosocial interventions with older adults? A meta-analysis.
Hildebrand, P. (1982). Psychotherapy with older patients. British Journal of Journal of Mental Health and Ageing, 7, 207–43.
Medical Psychology, 55, 19–28. Pollack, G.H. (1978). Process and affect: mourning and grief. International
Horney, K.J. (1950). Neurosis and human growth: the struggle toward Journal of Psychoanalysis, 59, 255–76.
self-realization. Norton, New York. Pollack, G.H. (1982). On ageing and psychotherapy. International Journal of
Hugo, V. (1862/1998). Les Misérables. Penguin Popular Classics, London. Psychoanalysis, 63, 275–81.
Improving Access to Psychological Therapies (2008). The IAPT pathfinders: Quinodoz, D. (2009). Growing old: a journey of self-discovery. Routledge,
achievements and challenges. Department of Health, London. London.
<http://www.iapt.nhs.uk/silo/files/the-iapt-pathfinders-achievement Salzberger-Wittenberg, I. (1970). Psycho-analytic insight and relationships: a
s-and-challenges.pdf> (accessed 31.01.2012). Kleinian approach. Routledge, London.
Jacques, E. (1965). Death and the mid-life crisis. International Journal of Sandler, A.M. (1978). Psychoanalysis in later life: problems in the
Psychoanalysis, 46, 502–14. psychoanalysis of an aging narcissistic patient. Journal of Geriatric
Johansson, P., et al. (2010). The mediating role of insight for long-term Psychiatry, 11, 5–36.
improvements in psychodynamic therapy. Journal of Consulting and Schmid, A.H. (1990). Dementia, related disorders and old age: psychodynamic
Clinical Psychology, 78, 438–48. dimensions in diagnosis and treatment. American Journal of Psychoanalysis,
Jung, C.G. (1931/1979). The stages of life. In: Read, H., Fordham, M., and Adler, 50, 253–62.
G. (eds) Collected works, Vol. 8. Routledge and Kegan Paul, London. Schore, A.N. (1997). A century after Freud’s project: is a rapprochement
Jung, C.G. (1933/2001). Modern man in search of a soul. Routledge, Abingdon. between psychoanalysis and neurobiology at hand? Journal of American
Jung, C.G. (1953). Two essays on analytical psychology. In: Read, H., Psychoanalytic Association, 45, 807–40.
Fordham, M., and Adler, G. (eds) Collected works, Vol. 7, p. 74. Routledge Segal, H. (1958). Fear of death: notes on the analysis of an old man.
and Kegan Paul, London. International Journal of Psychoanalysis, 39, 173–81.
Karlsson, H. (2011). Psychotherapy increases the amount of serotonin Settlage, C.F. (1996). Transcending old age: creativity, development and
receptors in the brains of patients with major depressive disorder. In: Levy, psychoanalysis in the life of a centenarian. International Journal of
R.A., Ablon, J.S., and Kächele, H. (eds) Psychodynamic psychotherapy Psychoanalysis, 77, 549–64.
research: practice based evidence and evidence based practice, pp. 233–8. Sinason, V. (1992). The man who was losing his brain. In: Mental handicap
Humana Press, New York. and the human condition: new approaches from the Tavistock, pp. 87–100.
King, P. (1974). Notes on the psychoanalysis of older patients. Journal of Free Association Books, London.
Analytical Psychology, 19, 22–37. Shedler, J. (2010). The efficacy of psychodynamic psychotherapy. American
King, P. (1980). The life cycle as indicated by the nature of the transference in Psychologist, 65, 98–109.
the psychoanalysis of the middle aged and elderly. International Journal Solms, M. (2004). Freud returns. Scientific American, 290(5), 83–8.
of Psycho-Analysis, 61, 153–60. Storr, A (1988). Bereavement, depression and repair. In: Solitude: a return to
Klein, M. (1946/1975). Notes on some schizoid mechanisms. In: Envy and the self. Free Press, London.
gratitude and other works 1946–1963, pp. 1–24. Hogarth Press, London. Terry, P. (1997). Counselling the elderly and their carers. Macmillan, London.
Klein, M. (1963/1975). On the sense of loneliness. In: Envy and gratitude and Terry, P. (2006). Terrors of growing old. Therapy Today, 17 (3), 4–8.
other works 1946–1963, pp. 300–13. Hogarth Press, London. Turner M.S. (1992). Individual psychodynamic psychotherapy with older
Knight, B.G. (1986). Psychotherapy with older adults. Sage, California. adults: perspectives from a nurse psychotherapist. Archives of Psychiatric
Knight, B.G. (1992). Older adults in psychotherapy: case histories. Sage, Nursing, 6, 266–74.
California. Van der Kolk, B.A. and McFarlane, A.C. (1996). The black hole of trauma. In:
Knight, B.G. (1996). Psychodynamic therapy with older adults: lessons van der Kolk, B.A., McFarlane, A.C., and Weisaeth, L. (eds) Traumatic
from scientific gerontology. In: Woods, R. (ed.) Handbook of the clinical stress: the effects of overwhelming experience on mind, body and society,
psychology of ageing. Wiley, Chichester. pp. 2–23. Guilford Press, New York.
Lipinska, D. (2009). Person-centred counselling for people with dementia: Viinamäki, H., et al. (1998). Change in monoamine transporter density
making sense of self. Jessica Kingsley, London. related to clinical recovery: a case-control study. Nordic Journal of
Martindale, B. (1989). Becoming dependent again: the fears of some elderly Psychiatry, 52, 39–44.
persons and their younger therapists. Psychoanalytic Psychotherapy, 4, Waddell, M. (1998). Inside lives: psychoanalysis and the growth of the
56–75. personality. Duckworth, London.
McKenzie-Smith, S. (1992). A psychoanalytical observational study of the Waddell, M. (2007). Only connect—the links between early and later life. In:
elderly. Free Associations, 3, 355–90. Davenhill, R. (ed.) Looking into later life: a psychoanalytic approach to
Menzies-Lyth, I.E.P. (1987). Containing anxiety in institutions. In: Selected depression and dementia in old age, pp. 187–200. Karnac, London.
essays, Vol. 1. Free Association Books, London. Wesby, R. (2004). Inpatient dynamics. In: Evans, S. and Garner, J. (eds)
Miesen, B.M.L. (1993). Alzheimer’s disease, the phenomenon of parent Talking over the years: a handbook of dynamic psychotherapy with older
fixation and Bowlby’s attachment theory. International Journal of adults, pp. 101–16. Brunner Routledge, Hove.
Geriatric Psychiatry, 8, 147–53. Winnicott, D.W. (1971). Playing and reality. Tavistock, London.
Wolff, H.H. (1977). Loss: a central theme in psychotherapy. British Journal of Zipes, J. (1993). The trials and tribulations of Little Red Riding Hood.
Psychology, 50, 11–19. Routledge, New York and London.
Woods, R.T. and Roth, A. (1996). Effectiveness of psychological interventions Zivian, M.T., et al. (1992). Psychotherapy for the elderly: psychotherapists’
with older people. In: Roth, A.D. and Fonagy, P. (eds) What works preferences. Psychotherapy, 29, 668–74.
for whom? A critical review of psychotherapy research, 2nd edition, Zoja, L. (1983). Working against Dorian Gray: analysis and the old. Journal of
pp. 425–46. Guildford, New York. Analytical Psychology, 28, 51–64.
Woodward, K. (1991). Aging and its discontents: Freud and other fictions.
Indiana University Press, Indianapolis.
CHAPTER 20
Family therapy
Jane Pearce
Family therapy is a psychological therapy frequently delivered What Is the Theory Underpinning
under the label of systemic therapy. The underlying principles are
drawn from understanding the family as a system and the focus of Systemic Therapy?
therapy is to explore problems in relational terms rather than to Communication cybernetics
treat them as residing in the patient alone. Early theoretical underpinning of systemic therapy was derived
Family therapy has a history of ongoing development of from communications research (Watzlawick, 1967), general sys-
theoretical ideas and clinical practices. This offers a diverse tems theory, and cybernetics (von Bertalanffy, 1968). The key idea
range of approaches (‘lenses’), summarized in Table 20.1, which is that cycles of feedback continuously create and recreate a basis for
can be used to tackle difficulties involving more complex situ- interaction, and hence for evolving relationships. Communication
ations in which mental health problems are presenting and through behaviour, expectations, and beliefs about behaviour, and
evolving. Whilst techniques were developed working within the reactions to behaviour, can be seen as ongoing circular processes.
family as part of child and adolescent services, they have been In these circular processes it is hard to distinguish the causal and
applied to a range of alternative ‘quasi-family’ constellations consequential events, since the consequences of one event may be
(including care networks and care settings) as well as to couples causal for another. Where a number of people are involved it is
(sometimes termed marital or couple therapy). For this reason arbitrary whether to draw the conclusion that something is a cause
we will use the more general term ‘systemic therapy’ throughout rather than a consequence, since this will depend upon the particu-
this chapter. This potential for application in a range of complex lar time at which the sequence of interactions is examined. These
settings was an initial challenge to research, but the evidence ongoing circular processes may often maintain the predictability of
base for systemic therapy has been building for the treatment ‘how things go on’ in that particular system. Interactions may be
of depressive disorders, schizophrenia, and bipolar disorder in shaped not only by the current context but also by what has gone on
adults, although the evidence base for older adults is not yet as before, how relationships have been conducted in the past, and by
strong. patterns of previous behaviours and daily practices, and thus may
Old age psychiatry routinely involves interacting with peo- inhibit appropriate change.
ple beyond patients themselves, whether this is with family The capacity for change is required across the lifespan (Carter
members, members of care networks, or homes, and certainly and McGoldrick, 1989). Both predictable (e.g. leaving school, start-
with members of social services and other agencies. All of us, ing work, birth of new child, death of old parent) and unpredict-
to a greater or lesser extent, have learned a range of ways to be able events (divorce, job loss, migration) will challenge a family as
effective in the wide variety of settings and predicaments within members move through the life-course from birth to death. Where
which we manage common psychiatric disorders. What is the events are ‘in phase’ (e.g. death of an older husband), they may be
value added of the family or systemic approach in everyday easier to cope with than when the events are ‘out of phase’, such as
work? This chapter will begin by summarizing the key theo- the death of a teenage granddaughter. The ways in which events and
retical ideas underpinning systemic therapy. Next, evidence on transitions in life have been negotiated at previous life stages may
the effectiveness of clinical application of systemic theory and influence those that come subsequently (Walsh, 1989). For exam-
therapy in adults is reviewed. The chapter will then use four ple, the progress made with separation and individuation during
clinical practice vignettes to illustrate the ‘systemic method’ in adolescence may have resulted in limited autonomy or some degree
action with older people. The settings for the vignettes include of enmeshment between family members. This may have been rela-
inpatient, community work, and the memory clinic. Common tively stable during mid life, masked by geographical distancing or
themes in the practice of old age psychiatry are touched upon, emotional detachment. However, dependency of the older person
including loss through bereavement and cognitive impairment, can unmask conflicts and anxieties.
and ways in which the past family history impacts upon caring A second major influence in systemic practice has been a recog-
abilities of children for their aging parents, in intractable or stuck nition of the limitation of hypotheses and formulation of problems
predicaments.
reliant upon observation of phenomena ‘from the outside’, as in Systemic Therapies

the standard scientific paradigm. With increasing understanding
of the ‘inside’ of a family, therapy teams came to see that their ini- Systemic therapy has evolved a range of ways of addressing the pre-
tial hypotheses may not always be helpful in facilitating change sented problems and working with them. The historical context of
(Selvini-Palazzoli et al., 1980). A further hypothesis would there- systemic therapy is of diversity of approaches developed over the
fore have to be developed using the ‘new information’ that had last 80 years. A wide variety of systemic models provides a selection
arisen whilst exploring the previous one. This entails the start of of approaches within which relational implications can be explored.
a circular endeavour, a continual process of developing and revis- These models differ with respect to their theoretical assumptions,
ing formulations based on information arising during therapy the central focus of therapy, and the interventions found to be most
(Selvini-Palazzoli et al., 1980) and ‘creative curiosity’ (Cecchin, suited to bring about change. There are, however, many similarities
1987). and overlaps between the different approaches, and often the dif-
ferences are primarily ones of emphasis and of terminology. Some
Postmodernism and social constructionism approaches are in a sense more ‘postmodern’ or ‘social construc-
tionist’ and place a greater emphasis on the meanings, stories, and
The third and major influence on systemic theory has been that of
beliefs that families have developed to make sense of their expe-
social constructionism, with its interest in the central position of
riences and patterns of interaction, while other approaches are
language and culture for meaning, values, and identity. Social con-
more behavioural, focusing directly on the patterns of interaction.
structionism refers to a constellation of theoretical approaches
However, this is not a hard and fast distinction and most approaches
with an underlying assumption that human beings construct their
take some account of patterns of communication and the meanings
‘social worlds’ through language. Cycles of communicative interac-
favoured by family members.
tions between people create and shape the meanings attributed to
There is also some variation between therapies in the extent to
their actions, choices, and responses.
which the formulation is made by the therapist or is evolved within
In everyday language, the words we have chosen create an inter-
the therapy (Dallos, 2006), although there is generally less dis-
pretation. We do not just describe the world through language but
tinction between assessment, formulation, and intervention than
use language to actively make sense of it and ‘construct’ it. ‘When
with other (nonsystemic) therapies. Particularly in postmodern
people talk to each other the world gets constructed’ (Burr, 1995).
approaches, the formulation may be jointly constructed between the
At the most general level, the dominant ideas of the time, the ‘dis-
therapist and family (McNamee and Gergen, 1992). Family thera-
courses’ of the culture and its current customs or norms, shape
pists may discuss their ideas with a family in the form of ‘reflective
the beliefs and expectations of any community, organization, or
conversations’ (Andersen, 1987). The therapy team might join the
nation.
family and discuss their ideas and formulations with family mem-
Social constructionism also further challenges the possibility
bers, who might also be invited to observe the therapists’ discussion
of objective observation upon which family therapists had tradi-
of these ideas and have their own conversation about what fits for
tionally relied (Gergen, 1994), since the therapist’s own culture
them or what they are interested in (Andersen, 1987).
and belief systems (both personal and professional) would surely
Table 20.1 summarizes the key features of some of the main
influence his or her account of the family system. Within a con-
approaches. (This is not an exhaustive list and a number of other
structionist perspective, one can reconceptualize the therapist’s
subtypes exist.)
role as that of collaborator with the family rather than as an outside
Some therapies, particularly ones where practitioners have been
observer, with further implications for the accounts that can be cre-
oriented towards formal empirical evaluation, have leant towards
ated within the family system.
a manualized treatment. One such example is the strategic-related
An alternative theory of problem development is also implicit
McMaster family therapy which employs a structured,
in the social constructionist perspective. Human problems can be
problem-centred, task-focused approach to help replace problem-
seen to arise and/or be maintained by the stories that dominate
atic interaction patterns and improve communication, problem
family life and in particular by ‘oppressive’ ones (White and Epston,
solving, and emotional connectedness (Ryan, 2005). Emotionally
1990). Thus the narratives told about an individual, a family, or
focused couple or family therapy has also been manualized. This
other system are the ways in which identity, lives, and problems are
therapy is based theoretically on an understanding of attachment
constituted, rather than vice versa.
within the relationship and postulates that, where this is insecure,
so will the relationship be affected by anxiety. Treatment is manual-
In summary
ized and the therapeutic goal is to help partners to develop ways
Systemic therapy focuses on an interactional understanding of to meet one another’s attachment needs in a secure way (Johnson,
wellbeing. It offers a theoretical basis on which to observe and 2004).
describe how a problem exists within its context. Dynamic,
interactive connections between people and their wider social
context are sought. Patterns in the relationships between the indi- Evidence Base for Systemic Therapy
viduals connected with a problem, their behaviours and beliefs, Reviews have suggested that a wide range of applications of sys-
and the broader collective social, political, and religious inter- temic therapy provide effective help across the age span (Asen,
pretations and practices can be explored. On this basis, systemic 2002b; Stratton, 2005; Carr, 2009). However, evidence from ran-
therapy offers a relational, dynamic, and contextual approach to domized controlled trials of systemic interventions with adults,
treating dilemmas and problems in families and quasi-family especially older adults, is as yet limited. Problems in the assessment
systems. of systemic therapy include the wide span of problems to which
CHAPTER 20 family therapy 261
Table 20.1 Key features of systemic family therapies

Family therapy model Central concern of the therapy Main theoretical ideas Distinctive features of therapist
activity
Structural family therapy Aims to challenge any The well-functioning family is seen to Therapist makes active interventions
problem-maintaining patterns of have a structure with clear hierarchies to challenge interactions between
family interaction or behaviours between generations, distinct spheres for generations and communications
communication, and distinct roles across related to the problem
the generations (Minuchin, 1974)
Solution-focused therapy Finding positives and building on Problem-saturated ways of talking can Therapist encourages the discussion of
them dominate, such that small positives get moments and times when things went a
missed when, as will happen from time to little better than usual. These exceptions
time, problems are slightly less severe or are used to devise ways of interacting, so
troubling than usual (de Shazer, 1985) that the positives occur more frequently
Strategic family therapy Helping the family to develop new Family problems may serve an important Therapist identifies interactions around
patterns of relationships that allow interpersonal function for some family the presenting problem and goals for
healthier relationships members change, and designs interventions
The family’s communication and to disrupt problematic patterns.
organizational patterns need to be Interventions can include reframing the
challenged (Haley, 1977) function of symptoms and paradoxical
interventions
MRI brief therapy Improve flexibility in approach to A symptom may be maintained by the A strategic approach but with smaller
finding solutions for problems attempted solution (Watzlawick, 1967) number of sessions and more limited
and well-defined goals
Psychoeducational family Aims to reduce family stress and The course of a mental illness, such as Help families to understand the factors
therapy enhance family support schizophrenia and depression, will be that affect the aetiology and course of
affected by the balance between levels illness (including illness management);
of stress and levels of support in the develop communication and
immediate psychosocial environment. problem-solving strategies
Emotionally charged, critical, or May be delivered in multiple family
overinvolved relationships will increase groups (‘multifamily’) (Asen, 2002a)
the risk of relapse or poor illness
management (Falloon, 1993)
Milan family therapy Belief systems and narratives that Hypotheses about the ‘problem’ can Interventions might include circular
constrain families’ patterns of be formulated around a family’s unique questioning (see Box 20.2), positive
interaction and may maintain relationships, beliefs, and interaction reconnotation, and paradoxical
presenting problems patterns. Exploring these hypotheses interventions
could lead to ideas of how change might
become possible (Selvini-Palazzoli, 1980)
Narrative family therapy Aims to find alternative stories or Individuals and families make sense of Family and therapist work together to
‘narratives’ that families can use to what goes on in their lives through what find new ways to make sense of their
make sense of their lives they talk about and how they talk about experiences, to evolve new stories that
it. This can be limited by broader societal, are not so negative and limiting to illness
historical, interpersonal, and familial May externalize the problem and see it
influences. Problems could arise from as separate from the person, rather than
or be maintained by ‘oppressive stories’ within the person
about their life, even if this contradicts
their experience or potential (White and
Epston, 1980)
Transgenerational family Aims to avert repetition of Family members may not be sufficiently Genogram (see Box 20.1) used to identify
therapy problematic parental behaviours differentiated from their family of origin. multigenerational problem-maintaining
and relationships by uncovering the This might lead to either ‘cutting off ’ from patterns, and families’ emotional
patterns from earlier generations their own family of origin or justifying responses to handling these. Past
acting in the way they experienced being legacies may be unfair, but exoneration
treated. The therapy aims to avoid the of hurtful parental behaviours through
replication of emotional relationship and understanding behaviours in the
behaviour patterns from the family of previous generation may be helpful. Idea
origin (Bowen, 1978) of a ‘ledger’ of entitlement to fairness
and good treatment
treatment has been directed, the range of treatment goals and A meta-analysis of these studies included four behavioural systemic
type of interventions, the range of outcomes measured, and how interventions, two emotionally focused interventions, one interper-
well outcome measures are matched to what was attempted in the sonal therapy (IPT) based and one cognitive based. Although no
therapy (Sprenkle, 2003). Here we will consider the evidence in the clear differences for improvement in mild depression (compared
treatment of major mental health problems, namely affective disor- to pharmacotherapy and individually based therapy) were found,
der (depression, bipolar disorder) and schizophrenia in adults. there was evidence of improved relationships and that marital ther-
The strongest level of evidence for treatment benefits is the sys- apy may be more acceptable than pharmacological treatment.
tematic review, which provides methodology to review the good The findings of these two Cochrane reviews in depression are
quality randomized controlled trials (RCT) that are relevant to the consistent with conclusions drawn from previous reviews, which
question in hand (OCEBM Levels of Evidence Working Group, considered the full range of RCTs (e.g. those that were excluded
2011). Cochrane Systematic Review methodology is recognized because of smaller sample size, or including questions outside
as providing the highest level of evidence because of the stringent the scope of the Cochrane review, such as prevention of relapse)
evaluation of studies’ methodology before they are eligible to be of the effectiveness of a range of family systemic interventions in
included in the review and rigorous statistical approaches to com- both outpatient and inpatient settings (Asen, 2002b; Carr, 2009).
bining the findings from individual studies (<www.thecochraneli- These data are helpful because they provide additional (albeit
brary.com>). Cochrane reviews are also important for identifying lower-strength) findings that can help clinical decision-making.
gaps and areas where research is still required. Interventions included systemic couple therapy, manualized
It is in the treatment of schizophrenia that Cochrane review evi- problem-centred therapy (see McMaster family therapy in the
dence is strongest and family interventions have been found effec- section Systemic therapies), emotionally focused couple therapy,
tive (Pharoah et al., 2010). Fifty-three RCTs met Cochrane criteria, behavioural marital therapy (aims to improve communication,
although only 13 of these included the patient in the family interven- improve problem-solving, increase frequency of satisfying expe-
tion. Primary outcome measures included patient symptom ratings riences), cognitive marital therapy (adds in a cognitive compo-
scores, relapse rates, and scores of family function in some studies. nent to modify cognitive factors that maintain dysphoria and
The family interventions reviewed have been predominantly psych- marital distress), and conjoint interpersonal therapy (promoting
oeducational and were found to decrease the frequency of relapse understanding of the interpersonal context of depression and rene-
and possibly also to reduce hospital admission. Data from second- gotiation of role relations between the partners). One RCT study
ary outcome measures also suggested that systemic therapy might that selected subjects with evident negative expressed emotion
encourage compliance with medication, that it was acceptable to found that depressed subjects improved significantly with a sys-
patients and carers (the attrition rates being no higher in the systemic family or couple therapy compared to antidepressant medi-
temic therapy than in the comparison groups), and in addition that cation (Leff et al., 2000). Where relationship issues are addressed
it seems to reduce social impairment and levels of expressed emo- in systemic family and couple therapy, there appears to be particu-
tion within the family. lar benefit when there is a mood problem (Carr, 2009).
In the treatment of depression a recent Cochrane review iden- In the treatment of bipolar disorder a Cochrane review has reported
tified only six studies of family interventions meeting criteria for on seven RCTs, which unfortunately provided insufficient evidence
analysis (Henken et al., 2009). The two studies involving adult for the authors to draw a conclusion (Justo et al., 2009). The family
subjects (four were of child and adolescent subjects) that met cri- interventions were semistructured and psychoeducational, except
teria for review provided data on problem-centred, multifamily, one that used a structured problem-centred systemic approach. The
and family psychoeducational interventions. These two studies of therapy was adjunct with pharmacological treatment. The five stud-
adult populations were not statistically combinable; both included ies comparing family treatment to no intervention did not show
people with depressive symptoms or depression occurring either clinical improvement post-treatment, nor differences in recovery at
in unipolar depressive illness or in bipolar disorder. In both stud- 28 months post-treatment. Three studies compared style or type of
ies, patients received concurrent medication. A study of inpatients, family therapy. It was not possible to use meta-analysis because of
randomized to receive a psychoeducational family intervention or the heterogeneity of outcomes. In the study that compared family
treatment as usual, concluded that women with bipolar depres- focused therapy (FFT is a manualized treatment approach combin-
sion receiving systemic therapy were better in terms of depressive ing psychoeducational and skills-training for families in which one
symptoms, global functioning, and family attitudes towards them (or more) member(s) has bipolar disorder) with a simpler psych-
at discharge and follow-up at 6 months than those offered treat- oeducational intervention, FFT was significantly superior to the
ment as usual (Glick et al., 1985). These effects were not found for brief intervention with respect to the primary outcome of relapse
unipolar depression and were not sustained at 18 months, and for prevention (number needed to treat (NNT) = 3 for one subject to
men there were no treatment benefits and more negative effects at benefit) (Miklowitz et al., 1996). However, anxiety, a secondary out-
18 months. The other study did not show any treatment effects for come measure, may be a side effect found in one study of ‘marital
family therapy or multifamily therapy (see Table 20.1) in recovery psychoeducation’ (van Gent and Zwart, 1991), a finding warranting
from the symptoms of low mood (Miller, 2004). Some suggestions further investigation.
were raised that the timing of family psychoeducation in the course Meta-analysis of randomized controlled studies of multimodal
of the illness was relevant and that ‘education’ can raise anxiety. programmes (involving medication, psychoeducation, individual
Further supporting evidence for the effectiveness of systemic CBT, and systemic therapy) suggest significant reduction in relapse
intervention comes from a Cochrane review of marital therapy rates; however, family therapy was no more or less effective than
for depression, which identified eight high quality RCTs specifi- other individual crisis and psychosocial therapies (Miklowitz and
cally of subjects with depression (Barbato and D’Avanzo, 2006). Craighead, 2007; Scott et al., 2007; Benyon et al., 2008). Data from
individual RCTs have suggested that interventions have more suc- This may reflect historical approaches to dementia and societal
cessful outcomes when intensive (over 20 sessions), and it is sug- perspectives on older peoples’ care. Contemporary medical and
gested that it is the less intensive family-based interventions that societal interest in early diagnosis of dementia brings urgency to
have not shown benefits (Miller et al., 2004). research to refine and evidence older adult systemic psychoeduca-
Among older adult patient populations, one study found that tional interventions that involve the patient.
goal-centred family counselling within a multicomponent inter-
vention was associated with lower rates of institutionalization The Place of Systemic Therapy in
compared to the control group in a small study of volunteer car-
ers (Mittelman et al., 1993). In a second study of older adults, a Old Age Psychiatry
cognitive-behavioural family intervention derived from models There are no inherent reasons why either systemic theory or therapy
for treatment of schizophrenia was found to reduce carer burden should be ‘age limited’. The family remains a significant context for
and depression significantly compared to information-giving to the older people with mental health problems and has been the system
carer alone (Marriott et al., 2000). most commonly described in accounts of theory and practice with
There is not much evidence on the cost effectiveness of family older adults. The specifics of life-stage are part of the context for the
interventions, but some secondary data support the argument that work in hand. Classic description of family development across the
systemic therapy can be no more expensive than antidepressants in lifespan brought recognition that the family develops throughout
the treatment of depression (Leff, 2000). its own life cycle and that later life developments take their place
Does it matter which types of theoretical ideas are applied? As within the cycle (Walsh, 1989). Descriptions of successful clini-
we have seen from the evidence considered so far, there is a bias cal outcomes in resolving problems around older adults living in
towards evaluation of psychoeducational interventions. There are a range of different contexts outside of traditional family units—
insufficient high-quality studies of nonpsychoeducational interven- including residential homes and hospitals—are also now available
tions to comment on whether some theoretical ideas are more help- (Anderson and Johnson, 2010).
ful than others. There is reason to be concerned about side effects if However, mental health problems in later life are frequently mul-
the treatment does not match the type of presentation and phases of tifactorial. Family composition, interaction, and concerns around a
the disorder, and different modes of family work might be needed 75-year-old with relapse in bipolar disorder will be different from
accordingly (Vieta, 2005). those of a 20-year-old with schizophrenia. Contributory factors
There is particular paucity of data for family interventions based among older adults include chronic organic and physical disor-
on principles of narrative therapy in adults. Common sense would ders, disability, and problematic relationships. Referrals to family
suggest that, since families play such a large role in the lives of people therapy within community mental health services for older adults
with mental illness, there is reason to think that social construction- are frequently for less clearly defined mental health problems, when
ist therapies might offer the type of collaboration needed to establish situations are ‘stuck’ or for problems such as disagreements over
the focus of work, fitting to the current issues that patients and their the best course of action, or chronic and intractable illness bring-
families are facing. These approaches have not yet been formally ing some dependency needs to the fore. Systemic therapy may be
studied. Systemic therapy does have a dilemma, given its postmod- particularly helpful in such ‘stuck’ situations.
ern theoretical developments and the eclectic style of therapists who The specific issue identified as the problem for which help is
move between models to use the best approach to fit the family and sought may depend upon what is most prominent or perceived
the problem. Moreover, it becomes harder still to develop the evi- as most worthy of attention at a certain point in time. This may
dence base given the continuing intermixing of theory and practice sometimes be a specific mental disorder; at other times, it may be
in the postmodern perspective (Stratton, 2005). These studies tend a problem with relationships or with the consequences of behav-
not to meet criteria for systematic review. The absence of good evi- iour around a chronic health disability. The practitioner therefore
dence on other types of systemic therapy can therefore be expected always needs to be sensitive to the wider range of issues that may be
to encourage commissioning of psychoeducational treatments. involved in addition to the ‘presenting problem’.
The application of systemic therapy with older adult families is
In summary well documented through case series and descriptive accounts for a
There is evidence to support clinical application of systemic theory range of mental health problems, including for people with demen-
and therapy: particularly from RCTs in depression and schizophrenia. tia and age-related issues. Commonly these indicate improvement
Psychoeducational family therapy has been the subject of the highest in the identified ‘presenting problem’.
quality research and there is some limited evidence of efficacy for this In this section, we illustrate how various ideas and approaches
method from Cochrane reviews. Therapies based on other theoreti- from systemic therapy can be helpful when dealing with a range
cal ideas have not been subject to systematic RCT research. However, of problems. We use some practice-based scenarios to illustrate
lower-level evidence does suggest that systemic therapies may be as the ways in which systemic ideas can fit into old age psychiatry
efficacious as treatment as usual, if not more effective. practice. The scenarios illustrate types of real-life situations and
There is no reason to exclude older adults from the findings of ways in which a variety of practices can fit different families and
data reported for adults, although there are some considerations different predicaments. We look first at therapeutic approaches
that need to be borne in mind: first, the highest level of evidence from the earlier cybernetics-based theoretical perspective (brief
for systemic therapy is available for the most structured interven- problem-focused and transgenerational therapies), then turn to
tions with a specific disorder. Second, with older adults, psychoed- approaches within the constructionism paradigm (solution-focused
ucational models have mainly focused on carer needs without the therapy), and lastly discuss the postmodern paradigm (narrative
patient present. approaches).
Problem-focused brief therapy independence of the older person through enhanced understand-
One of the first descriptions of systemic therapy for age-related ing of rules and balances of power at different life-cycle stages are
issues was the application of problem-focused brief therapy (Herr potentially productive (Benbow et al., 1990). These approaches
and Weakland, 1979). Each family member was asked how they build on ideas about the past contexts within which families have
saw the problem, and the attempted solutions were then explored. lived and the ways in which patterns of behaviour may be replicated
Inappropriate solutions that might in themselves have become, over generations (see transgenerational therapy in Table 20.1). This
or might perpetuate, the problem could be identified, and more account emphasizes the ‘intergenerational ledger’—an implicit bal-
successful alternatives could be sought. This model draws upon ance sheet of what has been given and what is owed between gen-
ideas from strategic therapy (see Table 20.1) and the model of erations. It is argued that a sense of integrity and justice can be a
brief ‘solution-focused’ therapy (de Shazer, 1985). Other brief successful outcome for the older person.
problem-focused interventions described involve family networks Scenario Bipolar Disorder: Contextual and Transgenerational
with a goal of mobilizing its resources, and an emphasis on positive Theoretical Ideas
reframing (see Table 20.1) (Pottle, 1984). The following scenario Mr Smith, aged 85, had been admitted to a psychiatric ward with
gives an example of the use of reframing. irritability and jocular mood, diagnosed as a brief hypomanic epi-
Scenario Depression: Using Ideas from MRI Brief and Strategic sode in his longer-standing bipolar 2 disorder. His daughter reported
Approaches he had not coped at all in the 6 months since moving into sheltered
housing to be near her and had been continually demanding of her
Mr Jones, a 75-year-old man with depression, had been given an
time and help. She is insistent that the inpatient team should sup-
antidepressant by his GP for low mood, loss of energy, and helpless-
port her wish to move him into residential care now it was obvious
ness—he has difficulty managing to cook and leaves various tasks
that he couldn’t cope without someone with him most of the time.
undone, letters unread, bills not paid. His son has stepped in to give
He shrugged his shoulders and agreed, but the ward team became
extra help. He was not responding to this antidepressant and had
concerned about his increasing hostility. Surely he would become
become more anxious, so his GP sought specialist help. Mr Jones
ill again, given the high emotions. They also felt stuck as to how to
was seen at home with his son who was frustrated and feeling fairly
move things on and discussed with a family therapist their ideas
exhausted with doing more and more ‘propping up dad’. As it hap-
about what was going on—they found it hard to know where to
pens, the son had recently lost his job, but with dad to look after it
start.
would be difficult to have the flexibility to get another job.
The ward had access to a visiting family therapist who listened
A member of the old age psychiatry team had undergone basic
to the team’s dilemma about how much to interfere here, fears of
training in systemic therapy and had access to a consultant systemic
‘opening a can of worms’ versus ‘the poor man, being forced into
psychotherapist in a monthly case discussion group. He developed
a home’. One idea was that unresolved feelings about the wife/
a formulation based on ideas about the ways that solutions tried so
mother’s death the preceding year were still affecting the responses
far became the problem (Herr and Weakland, 1979), rather than
of both father and daughter. Another was based on ideas of high
vice versa. Problems frequently arise from the failing solutions that
expressed emotion and that perhaps improving communication
are applied to difficulties. The approach in a brief therapy model
between father and daughter would help. A genogram (Carter and
to assessment would be to identify the ways in which the problem
McGoldrick, 1989) was constructed (see Box 20.1). This drew out
is linked to difficulties that the family has attempted to overcome:
the history that there had been an aunt with bipolar disorder. The
‘What is the problem?’ ‘When did it start?’ ‘To whom is it a prob-
existence of a second daughter was then brought into relief through
lem?’ ‘How is it a problem to them?’ ‘What are the solutions tried
talking about who there is to draw onto the map. The question arose
so far?’ ‘How did it come that you pursued this particular course
about her role and position in this current dilemma.
of action?’
Family systemic assessment could provide a method to approach
The old age team worker discussed his formulation with the con-
a meeting that would not be harmful to any member of the fam-
sultant systemic psychotherapist, using a description of the under-
ily, might remain constructive and fair for each person, and that
lying patterns and beliefs about the difficulties to describe how
might explore whether family work might be therapeutic. The ward
they played a part in the circular patterns between problems and
team decided to ask Mr Smith about having a meeting with both
solutions. In the next meeting, he planned to use reframing (see
his daughters, explaining they might both be helpful resources
Table 20.1) as a means to introduce a new suggestion to explain
to find the best way forward. He was dismissive of his youngest
the function served by the interactions and behaviours between Mr
Jones and his son. The goal was to interrupt the cycle of helpless-
ness–help more by presenting an alternative account of possible
meaning—dad was in fact propping his son up by giving him a role Box 20.1 Genogram
right now. This would be given, together with some information
about common ways in which depression gets people to behave in A genogram is a map of the structure and function of the
ways that maintain illness. patient’s family system. It is drawn starting with those involved
in the assessment, and is useful in assessment to explore the gen-
Transgenerational approaches erations, summarize the personal family history, roles, relation-
Multigenerational relationships have been the focus of accounts of ship patterns, ties and bonds, and family events—providing an
lifespan family therapy with older people (Boszormenyi-Nagy and account of how its members managed and adjusted to problems
Spark, 1984; Hargrave and Anderson, 1994). Interventions aimed and progression through earlier life stages. It can also be a map of
at improving relationships between generations and optimizing family strengths and resources.
daughter’s potential to help but agreed to the ward team inviting worried about how to help her mother cope since her step-father
her to a meeting. ‘seemed to have his head in the sand’. Mrs Brown had remarried 5
In this meeting the younger daughter introduced the idea that years ago when she was 60 years of age. This had brought three new
she had deliberately kept out of things. She felt very guilty but it had grown-up children and two new grandchildren into the reshaped
been her way of surviving when at home. She did care and was con- extended family.
cerned about what happened to her dad. With the family’s agree- In memory clinic it was thought that Mrs Brown would benefit
ment, the genogram was used to track the bipolar illness through from antidepressant treatment and that a family meeting might be
their family. Practical information on the ways in which an endur- the best way to assess ways of supporting Mr and Mrs Brown.
ing mental health problem affects relationships at different stages Rather than move directly into a psychoeducational model, the
in the lifespan was found helpful. The conversation moved to the meeting was framed as open enquiry aimed to ‘orientate’ the meet-
younger daughter talking about some times when it had been very ing: ‘What needs to be talked about?’ and ‘What would make this
frightening when her dad was unwell—she had always been afraid meeting a helpful one?’
after hearing that her grandfather had tried to kill himself. The main problem from the children’s perspective was that they
Both the older sister and father were able to listen to the younger worried they might be making too much demand of mother and
daughter’s experience as well as the experiences they both knew expect her to do too much. Mr Brown wanted to know about the
between themselves. This was helpful to Mr Smith in making him ways he could get help for his wife, and Mrs Brown wanted to talk
feel more for both his daughters and express his appreciation of about her feelings of being useless now that she couldn’t do things
them. This opened the way to both his daughters being able to for others in the way she had been able. Circular questioning (Penn,
acknowledge some of the effects the illness had had on his own 1982) was used to explore these concerns (see Box 20.2). The tech-
opportunities in life. nique of using circular questions can help each person (includ-
Here reference to contextual and transgenerational theoretical ing the therapist) orientate him- or herself to the problem and its
ideas was used. Current tensions can evoke old patterns, such as, effects. The aim is to bring into the open patterns and processes
e.g., when younger children had taken on more parental roles when that might be operating in family members’ relationships as they
their parents were suffering from mental illness. This and other adapted to the dementia.
‘historical contextual’ issues can remain as unresolved matters in Circular questions can be about sequences of interaction; for
the family history of living with the illness over the years. ‘Cut offs’ example: ‘So what happened next when your brother did x?’, or
from family may work for many years especially when living apart, comparisons: ‘Do your brothers feel the same way?’, or differences:
but when unexpected re-engagement occurs following a health ‘Which of your brothers is the most/least like you on this matter?’
change the old patterns of thinking, feeling, and behaving may kick Questions can be about behaviour, e.g. to clarify roles in the deci-
in again. sions that were made. For example, each member of the family
In this particular scenario/vignette the challenge was to facili- could be asked: ‘Who in the family is most likely to make the peace
tate a move away from these old patterns. It is possible that to be when there is a dispute?’ ‘Who is the least likely?’ ‘Who is most
heard and to start to understand influences on people who hurt likely to take the side of the person who makes the peace?’
you (through either illness or their behaviours) can increase capac- In the course of exploring the children’s concerns, a conversa-
ity to try out new ways of interacting in these new circumstances. tion took place about what happened in the kitchen when their
Similar ideas have been suggested for intervention in elder abuse step-brother visited with his son. Meal-time was often a point at
and that an abusive family might increase their ability to use sup- which the children got into conflict, Mrs Brown became anxious and
ports and practical help when dependency and family patterns of felt so unable to cope, everything went wrong, and they just couldn’t
time and across generations are illuminated for them (Richardson get on. There was a practical problem that Mrs Brown now found it
et al., 1994). harder to prepare family tea and she couldn’t do this and make sure
her grandson didn’t fall on the stairs. But behind this, the beliefs
Constructionist approaches that the other side of the family don’t pull their weight and that
When using more eclectic systemic therapy there have been no dif- they couldn’t handle the change in circumstances was very real. The
ferences in types of interventions and approaches proving helpful therapist was very struck by Mrs Brown’s commitment to making
with people with dementia as compared to those with functional things work better for everyone in the family—complicated enough
illness (Benbow et al., 1990). However, interventions in dementia even when there had not been a memory problem encroaching.
were shorter and more often for a crisis (e.g. around changes in
dependency and care needs). Common techniques found helpful
included circular questioning, positive reframing, task setting, and
drawing up a genogram. The following scenario gives an example of Box 20.2 Circular questions
the use of circular questioning. Circular questions are not addressed directly to the person con-
Scenario from the Memory Clinic: Using Social Construction cerned, but instead family members might in turn be invited to
and Solution-Focused Ideas comment on the thoughts and behaviours of, or about relation-
ships between, other family members. They can reduce auto-
Mrs Brown has recently been diagnosed with a vascular dementia,
matic answers that the person usually gives and may illuminate
and when referred to psychiatric memory services was found to be
differences of opinion, expectation, and understanding, and
anxious and experiencing both hopeless and suicidal thoughts. She
hence can bring new information to the fore or a new perspec-
was relieved not to have Alzheimer’s disease, from which her father
tive into the open.
had suffered. Her daughter had come to clinic with her and was
The therapist was, however, genuinely encouraged by the strength of the children was smiling and said it was a bit like when grandma
and optimism of the children that mum herself could regain her was alive as they were being so gentle. She had always been gentle
hope and that they could make a difference to the anxiety, despite when things went wrong. The mood in the room changed. Further
the newest ‘family member’—the mild dementia. She followed her questions invited more talk about the gentleness, the kind of person
developing formulation about the meaning of the problem with she was in her life, and what this meant to each of them. ‘When she
future-oriented questioning, which she hoped might provide ideas was concerned about you what did she do with that?’
for how to make changes for the better for the future: ‘Now that Exploration followed as to how they would like grandmother’s
your mum has mild memory problems, what will the house rules influence in their life to go on from here, i.e. to introduce the idea
become?’ ‘What matters most for the future now?’ Spontaneous of living with her in their futures. ‘How have you kept a connec-
examples of past small positives emerged within this future context tion with your grandmother since she died?’ ‘Has she retained a
conversation. This became the basis for more talk on taking these presence in your life?’ ‘What do you most want to keep of her in
ideas forward. your life?’
A second session was attended by two of the children. They Witness of re-membering has been successfully used in commu-
reported conversations they had had with their step-siblings. They nity settings, such as older peoples’ care homes when members of
had provided information on their mother’s memory changes and the community die (Andrews, 2007). It provides an opportunity for
come to an agreement how to try to start to collaborate over practi- residents to talk about the dead person’s life and then, in the here
cal arrangements for who visited when. They decided that the cen- and now, to consider how each of them feels about hearing this dis-
tral role played by their mother in supporting everybody to keep cussion about the dead person. Their connections with deceased
the peace could beneficially for all concerned be replaced by such persons could be acknowledged by the others in the group, or by
an agreement between themselves. how they have been moved to listen or touched by what they heard.
Descriptions of successful implementation of structural interven-
tions (with families following admission of a relative into long-term Conclusion
care) looking at boundaries, and communication patterns and
alignment of power have been described (Bogo, 1987). Systemic (family) therapy currently may be most usefully viewed as
Another model of delivery is through systemic (or family) thera- an intervention in situations in which the problem is not discrete
pist consultation to an old-age service (Stratton, 2005), and a recent and readily amenable to an individual therapy. There is evidence
detailed account of model of delivery demonstrates its role as an that addressing problems within the family of adults improves
educational tool for old age teams, as well as delivering older people symptoms, improves relapse prevention, and can be combined with
access to consultant systemic psychotherapist skills (Anderson and other therapeutic modalities. Systemic therapy offers a method for
Ekdawi, 2010). assessing complex situations, formulating and fitting interventions
with people in their contexts. The clinical ideas have a natural fit,
Narrative approaches and are in keeping, with the common ways in which profession-
als learn from their clinical experience. Clinicians are constantly
Ideas from narrative therapeutic approaches have been successfully
receiving feedback of various types about outcomes, preferences,
used with people of all ages as a means to help distressed bereaved
and opinions which come from many different perspectives for any
families in ‘re-membering’ practices (Hedtke and Winslade, 2004).
one clinical scenario. Meanwhile, they learn a repertoire of ways of
The focus here is on the stories that people tell about their lives in
proceeding that are most likely to work in ‘this situation with these
order to make sense of the current situation. This vignette suggests
resources, type of people involved, and this set of problems’. Family
that it may be possible to find an alternative, less oppressive story
therapy, as we have suggested in this chapter, can provide a range
that provides a more positive basis for family relationship. There
of theoretical understandings and practices to organize and extend
is evidence that this kind of life review that incorporates narrative
the range of ways of responding. This is the area that systemic ther-
therapy can reduce symptoms of depression (Korte et al., 2011).
apy has captured within services, but there are questions regarding
Scenario-Theme of Bereavement: Using Narrative Approaches the most pressing direction for research.
Mr Richards, a 70-year-old man, was referred with depression fol- There are resource implications for the research agenda on sys-
lowing his wife’s death 1 year previously. He came accompanied by temic therapy because of the importance of having an evidence base
his daughter who explained she was finding it hard to encourage in resource allocation. Research around the use of systemic therapy
him to keep the home clean and eat regular meals. Her own twin in ‘stuck situations’ might be beneficial to older adult service pro-
sons had tried hard too and were the only ones who got anywhere vision. If systemic interventions in the more complex situations
with him, but it never lasted. He expressed a number of bodily com- found in everyday practice are not researched, the more readily
plaints that prevented him doing anything. An initial conversation researchable alternatives will attract funding even though they may
gave a picture of close family relationships, but since grandmother not have the same potential benefits.
had died it was hard to enjoy being together. Mr Richards talked of
References
having lost ‘more than half ’ of himself when his wife died.
The twin grandchildren were invited to the next meeting and their Anderson, E. and Ekdawi, I. (2010) Introducing systemic approaches into our
services for older people. In: Being with older people: a systemic approach
mother called to say they were enthusiastic about this. After some lis-
(eds G. Fredman, E. Anderson, and J. Stott), pp. 211–33. Karnac, London.
tening and circular questioning, it still felt very depressed and hope- Anderson, E. and Johnson, S. (2010). Older people and their significant
less in the room. The therapist invited her coworker to discuss their systems: meeting with families and networks. In: Being with older
ideas in front of the family—a formulation was sketched. The family people: a systemic approach (eds G. Fredman, E. Anderson, and J. Stott),
members were then invited to talk about what they had heard. One pp. 113–38. Karnac, London.
Andersen, T. (1987) The reflecting team: dialogue and meta-dialogues in Korte, J., et al. (2011). Life review therapy for older adults with moderate
clinical work. Family Process, 26, 415–28. depressive symptomatology: a pragmatic randomised controlled trial.
Andrews, J. (2007). Honoring elders through conversations about their lives. Psychological Medicine, 14, 1–11.
In: Collaborative therapy (eds H. Anderson and D. Gehart), pp. 149–66. Leff, J., et al. (2000). The London Depression Intervention Trial: randomised
Routledge, London: controlled trial of antidepressants versus couple therapy in the treatment
Asen, E. (2002a). Multiple family therapy: an overview. Journal of Family and maintenance of people with depression living with a partner: clinical
Therapy, 24, 3–16. outcome and costs. British Journal of Psychiatry, 177, 95–100.
Asen, E. (2002b). Outcome research in family therapy. Advances in Psychiatric Marriott, A., et al. (2000). Effectiveness of cognitive-behavioural family
Treatment, 8, 230–8. intervention in reducing the burden of care in carers of patients with
Barbato, A. and D’Avanzo, B.B.D. (2006). Marital therapy for depression. Alzheimer’s disease. British Journal of Psychiatry, 176, 557–62.
Cochrane Database of Systematic Reviews, 2, CD004188. doi: McNamee, S. and Gergen, K. (1992). Therapy as social construction. Sage,
10.1002/14651858.CD004188.pub2. London.
Bateson, G. (1972). Steps to an ecology of mind. University of Chicago Press, Miklowitz, D.J. and Craighead, W.E. (2007). Psychosocial treatments for
Chicago. bipolar disorder. In: A guide to treatments that work (eds P. Nathan and J.
Benbow, S.M., Egan, D., and Marriott, A. (1990). Using the family life cycle Gorman), 3rd edition, pp. 309–23. Oxford University Press, New York.
with later life families. Journal of Family Therapy, 12, 321–40. Miklowitz, D.J., et al. (1996). Family risk indicators in the course of bipolar
Benbow, S.M., et al. (1993). Family therapy and dementia: review and affective disorder. In Mundt, C., et al. Interpersonal Factors in the Origin
clinical experience. International Journal of Geriatric Psychiatry, 8, and Course of Affective Disorders. Gaskell, London.
717–25. Miller, I., et al. (2004) Does adjunctive family therapy enhance recovery from
Benyon, S., et al. (2008). Psychosocial interventions for prevention of relapse bipolar 1 mood episodes. Journal of Affective Disorders, 82, 431–6.
in bipolar disorder: systematic review of controlled trials. British Journal Minuchin, S. (1974). Families and family therapy. Harvard University Press,
of Psychiatry, 192, 5–10. Cambridge, MA.
Bogo, M. (1987). Social work practice with family systems in admission to Mittelman, M.S., et al. (1993). An intervention that delays institutionalization
homes for the aged. Journal of Gerontological Social Work, 10, 5–20. of Alzheimer’s disease patients: treatment of spouse-caregivers. The
Boszormenyi-Nagy, I. and Spark, G. (1984). Invisible loyalties. Brunner/ Gerontologist, 33(6), 730–40.
Mazel, New York. OCEBM Levels of Evidence Working Group (2011). The Oxford 2011 Levels
Bowen, M. (1978). Family therapy in clinical practice. Aronson, New York. of Evidence. Oxford Centre for Evidence-Based Medicine, <http://www.
Burr, V. (1995). An introduction to social constructionism, pp. 31–4. Routledge, cebm.net/index.aspx?o=5653>.
London. Penn, P. (1982). Circular questioning. Family Process, 21(3), 267–80.
Carr, A. (2009). The effectiveness of family therapy and systemic interventions Pharoah, F., et al. (2010). Family intervention for schizophrenia. Cochrane
for adult-focused problems. Journal of Family Therapy, 31, 46–74. Database of Systematic Reviews, 2, CD000088. doi: 10.1002/1465 1858.
Carter, B. and McGoldrick, M. (1989). The changing family life cycle: a Pottle, S. (1984). Developing a network-orientated service for elderly
framework for family therapy. Allyn & Bacon, Boston. people and their carers. In: Using family therapy (eds A. Treacher and J.
Cecchin, G. (1987). Hypothesizing, circularity and neutrality revisited: an Carpenter). Blackwell, Oxford.
invitation to curiosity. Family Process, 26, 405–13. Richardson, C.A., et al. (1994).Working with older adults and their families—a
Dallos, R. and Stedman, J. (2006). Systemic formulation, mapping the family review. Journal of Family Therapy, 16, 225–40.
dance. In: Formulations in psychology and psychotherapy: making sense of Scott, J., Colom, F., and Vieta, E. (2007). A meta-analysis of adjunctive
people’s problems (eds R. Dallos and L. Johnstone), pp. 73–96. Routledge, psychological therapies compared to usual psychiatric treatment for
London. bipolar disorders. International Journal of Neuropsychopharmacology,
De Shazer, S. (1985). Keys to solutions in brief therapy. Norton, New York. 10, 123–9.
Falloon, I., et al. (1993). Managing stress in families. Routledge, London. Selvini-Palazzoli, M.S., et al. (1980). Hypothesising–circularity–neutrality:
George, E., Iveson, C., and Ratner, H. (1990). Problem to solution. Brief three guidelines for the conductor of the session. Family Process, 19, 3–12.
Therapy Press, London. Sprenkle, D.H. (2003). Effectiveness research in marriage and family therapy:
Gergen, K. (1994). Realities and relationships: soundings in social construction. introduction. Journal of Marital and Family Therapy, 29(1), 85–96.
Harvard University Press, Cambridge, MA. Stratton, P. (2005). Report on the evidence base of the systemic family therapy.
Glick, I.D., et al. (1985). Inpatient family intervention: a controlled evaluation Association for Family Therapy, <www.aft.org.uk>.
of practice, I. Preliminary results of the six month follow-up. Archives of Ryan, C.E., et al. (2005) Evaluating and treating families: the McMaster
General Psychiatry, 42, 882–6. approach. Routledge, New York.
Haley, J. (1977). Problem-solving therapy. Jossey-Bass, San Francisco. Van Gent, E.M. and Zwart, F.M. (1991). Psychoeducation of partners of
Hargrave, T.D. and Anderson, W.T. (1994). Aging and reparation in the bipolar-manic patients. Journal of Affective Disorders, 21(1), 15–18.
intergenerational family. Brunner/Mazel, New York. Vieta, E., et al. (2005). Evidence-based research on the efficacy of psychologic
Hedtke, L. and Winslade, J. (2004). Re-membering lives and conversations interventions in bipolar disorders: a critical review. Current Psychiatry
with the dying and bereaved. Baywood, New York. Reports, 7(6), 449–55.
Henken, T., et al. (2009). Family therapy for depression. Cochrane Database von Bertalanffy, L. (1968). General systems theory: foundations, development,
of Systematic Reviews, CD006728. applications. Penguin, Harmondsworth.
Herr, J.J. and Weakland, J.H. (1979). Counselling elders and their families: Walsh, F. (1989). The family in later life. In: The changing family life cycle
practical techniques for applied gerontology. Springer, New York. (eds B. Carter and M.C. McGoldrick), 2nd edition, pp. 311–22. Allyn
Johnson, S.M. (2004). The practice of emotionally focused marital therapy: and Bacon, Boston.
creating connection, 2nd edition. Bruner/Routledge, New York. Watzlawick, P., Jackson, D., and Beavin, J. (1967). Pragmatics of human
Justo, L., Garcia de Oliveira, B., and Calil, H. (2009). Family interventions communication. W.W. Norton, New York.
for bipolar disorder. Cochrane database , CD005167.pub2. doi: White, M. and Epston, D. (1990). Narrative means to therapeutic ends.
10.1002/14651858. Norton, New York.
CHAPTER 21
Nonpharmacological
interventions in care homes
Ian A. James and Jane Fossey
In England, a third of the people with dementia live in care homes. symptoms within individuals, as described, it helps to focus atten-
The main provider of the 18,083 homes in England is the pri- tion on a number of contributing factors to be considered.
vate sector, and residents tend to be accommodated in large 20- The complexity of the diagram can be reduced by demonstrat-
to 90-bedded facilities. In other countries, such as Sweden, the ing the core features within a grid (Table 21.2). In this grid we can
Netherlands, Austria, and Norway, smaller units are more common. see that there are eight broad groupings of presentation, each cell
Studies that have investigated the relative benefits of the different requiring a different treatment.
sizes of facilities are in the main inconclusive, but there are indica- It is worth noting that the table is a simplified representation as
tions that structural and cultural differences affect the amount of it does not portray the variability implied by the three intersecting
symptoms of agitation shown by residents and the antipsychotic continua. However, the table helpfully demonstrates the types of
drug use in care homes (Testad, 2010). presentations one finds in a care setting, from the least impaired
The English National Dementia Strategy has called for a residents (cell 1) to the most (cell 8). As one might expect, the needs
reduction in the percentage of people with dementia living at of people within each cell will differ greatly, as will the mechanisms
home seeking to support more people in their own homes for of providing psychosocial support. Indeed, those who are not suf-
longer. The impact of this policy, which is already detectable in fering from BPSD will not require specific clinical treatments, but
recent figures (Care Quality Commission, 2011), will change the active steps will be required to ensure they do not develop distress.
nature of people entering care, with care facilities being required This is a very important point as it suggests that nonpharmacologi-
to cater for residents with more advanced levels of cognitive cal input should be both preventative as well as reactive. Thus, even
and physical impairment. While dementia and physical frailty for residents in cell 1, care staff should be encouraged to provide
are common reasons for entering care, noncognitive symptoms, opportunities for them to meet the basic needs of companionship
such as agitation, psychosis, anxiety, and depression, are also and social activities to prevent the residents developing low affect.
determining factors. Agitation is a presentation within a wider Similar requirements will be needed for those in cell 2, although
cluster of symptoms referred to as either challenging behaviours these residents may require additional physical help to allow them
(CB) or behavioural and psychological symptoms of dementia to access such activities.
(BPSD); Table 21.1 outlines some of these common symptoms. Those in the other cells will require additional specialist input
Coping with these BPSD can be particularly taxing for families, depending on their mental and physical health needs. The people
leading to a point where families consider a move to a care home in the right-hand cells (5–8) will have the additional difficulties of
as the most appropriate action for themselves and their family cognitive problems, requiring trained staff to act on their behalf to
member. ensure their needs are met. Owing to the great diversity of pres-
In a study of 141 people with dementia over a 2-week period, the entations, those providing psychological support should ideally
combination of apathy and agitation within individuals was found be trained in a diverse range of interventions and management
to be the most common phenomenon, with behaviour fluctuat- strategies.
ing during the day (Buettner and Fitzsimmons, 2006). An under- It is important to recognize that these axes in Fig. 21.1 are not
standing of the different needs and how interventions may be best independent of one another, and a change on one may produce
tailored to meet these is essential. Simply focusing on a single symp- changes on others. As such, people can move between the cells,
tom (such as agitation) may compound the difficulties people face becoming better or worse with respect to a certain condition. For
in expressing their needs. Figure 21.1 maps out a range of needs in a example, take a fictitious client called Morris. He was referred
3-D model. It attempts to capture the diversity along three different to clinical services because of his depression, poor mobility, and
continua: an axis of good to poor physical health, an axis of good cognitive impairment (cell 8). Following successful treatment
to poor cognitive health, and one for good to poor mental health. of the arthritis in his knees, he became more independent, his
Whilst this model does not entirely account for the variations in mood improved significantly, and he can be reclassified as having
Table 21.1 List of common Challenging behaviours (CB) cell 5 status. In Morris’s case, the two latter features were both
‘reversible’ conditions, whereas his dementia was not significantly
Aggressive forms of CB Nonaggressive forms of CB impacted by the changes. In this case, his cognitive impairment
Hitting Apathy was seen as an ‘irreversible’ feature.
Kicking Depression In many respects, the work of therapist is to determine what aspects
Grabbing Repetitive noise of a resident’s presentation are reversible, and then use the correct
techniques and skills to reverse them. In contrast, the irreversible
Pushing Repetitive questions
conditions often require effective management strategies (palliative
Nipping Making strange noises
care, staff support, specialist equipment, etc.). It is important to note,
Scratching Constant requests for help however, that even when a condition is unlikely to be reversed, the
Biting Eating/drinking excessively person can benefit from identification and treatment of its psycho-
Spitting Overactivity social consequences (see the disability model of dementia; Ballard
Choking Pacing et al., 2001: 92). Consider Joan, who had experienced a stroke 5 years
Hair pulling General agitation ago that left her with severe right-sided weakness. Unfortunately, the
Tripping someone Following others/trailing resulting mobility problems prevented her leaving the home to go
Throwing objects Inappropriate exposure of parts of body to the day centre, go out with her family, walk in the home’s garden,
Stick prodding Masturbating in public areas
etc. These were all activities that previously gave her good wellbeing.
Although the stroke damage is irreversible, the problematic conse-
Stabbing Urinating in inappropriate places
quences of her poor mobility can be dealt with and, via the correct
Swearing Smearing
management, she should be able to go out and attend the day centre
Screaming Handling things inappropriately (training in use of walking frame, assisted-transport to the day cen-
Shouting Dismantling objects tre) and gain access to the garden (installation of a ramp).
Physical sexual assault Hoarding things In many situations, it is features on the Mental Health axis of
Verbal sexual advances Hiding items Fig. 21.1 that are the easiest to reverse and enhance (i.e. improve-
Acts of self-harm Falling intentionally ments in wellbeing and mood). For example, depressive symptoms
Eating inappropriate substances have been found to be associated with disruptive vocalization, one
Noncompliance of the forms of agitation often described as difficult to ameliorate in
Misidentifying
care homes (Dwyer and Byrne, 2000). Underlying depression has an
impact on many neuropsychiatric symptoms. In a recent study of
105 long-term care home residents, the use of tailored recreational
therapies addressing passive behaviour of people with dementia
with depression and agitation showed a significant change in pas-
Good sive behaviour and apathy related to significant changes in agita-
physical tion (Buettner et al., 2010). This is a key point, because they often
health can be the most neglected aspects in some of the ‘task-‘driven staff
practices operating in care facilities. It could also be argued that
good mental health is the most important status to achieve because
Good if contentment can be obtained, even if one has dementia and a
mental
health serious heart condition, a high quality of life can be experienced.
The remainder of this chapter will focus on the methods used to
Severe No improve people’s mental health status.
cognitive cognitive
deficits deficits
Current Treatment Regimens
Numerous UK reports have testified to problems with our exist-
ing service provision within care facilities (National Audit Office,
Poor 2007): e.g. poor leadership, inadequate staff training, and lack of
mental
health opportunities of residents to engage in activities and past-times. The
Poor
Alzheimer’s Society Home from Home report (2007) indicated that a
physical typical person in care spent just 2 min interacting with staff or other
health residents over a 6-h period (excluding time on supported care tasks).
Figure 21.1 Three-axes representation of health. Further, the English National Dementia Strategy criticised the lack
Table 21.2 Grid outlining the eight cells of the three axes model
No to little cognitive impairment Moderate to severe cognitive impairment
Good physical health 1. Good mental health 3. Poor mental health 5. Good mental health 7. Poor mental health
Poor physical health 2. Good mental health 4. Poor mental health 6. Good mental health 8. Poor mental health
CHAPTER 21 nonpharmacological interventions in care homes 271
of consistency in input from specialist mental health teams, and also therapy, staff training, structured activities, environmental interven-
raised concerns over the misprescribing of antipsychotics. Indeed, a tions, medical/nursing care interventions, and combination thera-
UK government commission report by Banerjee (2009) highlighted pies. She identified 83 nonpharmacological intervention studies via
this problem in detail. This watershed report stated that the antipsy- her search, although many were of a poor standard. Examining the
chotics, which were being used to treat CB with and without psychotic effectiveness of interventions for people who are often physically
features, were largely ineffective and had major problematic side frail and have dementia and/or anxiety and depression poses par-
effects, including death. It recommended the reduction of the use of ticular challenges. Studies address a range of different symptoms
antipsychotics by 66% within a 3-year period—a target not met. One of mood or behaviour and use a wide range of outcome measures,
of the reasons for this is the perceived lack of nonpharmacological which can make the findings difficult to compare. Specific prob-
alternatives to medication with respect to the symptoms described in lems relating to conducting research in the area include difficul-
Table 21.2 (Wood-Mitchell et al., 2008). The Department of Health in ties administering a conceptually discrete intervention package.
the UK has issued some guidance about how to reduce antipsychotic There is often overlap between the various forms of therapy, and
use (Alzheimer’s Society, 2011). The section Nonpharmacological many will include a mixture of environmental, orientation, and
Strategies for Older People will provide more detail on some of staff-training features. Hence, it is difficult to determine which
these alternatives to psychotropics, together with their evidence bases aspect of the package is the change element; such confounds also
A comprehensive account of these alternatives within a stepped-care mean it is problematic comparing the forms of therapy with each
framework has been produced (Brechin et al., 2013). other. It is often unfeasible to run double-blind studies due to the
rather overt nature of the interventions and difficulty in devising
Nonpharmacological Strategies placebo conditions. Despite this, there are a number of good quality
randomized controlled trials (RCTs) and other experimental trials
for Older People that are of high quality through ensuring fidelity to the interven-
There are an increasing number of therapies available for older peo- tion model (Kolanowski, 2006). Further published studies can be
ple (Pachana, et al., 2010). In Cohen-Mansfield’s (2001) systematic regarded as good practiced-based evidence. Table 21.3 provides an
review of psychological treatments, she classified eight types of overview of some of the nonpharmacological approaches used with
intervention: sensory, social contact (real or simulated), behaviour older people that have been reviewed systematically.
Table 21.3 Nonpharmacological approaches and their evidence base

I Preventative therapies Systematic reviews and empirical status in dementia Key articles
Reality orientation: uses rehearsal and physical prompts A Cochrane review by Spector et al. (2002) identified six RCTs. Verkaik et al. (2006)
to improve cognitive functioning related to personal The reviewers concluded there was evidence of improvements in
orientation. Tends to be used for people occupying cells terms of cognitive and behavioural features. RO is now assessed
5–8 of Table 21.1. under cognitive stimulation therapy.
Reminiscence therapy: involves discussion of A Cochrane review by Woods et al. (2005a, updated 2009) Gibson (1994); Bohlmeijer
past experiences individually or in a group format. identified five RCTs, four containing extractable data. The et al. (2003)
Photographs, familiar objects, and sensory items used to reviewers reported significant results in terms of cognitions,
prompt recall and discussion. Used with all presentations, mood, caregiver strain, and functional abilities. However, the
cells 1–8. quality of the studies was perceived to be poor.
Validation therapy: based on the general principle of A Cochrane review by Neal and Barton Wright (2003, updated Finnema et al. (2000);
acceptance of the reality of the person and validation of 2009) identified three studies, two showing positive effects. Schrijnemaekers (2002)
his/her experience. Used with people in cells 5–8. However, the reviewers concluded there was insufficient
evidence to view the approach as effective.
Psychomotor therapy: exercises (e.g. walking and ball A Cochrane review by Montgomery and Dennis (2002) Hopman-Rock et al. (1999);
games) are used to target depression and behavioural examining the impact of exercise on sleep problems identified Winstead-Fry and Kijek (1999)
difficulties. Used with all presentations, cells 1–8. one trial that demonstrated significant effects on a range of
sleep variables. Forbes et al. (2008) found limited evidence that
physical exercise slowed down cognitive decline.
Multisensory stimulation: stimuli such as light, sound, A Cochrane review by Chung and Lai (2002, updated 2009) Baker et al. (2001);
and tactile sensations, often in specially designed rooms, identified two RCTs. Despite some favourable results, the studies Van Weert et al. (2005)
used to increase the opportunity for communication were so different that they could not be pooled. As such, the Please update and supply
and improved quality of experience. Used with people in reviewers concluded there was insufficient evidence to view the
cells 5–8. approach as effective.
Cognitive stimulation therapy: derived from RO, focuses Awaiting findings of a new review by Woods et al. (in press). Spector et al. (2003, 2006);
on information processing rather than rehearsal of factual The two previous reviews (Clare et al., 2003; Woods et al., 2005b) Woods et al. (2005)
knowledge. Used with all presentations, cells 1–8, although concluded that despite positive evidence there was insufficient
developed for cells 5–8. In people with dementia, it tends evidence to view the approach as effective.
to improve skills only in those areas that training has been
given; otherwise it is of limited generalization.
(continued)

I Preventative therapies Systematic reviews and empirical status in dementia Key articles
Aromatherapy: use of essential oils to provide sensory A Cochrane review by Holt et al. (2009) identified two RCTs, but Ballard et al. (2002); Holmes
experiences and interactions with staff. The oils can be only the Ballard et al. (2002) trial was reviewed. This trial, despite et al. (2002)
administered via massage techniques or in patients’ baths. flaws, was viewed favourably in terms of reducing agitation and
Used with all presentations, cells 1–8. neuropsychiatric symptoms. Quynh-anh and Paton’s (2008)
work showed equivocal results.
Music therapy: includes playing and/or listening to A Cochrane review by Vink et al. (2009) identified five studies. Lord and Garner (1993);
music as a way of generally enhancing wellbeing. Can be However, the quality of the studies was poor. As such, the Gotell et al. (2002)
used in movement therapies. Used with all presentations, reviewers concluded there was insufficient evidence to view the
cells 1–8. approach as effective.
Environmental manipulation: use of environmental A Cochrane review by Forbes et al. (2009) on the use of bright Judd et al. (1997); Day et al.
cues, signage, and appropriate building layout in order to light therapy in terms of mood, sleep, and behaviour reviewed (2000)
facilitate communication, exercise, and pleasure and to three trials. However, the quality of the studies was poor. As
reduce disorientation. Developed for people in cells 5–8. such, the reviewers concluded there was insufficient evidence to
view the approach as effective. A Cochrane review by Price et al.
(2001, updated 2009) on the use of environmental and social
barriers to prevent wandering failed to identify suitable trials.
II Intervention strategies (formulation-led approaches)
Behavioural management techniques/functional A systematic review by Spira and Edelstein (2006) reported Moniz-Cook et al. (2011)
analysis (FA): based on learning theory and utilizing the 23 studies. These tended to be of poor to moderate quality,
antecedents and consequences of behaviour to devise and many were single case design. Moniz-Cooks et al.’s recent
and execute interventions. The approach has a long Cochrane review (2012) has identified 18 studies with weak but
therapeutic tradition and can be applied to people in all favourable evidence.
cells, but as FA it has been targeted at cells 5–8.
Psychotherapies: the use of CBT, IPT, and other standard Teri et al. (1997) demonstrated the positive impact of CBT on Teri et al. (1991); Miller (2009)
psychotherapeutic formats. These methods have a good mood and problem-solving abilities in people with dementia.
evidence base for people in cells 1–4. CBT has been Teri and Gallagher-Thompson’s (1991) RCT revealed significant
shown to be highly effective in people with comorbid reduction in depression for both people with dementia and their
mental health and physical problems. Adaptions to carers. Scholey and Woods (2003) undertook CBT with seven
people with dementia are recent developments, and will people with dementia and depression, and identified key themes
be limited to mild dementia. in such work.
The first group of approaches (I) are termed ‘preventative strate- Use of preventative approaches
gies’, and are generally designed to promote a positive therapeutic
These are designed to create an atmosphere and environment aimed
milieu and positive wellbeing. It is suggested that improving peo-
at promoting the wellbeing of residents. Many of the approaches
ple’s general levels of contentment serves to improve mood, and
in this category could be termed person-centred in that they are
reduces anxiety and the incidences of problematic behaviours. The
designed to (1) support the strengths of residents, (2) focus on their
second form of approach (II) is termed ‘intervention’, a reactive
difficulties, and (3) foster psychological health and validate their
strategy, responding to a difficulty that has already been diagnosed
daily experiences (Brooker, 2007). In this section, the approaches
(depression, anxiety) or observed (agitation, shouting, wander-
will be discussed in descending order of the strength of their
ing), and the procedure is specifically targeted to intervene with
evidence-base. Thus initially we will discuss the more established
the problem or its causes. These approaches routinely involve the
and empirically tested therapies, and then methods for which the
development of a formulation to help understand the triggering
evidence is emerging but as yet not definitive.
and maintaining features of the problem. For example, behaviour
management approaches pay particular attention to the function of Established preventative therapies
the behaviour, examining in great detail the antecedents, character- Reality orientation (RO)
istics, and consequences associated with its performance. The other RO is one of the most widely used management strategies in care
formulation-based approaches often include careful observation of homes, and particularly for people with dementia. The approach
the behaviour, but also examine the distal features associated with attempts to orientate residents to the ‘present’ via the use of cues
the behaviour (e.g. the patient’s history, personality, physical health, (clocks, calendars, newspapers) and/or discussion. The rationale
staff interactions) (James, 2011). underpinning this strategy suggests that owing to memory and ori-
In the following sections the various approaches will be described entation problems, people with dementia are often confused and
in more detail together with their evidence-base, including control- this may lead to social disengagement. However, if one is able to
led and noncontrolled studies and the implications of the findings provide cues that enable them to engage in what is happening in
for treatment use. the ‘here and now’, they are able to participate in conversations in a
more confident and fulfilling way. The provision of environmental She argues that people with dementia can retreat into an inner real-
cues (e.g. signs, picture boards) also has the advantage of assisting ity based on feelings rather than intellect as they find their present
residents with dementia to find their way around their setting. In reality too painful. VT therapists thus attempt to communicate
recent years, RO has become less popular as concerns have been with people with dementia through empathizing with the feelings
raised about the inflexible way in which it has been practised in and hidden meanings behind their confused speech and behaviour.
some services. There is debate regarding the efficacy of the approach It is the emotional content of what is being said that is therefore
(Verkaik et al., 2005), even in light of Spector et al.’s (2002) favoura- more important than the person’s orientation to the present. It is,
ble Cochrane review of six RCTs. The debate centres around claims however, suggested that therapists can become too focused on con-
that RO can remind the participants of their deterioration (Goudie fused communication and fail to identify simple explanations such
and Stokes, 1989), and it can lead to repeated confrontations with as pain and hunger. In a Cochrane review, Neal and Barton Wright
the person with dementia (Brooker, 2001). (2003) evaluated VT’s effectiveness across a number of controlled
trails, employing cognitive and behavioural measures (Finnema
Cognitive stimulation therapy (CST)
et al., 1998, 2000). They concluded that despite some positive indi-
CST has developed from RO and involves activating residents’ cators in terms of depression (Toseland et al., 1997), the jury was
remaining cognitive functioning through the presentation of still out with respect to its efficacy. Subsequent small controlled
orientation information (e.g. use of physical games; word and studies (Deponte and Missan, 2007; Tondi et al., 2007) have found
number games; everyday objects—see Spector et al., 2006). a reduction in behavioural disturbance and this approach may be
Recent studies show promise for this approach for people with helpful in addressing these signs of distress.
mild to moderate dementia (Knapp et al., 2006), and also dem-
onstrate its cost effectiveness due to the fact that it can often be Environmental modification
performed in a group setting. Livingston et al.s’ (2005) systematic Modifying environments to meet the needs of residents can be effec-
review of six studies of various quality (e.g. Romero and Wenz, tive in improving wellbeing and reducing unwanted behaviours for
2001; Spector et al., 2003) concluded that this approach consist- people with dementia (Bowie and Mountain, 1997). By developing
ently showed promise across a range of situations. This view was a psychosocial understanding of behaviour and its meaning for the
confirmed in a more recent expert review (Spector, 2008). There person, the environment can be changed to meet his or her needs.
is a great deal of ongoing work examining the long-term effects of The use of colour and structure in an environment can help with
CST, patient acceptability, and the mechanisms of change associ- orientation (Gibson et al., 2004). Designing an environment with a
ated with it. Much of this work is taking place under the Support more home-like atmosphere with good lighting and some environ-
at Home Interventions to Enhance Life in Dementia (SHIELD) mental stimulation can reduce unwanted and agitated behaviours
research programme. (Day et al., 2000). Access to safe gardens and outdoor spaces is also
beneficial and opens up the possibility to develop horticultural-type
Reminiscence therapy (RMT) therapies with residents. A comprehensive evidence-base has yet to
RMT involves residents reliving past experiences, especially those be established in this area, although a number of controlled studies
that might be positive and personally significant, such as family (Livingston et al., 2005) and a larger number of noncontrolled stud-
holidays and weddings. This therapy can be used as a group therapy ies have shown environment and design to be important (Hulme
or with individuals. Group sessions employ activities such as art et al., 2010; Zuidema et al., 2010).
and music and often use artefacts to provide stimulation. RMT is
seen as a way of increasing levels of wellbeing, providing pleasure Psychomotor therapy
and cognitive stimulation. When working with people with demen- Psychomotor therapy, sometimes referred to as activity therapy, is a
tia, care staff and families are often encouraged to jointly construct rather varied group of action-based activities, such as dance, sport,
historical reviews of the residents’ lives (i.e. life stories). Life story and drama. A recent study by Cohen-Mansfield and colleagues
work is helpful in promoting attachments between staff and resi- (2010) demonstrated the positive impact of a range of such activi-
dents, particularly in cases where residents have poor communities for residents in care. In this study, Cohen-Mansfield monitored
cation skills. There is growing evidence that RMT is an effective the impact of 25 different tasks undertaken over a 3-week period
treatment for older people with and without dementia (see Woods (conversations, interactions with animals, use of toys, reading, lis-
et al., 2005a and Bohlmeijer et al., 2003, respectively). Indeed, the tening to music, folding towels, flower arranging, puzzles, artistic
approach has many supporters (Gibson, 1994; Brooker and Duce, activities, etc.). For each of the activities, both the amount of per-
2000) due to its flexibility and adaptability to the individual’s needs ceived enjoyment obtained from the task and the length of ‘time
(e.g. a person with severe dementia can still gain pleasure from lis- spent’ engaging in the task were assessed. The findings revealed that
tening to an old record). However, within the care home setting, the most enjoyable tasks were those that involved engaging with liv-
studies continue to show mixed results for its effectiveness (e.g. ing things (people, real baby, animals), followed by tasks involving
Gudex, 2010; Hsieh, 2010) and careful consideration is needed of some form of social simulation (use of dolls, simulation presence
the aims for its use. videos). However, when length of engagement was assessed (i.e.
length of time spent on the task), the residents spent longer on tasks
Validation therapy (VT) that mimicked work-like activities (stamping envelopes, sorting
VT was developed to address the perceived lack of flexibility of the jewellery, folding towels). These results are likely to have important
RO approach with people with dementia. It was suggested by its consequences in terms of the sorts of activities one might suggest
originator, Naomi Feil, that some of the features associated with to carers to help occupy people with dementia in order to enhance
dementia such as repetition and retreating into the past were in wellbeing, increase self-worth, and relieve boredom. A study of a
fact active strategies to avoid stress, boredom, and loneliness. drama and movement group showed positive outcomes in terms
of increased communication (Hokkanen et al., 2003) and an ear- intervention (Buettner, 1999; Colling and Buettner, 2002). These
lier study showed moderate benefits in terms of relaxation and US studies led on to the evaluation and development of a series
orientation (Wilkinson et al., 1998). The therapeutic use of touch of theory-driven protocols addressing sensorimotor, physical,
occurring with activity programmes has also been found to reduce social, cognitive, and psychosocial needs of people with dementia
disruptive vocalizations (Woods et al., 2004). Despite these find- (Buettner and Fitzsimmons, 2009), for which there is growing evi-
ings, two of the three better quality controlled trials conducted in dence of effectiveness of the different elements. These have been
the area failed to find significant differences in terms of depression adapted and are being evaluated in the UK as part of a National
and apathy when compared to treatment as usual (Hopman-Rock Institute for Health Research-funded research programme called
et al., 1999) and low ‘activity’ sessions (Dröes, 1991). However, the Well-being and Health for People with Dementia (WHELD) which
third study (Montgomery and Dennis, 2002) showed a positive is running until 2015.
impact of exercise on a range of sleep variables.
Alternative preventative psychological strategies
Exercise Aromatherapy
It has been shown that physical exercise can have a number of health The two main essential oils used in aromatherapy for dementia are
benefits for older people in care settings. Vogel (2009) has reviewed extracted from lavender and lemon balm. There have been positive
the health benefits, which include a reduction in cardiovascular results from recent controlled trials that have shown significant
morbidity and mortality, optimization in blood pressure and lipid improvements in agitation symptoms with excellent compliance
profile, improved executive function, cardiorespiratory fitness, and and tolerability (Ballard et al., 2002; Holmes et al., 2002). It is rel-
insulin sensitivity, decreased Aß42 plasma levels, increased bone evant to note, however, that Thorgrimsen et al.’s, (2003) review
density, and prevention of falls. Other benefits such as a slower stated that there were flaws in both of these controlled studies, and
rate of decline of activities of daily living have been shown to result thus the findings must be treated with caution. Furthermore, the
from a simple exercise programme, 1 h twice a week (Rolland et al., latest study has found rather mixed results (Nguyen and Paton,
2007). Exercise has also been shown to increase older people’s 2008).
mood and confidence (Young and Dinan, 1994; Singh et al., 2005).
A major programme of research is currently underway in this area, Music therapy
and the studies are collectively known as the Seattle Protocols (Teri The poor quality of studies on this topic has been noted in a
et al., 2008). Cochrane review (Vink et al., 2009). Nevertheless, it has long been
accepted within the scientific literature that music can have a pro-
Multisensory therapy found effect on people’s mood and wellbeing (Sherratt et al., 2004).
Multisensory approaches usually involve using an activity room Indeed, music is used as a mood-inducing technique in a number
that has been designed to provide several types of sensory stimula- of clinical trials, where the music produces a state that temporarily
tion, such as light (often in the form of fibre optics which can move induces depression (Clark, 1983). Music is also considered useful
and be flexible), texture (cushions and vibrating pads), smell, and in reducing unwanted behaviour and improving communication,
sound. The use of these resources is tailored to the individual per- particularly when tailored to people’s taste. For example, Lord and
son and therefore all of the forms of stimulation are not necessarily Garner (1993) showed increased levels of wellbeing, better social
used in one session. Some of the reported benefits for those with skills, and improvements in autobiographical memory in a group of
more advanced dementia include a reduction in aggressive behav- residents interacting with bespoke music. Such improvements were
iours, apathy, and depression and an increase in interaction and not observed in a comparison group engaged in other activities. A
signs of wellbeing (van Weert et al., 2005a, 2005b). However, indi- sustained music therapy programme over 30 weeks has also dem-
viduals differ in their response to this treatment, with some stud- onstrated a reduction in scores on the Neuropsychiatric Inventory
ies failing to find an effect and some obtaining a negative one. The (Cummings et al., 1994)—in particular, reductions in agitation,
latter findings highlight the need for individualized assessments delusions, anxiety, apathy, irritability, aberrant motor activity, and
and planning (Hope, 1998). It is also possible to bring sensory night-time behaviour disturbances, with improvements maintained
experiences into the daily lives of people with dementia through at 1-month follow-up (Raglio et al., 2008). There have also been
the use of interesting and stimulating decoration, colour schemes, reported improvements in depressive symptoms through partici-
and textures in the environment, and the selection of personal care pation in reminiscence focused music groups (Ashida, 2002), and
items, such as toiletries that are scented (Wenbourn, 2003). Chung improvements in communication and irritability when engaged in
and Lai’s (2002) Cochrane report of this therapy stated that, over- structured playing of music (Suzuki et al., 2004). As with all strate-
all, the findings were, as yet, inconclusive. However, more recently, gies, care needs to be used when using this approach because music
Verkaik et al.’s, (2005a, 2005b) review has suggested that multisen- can cue problematic memories too (e.g. music of the 1940s remind-
sory approaches are particularly effective in the treatment of apathy ing people of wartime losses).
and depression both in care and in the community. The elements of
any programme can be difficult to evaluate, although this has been Art therapy
achieved in a study of ‘Simple Pleasures’, which investigated the Art therapy (drama, model making, drawing, and painting) is a
effects of 30 handmade recreational items on the behaviour of nurs- treatment in which people have the opportunity to explore new
ing home residents with dementia, as well as the impact on family skills (Mottram, 2003) and thereby enhance their self-esteem. In
visits, staff knowledge, and volunteer involvement in the homes. the case of people with dementia, art therapy has been shown to
Twenty-three items were found therapeutically valuable and there provide meaningful stimulation and improve social interaction
was some reduction in agitation. Family visits, use of recreational and levels of self-esteem (Killick and Allan, 1999). Whilst many
items, and satisfaction with visits significantly improved during the anecdotal studies suggest that this may be an activity that promotes
a) Triad for someone with a mild dementia Using dolls and toys
The use of dolls and toys in care settings is not new (Libin and
Challenging Behaviour: constantly
asking for reassurance Cohen-Mansfield, 2004), but has only recently been studied in a
systematic manner (James et al., 2005; Mackenzie et al., 2006a).
Investigations have involved the introduction of dolls and teddy
bears into care homes following a standard format (Mackenzie
Feeling (eg.
Anxious) et al., 2006b). Typically, staff are given information and guidelines
on their use prior to their introduction (Mackenzie et al., 2007).
The findings from these investigations have been favourable for
Physical Thoughts—I can’t do both residents and staff (James et al., 2006; Mackenzie et al., 2006a).
Sensations—, anything for myself. I However, this approach is controversial and an unresolved debate
raised heart- need my husband with
within the Journal of Dementia Care (Mackenzie et al., 2006c, 2007)
rate, shaking me at all times.
has demonstrated resistance to the technique as it can be viewed as
‘patronizing’ and promoting ‘infantilization’.
b) Triad for someone with a moderate-severe dementia
Challenging Behaviour: hitting a staff Tool-box approaches

member & trying to leave the building Some of the above strategies have been presented as stand-alone
techniques, but it is evident that often carers combine them and
use them in nonstandard ways. A good example of this is the vari-
Feeling (eg. ous forms of ‘tool-box’ techniques (Thwaites and Sara, 2010). This
Anger) widely used approach involves creating an individualized box of
items for each person, containing personalized material (e.g. photo-
graphs, postcards, camcorder recordings of family scenes or voices,
Physical Thoughts—he has no items of clothing, ornaments, relevant maps, aromatherapy oils).
Sensations— right to stop me getting
agitation, raised out of this place. I’m
Carers, particularly care staff, can use the items in the box to stimu-
heart-rate, hot. not in prison!! late and communicate with the individual, learning more about the
person’s history and promoting positive reminiscence.
Figure 21.2 Cognitive behaviour therapy triads
Intervention strategies
wellbeing, there is limited evidence of its impact-specific condi- In the following section, a number of ‘formulation-led’ approaches
tions or of longer-term benefit when sessions stop. are described. Here, typically a condition has already been observed
(depression, problematic behaviour, etc.) and the intervention pro-
Animal-assisted activities cedure is specifically targeted at its causes. These interventions
Animals introduced into nursing homes as regular visitors or as routinely involve the development of a formulation (i.e. a descrip-
home pets have been shown to have positive effects, including tive account of the problem in relation to the person and his or her
reducing blood pressure, agitation, strain, tension, and loneliness past) to help understand the triggering and maintaining features
and increasing life-expectancy (Churchill et al., 1999). Short-term of the condition (James, 2010). First, some standard psychothera-
interactions with dogs have been shown to increase social inter- peutic approaches will be described (cognitive behaviour therapy
action with, and between, older people with mental impairment (CBT), interpersonal therapy (IPT)), whose use is limited to those
(Greer et al., 2001). These forms of social contact have proven in cells 1–4 of Table 21.2 (i.e. with up to mild impairment) (James,
beneficial in the treatment of behaviour problems in dementia 2010). We will then go on to discuss therapies that can be used with
(Zisselman et al., 1996). The presence of other types of animal all presentations (cells 1–8), even with cases of severe dementia
have also demonstrated benefits. For example, the use of a fish (behaviour therapy (BT) and needs-led frameworks). It is relevant
tank in a dining area has been shown to reduce aggression and to note that the function of the formulation of the standard psy-
enhance the nutritional intake of home residents with dementia chotherapies differs with that of the two latter therapies. For exam-
(Edwards, 2004). However, the welfare of human and animal par- ple, CBT and IPT formulations are designed to aid ‘clients’ gain a
ticipants needs to be considered in any programme and guidance better understanding of their problems; this requires the clients to
is becoming available on both infection control (Lefebvre et al., have a degree of insight into their difficulties and the ability to initi-
2008) and on programme planning that incorporates consid- ate change with respect to their behaviour. In contrast, BT and the
erations needed for all participants to achieve positive outcomes needs-led frameworks employ formulations as vehicles to enable
(Fossey, in press). carers to gain a better understanding of the people with dementia’s
For situations when live animals are not appropriate, research- difficulties. This is particularly important because in cases of severe
ers have investigated the ability of robotic animals to provide dementia, it is the staff carers who are required to carry out the
companionship and increase quality of life in nursing home resi- interventions.
dents. Banks et al. (2008) examined the interactions of residents
with Sony’s robotic dog ‘Aibo’ and found that both Aibo and a live Standard psychotherapies
dog reduced agitation, and both were equally effective in doing so. Over the last 10 years there has been an increasing interest in apply-
Similarly, Libin and Cohen-Mansfield (2004) had promising pilot ing CBT and IPT to older people, including those with cognitive
results with both a robotic dog and a plush dog. impairment (Miller and Reynolds, 2007; James, 2010). In relation
to people with dementia, these therapies are used in cases of low actions leading up to it (A). The reactions to the behaviour (C)
mood and anxiety or when the person retains some insight and are also studied because they frequently serve to either increase or
problem-solving abilities. CBT examines people’s distress within a decrease the likelihood of the action occurring again. Such moni-
cycle (Fig. 21.2a and b); in order to alter the inner feeling, the outer toring helps to identify the function of CB, and the accurate obser-
features are worked on. The cognitive aspects (i.e. thoughts and vations are essential in generating better hypotheses. For example,
beliefs) receive particular attention, because when people are dis- consider the case of Joan, whose continual shouting was initially
tressed, they tend to engage in negative thinking (e.g. I am worth- attributed to arthritic pain. However, the functional assessment
less; everyone hates me; no-one wants me). CBT has developed revealed that she only shouted in the presence of men. This meant
strategies to help people re-evaluate their thinking, and to start her painkillers could be stopped, which relieved her constipation,
re-engaging in ‘helpful’ activities and relationships they have been and the new data allowed the therapist to focus on her issues with
avoiding (James, 2010). men, enabling her therapists to go on to develop a more appropriate
It is relevant to note that this triad is also helpful in understand- intervention.
ing the problematic behaviour of those with more severe dementia, Teri et al. (2005) developed and conducted an 11-session training
with their behaviours linked to their distorted sense of their current programme to teach 41 carers of individuals with dementia how
reality (Fig. 21.2b). However, their cognitive deficits prevent them to manage agitated behaviour using functional analysis (FA)-based
being able to make changes for themselves. behavioural interventions in their home environment. During ses-
Teri and Gallagher-Thompson (1991) reported positive find- sions, carers worked with a therapist to define problem behaviours,
ings from a clinical trial of CBT with people in the early stages of complete and interpret functional analyses, and develop interven-
Alzheimer’s disease. Single case and group CBT have also been tions. Teri et al. (2005) presented four case studies to illustrate the
used by other researchers with some favourable results (Koder, effectiveness of their programme. The antecedents to the four par-
1998; Kipling et al., 1999). ticipants’ agitation were found to be (1) lack of activity, (2) confron-
IPT, as the name suggests, examines the person’s distress within tation over confused statements, (3) inactivity specifically at dusk,
an interpersonal context. People are encouraged to look at how and (4) lack of social attention due to the caregiver completing
important relationships have changed since they have become dis- tasks, respectively. Consequently, antecedent control interventions
tressed, and they are also asked to examine their communication were designed (e.g. providing appropriately tailored activities).
styles to see if they can be improved. Such an approach is particu- For one participant, the interventions resulted in the elimina-
larly relevant for those therapists who see challenging behaviours as tion of verbal and physical aggression. For the other three partici-
a communication strategy in a form other people find socially unac- pants, reductions in agitated and aggressive behaviours were found.
ceptable. For example, a person with poor verbal skills and reduced Unfortunately no data were presented and limited qualitative feed-
insight is unable to ask for food, and so he or she takes it from back from the participants were reported as outcome measures,
someone else. There is some overlap of IPT with the person-centred making it difficult to interpret how successful the interventions
work of Kitwood (1997) and Stokes (2000). There is good empiri- actually were.
cal evidence for this form of treatment with older people (Miller The efficacy of BT has been demonstrated in the context of
and Reynolds, 2002); it has only recently been applied in the area dementia in a number of studies (Doyle et al., 1997; Baker et al.,
of dementia (James et al., 2003; Miller and Reynolds, 2007; Miller, 2006) and reviews (Spira and Edelstein, 2006) A rigorous Cochrane
2009; James, 2010). Owing to the demands IPT places on insight- review has recently been conducted by Moniz-Cook et al. (2012).
fulness and the ability to reflect, its use remains rather limited in the Fifteen trials were identified, ten from family care settings and five
treatment of people with dementia. from residential care settings. For the primary outcome of fre-
quency of problematic behaviours, only 10 studies (n = 1140 par-
Carer-centred, person-focused approaches
ticipants) had usable data. They concluded that whilst FA showed
The following approaches can be employed for a range of clinical promise, it is too early to provide an indication of its effectiveness.
presentations, even in situations where clients have poor communi- This is because the evidence rests largely on a handful of studies for
cation skills and/or poor insight; thus these methods are suitable for which its actual effects are unclear, since the intervention tended to
working with people with severe dementia. The key reason they can be delivered within broad multicomponent programmes of psycho-
be employed with this group is that, although the focus of the treat- social interventions. Furthermore, the studies were generally small
ment is directed at the person with dementia, much of the actual and the duration of interventions varied across studies.
work is done via carers. Due to this, working with these interven-
tions requires a systems approach, whereby much of the success Needs-led therapies
depends on supporting the carers to deliver the therapy effec-
These approaches have often been developed for the treatment of
tively. For this reason, the approaches are termed ‘carer-centred,
CB. Currently, there are a number of conceptual models that exam-
person-focused’ (James, 2011).
ine cases of CB in terms of people’s needs (Cohen-Mansfield, 2000;
Behaviour therapy James, 2011). These frameworks typically involve obtaining two
BT requires a detailed assessment period in which the triggers, types of information: (1) background features (history, premorbid
behaviours, and reinforcers (i.e. the antecedents, behaviour, and personality and coping style, cognitive status, mental health status,
consequences (ABC)) are identified and their relationships made physical health status, environmental and contextual status) and
clear. This process is frequently referred to as a functional assess- (2) a comprehensive description of the problematic behaviours—a
ment (Gormley et al., 2001; Moniz-Cook et al., 2012). The therapist functional assessment. By putting these two types of information
will often use some kind of chart or diary to gather information together, one is in a stronger position to accurately identify the per-
about the manifestations of a behaviour (B) and the sequence of son’s needs. The needs-based models highlight the fact that CB are
usually not unpredictable random actions; rather they are rational standpoint, and provide information about caregiver interactions—
activities with a high degree of predictability. Indeed, frequently, from which action plans for care improvement can be developed.
CB are manifestations of residents’ attempts to fulfil an unmet need. The DCM tool also acts as an audit of person-centred practice and
For example: has been used with positive results in some studies and had a more
◆ CB as a method of trying to achieve a need (e.g. breaking a win- equivocal impact in others. The fact that it can be both intervention
dow in order to get outside of the care facility and walk in the and outcome evaluation can make its true worth difficult to assess,
garden) but it has good face validity and is generally well received by prac-
titioners (Brooker, 2007).
◆ CB as a means of fulfilling one’s need (e.g. urinating in a sink in In general, evidence shows that staff can be trained to deliver bet-
order to relieve the pressure in one’s bladder) ter care (Opie et al., 2002; Schrijnemaekers et al., 2002), but the
◆ CB as an expression of frustration that a need is not being met effects do not tend to generalize across care activities and they tend
(e.g. hitting a member of staff who is insisting you go to bed when to fade with time. Owing to concerns about the limited long-term
you are enjoying a late-night TV programme). impact of training, it appears that employing training interven-
tions alone may not be the best use of clinical time unless the
Cohen-Mansfield (2000) provides one of the best descriptions
training is supported by continued supervision and organizational
of an unmet needs model, and her recent empirical study testi-
development.
fies to the efficacy of her approach (Cohen-Mansfield et al., 2007).
Of late, a number of studies have attempted to improve the sta-
The fundamentals of her work are based on Maslow’s hierarchy of
tus of care by intervening at the organizational level. For example,
needs: the need for physiological/physical wellbeing; safety; love
relatives have been encouraged to work alongside staff and get
and belonging; esteem; and self-actualization. She believes that CB
involved in care planning, delivery, and review. The results have
often occur when the ‘vulnerable and disadvantaged’ person strives
been mixed, with family members having difficulties in sustaining
to have some of these fundamental needs met. She believes that
their involvement (Gaugler et al., 2004; Train et al., 2005). On bal-
such striving is often problematic for people with dementia because
ance, the organizational studies show limited evidence of sustained
of the following difficulties: poor communication; inability to use
improvement in culture, attitudes, or the quality of care (Aylward
prior coping mechanisms; unable to meet own needs without the
et al., 2003).
assistance of others. There are often additional difficulties because
Programmes that have taken a broader organizational focus in
the person’s environment may not be able to provide the ‘need’, or
implementing person-centred care in combination with train-
the setting may simply not understand it.
ing and mentoring have been shown to have some benefits on
Bird et al. (2007) are also exponents of the needs-led perspec-
medication use in the Focused Intervention Training and Support
tive, and have provided numerous case examples of the benefits of
Programme (Fossey et al., 2006; Fossey and James, 2008), and some
this methodology. More recently, Bird has undertaken a controlled
aspects of quality of life in the VIPS model of care (Brooker, 2007)
study that illustrates the efficacy of his work (Bird et al., 2009).
and the Enriched Opportunities Programme (Brooker, 2011).
Other frameworks using a cognitive behavioural model to aid
Individual care home groups have also implemented their own
staff in their understanding of how they may contribute to resi-
programmes of organizational and cultural change; one exam-
dents’ behaviour through a communication cycle have also been
ple, the PEARL programme (Baker et al., 2009), is based on the
used with positive outcomes (Fossey et al., 2006; Fossey and James,
VIPS model (Brooker, 2007) and uses a cycle of audit, training,
2008).
and organizational change to achieve clearly defined standards of
care. In the UK, there has been support from the Department of
Staff Training Health and Alzheimer’s Society to test the feasibility of the roll-out
As demonstrated with carer-centred and person-focused of a structured programme (Fossey and James, 2008) to a larger
approaches, staff are a vital piece of the jigsaw in delivering good number of homes in order to address the general concerns that are
care in homes. De Vugt et al. (2004), using a longitudinal design, raised about care delivery.
showed that caregiving management strategies were associated
with incidences and types of behavioural problems. For example,
caregiver impatience, irritation, and anger towards residents were
Conclusion
more likely to result in greater resident agitation. It is important Having reviewed many of the treatments currently available, it is
that staff have an understanding of evidence for recreational and worth noting the commonalities between the various interventions.
therapeutic interventions, the rationale of reducing medication that Indeed, in many of the studies the treatments employed similar
has potential to interfere with therapy effectiveness, and the need mechanisms of change, often promoting activities, cognitive stimu-
to individualize activities to maximize their benefit (Kolanowski, lation, and person-centred approaches. From our perspective, and
et al., 2009). an examination of the modus operandi of many of the treatments
In terms of training, the methods of staff education have been presented, good communication skills and an ability to interact
varied and involved broad-ranging teaching programmes (Burgio positively are key to those studies that have demonstrated positive
et al., 2002) and use of reflective practice tools such as Dementia effects. This is because many residents in care homes find it an iso-
Care Mapping (DCM) (Kitwood, 1997) and operationalized lating experience, due to a combination of the losses experienced
person-centred care systems, defining practice in terms of value (social networks, familiar environment and routines) and cogni-
of people, individuality of care, perspective of service users, and tive (memory and orientation) problems resulting from dementia.
positive social psychology (VIPS) (Brooker, 2007). These frame- Therefore, when a therapist or staff member engages with a resi-
works assess the activity and level of wellbeing from a resident’s dent, using activities, photographs, music, an animal, doll, massage,
etc., their level of interpersonal contact rises considerably. For Baker, J.C., Hanley, G.P., and Mathews, R.M. (2006). Staff administered
many residents, this will result in an increase in mood and wellbe- functional analysis and treatment of aggression by an elder with
ing. It is also our view that the formats provided within the various dementia. Journal of Applied Behaviour Analysis, 39(4), 469–74.
Banerjee, S. (2009). The use of antipsychotic medication for people with
approaches provide staff with specific structures that aid the staff ’s
dementia: time for action. Department of Health, London.
ability to communicate. This feature was particularly evident in the
Ballard, C.G., et al. (2001). Dementia: management of behavioural and
studies using dolls (James et al., 2005), in which staff reported that psychological symptoms. Oxford University Press, Oxford.
the dolls gave them a focus of conversation. It is relevant to note Ballard, C.G., et al. (2002). Aromatherapy as a safe and effective treatment for
that in the above study, many staff admitted that they often found the management of agitation in severe dementia: the results of a double
speaking to people with dementia difficult. blind placebo controlled trial with Melissa. Journal of Clinical Psychiatry,
The present chapter has also shown that there is a general 63, 553–8.
move towards more person-centred and staff-centred foci of care. Banks, M.R., Willoughby, L.M., and Banks, W.A. (2008). Animal-assisted
Through the use of such perspectives, greater attempts are now therapy and loneliness in nursing homes: use of robotic versus living
dogs. Journal of the American Medical Directors Association, 9, 173–7.
made to understand the individual’s experience of his/her problem
Bird, M., et al. (2007) A controlled trial of a predominantly psychosocial
within an intrapsychic, interpersonal, and environmental context
approach to BPSD: treating causality International Psychogeriatrics,
(Bond and Corner, 2001). It is this perspective that leads to a fur- 19(5), 874–91.
ther shared feature with respect to the interventions: that is, the Bird, M., Llewellyn-Jones, R.H., and Korten, A. (2009). An evaluation of
systemic perspective. Onishi et al. (2006), in a large survey of three the effectiveness of a case-specific approach to challenging behaviour
different types of Japanese care facilities, found that the incidences associated with dementia. Aging and Mental Health, 13(1), 73–83.
and forms of problematic behaviours differed across settings. The Bohlmeijer, E., Smit, F., and Cuipers, P. (2003). Effects of reminiscence and
carers’ abilities to cope with the behaviours also varied. From their life review on late-life depression: a metal-analysis. International Journal
findings, the researchers suggested that such systemic knowledge of Geriatric Psychiatry, 18, 1088–94.
of the resilience of such settings was crucial in determining the Bond, J. and Corner, L. (2001). Researching dementia: are there unique
methodological challenges for health services research? Ageing and
appropriateness of placements and transfers. Therefore, it is clearly
Society, 21, 95–116.
important to recognize how the context of a person with demen- Bowie, P. and Mountain, G. (1997). The relationship between patient
tia’s relationships with family and caregivers affects both individu- behaviour and environmental quality for the dementing. International
als with dementia and their caregivers. Such a view emphasizes the Journal of Geriatric Psychiatry, 12, 718–23.
need to work with all of the systems involved in the residents’ care Brechin, D., Murphy, G., James, I., and Codner, J. (2013). Alternatives
(families, professional carers, organizations, etc.). Indeed, care staff to Antipsychotic Medication: Psychological Approaches In Managing
and families are integral to the treatment strategies and are essen- Psychological and Behavioural Distress in People with Dementia. British
tial in obtaining valid and reliable information and constructing Psychological Society Publications (www.bps.org.uk).
appropriate formulations and interventions. It is evident therefore Brooker, D. (2001). Enriching lives: evaluation of the ExtraCare Activity
Challenge. Journal of Dementia Care, 9(3), 33–7.
that carer/staff training is an integral part of most treatment pro-
Brooker, D. (2007). Person-centred dementia care: making services better.
grammes. Despite the relevance of this issue, there remain relatively Jessica Kingsley, London.
few ‘quality’ studies in the area (e.g. Moniz-Cook et al., 1998; Proctor Brooker, D. and Duce, L. (2000). Wellbeing and activity in dementia: a
et al., 1999; Marriot et al., 2000—see also Cohen-Mansfield et al., comparison of group reminiscence therapy, structured goal-directed group
1997). Clearly, training and support is an important area worthy activity and unstructured time. Aging and Mental Health, 4(4), 354–8.
of further work, and such studies need to be large and sufficiently Brooker, D.J., et al. (2011). The Enriched Opportunities Programme for
powered, with robust designs that include follow-up methods. The people with dementia: a cluster-randomised controlled trial in 10 extra
latter issue is given particular emphasis in the NICE guidelines on care housing schemes. Aging and Mental Health, 15(8), 1008–17.
dementia (2006), with specific mention of the need to develop and Buettner, L. (1999). Simple pleasures: a multilevel sensorimotor intervention
deliver person-centred educational programmes. Finally, it is evi- for nursing home residents with dementia. American Journal of
Alzheimer’s Disease, 14, 41–52.
dent from this review that the evidence base underpinning nonp-
Buettner, L. and Fitzsimmons, S. (2006). Mixed behaviours in dementia: a
harmacological interventions is growing and can offer alternatives new paradigm for treatment. Journal of Gerontological Nursing, 32(7),
to inappropriate medication and be used in combination with med- 15–22.
ication to improve care for residents with dementia. Buettner, L. and Fitzsimmons, S. (2009). NEST approach. Needs. Environment.
Stimulation. Techniques: dementia practice guidelines for disturbing
References behaviours. Venture Publishing, .
Alzheimer’s Society (2007). Home from home: a report highlighting Buettner, L., Fitzsimmons, S,. and Dudley, W.N. (2010). Impact of underlying
opportunities for improving standards of dementia care in care homes. depression on treatment of neuropsychiatric symptoms in older adults
Alzheimer’s Society, London. with dementia. Research in Gerontological Nursing, 3(3), 221–32.
Alzheimer’s Society (2011). Optimising treatment and care for people with Burgio, L.D., et al. (2002). Teaching and maintaining behaviour management
behavioural and psychological symptoms of dementia: a best practice guide skills in the nursing home. Gerontologist, 42, 487– 96.
for health and social care professionals. Alzheimer’s Society, London. Care Quality Commission (2011). Review of meeting the healthcare needs of
Ashida, S. (2002). The effect of reminiscence music therapy sessions on people in care homes. CQC, London.
changes in depressive symptoms in elderly persons with dementia. Chung, J.C.C. and Lai, C.K.Y. (2002). Snoezelen for dementia. Cochrane
Journal of Music Therapy, 37, 170–82. Database of Systematic Reviews: Issue 4, doi: 10.1002/14651858.CD00315.
Aylward, S., et al. (2003). Effectiveness of continuing education in long term John Wiley, Chichester.
care: a literature review. Gerontologist, 43, 259–71. Churchill, M., et al. (1999). Using a therapy dog to alleviate the agitation and
Baker, C. (2009). Introducing PEARL: rewarding good practice in dementia desocialisation of people with Alzheimer’s disease. Journal of Psychosocial
care. Journal of Dementia Care, 17(4), 31–3. Nursing and Mental Health Services, 37, 16–22.
Clark, D.M. (1983). On the induction of depressed mood in the laboratory: Gormley, N., Declan, L., and Howard, R. (2001). Behavioural management of
evaluation and comparison of the Velten and musical procedure. aggression in dementia: a randomised controlled trial. Age and Ageing,
Advances in Behaviour Research and Therapy, 5, 27–49. 30, 141–5.
Cohen-Mansfield, J. (2000). Use of patient characteristics to determine Goudie, F. and Stokes, G. (1989). Understanding confusion. Nursing Times,
non-pharmacological interventions for behavioural and psychological 85, 35–7.
symptoms of dementia. International Psychogeriatrics, 12(1), 373–80. Greer, K.L., et al. (2001). A comparison of the effects of toys versus live animals
Cohen-Mansfield, J. (2001). Nonpharmacologic interventions for on the communication of patients with dementia of the Alzheimer’s type.
inappropriate behaviors in dementia: a review, summary and critique. Clinical Gerontologist, 24, 157–82.
American Journal of Geriatric Psychiatry, 9, 361–81. Gudex, C., et al. (2010). Consequences from use of reminiscence—a
Cohen-Mansfield, J., et al. (1997). Evaluation of an in-service training randomised intervention study in ten Danish nursing homes. BMC
programme on dementia and wandering. Journal of Gerontological Geriatrics, 10, 33.
Nursing, 23, 40–7. Hokkanen, L.I., et al. (2003). Dance/movement therapeutic methods in
Cohen-Mansfield, J., Libin, A., and Marx, M. (2007). Nonpharmacological management of dementia. Journal of the American Geriatrics Society,
treatment of agitation: a controlled trial of systematic individualized 51(4), 576–7.
intervention. Journal of Gerontology: Medical Sciences 62(8), 906–18. Holmes, C., et al. (2002). Lavender oil as a treatment for agitated behaviour
Cohen-Mansfield, J., et al. (2010). The value of social attributes of stimuli in severe dementia: a placebo controlled study. International Journal of
for promoting engagement in persons with dementia. Journal of Nervous Geriatric Psychiatry, 17(4), 305–8.
and Mental Disease, 198(8), 586–92. Hope, K. (1998). The effects of multisensory environments with older people
Colling, K. and Buettner, L. (2002). Simple pleasures: interventions from with dementia. Journal of Psychiatric and Mental Health Nursing, 5,
the need-driven dementia compromised behavior model. Journal of 377–85.
Gerontological Nursing, 10, 16–20. Hopman-Rock, M., et al. (1999). The effects of psychomotor activation
Cummings J.L., et al. (1994). The Neuropsychiatric Inventory: comprehensive program for use in groups of cognitively impaired people in homes for
assessment of psychopathology in dementia. Neurology, 44, 2308–14. the elderly. International Journal of Geriatric Psychiatry, 14, 633–42.
Day, K., Carreon, D., and Stump, C. (2000). Therapeutic design of Hsieh, C.-J., et al. (2010). Reminiscence group therapy on depression and
environments for people with dementia: a review of the empirical apathy in nursing home residents with mild-to-moderate dementia.
research. Gerontologist, 40, 397–416. Journal of Experimental and Clinical Medicine, 2(2), 72–8.
Deponte, A. and Missan, R. (2007). Effectiveness of validation therapy (VT) Hulme, C., et al. (2010). Non-pharmacological approaches for dementia that
in group: preliminary results. Archives Gerontological Geriatrics, 44(2), informal carers might try to access: a systematic review. International
113–17. Journal of Geriatric Psychiatry, 25, 756–63.
De Vugt, M., et al. (2004). Do caregiver management strategies influence James, I.A. (2010). Cognitive behaviour therapy with older people: interventions
patient behaviour in dementia? International Journal of Geriatric for those with and without dementia. Jessica Kingsley, London.
Psychiatry, 19(1), 85–92. James, I.A. (2011). Understanding behaviour in dementia that challenges.
Doyle, C., et al. (1997). Efficacy of psychological treatments for noisemaking Jessica Kingsley, London.
in severe dementia. International Psychogeriatrics, 9, 405–22. James, I.A., Mackenzie, L., and Mukaetova-Ladinska, E. (2006). Doll use in
Dröes, RM. (1991). Effecten van psychosociale behandelingsvormen bij care homes for people with dementia. International Journal of Geriatric
SDAT-patiente (The effects of psychosocial treatment methods in Psychiatry, 21, 1093–8.
SDAT patients). In: In Beweging. Over psychosociale hulpverlening aan James, I.A., Postma, K., and Mackenzie, L. (2003). Using an IPT
demente ouderen (Psychosocial care for demented elderly). Dissertation. conceptualisation to treat a depressed person with dementia. Behaviour
Uitgeverij Intro, Nijkerk. and Cognitive Psychotherapy, 31(3), 451–6.
Dwyer, M. and Byrne, G.J. (2000). Disruptive vocalisations and depression in James, I.A., et al. (2005). The therapeutic use of dolls in dementia care. Journal
older nursing home residents. International Psychogeriatrics, 12, 463–71. of Dementia Care, 13(3), 19–21.
Edwards, N. (2004). Using aquariums in managing Alzheimer’s disease: Killick, J. and Allan, K. (1999). The arts in dementia care: tapping a rich
influence on resident nutrition and behaviours and improving staff resource. Journal of Dementia Care, 7, 35–8.
morale. In: SCAS (ed.) People and animals: a timeless relationship. Kipling, T., Bailey, M., and Charlesworth, G. (1999). The feasibility of a cognitive
IAHAIO, Glasgow. behavioural therapy group for men with a mild/moderate cognitive
Finnema, E. (2000). Emotion-oriented care in dementia. A psychosocial impairment. Behavioural and Cognitive Psychotherapy, 27, 189–93.
approach. Dissertation. Regenboog, Groningen. Kitwood, T. (1997). Dementia reconsidered: the person comes first. Open
Finnema, E., et al. (1998). The design of a large-scale experimental study into University Press, Buckingham.
the effect of emotion oriented care on demented elderly and professional Koder, D. (1998). Treatment of anxiety in the cognitively impaired elderly:
carers in nursing homes. Archives Gerontological Geriatrics, 6, 193–200. can CBT help? International Psychogeriatrics, 10(2), 173–82.
Fossey, J. (in press). Pet friendly care kit: developing pet friendly practices in Knapp, M., et al. (2006). Cognitive stimulation therapy for people with
care homes. Society for Companion Animal Studies, Burford. dementia: cost-effectiveness analysis. British Journal of Psychiatry, 188,
Fossey, J. and James, I. (2008). Evidenced based approaches for improving 574–80.
dementia care in care homes. Alzheimer’s Society, London. Kolanowski, A., Buettner, L., and Moeller, J. (2006). Treatment fidelity plan for
Fossey, J., et al. (2006). Effect of enhanced psychosocial care on antipsychotic an activity intervention designed for persons with dementia. American
use in nursing home residents with severe dementia: cluster randomised Journal of Alzheimer’s Disease and Other Dementias, 21, 326–32.
trial. Bristish Medical Journal, 332, 756–8. Kolanowski, A., Fick, D.M., and Buettner, L. (2009). Recreational activities
Gaugler, J.E., et al. (2004). Family involvement in nursing homes: effects on to reduce behavioural symptoms in dementia. Geriatric Aging, 12(1),
stress and well-being. Aging and Mental Health, 8, 65–75. 37–42.
Gibson, F. (1994). What can reminiscence contribute to people with dementia? Lefebvre, S.L., et al. (2008). Guidelines for animal assisted interventions
In: Bornat, J. (ed.) Reminiscence reviewed: evaluations, achievements, in health care facilities. American Journal of Infection Control, 36(2),
perspectives, pp. 46–60. Open University Press, Buckingham. 78–85.
Gibson, M.C., et al. (2004). Orientation behaviors in residents relocated to a Libin, A. and Cohen-Mansfield, J. (2004). Therapeutic robocat for nursing
redesigned dementia care unit. American Journal of Alzheimer’s Disease home residents with dementia: preliminary enquiry. American Journal of
and Other Dementias, 19, 45–9. Alzheimer’s Disease and Other Dementias, 19, 111–16.
Livingston, G., et al. (2005). Systematic review of psychological approaches to Romero, B. and Wenz, M. (2001). Self-maintenance therapy in Alzheimer’s
the management of neuropsychiatric symptoms of dementia. American disease. Neuropsychological Rehabilitation, 11, 333–55.
Journal of Psychiatry, 162(11), 1996–2021. Schrijnemaekers, V., et al. (2002). Effects of emotion oriented care on
Lord, T. and Garner, E. (1993). Effects of music on Alzheimer patients. elderly people with cognitive impairment and behavioural problems.
Perceptual and Motor Skills, 76, 451–5. International Journal of Geriatric Psychiatry, 17, 926–37.
Mackenzie, L., et al. (2006a). A pilot study on the use of dolls for people with Sherratt, K., Thornton, A., and Hatton, C. (2004). Music interventions for
dementia. Age and Ageing, 35(4), 441–4. people with dementia: a review of the literature. Aging and Mental
Mackenzie, L., Wood-Mitchell, A., and James, I.A. (2006b). Guidelines on the Health, 8, 3–12.
use of dolls in care settings. Centre for the Health of the Elderly, Newcastle Singh, N., et al. (2005). A randomized controlled trial of high versus low
General Hospital, Newcastle upon Tyne intensity weight training versus general practitioner care for clinical
Mackenzie, L., Wood-Mitchell, A., and James, I.A. (2006c). Thinking about depression in older adults. Journals of Gerontology. Series A Biological
dolls. Journal of Dementia Care, 14(2), 16–17. Sciences and Medical Sciences, 60, 768–76.
Mackenzie, L., Wood-Mitchell, A., and James, I.A. (2007). Guidelines on Spector, A., et al. (2002). Reality orientation for dementia. Cochrane Library:
using dolls. Journal of Dementia Care, 15(1), 26–7. Update Software, Issue 2, doi: CD001119.
Marriott, A., et al. (2000). Effectiveness of cognitive-behavioural family Spector, A., et al. (2003). Efficacy of an evidence-based cognitive stimulation
intervention in reducing the burden of care in carers of patients with programme for people with dementia: randomised controlled trial.
Alzheimer’s disease. British Journal of Psychiatry, 176, 557–62. British Journal of Psychiatry, 183, 248–54.
Miller, M. (2009). Clinician’s guide to interpersonal psychotherapy in late Spector, A., et al. (2006). Making a difference: an evidence-based group
life: helping cognitively impaired or depressed elders and their caregivers. programme to offer cognitive stimulation therapy (CST) to people with
Oxford University Press, Oxford. dementia. Hawker Publications, London.
Miller, M. and Reynolds, C.F. (2002). Interpersonal psychotherapy. In: J. Spector, A., Woods, B., and Orrell, M. (2008). Cognitive stimulation for the
Hepple, J. Pearce, and P. Wilkinson (eds) Psychological therapies with treatment of Alzheimer’s disease. Expert Review of Neurotherapeutics,
older people. Bruner-Routledge, Hove. 8(5), 751–7.
Miller, M. and Reynolds, C.F. (2007). Expanding the usefulness of interpersonal Spira, A. and Edelstein, B. (2006). Behavioral interventions for agitation
psychotherapy (IPT) for depressed elders with comorbid cognitive in older adults with dementia: an evaluative review. International
impairment. International Journal of Geriatric Psychiatry, 11, 97–102. Psychogeriatrics, 18(2), 195–225.
Moniz-Cook, E., et al. (1998). Can staff training reduce carer stress and Stokes, G. (2000). Challenging behaviour in dementia: a person-centred
behavioural disturbance in the elderly mentally ill? International Journal approach. Speechmark, Bicester.
of Geriatric Psychiatry, 13, 149–58. Suzuki, M., et al. (2004). Behavioral and endocrinological evaluation of
Moniz-Cook, E., et al. (2012). Functional analysis based interventions for music therapy for elderly patients with dementia. Nursing and Health
challenging behaviour in dementia. Cochrane Database of Systematic Sciences, 6, 11–18.
Reviews, doi: 10.1002/14651858.CD006929.pub2. Teri, L. and Gallagher-Thompson, D. (1991). Cognitive-behavioral
Montgomery, P. and Dennis, J. (2002). Physical exercise for sleep problems in interventions for treatment of depression in Alzheimer’s patients.
adults aged 60+. Cochrane Database of Systematic Reviews, 4, CD003404, Gerontologist, 31(3), 413–16.
doi: 10.1002/14651858. Teri, L., et al. (1997). Behavioural treatment of depression in dementia patients:
Mottram, P. (2003). Art therapy with clients who have dementia. Dementia, a controlled clinical trial. Journal of Gerontology, 52(B), 159–66.
2, 272–7. Teri, L., et al. (2005). Training community consultants to help family members
National Audit Office (2007). Improving services and support for people with improve dementia care: a randomized controlled trial. Gerontologist,
dementia. HC 604 (July). National Audit Office, London. 45(6), 802–11.
National Institute For Health and Clinical Excellence (NICE) (2006). Teri, L., Logsdon, R.G., and McCurry, S.M. (2008). Exercise interventions
Dementia: supporting people with dementia and their carers. Clinical for dementia and cognitive impairment: the Seattle Protocols. Journal of
Guideline 42. <www.nice.org.uk>. Nutrition and Health Aging, 12, 391–4.
Neal, M. and Barton Wright, P. (2003). Validation therapy for dementia. Testad, I., et al. (2010). Nursing home structure and association with
Cochrane Database of Systematic Reviews, 3, CD001394, doi: agitation and use of psychotropic drugs in nursing home residents in
10.1002/14651858. three countries: Norway, Austria and England. International Journal of
Nguyen, Q.A. and Paton, C. (2008). The use of aromatherapy to treat Geriatric Psychiatry, 25(7), 725–31.
behaviour problems in dementia. International Journal of Geriatric Thorgrimsen, L., et al. (2003). Aroma therapy for dementia. Cochrane Database
Psychiatry, 23(4), 337–46. of Systematic Reviews, 3, CD003150. doi: 10.1002/14651858.
Onishi, J., et al. (2006). Behavioral, psychological and physical symptoms Thwaites, S. and Sara, A. (2010). Tailor made cognitive stimulation. Journal of
in group homes for older adults with dementia. International Dementia Care, 18(5), 19–20.
Psychogeriatrics, 18(1), 75–86. Tondi, L., et al. (2007). Validation therapy (VT) in nursing home: a
Opie, J., Doyle, C., and O’Connor, D.W. (2002). Challenging behaviours in case-control study. Archives of Gerontology and Geriatrics, 44 Suppl 1,
nursing home residents with dementia: a randomised controlled trial 407–11.
of multidisciplinary interventions. International Journal of Geriatric Toseland, R.W., et al. (1997). The impact of validation group therapy on
Psychiatry, 17, 6–13. nursing home residents with dementia. Journal of Applied Gerontology,
Pachana, N., Laidlaw, K., and Knight, B. (2010). Casebook of clinical 16, 31–50.
geropsychology. Oxford University Press, Oxford. Train, G., et al. (2005). A qualitative study of the views of residents with
Proctor, R.,. et al. (1999). Behavioural management in nursing and residential dementia, their relatives and staff about work practice in long term care
homes: a randomised control trial. Lancet, 354, 26–9. settings. International Psychogeriatrics, 17, 237–51.
Raglio, A., et al. (2008) Efficacy of music therapy in the treatment of Van Weert, J.C., et al. (2005a). Behavioral and mood effects of snoezelen
behavioural and psychiatric symptoms of dementia. Alzheimer Disease integrated into 24-hour dementia care. Journal of the American Geriatrics
and Associated Disorders, 22(2), 158–62. Society, 53, 24–33.
Rolland, Y., et al. (2007). Exercise program for nursing home residents with Van Weert, J.C., et al. (2005b). Effects of snoezelen, integrated in 24 h
Alzheimer’s disease: a 1-year randomized, controlled trial. Journal of the dementia care, on nurse–patient communication during morning care.
American Geriatrics Society, 55, 158–65. Patient Education and Counseling, 58, 312–26.
Verkaik, R., Van Weert, J., and Francke, A. (2005). The effects of psychosocial Woods, D.L., Rapp, C.G., and Beck, C.E. (2004). Escalation/de-escalation
methods on depressed, aggressive and apathetic behaviours of people patterns of behavioral symptoms of persons with dementia. Aging and
with dementia: a systematic review. International Journal of Geriatric Mental Health, 8(2), 126–32.
Psychiatry, 20, 301–14. Woods, B., et al . (2005a). Reminiscence therapy for dementia.
Vink, C., Bruinsma, M., and Scholten, R. (2009). Music therapy for people Cochrane Database of Systematic Reviews, 2, CD001120.pub2, doi:
with dementia. Cochrane Dementia and Cognitive Improvement Group, 10.1002/14651858.
doi: 10.1002/14651858.CD003477.pub2. Woods, B., et al. (2005b). Cognitive stimulation to improve cognitive
Vogel, T., et al. (2009) Health benefits of physical activity in older patients: a functioning in people with dementia. Cochrane Database of Systematic
review. International Journal of Clinical Practice, 63(2), 303–20. Reviews, Protocols Issue, 4, CD005562, doi: 10.1002/14651858.
Wenbourn, J. (2003). Using a sensory approach to improve well-being. Young, A. and Dinan, S. (1994). ABC of sports medicine: fitness for older
Nursing and Residential Care, 5, 431–2. people. British Medical Journal, 309, 331–4.
Wilkinson, N., et al. (1998). Drama and movement therapy in dementia: a Zisselman, M.H., et al. (1996). A pet therapy intervention with geriatric psychiatry
pilot study. Arts in Psychotherapy, 25, 195–201. inpatients. American Journal of Occupational Therapy, 50, 47–51.
Wood-Mitchell, A., et al. (2008). Factors influencing the prescribing Zuidema, S., et al. (2010). Environmental correlates of neuropsychiatric
of medications by old age psychiatrists for behavioural and psychological symptoms in nursing home patients with dementia. International Journal
symptoms of dementia: a qualitative study. Age and Ageing, 3, 1–6. of Geriatric Psychiatry, 25(1), 14–22.
CHAPTER 22
Principles of service
provision in old
age psychiatry
Tom Dening
The past decade has seen considerable changes in the shape of men- London, was also highly influential, for example addressing the
tal health services for older people. This has sometimes been so Annual Meeting of the Royal Medico-Psychological Association
radical as to include the disappearance of the service as a separate in 1945 (Lewis, 1946) and being active through the Mental Health
entity, and yet this is at a time when the population of older people Standing Advisory Committee of the new National Health Service
is growing and therefore it might be assumed that the demand for (Hilton, 2005a).
specialist services would also be increasing. How this can be so is Several accounts are available of further developments of old age
the story of this chapter. psychiatry (e.g. Arie, 1989; Shulman and Arie, 1991). Important
There are two main strands to service provision to support men- advances included seminal clinical studies, especially those by Felix
tal health in older people. One is the general question of how men- Post (e.g. Post, 1962, not to mention his seminal textbook; Post,
tal health needs in this section of the population are to be met, and 1965) and Martin Roth (1955). These provided evidence that not
the other topic is the organization of specialist mental health serv- all mental disorders could be ascribed to senility and opened the
ices for older people. These strands are interwoven but also at risk possibility of offering effective treatments for certain disorders,
of unwinding from each other. especially affective states. Services dedicated to older people began
In this chapter, the focus is on the principles underlying serv- to open in a patchy fashion across the UK (e.g. Arie, 1970). The
ices, rather than on much detail, partly because the situation will common interests of psychiatrists working in this area led first
vary in different parts of the world but also partly because the way to the formation of an informal gathering, followed by an official
services are organized will continue to change, rendering any- Group within the Royal College of Psychiatrists in 1973, and then
thing too detailed obsolete in a short space of time. The chapter a full Specialist Section in 1978. Old age psychiatry was recognized
takes a largely UK, specifically England, focus. Many of the gen- by the UK Department of Health as a separate speciality in 1989.
eral issues are, however, global and also, as the UK has been one Numbers of consultant old age psychiatrists have continued to rise,
of the leaders in developing service models for old age psychiatry, the most recent estimate for England being 508 (<www.ic.nhs.uk/
it is likely that what happens there will be of interest to readers pubs/nhsstaff>).
internationally. For several years, the development of old age psychiatry pro-
ceeded apart from that of geriatric medicine, despite the obvious
overlap in clientele. The reasons for this were complex, not least
History a dismissive attitude of the physicians. More common ground
It is generally accepted that old age psychiatry originated in was reached by a joint Royal Colleges document on the care
the UK. The earliest traces of the speciality are to be found in of older people with mental illness, produced in 1989 and more
the early 1940s (Hilton, 2005a). Before the Second World War, recently updated (Royal College of Psychiatrists/Royal College of
there was virtually no interest in older people’s mental or physi- Physicians, 1998). Another area that lagged behind the developing
cal health issues, most conditions being assumed to be degenera- clinical services has been policy. Until the publication of Services
tive, ‘senile’, and not amenable to treatment. By the early 1940s, for Mental Illness Related to Old Age (Department of Health and
there were large numbers of older people with mental disorders Social Security, 1972), there had been no conceptual progress for
in hospitals like the Royal Edinburgh Hospital and Tooting Bec in over 20 years (Hilton, 2005b).
London. Felix Post, at first working in Edinburgh and encouraged Developments in other countries have been influenced to
by Professor Henderson, published one of the first clinical stud- varying degrees by the UK experience. Patterns of service reflect
ies of mental disorders in old age (Post, 1944). Aubrey Lewis, in national factors such as the nature of the healthcare system, the
degree to which old age psychiatrists are generalists or dementia

Box 22.1 National Dementia Declaration for England (Dementia
specialists, and of course the level of resources in the individual Action Alliance, 2010)
country. Snowdon and Arie (2005) have described service devel-
opments and also the establishment of national and international
1. I have personal choice and control or influence over decisions
organizations for old age psychiatry. Particularly influential
about me.
has been the International Psychogeriatric Association (IPA),
founded in 1982. The World Psychiatric Association (Camus et 2. I know that services are designed around me and my needs.
al., 2003) found that 40 out of 48 countries responding provided 3. I have support that helps me live my life.
some specific mental health services for older people, and 13 of
these countries recognized old age psychiatry as a separate speci- 4. I have the knowledge and know-how to get what I need.
ality. Over 20 countries had at least one academic chair in old age 5. I live in an enabling and supportive environment where I feel
psychiatry. The World Health Organization Atlas records 51% of valued and understood.
185 countries as having a national mental health programme for 6. I have a sense of belonging and of being a valued part of fam-
older people (<www.who.int/mental_health/evidence/atlas>). ily, community, and civic life.
Many of the countries that do not are in Africa and the Middle
East. 7. I know there is research going on that delivers a better life for
me now and hope for the future.
Principles and Values

One of the strengths of old age psychiatry, besides its consideration Policy
of epidemiological factors, has been a keen interest in the values
underlying its practice. For example, Arie and Jolley (1982) listed United Kingdom
five principles: flexibility; responsiveness and availability; unhier- In England, there is now no single strand of policies relating to
archical use of staff; domiciliary assessment; and willingness to col- older people’s mental health. There is instead a separation between
laborate with other services and agencies. The list has been amended dementia, on the one hand, and other mental disorders, on the
in various places, but the most general statements of values are to other. This has come about because under the Labour administra-
be found in series of consensus statements, jointly issued by the tion (1997–2010), as part of a series of National Service Frameworks
World Health Organization and the Geriatric Psychiatry Section of (NSFs) for various conditions, there were separate NSFs for Mental
the World Psychiatric Association. Health (Department of Health (DH), 1999) and Older People (DH,
Altogether, there are five consensus statements, concerning psy- 2001a). The first of these covered mental health up to age 65, and
chiatry of older people (WHO, 1996), organization of care (WHO, the latter incorporated mental health issues in older people along
1997), education (WHO, 1998), reducing stigma and discrimina- with physical illnesses like stroke. The NSF for Older People (NSF
tion against older people with mental disorders (WHO, 2002), and OP) emphasized depression and dementia as the most important
ethical practice (Katona et al., 2009). Of these, that on organiza- mental disorders of later life, but it drew some criticism as it was
tion of services has probably been most influential and it is the not resourced to bring about the service improvements that were
most relevant to this chapter. As well as setting out seven underly- needed to achieve its aims.
ing principles (forming the acronym CARITAS; Comprehensive, Subsequently, because of concerns about possible age discrimina-
A ccessible, R esponsive, I ndividualized, Transdisciplinary, tion, mental health policy has been extended across the age range.
Accountable, and Systemic), the document describes the various On the other hand, it is inescapably true that dementia is a major
types of care that may be provided and the components of special- public health and social challenge, so it has had attention in its own
ist services. The most recent consensus statement (Katona et al., right. Therefore, to understand the current picture, it is necessary
2009) discusses the needs for the application of high ethical stand- to look at both these strands. The extent to which they are joined
ards in clinical practice, especially in relation to mental capacity together is questionable. Mental health has its own branch within
and decision-making. the DH, and dementia is the responsibility of Social Care.
Values in relation to dementia have been set out by the National Mental health policy in England is currently set out in No Health
Dementia Declaration for England, produced by the Dementia Without Mental Health (NHWMH) (HM Government, 2011). It
Action Alliance (2010), a partnership with 41 signatories repre- focuses on six major outcomes, which include good mental and
senting government, professional bodies, and voluntary and inde- physical health, recovery, good experiences of care and support,
pendent sector organizations. The Declaration, which is part of the and reduced harm, stigma, and discrimination. Improving men-
implementation of the National Dementia Strategy for England tal health outcomes for older people is discussed in the section
(described in the section on Policy), sets out seven outcomes for of NHWMH that concerns equality and occupies just four para-
people with dementia (Box 22.1). There is a welcome focus on peo- graphs. These deal with: depression and access to services; demen-
ple with dementia and enabling them to make their own choices tia; depression in primary care; and ending age discrimination.
and decisions wherever possible. It reflects the ethos within which NHWMH does not anywhere address what mental health services
mental health services for older people are operating within a wider for older people might look like: indeed, there is no mention of
context of personalized care and support. The Declaration calls on whether specialist services for this population should be part of the
individuals and organizations to sign up to it and publicize the picture.
work being done to deliver the National Dementia Strategy and As regards dementia, policy in England is set out in the National
these outcomes in particular. Dementia Strategy (NDS) (DH, 2009) and subsequent offerings
CHAPTER 22 principles of service provision in old age psychiatry 285
from the DH. There had been calls for such a strategy for several developments, providing information about dementia and availa-
years (e.g. Dementia UK (Alzheimer’s Society, 2007)). The NDS sets ble care services, as well as allocating funds for research and evalu-
out 17 objectives, including improved public awareness, early diag- ation of the Initiative itself. The strategy is currently under review
nosis and intervention, improved quality of care in various settings, with the intention of issuing a new strategy from 2012 onwards (see
research, and implementation. Subsequently, implementation has <http://www.health.gov.au/> for further updates).
focused on a smaller number of priority objectives (DH, 2010),
on the basis of their high importance and potential for successful The Nature and Scope of Old Age
attainment. These are early diagnosis and intervention; improved
quality of care in general hospitals; living well with dementia in care
Psychiatry Services
homes; and reduced use of antipsychotic medication. Although the This chapter is mainly concerned with specialist services for older
last of these was not a specific objective of the NDS, it has become people with mental health problems. The boundaries of this activity
a central component of policy following concerns about excessive are rather imprecise, and most interactions with older people and
prescribing of antipsychotic drugs and the attendant risks (Banerjee, mental disorders will be in social care settings, such as day centres
2009). It has been estimated that there were about 180,000 people and home care services, or in primary care, where most common
with dementia in the UK on antipsychotic drugs. In only about disorders, such as depression, frequently present. The contributions
one-third of these were the drugs likely to be beneficial, with about of primary care and social care are discussed in Chapters 23 and 26,
1800 excess deaths per year as a result of their prescription. respectively.
The plan for implementation also set out several further initiatives. There are also variations as to the boundaries of specialist serv-
These included the appointment of a National Clinical Director for ices. In some countries, especially if old age psychiatry is not a
Dementia (Professor Alistair Burns), supported by three National recognized speciality, this work may be undertaken by general psy-
Dementia Champions, to cover the fields of NHS, social care, and chiatrists, geriatric physicians, or neurologists. Some services limit
the independent sector. Another area for new investment has themselves to dementia, while others have a more comprehensive
been research and, following the recommendations of a ministe- basis.
rial group, four new Biomedical Research Units for Dementia have The term used for the speciality can also vary. In the UK, ‘psycho-
been funded, the emphasis being on translational research, that is, geriatrics’ has largely been superseded by ‘old age psychiatry’, as its
putting research findings into practice. Other initiatives appear at practitioners have emphasized their links with the rest of psychia-
<http://www.dh.gov.uk/en/SocialCare/NationalDementiaStrategy/ try, rather than with geriatric physicians. In the US, the relevant
index.htm>, including compendia of good practice, workforce speciality is ‘geriatric psychiatry’, though this is often a rather dif-
information, and information for patients and carers. ferent activity from UK old age psychiatry, as the integral links with
Space precludes a fuller discussion of the whole policy frame- community teams are not always present to the same degree, and
work of relevance to older people. Obviously they are particularly dementia is more the concern of neurologists. Blazer (2000) criti-
affected by reforms to health services and changes in social care cized geriatric psychiatry in the US for moving away from the prin-
policy and funding. In the wider economy, the current debate over ciples of comprehensive, interdisciplinary assessment and therapy,
pensions will affect the material prosperity of the next generations and tending to abandon frail, very old people in favour of those
as they grow older, with people likely to continue working longer. nearer 65 with less complex needs.
Some more relevant issues, e.g. the roles of public bodies in regulat- However, despite these problems of terms and boundaries, the
ing services and promoting best practice, are discussed later in this following section is mainly about the model of old age psychiatry
chapter. that has developed from these various influences. There are two
The other devolved nations of the UK—Scotland, Wales, and accounts to consider: one is the ‘traditional’ model of old age psy-
Northern Ireland—all have similar initiatives to those in England chiatry, and the other is to look at more recent developments that
(see Scottish Government, 2009, 2010; Department of Health, Social move away from this simpler pattern to a more diverse array or
Services and Public Safety, 2011; Welsh Assembly Government and services. The manner in which services have developed owes much
Alzheimer’s Society, 2011). to the energy and innovation of early old age psychiatrists, such
as Tom Arie and Raymond Levy. In more recent years, there has
Other countries been increasing attention to more formal evaluations of services,
At the time of writing, there are four other European countries with with randomized controlled trials (RCTs) and several systematic
a national dementia strategy, namely Norway (published 2007), reviews, including overviews of the whole enterprise (e.g. Draper,
France (2008), the Netherlands (2008), and Denmark (2010). Several 2000; Bartels et al., 2002, 2003; Van Citters and Bartels, 2004;
other countries are committed to developing national responses and Draper and Low, 2005a, 2005b).
work is already underway in Switzerland and Belgium (Alzheimer
Europe, 2011). The European Commission (2009) has also issued Components of Old Age Psychiatry Services:
advice to the European Parliament and its Council, highlighting The ‘Traditional’ Model
four issues—early diagnosis and prevention, research, sharing of
good practice across member states, and respect for the rights of Community mental health teams
people with dementia. The NSF OP (DH, 2001a) described a service model for commu-
Probably the first country with a national dementia strategy was nity mental health teams (CMHTs) that reflected agreed norms at
Australia, which launched its Dementia National Health Priority that time. The core members of the specialist mental health serv-
Initiative in 2005. Now known as the Dementia Initiative, this ice include consultant old age psychiatrists, community mental
was a detailed strategy that made provision for national and local health nurses, clinical psychologists, occupational therapists, and
social workers. Additional untrained staff, such as health and social six reported that the team input was more effective than the control
care assistants, can make innovative contributions (McCrae et al., intervention, findings supported by data from several uncontrolled
2008), and adequate administrative support is essential. Agreed studies.
working and referral arrangements with other professionals, such It is often held that assessment and management should take
as speech and language therapists, physiotherapists, and dieticians, place at home where possible, rather than, say, in outpatient clinics.
are also required, though these disciplines are not usually mem- The advantages of this include being able to take in aspects of the
bers of the team itself. Most CMHTs have this structure, although patient’s social surroundings in one visit, as well as being highly
variations are often seen in practice. Most frequently, one or more acceptable to many older people. It has been demonstrated that ini-
professional groups may be lacking—often clinical psychology as tial assessments by nonmedical members of the multidisciplinary
there is a lack of practitioners working with old people. There may team can be effective (Collighan et al., 1993). Case management,
just be a single consultant psychiatrist for a team, and the consult- in various forms, often known in the UK as the Care Programme
ant may work somewhat detached from the team, perhaps taking Approach (CPA), is an effective way of ensuring that patients are
separate referrals but also having other commitments elsewhere, kept in contact with the service. Care management seems to be
e.g. to inpatient units. effective at keeping people with dementia at home for longer peri-
CMHTs characteristically cover an agreed sector, which may be ods (Challis et al., 2002) and, in general, outcomes seem to be better
defined geographically or else by which general practices refer to for team management approaches than for simply providing consul-
the team. In the UK, teams usually deal with the full range of men- tations and advice (Woods et al., 2003). It is accepted that specialist
tal disorders, not solely dementia. There will be an agreed route teams for older people are more effective at managing depression
of referral, with most referrals in practice coming from primary in older people than general adult mental health teams, though this
care, but also from other sources, including social services, residen- has not been formally evaluated (Draper and Low, 2005a).
tial homes, and sometimes directly from families. Patients appear As regards the specific role of psychiatry, consultant domiciliary
reluctant to refer themselves, even to teams with open referral poli- visits are probably best targeted at more complex cases, requiring
cies. Teams are of variable size, often smaller in rural areas. Within diagnosis and the initiation of treatment: studies of the outcome
a larger team, there is scope for some subdivision of responsibilities of consultant home visits are not easy to interpret because of this
or for specialization. For example, each member of the team may be (Orrell and Katona, 1998). Home-based practice lessens the rate of
linked to an individual GP surgery or have links with one or more failed appointments and also may cost less than running an outpa-
residential or nursing homes in the patch. Or they may develop tient clinic. However, many old age psychiatrists continue to hold
special roles, such as monitoring patients on antidementia drugs. outpatient clinics in addition to seeing patients at home. Such clin-
Leadership in teams is a commonly discussed theme, but there ics may be very useful for following up relatively mobile patients,
are several aspects to how a team is led. For example, the CMHT often those with functional disorders. For example, they can assist
will be managed as part of the service to which it belongs; the staff with managing certain patients on lithium treatment using shared
in it will be line managed; they will also have their own professional protocols with primary care (Head and Dening, 1998).
group hierarchies, which may include supervision of their clini- Thus, CMHTs for older people are well established in various
cal work; and there may a designated team leader, whose role will countries, and there is good evidence to support this style of serv-
include such matters as chairing the referral allocation meeting. As ice. There remain several issues where further research evidence
well as this, consultant psychiatrists provide clinical leadership to would be helpful. For example, there are no recent studies of crisis
the CMHT, though this is a complex construct. It seems to include intervention services for older people (Ratna, 1982), and there are
a commitment to be involved in the assessment and management no studies that evaluate the outcomes from different members of
of the more complicated cases and to act as one of the main con- the multidisciplinary team.
duits between the CMHT and other parts of the service, especially As has been emphasized in earlier chapters, numbers of older
inpatient care. people in developing countries, and consequently numbers of peo-
There is good evidence that CMHTs for older people offer effective ple with dementia, are increasing more sharply than in the devel-
interventions. One of the first evaluations of the impact of a com- oped world. This, combined with the limited resources that may be
munity team was by O’Connor et al. (1991) who found that early available for mental health services, requires new methods of case
intervention led to increased admissions to long-term care within a finding and service provision (e.g. Shaji et al., 2002). Services need
2-year period for people with dementia living alone. This may have to have realistic goals, they should be based on careful mapping of
reflected that this group was in fact living in conditions of high risk all local resources that may be supportive of older people, and they
and relative neglect. More recently, Callahan et al. (2006) found can work through general health workers with suitable brief train-
that an interdisciplinary team working collaboratively with primary ing (Dening and Shaji, 2005).
care provided significant improvements in the quality of care and
in behavioural and psychological symptoms of dementia among Memory assessment services and memory clinics
primary care patients and their caregivers. Rothera et al. (2008) These are discussed in Chapter 23. It has always been a function
found that a multiagency specialist support team for older people of old age psychiatry services to provide diagnostic assessments
with dementia achieved better outcomes than standard services. for people with suspected dementia, but it is only more recently
Occupational therapy interventions have also been the subject of that there has been more attention paid to this activity in its own
randomized trials in Europe, with rather conflicting results despite right. This reflects a body of evidence that demonstrates the ben-
initial promise (Graff et al., 2006; Voigt-Radloff et al., 2011). efits and cost effectiveness of early diagnosis in dementia (National
However, most RCTs of interventions by CMHTs have concen- Collaborating Centre for Mental Health, 2007; Banerjee and
trated on depression (Draper and Low, 2005a). Of these, four out of Wittenberg, 2009).
Memory clinics are often, though not necessarily, hospital clinic patients being treated for shorter episodes of depression. Moreover,
based. They were first established in the late 1980s and early 1990s, most day hospitals will have a small number of patients, whose
often by geriatricians and neurologists as well as psychiatrists. They exact diagnosis is difficult to pinpoint but whose personalities are
became more widespread and increasingly the domain of old age so difficult that they do not fit into ordinary day centres. In rural
psychiatry (Lindesay et al., 2002). They are an efficient vehicle for areas, because of distance and travelling time, it may be necessary
diagnostic assessment and investigation of possible dementia and to hold the day hospital in different localities on different days,
mild cognitive impairment, and were often set up for research pur- which means that on each day there will be a more heterogeneous
poses such as recruiting patients to trials of antidementia drugs. group of patients.
However, the initial conception of memory clinics has some limi- A national survey of day hospitals, coordinated by the Royal
tations and has attracted some criticisms (Pelosi et al., 2006). These College of Psychiatrists, confirmed that there is wide diversity in
include waiting lists caused by the complex protocols of some clin- the style and use of day hospital facilities across the UK (Audini
ics, an overemphasis on medication, and a lack of integration with et al., 2001). The majority of old age psychiatrists had access to day
social care and other community support. It remains the case, how- hospitals, over half the patients (56%) had dementia, and there
ever, that memory clinics can be an effective means of recruiting were variations in treatments offered and lengths of stay. This diver-
patients for trials of new drugs, and they can provide rich sources sity stimulated the formation of a day hospital network to encour-
of longitudinal data that have done much to illuminate the variable age communication and share experiences between day hospital
natural history of dementia. providers.
In response to these issues, there has been a tendency for memory Thus, given the mixed nature of day hospitals, and the hetero-
clinics to broaden their scope, focusing less on academic research geneity of patients attending, it is scarcely surprising that the sim-
and more on early cognitive problems (Ramakers and Verhey, ple question ‘Do day hospitals work?’ is virtually unanswerable.
2011). However, the emphasis is now more on memory services, Different patients will be aiming at different outcomes, so using
which can be more population based, providing assessments and simple outcome measures could miss significant benefits. There are
interventions closer to home. An example is the Croydon Memory relatively few studies of outcomes from day hospital attendance. The
Service in south London, which has been well described and evalu- only RCT (Ashaye et al., 2003) was not primarily about outcomes
ated (Banerjee et al., 2007). This service achieved six preset goals, but was an evaluation of a standardized needs assessment (the
namely high acceptability; high appropriate referral rate; success- Camberwell Assessment of Need for the Elderly (CANE); Orrell
ful engagement with people from minority ethnic groups; success- and Hancock, 2004). However, both groups, i.e. those assessed by
ful engagement with people with young onset dementia; focus on the CANE and the controls, showed improvements in unmet needs
engagement with mild cases to enable early intervention; and an and Health of the Nation Outcome Scale (HoNOS 65+) scores over
increase in the overall number of new cases of dementia seen. a 3-month period, even though physical dependency and behav-
iour problems increased. Other studies provide evidence that day
Day hospitals hospitals are effective in improving mental health outcomes, espe-
Day hospitals (often referred to in the US as partial hospitalization cially for depression (e.g. Mackenzie et al., 2006). As a form of res-
programs) have historically been regarded as a key component of pite care, day care appears to be more expensive than usual care
old age psychiatry services, though their efficacy is often called and provides only slight benefits at best for carers (Mason et al.,
into question. In the UK, day hospitals are generally distinguished 2007a, 2007b). Furthermore, for older people with physical health
from day centres, the latter being run by social care providers and problems, such as rehabilitation and falls prevention, day hospitals
the voluntary sector, whereas day hospitals are part of the health confer no advantages over care at home (Parker et al., 2009; Conroy
service. There are therefore differences in the types of staffing to be et al., 2010).
found in each type of facility. Day hospitals are more likely to focus There is a distinct lack of comparisons between day hospitals and
on assessment and therapeutic activities (Reilly et al., 2006), and either inpatient or community treatment, so their role in provid-
are generally looking to discharge patients elsewhere, either after ing acute care is so far unresolved. It is more likely, however, that
an agreed period of attendance or once a suitable outcome has been day hospitals are a useful adjunct to discharge from inpatient care,
achieved. Day centres do not have the same emphasis on assess- rather than in preventing admissions. Certainly, despite indications
ment, activities are generally diversionary rather than therapeutic, of their effectiveness, a more robust evidence base for day hospitals
and indefinite attendance is not usually a problem. An important is needed (Hoe et al., 2005).
difference, certainly in the UK, is that day hospitals, as part of the
National Health Service, do not charge for attendance, and often Respite care
transport is easier to arrange (though sometimes equally unreli- Respite care comprises a mixture of activities, aimed at giving car-
able!). In practice, however, there is considerable overlap. Critics of egivers temporary opportunities to be away from the patient in
day hospitals have pointed out a lack of therapeutic activities, ina- order to reduce their stress and to delay or prevent nursing home
bility to move patients on to other facilities, and a lack of evidence admission. There are broadly three types of respite: short residential
that day hospitals prevent hospital admissions (Fasey, 1994). placements, day care, and home care. Nowadays, short-stay respite
Day hospitals themselves are quite variable, especially depend- care is generally provided in residential and nursing homes rather
ing on the relative proportions of patients with functional disorders than in hospitals. Respite care may be planned, offered in a crisis, or
and dementia. Many organize for these groups to attend largely on at the request of the carer.
separate days, as the type of programme appropriate to each may be Several evaluations of respite care were conducted in the 1980s
quite different. Some day facilities cater for patients with long-term and 1990s to see if the service was effective at delaying or avoid-
severe mental illness and, again, this group has different needs from ing admission to long-term care (reviewed by Melzer et al., 1999),
but there was no conclusive evidence in terms of direct benefits to roles. More formal programmes have been developed and evalu-
patients, relief of carer burden, or delay in admission to long-term ated (Mittelman et al., 1996; Brodaty et al., 1997), which comprise
care. More recent systematic reviews of ‘frail older people’ (a cat- various elements, such as improving carers’ knowledge of mental
egory that includes many people with dementia as well as physical health problems and teaching exercises to deal more effectively
health problems) (Mason et al., 2007a, 2007b; Shaw et al., 2009) with challenging behaviours. Less intensive forms of carer educa-
found evidence of some benefit to carers in terms of decreased tion programmes are commonly provided, but these have not so far
burden and improved mental and physical health. There was no been shown to be as effective in reducing carer stress.
reliable evidence that respite either benefits or adversely affects the One particular model of working with carers is that of Admiral
care recipient, or that it delays entry into residential care. There is Nurses. These are specialist mental health nurses working in the
little evidence as to the most effective forms of respite and the only field of dementia. The emphasis in their work is on working with
economic evaluations have been of day care, which appears rela- family carers, as well as with the person with dementia. The concept
tively costly as a form of respite. A review of the effects of residen- of Admiral Nurses belongs to the charity Dementia UK (<http://
tial respite care on the behaviour of older people with dementia www.dementiauk.org/>), which was founded to respond to the
found variable results, with a modest but short-term reduction in unmet needs of family carers. It is a popular model with carers and
symptoms in some studies (Neville and Byrne, 2007). families as the nurse remains in contact with them throughout the
However, in practice, carers do use respite care when it is availa- patient’s journey. There is a challenge for the model to demonstrate
ble and they often describe high levels of satisfaction with it. Respite improved outcomes, but, more pressingly, as the nurses are often
care may be used by some carers to help them keep the patient at funded on a start-up basis, they are vulnerable to cost saving meas-
home for longer, whereas, for others, respite care functions as a ures in local services.
stepping stone on the way to long-term care, so it is not surprising Carers also have an important role to play in the formal and infor-
that there is no demonstrable effect in delaying admission into resi- mal education of personnel training in old age psychiatry and other
dential care. Sometimes, perhaps, respite care is offered when what professional groups; and they can provide a valuable independent
is needed is really long term, in which situation it may be regarded perspective in the evaluation of mental health services (Dening and
as a form of rationing. Overall, it does appear that providing respite Lawton, 1998). They are often the most passionate and effective
in various forms, in partnership with other agencies, is a legitimate lobbyists on behalf of people with dementia, especially. Carers are
activity for old age psychiatry services. important forces in organizations such as the Alzheimer’s Society
in England and equivalent organizations elsewhere.
Support to carers
Old age psychiatry was perhaps one of the first areas where carer Inpatient care
issues were first recognized (e.g. Argyle et al., 1985). Carers of peo- In general, wards for older patients are separate from those for
ple with dementia seem to have particularly high levels of stress adults under 65. Experience of wards for mixed ages is that they
and lower levels of satisfaction with services than other carers (e.g. disadvantage older people, who may be relatively neglected in
Bedford et al., 1996). Consequently, assessments of patients also the face of demands placed by younger, psychotic, and disturbed
routinely include assessment of the carers’ mental state and coping patients. Perhaps surprisingly, the evidence from a small number
resources. Indeed, in the UK, carers are legally entitled to their own of studies of older patients on general psychiatry wards suggests
separate assessments. More recent literature has been critical of the that clinical outcomes are acceptable, but in most of these studies
emphasis on the problems of caregiving, which can be overstated an old age psychiatrist was attached to the unit (Draper and Low,
in clinical samples. Instead, we now have a more balanced view of 2005a). It is also customary to separate dementia and functional
caregiving as a normal part of life, something that brings joy and illness, since experience suggests that depressed patients find the
rewards as well as sorrow and burden (Kramer, 1997). The previous presence of patients with dementia and behaviour problems quite
quality of the relationship with the caregiver is also an important difficult to bear.
influence on the experience of caregiving (Quinn et al., 2009). Although joint assessment wards with geriatric medicine were
Most services provided for patients will also directly or indirectly at one time recommended, they are now uncommon. Various fac-
benefit carers, the simplest example being that treating the patient’s tors have contributed to this, but it appears that geriatric physicians
symptoms will make life more tolerable for the carer. Many of the are more comfortable with a responsive psychiatric liaison service,
services for people with dementia, e.g. day care and the manage- and geriatric psychiatrists are satisfied if there is adequate access to
ment of difficult behaviour, are often aimed primarily at improv- medical and diagnostic facilities for their physically unwell patients.
ing matters for the carer. Respite care is provided to give carers a It can, however, leave something of a lottery as to whether a person
much-needed break. Several systematic reviews have shown that with dementia lands up in a medical or a psychiatric bed. Certainly
psychosocial interventions for carers of people with dementia can there is evidence that the mental health of older people with depres-
be effective across a range of outcomes (e.g. Brodaty et al., 2003; sion or cognitive impairment does not benefit from admissions to
Selwood et al., 2007). Perhaps sadly, befriending interventions for general medical wards (Cole, 1993), and patients with depression
carers do not seem to be effective, due mainly to low take-up rates fare better in psychiatric rather than general wards (Norquist et al.,
(Charlesworth et al., 2008). 1995).
As well as this, there is considerable interest in supporting car- Inpatient care for dementia has changed more radically in recent
ers (especially carers of people with dementia) through educational years than that for functional disorders. The most obvious change
and other means. At the most basic level, this includes carer sup- is the large decrease in long-stay hospital beds for dementia. The
port groups, where carers can provide mutual support by sharing amount of respite care provided in hospital has also decreased.
experiences and passing on skills they have learned from their own A second change is that early psychogeriatric services used to
designate beds for dementia assessment, the idea being to generate be funded from the mental health or the general hospital budget.
a comprehensive care plan that could be followed through in the This is amplified by the separation of general medical and mental
community. However, it has become clear that such assessment is health services into different organizations (Holmes et al., 2003).
more appropriately carried out in the community, and acute demen-
tia beds are now used for particularly difficult behaviour problems. Relationships to other agencies
A third change is that there was a vogue for ‘challenging behaviour There are several key relationships to consider, several of which
units’, which proposed to offer intensive behavioural treatments for are discussed in more detail in other chapters of this book. They
the most agitated and aggressive individuals. However, as behav- include working with primary care (Chapter 23), other branches
iour disturbance is by far the most common reason for psychiatric of medicine (Chapter 25), social services (Chapter 26), nongovern-
admission in dementia, it seems pointless to designate some units ment agencies, and residential and nursing homes (Chapters 21 and
in this way. In practice too, the evidence for challenging behaviour 27). End of life care, especially in relation to people with dementia,
units is scanty, and it seems that the best approach is to adhere to has become an increasingly important area of research and practice
best practice rather than establishing special units (Lai et al., 2009). in recent years. Palliative care and end of life issues involve all of
More pertinent is the challenge posed by men with sexually dis- these different agencies (see Chapter 28 and also Chapter 57 for end
inhibited behaviour, for whom single-sex facilities seem to be the of life issues in dementia).
most appropriate approach.
Published evidence certainly suggests that good links between The Changing Nature of Modern Services
inpatient units and their respective community teams are beneficial
for patients, e.g. as occurs when the same consultant is responsible Age-inclusive services
for providing care in both settings. Community follow-up, includ- There is justifiable concern that people should not be discriminated
ing outpatient attendance and community psychiatric nurse visits, against on the basis of their age. Indeed, age discrimination has
reduces readmission for patients with depression (Philpot et al., now become illegal in the UK through the Equality Act 2010 (see
2000). Longer lengths of stay in hospital are also associated with also Carruthers and Ormondroyd, 2009). Part of the reason for this
fewer readmissions, but the relationship between length of stay and legislation was evidence that older people were denied access to
other outcomes is less clear (Draper, 2000). Uncontrolled studies of certain health services solely on the grounds of their age (Centre for
inpatient treatment suggest quite good treatment outcomes, both Policy on Ageing, 2009). For example, in the field of mental health
for depression and for behavioural and psychological symptoms in this often included access to psychological treatments. How age
dementia (BPSD) (Draper and Low, 2005b). equality should be translated into mental health services remains
Long-term psychogeriatric beds were provided in much greater a matter of debate. The two opposing viewpoints are, on the one
numbers in the past, with considerable criticism of the care pro- hand, that age inclusiveness means that a separate specialist service
vided in long-stay wards in old hospitals. In recent years, stud- for older people is in itself discriminatory, and, on the other, that
ies have examined outcomes in newer, purpose-built facilities or the needs of older people are distinct and therefore age equality
among patients discharged from hospital to community settings. does not mean scrapping the older people’s mental health services
Generally, patients discharged to community settings do better as a separate entity.
than those who remain in hospital, with greater staff satisfaction The author’s view is that the second viewpoint is correct. In some
(Trieman et al., 1999), but it remains unclear if this applies to those parts of the country, services for older people have indeed been
with most severe behaviour disorders (Draper and Low, 2005b). merged with those for younger adults. There may or may not remain
a separate service for the assessment and management of demen-
Consultation-liaison psychiatry tia. This has been as much a cost saving measure as for the clinical
There is ample evidence to demonstrate that older people in gen- benefit of patients. However, in such general services, it is always
eral hospitals have high levels of psychiatric morbidity, especially likely that the more immediate physical risks posed by younger
dementia and depression. For most old age psychiatry services, patients, especially those with psychosis, substance misuse prob-
patients in general hospital beds form a high proportion of the total lems, and personality disorders (often in combination), will take up
number of cases referred, about one-third being a typical figure. much of the attention of the team to the detriment of older people.
This important topic is discussed in Chapter 25. From the point Mixed age, ‘age inclusive’, inpatient units are often characterized by
of view of this section, it is worth noting how various models of depressed older people cowering in corners and being generally
consultation and liaison have developed. Consultant psychiatrists neglected. Even within community teams, it is difficult to maintain
have an important role in diagnosis and advice about treatment in the specific expertise required for supporting older people.
medically complex situations. Mental health liaison nurses appear Of course, in present times, most people in their late 60s do not
to be effective in assessing and managing many cases, as well as regard themselves as old, especially if they have good physical health
forming close links with their general nursing colleagues, but it is and either employment or a viable pension. There may be a case for
clear that more evidence is required as to how these services should a more flexible boundary between younger adult services and older
be most effectively provided (Baldwin et al., 2004; Cullum et al., people’s services, based on need rather than exact chronological
2007). Commitment from geriatric medicine is also required to age. However, beyond this stage in life, cognitive impairment and/
develop a successful service, and the joint document Who Cares or physical ill health become much more important features, and
Wins (Royal College of Physicians and Surgeons of Canada, 2005) so the approach to assessment and diagnosis is very different for
sets out how this relationship should work. Service development in someone in their 80s or 90s to that for a 30-year-old. There are spe-
the UK is currently somewhat hampered by uncertainty about the cial skills in either case and it is important that these are maintained
commissioning arrangements—i.e. whether liaison services should and not lost in the pursuit of political correctness.
An argument is sometimes presented that services for ‘functional’ have continued to work by serving a catchment area in the com-
disorders can be provided separately from dementia, and therefore munity but also keeping responsibility for their patients when they
that the functional part of the service can safely be amalgamated are admitted to hospital. Anecdotally at least, this seems to be a
with the general adult service. However, this overlooks one of the more satisfying role, with advantages in terms of bed management
central diagnostic challenges for old age psychiatry, which is the and discharge planning. It does mean that several consultants may
assessment of states (such as mild cognitive impairment, late onset admit to the same ward, so it is necessary to think about efficiency,
depression, and depression with a physical illness) where it is not e.g. sharing ward rounds.
clear whether the problem is mainly organic and, if so, to what Among the newer types of functional teams, there is little evi-
extent. Many of these cases cannot be easily assigned to ‘functional’ dence to date to suggest that they are especially effective for older
or ‘organic’ categories. people, for example in preventing hospital admissions (Jacobs and
A recent report (National Development Team for Inclusion, Barrenho, 2011) or in achieving better outcomes. Toot et al. (2011)
2011) identified several major challenges to achieving equality in reviewed the literature for crisis resolution and home treatment
mental health services. These included such issues as a lack of data teams for older people and found modest evidence of reduced hos-
on outcomes for older people; poor experience of services, nota- pital admissions, but no significant effects on lengths of hospital
bly in general hospitals; inequity of access to treatments and other stay or subsequent duration of residence in the community. It is
services; and low expectations and pessimistic assumptions about noteworthy that, despite later developments being mainly with
outcomes for older people. To bring about improvements requires younger adult patients, the first crisis resolution team reported in
leadership and a clear vision as to what truly constitutes age equal- the UK was an old age psychiatry team (Ratna, 1982). However,
ity, making more use of the voices of service users and carers, and none of the studies reviewed was a true RCT and the nature of the
demonstrating the use of evidence-based practice. comparison groups may explain many of the more positive findings.
In summary, it matters not so much how the services are organ- A Dutch trial is underway to examine the effectiveness of assertive
ized as that there is a clear philosophy and a genuine commitment community treatment for older people with severe mental illness
to equality of access to appropriate support irrespective of age. (Stobbe et al., 2010). Certainly, however, services can be success-
However, this does require recognition that older people’s issues fully redeveloped, especially if attention is paid to the needs of the
are different in certain ways and therefore specialist expertise is staff implementing the programme (McCrae and Banerjee, 2011).
relevant and necessary. There is a real risk that in ‘age-inclusive
services’, older people may be neglected because of the pressures Primary care mental health services
generated by younger patients, and the leadership of such teams It has been long recognized that most consultations related to men-
may not have the necessary expertise and understanding of older tal health take place in primary care and most patients are not, and
people’s issues. do not require to be, referred to specialist mental health services.
In recent years, there has been much more interest in the actual
New Ways of Working (NWW) services that might be offered to people attending primary care,
This term was used by the Royal College of Psychiatrists and the apart from seeing a GP or the practice counsellor. A primary care
Department of Health for a review of the working patterns of psy- setting may have advantages over being seen by a specialist men-
chiatrists (Royal College of Psychiatrists and National Institute of tal health team, e.g. familiarity, convenience, a broader focus on
Mental Health in England, 2005). It arose because of increasing physical health matters, availability of blood tests, and prescribing.
problems with morale and recruitment in general adult psychiatry, This does place certain obligations upon primary care to become
where psychiatrists were struggling with the increasing burdens of more psychologically minded and to play a larger part in the care
roles that encompassed both inpatient and community responsi- of patients with depression and long-term mental illnesses (Royal
bilities. It built on the changes set out in the NSF for Mental Health College of General Practitioners, 2005, 2011).
(DH, 1999), which introduced new types of clinical teams across Another big stimulus for primary care mental heath has been the
England—for crisis resolution and home treatment; early inter- recognition of the huge impact of depressive and anxiety symptoms
vention in psychosis; and assertive outreach (for patients who are on the economy. In response to a report by Lord Layard, a lead-
hard to engage with conventional services). The initiative led to ing economist, the programme Improving Access to Psychological
changes in working patterns for many general adult psychiatrists in Treatments (IAPT) (<http://www.iapt.nhs.uk/>) was established in
England, promoting more delegation to other professional groups 2006. Although this was initially aimed at working-age adults, it
within multidisciplinary teams for assessment and management of has been extended to older adults since 2010, though whether older
cases, and also placing consultants either with inpatient units or people are yet fairly represented in referrals to the service is unclear.
else working with specific teams in the community. The advan- IAPT provides cognitive behaviour therapy at two levels of inten-
tages of the changes are that psychiatrists have a more manageable sity. There is a standard approach, and routine outcome measure-
job, and as a result recruitment into general adult psychiatry has ment forms part of the package. Although IAPT is not technically
improved. The downside is that the creation of numerous teams has a primary care mental health service, as referral is usually required,
led to patients passing between teams, with little continuity or sense it sits closer to primary care than traditional mental health services,
of overall direction. and clinics are often held in primary care or close at hand.
NWW was taken up to variable degrees in old age psychiatry. In Other workers from mental health teams may be based in pri-
some places, especially those where the specialist old age psychiatry mary care, such as so-called gateway workers, who will advise about
service has been merged into a general adult service, it is logical for whether specialist referral should be made or else see some patients
the old age psychiatrists to work in similar fashion. However, in themselves. There may also be community psychiatric nurses based
other areas, a greater weight is placed on continuity and consultants within GP surgeries, but generally they have remained part of a
CMHT in order to prevent professional isolation. In addition, there and local ICPs proliferate. The National Commissioning Pack for
are now developments to provide more mental health workers in Dementia (DH, 2011) is, however, based on an ICP, with six stages
primary care. For older people, this will often comprise a mixture of from the initial awareness of memory problems to the end of life.
qualified staff and support workers, dealing, for example, with mild From this, four key areas are picked out, three of which relate to
to moderate cases of depression and offering support to people with stages along the pathway (early diagnosis and interventions; at
dementia and their carers. home and in care homes; in hospital) and the other (reducing
The importance of primary mental healthcare is reflected in the inappropriate prescribing of antipsychotics) can apply at any stage
establishment of a joint forum for Mental Health in Primary Care of the journey.
by the Royal Colleges of General Practitioners and of Psychiatrists The importance of ICPs is likely to increase, at least in England,
(<http://www.rcgp.org.uk/mental_health.aspx>). The forum works as they will form the basis of payments to providers once a tariff
in partnership with other bodies, including the Department of system (Payment by Results) is introduced for mental health in the
Health and the Royal College of Nursing, and it has produced sev- near future.
eral factsheets, of which the most relevant deals with the manage- Work on integrated pathways leads naturally to considering who
ment of depression in older people (see also Chew-Graham et al., should provide different parts of the pathway or indeed what the
2008). boundaries of the pathway might be. So, for example, consider-
Several important studies from the US have demonstrated ing the support and information needs of people with dementia,
the advantages of collaborative care between specialist serv- it is evident that these will not all be met from the same source.
ices and primary care for older people, both those with depres- Information about the diagnosis and about drug treatment may
sion (Unützer et al., 2002; Von Korff, et al., 2011) and those with come from the specialist mental health service, but after the ini-
dementia (Callahan et al., 2006, 2011). What is not yet evaluated tial flurry of clinical activity will come a longer period when other
is whether such collaborative arrangements will confer simi- forms of help will be required. These may include advice about legal
lar benefits in other health systems with more of a tradition of and financial matters or arranging practical care and support. The
community-oriented old age psychiatry, so trials in other coun- providers of these different aspects are likely to be different organi-
tries are needed. zations—social services, independent care providers, and voluntary
organizations perhaps—but from the patient and family perspec-
The integration agenda tive the service received is regarded as a whole. Integration could be
This section discusses the development of integrated care pathways improved by better information sharing, including access to social
(ICPs) and goes on to consider various ways in which services for care records for CMHTs, and by enabling them to commission
older people may work in a more coordinated way. social care directly (Wilberforce et al., 2011).
An ICP is a multidisciplinary outline of anticipated care, placed Another example is the discharge of older people with mental
in an appropriate timeframe, to help a patient with a specific con- health problems from general hospitals after an episode of physi-
dition or set of symptoms move progressively through a clinical cal illness. Here they will pass from the responsibility of the NHS
experience to positive outcomes (Middleton et al., 2001). Although hospital, usually either back to their own home or into a care
variations from the pathway may be required for individual patients home. This may be straightforward, but there may be many people
with differing needs, the strength of an ICP is that it can help to involved if their needs are complex. There may be a hospital dis-
reduce unnecessary variation in the care offered. It does this by charge planning team and, while in hospital, they may have been
making explicit what is on offer, setting standards, and empowering seen by the older people’s mental health liaison team. There may
patients and carers. It can be used as a means of incorporating local well be a team in the community that provides short-term support
and national guidelines into everyday practice, thereby helping to (reablement) to people for a week or two after they return home.
reduce risk and promote effective clinical governance. ICPs were Alternatively, patients may be referred to a rehabilitation unit for a
originally developed for surgical procedures and more ‘predictable’ few weeks to improve their level of functioning. Once they return
medical conditions, but it is clear that the approach is applicable to home, the CMHT may need to be involved. If they were previ-
mental health problems, such as depression and dementia, that are ously in receipt of personal care, this will need to be reinstated or
prevalent in older people. even increased if needed. If they live with an aged spouse, they too
An ICP can be developed within a single organization, but for may have needs to be assessed and catered for. Should individuals
mental disorders it is more sensible for this work to be shared not return home, but instead need to go into residential care, then
between stakeholders, such as commissioners and providers, with there will be a complex assessment, including of their financial
appropriate public and patient contributions. The output of the affairs, to determine their level of need and who is going to pay
development work will be a chart or set of charts summarizing the for it, all of which suggests that this is fertile ground for an ICP
decisions and actions that need to be taken in given situations. For but also for some simplification of the pathway. Besides, are there
example, for dementia, there might be a single chart encompassing fundamental differences between dementia and other long-term
the whole patient journey, but it is more likely that this will be con- conditions, such as stroke and immobility? Perhaps the distinc-
sidered in separate stages—these could be assessment and diagno- tion between physical and mental health needs and the existence
sis of early dementia; living well with dementia; and end of life care. of separate community teams to deal with them is unhelpful and
The Scottish Dementia Strategy (Scottish Government, 2010) has more could be done to join them up. This could be by adopting
the development of a national dementia ICP as a priority, and the a common ICP and does not necessarily require organizational
work so far has concentrated on three aspects: postdiagnostic sup- change (i.e. merger) to achieve. There is already some evidence of
port; assessment for therapies; and managing challenging behav- community services including dementia care in their provisions
iour. In England, development of ICPs is much less coordinated for long-term conditions.
Support to care homes may be high in certain groups (Livingston et al., 2001), though the
The catchment areas of most old age psychiatry services will con- underlying factors may be complex (Weiner, 2008), rather than due
tain one or more care homes, providing residential and nursing to immigration per se. Comparisons suggest that older people from
care to older people, as well as sometimes younger people with minority ethnic groups tend to present later with dementia, and
dementia. Because of the high levels of dementia and other men- once diagnosed they are less likely to receive antidementia drugs or
tal health problems in care homes (Matthews and Dening, 2002; to participate in research (Cooper et al., 2010). Failure to recognize
Dening and Milne, 2011), there are likely to be a lot of residents who dementia as an illness about which something can be done creates
may be referred for specialist assessment. There are various models significant barriers to obtaining services (Mukadam et al., 2011),
for responding to this demand, ranging from simply providing a but there is evidence to suggest that negative cultural assumptions
consultation service to establishing a specialist team to liaise with can be tactfully challenged (Lawrence et al., 2011). Some recent
care homes. This subject is reviewed by Thompsell (2011); see also examples of good practice are described in a Royal College of
Chapter 27. One problem has been that specialist teams are often Psychiatrists report (Shah et al., 2009).
established with short-term funding and may be vulnerable to cost
Sexuality
saving measures, so it appears that the most effective interventions
are probably educational ones to support leadership and the acqui- Lesbian, gay, bisexual, and transgender (LGBT) older people may
sition of skills in dementia care (Loveday, 2011). have significant difficulties, as social attitudes towards same-sex
relationships have been much less tolerant in the past. Also gay and
lesbian partners are not necessarily recognized as the next of kin by
Equality issues statutory services and this may cause considerable distress. Issues
The UK Equality Act 2010 makes it illegal to discriminate against around discrimination or disclosure remain problematic and affect
a person on the grounds of what are referred to as protected char- the uptake of services by LGBT older people and carers (Heaphy
acteristics. These include age, sex, disability, sexual orientation et al., 2003; Addis et al., 2009; Price, 2010). Few older people’s serv-
including gender reassignment, race, religion or belief, and mar- ices are targeted towards the needs of this group (Warner et al.,
riage or civil partnership. Age has been considered above, but the 2003), and with smaller social networks than their heterosexual
other characteristics are also important, in addition to which cer- peers, LGBT older people may need help to access statutory serv-
tain other groups may also be at risk of unequal access to services. ices, despite their low levels of trust and confidence in these services
(Stonewall, 2011).
Gender
Women live longer than men, the ratio of women to men increasing Rural areas
with advanced age. They are more likely to be bereaved and to live Rural areas present particular challenges to older people’s mental
alone. Depression is more common in women at all ages. Women health services, but they are important as often the proportion of
are more likely to be spouse carers than men. The majority of older older people in rural communities is higher than elsewhere. In
people presenting to health and social care are women, so this area some parts of the world, rural areas are extremely remote or at
may legitimately be regarded as a feminist issue. great distances from centres of population. Studies of rural health
In contrast, older men are in a declining section of the popula- and social care consistently highlight certain themes, such as poor
tion and may have difficulty finding male company and support in access to assessment and diagnostic services, sparse public facili-
such facilities as day centres. They are more likely to be supported ties such as daytime activities and paid caregivers, and difficulties
by a coresident carer, which greatly reduces their risk of moving with isolation compounded by poor and expensive transport. All
into institutional care (Banerjee et al., 2003). Suicide rates are, how- of these apply to people with dementia and other mental health
ever, higher among older men (see Chapter 43). problems, and to their carers. Other obstacles, such as denial of ill-
ness and uncooperative or absent family members, seem to be more
Ethnicity frequent in rural settings (Teel, 2004), although, conversely, satis-
Just about all countries of the world have significant numbers of faction with local doctors and health services seems to be higher
people from different ethnic and cultural backgrounds. Developed (Farmer et al., 2005).
countries often have large numbers of people who have immigrated A few studies have examined the circumstances of rural older
from elsewhere, for various reasons including looking for work and people, mainly with dementia, in relation to the care they receive.
to escape persecution. In general, minority ethnic groups have a Innes et al. (2005) found that certain aspects of rural life in Scotland
younger age profile than the majority population, but, even so, sig- balanced to some extent the gaps in services, and a Canadian study
nificant numbers of older people may be involved. In England and (Bedard et al., 2004) found that rural caregivers did not experience
Wales, there were estimated to be 675,000 older people from ethnic higher levels of burden even though rural patients had a higher
minority groups in 2007, a figure predicted to rise to 1.3 million by level of behaviour disturbance. In North Wales, carers received
2026 and 3.8 million by 2051 (Lievesley, 2010: 59). most essential services even though levels of services were low, but
Minority ethnic groups vary greatly in their access to material the main problems were a lack of crisis support and a reluctance to
and social resources. For example, Indian older people are less likely accept long-term care when it was needed (Wenger et al., 2002).
to experience multiple deprivations, with similar levels to white Rural healthcare is perhaps more likely to require novel
older people. Levels of deprivation affect the wellbeing of older approaches to providing services (Chalifoux et al., 1996). Some
people. Some illnesses are more common among specific groups, work has used family and care staff focus groups to identify ways of
e.g. hypertension and stroke among African Caribbean people and providing accessible services (Morgan et al., 2002). Small, flexible
diabetes among South Asians. Rates of dementia and depression day centres can be valued and effective, though there are attendant
problems about storing and transporting records and materials, death, but the age-specific incidence and mortality rates of most
and about effective supervision and management of staff (Gibson cancers decline beyond age 90. With regard to mental health, cen-
et al., 1995). Training programmes for caregivers that have been tenarians scored lower on cognitive tests and were more likely to
designed to be transportable may be useful (Hepburn et al., 2003) feel useless and that life was routine, not exciting (Martin et al.,
and advances in telecommunications have obvious potential to 1992), and more likely to report a lack of social resources compared
reach people in remote places (Sumner, 2001), though some issues to people in their 80s (Randall et al., 2010). Variable estimates of
such as funding of equipment, confidentiality, and ethical conduct the prevalence of dementia in centenarians have been reported.
of teleconsultations are not fully worked out (Goins et al., 2001). Although prevalence studies of dementia suggest that there may
There also seems to be a tension between technological solutions be a survivor effect at extreme old age, incidence studies show an
and what older people actually value, which is more to do with increasing incidence of dementia in the population up to at least
personal contact and effective services (King and Farmer, 2009). age 90, making this unlikely (Matthews et al., 2005). Cohort studies
Nonetheless, there is still a need for new ideas, perhaps making of centenarians also suggest that decline continues to develop with
more use of generic services or everyday environments such as greater age (Miller et al., 2010).
pubs, supermarkets, and hairdressers, to provide flexible and effec- Within the everyday practice of old age psychiatry, there is a
tive support to patients and carers. noticeable increase in the age of many of the patients referred, with
consequent increases in medical comorbidity, physical disability,
Prisons and complex social care needs. This underscores the importance of
There are significant numbers of older people among the inmates providing mental health services for very old people that are com-
of prisons and special hospitals, almost 7000 in England and Wales prehensive, interdisciplinary, and allied to provision of physical
in 2008 (Prison Reform Trust, 2009). Of these, just over 450 were healthcare, and for the policy and research agendas to address the
aged over 70 and a few were over 80. The numbers of both older needs of this group (Dening and Gabe, 2000).
men and women in prison has trebled since 1996. Over half of older
prisoners have some form of mental disorder, usually depression Elder abuse and safeguarding
(Kingston et al., 2011). Identification and treatment of mental dis- The mistreatment of older people is considered in more detail in
orders in prisoners can be readily overlooked in older prisoners, Chapter 59. From the perspective of old age psychiatry services,
as they are less disruptive. Many prisoners are in unsuitable condi- there are perhaps two main issues. One of these is a concern with
tions with only limited facilities to support them after release. The the frequency of abuse of older patients that can occur in institu-
same applies to special hospital patients who may remain detained tional settings including hospitals. There is a sad catalogue of hospi-
longer than necessary because there are no alternatives (Yorston, tal inquiries, which highlight similar themes, such as geographical
1999; Fazel et al., 2001). Aspects of crime in relation to old age psy- isolation, low staffing levels, lack of training, lack of nursing lead-
chiatry are discussed further in Chapter 60. ership, and lack of clinical governance. In England, regulation of
health and social care services has been brought together under the
Learning disabilities Care Quality Commission (CQC), which monitors performance
People with learning disabilities have higher rates of early-onset against a set of essential care standards. At the time of writing, the
dementia than the general population. They may also be excluded CQC is seen as a tough regulator by many health and social provid-
from consideration by mainstream health and social facilities. ers, yet it has faced criticism at times for overlooking some institu-
Current UK policy (DH, 2001b) emphasizes the need for people tions where there was clear evidence of abuse.
with learning disabilities to access normal services wherever pos- The other issue for old age psychiatry services is to participate
sible. Building on this, the Foundation for People with Learning as part of the multiagency, interdisciplinary response to cases that
Disabilities (2002) developed an extensive programme of work, arise in all settings. This may include various roles, such as psychi-
Growing Older with Learning Disabilities (GOLD), which ranges atric assessment of the victim and/or the perpetrator, assessments
from issues of healthy ageing to the services required for dementia of mental capacity especially when financial abuse is suspected, and
and palliative care for people with learning disabilities. participating in vulnerable people’s procedures. In England, policy
In general, however, specialist services for people with learn- has been driven by the government’s No Secrets document (DH,
ing disabilities cover the whole age range, and people who develop 2000), but the current position is more recently set out in a frame-
dementia continue to be served by facilities with expertise in learn- work for safeguarding (Association of Directors of Social Services,
ing disabilities. In future, there will need to be more closely defined 2005). Thus far, physical abuse and neglect have been highlighted;
working arrangements between primary care, learning disabilities it is likely that the scale of financial abuse against older people will
services, and older people’s mental health services. receive more recognition and public concern over the next few
years.
Very old people
Definitions of the ‘oldest old’ or ‘very old people’ are variable and
the thresholds have increased over time with growing numbers of Quality in Old Age Psychiatry Services
people aged 90 plus. Several cohort studies of very old people and There is no single definition of what constitutes quality, so it is usu-
centenarians have been conducted (e.g. Baltes and Mayer, 1999). ally interpreted pragmatically. In business discourse, the viewpoint
Only 25–30% of centenarians live independently; most are in resi- of the customer is clearly the most important in assessing the qual-
dential or nursing home care. ity of the product or service provided. There are differing views on
With great age, the epidemiology of common illnesses changes. whether that is the sole aspect of quality, or whether other perspec-
Circulatory diseases become an increasingly common cause of tives may be considered. For example, from the point of view of a
manufacturer, this could include how well they procure their raw specific aspects of treatment, e.g. the use of antidementia drugs
materials, how smoothly their production lines run, and keeping or psychological treatments.
the proportion of defective products to a minimum. ◆ Care management. As mental health problems in old age often
In healthcare, the thorny question of defining quality is usually require input from a range of professionals and from different
avoided. Most approaches to quality emphasize that healthcare agencies, ways of recording, storing, and sharing information are
should be safe, effective, patient-centred, efficient, equitable, and extremely important in ensuring a quality service. In the UK, this
timely (Committee on Quality Health Care in America, 2001). In is provided through the CPA, which uses standard documenta-
the NHS in England, the emphasis has been mainly on effective- tion and provides a comprehensive record that can be regularly
ness, safety, and patient experience. The focus on clinical quality updated and shared with patients and carers. Integrating differ-
sharpened after the report High Quality Care for All (Darzi, 2008), ent systems of care management has been the biggest challenge
which insisted that quality should be placed at the heart of the NHS to successful joint working between health and social services.
and provided a strong case for clinical leadership in delivering the There is obvious potential to use electronic versions of CPA so
quality agenda. This led to a requirement for all NHS Trusts to pro- that the same patient data can be readily available to different
duce annual quality reports alongside their financial accounts, and professionals at different sites.
to the establishment of a National Quality Board, chaired by the
NHS Medical Director. ◆ Outcomes. In theory at least, measuring outcomes should be
From the point of view of mental health services for older people, the gold standard for quality, as it has more direct relevance to
we might consider first the ways in which providers can measure patients than do the proxy alternatives of structure and process
and enhance the quality of the services that they provide, and then that are usually measured instead. However, outcome measure-
discuss the wider context, e.g. the framework of regulation within ment has long been a problem for mental health and this is espe-
which services operate. cially so in old age psychiatry, where often ‘bad’ outcomes such as
Quality relates to the care and treatment provided to an individ- death or institutionalization may actually follow episodes of very
ual and to the overall provision of services to a population. Much high quality care. Routine measurement of outcomes, especially
work in improving quality is about reducing variation in clinical using the HoNOS 65+ (Burns et al., 1999), has become more
practice so that patients can expect a consistent approach across the widespread, particularly as clinical clusters derived from HoNOS
whole service. This involves consideration of the whole care path- 65+ will be used as the basis for a tariff-based funding system in
way or ‘the patient’s journey’. Examples of this include: mental health.
◆ Referral criteria. Ensuring that the service sees those patients ◆ Evaluation of services. Research plays a part in this, but health
who are in most need within a reasonable time scale requires services research may be complex and costly. RCTs are relatively
collaboration with referrers, especially primary care. Agreeing rare and it is impossible for an individual service to investigate
thresholds not only for referral but also for discharge from a spe- most aspects of service delivery in this way. Instead, a range of
cialist service is an important way of controlling the caseloads of approaches are used, of which clinical audit and surveys are
team members. There may need to be agreed arrangements for probably the most important. Clinical audit is a tool to measure
the transfer of patients who have been with services for younger the way in which things are being done assessed against agreed
adults. standards, leading to a cycle of making the necessary changes and
then re-auditing. The selection of topics for audit may be either
◆ Standardized assessments. The Single Assessment Process, ‘top-down’ (imposed) or ‘bottom-up’ (chosen) and this may cre-
introduced by the NSF OP, underlines the importance of mak- ate tensions. There may be limited administrative resources to
ing comprehensive and standard assessments. Local authorities support the work and the evidence that audit leads to better out-
are obliged to have a single assessment process and this can be comes for patients is sometimes quite thin.
used jointly with health services (examples are available on local
authority websites). In addition, several versions of more special- Effectiveness, safety, and patient experience
ized, comprehensive assessment are available, such as the CANE Clinical effectiveness is concerned with the extent to which a serv-
(Orrell and Hancock, 2004), but there are also many standard ice follows best practice. In England, the most important source of
rating scales available for more specific purposes, such as assess- guidance is the National Institute for Health and Care Excellence
ment of cognition and of mood (compiled by Burns et al., 2003). (NICE), which produces clinical guidelines on important conditions
◆ Guidelines and protocols for treatment. Clinical guidelines are sys- such as dementia and depression, as well as technical appraisals of
tematically developed statements to assist clinicians and patients new treatments (drugs, therapies, operative procedures, etc.). The
in making decisions about appropriate treatment for specific most important technical appraisal for old age psychiatry has been
conditions. Protocols are written plans specifying the procedures that concerning drugs for treating dementia (for the latest version
to be followed in providing care for a particular condition. ICPs see NICE, 2011). Other aspects of ensuring best practice include
are locally agreed summaries of practice, based on the best avail- clinical audit and medicines management. Audits may be local,
able evidence, for a specific condition or patient group. A care but there are also national audits, some of which are mandatory,
pathway may therefore be the application of a guideline in a prac- e.g. the National Confidential Inquiry into Suicide and Homicide
tical situation. Clinical guidelines so far available to inform old by People with a Mental Illness (Appleby et al., 2011). The Royal
age psychiatry tend to concentrate on dementia rather than other College of Psychiatrists has established a College Centre for Quality
mental disorders (Burns et al., 2002) and they vary as to how sys- Improvement (CCQI), which currently has nine project areas
tematically derived they are. Guidelines and protocols may cover across mental health, of which the most relevant to old age psychia-
the whole management of a condition or they may emphasize try are the Memory Services National Accreditation Programme
(MSNAP), the ECT Accreditation Service (ECTAS), and the influenced GP behaviour and encourages GPs to seek a diagnosis
Accreditation for Inpatient Mental Health Services (AIMS). Each for people with possible dementia.
of these provides inspection visits, feedback, and accreditation for Finally, note that quality in the sense used here relates to qual-
participating services. The CCQI also runs national audits, includ- ity of care, but there is also an important strand of work regarding
ing one on dementia in general hospitals. Another, the Prescribing quality of life that is beyond the scope of this account. For an up-to-
Observatory for Mental Health (POMH-UK), benchmarks predate discussion of the importance of and how to measure quality of
scribing practice across member Trusts in order to improve stand- life in dementia, see Banerjee et al. (2009).
ards and consistency. POMH-UK has recently developed an audit
of the use of antipsychotic drugs in dementia. Future challenges
There are numerous indicators relevant to patient safety, includ-
ing suicide data, incident rates and incident reporting, and informa- Population change
tion about falls. Workforce information, such as staff satisfaction, It is generally accepted that the biggest challenge to services for old
sickness rates, and the proportion of temporary staff working in a people across the world is posed by demographic change and the
team, is also highly relevant to patient safety. In recent years there ageing population. This takes different forms in different countries.
has been more attention paid to safeguarding vulnerable adults, At present, the developed countries of Europe, Japan, and the US
which is of obvious relevance too. have the highest percentages of people aged 65 and over in their
Patient experience is best directly measured by obtaining feed- populations. Over the next 20–30 years, numbers of older people
back from people using the service. This may include questionnaire will continue to increase, but the increase will be especially in very
surveys, but also information from complaints and patient enquir- old people, those aged 85 and above.
ies gives an important indication of the issues that concern patients In contrast, middle- to low-income countries have lower per-
and families. In old age mental health, carers often act as proxies for centages of old people, but in general their populations are ageing
patients, especially for people with dementia, but it is important to at a faster rate. That is, numbers of people aged over 65 will increase
recognize that the interests of patients and carers do not necessar- by a greater proportion than in Europe and the US. Moreover, as
ily coincide (Dening and Lawton, 1998). Other objective measures, certain countries (China, India, and Indonesia, for example) have
such as environmental inspections, length of stay, and delayed dis- very large total populations, the absolute numbers of older people
charges (for inpatients), and clinical outcomes are also important. in these countries will grow very rapidly.
Organizations, such as NHS Trusts, that run mental health serv- It is estimated that there are currently around 36 million people
ices will have in place a structure for healthcare governance to report in the world with dementia (Prince et al., 2011), a figure predicted
at board level on issues related to quality. There will be board level to double every 20 years to 115 million by 2050. Already, almost
responsibility for quality and governance and this will also include 60% of people with dementia live in low- and middle-income coun-
responsibility for producing the Trust’s annual quality account. tries and this proportion will grow to 70% by 2050. The worldwide
costs of dementia (US $604 billion in 2010) amount to more than
Regulation 1% of global GDP, leading to the often quoted observation that if
The regulatory framework for NHS Trusts in England is pro- dementia care were a country, it would be the world’s 18th largest
vided by two public bodies, Monitor and the CQC. The former is economy (Prince et al., 2011) or, if it were a company, it would be
the regulator of the business activities of NHS Foundation Trusts, by far the world’s largest. Most people with dementia, an estimated
though Monitor does also have a quality framework for Trusts to 28 million, have not received a diagnosis.
report against. However, the CQC has the main role in relation to An ageing population will be associated with not only more cases
quality. It is the regulator not only of NHS services but also of the of dementia, but also increased levels of other forms of illness and
whole of adult social care (community and residential services), as disability, and comorbidity of physical and mental disorders. At the
well as having a remit for compliance with the Mental Health Act. same time, an increasing proportion of older people will affect the
CQC has a framework of standards based on the legislative regula- national economy, especially if the working-age population is not
tions that established it. In practice, there are 16 key standards that expanding. This has both economic and manpower effects on the
apply to health and social care, covering a full range of issues, from resources that are available for health and social services available
treating people with respect to environmental and staffing issues. to older people. There will of course be changes in retirement ages
Organizations are inspected against these standards and required to and pensions, which may serve to keep people working for longer.
amend any failures or concerns that are raised. CQC reports are pub- In the light of these profound changes, every country should be
lic documents, as too are the quality reports produced by Trusts. developing a dementia strategy and putting in place systems that
Another important part of the regulatory framework is the will improve access to diagnosis, information, and help. Even in
Quality and Outcomes Framework that operates in general practice the poorest countries, it should be possible to introduce a coherent
and forms part of the basis for payments to GPs and practices. This strategy using simple interventions based in primary care.
sets out a total of 86 indicators across 20 major clinical areas that
GPs are expected to meet to obtain the requisite payments. Two of Scientific advances
these are relevant to old age psychiatry. There are three indicators It is of course impossible to predict what scientific discoveries
for dementia, around maintaining a dementia register, perform- may influence the future course of mental disorders in old age.
ing appropriate investigations for possible dementia, and keeping At present, no single development seems likely to have a startling
patients with dementia under review. The indicators for depres- impact upon conditions such as Alzheimer’s disease. Better gen-
sion apply to all ages but are concerned with diagnosis, assessment eral health may affect dementia and depression too, though this is
of severity, and review. The introduction of this framework has difficult to demonstrate. The future availability of presymptomatic
diagnosis of dementia would have an impact upon the demand for in old age psychiatry (Gustafson et al., 2003). The curriculum
services. Clinicians will become adept at using currently available comprises 22 areas of competence and learning objectives. These
treatments more efficiently. For example, perhaps Alzheimer’s dis- include clinical skills, such as recognition of mental health prob-
ease may respond better to a combination of several treatments, lems, history taking, and physical examination; relevant areas of
rather than a single drug such as a cholinesterase inhibitor. It will, knowledge, including ageing and the mental disorders of old age;
however, be difficult to do RCTs to investigate combined treatments treatment, management, and care; and the organization of services.
as large numbers of patients will be required. In the treatment of There are some rather specific domains, including end of life issues
depression and psychotic illnesses, improved use of newer antide- and the abuse of older people. The last few areas of competence in
pressant, mood stabilizing, and antipsychotic drugs, together with the list are also general topics, such as prevention, teaching, knowl-
use of combined treatments, may bring better outcomes with fewer edge management, and research. The curriculum is likely to help
side effects for many patients. the development of old age psychiatry programmes in countries
It is encouraging to see that dementia is increasingly identified where this is under consideration, though the areas of competence
as a priority for research, though the sums of money spent on it are not operationally defined and the curriculum does not provide
remain trivial besides the amounts spent on cancer and heart dis- much guidance as to how they should be assessed.
ease. Similarly, there is not yet much of a culture for recruiting Details and requirements for training vary between countries. For
people with dementia into research such as treatment trials, again example, the entry point into specialization in old age psychiatry in
a marked contrast to the situation with cancer. The way forward the UK is after 3 years of core psychiatry training, and consists of
involves establishing registers of patients with dementia and ascer- a further 3 years, of which at least two must be in old age psychia-
taining their willingness to participate in research at the outset, try. In the US, doctors acquire the certificate in psychiatry of the
together with the further development of clinical and research American Board of Psychiatry and Neurology and then undertake
networks. an additional year of geriatric psychiatry before sitting the exam
In the immediate future, however, it is most likely that older peo- for the additional certificate. In general, however, there has been a
ple will be most helped by the more effective provision of services. trend towards old age psychiatry/geriatric psychiatry both becom-
Simple routes of referrals, prompt and efficient assessments, access ing more central to the core curriculum of psychiatric training and
to specialized treatment when needed, care and support at home acquiring its own identity as a speciality. A good example is pro-
wherever possible, all backed up by effective care coordination, vided by Canada, where geriatric psychiatry has recently become
are all elements of good services that are supported by evidence. required as part of core training for psychiatry, and in 2009 geriat-
In keeping with trends across the whole of medicine (and, indeed, ric psychiatry became a recognized subspecialty within the Royal
society in general), it can be predicted that the voice of patients and College of Physicians and Surgeons of Canada. At the same time,
carers as consumers will become more powerful and this too will the Canadian Academy of Geriatric Psychiatry has won representa-
influence the type of services available and how they are provided. tion on the board of the Canadian Psychiatric Association.
The quality of services can also be improved by the application of Probably the most detailed curriculum is that developed by the
information and communications technology, in various ways, such UK Royal College of Psychiatrists (2010), which covers the com-
as improved access to information for users and carers through the petencies required and the relevant assessments for both core and
internet and the application of (often simple) devices in the indi- specialist training within psychiatry and old age psychiatry, respec-
vidual’s home to provide prompts and reminders or to activate tively. It draws on the Canadian model, CanMEDS (Royal College
alarms. Technology is likely to be used increasingly to provide con- of Physicians and Surgeons of Canada, 2005), which describes the
tact with services across long distances, e.g. in rural areas. Finally, role of physicians in seven related domains: medical expert, com-
the potential of electronic care records is yet to be fully realized municator, collaborator, manager, health advocate, scholar, and
as systems are only now being established, but it is probable that professional.
individuals will take a much more active role in their healthcare There are perhaps three big challenges in this area. One is that
by taking responsibility for the record. This could transform the the curricula usually just relate to old age psychiatry and not other
current relationship between consumers and providers of services, mental health professions, for whom training arrangements and
although it may also raise important issues regarding information competencies are far less developed. A second issue is that there
governance and confidentiality. needs to be specific attention on developing clinical researchers in
old age psychiatry so that the clinical aspects of mental disorders in
Specialist training in old age psychiatry old age continue to receive adequate attention alongside basic bio-
One important sign of a robust speciality is the development of logical and epidemiological research (Bartels et al., 2010). The third
specialist training programmes and their recognition by official area is that recruitment continues to be problematic, since profes-
bodies. For old age psychiatry, this has closely matched the devel- sionals in training may find working with older patients less attrac-
opment of older people’s mental health services. Various curricula tive than with other groups. In the US, the Institute of Medicine
and training guidelines have been developed since the late 1970s, (2012) has reported on the mental health workforce that will be
but formal criteria for old age psychiatry training were first adopted required for geriatric populations.
in the UK in 1989, followed by programmes in the US, Canada, and
Australasia (Draper, 2003).
The European Association of Geriatric Psychiatry, in collabora- Conclusion
tion with the World Health Organization and the World Psychiatric Increasing numbers of older people across the world pose a major
Association Section of Psychiatry of the Elderly, published a con- challenge to health and social care services, including those involved
sensus curriculum of skill-based objectives for specialist training in mental health. Old age psychiatry has become a mature speciality
and there is a considerable evidence base for many of the things Bartels, S.J., et al. (2002). Evidence-based practices in geriatric mental health
that it has to offer. However, there is a paradox in that, at the time care. Psychiatric Services, 53, 1419–31.
that it would seem to be most needed, it is also perhaps most under Bartels, S.J., et al. (2003). Evidence-based practices in geriatric mental health
care: an overview of systematic reviews and meta-analyses. Psychiatric
threat. Nonetheless, in whatever way services are configured in the
Clinics of North America, 26, 971–90.
future, there will still be a need for well-trained, skilled practition-
Bartels, S.J., et al. (2010). Programs for developing the pipeline of early-career
ers who are able to assess and manage conditions such as dementia geriatric mental health researchers: outcomes and implications for other
and depression in older people. Anyone who has these competen- fields. Academic Medicine, 85, 26–35.
cies or is in the process of acquiring them should feel confident that Bedard, M., Koivuranta, A., and Stuckey, A. (2004). Health impact on
they have a major part to play. caregivers of providing informal care to a cognitively impaired older
adult: rural versus urban settings. Canadian Journal of Rural Medicine,
References 9, 15–23.
Addis, S., et al. (2009). The health, social care and housing needs of lesbian, Bedford, S., et al. (1996) What becomes of people with dementia referred
gay, bisexual and transgender older people: a review of the literature. to community mental health teams? International Journal of Geriatric
Health and Social Care in the Community, 17, 647–58. Psychiatry, 11, 1051–6.
Alzheimer Europe (2011). National dementia plans. Alzheimer Europe, Blazer, D. (2000). Psychiatry and the oldest old. American Journal of
Luxembourg . <http://www.alzheimer-europe.org/index.php/EN/ Psychiatry, 157, 1915–24.
Policy-in-Practice2/National-Dementia-Plans>. Brodaty, H., Gresham, M., and Luscombe, G. (1997). The Prince Henry
Alzheimer’s Society (2007). Dementia UK: the full report. Alzheimer’s Society, Hospital dementia caregivers’ training programme. International Journal
London. of Geriatric Psychiatry, 12, 183–92.
Appleby, L., et al. (2011). The National Confidential Inquiry into Suicide and Brodaty, H., Green, A., and Korschera, A. (2003). Meta-analysis of
Homicide by People with a Mental Illness: annual report—England, Wales psychosocial interventions for caregivers of people with dementia.
and Scotland. <http://www.medicine.manchester.ac.uk/mentalhealth/ Journal of the American Geriatrics Society, 51, 657–64.
research/suicide/prevention/nci/inquiryannualreports/Annual_ Burns, A., et al. (1999). Health of the Nation Outcome Scales for Older People
Report_July_2011.pdf>. (HoNOS 65+). British Journal of Psychiatry, 174, 424–7.
Argyle, N., Jestice, S., and Brook, C.P. (1985). Psychogeriatric patients: their Burns, A., Dening, T., and Lawlor, B. (2002). Clinical guidelines in old age
supporters’ problems. Age and Ageing, 14, 355–60. psychiatry. Dunitz, London.
Arie, T. (1970). The first year of the Goodmayes psychiatric service for old Burns, A., Lawlor, B., and Craig, S. (2003). Assessment scales in old age
people. Lancet, ii, 1179–82. psychiatry, 2nd edition. Dunitz, London.
Arie, T. (1989). Martin Roth and the ‘psychogeriatricians’. In: K. Davison and Callahan, C.M., et al. (2006). Effectiveness of collaborative care for older
A. Kerr (eds) Contemporary themes in psychiatry: a tribute to Sir Martin adults with Alzheimer disease: a randomized controlled trial. Journal of
Roth, pp. 231–8. Gaskell, London. the American Medical Association, 295, 2148–57.
Arie, T. and Jolley, D. (1982). Making services work: organisation and style Callahan, C.M., et al. (2011). Implementing dementia care models in primary
of psychogeriatric services. In: R. Levy and F. Post (eds) The psychiatry of care settings: the aging brain care medical home. Aging and Mental
late life, pp. 222–251. Blackwell, Oxford. Health, 15, 5–12.
Ashaye O.A., Livingston, G., and Orrell, M. (2003). Does standardised needs Camus, V., et al. (2003). Teaching and training in old age psychiatry: a
assessment improve the outcome of psychiatric day hospital care for general survey of the World Psychiatric Association member societies.
older people? Aging and Mental Health, 7, 195–9. International Journal of Geriatric Psychiatry, 18, 694–9.
Association of Directors of Social Services (2005). Safeguarding adults: a Carruthers, I. and Ormondroyd, J. (2009). Achieving age equality in health
national framework of standards for good practice and outcomes in adult and social care. <http://www.dh.gov.uk/prod_consum_dh/groups/
protection work. <http://www.adass.org.uk/old/publications/guidance/ dh_digitalassets/documents/digitalasset/dh_107398.pdf> (accessed
safeguarding.pdf> (accessed 21.05.2012). 21.05.2012).
Audini, B., et al. (2001). Old age psychiatric day hospital survey. Royal College Centre for Policy on Ageing (2009). Ageism and age discrimination
of Psychiatrists Research Unit, London. in mental health care in the United Kingdom: a review from the
Baldwin, R., et al. (2004). Does a nurse-led mental health liaison service literature. <http://www.cpa.org.uk/information/reviews/CPA-ageism_
for older people reduce psychiatric morbidity in acute general medical and_age_discrimination_in_mental_health_care-report.pdf> (accessed
wards? A randomised controlled trial. Age and Ageing, 33, 472–8. 21.05.2012).
Baltes, P.B. and Mayer, K.U. (eds) (1999). The Berlin Aging Study: aging from Chalifoux, Z., et al. (1996). Mental health services for rural elderly: innovative
70 to 100. Cambridge University Press, Cambridge. service strategies. Community Mental Health Journal, 32, 463–80.
Banerjee, S. (2009). The use of antipsychotic medication for people with Challis, D., et al. (2002). Care management, dementia care and specialist
dementia: time for action. Department of Health, London. mental health services: an evaluation. International Journal of Geriatric
Banerjee, S. and Wittenberg, R. (2009). Clinical and cost effectiveness of
services for early diagnosis and intervention in dementia. International Charlesworth, G., et al. (2008). Befriending carers of people with dementia:
Journal of Geriatric Psychiatry, 24, 748–54. randomised controlled trial. British Medical Journal, 336, 1295.
Banerjee, S., et al. (2003). Predictors of institutionalisation in people with Chew-Graham, C., Baldwin, R., and Burns, A. (eds) (2008). Integrated
dementia. Journal of Neurology Neurosurgery and Psychiatry, 74, management of depression in the elderly. Cambridge University Press,
1315–16. Cambridge.
Banerjee, S., et al. (2007). Improving the quality of care for mild to moderate Cole, M.G. (1993). The impact of geriatric medical services on mental state.
dementia: an evaluation of the Croydon Memory Service Model. International Psychogeriatrics, 5, 91–101.
International Journal of Geriatric Psychiatry, 22, 782–8. Collighan, G., et al. (1993). An evaluation of the multidisciplinary approach
Banerjee, S., et al. (2009). What do we know about quality of life in dementia? to psychiatric diagnosis in elderly people. British Medical Journal, 306,
A review of the emerging evidence on the predictive and explanatory 821–24.
value of disease specific measures of health related quality of life in Committee on Quality Health Care in America/Institute of Medicine (2001).
people with dementia. International Journal of Geriatric Psychiatry, 24, Crossing the quality chasm: a new health system for the 21st century.
15–24. National Academy Press, Washington, DC.
Conroy, S., et al. (2010). A multicentre randomised controlled trial of day Farmer, J., et al. (2005). Urban versus rural populations’ view of health care in
hospital-based falls prevention programme for a screened population of Scotland. Journal of Health Service Research and Policy, 10, 212–19.
community-dwelling older people at high risk of falls. Age and Ageing, Fasey, C. (1994). The day hospital in old age psychiatry: the case against.
39, 704–10. International Journal of Geriatric Psychiatry, 9, 519–23.
Cooper, C.G., et al. (2010). A systematic review and meta-analysis of ethnic Fazel, S., et al. (2001). Hidden psychiatric morbidity in elderly prisoners.
differences in use of dementia treatment, care, and research. American British Journal of Psychiatry, 179, 535–9.
Journal of Geriatric Psychiatry, 18, 193–203. Foundation for People with Learning Disabilities (2002). Today and
Cullum, S., et al. (2007). Effectiveness of liaison psychiatric nursing in older tomorrow: the report of the Growing Older with Learning Disabilities
medical inpatients with depression: a randomised controlled trial. Age (GOLD) Programme. Foundation for People with Learning Disabilities,
and Ageing, 36, 436–42. London.
Darzi, A. (2008). High quality care for all: NHS next stage review final report. Gibson, F. and Whittington, D. (1995). Day care in rural areas. Social care
The Stationery Office, London. research report 7. Joseph Rowntree Foundation, York
Dementia Action Alliance (2010). National Dementia Declaration for Goins, R.T., Kategile, U., and Dudley, K.C. (2001). Telemedicine, rural elderly,
England: a call to action . <http://www.dementiaaction.org.uk/ and policy issues. Journal of Aging and Social Policy, 13, 53–71.
dementiaaction/downloads/file/1/national_dementia_declaration> Graff, M.J., et al. (2006). Community based occupational therapy for patients
(accessed 21.05.2012). with dementia and their care givers: randomised controlled trial. British
Dening, T. and Gabe, R. (2000). Disability and contact with services in very Medical Journal, 333(7580), 1196.
old people. Reviews in Clinical Gerontology, 10, 291–309. Gustafson, L., et al. (2003). Skill-based objectives for specialist training in old
Dening, T. and Lawton, C. (1998). The role of carers in evaluating mental age psychiatry. International Journal of Geriatric Psychiatry, 18, 686–93.
health services for older people. International Journal of Geriatric Head, L. and Dening, T. (1998). Lithium in the over-65s: who is taking it
Psychiatry, 13, 863–70. and who is monitoring it? International Journal of Geriatric Psychiatry,
Dening, T. and Milne A. (eds) (2011). Mental health and care homes. Oxford 13, 164–71.
University Press, Oxford. Heaphy, B., Yip, A., and Thompson, D. (2003). Lesbian, gay and bisexual
Dening, T. and Shaji, K.S. (2005). Psychogeriatric service delivery with lives over 50. York House Publications, Nottingham Trent University,
limited resources. In: B. Draper, P. Melding and H. Brodaty (eds) Nottingham. <http://ess.ntu.ac.uk/heaphy/lgb50+.doc> (accessed
Psychogeriatric service delivery: an international perspective, pp. 327–44. 21.05.2012).
Oxford University Press, Oxford. Hepburn, K.W., et al. (2003). The Savvy Caregiver Program: developing and
Department of Health (1999). National Service Framework for Mental Health. testing a transportable dementia family caregiver training program.
Department of Health, London. Gerontologist, 43, 908–15.
Department of Health (2000). No secrets. Department of Health, London. Hilton, C. (2005a). The origins of old age psychiatry in Britain in the 1940s.
Department of Health (2001a). National Service Framework for Older People. History of Psychiatry, 16, 267–89.
Department of Health, London. Hilton, C. (2005b). The clinical psychiatry of late life in Britain from 1950
Department of Health (2001b). Valuing people: a new strategy for learning to 1970: an overview. International Journal of Geriatric Psychiatry, 20,
disability for the 21st century. Department of Health, London. 423–8.
Department of Health (2009). Living well with dementia: a national dementia HM Government (2011). No health without mental health: a cross-government
strategy. Department of Health, London. mental health outcomes strategy for people of all ages. <http://www.dh.gov.
Department of Health (2010). Quality outcomes for people with dementia: uk/prod_consum_dh/groups/dh_digitalassets/documents/digitalasset/
building on the work of the National Dementia Strategy. <http://www. dh_124058.pdf> (accessed 21.05.2012).
dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/@dh/@en/@ps/ Hoe, J., Ashaye, K., and Orrell, M. (2005). Don’t seize the day hospital! Recent
documents/digitalasset/dh_119828.pdf> (accessed 21.05.2012). research on the effectiveness of day hospitals for older people with
Department of Health (2011). Commissioning services for people with mental health problems. International Journal of Geriatric Psychiatry, 20,
dementia . <http://www.dh.gov.uk/en/Publicationsandstatistics/ 694–8.
Publications/PublicationsPolicyAndGuidance/Browsable/DH_127381> Holmes, J., Bentley, K., and Cameron, I. (2003). A UK survey of psychiatric
(accessed 21.05.2012). services for older people in general hospitals. International Journal of
Department of Health and Social Security (1972). Services for mental illness Geriatric Psychiatry, 18, 716–21.
related to old age. HMSO, London. Innes, A., et al. (2005). Dementia care provision in rural Scotland: service
Department of Health, Social Services and Public Safety (2011). Improving users’ and carers’ experiences. Health and Social Care in the Community,
dementia services in Northern Ireland: a regional strategy. <http://www. 13, 354–65.
dhsspsni.gov.uk/improving-dementia-services-in-northern-ireland-a-re Institute of Medicine (2012). The Mental Health and Substance Misuse Workforce
gional-strategy-november-2011.pdf> (accessed 14.02.2013). for Older Adults: in whose hands? <http://www.iom.edu/Reports/2012/
Draper, B. (2000). The effectiveness of old age psychiatry services. International The-Mental-Health-and-Substance-Use-Workforce-for-Older-Adults.
Journal of Geriatric Psychiatry, 15, 687–703. aspx> (accessed 14.02.2013).
Draper, B. (2003). Training in old age psychiatry. International Journal of Jacobs, R. and Barrenho, E. (2011). Impact of crisis resolution and home
Geriatric Psychiatry, 18, 683–5. treatment teams on psychiatric admissions in England. British Journal
Draper, B. and Low, L.F. (2005a). Evidence-based psychogeriatric service of Psychiatry, 199, 71–6.
delivery. In: B. Draper, P. Melding and H. Brodaty (eds) Psychogeriatric Katona, C., et al. (2009). World Psychiatric Association Section of Old Age
service delivery: an international perspective, pp. 75–123. Oxford Psychiatry consensus statement on ethics and capacity in older people
University Press, Oxford. with mental disorders. International Journal of Geriatric Psychiatry, 24,
Draper, B. and Low, L.F. (2005b). What is the effectiveness of acute hospital 1319–24.
treatment of older people with mental disorders? International King, G. and Farmer, J. (2009). What older people want: evidence from a
Psychogeriatrics, 17, 539–55. study of remote Scottish communities. Rural and Remote Health, 9, 1166.
European Commission (2009). Communication from the Commission to <http://www.rrh.org.au> (accessed 21.05.2012).
the European Parliament and the Council on a European Initiative on Kingston, P., et al. (2011). Psychiatric morbidity in older prisoners: unrecognized
Alzheimer’s Disease and Other Dementias. <http://ec.europa.eu/health/ and undertreated. International Psychogeriatrics, 23, 1354–60.
archive/ph_information/dissemination/documents/com2009_380_ Kramer, B.J. (1997). Gain in the caregiving experience: where are we? What
en.pdf> (accessed 21.05.2012). next? Gerontologist, 37, 218–32.
Lai, C.K., et al. (2009). Special care units for dementia individuals with National Institute for Health and Clinical Excellence (2011). NICE
behavioural problems. Cochrane Database of Systematic Reviews, 7(4), technology appraisal guidance 217: donepezil, galantamine, rivastigmine
CD006470. and memantine for the treatment of Alzheimer’s disease (review of NICE
Lawrence, V., et al. (2011). Threat to valued elements of life: the experience of technology appraisal guidance 111). NICE, London.
dementia across three ethnic groups. Gerontologist, 51, 39–50. Neville, C.C. and Byrne, G.J. (2007).The impact of residential respite care
Lewis, A. (1946). Ageing and senility: a major problem of psychiatry. Journal on the behaviour of older people with dementia: literature review.
of Mental Science, 92, 150–70. International Journal of Older People Nursing, 2, 2–8.
Lievesley, N. (2010). The future ageing of the ethnic minority population of England Norquist, G., et al. (1995). Quality of care for depressed elderly patients
and Wales. Runnymede and Centre for Policy on Ageing, London. hospitalized in the specialty psychiatric units or general medical wards.
Lindesay, J., et al. (2002). The second Leicester survey of memory clinics in Archives of General Psychiatry, 52, 695–701.
the British Isles. International Journal of Geriatric Psychiatry, 17, 41–7. O’Connor, D.W.M., et al. (1991). Does early intervention reduce the number
Livingston, G., et al. (2001). Mental health of migrants—the Islington study. of elderly people with dementia admitted to institutions for long term
British Journal of Psychiatry, 179, 361–6. care? British Medical Journal, 302, 871–5.
Loveday, B. (2011). Dementia training in care homes. In: Dening, T. and Orrell, M. and Hancock, G. (2004). CANE: Camberwell Assessment of N eed
Milne A. (eds) Mental health and care homes, pp. 327–44. Oxford for the Elderly. Gaskell, London.
University Press, Oxford. Orrell, M. and Katona, C. (1998). Do consultant home visits have a future
Mackenzie, C.S., Rosenberg, M., and Major, M. (2006). Evaluation of in old age psychiatry? International Journal of Geriatric Psychiatry, 13,
a psychiatric day hospital program for elderly patients with mood 355–7.
disorders. International Psychogeriatrics, 18, 631–41. Parker S.G., et al. (2009). Rehabilitation of older patients: day hospital
Martin, P., et al. (1992). Personality, life events and coping in the oldest-old. compared with rehabilitation at home. A randomised controlled trial.
International Journal of Aging and Human Development, 34, 19–30. Health Technology Assessment, 13 (39), 1–143.
Mason, A., et al. (2007a). A systematic review of the effectiveness and Pelosi, A.J., McNulty, S.V., and Jackson, G.A. (2006). Role of cholinesterase
cost-effectiveness of different models of community-based respite care inhibitors in dementia care needs rethinking. British Medical Journal,
for frail older people and their carers. Health Technology Assessment, 11 333, 491–3.
(15), 1–157. Philpot, M., et al. (2000). The prognosis of late-life depression in two
Mason, A., et al. (2007b). The effectiveness and cost-effectiveness of respite contiguous old age psychiatry services: an exploratory study. Aging and
for caregivers of frail older people. Journal of the American Geriatrics Mental Health, 4, 72–8.
Society, 55, 290–9. Post, F. (1944). Some problems arising from a study of mental patients over
Matthews, F.E and Dening, T.R. (2002). Prevalence of dementia in institutional the age of 60 years. Journal of Mental Science, 90, 554–65.
care. Lancet, 360, 225–6. Post, F. (1962). The significance of affective symptoms in old age: a follow up
Matthews, F., Brayne, C., and MRC Cognitive Function and Ageing Study study of 100 patients. Maudsley Monograph No. 10. Oxford University
Investigators (2005). The incidence of dementia in England and Wales: Press, Oxford.
findings from the five identical sites of the MRC CFA study. Public Post, F. (1965). The clinical psychiatry of late life. Pergamon, Oxford.
Library of Science Medicine, 2, e193. Price, E. (2010). Coming out to care: gay and lesbian carers’ experiences of
McCrae, N. and Banerjee, S. (2011). Modernizing mental health services for dementia services. Health and Social Care in the Community, 18, 160–8.
older people: a case study. International Psychogeriatrics, 23, 10–19. Prince, M., Bryce, R., and Ferri, C. (2011). Alzheimer’s Disease International
McCrae, N., et al. (2008). An extra pair of hands? A case study of the World Alzheimer Report 2011: the benefits of early diagnosis and
introduction of support workers in community mental health teams for intervention. Alzheimer’s Disease International, London.
older adults. Journal of Nursing Management, 16, 734–43. Prison Reform Trust (2009). Doing time: the experiences and needs of older people
Melzer, D., et al. (1999). Epidemiologically based needs assessment: in prison. <http://www.prisonreformtrust.org.uk/Portals/0/Documents/
Alzheimer’s disease and other dementias. In: A. Stevens et al. (eds) Health Doing%20Time%20the%20experiences%20and%20needs%20of%20
care and needs assessment: the epidemiologically based needs assessment older%20people%20in%20prison.pdf> (accessed 21.05.2012).
reviews—first series update, pp. 235–48. Radcliffe, Abingdon. Quinn, C., Clare, L., and Woods, B. (2009). The impact of the quality of
Middleton, S., Barnett, J., and Reeves, D. (2001). What is an integrated care relationship on the experiences and wellbeing of caregivers of people with
pathway? Bandolier, 3, 1–8. <http://www.medicine.ox.ac.uk/bandolier/ dementia: A systematic review. Aging and Mental Health, 13, 143–54.
painres/download/whatis/What_is_an_ICP.pdf> (accessed 21.05.2012). Ramakers, I.H. and Verhey, F.R. (2011). Development of memory clinics in
Miller, L.S., et al. (2010). Cognitive performance in centenarians and the the Netherlands: 1998 to 2009. Aging and Mental Health, 15, 34–9.
oldest old: norms from the Georgia Centenarian Study. Neuropsychology Randall, G.K., et al. (2010). Social resources and longevity: findings from the
Development and Cognition, Section B Aging Neuropsychology and Georgia Centenarian Study. Gerontology, 56, 106–11.
Cognition, 17, 575–90. Ratna, L. (1982) Crisis intervention in psychogeriatrics: a two-year follow-up
Mittelman, M.S., et al. (1996). A family intervention to delay nursing home study. British Journal of Psychiatry, 141, 296–301.
placement of patients with Alzheimer disease. A randomized controlled Reilly, S., et al. (2006). Standards of care in day hospitals and day centres:
trial. Journal of the American Medical Association, 276, 1725–31. a comparison of services for older people with dementia. International
Morgan, D.G., et al. (2002). Rural families caring for a relative with dementia: Journal of Geriatric Psychiatry, 21, 460–8.
barriers to use of formal services. Social Science and Medicine, 55, Roth, M. (1955). The natural history of mental disorder in old age. Journal of
1129–42. Mental Science, 101, 281–301.
Mukadam, N., Cooper, C., and Livingston, G. (2011). A systematic review Rothera, I., et al. (2008). An evaluation of a specialist multiagency home
of ethnicity and pathways to care in dementia. International Journal of support service for older people with dementia using qualitative methods.
Geriatric Psychiatry, 26, 12–20. International Journal of Geriatric Psychiatry, 23, 65–72.
National Collaborating Centre for Mental Health (2007). Dementia: the Royal College of General Practitioners (2005). Position statement: mental
NICE-SCIE guideline on supporting people with dementia and their carers health and primary care. <http://www.rcgp.org.uk/PDF/clinspec_
in health and social care. British Psychological Society/Royal College of printed%20version%20mental%20health.pdf> (accessed 21.05.2012).
Psychiatrists, Leicester/London. Royal College of General Practitioners (2011). Management of depression in
National Development Team for Inclusion (2011). A long time coming: part older people: why this is important in primary care. <http://www.rcgp.org.
1—strategies for achieving age equality in mental health services. National uk/pdf/NMH_10095_OPMH%20%20depression%20%20Feb%202011.
Development Team for Inclusion, Bath. pdf> (accessed 21.05.2012).
Royal College of Physicians and Surgeons of Canada (2005). CanMEDS Toot, S., Devine, M., and Orrell, M. (2011). The effectiveness of crisis
2005 Framework. <http://rcpsc.medical.org/canmeds/bestpractices/ resolution/home treatment teams for older people with mental health
framework_e.pdf>. problems: a systematic review and scoping exercise. International Journal
Royal College of Psychiatrists (2010). A competency based framework for of Geriatric Psychiatry, 26, 1221–30.
specialist training in psychiatry: specialists in old age psychiatry. <http:// Trieman, N., Leff, J., and Glover, G. (1999) Outcome of long stay psychiatric
www.gmc-uk.org/OAP_submission_content_final_May.pdf_33163658. patients resettled in the community: a prospective cohort study. British
pdf> (accessed 21.05.2012). Medical Journal, 319, 13–16.
Royal College of Psychiatrists and National Institute of Mental Health in Unützer, J., et al. (2002). Collaborative care management of late-life depression
England (2005). New ways of working for psychiatrists: enhancing effective, in the primary care setting: a randomized controlled trial. Journal of the
person-centred services through new ways of working in multidisciplinary American Medical Association, 288, 2836–45.
and multi-agency contexts. Department of Health, London. Van Citters, A.D. and Bartels, S.J. (2004). A systematic review of the
Royal College of Psychiatrists/Royal College of Physicians (1998). The care effectiveness of community-based mental health outreach services for
of older people with mental illness: specialist services and medical training. older adults. Psychiatric Services, 55, 1237–49.
RCPsych/RCP, London. Voigt-Radloff, S., et al. (2011). A multicentre RCT on community
Scottish Government (2009). Towards a mentally flourishing Scotland: policy occupational therapy in Alzheimer’s disease: 10 sessions are not better
and action plan 2009–2011. Scottish Government, Edinburgh. than one consultation. British Medical Journal Open, 1(1), e000096.
Scottish Government (2010). Scotland’s National Dementia Strategy. Scottish Von Korff, M., et al. (2011). Functional outcomes of multi-condition
Government, Edinburgh. collaborative care and successful ageing: results of randomised trial.
Selwood, A., et al. (2007). Systematic review of the effect of psychological British Medical Journal, 343, 6612
interventions on family caregivers of people with dementia. Journal of Warner, J.P., et al. (2003). The psychological health and quality of life of older
Affective Disorders, 101, 75–89. lesbians and gay men: a snowball sampling pilot study. International
Shah, A., Adelman, S., and Ong, Y.L. (2009). Psychiatric services for black and Journal of Geriatric Psychiatry, 18, 754–5.
minority ethnic older people. Royal College of Psychiatrists, London. Weiner, M.F. (2008). Perspective on race and ethnicity in Alzheimer’s disease
Shaji, K.S., et al. (2002). Revealing a hidden problem: an evaluation of a research. Alzheimers Dementia, 4, 233–8.
community dementia case-finding program from the Indian 10/66 Welsh Assembly Government and Alzheimer’s Society (2011). National
dementia research network. International Journal of Geriatric Psychiatry, dementia vision for Wales: dementia supportive communities. Welsh
17, 222–5. Assembly Government, Cardiff.
Shaw, C., et al. (2009). Systematic review of respite care in the frail elderly. Wenger, G.C., Scott, A., and Seddon, D. (2002). The experience of caring for
Health Technology Assessment, 13(20), 1–224. older people with dementia in a rural area: using services. Aging and
Shulman, K. and Arie, T. (1991). UK survey of psychiatric services for the Mental Health, 6, 30–8.
elderly: direction for developing services. Canadian Journal of Psychiatry, Wilberforce, M., et al. (2011). Towards integrated community mental
36, 169–75. health teams for older people in England: progress and new insights.
Snowdon, J. and Arie, T. (2005). A history of psychogeriatric services. In: International Journal of Geriatric Psychiatry, 26, 221–8.
B. Draper, P. Melding and H. Brodaty (eds) Psychogeriatric service Woods, R.T., et al. (2003). Support in the community for people with
delivery: an international perspective, pp. 3–20. Oxford University dementia and their carers: a comparative outcome study of specialist
Press, Oxford. mental health service interventions. International Journal of Geriatric
Stobbe, J., et al. (2010). Assertive community treatment for elderly people Psychiatry, 18, 298–307.
with severe mental illness. BMC Psychiatry, 10, 84. World Health Organization (1996). Psychiatry of the elderly: a consensus
Stonewall (2011). Lesbian, gay and bisexual people in later life. <http://www. statement. WHO/WPA, Geneva.
stonewall.org.uk/documents/lgb_in_later_life_final.pdf> (accessed World Health Organization (1997). Organization of care in psychiatry of the
21.05.2012). elderly: a technical consensus statement. WHO/WPA, Geneva.
Sumner, C.R. (2001). Telepsychiatry: challenges in rural aging. Journal of World Health Organization (1998). Education in psychiatry of the elderly: a
Rural Health, 17, 370–3. technical consensus statement. WHO/WPA, Geneva.
Teel, C.S. (2004). Rural practitioners’ experiences in dementia diagnosis and World Health Organization (2002). Reducing stigma and discrimination
treatment. Aging and Mental Health, 8, 422–9. against older people with mental disorders: a technical consensus statement.
Thompsell, A. (2011). Support to care homes. In: Dening, T. and Milne A. WHO/WPA, Geneva.
(eds) Mental health and care homes, pp. 221–36. Oxford University Press, Yorston, G. (1999). Old age forensic psychiatry. British Journal of Psychiatry,
Oxford. 174, 193–5.
CHAPTER 23
Primary care management
of older people with
mental health problems
Louise Robinson and Carolyn Chew-Graham
This chapter discusses the presentation and primary care manage- in the UK. The contract defines essential primary care services and
ment of the commonest mental health problems in older people. optional enhanced services that are additionally remunerated. The
We present cases drawn from our real-life practice, to represent contract links achievements in clinical and nonclinical care quality
clinical presentations and management within primary care and to financial rewards, through a Quality and Outcomes Framework
liaison with secondary care and the wider team. The management derived from evidence-based care (NHS Confederation, 2011).
of patients is discussed largely within reference to UK primary care The system encourages the delivery of optimum care in clinical
systems and policy, but the international readership should find domains, with emphasis on chronic disease management (Lester
parallels within their own healthcare systems. et al., 2006).
Common mental health problems in older people in

Primary Care in the UK primary care
In the UK, primary care services are an integral part of the National Primary care is on the front line in dealing with older people’s men-
Health Service (NHS) in which general practitioners (GPs) work as tal health, supporting families and managing people with complex
independent contractors. People are required to register as patients comorbidities. Older people consult almost twice as often as other
with a general practice; currently, a practice determines its bound- age groups: 22% of older people will have attended their GP within
aries and only accepts patients who reside within this area. The the last 2 weeks and 40% may have a mental health problem (The
GP works as a generalist and a provider of personal, primary, and Information Centre, 2007). We will now consider in detail the ini-
continuing care to individuals, families, and a practice population, tial management of the most common mental health problems
irrespective of age, gender, ethnicity, and problem. In England, the encountered in older people in primary care. These include delir-
Health and Social Care Act 2012 requires GPs to undertake a com- ium, delusions, depression and anxiety, and dementia.
missioning role with greater involvement in the purchasing of local
services for their patients. Delirium
GPs increasingly work with a range of healthcare professionals Delirium is common in hospital patients (10–20% in all medical
in a multidisciplinary primary healthcare team. The team includes patients but up to 40% in people with cancer), but it is becom-
a practice manager and administrative staff, practice nurses, and ing increasingly common in older people in the community, with
nurse practitioners or specialist nurses. Community nurses, such as an incidence of 13.5% in people over 85 years old (Folstein et al.,
district nurses, active case managers, Macmillan nurses, and health 1991).
visitors may be colocated or linked with a group of practices. Team
working is essential in order to manage the complex demands Clinical presentation
placed on general practice, which are partly due to caring for an Delirium is a syndrome comprising disturbance of consciousness
increasingly ageing population with chronic and multiple health (often manifest as impaired attention or concentration), cognitive
problems. The greater emphasis on preventative care, the transfer deficits (such as memory, orientation, and language problems), and
of clinical responsibility for some chronic diseases from secondary disturbed sleep–wake cycle; perceptual problems such as visual
to primary care, and the shift in service provision in order to deliver and auditory hallucinations and delusions; behavioural distur-
care closer to patients’ homes has contributed to these demands. bance (such as restlessness, agitation, and apathy) and alterations
The implementation of a new General Medical Services (GMS) in affect (such as fear), in addition to alterations in social behaviour
contract in 2004 fundamentally changed the way in which GPs work (e.g. lack of cooperation with reasonable requests, withdrawal, and
Table 23.1 How to differentiate between delirium, depression, and tested. The practitioner should consider the need for other investi-
dementia gations such as electrocardiogram (ECG) and chest X-ray. If more
intensive investigation is thought necessary then admission to hos-
Delirium Depression Dementia
pital should be considered.
Onset Acute Variable Insidious
Principles of management
Duration Days Variable Months to years
There are two treatment principles in the management of a patient
Course Fluctuates Possible diurnal Slowly with delirium: to identify and treat the underlying disorder and to
variation (worse in progressive ameliorate symptoms (National Institute for Health and Clinical
morning) (though may be Excellence, 2010a). The latter can be subdivided into: nonpharma-
stepwise) cological strategies and medical management. Nonpharmacological
Consciousness Impaired and Unimpaired Clear at onset approaches help keep the patient in touch with reality and therefore
fluctuating less likely to become distressed and agitated:
Attention and Inattentive Poor concentration, Poor memory ◆ Ensure effective communication and reorientation (e.g. explain-
memory Poor memory sometimes but without ing where the person is, who he/she is, and what your role is)
complaining of poor inattention and provide reassurance for people diagnosed with delirium.
memory
Consider involving family, friends, and carers to help with this.
Affect Variable Depressed, loss of Variable
interest and pleasure
◆ Managing the environment by avoiding sensory deprivation
in usual activities (e.g. a windowless room) or sensory overload (e.g. noisy envi-
ronment); ensuring day–night variation in the environment is
preserved; providing a large clock; avoiding unnecessary room
changes; and trying to have some familiar objects around the
alterations in communication) (Table 23.1). The term delirium is patient.
synonymous with ‘acute confusional state’. The onset is often sud- ◆ Ensure the patient’s safety by anticipating and taking steps to
den (hours or days) and fluctuation is a hallmark. The differential prevent complications such as pressure sores, falls, further infec-
diagnosis includes depression, which can be associated with agita- tions, constipation, and reduction in mobility.
tion or withdrawal, and dementia (see Table 23.1).
Two main presentations of delirium have been described: hyper- Pharmacological interventions aim to reduce symptoms that
active delirium (hallucinations, delusions, agitation, and disorien- distress the patient and/or add to risk. They are targeted at spe-
tation) and hypoactive delirium (cognitive impairment with apathy cific symptoms, e.g. pain, psychotic phenomena, and agitation.
or withdrawal, and less often accompanied by hallucinations and Psychotropic medications should be reserved for older persons in
delusions). The latter can easily be overlooked in older patients, distress or with psychotic symptoms, to prevent them endanger-
particularly those in residential (care) and nursing homes. ing self or others, with the short-term use of haloperidol or olan-
Delirium can be triggered by infection, dehydration, pain, bone zapine being suggested (National Institute for Health and Clinical
fracture, and side effects of medication, but those with pre-existing Excellence, 2010a), but neither are currently licensed in the UK
cognitive impairment or dementia are at increased risk. As age is for this indication. The use of psychotropic medication to manage
a risk factor in itself, people in nursing and residential homes are wandering on its own should be avoided.
at increased risk. Factors frequently combine, e.g. a person with For patients with psychotic symptoms, antipsychotic medications
dementia who becomes constipated or develops a urinary tract (‘major tranquillizers’) are often the pharmacological treatment of
infection. Delirium is common in end-of-life care and poor man- choice. Haloperidol is most frequently recommended because it has
agement will contribute to a poor-quality death and distress for few anticholinergic side effects, few active metabolites, and a small
carers. likelihood of causing sedation and hypotension. The initial dose
should be low—0.5–1 mg repeated if necessary every 2–4 h—with
Assessment careful titration of the dose against symptoms. Vigilance is needed
Delirium is a symptom of underlying problems, not a diagnosis for side effects such as excessive sedation, extrapyramidal side
in itself. If delirium is suspected, the practitioner should carry effects, and akathisia (motor restlessness), which may arise quite
out a clinical assessment based on the Diagnostic and Statistical rapidly after several days, so sedative medication should never
Manual of Mental Disorders (DSM-IV) criteria or short Confusion be prescribed without an arrangement for review. Patients with
Assessment Method (short CAM) to confirm the diagnosis (Inouye underlying dementia with Lewy bodies (see section on Dementia)
et al., 1990). It is important to assess cognitive function both as a may react adversely to antipsychotics so drugs like haloperidol are
baseline and to assess fluctuation. If there is difficulty distinguish- best avoided; lorazepam 0.5–1 mg orally which can be given up to
ing between the diagnoses of delirium, dementia, and delirium 2 hourly (maximum 3 mg in 24 h) is an alternative.
superimposed on dementia, treat for delirium first and ensure that
the diagnosis of delirium is documented in the patient’s clinical Referral and liaison
notes record. Information from patient and carers and a focused Patients admitted to a care home or hospital should be assessed for
physical examination are essential. Investigations in primary care their risk of delirium (National Institute for Health and Clinical
will include routine bloods such as full blood count (FBC), urea Excellence, 2010a). Patients with delirium should, where possible,
and electrolytes (U&Es), blood glucose, liver and thyroid function, be treated in their usual environment, but if the cause of the delir-
and C-reactive protein; a midstream urine sample (MSU) should be ium is unknown, if confusion and wandering are not controlled, or
CHAPTER 23 primary care management of older people with mental health problems 303
if the patient is at risk of developing dehydration, then admission to presentation. Whether these conditions are distinct from schizo-
hospital may be needed. Intermediate care may be an alternative if phrenia is unresolved. Aetiological factors include a genetic
available. This system of care allows for either an increase in nursing component, sensory deprivation (particularly deafness), and
care support for patients who are acutely ill but can be managed at long-standing social isolation. Psychotic symptoms can be associ-
home, or a brief admission to a facility that is intermediate between ated with aggressive or disruptive behaviour (Gilley et al., 1997)
hospital and home, that offers intensive nursing and physiotherapy and are often a source of distress to caregivers (Schneider et al.,
support, with a view to early discharge home. 1997). They can result in neglect and abuse of older patients (Steele
et al., 1990) and persistent symptoms often result in admission to a
residential or nursing home, which imposes a heavy financial bur-
Case 1: an older person becomes confused Mrs S is an den (Stern et al., 1997).
83-year-old lady who lives in a residential home and who over-
night became confused and withdrawn and was wandering around
Assessment
the corridors in the night. She was reported to be irritable and In principle, the assessment of psychotic symptoms in older people
argumentative when staff tried to get her up the next morning. is the same as in younger adults, but there are some nuances and the
The staff member in charge rang the practice to request a home differential diagnosis, especially where onset is new, should always
visit for Mrs S. The GP noted that Mrs S has a history of diabetes, include consideration of an organic (medical) cause, delirium, side
hypothyroidism, arthritis, hiatus hernia, and recurrent urinary effect of a drug, or an early dementia.
tract infections, and her medications are metformin, ramipril, Management
simvastatin, omeprazole, levothyroxine, and paracetamol. It is interesting to consider whether isolated psychotic symptoms
The GP visited Mrs S and established with the care home staff not distressing or adversely affecting the patient require treatment
that she had become suddenly unwell the previous evening. Mrs with medication (Kidder, 2003); certainly if the person is distressed
S was reported to have eaten breakfast and was drinking plenty. or at risk, pharmacological treatment is required.
She had been incontinent of urine, which was unusual for her. As older people are at particular risk of tardive dyskinesia, atypi-
On examination, she was disoriented in time and place and more cal antipsychotics may be more appropriate, although recent evi-
confused than usual. She was reluctant to be examined. She was dence (Douglas and Smeeth, 2008) suggests that the side effects
apyrexial and her pulse and BP were stable (compared to usual from atypical antipsychotics might have been underestimated and
readings), her chest was clear, and there were no focal neurologi- there is an increased risk of stroke and impaired glycaemic con-
cal signs. The staff had been unable to collect a urine specimen. trol. A careful assessment of potentially remediable environmental
A provisional diagnosis of urinary tract infection was made causes such as sensory deprivation, poor lighting, and social iso-
and she was commenced on trimethoprim, with a request made lation improve symptoms. Simple practical measures can reduce
to collect a urine specimen before and after treatment if possi- stimulations that produce psychotic symptoms, e.g. removing mir-
ble. The GP also took blood for FBC, U&Es, and blood glucose rors if reflections cause the delusion of having ghosts in the house,
(recent HbA1C and thyroid function tests had been satisfactory), or drawing curtains over windows if the patient has a delusion of
and requested that the staff monitor Mrs S carefully and dis- being spied upon. Addressing other contributory and causal factors
cussed how to manage the situation should Mrs S wander about such as physical illness and side effects of medication (particularly
during the night. The GP suggested that the staff should contact benzodiazepines, antiParkinson’s drugs, some cardiac drugs, nons-
the practice again if Mrs S did not improve. teroidal anti-inflammatory drugs) is equally important. Treatment
of older people with schizophrenia can be difficult because of lack
of insight and capacity, but response to antipsychotics is usually
Delusions and other psychotic symptoms good, especially if given as a long-acting ‘depot’ neuroleptic. Good
Psychotic symptoms are not uncommon in the older population negotiating skills will help, as patients will sometimes deny mental
and prevalence figures in community samples range from 0.2– disorder but agree to take ‘nerve medicine’.
4.7% (Targum and Abbott, 1999). In nursing homes, prevalence
rates from 10% to as high as 63% have been reported (Zayas and
Referral and liaison
Grossberg, 1998). If psychotic symptoms are suspected, an assessment of the risk to
Among older patients, psychotic symptoms can be seen in a wide the patient (and others) is required, and whether the patient can
range of conditions. The causes and clinical manifestations of the remain at home. Referral for specialist opinion from old age psy-
symptoms usually vary with the underlying condition. Psychotic chiatry will be indicated for new-onset symptoms, the urgency
symptoms of acute onset are usually seen in delirium secondary depending on risk and the availability of family and social support.
to a medical condition (see section on Delirium), drug misuse, With the population ageing, more people with psychosis diagnosed
and drug-induced psychosis. Chronic and persistent psychotic in earlier adult life are entering old age and may require ongoing
symptoms may be due to a primary psychotic disorder (chronic input from specialist services.
schizophrenia, late-onset schizophrenia, delusional disorders,
affective disorders), psychosis due to neurodegenerative disorders
Case 2: a difficult hospital discharge A patient, Deirdre R, was
(Alzheimer’s disease, vascular dementia, dementia with Lewy bod-
discharged from hospital to a local nursing home and is newly
ies, Parkinson’s disease), or chronic medical conditions.
registered with the practice. Her discharge letter stated that she
Delusional disorders in older people range from highly cir-
was admitted to hospital from a residential home with a chest
cumscribed persecutory delusions, through the presence of both
infection and that the previous home refused to take her back
delusions and hallucinations, to a full-blown schizophrenia-like
because of ‘difficult behaviour’. The discharge letter mentioned present with nonspecific symptoms such as malaise, tiredness, and
‘chronic schizophrenia’. Discharge medication was risperidone, insomnia, rather than disclosing depressive symptoms (Unützer
temazepam, lansoprazole, ramipril, and bendroflumethazide. et al., 1999), and these symptoms fit poorly with current classifica-
The day after discharge, the nurse in charge of the home rang tions of mood disorders. Such classifications have been generated
the practice and asked for an urgent visit because Deirdre was to reflect symptoms observed in younger people and have inher-
reported to be aggressive, had urinated in another person’s room, ent limitations for diagnosis of depression in older people, whose
and was talking to herself. presentation may differ because of ageing, physical illness, or both
The GP agreed to visit Deirdre, but first contacted the hospital (Chew-Graham et al., 2004). In addition, physical symptoms, in
ward to ask for further information about Deirdre’s usual behav- particular pain, are common and the primary care clinician may
iour. Deirdre had been seen by the old age psychiatry team whilst feel they represent organic disease. Similarly, forgetfulness may lead
she was on the ward and they were arranging follow-up, although health professionals to be concerned that the patient has cognitive
it was unclear when this might be. At the visit, the staff com- impairment and early dementia rather than depression (Unützer
plained that they had not been made aware of how difficult they et al., 1999). Older adults may have beliefs that prevent them from
were going to find managing Deirdre and asked if she could be seeking help for depression, such as a fear of stigmatization and
prescribed night sedation. The GP examined Deirdre to ensure that antidepressant medication is addictive (Givens et al., 2006), or
that there were no acute physical conditions to account for her may not consider themselves candidates for care because of pre-
increased confusion and arranged for an MSU to be sent. The GP vious experience of help-seeking (Chew-Graham et al., 2012). In
suggested that a joint visit with the old age psychiatrist might be addition, older people are reluctant to recognize and name ‘depres-
useful in order to plan her care and management within the care sion’ as a set of symptoms that legitimizes attending their GP
home, and arranged for this to be held later in the week. This (Chew-Graham et al., 2012), or they may misattribute symptoms of
seemed to be acceptable to the staff. major depression for old age. ill-health, or grief (Burroughs et al.,
2006) and use normalizing attributional styles that see depression
as a normal consequence of ill health.
Depression Primary care practitioners may lack the necessary consultation
skills and confidence to correctly diagnose late-life depression.
Depression is a major contributor to healthcare costs and is pro-
They may be wary of opening a ‘Pandora’s box’ in time-limited con-
jected to be the leading cause of disease burden in middle and
sultations and instead collude with the patient in what has been
higher income countries by the year 2030 (Mathers and Loncar,
called ‘therapeutic nihilism’ (Burroughs et al., 2006). In addition,
2006). Depression in later life, traditionally defined as age older
due to the time-limited nature of primary care consultations, clini-
than 65, is associated with disability, increased mortality, poorer
cal decision-making is often centred around prioritizing competing
outcomes from physical illness, and increased use of primary and
patient demands that may focus on medical comorbidities. This is
secondary care resources (Pearson et al., 1999). Population stud-
especially true in health settings like the NHS where the manage-
ies have demonstrated that depression severe enough to warrant
ment of people with long-term conditions is driven by guidelines
intervention is one of the commonest mental health problems fac-
and treatment algorithms that focus on single diseases. In these
ing older people, affecting around one in 10 older people in the
time-limited and highly structured environments, competing
community (Copeland et al., 1999). Major depression is a recurring
demands on health professionals’ time often lead to prioritization
disorder and older people are more at risk of recurrence than the
of physical health problems (Rost et al., 2000; Coventry et al., 2011).
younger population (Mitchell and Subramaniam, 2005).
Additionally, a lack of congruence between patients’ and profes-
According to WHO data, proportionately more people aged over
sionals’ conceptual language about depression, along with deficits
65 commit suicide than any other age group, and most have major
in communication skills on the part of both patients and profes-
depression. Older people who attempt suicide are more likely to
sionals, can lead to uncertainty about the nature of the problem
die than younger people, while, in those who survive, prognosis
and reduce opportunities to develop appropriate treatment strate-
is worse for older adults (Manthorpe and Iliffe, 2010). A recent
gies (Coventry et al., 2011).
comprehensive meta-analysis using studies with moderate to high
methodological quality showed that the point prevalence of major
Case-finding for depression in older people
depression in over 75s ranged from 4.6–9.3%, whereas rates for sub-
Table 23.2 shows the risk factors for depression in older people,
threshold depressive symptoms (those failing to reach diagnostic
identified from large community studies (Colasanti et al., 2010).
criteria) ranged from 4.5–37.4%. (Meeks et al., 2011). Most depres-
Thus, case-finding for depression in patients with such risk factors
sive episodes in late life will be a recurrence rather than a first-ever
has been advocated using the Whooley questions (see Table 23.3)
episode and the increased women to men ratio is in line with that
(Whooley et al., 2000); has until recently been rewarded under the
in younger adults. Prevalence rates of depression are increased in
GMS contract for diabetes and coronary heart disease (Department
brain disorders including dementia, Parkinson’s disease, and stroke,
of Health, 2010).
and also in people with long-term conditions, e.g. diabetes mellitus
There is some evidence (Arroll et al., 2005) that a further ques-
and cardiovascular disease (Moussavi et al., 2007).
tion, ‘Is this something you want help with?’ increases the useful-
Why does depression in older people go underdetected and ness of the screening questions in practice. This may help the flow
undertreated? of the conversation and the negotiation of discussion of distress
Older people consult their primary care practitioner more fre- (Coventry et al., 2011). An assessment of severity of the depres-
quently than younger people, yet depression is diagnosed less often sion should be made by the GP using a schedule such as PHQ-9
in older people (Chew-Graham et al., 2004). Older people can (Kroenke et al., 2001). In addition, it is vital that the GP explores
Table 23.2 Risk factors for depression in older people Table 23.4 Diagnostic and Statistical Manual of Mental Disorders
(DSM-IV) criteria for diagnosis of a major depressive episode. Depressed
Physical factors mood for ≥2 weeks not meeting these criteria is defined as a minor
Chronic disease: diabetes, ischaemic heart disease, heart failure, chronic depressive episode
obstructive pulmonary disease
Organic brain disease: dementia, Parkinson’s disease, cerebrovascular disease Nearly every day for the preceding 2 weeks the patient has experienced five or
more of the following:
Endocrine/metabolic disorders: thyroid disease, hypercalcaemia, B12 and folate
◆ Depressed mood for most of the daya
deficiency
◆ Decreased interest or pleasure in nearly all activities for most of the daya
Malignancy
◆ Marked loss or gain of weight or markedly increased or decreased appetite
Chronic pain and disability
◆ Excessive sleep or not enough sleep
Psychosocial factors
◆ Observable psychomotor agitation or retardation
Social isolation
◆ Tiredness or loss of energy
Change in financial circumstances
◆ Feelings of guilt or worthlessness
Being a carer
◆ Poor concentration or indecisiveness
Change of role and loss of social status
◆ Thoughts of dying or suicide, or suicide attempt
Bereavement and loss
a One of these features must be present.
Difficulty in adapting to illness/pain/disability
(Diagnostic and Statistical Manual of Mental Disorders (DSM-IV), American Psychiatric
History of depression Association, 1994.)
Being in institutional care
usually takes the form of executive dysfunction. meaning difficulty

Table 23.3 Case-finding questions for depression (Whooley et al., 2000) in initiating and persevering with mental tasks. Anxiety is a common
presenting or accompanying symptom and may mask the underly-
During the past month, have you often been bothered by feeling down,
depressed, or hopeless? ing depression. Dementia may alter the presentation of depression,
and primary care clinicians should be aware that sustained irritable
During the past month, have you often been bothered by having little interest
or pleasure in doing things? disruptive outbursts in patients with dementia may signify coexistent
depression. Apathy and withdrawal are not uncommon symptoms
A ‘yes’ to either question is considered a positive test.
and it is thought that apathetic presentations in later life with execu-
A ‘no’ response to both questions makes depression highly unlikely.
tive dysfunction may be due to vascular disease of the brain, to which
older people are prone. Alcohol misuse is often linked to depression.
Lastly, a persistent complaint of loneliness in an older person not
with the patient ideas and plans for self-harm, and factors prevent-
usually prone to such remarks, often accompanied by a request to be
ing the patient from acting on such ideas or plans.
rehoused, should prompt enquiry into mood, feelings, views on the
In terms of screening for depression in later life, the Geriatric
future, and assessment of risk, and a more systematic enquiry about
Depression Scale (GDS) (Yesavage et al., 1983) is widely used. False
biological symptoms of depression, along with a formal assessment.
positives mean that the use of the GDS in entire practice popula-
tions of older people is not justified, but opportunistic screening Assessment
or screening of at-risk groups may be useful, although there is little Diagnosis is based on clinical interview, observation of the patient’s
evidence to support this. The GDS has the advantage of having a behaviour, and, if possible, a collateral history from relatives and car-
comprehensive website listing short and long versions in a variety egivers. The GP should cover five areas in the primary care consulta-
of languages, many of which have cross-cultural validity (Rait et al., tion when depression is suspected in an older person: history, mental
1999). Other authors question the usefulness of such reductionist state, risk assessment, focused physical examination, and appropriate
approaches (Dowrick, 2009) and emphasize the value of professional investigations including FBC, U&E, liver and thyroid function tests,
judgment over the narrow use of schedules (Van Weel et al., 2009). vitamin B12 and folate, HbA1C, bone profile, and any further tests
dictated by clinical presentation. The risk of self-harm must be estab-
Symptoms of depression in older people
lished. Even seemingly medically trivial attempts at self-harm in older
Table 23.4 lists the DSM-IV criteria for diagnosis of a major depres-
people cannot be ignored (Manthorpe and Iliffe, 2010) and should be
sive episode. When taking a background history it is important
assumed to be due to depression unless proven otherwise. It is impor-
to identify factors that may precipitate and maintain depression
tant to check carefully for delusional ideas, as assessment and treat-
(Table 23.2).
ment of such patients require early referral for specialist opinion.
The following nuances are important in older people: somatic pre-
Severity of depression can be assessed using a validated assessment
occupation, hypochondriasis, and the morbid fear of illness, which
schedule such as PHQ-9 (Kroenke et al., 2001) and this is included
are more common presentations than the complaint of low mood or
in the QOF of the new GMS contract (in England and Wales).
sadness. This can cause problems for the GP, as a depressed patient’s
hypochondriacal complaints can be quite different from the bod- Management of depression in older people
ily symptoms one might expect from a knowledge of the patient’s Most patients with depression are managed in primary care set-
medical history. Subjective memory disturbance may be a prominent tings; however, a substantial number of patients are not recognized,
symptom and lead to a differential diagnosis of dementia, but true and those who are diagnosed often do not receive effective treat-
cognitive disturbance is also common in late-life depression. This ment (Chew-Graham et al., 2004). There is a good evidence base
for the management of depression in older people (Baldwin et al., reinforce. The unconscious goal may be to live under the same
2003). The treatment goals are to treat the whole person, reduce roof as one’s children. More positively, the family is often critical
risk, and achieve remission of depressive symptoms (because par- in ensuring a successful outcome in treatment, e.g. reinforcing
tial recovery is associated with later risk of relapse and chronic- messages about treatment concordance, exercise and activity, and
ity). A longitudinal primary care cohort study in the Netherlands goal-setting, and the primary care team should work with the fam-
reported that the median duration of a major depressive episode in ily as well as the individual patient.
late life was 18 months, with two-thirds of patients taking 3 years to The UK NHS has attracted significant resource to improve access
recover (Licht-Strunk et al., 2009). to routine, evidence-based, first-line treatment of common men-
There are effective treatments for the management of late-life tal health problems for adults of working age by rolling out the
depression (Wilson et al., 2001); however, only one in four Improving Access to Psychological Therapies (IAPT) programme
depressed older people receive effective pharmacological treatment across England. Initial evaluation from the demonstration sites sug-
and less than 10% a talking therapy (Singleton et al., 2001), despite gested that older people are not referred as readily as those under
the fact that many people express a preference for such treatments the age of 65 (Clark et al., 2008).
(Givens et al., 2006).
It is vital that the GP explores the patients’ views of their problem Stepped care and NICE guidelines
and the options that might be available to them. So, for example, a Stepped care for common mental health problems is recommended
patient’s views on antidepressants and talking through treatments in the primary mental healthcare setting within the UK (National
or psychosocial interventions, what to expect, and waiting times is Institute for Health and Clinical Excellence, 2009; National
vital. In addition, the GP has a role in signposting to third sector Collaborating Centre for Mental Health, 2010). The basic principle
agencies (if acceptable to the patient) and thus should be familiar of stepped care is that patients presenting with a common mental
with groups and local networks in his or her area. Exercise and health disorder will ‘step through’ progressive levels of treatment
activity are important. Behavioural activation is a technique that can as necessary (Bower and Gilbody, 2005). There is limited evidence
overcome the withdrawal and apathy that so often exists in late-life for this approach, particularly in older people, and it is suggested
depression. It works by helping the patient develop a schedule of (Unützer, 2002) that systematic models of care dedicated to proac-
activities (which the patient used to enjoy doing but has stopped tively managing depression as a chronic illness are required.
due to to low motivation and apathy), agreed with the patient, with Initial management by the GP should include information about
or without a written diary to support implementation. The GP can depression and referral to third sector resources such as Age UK.
use some of these techniques within the primary care consultation, Exercise is recommended by the NICE guidelines as a treatment
particularly advising about diet, exercise, and alcohol, encouraging for mild to moderate depression (National Collaborating Centre
behaviour change and goal setting, challenging negative thinking, for Mental Health, 2010), but there is only limited evidence for this
and teaching relaxation techniques. GPs can give self-help leaflets approach in older people (Singh et al., 1997). Training in the use of
to patients that reinforce the content of their discussion within con- the internet to increase social support has been shown to reduce
sultations. Regular follow-up and support is important and is called complaints of loneliness and depression (White et al., 2002) and
‘active monitoring’ in the NICE guideline for depression (National there is recent evidence that befriending has a useful function in
Collaborating Centre for Mental Health, 2010). the management of mild depression in older people (Lester et al.,
2011). Interventions offered need to encourage social engagement,
Psychosocial interventions such as befriending, and enhancement of creative, physical, and
Cognitive behavioural therapy (CBT), interpersonal psychother- social activity (Chew-Graham et al., 2012).
apy (IPT), problem-solving treatment (PST), and psychodynamic Collaborative care is recommended by NICE (National
psychotherapy are the most widely researched forms of psycho- Collaborating Centre for Mental Health, 2010) guidelines in step 3
therapy in later-life depression (Gatz et al., 1998; Pinquart and for people with depression and long-term conditions. There is evi-
Sörensen, 2001). In moderate to severe (nonpsychotic) depressive dence that this approach is effective (Katon et al., 2010). The inter-
episode, combining antidepressant medication with a psychologi- vention should involve (Gunn et al., 2006):
cal intervention such as CBT or IPT can improve outcomes further
◆ a multiprofessional approach to patient care delivered by a GP
(Reynolds III et al., 1999).
and at least one other health professional (e.g. a nurse, psycholo-
Behavioural activation has been shown to be an effective treat-
gist, psychiatrist, or pharmacist)
ment for depression (Cuijpers et al., 2007) and results from a 2009
metaregression analysis suggest that psychological therapy, partic- ◆ a structured patient management plan that facilitates delivery
ularly CBT, IPT, and PST, is equally effective in older and younger of evidence-based interventions (either pharmacological or
adults with depression (Cuijpers et al., 2009b). Combined psycho- nonpharmacological)
logical therapy and pharmacological therapy is more effective than ◆ scheduled patient follow-ups on one or more occasion, either
psychological treatment alone for older people with depression face-to-face or by remote communication (e.g. telephone)
(Cuijpers et al., 2009a). Anxiety management can be a highly effec-
◆ enhanced interprofessional communication between the mul-
tive adjunctive treatment for depressed patients, especially where
tiprofessional team who share responsibility for the care of
patients who are recovering from depression are left with residual
the depressed patient (e.g. team meetings, case conferences,
anxiety, low confidence, or phobic avoidance, which can under-
supervision).
mine functional improvement.
Family work is important: the patient may unconsciously use the A key component of collaborative care is the introduction of
family to foster invalidism, which the family may then unwittingly a care (or case manager) in primary care. The case manager acts
as the conduit between patients and professionals in primary and Referral and liaison
specialist care and works as the patient’s support or advocate to The role of specialist care in the collaborative care model has been
jointly determine problems, set goals and action plans, and offer described in the section on Stepped care and NICE guidelines.
education and problem-solving skills as ways to promote better Referral to an old age psychiatrist is necessary if the patient is at
patient self-care. The PROSPECT study (PRevention Of Suicide risk of self-harm or suicide, or neglect. Referral should also be con-
in Primary care Elderly: Collaborative Trial) studied the effect of a sidered when treatment with two antidepressants and talking treat-
depression care manager offering recommendations to GPs accord- ment has not led to an improvement of the patient’s symptoms. At
ing to a guideline and helping patients with adherence to medica- this stage, discussion with and/or referral to an old age psychiatrist
tion (Bogner et al., 2007) The IMPACT study (Improving Mood: is indicated. Treatment resistance may require a second antidepres-
Promoting Access to Collaborative care Treatment) utilized a case sant or addition of a mood stabilizer, both of which should only
manager coordinating care and delivering a specific psychosocial be initiated in specialist care, or consideration of electroconvul-
intervention (behavioural activation or problem-solving treatment) sive therapy (ECT). Guidance is outlined in the NICE guideline
with or without medication management, and liaising with both
the GP and the specialist mental health services (Unützer et al.,
2002). Both studies suggest the effectiveness of such a collabora- Case 3: multimorbidity and depression Sirfraz Ali consults his
tive care approach that has been shown to be acceptable to family GP to discuss the pain in his knees. He is reluctant to take pain-
physicians (Levine et al., 2005), and initial evidence from the UK killers as he says he is already taking so many tablets. He admits
(Chew-Graham et al., 2007) is promising. that he feels his diabetes has not been well controlled since his
Treatment with antidepressants wife died and says he feels lonely even though his grown-up
Selective serotonin reuptake inhibitors (SSRIs) are well established children all live nearby. He is tearful during the consultation
as first-line treatment for depression in older adults. A Cochrane and says he feels ashamed of how he feels. He says he is fright-
review included 32 randomized controlled trials of antidepres- ened of what his future holds. Sirfraz admits that he feels low but
sant treatment in people aged 55 or over and reported that SSRIs insists that this is understandable as his wife died and he feels
and tricyclic antidepressants (TCAs) had similar efficacy, but that a burden on his children. He says that he should be strong and
TCAs were associated with more side effects and withdrawal from feels angry with himself for being ‘like this’. He denies thoughts
treatment (Mottram et al., 2006). Consideration of a patient’s other of harming himself, although admits reluctantly that he wishes
medication is required when initiating treatment with antidepres- every night that he won’t wake up in the morning. He has had
sants, although an SSRI will usually be first line. If antidepressants recent blood tests which confirm a high HbA1C, but all other
are considered, it is vital that the GP explores the patient’s views tests are normal. A PHQ-9 is 18 and the GP suggests that Sirfraz
and concerns about such medication, and a full discussion about may be depressed. He is not happy to use this word and says that
these drugs, the time taken to work, and the possible side effects he ‘shouldn’t have mental problems’. The GP acknowledges his
are covered. feelings about the word ‘depression’ and suggests that there are
Ongoing support and review from the GP, accompanied by writ- ways of improving how he feels and that he might wish to attend
ten information, particularly when antidepressants are prescribed, is a local support group (the GP is aware of two groups in a local
important. The patient should be warned about common side effects community centre that are for South Asians, and one group in a
that can occur when tablets are started (such as nausea, fatigue, local church for people with diabetes), but also gives Sirfraz some
headache, and increased anxiety), including longer-term side effects written information about depression. They discuss things that
(such as reduced libido and weight gain), and doctors should be Sirfraz used to like doing and established that he likes cricket and
aware of the risk of GI bleeding (particularly if the patient is taking photography, so the GP asked him to think about how he might
aspirin) and hyponatraemia. The patient should be warned about get back to these hobbies.
the side effects of stopping an antidepressant abruptly and that it is On review 2 weeks later, Sirfraz is no better, he hadn’t felt able
vital to continue the antidepressant for at least 9 months (longer if to contact the support groups, and said most days he was just sit-
this is an episode of recurrent depression) to achieve remission of ting at home, was not answering the telephone, and had stopped
symptoms (National Collaborating Centre for Mental Health, 2010). his family coming round on two occasions. The GP asked if he
For further details of antidepressant prescribing, see Chapter 42. wished to talk more about how he felt and how this was affecting
NICE guidelines (National Collaborating Centre for Mental his life. Sirfraz said that talking wasn’t something he felt com-
Health, 2010) suggest that an SSRI should be tried for at least 2 fortable with, so the GP suggested that antidepressants might
weeks, possibly up to 6 weeks in older people, before a change in help, and discussed how it is thought they work, how long they
antidepressant is considered. It is appropriate to start at the thera- would take to work, and what the potential side effects might be.
peutic dose of an SSRI for older people (care being taken to ensure Although reluctant, Sirfraz agreed to take a prescription, but also
there are no drug interactions). An increase in dose can be tried agreed that he could be referred to the local IAPT service, saying
before an alternative SSRI is prescribed. Alternatively, a TCA could that he would be willing to be seen, although he did not want to
be prescribed, particularly if the patient has chronic pain or sleep participate in a group. The GP asked Sirfraz if he would return
problems. The GP should attempt to tailor the drug to the patient’s in a further 2 weeks. The GP contacted the local IAPT service to
symptoms. If there is no improvement in depressive symptoms, it see if they had therapists who work with people with long-term
is vital that the GP considers comorbidities, concurrent prescrib- conditions (LTCs) and depression, and also talked to the practice
ing, alcohol excess, continuing loss and loneliness, or a diagnosis of nurse about how the practice could better offer help to people
vascular depression. with depression and LTCs.
for depression (National Collaborating Centre for Mental Health, or self-help groups (Anxiety UK), and discussion of behavioural
2010). activation techniques, with an arrangement to follow the patient
up. This constitutes ‘active monitoring’ in the NICE guideline for
Anxiety anxiety (National Collaborating Centre for Mental Health, 2011).
Epidemiology Appropriate advice about alcohol and activity as in the manage-
Anxiety disorders are more common than depression in all age ment of depression should be given.
groups and as common in older people, and are accompanied by If there is no improvement of symptoms, then discussion with the
significant morbidity; the 2007 UK household survey reported a patient about the acceptability of referral for ‘low-intensity psycho-
9% prevalence of mixed anxiety and depression (McManus et al., logical interventions’ should take place. If this is unacceptable to the
2009). ‘Anxiety’ covers the terms generalized anxiety disorder patient, then the GP should discuss with the patient how the patient
(GAD), panic disorder, obsessive-compulsive disorder, and phobia would feel about taking antidepressants (SSRIs) and the rationale for
(National Collaborating Centre for Mental Health, 2011). GAD is this suggestion. Management of anxiety in older people should be
a common disorder, of which the central feature is excessive worry no different from management in younger adults (see Chapter 45),
about a number of different events associated with heightened ten- although the link with chronic physical health problems means that
sion. A formal diagnosis using the DSM-IV classification system a collaborative care approach (see Stepped care and NICE guidelines)
requires two major symptoms (excessive anxiety and worry about may be indicated, although there is limited evidence for this and fur-
a number of events and activities, and difficulty controlling the ther research is advocated (Buszewicz and Chew-Graham, 2011).
worry) and three or more additional symptoms from a list of six Referral and liaison
(feeling wound-up, tense, or restless; becoming easily fatigued, con- Occasionally, patients with GAD suffer severe functional impair-
centration problems, irritability, tension in muscles, and difficulty ment, or are at risk of self-neglect or self-harm (National
with sleep). Symptoms should be present for at least 6 months and Collaborating Centre for Mental Health, 2010, 2011). In such cases,
should cause clinically significant distress or impairment in social, urgent referral for a specialist opinion is required. Persistence of
occupational, or other important areas of functioning.
Symptoms of anxiety Case 4: a frequent attender Miss Peters is 67 years old and has
Patients complain of worry, irritability, tension, tiredness, or ‘nerves. presented at the desk asking to be seen. The receptionist asks the
but older people may complain of somatic symptoms that may cause GP on duty what she should do as Miss Peters seems quite agi-
difficulty for the GP and may result in unnecessary investigations tated. She informs the GP that Miss Peters has already been seen
for the patient. Anxiety often coexists with depression, and the GP in the practice once this week, and the GP sees from the compu-
also needs to be aware of the link with alcohol misuse and should ter records that she was complaining of stomach-ache and nau-
always explore alcohol consumption in older people who present sea. Today she is apparently complaining of chest discomfort and
with symptoms of anxiety. shortness of breath. The GP notes she is taking mirtazapine, rami-
pril, bisoprolol, calcium and vitamin D tablets, and paracetamol.
Assessment The GP asks Miss Peters to take a seat in the waiting room and
Some clinicians consider anxiety and depression to be part of a con- tells her that she will be seen at the end of surgery. Miss Peters
tinuum, and labelling the patient as having one disorder or the other is calm when she enters the consulting room, but soon becomes
may be less important than assessing the severity and impact on the agitated and tearful and talks about her constant fear that ‘some-
patient’s life. This may be a valid perspective in primary care, where thing awful is going to happen’. She is distressed and clutches her
people often present with mixed or comorbid problems, but it is impor- chest. The GP establishes that her chest pain is not exertional and
tant to distinguish which symptoms are most prominent in order to does not sound cardiac. The GP reflects to Miss Peters about her
focus explanations, identify appropriate management and resources, repeated attendances and her symptoms and asks her what she
and ensure referrals are made (Buszewicz and Chew-Graham, 2011). thinks might be wrong. Miss Peters insists that she must have a
Thus, the use of the GAD-7 has been advocated as a useful instrument heart problem and thinks she needs to go to hospital.
to assess severity of anxiety and thus direct management (Spitzer et The GP acknowledges her worries and suggests that there
al., 2006). The need to assess risk to the patient is often overlooked might be other ways of managing her problems and asks her
when the predominant symptom is anxiety, but patients are at risk of if she would complete a GAD-7. She agrees and scores 16. The
self-harm and self-neglect and the GP should be aware of this. GP suggests she might be suffering from anxiety and that there
Management of a patient with anxiety is help available. Miss Peters does not seem sure about this and
The NICE guideline for anxiety (National Collaborating Centre for insists that it is her body that has the problem and she says she
Mental Health, 2011) offers a stepped care model for the manage- has ‘nothing to be anxious about’. Initially, the GP gives Miss
ment of people with anxiety and this approach is appropriate for P some reading material about anxiety and provides a relaxa-
older people with anxiety symptoms, with or without depression. tion CD, along with the suggestion that setting herself a goal to
Thus, discussion with the patient about the symptoms and their achieve each day might be helpful. The GP also mentions referral
meaning should occur, followed by negotiation of a management to the local primary care mental health service and antidepres-
plan acceptable to the patient. When physical symptoms are the sants as options, and asks Miss P to consider both these before
presenting problem, appropriate physical examination may help to the next appointment, which they arrange in 2 weeks.
reassure the patient that their symptoms are being taken seriously, The GP is careful to document this consultation in detail in the
but repeated investigations should be avoided. Initial manage- records, so that if Miss Peters contacts the practice again before
ment should involve verbal and written information about anxiety the next scheduled appointment, the consulting GP is aware of
and signposting to age-appropriate support groups (e.g. Age UK) the plan negotiated at this consultation.
symptoms should also lead to discussion with a specialist, and are causing intellectual impairment. These may include sedatives,
research is required to assess the place of a collaborative care model tranquillizers, antiparkinsonian agents, and hypotensives; adjust-
for the management of people with anxiety. ment or withdrawal of these will clarify the diagnosis. Normal pres-
sure hydrocephalus, hypothyroidism, syphilis, hypercalcaemia, and
Dementia vitamin deficiencies as the cause of, rather than result of, dementia
In the UK, about two-thirds of people with dementia live in private are rarities in practice. For GPs, depression masquerading as demen-
households, with the majority of their care provided by primary tia (‘pseudodementia’) is probably the most common differential
and community care teams (Alzheimer’s Society, 2007). Although
a GP may only see a few new cases of dementia per year, these
numbers are predicted to double in the next 40 years (Alzheimer’s Case 5: a patient with memory loss The warden of a sheltered
Society, 2007); notwithstanding, dementia is currently one of the housing unit requested a GP to visit Mrs W, a 78-year-old female
major causes of care burden and disability in later life, greater than resident who had been ‘going downhill’ mentally for the past few
heart disease, cancer or stroke (Mathers and Leonardi, 2000). months with memory loss, intermittent confusion, incontinence,
and difficulty with self-care. The patient had a past history of atrial
Symptoms and initial presentation fibrillation and transient ischaemic episodes from cerebral emboli,
The diagnosis of dementia can be delayed by the insidiousness of and had been commenced on warfarin. The warden recalled a
the symptoms and the perceptions of both lay and professional peo- fall 6 weeks earlier, after which the patient had been examined
ple that this may be normal ageing (Bamford et al., 2007). General at hospital and reassured that no injury had occurred. Rereferral
practice is often the first point of contact for patients with mem- for further investigation resulted in the detection of chronic bilat-
ory problems or other suspicious symptoms; however, GPs appear eral subdural haematomas on CT scan, with no cerebral atrophy.
reluctant to use brief cognitive tests and also to refer people for Following surgery, her confusion cleared. Despite only partial
early assessment (Bamford et al., 2007). In UK practice, the early recovery of memory, she was again able to cope alone.
recognition and the diagnosis of dementia by GPs is variable, with
widespread underdetection reported (National Audit Office, 2007;
Ahmad et al., 2010). Individual GPs will have different abilities and diagnosis and should always be considered; if suspected, a trial of
confidence in diagnosing and managing dementia; however, educa- antidepressants may be indicated.
tional interventions can significantly improve dementia detection A physical examination of the patient, particularly to detect
rates by GPs (Downs et al., 2006). anaemia, cardiac disorders, neurological evidence of previ-
Initially, the patient may present with memory problems, com- ous stroke, and undiagnosed malignancy, is essential. Baseline
munication difficulties, behavioural issues, and/or personality investigations carried out by the GP should include FBC, eryth-
changes; they are often accompanied by a concerned family carer rocyte sedimentation rate, biochemistry, thyroid function tests,
(Bamford et al., 2007). As the illness moves into the moderate and syphilis serology, although results are more likely to detect
stages, memory and communication difficulties worsen and there coexistent medical pathology than to aid the differential diag-
may be concerns around personal care and/or safety issues, such as nosis of dementia. The correction of iron- or vitamin-deficient
nonconcordance with treatment in a previously compliant patient, anaemias, the control of cardiac arrhythmias, and treatment of
hoarding of medication, missed appointments, difficulty manag- hypothyroidism are good medical practices, but there is a risk
ing money, and poor coping at home. Suspicions of family carers of the GP diverting clinical interest into ‘medicalization of the
recorded in patients’ notes help prompt further evaluation at subse- patient’ rather than confronting the management of the underly-
quent contacts, and aid earlier diagnosis (Bamford et al., 2007). ing dementia.
Indirect presentation of a patient with possible dementia to other Use of cognitive assessment tools in primary care
members of the primary care team or other community profession-
A number of simple tools are available for use in the community
als is common. Practice receptionists or community pharmacists
to make an initial assessment of a person’s cognitive function
may report missed appointments, difficulties in patients’ manage-
(National Collaborating Centre for Mental Health, 2007). The most
ment of repeat prescribing, or obvious confusion or distress over
commonly used cognitive assessment tool is the Mini-Mental State
medication. The community nurse or social worker may observe
Examination (MMSE) (Folstein et al., 1975) marked out of 30: a
disorientation of a patient visited for other needs, or a health visi-
score of less than 25 is suggestive of dementia. However, this can
tor may report a family crisis precipitated by failure to cope with a
take up to 20 min to complete and may not be practical for use
person suffering from undiagnosed dementia.
within a primary care consultation, which is usually 7–10 min
Assessment in length. The General Practitioner Assessment of Cognition
The initial assessment should include a careful history from the (GPCOG) (Brodaty et al., 2002), and two other cognitive screening
patient and their main carer, with relevant information from the tests, the Mini-Cog Assessment Instrument (Borson et al., 2000)
patient’s notes; a physical examination with baseline investiga- and the Memory Impairment Screen (MIS) (Buschke et al., 1999),
tions and an objective cognitive assessment should be also carried have been found to be as clinically and psychometrically robust and
out by the GP (National Collaborating Centre for Mental Health, more appropriate for use in primary care than the MMSE (Milne
2007; Young et al., 2011). The first step for the GP is to exclude a et al., 2008). The GPCOG is estimated to take 5–7 min to com-
potentially treatable illness or reversible cause of the dementia, as in plete, with questions for both the patient and family carer, and thus
Case 5. A history of a recent fall suggests the possibility of a subdural may be more relevant for primary care physicians (Brodaty et al.,
haematoma and, if suspected, immediate referral to a geriatrician or 2002). An alternative, developed in primary care, is the six-item
neurologist should be made for an urgent CT brain scan. A review of Cognitive Impairment Test (6CIT), which performs as well as the
medication may identify incorrect dosage or drug interactions that MMSE and is easier to use (Brooke and Bullock, 1999). Addition of
a clock-drawing test may also be a useful quick and simple test for al., 2004; Pettiti et al., 2005). Prescribing of cholinesterase inhibi-
the GP to use (Shulman, 2000). tors to people with mild/moderate dementia is recommended
(National Institute for Health and Clinical Excellence, 2011) in
Specialist assessment and confirmation of the diagnosis
order to slow the rate of deterioration. Shared care protocols pro-
If the GP is highly suspicious of a possible diagnosis of dementia, vide a framework for GPs to continue prescribing such drugs once
he/she should refer the patient for a specialist multidisciplinary a specialist decision to initiate treatment has been made (O’Brien
assessment preferably to old age psychiatry or, if locally available, a et al., 2001).
memory clinic (National Collaborating Centre for Mental Health, In the UK, national guidance (National Collaborating Centre
2007). Increasingly, old age psychiatry and memory clinics have for Mental Health, 2007) and the National Dementia Strategy
access to CT and specialist scanning facilities to enable not only (Department of Health, 2009) both propose a systematic approach
a diagnosis of dementia to be made but also a specialist subtype of to the long-term support of people with dementia in the commu-
dementia to be confirmed. As illustrated in Case 6, the advantages of nity, with care plans tailored to the needs and circumstances of
early referral include: relief at acquiring an explanation for distress- individual patients and carers. Such an approach should include:
ing symptoms; access to anticholinesterase inhibitor drugs and spe- the provision of psychosocial interventions to help reframe
cialist cognitive rehabilitation and/or psychological interventions; dementia and reduce the rate of cognitive decline, the provision
of relevant information on prognosis, care options and financial
and legal issues; the recognition of personal and social needs;
Case 6: a patient with dementia George P, a 77-year-old man the assessment and support of key carers (Cameron et al., 2011);
with a history of hypertension, hyperlipidaemia, and polycythae- the treatment of underlying medical conditions or disability; the
mia, presented to his GP with concerns about his memory but management of behavioural problems and relationship difficul-
no disturbance of function. An initial MMSE score was 27/30. ties; and ensuring where possible, that the patient remains in their
The GP arranged to review him at 6-monthly intervals. One preferred place of care for as long as is feasibly possible (Robinson
year later, he attended with his wife who was very concerned. et al., 2010).
She stated he was becoming confused and disorientated when
outside; his MMSE was now 24. He was referred to the memory Nonpharmacological Interventions for the Person with
clinic, where a scan revealed changes consistent with vascular Dementia
dementia. He and his wife chose to complete a Lasting Power of There is increasing evidence that nondrug interventions can
Attorney for finance and to visit the local dementia café for peer improve short-term performance or reduce the rate of cognitive
group support. decline, especially if combined with anticholinesterase medica-
tion (Alzheimer’s Disease International, 2011). Such interventions
include cognitive training (interventions targeted at improv-
advice on appropriate information and support services, including ing standard tasks of cognition, memory, and problem-solving);
day care; carer assessment and support; and the ability to undertake cognitive rehabilitation (a more individualized approach to help
advance care planning (Robinson et al., 2010). people focus on residual cognitive skills and improve their coping
If patients are in the moderate to severe stages of the illness, strategies); cognitive stimulation (an approach that targets cogni-
it may be less distressing for them to have an initial specialist tive and social functioning through reality orientation and other
assessment in their home in the presence of relatives; this can also activities); and reminiscence therapy (discussion of past activities,
facilitate a joint assessment with the local authority social serv- events, and experiences). To date, evidence is strongest for cog-
ices care coordinator to discuss both diagnosis and a future care nitive stimulation therapies (Alzheimer’s Disease International,
plan. Despite the need to inform patients and their families sen- 2011); unfortunately, due to a shortage of such specialist services,
sitively and accurately of a diagnosis of dementia, disclosure rates psychological therapies may not be easily accessible to people with
vary considerably (Robinson et al., 2010). Systematic reviews dementia in primary care. Studies have explored the potential of
reveal that carers may be told the diagnosis more frequently than other community-based professionals, such as community mental
patients and that there is still considerable use of euphemistic health nurses, to successfully deliver psychosocial interventions
terms by health professionals (Bamford et al., 2004; Robinson et with subsequent improvements in carer coping and behavioural
al., 2010). and psychological symptoms of dementia (BPSD) management
Long-term management in the community (Moniz-Cook et al., 2008).
People with dementia live an average of 5 years from the emer- Management of Behavioural Problems in the Community
gence of symptoms (Xie et al., 2008). Together with their old age The vast majority of people with moderate to severe dementia will
psychiatry colleagues and/or community mental health teams, the experience BPSD at some time, particularly in the middle and later
GP and the primary healthcare team is responsible for coordinat- stages (Robinson et al., 2006). However, while the risks, such as
ing and directing the long-term clinical management of patients wandering and getting lost, from such behaviour are often not as
with dementia; local authority social services are responsible for high as carers fear (Robinson et al., 2007), they can lead to high lev-
the social care and support of the patient. Treatment of medi- els of carer stress, curtailment of the person with dementia’s activi-
cal conditions coexisting with dementia should be continued, if ties, and may be the crucial factor that leads to care home moves
supervision is available, until a decision is made for palliative care. (Balestreri et al., 2000).
However, hypertension treatment in patients with vascular demen- National guidance in the UK recommends nondrug approaches
tia may no longer be needed, as blood pressure has been shown to as the first-line management of such symptoms (National
decline with the onset of dementia (den Heijera et al., 2003; Qiu et Collaborating Centre for Mental Health, 2007), but in the
community this can be very difficult to achieve due to (1) a lack introduction of the Mental Capacity Act 2005 has provided a statu-
of access to specialist nursing and psychological services to enable tory framework for acting and making decisions on behalf of adults
an individualized assessment of patients and their environment, to who lack the mental capacity to do so for themselves. The code of
determine the antecedents, behaviours, and possible causes (‘ABC’ practice associated with the Act provides guidance on good practice
approach); and (2) the limited high quality evidence for the clinical for those acting under the terms of the Act. It contains (1) guidance
and cost effectiveness of nonpharmacological treatments (Robinson on how to assess capacity; (2) a statutory framework for an advance
et al., 2006). directive; and (3) advice on how to manage patient decisions, once
In the UK, a multidisciplinary pathway for the optimized man- the advance directive is applied (Department of Constitutional
agement of BPSD has recently been developed. This provides Affairs, 2007). The Act has also established an Independent Mental
evidenced-based guidance to primary, secondary, and commu- Capacity Advocate (IMCA) service; this provides representation
nity care professionals, including care home staff, on the holis- and support for individuals who lack capacity but who have no one
tic management of these challenging behaviours (Alzheimer’s else to support them when major decisions are being made about
Society, 2011). If the behaviour persists and becomes more severe their lives.
despite the use of nondrug approaches, the GP can prescribe a The Act has also seen the introduction of a new formal Power
short course of an antipsychotic drug, such as risperidone; how- of Attorney, the Lasting Power of Attorney (LPA), which has
ever, prescribing of these drugs should be time limited and only replaced the provision for existing Enduring Powers of Attorney
reserved for severe and distressing symptoms after careful assess- (EPA), although the latter will remain valid if executed before the
ment of the risks and benefits of their use and consideration of implementation of the new Act. LPAs will extend the areas in which
the type of dementia (O’Brien, 2008; Burns and Iliffe, 2009). There donors can authorize others to make decisions on their behalf, to
is some evidence for the use of other drugs for BPSD includ- include personal care and medical treatment, i.e. any matter of per-
ing antidepressants (citalopram; trazodone) and antiepileptics sonal welfare, in addition to LPA for property and financial affairs.
(carbamazepine). Advance care planning is a term used to refer to a series of discus-
sions between patients, their families, and care professionals about
Providing Information in Dementia
their future wishes for their care should they lose capacity. Following
People with dementia and their carers do not appear to receive these discussions, patients can formally record their wishes in a
either sufficient information or information in an acceptable number of ways, including the completion of an advance directive,
format (van der Roest et al., 2009). Family members and other or living will, as it was previously known (see Table 23.5). In a study
supporters should have access to information and advice from from the US, only 11% of people with dementia had made a living
sources that are accurate and allow them to address their indi- will (Mitchell et al., 2004). In the UK, the National End of Life Care
vidual concerns, which may be different to those anticipated by Programme website provides advice for both patients and profes-
clinicians (Wald et al., 2003; van der Roest et al., 2009). Accurate sionals about advance care planning and also examples of docu-
and up-to-date information should be available in primary and mentation (<http://www.endoflifecareforadults.nhs.uk/>).
secondary care, social care, and voluntary and community set-
tings (Audit Commission, 2000). Voluntary or third sector
organizations are a major source of information for people with Caring for the Carers: The Role of the GP
dementia and their families; in the UK, organizations such as the
Carers of people with dementia are more likely to experience
Alzheimer’s Society (<www.alzheimers.org.uk>), Age Concern
depressed mood, to report a higher burden and to have worse gen-
(<www.ageuk.org>), and Dementia Services Development
eral health compared with carers of people with other long-term
Centres offer a wide range of information and practical advice,
conditions (Brodaty et al., 2002). In dementia, carers may grieve
such as the 10 Helpful Hints for Carers book (see <http://demen-
or experience loss as their relative loses functional and cognitive
tia.stir.ac.uk>).
abilities and the tangible benefits of companionship, affection, and
People with financial problems may find caregiving particularly
intimacy (Zarit and Zarit, 2008); this experience has been likened
difficult and may benefit from early signposting to financial or debt
to ‘coping with a living death’ (Woods, 1989). Depressed mood in
advice agencies (Schneider et al., 1999). Welfare benefits provide
the carer is one of the factors that determine the move of the per-
some financial remuneration for people with a chronic illness and
son with dementia to residential care (Grasel, 2002). There is some
their family carers. In the UK, these include: Disability Living
evidence that carers are reluctant to seek professional help in such
Allowance, with care and mobility components for those aged under
circumstances (Toseland et al., 2002) and when they can no longer
65 years; Attendance Allowance if aged over 65, with special rules
cope (Bruce et al., 2002). In view of the above evidence, primary
if terminally ill; Housing Benefit; and the Invalid Care Allowance
care physicians should endeavour to provide proactive carer sup-
for carers aged between 16 and 65 who are spending at least 35 h
port and to monitor their health and wellbeing, in addition to car-
weekly caring for a severely disabled person. The Disability Parking
ing for the person with dementia (Cameron et al., 2011).
Scheme allows people with severe mobility impairment or their
In England, the 2006 General Practitioner Contract has encour-
carers to park close to the places they want to reach, and is available
aged a more proactive approach in supporting carers. Practices are
also to those with severe memory impairment.
rewarded for implementing a management system that includes a
Legal and Ethical Issues in Dementia protocol for the identification of carers and a mechanism for the
The GP is well placed to provide sensitive and timely discussion referral of carers for social services assessment in accordance with
of legal and financial matters, including information and advice the Carers (Equal Opportunities) Act 2004. At a personal level, car-
about seeking Power of Attorney and advance care planning for ers expect realistic information about dementia, the implications
future health or care preferences (see Chapter 56). In England, the of the diagnosis and its prognosis, and how to make best use of
Table 23.5 Outcomes of advance care planning discussions: Some areas of England have access to an Admiral Nursing
international and national terminology Service; these are specialist nurses with expertise in dementia care
and in particular carer support (Greenwood and Walsh, 1995).
◆ Statement of wishes and preferences: documents an individual’s wishes
However, to date, evidence shows little difference in patient or carer
and preferences for future care and is not legally binding. In the UK, this is
known as an advance statement. outcomes between families receiving this specialist service and
those supported by existing community teams (Woods et al., 2003).
◆ An advance directive for refusal of treatment (originally known as a ‘living
will’): this comprises a statement of an individual’s refusal to receive specific Although respite or short-break care is often offered as a means of
medical treatment in a predefined potential future situation. It is legally providing support to the carer and is positively received by them,
binding and comes into effect when a person loses mental capacity. In the the long-term benefits remain unclear (Mason et al., 2007).
UK, this is known as an advance decision to refuse treatment (ADRT).
◆ A proxy decision-maker or Power of Attorney (POA): this is a legally Ensuring Preferred Place of Care and
binding document whereby the person nominates another (‘attorney’) to
make decisions on his or her behalf should he/she lose capacity. In England, Care at Home
following the introduction of the Mental Capacity Act, this is now known The success of maintaining the patient at home depends on the
as a Lasting Power of Attorney (LPA); there are two separate aspects to carer’s reason for caring, the quality of the relationship with car-
LPA, one in relation to a person’s health and welfare and another in relation ers, and identifying and addressing the carers’ needs (National
to property and affairs.
Collaborating Centre for Mental Health, 2007). Regular visits by the
community psychiatric nurse, the community nurse, and a named
GP, regardless of whether the patient lives alone or with carers,
available facilities. Services should be adapted to the individual provides support and a coordinated, multidisciplinary approach
circumstances of the carer and the patient, with regard to differ- between primary and secondary care. Support services from local
ing racial, cultural, and religious backgrounds. The carer should be authority social services can include home care assistants, delivery
accepted as a contributing member of the care team, and effective of ready prepared meals, advice on financial benefits, and respite
communication with carers should be encouraged through discus- care. In the UK, voluntary organizations such as Age UK and the
sion and the keeping of clear, concise records. For housebound Alzheimer’s Society can provide a wide range of support, includ-
patients, a shared care record, held in the patient’s home, will main- ing information and advice (legal and financial) as well as practi-
tain continuity and improve communication between the different cal support via peer support groups, dementia cafes, carer support
professionals involved. groups, sitting services, and respite care.
The mental and physical health of carers is paramount if commu- In addition to behavioural disturbances, other distressing prob-
nity care is to be a realistic, long-term possibility. A meta-analysis lems arise with the progression of dementia and include sleep distur-
of psychosocial interventions for carers of people with dementia bance, swallowing difficulties, food and drink refusal, incontinence,
revealed that such interventions can reduce psychological morbid- and immobility. Advice on regular toileting to encourage continence
ity for carers and help their relatives stay at home longer; interven- and, if incontinence is a problem, the provision of aids such as pads,
tions that were intensive, individualized, and also included people pants, and protective mattress sheets, are available through the com-
with dementia, as well as their carers, were the most successful munity nursing services. Referral to appropriate allied health pro-
(Brodaty et al., 2003). A systematic review explored in detail the fessionals, such as the community physiotherapist, speech therapist,
effects of combined carer/patient interventions on the health and and occupational therapist for their specialist assessment should be
wellbeing of both groups (Smits et al., 2007). These programmes considered to enhance a patient’s functional health and also facili-
improved the general mental health of carers and were often effec- tate access to aids and adaptation to help the family cope at home.
tive in delaying admission to care homes; however, effects on carer
burden and coping, physical health, and survival were less conclu- The Patient with Dementia
sive (Smits et al., 2007).
The carer may or may not be registered with the same practice
Who Lives Alone
as the patient. Overall, the evidence suggests that clinicians need Providing primary care services for older, demented patients who
to tailor interventions to the specific needs of individual carers and live alone poses special problems, and may only be possible through
systematically assess carer needs (Pinquart and Sorenson, 2006; the goodwill of neighbours and external social care support. If
Sörensen et al., 2006). As it is the family that bears the major bur- access to the home is unreliable, arrangements should be made
den of caring for older people, the GP is the pivotal support for the with neighbours to hold emergency keys. Neighbours are reassured
family unit. The GP’s role should include offering advice to carers by contact with the GP and by discussion about when to request
on their personal health, acknowledging the carers’ crucial contri- help. If there is no available supervision from family carers or the
butions and the problems they experience, and giving emotional community nurse team, the taking of medication can be simplified
support by counselling the carer on their attitudes and expecta- by use of a ‘Dosett’ box in which medication is placed into daily
tions (Cameron et al., 2011). Vigilance by the GP in detecting early dosage compartments dispensed on a weekly basis by a pharmacist.
signs of strain or poor coping will help to avert crises; high rates of Single demented patients may cause concern over security, and per-
carer abusive behaviours towards care recipients have been recently sonal road and home safety. The local police welcome information
reported in the UK (Cooper et al., 2009). about these vulnerable residents and this is particularly encouraged
The community nurse team may also have a valuable supporting if there is a local neighbourhood security scheme. When eventu-
role, teaching carers the practical skills needed for home nursing ally these patients require admission to a protected environment,
and providing emotional support. a residential care or nursing care home is more appropriate than a
Case 7: a person with dementia who lives alone An 84-year-old respite beds in residential care or hospitals. The GP should be pre-
woman with dementia who had no relatives lived alone with pared to advise carers and families on respite care, and offer infor-
difficulty. The social services arranged admission to a nearby mation on the range of options and the referral process appropriate
sheltered housing unit. Within a short time she appeared more to the patient’s level of dependency.
confused, her room became squalid, and she roamed the streets For support in the home, domiciliary care may be offered by vol-
and was frequently returned by the police. She developed para- untary organizations or home care agencies such as Age UK, the
noid behaviour, attacked several residents, and was formally Alzheimer’s Society, and local authority social services. Families and
admitted to hospital under Section 2 of the Mental Health Act. carers are reassured to know that voluntary organizations are regu-
Following assessment and treatment, she was stable enough to be lated and required to work to specific care standards. Short-term
transferred to a nursing home for people with dementia. care can be arranged in residential care homes for patients who
are mobile and only mildly confused; short-term care in nursing
homes or hospitals should be arranged if the patient needs more
intensive nursing care, is seriously confused, or doubly inconti-
sheltered housing unit where the need to become familiar with new nent. If the patient is self-funding, private care arrangements are
surroundings whilst still living independently is liable to accentuate made, but if financial assistance is needed, referral for a commu-
confusion, as illustrated in Case 7. nity care assessment should be made to the local authority social
services by the carer or GP well before the planned admission date
Social Care for People with Mental Health (see Chapter 26). Patients are assessed against eligibility criteria for
funded care. When a patient requires nursing care or supervision,
Problems or needs multidisciplinary assessment and rehabilitation, admis-
Social care covers a range of services and support to help people sion to hospital or a nursing home is the responsibility of the NHS.
maintain their health and independence in the community. Such Since April 2004, the NHS and local authority councils have been
services include: home care (personal care, meals, laundry, shop- required to identify people who are likely to require short-stay res-
ping), day services, respite care, and, if available, residential and/ pite care, and to agree arrangements for their management.
or nursing care. In the UK, the White Paper on health and social Although respite admission provides an opportunity for multi-
care aimed to improve the current provision of community care disciplinary assessment of the patient, and may delay institutionali-
through increasing the choices available to users of these serv- zation, it may not reduce the overall burden on the carer. Heavily
ices and creating more integrated services. Initiatives include the dependent patients may need regular admissions, e.g. for 2 out of
inclusion of a Personal Health and Social Care plan in a patient’s every 8 weeks, to ease the strain on carers. Relatives appreciate the
records to integrate health and social care provision, the estab- opportunity of visiting the home or hospital beforehand to gain
lishment of joint health and social care teams, and the introduc- confidence in the staff and the surroundings, as the family may
tion of personal budgets, whereby people who require social care have reservations about delegating care to staff who are less famil-
can opt to receive payments directly and purchase services them- iar with the patient’s condition than they are. Disadvantages of res-
selves, potentially enabling greater choice and control. Evidence pite care for the carer include the difficulties of visiting patients in
has revealed that personal budgets have the potential to achieve hospital or nursing homes, feelings of sadness, loneliness, and guilt
greater efficiency whilst giving people greater control and satis- on being separated from their relative, and the possibility of insti-
faction (Alakeson, 2007; Carr, 2010). In dementia care, social care tutional cross infections. Some carers, however, refuse offers of help
also has a safeguarding role in terms of the increased vulnerability or indeed any support; others may stoically underplay the strain of
of people who may not have the mental capacity to make their caring, and deny the seriousness of the illness in their relative.
own decisions.
In England, the Health and Social Care Act 2012 has introduced Situations requiring hospital admission
changes that adapt the organization of care to meet the demands A point may be reached when the carer is no longer willing, or able,
of an ageing population. A key component of the Act is the intro- to cope. Increasing debility of the patient outstrips the tolerance
duction of clinical commissioning groups whereby clinicians will of the carer and is particularly likely to occur when sleep distur-
be in charge of shaping and purchasing local services to meet the bance, wandering, incontinence, and immobility are problems. The
needs of their local populations. It is hoped this will provide a patient then requires long-term care. It is important to be aware of
more effective and more integrated health and social care system. the abuse of older people, which may be a symptom of strain in the
An overview Health and Wellbeing Board will bring together carer (see Chapter 59). If this is suspected, the GP should refer the
both local commissioners of health and social care and patient patient to social services for an assessment, and perhaps for transfer
representatives to ensure ICPs for those living with long-term to a place of safety in hospital or residential care.
illnesses. Acute admissions mostly occur when there is unexpected illness
in the carer, or when a mildly demented patient who lives alone
Respite care suffers a self-limiting illness that, under normal circumstances,
Carers are at risk of increasing isolation and exhaustion through the would not require admission. In some areas, local schemes such
intensity of their 24-h responsibility. To prevent this, and to delay as ‘Hospital at Home’ initiatives and rapid response nursing teams
institutionalization, the GP should emphasize the value of relief are available; these provide comprehensive 24-h nursing at home
care, which can be organized on a daily basis, through day centres for people who would traditionally have been cared for in hospital.
or sitting services, or on a longer-term basis through rotation of Out-of-hours emergency calls by relatives who infrequently visit
care with other family members, or planned regular admissions to a mildly demented patient living alone, and insist that ‘something
must be done’ before their return home, is not an uncommon crisis, have recourse to specialist advice. This is a useful option when
and may express their own guilt. insufficient home care or community support makes continu-
It is vitally important to provide adequate and timely information ing home care untenable, particularly when mentally impaired
between hospital staff and primary care teams involved in caring patients require acute generalist care or rehabilitation. Continuing
for older people with dementia or other mental health conditions, care schemes, rehabilitation, and palliative care may be provided
as their needs are often multiple and complex. In the UK, current by community outreach teams, or by domiciliary and outpatient
commissioning guidance for GPs on dementia care recommends therapy services as alternatives to day hospital or inpatient care.
that initial notification of the diagnosis is made by facsimile (fax)
to improve communication and allow the GP to provide immedi- Home care services and residential care
ate support and further discussion with the family (Department Home care provision by local authority social services and inde-
of Health, 2011). For older people with mental health issues who pendent voluntary sector organizations has allowed people to
have been admitted to hospital, prompt discharge letters should remain in their own homes for longer than was previously pos-
include information on the patient’s care needs, particularly in rela- sible before admission to a residential care home or nursing care
tion to community services and carer support, medication, general facility. Intensive home care services and increased contact hours
condition, and follow-up requirements. Sending discharge letters by care staff have contributed to improvements in the quality of life
with patients to be handed to the GP is unreliable, especially when and independence of older people (NHS Health and Social Care
patients have poor mobility, live alone, or suffer memory loss. First Information Centre, 2006). The GP has a coordinating role and acts
attendances, hospital discharges, and a change in the treatment or as a point of contact for community care staff. The decision to trans-
condition of a patient should be reported, as should annual progress fer a patient into residential care is difficult and should only be taken
of patients attending for long-term care. Letters should summarize after all possible community support services in the home have been
the findings at consultation, an assessment and management plan, explored with the patient’s relatives and the community planning
and have educational value for the GP. team. The latter consists of community nurses, social services, and,
When patients are discharged or managed in outpatients, effec- in patients with complex needs, the community matron employed
tive liaison between hospital staff and GPs is essential, especially by the local primary care team. Patients or their relatives seeking
when patient medication is initiated or changed; the consultant admission to a home are advised to discuss the situation with the
should give the GP timely notification of the patient’s diagnosis and community nurses caring for the patient and with the patient’s GP
details of the drug therapy. The doctor who signs the prescription before requesting assessment for a nursing home placement.
has legal responsibility for the consequences of drug treatment, Nursing homes lie predominantly within the independent sector
and must not prescribe without adequate clinical information; it is and are subject to registration with the NHS (see Chapter 27). Medical
therefore unacceptable for hospital staff to expect GPs to do so. For care in nursing homes may be provided by an attending medical officer
patients under consultant supervision, medication should normally appointed by the home, or more usually by a local general practice
be prescribed in hospital and dispensed in the hospital pharmacy under contract for the provision of general medical care. Patients in
for not less than 14 days at discharge, or after referral to and treat- residential care homes register with a local GP and have access to gen-
ment in outpatients to ensure they have enough medication. eral medical care in the same way as the general population. Studies
have shown that in the population living in residential care, there is a
Intermediate care high prevalence of physical dependency and depression (McDougall
The development of rehabilitation and care schemes that are inter- et al., 2007; Seitz et al., 2010) and even higher prevalence rates for
mediate between acute hospital care and long-term institutional dementia (Matthews and Dening, 2002). This results in high rates of
care has led to a range of alternative provision termed intermedi- consultation, prescribing, and referral for the attending GP, with the
ate care. Such services are designed to maximize independence and workload for patients in nursing homes being double that for other
prevent unnecessary hospital admissions. In the UK, the National patients over the age of 74 years (Carlisle, 1999).
Service Framework for Older People (Department of Health, 2001)
set clear targets for the future expansion of intermediate care serv-
ices. Most schemes offer short-term interventions (1–6 weeks) and
Managing terminal illness at home
involve cooperative working with other agencies including primary A GP can expect an average of 20 deaths from their list each year,
care teams, local authority social services, and the voluntary and of which about one-quarter will be at home (Gold Standards
private sectors. For acute care, recent developments aim to bridge Framework, 2006). Most of these will be among older patients, some
the divide between hospital trusts and local authorities. Examples of whom will suffer from mental illness. In the UK, over 40% of peo-
include innovative services such as ‘hospital at home’ or ‘supported ple with dementia die in the community (Houttekier et al., 2010). It
discharge’ schemes in which the patient is discharged early from is the GP who coordinates their care, both for those who die at home
hospital and supported by intensive home nursing; these are accept- and, in the time preceding admission, for those who die in hospital
able to both patients and carers and provides care as effectively as or hospice care. About 70% of older people with dementia die from
hospital admission (Wilson et al., 1999). However, the cost effec- bronchopneumonia, a finding confirmed on post-mortem (Burns et
tiveness of intermediate care schemes in general remains uncer- al., 1990), whilst cardiovascular disease, pulmonary embolus, septi-
tain (Steiner et al., 2001) and the availability of hospital at home caemia, and renal failure account for the remainder of deaths.
schemes is variable (Corrado, 2000). When patients with dementia become ill with a respiratory infec-
An alternative to acute hospital admission is the use of small tion, the GP should discuss the poor prognosis with relatives and
community hospitals or nursing homes, in which inpatient care discover their preferences and their knowledge of their relative’s
is provided by GPs and their multidisciplinary teams who usually wishes for the preferred place of care. Wherever the patient is cared
for and whatever the mental status, the principles of good termi- Having enlisted the cooperation of an interested general practice,
nal care apply. These include: mouth care; the control of distressing it is essential to gain consent from participating GPs and, at the
symptoms such as pain and constipation; the involvement of com- outset, to visit the practice to explain the project to all staff. The
munity nurses and specialist palliative care services; and attention acceptability of protocols, letters, questionnaires, and pilot stud-
to the need for emotional, practical, and spiritual support for both ies should be confirmed with staff, before seeking approval from a
patients and their families. local research ethics committee. Establishing a link with a key prac-
People with dementia and their families require more input from tice administrator and a GP from each practice facilitates effective
healthcare and social services prior to death than do people with organization. It is recommended practice, according to the require-
terminal cancer (McCarthy et al., 1997). In the US, specialist units ments of the Data Protection Act, 1998, that researchers should
similar to hospices have been introduced for people with demen- obtain the consent of the patients or subjects before the practice is
tia. These have been shown to be more effective clinically and eco- able to release their contact details to the research team.
nomically than traditional long-term care (Volicer et al., 1994). In communicating with patients, it is essential to provide infor-
However, such examples of specialist care for people with dementia mation on the research project and its potential benefits. The
are limited in the UK (Sampson and Robinson, 2009; Treloar et al., identity of the research clinician and his/her relationship with the
2009). The National Institute for Health and Clinical Effectiveness practice should be explained. Patients appreciate a choice of venue
(NICE, 2010b) has recommended the use of end of life care path- for an interview, and usually the preferences range between home,
ways (Ellershaw et al., 1997; Thomas, 2003) to improve the quality the surgery, and the hospital. The researcher should provide an
of care at the end of life. These care pathways contain detailed guid- identity card, a letter of introduction from the link GP, and a con-
ance on how to transfer the principles of specialist palliative care, as tact telephone number for security reasons and to reassure patients,
given in hospices, to other settings in which patients are cared for relatives, and neighbours. Where door-to-door surveys are being
during their last few days of life. Practices are required to maintain carried out, it is essential to inform the local police, because neigh-
a palliative care register and regularly to review the listed patients as bours will rightly be highly suspicious. There is usually a long
indicated in the QOF (NHS Confederation, 2006). lead-in time when setting up research and the actual interviewing,
Death from dementia is not unexpected and in theory the carer so a check on a patient’s health status should be made by practice
of a person with dementia has time to adjust to the prognosis of staff prior to visiting so as to avoid the distress of including acutely
this fatally progressive disease. Nevertheless, the intensity and inti- ill or recently deceased subjects. A common problem when inter-
macy of caring leaves the main carer bereft of a role after a patient’s viewing in the community is how to respond to patients’ requests
death. The carer’s expected sense of relief may be overshadowed for medical advice. As the interviewer is not providing general
by the bereavement, and he or she will require much consolation medical care, any difficulties with the patient’s consent should be
and counselling; yet such support from primary care is not always referred to the patient’s doctor.
forthcoming. On completion of the research, feedback to the general practices
involved helps to integrate the results with the clinical needs of
Primary care: opportunities for research in patients, and to foster a sense of purpose and satisfaction. It is cour-
teous to acknowledge the contributions of participating practices
older people’s mental health in papers submitted for publication. Indeed, if a GP has personally
General practice provides a wealth of opportunities for research in the collected research data, joint authorship should be considered. This
area of older people’s mental health. To achieve a whole-population will provide a sense of goodwill which may lead to further fruitful
approach in epidemiological surveys or in studies of community collaboration with general practice.
care-based interventions, it is worth considering the potential of gen-
eral practice; a successful example is the Medical Research Council Conclusion
Cognitive Function and Ageing Study (MRC CFAS) which has recruited
a cohort of 13,000 people aged 65 years and over to explore their cogni- Recognition by the primary care physician (PCP) or GP of early
tive, physical, and psychological health. Advantages of recruitment of changes, use of clinical intuition, and acting on family worries are
patients from primary care include: the possibility of a more accurate vital in the earlier detection of mental health problems in older peo-
database compared to the electoral register; a personal approach to ple. An awareness of the increased risk of depression in older people
the patient through the GP, which may lead to higher response rates; with comorbid physical conditions is vital. Members of the primary
access to the patient’s medical records and up-to-date chronic disease care team should provide holistic care for older people with mental
registers; and the GP’s personal knowledge and contact with patients, health problems, and their carers, both at the point of diagnosis and
which facilitates data collection and interpretation of results. in the longer term. Appropriate referral to secondary care and good
The question of whether to involve multiple small practices or liaison and collaboration between primary and secondary care are
one large group practice should be considered, along with the vital in the management of older people with mental health prob-
demographic balance of the practice populations; this avoids the lems. Knowledge of third sector services and appropriate referral
use of skewed population samples and ensures the recruitment of can improve the support available to older people and their carers.
sufficient patient numbers. In the UK, the introduction of national
research networks, including the Primary Care Research Network, References
and disease-specific research networks, such as the Mental Health Ahmad, S., et al. (2010). GPs’ attitudes, awareness, and practice regarding early
Network, and the Dementia and Neurodegenerative Diseases diagnosis of dementia. British Journal of General Practice, 60, 666–74.
Research Network, have been successful in facilitating primary Alakeson, V. (2007). Putting service users in control: the case for extending
care-based research. self-direction into the NHS. Social Market Foundation, London.
Alzheimer’s Disease International (2011). World Alzheimer Report 2011. The Chew-Graham, C.A., et al. (2012). Why may older people not present to
benefits of early diagnosis and intervention. Alzheimer’s Society, London. primary care? Messages from secondary analysis of qualitative data.
Alzheimer’s Society (2007). Dementia UK: the full report. Alzheimer’s Society, Health and Social Care in the Community, 20(1), 52–60.
London. Clark, D.M., Layard, R., and Smithies, R. (2008). Improving access to
Alzheimer’s Society (2011). Optimising treatment and care for people with psychological therapy: Initial evaluation of the two demonstration sites.
behavioural and psychological symptoms of dementia. A best practice guide LSE Centre for Economic Performance, London.
for health and social care professionals. Alzheimer’s Society, London. Colasanti, V.C., et al. (2010). Tests for the evaluation of depression in the
Arroll, B., et al. (2005). Effect of the addition of a “help” question to two screening elderly: a systematic review. Archives of Gerontology and Geriatrics, 50,
questions on specificity for diagnosis of depression in general practice: 227–30.
diagnostic validity study. British Medical Journal, 331(7521), 884. Cooper, C., et al. (2009). Abuse of people with dementia by family carers:
Audit Commission (2000). Forget me not. Audit Commission, London. representative cross sectional survey. British Medical Journal (Research),
Baldwin, R., et al. (2003). Guideline for the management of late-life depression British Medical Journal, 338, b155.
in primary care. International Journal of Geriatric Psychiatry, 18, 829–38. Copeland, J.R., et al. (1999). Depression in Europe. Geographical distribution
Balestreri, L., Grossberg, A., and Grossberg, G.T. (2000). Behavioural and among older people. British Journal of Psychiatry, 174, 312–21.
psychological symptoms of dementia as a risk factor for nursing home Corrado, O.J. (2000). Caring for older hospital-at-home patients. Age and
placement. International Psychogeriatrics, 12, 59–62. Ageing, 29, 97–8.
Bamford, C., et al. (2004). Disclosing a diagnosis of dementia: a systematic Coventry, P.A., et al. (2011). Talking about depression: a qualitative study of
review. International Journal of Geriatric Psychiatry, 19, 151–69. barriers to managing depression in people with long term conditions in
Bamford, C., et al. (2007). Can primary care record review facilitate earlier primary care. BMC Family Practice, 12, 10.
diagnosis of dementia?. Family Practice, 24, 108–16. Cuijpers, P., van Straten, A., and Warmerdam, L. (2007). Behavioral activation
Bogner, H.R., et al. (2007). The effect of memory, attention and executive treatments of depression: A meta-analysis. Clinical Psychology Review,
dysfunction on outcomes of depression in a primary care intervention 27, 318–26.
trial: the PROSPECT study. International Journal of Geriatric Psychiatry, Cuijpers, P., et al. (2009a). Psychotherapy versus the combination of
22, 922–9. psychotherapy and pharmacotherapy in the treatment of depression: a
Borson, S., et al. (2000). The mini-cog: a cognitive ‘vital signs’ measure for meta-analysis. Depression and Anxiety, 26, 279–88.
dementia screening in multi-lingual elderly. International Journal of Cuijpers, P., et al. (2009b). Is psychotherapy for depression equally effective
Geriatric Psychiatry, 15, 1021–7. in younger and older adults? A meta-regression analysis. International
Bower, P. and Gilbody, S. (2005). Stepped care in psychological therapies: Psychogeriatrics, 21, 16–24.
access, effectiveness and efficiency: narrative literature review. British den Heijera, T., et al. (2003). Association between blood pressure levels
Journal of Psychiatry, 186, 11–17. over time and brain atrophy in the elderly. Neurobiology and Aging, 24,
Brodaty, H., et al. (2002). The GPCOG: a new screening test for dementia 307–13.
designed for general practice. Journal of the American Geriatrics Society, Department of Constitutional Affairs (2007). Mental Capacity Act 2005: code
50, 530–4. of practice. <http://www.webarchive.nationalarchives.gov.uk/ + http://dca.
Brodaty, H., Green, A., and Koschera, A. (2003). Meta-analysis of psychosocial gov.uk/legal-policy/mental-capacity/mca-cp-> (accessed 20.01.2012).
interventions for caregivers of people with dementia. Journal of the Department of Health (2001). National Service Framework for Older People.
American Geriatrics Society, 51, 657–64. HMSO, <http://www.dh.gov.uk/prod_consum_dh/groups/dh/groups/
Brooke, P. and Bullock, R. (1999). Validation of a 6-item cognitive impairment dh_digitalassets/@dh/@en/documents/digitalasset/dh_4071283>
test with a view to primary care usage. International Journal of Geriatric (accessed 20.01.2012).
Psychiatry, 14, 936–40. Department of Health (2009). Living well with dementia: a national strategy.
Bruce, D.G., et al. (2002). Communication problems between dementia <http://www.dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/@
carers and general practitioners: effect on access to community support dh/@en/documents/digitalasset/dh_094051> (accessed 20.01.2012).
services. Medical Journal of Australia, 177, 186–8. Department of Health (2010). Standard general medical services-model
Burns, A. and Iliffe, S. (2009). Dementia. British Medical Journal, 338, 405–9. contract and variation document . <http://www.dh.gov.uk/en/
Burns, A., et al. (1990). Cause of death in Alzheimer’s Disease. Age and Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/
Ageing, 19, 341–44. DH_116299> (accessed 20.01.2012).
Burroughs, H., et al. (2006). ‘Justifiable depression’: how health professionals Department of Health (2011). Dementia commissioning pack. <http://www.
and patients view late-life depression; a qualitative study. Family Practice dh.gov.uk/en/Publicationsandstatistics/Publications/Publications
23, 369–77. PolicyAndGuidance/Browsable/DH_127381> (accessed 20.01.2012).
Buschke, H., et al. (1999). Screening for dementia with the Memory Douglas, I. and Smeeth, L. (2008). Exposure to antipsychotics and risk of
Impairment Screen. Neurology, 52, 231–8. stroke: self controlled case series study. British Medical Journal, 337,
Buszewicz, M. and Chew-Graham, C.A. (2011). Improving detection and 1227.
management of anxiety disorders in primary care. British Journal of Downs, M., et al. (2006). Effectiveness of educational interventions in
General Practice, 589, 489–90. improving detection and management of dementia in primary care:
Cameron, I.D., et al. (2011). Assessing and helping carers of older people. a cluster randomised controlled study. British Medical Journal, 332,
British Medical Journal Clinical Review, 343, 1–5. 692–6.
Carlisle, R. (1999). Do nursing home residents use high levels of general Dowrick, C. (2009). Beyond depression, 2nd edition. Oxford University Press,
practice services? British Journal of General Practice, 49, 645–6. Oxford.
Carr, S. (2010). SCIE Report 37: Personalisation, productivity and efficiency. Ellershaw, J., et al. (1997). Developing an integrated pathway for the dying
SCIE, London. patient. European Journal of Palliative Care, 4, 203–7.
Chew-Graham, C.A., Burns, A., and Baldwin, R. (2004). Treating depression Folstein, M.F., Folstein, S.E., and McHugh, P.R. (1975). Mini-Mental State:
in later life: we need to implement the evidence that exists. British Medical a practical method for grading the cognitive state of patients for the
Journal, 329, 181–2. clinician. Journal of Psychiatry, 12, 189–98.
Chew-Graham, C.A., et al. (2007). A randomised controlled trial to test the Folstein, M., et al. (1991). The epidemiology of delirium in the community:
feasibility of a collaborative care model for the management of depression The eastern baltimore mental health. International Psychogeriatrics, 3,
in older people. British Journal of General Practice, 57, 364–70. 169–76.
Gatz, M., et al. (1998). Empirically validated psychological treatments for Milne, A., et al. (2008). Screening for dementia in primary care: a review of
older adults. Journal of Mental Health and Aging, 4, 9–46. the use, efficacy and quality of measures. International Psychogeriatrics,
Gilley, D.W., et al. (1997). Psychotic symptoms and physically aggressive 3, 431–58.
behaviour in Alzheimer’s disease. Journal of the American Geriatrics Mitchell, A.J. and Subramaniam, H. (2005). Prognosis of depression in old
Society, 45, 1074–9. age compared to middle age: a systematic review of comparative studies.
Givens, J., et al. (2006). Older patients’ aversion to antidepressants: a American Journal of Psychiatry, 162, 1588–601.
qualitative study. Journal of General Internal Medicine, 21, 146–51. Mitchell, S.L., Kiely, D., and Hamel, M.B. (2004). Dying with advanced
Gold Standards Framework (2006). Palliative care and the GMS contract. dementia in the nursing home. Archives of Internal Medicine, 164,
Prognostic indicator guidance. <http://www.goldstandardsframework. 321–6.
nhs.uk/content/gp_contract/QOF_Introduction_Paper_1> (accessed Moniz-Cook, E.D., et al. (2008). Can training community mental health nurses
29.08.06). to support family carers reduce behavioural problems in dementia? An
Grasel, E. (2002). When home care ends—changes in the physical health of exploratory pragmatic randomised controlled trial. International Journal
informal caregivers caring for dementia patients: a longitudinal study. of Geriatric Psychiatry, 23, 185–91.
Journal of the American Geriatric Society, 50, 843–9. Mottram, P., Wilson, K., and Strobl, J. (2006). Antidepressants for depressed
Greenwood, M. and Walsh, K. (1995). Supporting carers in their own right. elderly. Cochrane Database of Systematic Reviews, 1, CD003491.
Journal of Dementia Care, 3, 14–16. Moussavi, S., et al. (2007). Depression, chronic disease, and decrements in
Gunn, J., (2006). A systematic review of complex system interventions health: results from the World Health Surveys. Lancet, 370, 851–8.
designed to increase recovery from depression in primary care. BMC National Audit Office (2007). Improving services and support for people with
Health Services Research, 6, 88. dementia. The Stationery Office. London.
Houttekier, D., et al. (2010). Place of death of older persons with dementia. National Collaborating Centre for Mental Health (2007). Dementia: the
A study in five European countries. Journal of American Geriatric Society NICE/SCIE guideline on supporting people with dementia and their carers
58, 751–6. in health and social care. National Clinical Practice Guideline 42. British
Inouye, S., et al. (1990). The confusion assessment method. Annals of Internal Psychological Society and Royal College of Psychiatrists, Leicester and
Medicine, 113, 941–8. London.
Katon, W.J., et al. (2010). Collaborative care for patients with depression and National Collaborating Centre for Mental Health (2010). Depression: the
chronic illness. New England Journal of Medicine, 363, 2611–20. treatment and management of depression in adults (updated edition).
Kidder, S.W. (2003). Psychosis in the elderly—whose delusion is it?. Geriatric National Clinical Practice Guideline 90. British Psychological Society
Times, 4, 25–6. and Royal College of Psychiatrists, Leicester and London.
Kroenke, K., Spitzer, R.L., and Williams, J.B. (2001). The PHQ-9: validity of a National Collaborating Centre for Mental Health (2011). Generalised anxiety
brief depression severity measure. Journal of General Internal Medicine, disorder in adults: management in primary, secondary and community
16, 606–13. care. National Clinical Guideline 113. British Psychological Society and
Lester, H., et al. (2006). The quality outcomes framework of the GMS Royal College of Psychiatrists, Leicester and London.
Contract. British Journal of General Practice, 56, 245–6. National Institute for Health and Clinical Excellence (2009). Depression in
Levine, S., et al. (2005). Physicians’ satisfaction with a collaborative disease adults with a chronic/physical health problem. NICE Clinical Guideline
management program for late-life depression in primary care. General 91. National Collaborating Centre for Mental Health, London.
Hospital Psychiatry, 27, 383–91. National Institute for Health and Clinical Excellence (2010a). Delirium:
Licht-Strunk, E., et al. (2009). Outcome of depression in later life in primary diagnosis, prevention and management. NICE Clinical Guideline 103.
care: longitudinal cohort study with three years’ follow-up. British National Clinical Guideline Centre for Acute and Chronic Conditions,
Medical Journal, 338, 3079. London.
Manthorpe, J. and Iliffe, S. (2010). Suicide in later life: public health and National Institute for Health and Clinical Excellence (2010b). End of life
practitioner perspectives. International Journal of Geriatric Psychiatry, care for people with dementia. <http://www/nice.org.uk/media/8E6/C4/
25, 1230–8. commissionningGuideEoLDementia> (accessed 20.01.2012).
Mason, A., et al. (2007). The effectiveness and cost effectiveness of respite for National Institute for Health and Clinical Excellence (2011). Donepezil,
caregivers of frail older people. Journal of the American Geriatric Society, galantamine, rivastigmine and memantine for the treatment of Alzheimer’s
55, 290–9. disease. NICE Technology Appraisal Guidance 219. <http://www.nice.
Mathers, C. and Leonardi, M. (2000). Global burden of dementia in the year org.uk/nicemedia/live/13419/53619/53619> (accessed 20.01.2012).
2000: summary of methods and sources. <http://www.who.int/entity/ NHS Confederation (2006). Revisions to the GMS contract 2006/7. <http://
healthinfo/statistics/bod_dementia> (accessed 20.01.2012). www.nhsemployers.org/pay-conditions/primary-902.cfm> (accessed
Mathers, C.D. and Loncar, D. (2006). Projections of global mortality and 31.01.2008).
burden of disease from 2002 to 2030. PLoS Medicine, 3(11), e442. NHS Confederation (2011). Quality and Outcomes Framework. <http://www.
Matthews, F. and Dening, T. (2002). Prevalence of dementia in institutional nhsemployers.org/PayAndContracts/GeneralMedicalServicescontract/
care. Lancet, 360, 225–6. gof/Pages/QualityOutcomesFramework.aspx> (accessed 20.01.2012).
McCarthy, M., Addington-Hall, J., and Altmann, D. (1997). The experience NHS Health and Social Care Information Centre (2006). Community care
of dying with dementia: a retrospective study. International Journal of statistics 2005: referral, assessments and packages of care for adults,
Geriatric Psychiatry, 12, 404–9. England. Information Centre, Knowledge for Care, Leeds.
McDougall, F.A., et al. (2007). Prevalance and symptomatology of depression O’Brien, J. (2008). Antipsychotics for people with dementia. British Medical
in older people living in care homes in England and Wales. Age and Journal, 337, 64–5.
Ageing, 36, 562–8. O’Brien, J., Robinson, L., and Fairbairn, A. (2001). Proposed shared
McManus, S., Meltzer, H., and Bughra, T. (2009). Adult psychiatric morbidity care protocol between primary and secondary care for the ongoing
in England, 2007: results of a household survey. NHS Information Centre management of those on anti-dementia medication. Journal of Primary
for Health and Social Care, Leeds. Care Psychiatry, 7, 111–13.
Meeks, T.W., et al. (2011). A tune in ‘a minor’ can ‘b major’: a review Pearson, S., et al. (1999). Depression among high utilizers of medical care.
of epidemiology, illness course, and public health implications of Journal of General Internal Medicine, 14, 461–8.
subthreshold depression in older adults. Journal of Affective Disorders, Pettiti, D., et al. (2005). Blood pressure levels before dementia. Archives of
129, 126–42. Neurology, 62, 112–16.
Pinquart, M. and Sörensen, S. (2001). How effective are psychotherapeutic Targum, S.D. and Abbott, J.L. (1999). Psychosis in the elderly: a spectrum of
and other psychosocial interventions with older adults? A meta-analysis. disorders. Journal of Clinical Psychiatry, 60, 4–10.
Journal of Mental Health and Aging, 7, 207–43. The Information Centre (2007). Health Survey for England 2005: health of
Pinquart, M. and Sorenson, S. (2006). Helping caregivers of persons with older people. Report by R. Craig and J. Mindell (eds). NHS, Leeds.
dementia: which interventions work and how large are their effects?. Thomas, K. (2003). Caring for the dying at home. Companions on a journey.
International Psychogeriatrics, 18, 577–95. Radcliffe Medical Press, Oxford.
Qiu, C., et al. (2004). Decline in blood pressure over time and risk or dementia; Toseland, R., et al. (2002). Predictors of health and human services use by
a longitudinal study from the Kungsholmen Project. Stroke, 35, 1810–15. persons with dementia and their family caregivers. Social Science and
Rait, G., et al. (1999). Screening for depression in older Afro-Caribbeans. Medicine, 55, 1255–66.
Family Practice, 16, 591–5. Treloar, A., Crugel, M., and Adamis, D. (2009). Palliative and end of life care
Reynolds III, C.F., et al. (1999). Treatment of bereavement-related of dementia at home is feasible and rewarding: Results from the ‘Hope
major depressive episodes in later life: a controlled study of acute for Home’ study. Dementia, 8, 335–47.
and continuation treatment with nortriptyline and interpersonal Unützer, J. (2002). Diagnosis and treatment of older adults with depression in
psychotherapy. American Journal of Psychiatry, 156, 202–8. primary care. Biological Psychiatry, 52, 285–92.
Robinson, L., et al. (2006). Wandering in dementia: a systematic literature Unützer, J., et al. (1999). Treating depressed older adults in primary care:
review of the effectiveness of non-pharmacological interventions to prevent narrowing the gap between efficacy and effectiveness. Mill-bank
wandering in dementia and evaluation of the ethical implications and Quarterly, 77, 225–56.
acceptability of their use. Health Technology Assessment, 10(26), iii, ix–10. Unützer, J., et al. (2002). Collaborative care management of late life depression
Robinson, L., et al. (2007). Balancing rights and risks – conflicting perspectives in the primary care setting: a randomized controlled trial. Journal of the
in the management of wandering in dementia. Health, Risk and Society, American Medical Association, 288, 2836–45.
9, 389–486. van der Roest, H.G., et al. (2009). What do community-dwelling people with
Robinson, L., et al. (2010). Patient preferences for future care—how can dementia need? A survey of those who are known to care and welfare
advance care planning become embedded into dementia care: a study services. International Psychogeriatrics, 21, 949–65.
protocol. BMC Geriatrics, 10, 1471–6. Van Weel, C., Van Weel-Baumgarten, E., and Van Rikswijk, E. (2009).
Rost, K., et al. (2000). The role of competing demands in the treatment Treatment of depression in primary care. Incentivised care is no substitute
provided primary care patients with major depression. Archives of Family for professional judgment. British Medical Journal, 338, b934.
Medicine, 9, 150–4. Volicer, L., et al. (1994). Impact of special c1are units for patients with
Sampson, E. and Robinson, A.L. (2009). Editorial: end of life care in dementia: advanced Alzheimer’s disease on patient discomfort and costs. Journal of
building bridges for effective multidisciplinary care. Dementia, 8, 331–4. the American Geriatrics Society, 42, 597–603.
Schneider, J., et al. (1999). Eurocare: a cross-national study of co-resident Wald, C., et al. (2003). What to tell dementia caregivers—the rule of threes.
spouse caters for people with Alzheimer’s disease: i-factors associated with International Journal of Geriatric Psychiatry, 18, 313–17.
carer burden. International Journal of Geriatric Psychiatry, 14, 651–1. White, H., et al. (2002). A randomized controlled trial of the psychosocial
Schneider, L.S., et al. (1997). Validity and reliability of the Alzheimer’s disease impact of providing internet training and access to older adults. Aging
cooperative study—clinical global Iimpression of change. The Alzheimer’s and Mental Health, 6, 213–21.
Disease Cooperative Study. Alzheimer Disease and Associated Disorders, Whooley, M., Stone, B., and Sogikian, K. (2000). Randomized trial of
11, S22–32. case-finding for depression in elderly primary care patients. Journal of
Seitz, D., Purandare, N., and Conn, D. (2010). Prevalence of psychiatric General Internal Medicine, 15, 293–300.
disorders among older adults in long-term care homes. International Wilson, A., et al. (1999). Randomised controlled trial of effectiveness of
Psychogeriatrics, 22, 1025–9. Leicester hospital at home scheme compared with hospital care. British
Shulman, K.I. (2000). Clock-drawing: is it the ideal cognitive screening test?. Medical Journal, 19, 1542–6.
International Journal of Geriatric Psychiatry, 15, 548–61. Wilson, K., et al. (2001). Antidepressant versus placebo for depressed elderly.
Singh, N.A., Clements, K.M., and Fiatarone, M.A. (1997). A randomized Cochrane Library, Issue 2, Update Software, Oxford.
controlled trial of progressive resistance training in depressed elders. Woods, R.T. (1989), Alzheimer’s disease: coping with a living death. Souvenir
Journals of Gerontology. Series A. Biological Sciences and Medical Sciences, Press, London.
52, M27–35. Woods, R.T., et al. (2003). Support in the community for people with
Singleton, N., et al. (2001). Office of National Statistics: psychiatric morbidity dementia and their carers: a comparative outcome study of specialist
among adults living in private households, 2000. HMSO, London. mental health service interventions. International Journal of Geriatric
Smits, C.H.M., et al. (2007). Effects of combined intervention programmes for Psychiatry, 18, 298–307.
people with dementia living at homes and their caregivers: a systematic Xie, J., Brayne, C., and Matthews, F.E. (2008). Survival times in people with
review. International Journal of Geriatric Psychiatry, 22, 1181–93. dementia: analysis from population based cohort study with 14 year
Sörensen, S., et al. (2006). Dementia care: mental health effects, intervention follow-up. British Medical Journal, 336, 258–62.
strategies, and clinical implication. Lancet Neurology, 5, 961–73. Yesavage, J.A., et al. (1983). Development and validation of a geriatric
Spitzer, R.L., et al. (2006). A brief measure for assessing generalized anxiety depression screening scale: a preliminary report. Journal of Psychiatric
disorder: the GAD-7. Archives of International Medicine, 166, 1092–7. Research, 17, 37–49.
Steele, C., Rovner, B., and Chase, G.A. (1990). Psychiatric symptoms and Young, J., Meagher, D., and MacLullich, A. (2011). Cognitive assessment of
nursing home placement of patients with Alzheimer’s disease. American older people. British Medical Journal (Clinical Review), British Medical
Journal of Psychiatry, 147, 1049–51. Journal, 343, 1–7.
Steiner, A., et al. (2001). Therapeutic nursing or unblocking beds? A Zarit, S.H. and Zarit, J.M. (2008). Flexibility and change: the fundamentals
randomised controlled trial of a post-acute intermediate care unit. for families coping with dementia. In: M. Downs and B. Bowers (eds)
British Medical Journal, 322, 453–60. Excellence in dementia care, pp. 85–102. Open University Press, Milton
Stern, Y., et al. (1997). Predicting time to nursing home care and death in Keynes, and McGraw Hill, New York.
individuals with Alzheimer’s disease. Journal of the American Association, Zayas, E.M. and Grossberg, G.T. (1998). The treatment of psychosis in late
277, 806–12. life. Journal of Clinical Psychiatry, 59, 5–12.
CHAPTER 24
Memory assessment services
Sube Banerjee
Of Clinics and Services about where they were ‘the clinic’ but were instead about what they
did (for people with dementia and their carers)—‘the service’.
Memory clinics were first established in the US in the 1970s in spe- This chapter is not about how to run a memory service or what
cialist centres interested in research into Alzheimer’s disease (AD). tests to use. This book is full of information on how best diagnoses
These clinics offered outpatient diagnostic, treatment, and advice can be made and what constitutes best treatment and care. Instead,
services for people concerned about their memory. In addition this chapter is about how this knowledge can become action that is
to general research into AD, Fraser (1990), describing these early of use to all those who would benefit from help: how we can deliver
services, identified four core functions: care to those with dementia that need it.
1. to forestall deterioration in dementia by early diagnosis and So, in this chapter we will consider who memory assessment serv-
treatment ices are for and what they should aim to achieve. We will consider
the evidence base and discuss ‘what good looks like’. But, before
2. to identify and treat disorders other than dementia that might be
considering what we need to do, we need to understand the specific
contributing to the patient’s problems
problem that such services are designed to address. We will there-
3. to evaluate new therapeutic agents in the treatment of dementia fore first refresh the rationale for considering the establishment of
4. to reassure people who are worried that they might be losing such services.
their memory, when no morbid deficits are found.
The thing that set these memory clinics apart from normal neu- The Current System—Most Undiagnosed
rological, neuropsychiatric, and old age psychiatric clinics was their
specific focus on dementia. None was a full-time operation, most
and When Diagnosed, Diagnosed Late
were exactly what they said they were—a clinic, a room in a specific In terms of service provision for dementia, one major issue is that, in
hospital where once or twice a week there was a specific focus. The current systems, less than a half of people with dementia in the UK
term ‘memory’ seems to have been used in order to avoid the nega- have a formal diagnosis made, or contact with specialist services, at
tive connotations of the word ‘dementia’ and to denote a broader any time in their illness (National Audit Office (NAO), 2007). Such
interest in cognition. The first memory clinic in the UK was prob- diagnosis and contact often only occurs late in the illness and in cri-
ably that opened at St Pancras Hospital in London in 1983. The sis, and usually not at all. The UK has no monopoly on poor practice
numbers of such clinics grew, so that it was possible to identify 20 in this area; this is a worldwide phenomenon. The World Alzheimer’s
clinics in 1995 (Wright et al., 1995) and 102 in 2002 (Lindesay et al., Report 2011 (Alzheimer’s Disease International (ADI), 2011) shows
2002), with the authors noting that the orientation of these clinics that most people in the world with dementia have not received a
was moving more from being research bases to ways of providing a formal diagnosis. In high-income countries, between 20 and 50% of
diagnostic service for dementia. dementia cases are recognized. This ‘treatment gap’ is even greater in
The term ‘memory service’ appears to have first been used in low- and middle-income countries, with 90% unidentified in India.
naming the Croydon Memory Service (Banerjee et al., 2007), If these data are extrapolated worldwide, 28 million (78%) of the
when the author, together with David Matthews, the team’s leader, 36 million people with dementia have not received a diagnosis, and
coined the term to describe a new team, working out of a dilapi- therefore do not have access to treatment, care, and organized sup-
dated 1970s psychiatric day hospital in the grounds of the former port that getting a formal diagnosis can provide.
Croydon Workhouse. Our intentions were the same as when using If there is value in diagnosing people with dementia, and extra
the term in drafting the National Dementia Strategy (NDS) for value in diagnosing them early, then this is a system that is ‘broke’
England. There were two aims. First, we wanted to indicate that this and needs fixing. By diagnosing people late or not at all, opportu-
should be a service whose job was to serve people with dementia nities for harm prevention at an individual and a societal level are
and their carers, rather than a clinic organized for the convenience missed. We will consider the potential benefits of diagnosis and early
of service providers to meet their interests and needs (e.g. to recruit diagnosis in more detail below. But if dementia is diagnosed, then
research participants or to be something fun to do on a Tuesday at the very least people with dementia and their family carers have
afternoon). Second, we wanted to stress that such services were not the possibility of planning for their future and availing themselves
of the help, support, and treatments (social and psychological, as 3. Maximizing decision-making autonomy—I need to know to
well as pharmacological) that are available. decide what to do. Early diagnosis and early intervention enables
people with dementia and their families to plan their finances,
Early Diagnosis Is Good, Let’s Stop Talking lives, and care in the future by identifying the condition at a time
when the person with dementia can fully participate in medical,
About it and Just Get on with it legal, and financial decisions and make plans for future care.
It is remarkable, but there are still some who would wish to deny 4. Access to services—Without diagnosis I can’t get any sort of treat-
people with dementia an early and accurate diagnosis. This case is ment for dementia. If you don’t know you have it, you can’t get
often made on the basis of an asserted lack of evidence, rather than the treatments for it. Early diagnosis allows access to treatments,
evidence of disbenefit, and often when that person wishes to spend programmes of care, and services for dementia.
resource on something other than dementia. Surely this is not a dis-
cussion that we need entertain any more? No-one would advocate 5. Risk reduction—Diagnosis allows for the prevention of harm. We
this sort of therapeutic nihilism or medical paternalism in cancer, can act to make the person with dementia, their carers, and the
heart disease, or HIV/AIDS, and nobody should do so in dementia. world safer. Dementia places older adults at risk for relation-
ADI, in their 2011 World Alzheimer Report, reviewed in detail all ship problems, delirium, motor vehicle accidents, medication
the data available and generated a categorization of the rationale for errors, and financial difficulties and exploitation, just for a start.
early diagnosis into nine broad themes (Box 24.1). This is a useful Knowing creates time to address safety issues before accidents or
categorization that is worth reproducing, and each element will be emergencies occur. Again, we are preventing harm.
considered briefly in order to articulate the points made. The state- 6. Planning for the future—I need to know to plan. The early rec-
ments italicized are those of the author. ognition and detection of dementia enables people with demen-
1. Optimizing current medical management—It’s bad medicine tia, their families, and clinicians to plan more effectively for the
not to. The diagnostic process allows for the identification of future, including withdrawing from work if still at work and
potentially treatable or reversible disorders, including infections, ensuring child safety.
depression, medication side effects, and nutritional deficien- 7. Improving clinical outcomes—Everyone, me, carers, society,
cies. The diagnosis allows physicians to adjust treatment plans does better if I know. Early diagnosis enables early intervention;
for other health conditions, including avoiding drugs with anti- early intervention enables improved outcomes for people with
cholinergic effects, and to factor in comprehension and compli- dementia and their family carers. Data suggest that interventions,
ance challenges. Cognitive health can be maximized by control of be they drugs, psychological, social, or educational, are of most
behavioural and psychological symptoms, vascular risk factors, benefit if given early in the disease process (NICE/SCIE, 2007).
nutrition, and social activity.
8. Avoiding or reducing future costs—Early diagnosis delivers better
2. Relief gained from better understanding of symptoms—It’s value, saving money. The driver of long-run costs in dementia is
better to know than not. If we ourselves are to have any chance time spent as a resident of a care home, with acute hospital bed
of managing any problem, we need to know what is going on. days determining medium-run costs. The data reviewed suggest
People with early dementia very often know something is going that interventions aimed at people with dementia and also their
on but do not know what, and that is a powerful source of carers can reduce total care expenditure by preventing acute hos-
strain. Early diagnosis enables people with dementia and their pital admission and delaying the time to care home admission.
families to understand and adapt to cognitive and behavioural Better outcomes and lower cost equate to better value. This is the
changes in dementia, reducing blame and impatience. Thus, fundamental golden equation in health services development, and
harm is reduced by knowledge. Knowing the causes of prob- evaluation and early diagnosis and intervention delivers this.
lems can resolve anxiety felt by people with dementia and their 9. Diagnosis as a human right—Whose diagnosis is it anyway? It is
families. still possible to talk to clinicians, who do not specialize in demen-
tia, who will aver that ‘it is better not to know the diagnosis of
dementia than it is to be told’. What is striking is how very lit-
Box 24.1 World Alzheimer Report 2011: nine reasons for early
tle support there is for such paternalism and unwanted protec-
identification and treatment of dementia
tionism. This is not a shield that people with dementia and their
families either need or want. In many cases, this unintention-
1. Optimizing current medical management
ally toxic position is a reaction in those clinicians of their being
2. Relief gained from better understanding of symptoms uncomfortable in making the diagnosis and communicating it
3. Maximizing decision-making autonomy to their patients. The last 20 years has seen increasing attention
being paid to ensuring that people with dementia are treated as
4. Access to services autonomous individuals with their rights respected. The Fairhill
5. Risk reduction Guidelines (Post and Whitehouse, 1995) stated, ‘because indi-
viduals have a right to control their own lives, and because true
6. Planning for the future
control depends on knowing about oneself, individuals have a
7. Improving clinical outcomes right to full disclosure regarding a dementia diagnosis’. This is a
8. Avoiding or reducing future costs fundamental truth, and the medical profession needs to deal with
it and design systems to enable it. A moment spent putting one-
9. Diagnosis as a human right
self in the position of a person with dementia makes this clear.
CHAPTER 24 memory assessment services 321
Early Diagnosis and Early Intervention in will not be possible with the traditional once a week memory clinic
that sees a couple of new cases per week. We need to have a grander
Dementia—an Emerging Health Policy and broader ambition for our memory services.
Priority As envisaged by the UK NDS, memory services are powerful
Responding to this, early diagnosis and early intervention have local agents and engines for change. They are there to empower
consistently emerged as key policy priorities in the recent series of people with dementia and their carers by providing them with the
National Dementia Strategies. The fourth objective of the French benefits that come from good quality early diagnosis and treatment.
Plan Alzheimer is ‘improving access to diagnosis and care pathways’. They are also there to enable the whole health and social care sys-
In the US, early diagnosis of AD is one of the six main purposes of tem to work well for dementia. Memory services are therefore at the
the US National Alzheimer’s Project Act (NAPA), which was passed heart of a win-win-win scenario, where there is quality and value
with bipartisan support and signed into law by President Obama in improvement for people with dementia and carers, for health serv-
January 2011. The Secretary of State for Health for England, launch- ices, and for social services.
ing the country’s first NDS (Department of Health, 2008), said: More detail is given in the NDS of what a memory service would
be expected to do. The full text of its objective 2 reads:
Current best estimates are that only one-third of people with dementia
ever receive a diagnosis of their illness. We can’t hope to address their Objective 2: Good-quality early diagnosis and intervention for all. All
needs fully, or those of their carers, without a diagnosis being made, people with dementia to have access to a pathway of care that deliv-
appropriate information being given and effective intervention at an ers: a rapid and competent specialist assessment; an accurate diagno-
early stage. Some have argued in the past that it is best not to let people sis sensitively communicated to the person with dementia and their
know. We have long accepted that this should not occur with cancer carers; and treatment, care and support provided as needed following
sufferers. The same should be true for those with dementia. This was diagnosis. The system needs to have the capacity to see all new cases
one of the most consistent messages emerging from the consultation of dementia in the area.
process from people with dementia. This partial operationalization gives us further information on
what such services would be expected to do in order to be consid-
ered of ‘good quality’. This requires:
What Is a Memory Assessment Service?
◆ a rapid assessment
As we have discussed, there is a lack of clarity of definition of what
constitutes a memory assessment service, a memory service, and ◆ a competent specialist assessment
a memory clinic. What they all share is an interest and skill in ◆ an accurate diagnosis to have been made
the diagnosis of dementia and particularly in the early diagnosis
◆ the communication of the diagnosis in a sensitive manner to the
of dementia. An excellent and commendably simple statement of
person with dementia and their carer
what a memory service is for (i.e. its primary aim) is that given in
the NDS for England: ◆ provision of treatment, care, and support needed following the
Good-quality early diagnosis and intervention for all. diagnosis
This requires the service to do three things: ◆ to do this for all incident cases in the area served.
1. to provide a good quality diagnosis early in the illness We start to have a checklist with which we can work out if the
services we have at the moment are up to the challenge, or if we
2. to provide good quality early intervention in dementia need to make the case for improvement by service development.
3. to provide this good quality service to all who might need it in a
given population. Communicating the Diagnosis Well
So a service without the skills to subtype dementia and to have Communicating the diagnosis to people with dementia and their
a high degree of clinical certainty that when they say that a person families in a sensitive way that enables people to take on board this
does not have dementia, they do not have dementia, and when they vital but challenging information is at the heart of what a memory
say that they do, they do, would not meet these criteria. Equally, the service needs to be able to achieve. This process has been termed
service needs to have the skill to identify that small group where ‘breaking the diagnosis well’, following on from ‘making the diag-
there is real clinical uncertainty, even after a full multifaceted, nosis well’ (Banerjee, 2010).
multidisciplinary assessment, and where a longitudinal approach It will always be possible to find specific cases where the person
to diagnosis is needed to be sure. Using these tests, a service that with dementia may not want to know, though in reality such individ-
provides diagnosis but not treatment would also be seen to fail. The uals are unlikely to have cooperated with the diagnostic process. But
issues of what we do to make a diagnosis well and the effectiveness basing practice on such hard cases makes for bad medicine. There is
of intervention in terms of clinical outcomes are covered through- no substitute for a personalized and sensitive approach to the disclo-
out this book and will not be considered further here. sure of the diagnosis, but there is also no avoiding this task.
Finally there is the stipulation that it should work for all in a It is natural that carers should feel protective to family members
population. This is based on equity and access; these should not be with dementia. The finding that 83% of carers from an Irish mem-
services for the lucky few, as many memory clinics have been in the ory clinic did not wish the patient to be told of the diagnosis is
past. They need to have the capacity to deal with all incident cases often quoted. But the most salient fact from this study is that 71%
in the area they serve. An average UK Clinical Commissioning of the same carers wanted to be told the diagnosis if they devel-
Group (CCG) population has 50,000 older people; this means that oped dementia (Maguireet al., 1996). It is also important to note
the service needs to be able to see around 1000 people a year. This that this snapshot was taken two decades ago and public attitudes
and understanding are changing. In a more recent study of 50 peo- be a discussion about the value brought, by different approaches, to
ple with mild dementia, 74% of the carers were willing for patients people with dementia and their carers primarily, and to the system
to be informed of the diagnosis of dementia (Pinner et al., 2003), as a whole secondarily. The equation will vary between countries in
98% of carers wanted to know themselves if they had it, and 92% of terms of what can and should be done in primary care and what is
those with dementia wanted to be told their diagnosis. best done in secondary care. Different systems have different train-
A memory service needs to treat communication of the diagno- ing available and different time packages that can be spent in assess-
sis as a process, not an event. The process is for a period of reflec- ment. What is possible in Canada may not be possible in the UK.
tion and adaptation to follow immediate reactions to the initial In terms of the UK system, the strong and consistent message
diagnosis-giving. Research carried out in this later period suggests that emerged from Department of Health consultation on the NDS
that there are often feelings of regret on the part of carers that they was that the diagnosis of dementia, and in particular mild dementia
had not received the diagnosis earlier so that they could have been where the diagnosis is more complex, should be carried out by a cli-
more patient, understanding, and less blameful of the person with nician with specialist skills. With a disorder as common as dementia
dementia (Connell et al., 2004). Disclosure of the diagnosis is a it is tempting to assume that this should be completed by primary
necessary prerequisite for future care planning including (with sen- care. However, this is in effect the status quo that has delivered the
sitivity) end of life care decisions (Derksen et al., 2006; Lawrence low levels of diagnosis that we have now. A review of the evidence
et al., 2011). It is self-evident that response to the disclosure of a confirms that there is a marked reluctance on the part of primary
diagnosis of dementia will depend on the way that this is done. It is care to be directly involved in the diagnosis of dementia for reasons
an absolute requirement that memory services should do this well that include: the belief that nothing can be done for dementia; risk
from the perspectives of patients and carers. The consultation for avoidance; concerns about competency; and concerns about the
the UK NDS found distressingly frequent instances when doctors, availability of resources (Iliffe et al., 2006). This can be particularly
thinking that they were doing this well, were in fact doing this badly problematic for people from specific groups, such as people with
and potentially doing harm. The following are indicative quotes learning disabilities and younger people with dementia.
taken from the NDS (Department of Health, 2008): Interestingly, the message from the GPs responding to the NDS
I’ve just been told ‘You’ve got Alzheimer’s,’ and they walk out; [it] is consultation was that these are reasonable concerns, and that it
absolutely bloody disgusting. (person with dementia) is not a reasonable expectation of primary care that they should
I got the diagnosis on the phone by somebody I had never met make the diagnosis of dementia, communicate it, and then provide
telling me, ‘Your husband has Alzheimer’s and vascular dementia’. treatment, all by themselves. It was seen as a legitimate role for sec-
That was probably the worst possible way. That was absolutely infu- ondary care to do the heavy lifting in terms of the elements of the
riating. (carer) memory service noted above. They should communicate the diag-
I think they need [to be] diagnosing much earlier and take notice nosis to the primary care team, who can then get on with managing
of it because I think there are lots of channels that could be avoided
the patients with all their problems as before, but enhanced by the
if the first time they saw you, they got down to business and meant it.
Not just fob you off. (person with dementia) power that comes from the accurate diagnostic information and
They didn’t give me enough information. I came away thinking, that having been communicated to the family.
‘What do we do now, where do we go from here?’ I have a prescription This then defines the primary care role as one of identifying those
in one hand and a note for blood tests in the other and nobody has said with worrisome symptoms that might mean that their patient has
what the CAT scan showed . . . nobody has given me that information. dementia, of excluding any other explanatory disorder, and then of
I am the person who is going to deal with [my husband]. (carer) referring on to a specialist service for that individual to receive a
All memory services should develop standard operating proce- definitive diagnosis and management plan. This would require that
dures that are owned by the whole team and that are delivered con- such services were available for the GP to refer to, and in the major-
sistently. Recommendations for good practice include the following ity of the country, this remains questionable.
elements (Lecouturier et al., 2008):
1. Preparation A Separate Service or One Part of General
2. Involving family members Provision?
3. Exploring the patient’s perspective It is legitimate to ask whether the memory service as a function
might be most effectively provided as a special stand-alone service,
4. Disclosing the diagnosis
commissioned in a way that is complementary to existing services,
5. Responding to patient reactions or as an extension of a current team or service. Memory services can
6. Focusing on quality of life be (and are) provided as stand-alone distinct services, such as the
Croydon Memory Service (Banerjee et al., 2007).But the work of a
7. Future planning memory service could also be provided as part of the menu of possi-
8. Effective communication bilities available from ‘traditional services’, where general medical or
psychiatric services (e.g. general practices, old age psychiatric clinics,
geriatric clinics, and neurology clinics) assess people who may have
Role of Primary Care dementia and may make the diagnosis as part of a general service.
There are an important set of questions concerning the role of pri- This boils down to a decision between a specific service (a
mary care in the diagnosis of dementia and, in particular, early stand-alone memory service) or an extension of a general service
dementia. This should not be a conflict about professional protec- (e.g. widening the remit of the neurology service, or a geriatric day
tionism between specialities, or cost containment. Instead, it should unit or a community mental health team for older people to do the
CHAPTER 24 memory assessment services 323
memory service role as well as their existing work). In the English care homes increases over a 10-year period. Total annual savings to
context, taking into account the NDS consultation, the need for society from a 6, 10, and 20% reduction in the numbers of people
accountability, data from the piloting of the Croydon Memory with dementia entering care homes would amount to around £150,
Service (Banerjee et al., 2007), and the DH cost-effectiveness case £245, and £490 million, respectively by year 10 from the nationwide
(Banerjee and Wittenberg, 2009), it appears that separate services introduction of the early diagnosis and intervention service.
would be best. This is reflected in the commissioning guidance In terms of cost-effectiveness, in the 10th year of the service’s oper-
issued by DH in 2011. This stipulated that such services would need ation, its estimated cost would be around £265 million (in 2007/8
to provide a simple single focus for referrals of those with possible prices), taking account of real rises in care costs. If quality of life is fac-
mild to moderate dementia from primary care, and would work tored in, the estimated Net Present Value (NPV) over 10 years would
locally to stimulate understanding of dementia and referrals to the be positive, with a gain of around 6250 Quality Adjusted Life Years
service. It would provide an inclusive service, working for people of (QALYs) in the 10th year, where a QALY is valued at £40,000 or 12,500
all ages and from all ethnic backgrounds. QALYS if a QALY is valued at £20,000. A gain of 12,500 QALYS would
Such services would not replace the work currently completed amount to only around 0.02 QALYs per person year. These relatively
by old age psychiatry, geriatrics, neurology, or primary care, but small improvements seem very likely to be achievable with ease in
would be complementary to their work. The aim is to conduct work view of the rise of 4% achieved in the pilot of the Croydon Memory
not currently associated with any service. Such a service might be Service (Banerjee et al., 2007). This intervention would therefore
provided by any of a number of types of specialist with diagnostic meet stringent accepted definitions of cost-effectiveness (NICE, 2004;
skills in dementia (e.g. old age psychiatrists, geriatricians, neurolo- Rawlins and Culyer, 2004). Two other cost–benefit analyses provide
gists, and GPs with a specialist interest in dementia) or combina- confirmatory evidence of the potential economic benefits associated
tions thereof. Local decisions on what sort of service to commission with earlier diagnosis and intervention. Both analyses are based on
should be based on existing service provision and where local skills screening programmes in primary care, with savings estimated of
and enthusiasm lie. Those referred with needs other than dementia a similar order of magnitude at US $4,000 and $7,700 per person,
would need to be referred on appropriately. The service could be respectively (Weimer et al., 2009; Getsios et al., 2012).
commissioned as a joint health and social care venture, with core
involvement of local third sector organizations. References
Such services are about function not place; they can see people Alzheimer’s Disease International (2011). World Alzheimer Report 2011.
in their own homes, or in primary care settings, rather than being ADI, London.
necessarily hospital-based. The provision of such services locally Banerjee, S. (2010). Living well with dementia: development of the National
can simplify the care pathway for the majority, locating responsibil- Dementia Strategy for England. International Journal of Geriatric
ity and so enabling easy referral, simple communication, and clear
Banerjee, S., and Wittenberg, R. (2009). Clinical and cost effectiveness of
performance monitoring. Where there are existing memory clinics
services for early diagnosis and intervention in dementia. International
that have resources associated with them, they may form the core Journal of Geriatric Psychiatry, 24, 748–54.
of such a new service, and those resources could be used as part Banerjee, S., et al. (2007). Improving the quality of care for mild to moderate
of this service reconfiguration. This is an emerging discipline with dementia: an evaluation of the Croydon Memory Service Model.
different models of service provision; there is a need for fully opera- International Journal of Geriatric Psychiatry, 22, 782–8.
tionalized service definitions and classification and evaluations of Connell, C.M., et al. (2004). Attitudes toward the diagnosis and disclosure
the services provided. of dementia among family caregivers and primary care physicians.
Gerontologist, 44, 500–7.
Department of Health (2008). Living well with dementia: a National Dementia
The Economic Case for Memory Services Strategy. The Stationery Office, London.
We finish addressing the issue of cost. Services must bring value Derksen, E., et al. (2006). Impact of diagnostic disclosure in dementia on
to the people they serve. This value should be measured in terms patients and carers: qualitative case series analysis. Aging and Mental
of the improvements in outcomes that accrue from this service Health, 10, 525–31.
compared to another divided by the cost of the service. If we look Fraser, M. (1992). Memory clinics and memory training. In: Arie, T. (ed.)
Recent advances in psychogeriatrics, 2nd edition, pp. 105–16. Churchill
at memory services, then it seems very likely that they meet this
Livingstone, Edinburgh.
vital test.
Getsios, D., et al. (2012). An economic evaluation of early assessment for
A core aim of the NDS was to ensure that effective services for Alzheimer’s disease in the United Kingdom. Alzheimer’s Dementia, 8,
early diagnosis and intervention are available for all on a nation- 22–30.
wide basis. The Strategy accepted that this would require local Iliffe, S., Wilcock, J., and Haworth, D. (2006). Obstacles to shared care for
investment and stated its assessment, and that of HM Treasury, that patients with dementia: a qualitative study. Family Practice, 23, 353–62.
such a ‘spend to save’ approach, as advocated by the NAO in its Lawrence, V., et al. (2011). Dying well with dementia: a qualitative
value for money report (2007), could both increase the quality of examination of good end of life care for people with dementia. British
care and save hundreds of millions of pounds of expenditure over Journal of Psychiatry, 199, 417–22.
a 10-year period. This is an external validation and endorsement of Lecouturier, J., et al. (2008). Appropriate disclosure of a diagnosis of
dementia: identifying the key behaviours of ‘best practice’. BMC Health
the clinical and cost effectiveness of such services.
Services Research, 8, 95.
Banerjee and Wittenberg (2009) modeled the impact of national Lindesay, J., et al. (2002). The second Leicester survey of memory clinics in the
provision of memory services in preventing admissions to care British Isles. International Journal of Geriatric Psychiatry, 17, 1741–7.
homes. They estimated the overall savings to society, including Maguire, C.P., et al. (1996). Family members’ attitudes toward telling the
savings accruing to public funds and private individuals. The sav- patient with Alzheimer’s disease their diagnosis. British Medical Journal,
ings increased as the numbers of people prevented from entering 313, 529–30.
NAO (2007). Improving services and support for people with dementia. The Post, S.G. and Whitehouse, P.J. (1995). Fairhill guidelines on ethics of the care
Stationery Office, London. of people with Alzheimer’s disease: a clinical summary. Journal of the
NICE (2004). National Institute for Clinical Excellence guide to the methods American Geriatrics Society, 43, 1423–9.
of technology appraisal. National Institute for Health and Clinical Rawlins, M.D. and Culyer, A.J. (2004). National Institute for Clinical
Excellence, London. Excellence and its value judgments. British Medical Journal, 329, 224–7.
NICE/SCIE (2007). Dementia: a clinical guideline. The Stationery Office, Weimer, D.L. and Sager, M.A. (2009). Early identification and treatment of
London. Alzheimer’s disease: social and fiscal outcomes. Alzheimer’s Dementia,
Pinner, G. and Bouman, W.P. (2003). Attitudes of patients with mild dementia 5, 215–26.
and their carers towards disclosure of the diagnosis. International Wright, N. and Lindesay, J. (1995). A survey of memory clinics in the British
Psychogeriatrics, 15, 279–88. Isles. International Journal of Geriatric Psychiatry, 10, 379–85.
CHAPTER 25
Liaison old age psychiatry
John Holmes
Since the last edition of this book, there has been a substantial mix of physical problems can vary day to day on a general medi-
increase in interest in liaison psychiatry services for older people cal or surgical ward, leading to fluctuations in the prevalence of
in the UK. This is not surprising, since most patients in general associated mental health problems, whereas areas dealing with
hospitals, where liaison psychiatry services operate, are aged 65 single conditions (such as coronary care units and stroke wards)
years or older, and there is now a much wider appreciation of the will have more consistent prevalences.
high prevalence and adverse impact of mental health problems in ◆ Case-finding can be difficult. Many people in hospital report
this population. This chapter sets out the case of need for specific depressive symptoms, but when viewed in the context of their
liaison psychiatry services for older people, discusses what services physical illness and levels of distress the use of symptom check-
look like and what they do, and advises on how services can be lists may lead to overdiagnosis of formal depressive episodes,
established successfully. when many people have adjustment disorders that will resolve
Two-thirds of general hospital beds are occupied by people aged spontaneously on discharge or as their physical problem resolves
65 years or older in the UK (Department of Health, 2001; Scottish (Winrow and Holmes, 2005). Delirium by definition waxes and
Office, 2001). Older people are found almost everywhere in the wanes in severity, meaning that studies examining participants
general hospital; care of older people departments by definition frequently over several days will find higher rates of delirium
have older patients in their beds, but general medical wards, gen- than those looking only once. Delirium shares many features
eral surgery, orthopaedics, respiratory wards, cardiology, and many with dementia but has a different aetiology and time-course, and
other specialities also look after older people. Figure 25.1 shows the delirium commonly occurs superimposed on dementia; perhaps
distribution of unplanned admissions of older people to a range this is why some researchers have not differentiated between the
of specialities in a large teaching hospital, and demonstrates that, two and have simply measured cognitive impairment, which can
apart from paediatrics and obstetrics, the care of older people is also be present in depression. Moreover, it is perfectly possible
core business for most medical and surgical specialities. to have depression and dementia together, yet most case-finding
Why should this be of interest to psychiatrists? There are five instruments (e.g. the Geriatric Mental State Schedule (Gurland
reasons: (1) the prevalence of mental health problems in the gen- et al., 1976)) have only a single diagnostic output, usually based
eral hospital setting is higher than in community settings; (2) these on a hierarchical system where organic diagnoses override func-
mental health problems independently predict poor outcomes; tional ones.
(3) the management of mental health problems in older people in
this setting is often suboptimal; (4) referrals from general hospitals ◆ Recruitment. Studies excluding the 30% of patients in general
comprise at least a quarter of referrals received by old age psychia- medical wards who lack the capacity to consent to participation
try services; and (5) the redesign of mental health input into general (Raymont et al., 2004) will fail to recognize the dementia and
hospitals can help to meet the challenge of optimizing management delirium that has produced this incapacity. The timing of recruit-
and improving clinical and organizational outcomes (Royal College ment is also important, since length of hospital stay is positively
of Psychiatrists, 2005). skewed, meaning that recruitment after 1 week in hospital will
miss large numbers of potential participants who have already
been discharged or died.
What Is the Scale of the Problem? ◆ Sampling methods. People with specific conditions, such as hip
Attempting to determine the prevalence of mental health problems fracture, present at a rate manageable by researchers, meaning
in general hospital settings can be difficult due to a wide range of that recruitment to a prevalence study can be comprehensive.
problems: In contrast, a busy medical assessment unit can turn over its
◆ The hospital setting. The prevalence of depression, dementia, and entire population in a couple of days, requiring a large number of
delirium is different in different areas of the hospital. Older peo- researchers (bringing issues of inter-rater reliability) or a suitable
ple on orthopaedic wards have much higher levels of delirium method of randomly sampling for potential participants.
than in the general medical setting because of the higher levels of Despite these problems, a systematic review (Royal College of
risk factors for delirium in the older trauma population. The case Psychiatrists, 2005) has revealed 97 studies that met predefined
8000
7000
6000
5000
4000
3000
2000
1000
ics
ne
T
ly
ry
y
y
y
y
og
og
og
r
og
g
EN
r
ge
to
lo
ici
ed
de
ol
ol
ol
ol
ro
ira
r
ed
pa
El
Su
di
at
lm
te
U
sp
ar
ho
m
e
ha
al
en
th
Re
eu
al
rt
er
pt
o
er
of
Rh
O
en
tr
O
en
as
e
G
ar
G
C
Fig. 25.1 The distribution of emergency admissions of older people within a general hospital over a 1-year period.
(Data from Leeds Teaching Hospitals NHS Trust.)
Table 25.1 The prevalence of mental health problems in older people in general hospitals
Diagnosis No. of studies Total no. of participants Mean sample size Prevalence range (%) Mean prevalence (%)
Depression 47 14,632 311 5–58 29
Delirium 31 9601 309 7–61 20
Dementia 17 3845 226 5–45 31
Cognitive impairment 33 13,882 421 7–88 22
Anxiety 3 1346 449 1–34 8
Schizophrenia 4 1878 376 1–8 1.4
Alcohol misuse 4 1314 329 1–5 3
(From Royal College of Psychiatrists, 2005.)
quality criteria (out of a total of 576 studies). These 97 studies hospital is providing inpatient care to about six times as many older
reveal a large amount of evidence for higher levels of depression, people with mental health problems than the local mental health
dementia, and delirium in a variety of hospital settings than found services.
in the community. Alcohol misuse and anxiety are also present, as
is schizophrenia, though at rates no higher than in community sur- The Impact on Outcomes
veys. Further details are found in Table 25.1. Much of the stroke lit-
erature is not exclusive to older people, but there are high levels of Having established that mental health problems are common in
delirium, dementia, and depression reported in people after stroke older people in general hospitals, we now turn to the effect on
(Ferro et al., 2002; Merino andand Hachinski, 2002; Turner-Stokes outcomes. This leads to the question of which outcomes and for
and Hassan, 2002). We are also beginning to appreciate that mental whom. Individual patients may be interested in their survival, inde-
health problems exist in other conditions previously unresearched, pendence, and quality of life. Carers want to know about the level
such as chronic obstructive pulmonary disease (Yohannes et al., of carer burden and strain. Clinicians will share the aims of their
2000). The low levels of medically unexplained symptoms (Wijeratne patients but will also be interested in quality of care. Health service
et al., 2003) and self-harm in older people, who comprise only 2.7% commissioners and managers, whilst aware of the importance of
of the self-harm population presenting to accident and emergency the above, are also concerned about financial and organizational
departments (Horrocks et al., 2003), serve to highlight the different costs. Furthermore, this is not just about healthcare: managers and
needs of the populations served by liaison mental health services for commissioners in adult social care are interested in institutionaliza-
older people and for adults of working age (where the focus is on tion that they may end up funding. However, as with determining
medically unexplained symptoms and self harm). prevalence, carrying out research in this area is not easy. Outcome
These prevalence rates mean that some 60% of older people in a studies in this setting must first have all the features of a good
typical general hospital have a mental health problem, either asso- prevalence study, detailed in the section What Is the Scale of the
ciated with a physical problem or sometimes as the sole reason for Problem? Then they need to address the following:
presentation. This represents a third of the total number of occu- ◆ Timing of recruitment. Because of the changing nature of expo-
pied beds in that hospital, meaning that at any one time the general sure to different risks at different times of a hospital admission,
CHAPTER 25 liaison old age psychiatry 327
and the skewed length of stay, an inception cohort study with a may itself contribute to adverse outcomes. There are several pos-
common time of recruitment early in the admission is required. sible explanations:
This, however, brings ethical issues related to allowing enough ◆ Mental health problems are poorly detected. General ward staff are
time for informed consent to be sought. Additionally, the severity not tuned in to mental health problems, and do not routinely
of physical illness may preclude entry into a research study at a screen for them in the same way that they screen for physical
common inception point. problems, e.g. by taking temperatures, pulse rate, and blood pres-
◆ Confounding variables. There are many variables that can affect sure. Where mental health problems are identified in older peo-
outcome in the general hospital population. These include age, ple, it is usually through their behaviour—or, more accurately,
gender, severity and number of physical illnesses, preadmission perceived misbehaviour such as aggression, wandering, interfer-
abilities of activities of daily living, social support, being cared ing with other patients, and refusing medication (Atkin et al.,
for by a specialist team, what type of anaesthetic is used dur- 2005). In the hip fracture population, where the delirious are in
ing surgery, and many others. Add to these dementia, delirium, the majority, clinicians only detect half of the cases of delirium
and depression, and indeed delirium superimposed on demen- found by researchers (Gustafson et al., 1991), and in the broader
tia which may have a particularly bad prognosis (Andrew et al., medical population detection rates are only between 32 and 67%
2006), and we have a complex dataset to acquire and analyse. (Inouye, 1994). The same is true of depression, with a systematic
◆ Sample size and analysis. Whereas studies of prevalence can be review reporting a median detection rate of only 10% (Cole and
carried out with relatively small numbers, since the three main Bellavance, 1997). One study describing older people referred to
conditions being examined are common, the large number of a liaison psychiatry service for assessment of mood found that
confounding variables for outcomes means that a much larger delirium was the cause in 40% (Farrell and Ganzini, 1995).
sample size is required. For example, in order to determine the ◆ Physical care needs are prioritized over mental health needs. The
impact of psychiatric illness in an orthopaedic population with focus is on physical interventions for physical problems, with
90% power and a P value of 0.05, a sample size of 660 is required mental health problems not seen as life-threatening. Even when
(Holmes and House, 2000a). The correct statistical methods are staff have the time to talk to a confused patient, they are uncom-
also important; for endpoints such as discharge, death, readmis- fortable doing so and would rather tidy up the ward (Atkin et al.,
sion, and institutionalization, survival analysis is usually the 2005). In some cases, mental health needs are ignored completely
most appropriate technique, with identical follow-up periods for in the expectation that they will go away spontaneously (Holmes
all participants to ensure equal exposure to risks and adequate et al., 2002).
accounting for those participants who have died as they are no ◆ General hospital staff lack the knowledge and skills to manage men-
longer at risk of other outcomes. tal health problems. This leads to treatment rates for depression
With this in mind, a further systematic review of outcomes in as low as 25% (Holmes and House, 2000a), to an overreliance on
this population has been carried out (Royal College of Psychiatrists, psychotropic medication in delirium rather than recommended
2005). This found 27 studies meeting predefined quality criteria. methods such as de-escalation and environmental manipula-
Findings were mixed, with many studies reporting no impact on tion (Holmes et al. 2003b; NICE, 2010), and to wide variations
outcomes such as mortality, length of stay, and independence. in the choice of medication, dose, and route of administration,
Heterogeneity of case-determining instruments, outcomes meas- with some doctors prescribing doses that are well above British
ured, duration of follow-up, and statistical techniques means that National Formulary maximums and that could be described as
a meta-analysis of these studies was not feasible. It is noteworthy toxic rather than therapeutic (Hally and Cooney, 2005). These
that only three studies had sample sizes larger than the 660 sug- are all staff who have been through professional training sup-
gested by the above power calculation, meaning that the likelihood posed to equip them for their future careers, yet examination of
of a type II statistical error is high, i.e. that an important effect may the content of many of these training programmes reveals a pau-
have not reached the statistical significance it may have done with city of mental health experience and training, with what training
a larger sample. This is borne out by the fact that the larger studies there is often being questionable—one focus group participant
in this review show robust adverse effects for depression, delirium, revealed that his mental health experience consisted of 4 weeks
dementia, and unspecified cognitive impairment on mortality, in a day hospital where he played bingo, did exercises, and threw
length of hospital stay, institutionalization, physical dependence, a ball at each other (Atkin et al., 2005), hardly helpful when hav-
and general health status. This means that as well as being coming to manage complex problems in a general hospital setting.
mon, mental health problems in older people in general hospitals This means that the skill-mix on general wards does not address
are independent predictors of poor outcomes that are of interest mental health needs adequately (Norquist et al., 1995).
to patients, carers, clinicians, commissioners, and managers across ◆ Mental health services are seen as slow to respond. Referrals are
health and social care. not made, as they will slow down the discharge process at a time
when there is marked pressure on hospital beds and hospital
lengths of stay are reducing (Holmes et al., 2002). Only 5% of
Why Are Outcomes So Bad? old age psychiatrists surveyed in 2002 felt that their service was
There are several possible factors contributing to poor outcomes. generally able to respond to general hospital referrals within one
The index mental health problem itself may bring worse outcomes, working day, and 60% felt that a response time of five or more
no matter how well it is managed. However, there is evidence to days was a realistic target (Holmes et al., 2003a). A more recent
suggest that management of mental health problems in older peo- survey showed some improvement in response times as services
ple in the general hospital setting is far from optimal, and that this have changed (Holmes et al., 2010).
◆ There are low referral rates from particular specialities. Consultant healthcare needs that would be best met in an environment where
old age psychiatrists considered that there were several medi- they were working side by side with mental health nurses on the
cal and surgical specialities that had inappropriately low refer- same ward, and noted that such a resource would be invaluable for
ral rates, including orthopaedics, general surgery, and neurology training (Atkin et al., 2005). This means that part of the answer to
(Holmes et al., 2003a), the former a particular worry given the the problem of mental health problems in general hospitals is likely
extremely high prevalence and adverse impact of mental health to lie in old age psychiatry service configuration and activity rather
problems in the hip fracture population (Holmes and House, than antidepressants or antipsychotics alone.
2000a, 2000b). It is now time to examine the kind of input that old age psy-
Despite these problems, general nursing staff and other colleagues chiatry services provide to general hospitals, to see if it meets the
often recognize that they are not adequately trained to manage requirements of our nursing colleagues. Before we do this, we need
older people with mental health problems; they feel that they are to consider the types of service model that could possibly operate,
doing their patients a disservice, and that similar mismanagement together with the pros and cons of each model. We also need to
of physical problems would not be tolerated (Atkin et al., 2005). understand the meaning of the two terms, consultation and liaison,
Overarching all of this is the organizational and managerial in this context.
structure of the National Health Service, particularly in England,
where providers of acute hospital care and mental healthcare sit in
different organizations with different organizational objectives and
Consultation or Liaison?
what seems like a silo mentality. If the attitudes of general hospital As a trainee psychiatrist I found that any referral originating in the
staff are reflected by their own managers, the lack of ownership of general hospital setting was called a liaison referral, and colleagues
older people with mental health problems in the general hospital would talk about liaison activity when talking about their responses
setting could be a further reason why outcomes for this group are to these referrals. The response entailed a visit to the referring ward,
so poor. This is reflected at a higher level too; at the time of writing, scrutiny of the notes, discussion with whichever clinicians were
a search of the English Department of Health website reveals only available, assessment of the patient, and the writing of a manage-
one hit for delirium, one of the commonest conditions found in ment plan in the medical notes. Sometimes further discussion with
older people in general hospitals. ward staff would take place. However, it was not unusual to find few
It seems therefore that there are deficiencies at several stages con- clinicians to speak to, particularly as many assessments were carried
tributing to poor outcomes, including at the level of the individual out late in the day on the way home, meaning that little true liaison
practitioner and at an organizational level too. We will next con- took place. It was also often difficult to clarify exactly what question
sider how these deficiencies can be addressed. the referrer had in mind, and to obtain further relevant information
about patients and their circumstances. In addition, recommenda-
tions made in medical notes were not often followed, something
Improving Outcomes that was not often appreciated as few patients were followed up on
There have been several intervention studies in physically ill older the ward. A more correct term for this type of work is consulta-
people with mental health problems, but few of high quality. For tion, a model that relies on general hospital staff to detect mental
example, a Cochrane review has revealed only one randomized con- health problems, refer on a case-by-case basis, and follow the advice
trolled trial for the treatment of depression in this population (Gill offered, and we have already highlighted deficits in these areas.
and Hatcher, 2006). Examination of the multicomponent inter- What are the alternatives? One possibility is a true liaison
vention studies to reduce the incidence and severity of delirium approach, where the service is more proactive, working collabora-
reveals that most of the interventions could simply be described tively with general hospital colleagues to train and educate them so
as good nursing and medical care (NICE, 2010), e.g. ensuring that that they are confident in the basics of the management of the com-
patients are hydrated, nourished, and have optimal sensory input. mon mental health problems they come across, and so they know
However, if the reasons for poor outcomes explored in the section who, when, and where to refer. The response to referrals is more
Why Are Outcomes So Bad? are to be believed, there needs to be prompt, and a frequent presence on general wards allows for the
a mechanism for addressing deficits in knowledge, skills, and atti- delivery of mental healthcare (including brief psychological thera-
tudes of a wide range of healthcare professionals, as well as ena- pies) routinely, including opportunities for modelling good care to
bling the routine delivery of mental health assessment and care by general staff. This has much more potential to change practice, to
mental health professionals in the general hospital setting. This is improve outcomes, and in particular to challenge the negative atti-
echoed by nursing colleagues, who want help to be able to identify tudes to older people with mental health problems that seem to be
and manage less complex mental health problems. For dementia, at the heart of the problem. The liaison approach requires a consul-
this issue is highlighted by the English National Dementia Strategy tation service operating in the background, able to respond to refer-
(Department of Health, 2009), where workforce training is one of rals received, so these services are known as consultation-liaison
the key workstreams. services in some parts of the world. Table 25.2 shows the pros and
General nurses also feel that better signposting to old age psy- cons of consultation and liaison.
chiatry services would help them, with clear referral routes, a more Liaison psychiatry services already exist in the UK for adults of
rapid response, direct access to a mental health professional for tel- working age, and have undergone some expansion since a 2003
ephone advice on management, any member of the ward team able survey revealed 93 funded consultant posts (Swift and Guthrie,
to refer for an old age psychiatry assessment, and more ward-based 2003). There is a specialist Faculty of Liaison Psychiatry of the
follow-up and review by mental health staff. General nurses also Royal College of Psychiatrists and training with specialist accredi-
identified a group of patients with complex physical and mental tation. They have developed despite the lack of a robust evidence
Table 25.2 The differences between consultation and liaison for the of confusion to general hospital colleagues. This model also assumes
general hospital setting that the psychiatrist is in the best position to assess and offer advice,
although there are occasions when input from other mental health
Consultation Liaison
professionals may be much more appropriate. Opportunities for
Reactive Proactive teaching and training are limited with this model.
Low cost Higher cost
The enhanced sector model
Professionally isolated Collaboration with other professions
Where this model operates, a community mental health team
General hospital referrals a low General hospital referrals a high priority receives additional staffing (usually nursing) ring-fenced to provide
priority input to the general hospital. This creates more opportunities for
Slow response to referrals Rapid response to referrals nonmedical assessments and more reviews, and continuity of care
Low review rates Frequent review rates is good. However, most limitations of the standard sector model
apply, and staff time intended for general hospital work may be
Some poor quality referrals Fewer poor referrals eroded by pressures of work in the community.
Poor adherence to Improved adherence to
recommendations recommendations Outreach from mental health wards
No influence on practice of Influences practice of general hospital With the outreach model, staff from mental health wards provide
general hospital staff staff input to general hospital wards. This is usually on a consultation
Mental health managerially Mental health managerially integrated basis, although often staff will review a patient who has been trans-
separate from general services with general services ferred from the psychiatric ward to the general hospital ward, or
who is about to be transferred the other way. There is the poten-
(Adapted from Royal College of Psychiatrists, 2005.)
tial for training and education with this model, but it does depend
on psychiatric ward staff not being needed on their own ward,
and psychiatric ward staff are as busy as their general hospital col-
base for their effectiveness (Ruddy and House, 2005) and the bulk leagues. For this model to work at all, the psychiatric ward needs to
of their workload relates to self-harm and medically unexplained be on the general hospital site and this is not necessarily the case in
symptoms. Some liaison psychiatry services for adults of working many places. The response to a referral may be slow when an urgent
age see older people, often solely after self-harm or in emergen- assessment is necessary.
cies, but this is far from universal (Ruddy and House, 2003). Ten
years ago, specialist liaison services for older people were few and The liaison mental health nurse
far between; more recently they have increased in number, size, and
Here, a specialist mental health nurse is based in the general hospi-
activity (Holmes et al., 2010).
tal and provides a responsive liaison mental health service to gen-
eral hospital wards. Referrals are seen quickly, and more patients
Models of Liaison are reviewed to monitor their mental state and check that advice
Many old age psychiatry services still offer the consultation model is being followed. Some patients may subsequently need a psychi-
to general hospitals, although many old age psychiatrists would atric review, with the nurse acting at least in part as a triage point.
prefer to offer a liaison model (Holmes et al., 2010). Possible mod- This means that there should be access to a psychiatrist as part of
els providing general hospital input are as follows. this model. A particular benefit of this model is the possibility of
offering advice on the nonpharmacological management of diffi-
The standard sector model cult behaviour. A liaison nurse has time for teaching and training,
This is currently the prevalent service model, providing comprehen- although the hierarchical nature of general hospital professionals
sive mental health services for a population of older people defined may mean that some professional groups may be difficult for the
by either geography or general practitioner. General hospital input is nurse to access. However, through good training programmes,
on a consultation basis and referrals are usually seen by medical staff, protocols for screening and treating can be developed with general
although other members of the community mental health team may hospital staff as true partners. Interfaces with the other old age psy-
visit people already on their caseload. In order to access this serv- chiatry services (and, in particular, the community mental health
ice, referrers need to know which sector service to refer to and how teams) need to be clarified and agreed so that continuity of care can
to make contact. Old age psychiatry services are increasingly based be delivered, albeit through more than one healthcare professional.
away from general hospital settings, so there may be inefficiencies in Many liaison nurses work in isolation and this leads to a high work-
travelling and parking built into this model. However, there is good load and the possibility of burnout.
communication about clinical cases within the community team,
and continuity of care is perceived to be better. Response to gen- The liaison psychiatrist
eral hospital referrals can be slow, particularly as referrals of people Old age liaison psychiatrists have dedicated time for general hos-
in the community are seen as more at risk and so are prioritized pital work. Their activity is similar to that of liaison mental health
(Holmes et al., 2002). Review of patients does not often happen. nurses, with a rapid response to referrals and an emphasis on teach-
Because several different psychiatrists from different sectors and of ing and training. A medical background brings an understanding
different grades can potentially respond to referrals, opinions and of the complexity of some medical problems, but there may be less
advice offered may not be consistent, providing a potential source expertise in the area of behavioural management. As with the liaison
nurse, interfaces and communication with other parts of the serv- skills of other team members as appropriate. Teaching and training
ice are important, and increasing workloads can lead to burnout. of general hospital staff are core business for a hospital mental health
Both the liaison nurse and liaison psychiatrist operate in a unidisci- team, resulting in the ward staff taking ownership and responsibility
plinary way, unlike mental health services in most settings. for mental health problems on that ward and providing the basics
of management, with specialist referral to the mental health team as
The shared care ward required. There is a single point of access and referrals are responded
In this model, a ward on the general hospital site has psychiatric to promptly (often the same day). There is also the possibility of
and general nurses, psychiatrists, physicians, and therapy staff work- introducing staff support in areas where stress is high, such as in
ing together delivering care to patients who have both physical and intensive care. As with other liaison models, good communication is
mental healthcare needs who would otherwise fall between services. important. The hospital mental health team is the model of care rec-
It is an add-on to other services, having a small bed base (12–16 ommended for old age psychiatry services in England (Department
beds are adequate) but able to cope with complex care needs pre- of Health and Care Services Improvement Partnership, 2005).
sented by, for example, someone with an agitated delirium, and also
able to provide a haven for those detained under the Mental Health Other teams and disciplines
Act who can then receive the psychiatric care that they are detained In some places, teams exist to provide specialist input for specific
to the general hospital for. The shared care ward can act as a train- psychiatric illnesses in older people. Examples include accelerated
ing resource for many staff, who are able to learn new knowledge discharge teams for people with dementia. These teams are usually
and skills from colleagues of other disciplines whilst working side organized in a similar way to a hospital mental health team but are
by side with them. Existing examples of shared care wards are var- not designed to offer a comprehensive service. Clinical psycholo-
ied, but those that are successful have clear admission and discharge gists also work in general hospitals but do not often work along
criteria and are explicitly not used to accommodate patients waiting with other mental health professions (Holmes et al., 2002).
for placement. Systematic evaluations of effectiveness are, however,
lacking, due in part to the complexity of the evaluation and the low
numbers of patients an individual ward will admit over time. The UK Picture
In the UK, the predominant model for providing old age psychiatry
The hospital mental health team input to general hospitals has moved from the traditional sector
In this, the most complete model of true liaison, a multidisciplinary model, used by 73% of old age psychiatrists surveyed, to a range of
team with a similar professional mix to a community mental health liaison models (Holmes et al., 2003a, 2010). Figure 25.2 shows how
team (psychiatrists, mental health nurses, clinical psychologists, service provision has changed in a relatively short space of time.
social workers, occupational therapists, etc.) works with the general This description of the change in service provision does not
hospital population as its sector, with a large and transient popula- reflect the wide differences in staffing levels, activity, manage-
tion. Individual team members can build up affiliations with par- ment structures, administrative support, handling of referrals, and
ticular parts of the general hospital, and can call upon the specialist recording of clinical and process-related information that we found
Service models 2002 and 2006 2002

2006
130
120
110
100
Number of hospitals
90
80
70
60
50
40
30
20
10
0
ea ic
r
e
r
w MH
am l
er
Te nta
N tric
to
to
ar
l T atr
th
ds
se
m
c
ec
C
lth e
a
se
O
h
ica hi
ea l M
ur
ar
hi
-s
ed
ac
ed yc
al-
yc
d
re
H pita
ar
M n Ps
ce
on
Ps
ut
Sh
an
iti
os
on
O
o
h
ad
ais
H
En
ais
Tr
Li
Li
Service model
Fig. 25.2 The change in service provision between 2002 and 2006.
(From Holmes et al., 2010.)
Table 25.3 The range in activity for nine different liaison mental health services for older people
TSM 1 TSM 2 TSM 3 LPN 1 LPN 2 LPN 3 HMHT 1 HMHT 2 HMHT 3
Referrals/year 90 62 67 360 375 300 264 859 1450
Referrals/admission (%) 0.5 0.3 1.2 0.4 2.7 1.9 0.9 3.6 1.6
Time between referral and receipt 0 3.2 2.2 0.5 0.2 0.8 0.9 0.8 0.2
(mean number of working days)
Time between receipt and 4 7 6 3 4 2 2 6 1
assessment (median number of
working days)
Number of reviews 1 0 1 2.2 1.8 1.2 2.1 1.7 1.5
TSM, Traditional Sector Model; LPN, Liaison Psychiatry Nurse; HMHT, Hospital Mental Health Team.
(From Holmes et al., 2010.)
during a more indepth study (Holmes et al., 2010). It is evident that liaison resource; other places have in place or may prefer a sepa-
the term liaison, as mentioned earlier, means different things to dif- rate team for older people, since different knowledge and skills are
ferent people. The broad range of service models in existence have required for them.
evolved through local interest rather than with a strong national Until recently, liaison psychiatry services for older people have
direction, and a one-size fits all solution will not work everywhere, existed in a policy vacuum. Where they did exist, it was been due
as there are many local factors that influence what services look to opportunistic developments by individual enthusiasts rather
like. Despite this, examination of Table 25.3 reveals unacceptable than because of policy directives. The lack of joint working by
variation in the activities of different services. Particularly striking providers of physical care and psychiatric care and their com-
are the range of referrals each year (a 23-fold difference), the dif- missioners has not helped. Neither does the fact that liaison psy-
ference in the percentage of all admissions of older people referred chiatry work is not costed or counted at an organizational level,
(nine-fold), and the range in median response times from 1–7 days. with no changes in income streams for increases or decreases in
The variation in service provision and activity suggests a need for activity. It is also unclear whose responsibility it is to fund liai-
service benchmarks that do not currently exist. son psychiatry—is it the general hospital or is it the commission-
So, what evidence is there that introducing a liaison service can ers? All parties, including the mental health provider services,
help? Unfortunately there is little high quality evidence to show us seem to perceive this crucial activity as a potential diversion of
that the introduction of a liaison service can make a difference to their scarce resources. There is no direct link to NHS perform-
outcomes. The evaluation of the efficacy and effectiveness of a liai- ance indicators that healthcare providers are measured by, and
son service is best carried out in a large, multicentre, randomized specifically no link at all to the performance indicators of mental
controlled trial. The unit of randomization should be at the level of health providers. This is particularly important since the organi-
the hospital, to avoid the control group receiving care from a group zation that benefits most from a liaison service is not the mental
of staff already educated by the liaison team. Such an evaluation is health provider (trust or equivalent organization) that provides
complex and not as scientific as a straightforward randomized con- the service, but the general hospital provider trust in which the
trolled trial of a drug. It is also expensive and perceived by research service sits. This is a difficult message to get across to some man-
funders to be high risk. A systematic review of liaison service eval- agerial colleagues, and only a whole-systems view can resolve the
uations (Holmes et al., 2010) revealed that the best evidence that problem.
there is comes from the US, where a controlled study of the impact One approach that will help in England and elsewhere is the
of a liaison service suggests that the introduction of a liaison psy- recent development of national dementia strategies. These high-
chiatry screen–treat intervention reduced the overall median length light people with dementia as a significant issue for general hos-
of stay of a hip fracture population by 2 days and produced a cost pitals and also underline the importance of workforce training.
benefit (Strain et al., 1991). There are doubts over the generalizabil- However, there are two cautions: dementia is only the second most
ity of one single study in an orthopaedic setting, perhaps explaining common mental health problem in old age, with depression being
the slow uptake of liaison services in the UK. More recent evidence more common; and a focus on dementia may lead to the continued
from the UK, using a before and after methodology, suggests that neglect of delirium.
the introduction of a liaison service produced reductions in length
of stay in older people that produced substantial cost savings. This
service, called Rapid Assessment Interface and Discharge (RAID, What an Old Age Liaison Service Should Do
see <http://www.bsmhft.nhs.uk/raid/>), delivers an in-reach serv- There are several domains to a liaison service’s activity. These are
ice so that all patients over the age of 16 can be assessed and treated mentioned in detail elsewhere (Royal College of Psychiatrists,
or referred appropriately much earlier. Although for adults of all 2005), but essentially consist of the following:
ages, it is noteworthy that older people represent only about a third
of the studied patient samples but account for around 90% of total ◆ Clinical activity
benefits in terms of reduced bed use (Parsonage and Fossey, 2011). • The prompt assessment, diagnosis, and management of referred
This all-age service was developed in a hospital with little existing older people, including all those who have harmed themselves
• Risk assessment and risk moderation The Liaison Curriculum

• Incorporation of advice and treatments into care plans If teaching general hospital staff is a key component of liaison psy-
• Regular review to monitor response to treatment and adher- chiatry activity, then what is to be taught? There are several impor-
ence to advice tant areas to cover:
• Engagement with carers and relatives ◆ The general approach. Those with psychiatric conditions are peo-
• Arranging suitable mental health aftercare (including transfer ple too, and should be treated as such. A holistic, person-centred
to a mental healthcare setting and signposting to community approach should be the cornerstone of the liaison curriculum,
services where appropriate) with dignity and respect promoted to all staff. Talking to people
with a mental health problem is a particular skill that will need to
◆ Educational and promotional activity be developed.
• Provide educational programmes to improve detection and ◆ Specific conditions. Dementia, delirium, and depression are the
management of common psychiatric disorders in the general commonest mental health problems found in general hospitals
hospital setting and all qualified staff should be competent in their management
• Develop treatment protocols and care pathways in conjunction as applied to their own practice. Staff should understand that, for
with general hospital colleagues to improve ownership and example, someone who is depressed lacks motivation and energy
uptake due to their depression and is not simply being lazy. Alcohol mis-
use, anxiety, and substance abuse are also important. All mental
• Develop training posts within the service for a range of
health problems, such as schizophrenia, can be found in general
disciplines
hospitals and staff should be aware of the basics of management.
• Raise awareness of the importance of mental health and chal-
◆ Problem behaviours. The most obvious is the challenging behav-
lenge and reduce stigma
iour associated with agitated dementia and delirium, and an
• Advocacy for vulnerable groups introduction to de-escalation techniques will be necessary. Staff
◆ Operational activity need to understand that someone with dementia who wanders
round the ward is likely to be looking for something purposefully
• Develop operational policies and clinical governance structures rather than deliberately misbehaving. Motivational techniques
• Establish clear lines of management for all professionals can be helpful in some patients, and staff (particularly rehabilita-
• Use clear signposting for referrers, including a single point of tion staff ) will benefit from knowing about these.
access ◆ Psychotropic medications. Information about the effects and side
• Record clinical and service related data for audit purposes effects of common medication is useful, together with explana-
tion of the basic modes of action and speed of onset of effective-
• Work collaboratively with general hospital colleagues to develop ness. Mood stabilizers and depot antipsychotics may be stopped
shared objectives and outcomes. in hospital, and staff need to know that they should be restarted
In order to deliver the above, liaison practitioners need a certain or a liaison psychiatry opinion sought.
set of knowledge, skills, and attitudes. Most important are the clini- ◆ Legal issues. There is increasing awareness of issues of capacity
cal knowledge skills required to assess complex cases in what can and consent in the general hospital setting, from consent for
feel like an alien and sometimes hostile setting. These skills may be major surgical procedures through to discharge planning. All
difficult to obtain, and many liaison practitioners have had to learn healthcare professionals should be aware that consent is an issue
their clinical skills on the job, there being no other route of acquisi- for their daily practice, and be conversant with the law relating to
tion. Excellent written and oral communication skills are necessary, capacity and consent in their particular legislature. General hos-
both to carry out and communicate the findings of clinical assess- pital staff are particularly confused by mental health legislation,
ments and management plans and to deliver teaching and train- and this should be a component of any liaison curriculum.
ing that is accessible to all. Excellent liaison practitioners will be
able to develop a relevant curriculum for the general hospital team
and will not only impart their knowledge but also promote good
Steps Towards Establishing a Liaison Service
communication skills, along with a sense of their enthusiasm for Despite the potential barriers to service development, several places
the work. Leadership skills are important, and liaison practitioners in the UK have successfully established liaison psychiatry services
often find themselves championing the cause of mental health to for older people. The final part of this chapter examines the steps
many stakeholders across health and social care. Good negotiating necessary to establish and perpetuate a successful liaison psychiatry
skills are useful when issues such as funding arise. One less recog- service for older people.
nized but equally important skill is that of diplomacy; sometimes it
is necessary to stay calm and politely point out that mistakes have Assessment of need
been made, rather than resort immediately to a critical incident or It is necessary to provide evidence that a service is required.
complaints procedure. This is particularly important since liaison Establish how many older people with mental health problems
psychiatry is a long game with few quick wins, and shifting staff there are likely to be in your local general hospital, assuming the
attitudes can take years rather than months or weeks in some parts prevalences in Table 25.1 are correct, and link that to what is known
of the hospital. about the poor outcomes. Draw on the experience of patients and
carers: the local branch of the Alzheimer’s Society will have people starting at the same time? The decision on the launch may be
with adverse experiences of general hospital care. What national closely related to recruitment; if only a few people are likely to meet
policies may be relevant? The internet is a useful resource of policy your person specification, then a slower, incremental launch with
and related documents to brief you. Become a salesman to get your ongoing recruitment may be better. Don’t forget accommodation;
message across. a base in the general hospital will allow the service to run at peak
efficiency. Adequate administrative support is also essential.
Scoping the project
Who are the key stakeholders and what motivates them? What are Evaluation
their must-dos and how can you link to them? You will need to Once you have established a service, you will be expected to show
engage a wide range of supporters across health, social care, and that it works. Although evaluation is a complex process, and out-
voluntary agencies who will all need briefing about why they should comes are subject to influence by a large number of confound-
support you. Look for opportunities to change, including sources of ing factors, it is still possible to show that the service is having an
funding that are briefly available for pilot or similar projects. What effect. One thing always seen is an increase in referrals, resulting
capacity is there for change, and are there reorganizations of serv- in more older people accessing specialist mental health assessment
ices that you can link with? and treatment, and all liaison services should prospectively harvest
this data, since no-one else will do it for them. Outcomes examined
Mapping the process should be measured in individual patients and carers, e.g. through
What will your service look like? A range of factors influence this, a satisfaction questionnaire. Similar tools can be used with gen-
including hospital size, geography, current service resourcing and eral ward staff so that it can be demonstrated that the service is
provision, and local opinion. Think of it as an evolutionary proc- well received and perceived as helpful. As well as these qualitative
ess if resources are scant at the moment. Use national standards approaches, it is possible to obtain hard outcome measures, such
for staffing of services (Faculty of Old Age Psychiatry, 2006). What as length of stay for those patients coded by patient administra-
is already there that can be built on? A single liaison nurse can be tion system databases as having a primary or secondary mental
used as the foundation for a larger multidisciplinary team. What health problem; although coding may not be particularly accurate
blocks to progress are there? Individual clinicians may be unsym- and probably underestimates the level of mental health problems,
pathetic to your cause, in which case sell it to them better, or if that this is not likely to be systematically biased. It is worth bearing in
fails find a way around them. Find out where others have been suc- mind that patients referred to the service are by definition likely to
cessful and collaborate, rather than reinventing the wheel. have increased length of stay compared to those not referred. Other
areas to measure may include delayed transfers of care, psycho-
Service design tropic prescriptions (pharmacy colleagues may be able to help with
This includes the design of clinical pathways that cross interfaces, this), and referrals to community teams and other mental health
together with services to ensure the optimal working of these services for follow-up after discharge. Teaching activity should be
pathways. An operational policy will be needed, together with recorded, and its impact measured by regular checks on general
support systems to deliver the policy. Decisions about paperwork hospital staff knowledge and skills. Audit cycles can repeatedly
are required—which trust’s medical records are written in and by report rapid responses to referrals and prompt written communi-
whom? Which trust logo is on the letterhead? Line management cations with relevant teams. A successful evaluation process needs
of clinicians and administrative staff will be required, and this can to be built in to the day-to-day working of the liaison service, and
be from the general hospital management structure or from the should be linked to performance management process.
mental health provider. IT systems and support will be needed, and
decisions about which organization’s IT systems (both hardware Sustainability and spread
and software) will be used by the service; there are advantages and A successful evaluation is one good way of attaining sustainabil-
disadvantages to using either the mental health provider or the gen- ity, but it is important not to be complacent. Key stakeholders will
eral hospital IT systems (which will inevitably be incompatible with move on to other posts, bringing new people you will have to sell
each other). Developing and delivering the educational curriculum your service to. Difficult issues may arise, and it is important to be
may require liaison with local and national education and training open and honest, rather than papering over the cracks. Succession
bodies. strategies are vital in what are usually small teams, and enthusing
trainees in all professions is one way of ensuring successful succes-
Implementation sion. Keep asking ‘Can we do it better?’; strive for improvement and
Establishing a liaison service is complex and requires senior mana- be flexible within resources. Liaison can be lonely, so find a network
gerial input for project management if it is to succeed. The vision of like-minded people and get support and new ideas from them.
and aims of the service should be shared with clinical and manage- Tell others what you do and they may come to you for ideas too.
rial teams involved (both in the general hospital and the mental
health provider organization), so that teams understand what is
happening and sign up to the new arrangements. A robust com- Conclusion
munications strategy will ensure the sharing of relevant clinical This chapter has highlighted some of the challenges for research and
information so that appropriate follow-up can be arranged. Should service development in a complex and until recently under-recognized
the service be launched incrementally and rolled out gradually to area. Mental health problems are common in older people in gen-
different areas of the hospital, or is it a big bang launch, everything eral hospitals, and the poor outcomes they experience ought to be a
rallying call for better services and a clearer understanding of liaison SDO Project (08/1504/100). <http://www.netscc.ac.uk/hsdr/files/project/
service effectiveness. The information in this chapter is intended to SDO_FR_08–1504–100_V01.pdf> (accessed 21.05.2012).
stimulate more interest in the area, and to spark discussions about Horrocks, J., et al. (2003). Self-injury attendances in the accident and
emergency department: clinical database study. British Journal of
different service models that make things better for older people in
Psychiatry, 183(1), 34–9.
general hospitals.
Inouye, S.K. (1994). The dilemma of delirium: clinical and research
controversies regarding diagnosis and evaluation of delirium in
References hospitalized elderly medical patients. American Journal of Medicine,
Andrew, M., Freter, S., and Rockwood, K. (2006). Prevalence and outcomes 97(3), 278–88.
of delirium in community and non-acute care settings in people without Merino, J.G. and Hachinski, V. (2002). Stroke-related dementia. Current
dementia: a report from the Canadian Study of Health and Aging. BMC Atherosclerosis Reports, 4(4), 285–90.
Medicine, 4(1) 15. NICE (2010). Delirium: diagnosis, prevention and management. Clinical
Atkin, K., Holmes, J., and Martin, C. (2005). Provision of care for older Guideline 103. NICE, London.
people with co-morbid mental illness in general hospitals: general nurses’ Norquist, G., et al. (1995). Quality of care for depressed elderly patients
perceptions of their training needs. International Journal of Geriatric hospitalized in the specialty psychiatric units or general medical wards.
Psychiatry, 20(11), 1081–3. Archives of General Psychiatry, 52(8), 695–701.
Cole, M.G. and Bellavance, F. (1997). Depression in elderly medical Parsonage, M. and Fossey, M. (2011). Economic evaluation of a liaison
inpatients: a meta-analysis of outcomes. Canadian Medical Association psychiatry service. London School of Economics, London.
Journal, 157(8), 1055–60. Raymont, V., et al. (2004). Prevalence of mental incapacity in medical
Department of Health (2001). National service framework for older people. inpatients and associated risk factors: cross-sectional study. Lancet,
DH, London. 364(9443), 1421–7.
Department of Health (2009). Living well with dementia: a national dementia Royal College of Psychiatrists (2005). Who care wins: improving the
strategy. DH, London. outcome for older people admitted to the general hospital. Working
Department of Health and Care Services Improvement Partnership (2005). Group of the Faculty of Old Age Psychiatry, Royal College of
Everybody’s business. Integrated mental health services for older adults: a Psychiatrists London.
service development guide. DH, London. Ruddy, R. and House, A. (2003). A standard liaison psychiatry service
Faculty of Old Age Psychiatry (2006). Raising the standard: specialist services structure? A study of the liaison psychiatry services within six strategic
for older people with mental illness. Royal College of Psychiatrists, health authorities. Psychiatric Bulletin, 27(12), 457–60.
London. Ruddy, R. and House, A. (2005). Meta-review of high-quality systematic
Farrell, K.R. and Ganzini, L. (1995). Misdiagnosing delirium as depression reviews of interventions in key areas of liaison psychiatry. British Journal
in medically ill elderly patients. Archives of Internal Medicine, 155(22), of Psychiatry, 187, 109–20.
2459–64. Scottish Office (2001). Adding life to years. Report of the Expert Group
Ferro, J.M., Caeiro, L., and Verdelho, A. (2002). Delirium in acute stroke. on Healthcare of Older People. Scottish Office Health Department,
Current Opinion in Neurology, 15(1), 51–5. Edinburgh.
Gill, D. and Hatcher, S. (2006). Antidepressants for depression in medical Strain, J.J., et al. (1991). Cost offset from a psychiatric consultation-liaison
illness (Systematic Review). Cochrane Database of Systematic Reviews, intervention with elderly hip fracture patients. American Journal of
3(3). Psychiatry, 148(8), 1044–9.
Gurland, B., et al. (1976). The geriatric mental status interview (GMS). Swift, G. and Guthrie, E. (2003). Liaison psychiatry continues to expand:
International Journal of Aging and Human Development, 7(4), 303–11. developing services in the British Isles. Psychiatric Bulletin, 27(9), 339–41.
Gustafson, Y., et al (1991). Underdiagnosis and poor documentation of acute Turner-Stokes, L. and Hassan, N. (2002). Depression after stroke: a review
confusional states in elderly hip fracture patients. Journal of the American of the evidence base to inform the development of an integrated care
Geriatrics Society, 39(8), 760–5. pathway. Part 1: diagnosis, frequency and impact. Clinical Rehabilitation,
Hally, O. and Cooney, C. (2005). Delirium in the hospitalised elderly: an audit 16(3), 231–47.
of NCHD prescribing practice. Irish Journal of Psychological Medicine, Wijeratne, C., et al. (2003). The neglect of somatoform disorders by old
22(4), 133–6. age psychiatry: some explanations and suggestions for future research.
Holmes, J. and House, A. (2000a). Psychiatric illness predicts poor outcome International Journal of Geriatric Psychiatry, 18(9), 812–19.
after surgery for hip fracture: a prospective cohort study. Psychological Winrow, A. and Holmes, J. (2005). Old age medical patients screening positive
Medicine, 30, 921–9. for depression. Irish Journal of Psychological Medicine, 22(4), 124–7.
Holmes, J. and House, A.O. (2000b). Psychiatric illness in hip fracture. Yohannes, A.M., Baldwin, R.C., and Connolly, M.J. (2000). Depression and
Systematic review. Age and Ageing, 29(6), 537–46. anxiety in elderly outpatients with chronic obstructive pulmonary disease:
Holmes, J., Bentley, K., and Cameron, I. (2002). Between two stools: psychiatric prevalence, and validation of the BASDEC screening questionnaire.
services for older people in general hospitals. University of Leeds, Leeds. International Journal of Geriatric Psychiatry, 15(12), 1090–6.
Holmes, J., Bentley, K., and Cameron, I. (2003a). A UK survey of psychiatric
services for older people in general hospitals. International Journal of
Geriatric Psychiatry, 18(8), 716–21. Further reading
Holmes, J., et al. (2003b). Trends in psychotropic drug use in older people in Draper, B. and Melding, P. (eds) (2001). Geriatric consultation liaison
general hospitals. Pharmaceutical Journal, 271(7272), 584–6. psychiatry. Oxford University Press, Oxford.
Holmes, J., et al. (2010). Liaison mental health services for older people: a Royal College of Psychiatrists (2005). Who cares wins: improving the outcome
literature review, service mapping and in-depth evaluation of service models. for older people admitted to the general hospital. Working Group of the
National Institute for Health Research Service Delivery and Organization, Faculty of Old Age Psychiatry, Royal College of Psychiatrists, London.
CHAPTER 26
Social care
Jo Moriarty
Medical professionals need to be educated in how the social care system

works—given the confusing array of approaches taken by different
[local] authorities. (Altmann, 2011)
At the heart of this fragmentation [of services and commissioning]
lies a key issue—the distinction that has been drawn between what is
health care (commissioned and largely delivered by the NHS), and what
is social care (mainly commissioned by local authorities and individuals,
and provided by many different sources). This distinction, much
discussed but little understood, arises from a succession of political
compromises stretching back to the 1920s. (House of Commons Health
Committee, 2012: 7)
Our . . . approach leads us to define social care as the activities and
relations involved in meeting the physical and emotional requirements of
dependent adults and children, and the normative, economic and social
frameworks within which these are assigned and carried out.
(Daly and Lewis, 2000: 285)
These opening quotations illustrate both the complexity of the social mental health problems and their families are subject to wider
care system in the UK and the challenges involved in summarizing political and funding priorities. At the time of writing, a number
what support it provides to older people with mental health prob- of major policy changes in England discussed in the chapter, such
lems. The effects of dementia and depression, the two most common as the policy of personalization advocated by both the previous
mental health conditions seen by old age psychiatrists, are such that Labour and current Coalition government, have yet to be fully
the overwhelming majority of patients and their families will need evaluated in terms of their impacts upon services for older peo-
social care support. This means that old age psychiatrists working ple with mental health problems and their families. Furthermore,
alongside those arranging or delivering social care require a broad the government has promised but has yet to pass legislation mak-
understanding of how these services are provided. Furthermore, ing changes to funding for long-term care. This may resolve some
they are often the first point of call for patients and their families of the anomalies outlined here, although the wider debates about
attempting to negotiate a complicated and diverse system of fund- whether social care is adequately funded are likely to continue.
ing and support, given that the general public is generally poorly At the same time, it is important not to exaggerate the rate of
informed about what constitutes social care and how it is funded change. While there are signs of improvement, many of the chal-
(Ipsos MORI Social Research Institute, 2011). lenges for social care services identified by pioneering old age
The chapter aims to describe some of the services that come under psychiatrists, such as the need to support family carers caring for
the social care ‘umbrella’, explain how they are funded, comment someone with dementia (Bergmann et al., 1978) or provide bet-
on their effectiveness, and discuss some of the key current policy ter treatment for older people in care homes with depression or
debates. Social care is what is termed a ‘devolved matter’, so the dementia (Ames et al., 1988), are by no means settled. In this sense,
Scottish and Welsh governments and the Northern Ireland Assembly the sector is characterized by both change and continuity.
can make decisions about how social care services will operate
under their jurisdiction. Since devolution, social care policy has
become increasingly divergent across the UK (Birrell, 2009). While Defining Social Care
it broadly concentrates on England, the chapter makes some refer- The term ‘social care’ is not widely used outside the UK and, indeed,
ence to developments in Scotland, Wales, and Northern Ireland. was first used to provide a generic label for the people who worked
An important caveat for readers is that social care is a rapidly in residential care and other social services but who were not social
changing landscape in which the preferences of older people with workers, rather than to describe the support that they provided
(Platt, 2007). Nevertheless, a number of attempts have been made people accessing places in care homes through the social security
to provide a framework for describing this rather amorphous con- system created a policy imperative to reduce welfare expenditure by
cept. Waine and colleagues (2004) describe social care as covering: controlling the demand for care.
all interventions provided or funded by statutory and/or independent The utility of means testing is, of course, dependent upon the
agencies which support older people, younger adults and children in extent to which revenue from charges outweighs the costs of col-
their daily lives, and provide services which they are unable to pro- lecting assets. Over the past 25 years, the income of people aged
vide for themselves, or which it is not possible for family members to over the state retirement age from occupational pensions, invest-
provide without additional support. They can be provided at home, in ments, earnings, and benefits has risen considerably. Between 1979
day centres or on a residential basis, including substitute family care and 1997, the income of so-called pensioner households grew by
and care homes. 68% in real terms compared with an average 36% increase in earn-
(Waine et al., 2004: 1)
ings across the economy as a whole (Department for Work and
Increasingly, as will be discussed in more detail later in this chap- Pensions, 2011). The number of older people with assets in the form
ter, these interventions now include ‘cash for care’ in the form of of housing equity has also increased, with around three-quarters of
personal budgets and direct payments that allow recipients to decide older people owning their own home (Terry and Gibson, 2010).
how, and on what, to spend the money allocated for their support. Set against this increase in the number of older people able to fund
The origins of the distinction between social and healthcare at least some of their care in old age, we must set aside the costs that
are generally attributed to the separate jurisdictions set out in the are needed to provide intensive care at home or in care homes for
National Health Act 1946, which established the National Health people who have very high support needs. Very broadly, under the
Service (NHS) in 1948, and the National Assistance Act 1948, which current system, people living in England with assets over £23,250
gave local authorities (LAs) responsibility for making ‘provision for (the limit in 2012–2013) receive no financial support from the state
the welfare of disabled, sick, aged and other persons’ (Means and and need to fund their own care until they have ‘spent down’ their
Smith, 1998; Glasby and Littlechild, 2004). These two pieces of leg- assets to the point at which they qualify for means-tested support.
islation assumed that it was possible to distinguish between people It is estimated that about 170,000 people (about 45% of all those
who were sick, and thus entitled to NHS healthcare free at the point living in care homes) are currently funding their own care. A fur-
of delivery, and those whose frailty and disability resulted in their ther 168,701 people are paying for their own home care, although
having social care needs, which would be met by the LA (Glasby, it is more difficult to estimate this number accurately (Institute of
2007: 66). In reality, differences between health and social care are Public Care, 2011). Depending upon the type of care needed and
often blurred, particularly where a person requires 24-h care on a the part of the country in which the person lives, someone with
long-term basis, and this has been the subject of various legal cases. dementia who lives in a nursing home for the last 2 years of his or
her life might spend over £100,000 (Simon, 2010).
A further complication for people making decisions about
How Social Care Is Funded whether to stay at home or move into a care home is that housing
At the heart of the ‘much discussed but little understood’ distinc- assets are not included in the financial assessments of people living
tion between social care and healthcare mentioned at the start at home. However, this is not the case should they choose to move
of this chapter (House of Commons Health Committee, 2012: 7) into a care home, provided they have no dependants living in their
is the difference between NHS care free at the point of delivery house (Commission on Funding of Care and Support, 2011). Many
and means-tested social care. Crucially, while the 1945–51 Attlee homeowners are reluctant to sell their home to pay for care, mean-
Government was reorganizing health and welfare services in the ing that they are less likely to move into long-term care than those
UK, they had to adopt a mixture of pragmatism and principle. A renting their accommodation (Hancock et al., 2002; McCann et al.,
decision was made to fund the NHS from central taxation but to 2012). Concerns have been expressed that this can lead to dispro-
fund support provided by LAs through a complicated mix of local portionate burdens being placed on family carers. A cross-European
taxation, charges, and centrally provided welfare benefits (Lowe, study (Schneider et al., 1999) has found that financial concerns were
2002; Thane, 2009). The end result was that the National Assistance one of the factors contributing to higher scores on the Zarit Burden
Act 1948 gave LAs the power to charge for services provided to Inventory (Zarit et al., 1980), a widely used self-report measure cov-
older and disabled people ‘in need of care and attention which is ering difficulties commonly faced by family carers.
not otherwise available to them’. In practice, while all those apply- Old age psychiatrists should be particularly alert to the needs of
ing for LA funding to support the cost of living in a care home were spouses caring for a person with dementia who may delay decisions
subjected to means testing, charges were rarely applied for services about long-term care, even when they feel physically and psycho-
provided to people living at home until the 1980s (Balloch, 1994; logically unable to carry on caring, through lack of understanding
Baldwin and Lunt, 1996). However, since then, LAs have come about their entitlements under the current system or options that
under increasing financial and political pressures to means test. might be available to them, such as deferred payments or equity
Depending upon a person’s assets, services such as home care are release. An emerging role for voluntary organizations and other
now generally charged at around their full economic cost. social care providers is to provide accurate and tailored financial
Lewis (2001: 349) has described this change as resulting from planning information for people with dementia and their families
a ‘pincer movement’ in which increased demand for NHS serv- (Manthorpe et al., 2011; Samsi and Manthorpe, 2011). Some coun-
ices and the closure of large numbers of NHS beds for so-called cils have piloted equity release schemes for older home owners, but
long-stay patients resulted in increasingly tight definitions of what it is not yet clear how feasible it would be to provide this on a large
constituted healthcare, while huge increases in the number of scale (Terry and Gibson, 2010, 2012).
CHAPTER 26 social care 337
Eligibility for Social Care throughout the country. At the time of writing (spring 2013), the
Coalition government’s draft Care Bill has adopted many of the
In addition to the complexities of the current funding system, Law Commission’s recommendations, including strengthened
access to social care arranged or funded by the LA depends upon rights to an assessment for individuals and carers, and a minimum
eligibility. Currently, there are various statutes governing how peo- national eligibility threshold across local authority areas. If a per-
ple living in England and Wales are eligible for social care. In gen- son moves to a new area, their new LA will have to continue to
eral, LAs base decisions on who is eligible for social care services meet needs met by their former authority pending reassessment.
on the results of an assessment. Depending upon whether a person Similar legislation is currently under way in Wales, although there
is assessed as having low, moderate, substantial, or critical needs are some differences, including a code of practice to be laid down
(Department of Health, 2010a), councils can then decide what sup- by the Welsh assembly on how local councils implement their
port he or she will be offered. This leads to such variation between social care responsibilities.
LAs that it has been suggested that:
There are currently 152 different adult social care systems—one for
each local authority in England. Entitlement to services differs across Reforming Social Care—Funding
the country and people complain of a ‘postcode lottery’ of care. Those responsible for planning and arranging social care recognize
Different people, with similar care needs, can receive very different the inconsistencies in, and unfairness of, the current system, but
levels of support from their local authorities. partly attribute it to difficulties in funding (Association of Directors
(Commission on Funding of Care and Support, 2011: 15)
of Adult Social Services, 2011). Adult social care has enjoyed an
This variability and inconsistency in eligibility and assessment average annual real-terms growth of 5.1% since 1994, but much
processes may contribute to the misplacement of frail older people of this has been absorbed by demographic pressures. There have
(Challis and Hughes, 2002). been proportionally greater increases in funding for people with
Another factor determining eligibility is the extent to which learning difficulties and physical disabilities, but spending on older
assessments take account of carers’ needs as well as those of the per- people has increased by less than 3% and has not kept pace with
son for whom they care. Guidance (Department of Health, 2010a) demographic change. There are fears that the difference between
suggests that preventive social care support, such as 1 or 2 h of home the numbers of people needing social care and the amount of
care or funding to pay for home adaptations or telecare could be money available to support them could lead to a funding gap of at
provided to carers of people with moderate or substantial needs— least £1.2 billion by 2014. This could increase the number of people
even if the council is only providing services to those assessed as in need of but not receiving social care and lead to additional pres-
having critical need—in the expectation that this will delay the sures on the NHS (Humphries, 2011).
point at which the needs of the person cared for become substan- The previous Labour government promised to look at how care
tial. However, the former social care regulator, the Commission for should be funded and set up a Royal Commission on Long-Term
Social Care Inspection, considered that while guidance emphasized Care. The Commission’s remit includes community nursing serv-
the importance of assessing carers’ needs, this has ‘been lost sight of ices and continuing care hospital beds, as well as social care services
in [the] implementation’ (Commission for Social Care Inspection, and assistive technology and supported housing. It recommended
2008: 23). that personal care (defined as nursing care provided under the
direction of a nurse and personal care that involved directly touch-
ing a person’s body, e.g. to wash or dress that person) should be free,
Reforming Social Care—Eligibility although the people living in care homes should still be expected to
In 2011 the Law Commission published the results of its 3-year pay for their ‘hotel’ costs (charges for food, laundry, and accommo-
enquiry into adult social care. It recommended that, rather than dation) (Royal Commission on Long-Term Care, 1999).
the complex and piecemeal patchwork of legislation and guidance The Labour government accepted many of the Commission’s rec-
that exists at the moment, there should be single separate statutes ommendations but rejected the free personal care option. However,
for adult social care in England and Wales. Social care services the Scottish Government (then Scottish Executive) chose to fund
would be provided at two levels, the first consisting of universal free personal care in people’s homes and care homes. Since then,
services to which everyone would have access. This would include although definitive empirical evidence on the impact of these
providing information and advice to those who did not want or changes is lacking, commentators have praised the decisiveness of
were not eligible for an assessment. The second level would con- the Scottish Government and drawn attention to the strong public
sist of targeted services provided through an assessment. Councils support for this decision (Bowes and Bell, 2007; Dickinson et al.,
would then have a duty to meet eligible needs. There would be a 2007; Vestri, 2007). However, they also recognize that those whose
new duty to assess all carers, not just those providing ‘substantial needs for personal care are met by their family carers have benefited
amounts of care on a regular basis’, as happens at the moment, and less from this decision. Here, concerns have been identified with
LAs would have new duties and powers to safeguard adults from flexibility (Bowes and Bell, 2007) and gaps in the amount of social
abuse and neglect (Law Commission, 2011). It also recommended care support that older people and carers living in Scotland receive
that assessments should be ‘portable’ so that people moving from (Innes et al., 2005; Vestri, 2007).
one part of the country to another do not have to undergo a new This highlights the need for funding systems that take account
assessment and a possible change in the amount or type of sup- of the needs of family carers as well as considering the needs of
port with which they are provided. Taken as a whole, these changes older people with mental health problems. The Northern Ireland
are designed to decrease the variability in social care support Assembly and Welsh Government have expressed support in
principle for providing free personal care, but have rejected the In 2005, the Department of Health invited 13 LAs with social
option as being too expensive. services responsibilities to pilot an individual budget scheme which
Upon its election in 2010, the Coalition Government set up the differed from direct payments in that it would combine different
Commission on Funding of Care and Support, headed by Andrew funding streams from social care, housing, and benefits, with the
Dilnot. Many of the Commission’s (2011) conclusions on the need exception of money for healthcare. The amount that individuals
for greater transparency and consistency in funding mirrored those would receive depended on what needs they were assessed as hav-
made by the Law Commission (2011), which had reported 2 months ing. The funds would then be spent in accordance with a support
earlier. The Dilnot Commission was especially concerned by what it plan agreed by the service user and a care manager or social worker
saw as the unfairness of a system in which a small proportion of peo- or an external ‘broker’. The aim of the pilots was to see if this system
ple were faced with what it termed ‘catastrophic care costs’ while the would increase choice for the service user—for instance, would an
majority might expect to pay only for comparatively small amounts older person choose to go out for a meal in a café or pub rather
of care in the last few weeks of their lives. It recommended that the than receive a hot meal delivery service (meals on wheels)? There
asset level at which people should be asked to pay for their social would also be opportunities to look at how assessments could be
care should be raised to £100,000 and that a cap of £35,000 should integrated to improve working across different agencies and reduce
be set on the amount that any individual is asked to pay for his or the number of assessments that an individual might be expected to
her care. It specifically noted that people with dementia and other undergo (Moran et al., 2011).
long-term neurological conditions who are likely to have significant Evaluation of the pilots was undertaken through a randomized
social care needs would benefit from this change (2011: 66). The multimethod comparison group study in which 510 individual
Commission’s proposals met with strong support from voluntary budget holders living in the 13 pilot authorities were compared
organizations representing people using services and their families with 449 people receiving conventional services in the same local-
and from various think-tanks, such as the King’s Fund. However, the ity (Manthorpe et al., 2009; Moran et al., 2011; Netten et al., 2011;
government has decided that the current economic difficulties mean Stevens et al., 2011). The overall conclusions from the evaluation
that they must set the cap at £72,000. This limit would only apply to were that individual budgets produced positive outcomes in terms
care costs and not to the costs of accommodation, meals, and so on. of service users’ quality of life and sense of control. Benefits could
It does not envisage implementing these changes until 2016. As a also be seen for their carers, who also reported a better quality of
general election is due to be held before this date, the final form of life and considered that they had more opportunities to have a
any funding changes remains uncertain. social life (Moran et al., 2012). However, the evaluation concluded
that individual budgets were not a ‘magic bullet’ (Netten et al.,
2011: 13). In particular, older people were less likely to use innova-
Reforming Social Care—Personalization tive support, and were more anxious about the process of planning
Until the last quarter of the twentieth century, the majority of social and managing support. These findings were thought to suggest that
care services were directly provided by the state and choice was not more work was needed on tailoring approaches to the needs of dif-
an important factor in how they were organized. However, from ferent groups of service users.
the late 1970s, choice of provider has been the main way in which Anticipating the results of the individual budgets pilot, the then
successive governments have sought to improve service quality Labour government issued its Putting People First strategy (HM
(Stevens et al., 2011). In the UK, as in the majority of the more Government, 2007) which stated that personal budgets (as they
developed countries, the option for people using services to choose were retermed) would be provided to everyone eligible for publicly
either services directly provided to them or to receive cash that they funded adult social care support other than in circumstances where
can use to purchase services (so-called cash for care schemes) has they required emergency access to provision. People could choose
become an important policy objective (Glendinning et al., 2004). to receive an individual budget in the form of a direct (cash) pay-
This development was particularly welcomed by members of the ment held directly by the person or, where they lacked capacity by
disability movement, mainly consisting of people aged 18–65 with a ‘suitable person’ such as a carer or advocate who could act as their
physical disabilities, for whom choice was an important way of representative, by way of an ‘account’ held and managed by the
achieving greater independence and autonomy (Prideaux et al., council or another third party in line with the person’s wishes, or by
2009; Stevens et al., 2011). While control and empowerment are a combination of the two. In November 2010, the Coalition govern-
important ends in themselves, a key policy rationale is that people ment confirmed that it too supported this system, stating that:
are better at identifying what will support them most effectively, and
The time is now right to make personal budgets the norm for everyone
can potentially draw on informal networks. This will, it is hoped,
who receives ongoing care and support—ideally as a direct cash pay-
enable formal social care resources to work in a more complemen- ment, to give maximum flexibility and choice.
tary way with individuals and their families (Netten et al., 2012). (Department of Health, 2010b: 16)
The option to choose a direct payment instead of directly provided
services was first made available to people living in the UK following It set a timetable of April 2013 for local councils to provide per-
implementation of the Community Care (Direct Payments) Act 1996. sonal budgets for everyone eligible for ongoing social care. However,
Payments were generally used to pay for a support worker or personal a key difference between these proposals and the individual budget
assistant to provide help with personal care and domestic tasks. The pilots was that there was no integration of different funding streams,
original legislation excluded older people and people who lacked thus potentially missing some opportunities to avoid duplication
capacity to manage their own money, although it was later extended and achieve greater flexibility (Moran et al., 2011).
to include these groups. However, take-up remained comparatively The Vision for Adult Social Care (Department of Health, 2010b)
low, particularly among older people (Leece and Leece, 2006). also distinguished between personal budgets and the wider policy
of personalization. Personalization was a term first coined in serv- have had access to personal budgets or direct payments, and a
ices for people with learning difficulties to convey attempts to place report by the Alzheimer’s Society (2011) expressed concern that few
the interests of service users above the interests of those arranging people with dementia have been offered them. Furthermore, people
or providing services, and to give them independence, choice, and with dementia and their carers were dissatisfied with the amount
control over the services they use. Personal budgets and direct pay- of information they were given and the amount of support they
ments are viewed as important ways of achieving personalization. were offered in managing payments, illustrating the importance
Underpinning the government’s vision are seven principles (some- of the advice expressed by Netten and colleagues (2011) that sys-
times referred to as the 7 Ps) (see Box 26.1). tems need to adapt to the needs of different groups of service users.
Research undertaken across the UK suggests that personal budg- Goodchild (2011) suggests that LAs must develop money manage-
ets (or, as they are termed in Scotland, self-directed support) have ment systems, so that it is not always the carer who is expected to
developed most extensively in England (Davey et al., 2007). In manage the personal budget, and ensure that advocacy services are
Wales, they are almost exclusively used by younger people, and in available to help people with dementia and their carers decide what
Scotland there are around 18 people using traditional home care is best for them. Manthorpe and Samsi’s (2012) interviews with
services for every person using self-directed support (Samuel, Adult Safeguarding Coordinators highlight the need for workers to
2011). At the time of writing, direct payments are available in be aware of the risks of financial abuse among those receiving a
Northern Ireland but not personal budgets. personal budget, but also identify opportunities for social workers
At this stage, there are no clear answers to the question whether and support planners to work with people with dementia and their
personal budgets will benefit older people with mental health prob- families to minimize the risks of abuse.
lems. It is only comparatively recently that people with dementia
Who Provides Social Care?
Box 26.1 The government’s principles for adult social care The policies towards marketization discussed in the section
Reforming Social Care—Personalization have created considera-
Our vision for a modern system of social care is built on seven ble variation across social care providers. In the past, LAs employed
principles: the majority of people working in the sector, although some care
Prevention: empowered people and strong communities will was also provided by voluntary organizations. Changes to social
work together to maintain independence. Where the state is security rules in the 1980s led to a rapid expansion in the number
needed, it supports communities and helps people to retain and of privately run care homes. This was followed by an expansion of
regain independence. in the number of home care providers. Taken as a whole, Eborall
and colleagues (2010) estimate that there are 40,600 units provid-
Personalization: individuals not institutions take control of their ing social care (excluding self-employed individuals). While some
care. Personal budgets, preferably as direct payments, are pro- of these include care homes owned by large national and multina-
vided to all eligible people. Information about care and support tional chains, the majority of social care enterprises are small or
is available for all local people, regardless of whether or not they small to medium businesses. Around three-quarters of employ-
fund their own care. ers have fewer than 20 employees and half have fewer than 10
Partnership: care and support delivered in a partnership between (Eborall et al., 2010). These figures exclude the increasing number
individuals, communities, the voluntary and private sectors, the of self-employed workers who work for people receiving personal
NHS, and councils—including wider support services, such as budgets and the rising number (albeit one that is hard to quantify)
housing. of people organizing and paying for their own care.
The final part of this chapter now considers what we know about
Plurality: the variety of people’s needs is matched by diverse
different social care workers and social care services providing
service provision, with a broad market of high quality service
support for older people with mental health problems and their
providers.
families.
Protection: there are sensible safeguards against the risk of
abuse and neglect. Risk is no longer an excuse to limit people’s
freedom. The Role of Social Workers
Productivity: greater local accountability will drive improve- Social workers and those who have similar responsibilities for
ments and innovation to deliver higher productivity and high arranging social care assessments, such as care/case managers
quality care and support services. A focus on publishing infor- and care coordinators, play an important role in achieving social
mation about agreed quality outcomes will support transparency care support that fits in with what people using services and their
and accountability. families want. Interviews with older people suggest that their
experiences of social workers are variable, but that they are most
People: we can draw on a workforce who can provide care and appreciative of those who offer the skills and knowledge to provide
support with skill, compassion, and imagination, and who are specialist advice and help ‘navigating’ the system (Manthorpe et al.,
given the freedom and support to do so. We need the whole 2008).
workforce, including care workers, nurses, occupational thera- It is perhaps significant that the clearest evidence for the effec-
pists, physiotherapists, and social workers, alongside carers and tiveness of social work with dementia comes from the Lewisham
the people who use services, to lead the changes set out here. Care Management Scheme which operated during the 1990s. This
(Department of Health, 2010b: 8.) involved basing care managers employed by the LA in a community
mental health team. People with dementia who were in the experi- been evaluated positively by care homes (Scottish Government/
mental group that received intensive care management remained at Alzheimer Scotland—Action on Dementia/Dementia Services
home for longer and experienced better quality of life (Challis et al., Development Centre, 2009).
2002). In contrast, social workers in ‘standard’ teams not providing Extra care housing in which people with dementia can live inde-
specialist mental health support could expect to spend more of their pendently but have access to support workers is emerging as an
time on assessments and less time considering how to refine the care additional option to moving into a care home. People with demen-
plans they had set up (Weinberg et al., 2003). tia value the extra independence and choice that living in extra care
Since these studies were undertaken, the changes to eligibil- housing offers (Evans et al., 2007), but there are no comparative
ity criteria mentioned earlier have meant that social workers are studies of the two options.
increasingly seeing people with very complex needs and there is A study of people in care homes with depression found that
evidence that safeguarding is becoming an increasingly central part after training from members of the local community mental health
of their role (Manthorpe et al., 2009). It is possible that the devel- team, care staff were able to work individually with residents on
opments in personalization will alter the dual emphasis on assess- activities such as re-establishing contact with friends and attending
ments and safeguarding, leading to the suggestion that it might lead social activities. These interventions produced significant reduc-
‘either to a reinvigoration or erosion of social work skills dependent tions in residents’ scores on a recognized measure of depression
upon the nature of local implementation’ (Jacobs et al., 2013: 18). (Lyne et al., 2006).
Another suggested potential role for social workers is for them to
become more involved in memory clinics where their experience Home-Based Care
in breaking bad news might mean that they could have a role in
supporting newly diagnosed people (Manthorpe and Iliffe, 2009). High rates of depression (Banerjee and Macdonald, 1996) and
Social workers could also become more involved in preliminary dementia have been found among people using home care
screening for identifying possible dementia among people living at (Livingston et al., 1997). Home-based care can cover a range of
home (Clarkson et al., 2012). support, from workers to help with personal care, to support
It is important to recognize that social work qualifying education is workers whose role is to enable the person with dementia to take
currently generic, and it is not entirely clear how much teaching stu- part in leisure activities and hobbies, to services designed to give
dents receive on the needs of older people with mental health prob- carers a break. One study of consecutive referrals to social work
lems. However, as the availability of dementia studies programmes teams in four areas found that people with cognitive impairment
in higher education increases, there are now more opportunities for receiving home care remained at home for longer (Andrew et al.,
social workers to undertake continuing professional development 2000). Accounts of specialist schemes have suggested that home
in this area, especially as these courses generally offer the opportu- care services specializing in supporting people with dementia can
nity to study flexibly, either part time or thorough distance learning. offer advantages in that they are more likely to offer a worker who
Social workers’ recognition of depression in older people is thought can build up a good relationship with the person with dementia,
to need improvement, so there are opportunities for continuing pro- and staff are more likely to have had access to specialist training
fessional development in this area too (Clarkson et al., 2012). (Rothera et al., 2008; Snayde and Moriarty, 2009). However, other
work (Challis et al., 2010) has suggested that what matters is not so
much whether a service is generic or specialist but that it conforms
Direct Care Workers to good standards in dementia care and that both types of service
In contrast to the comparatively small number of professionals can do this.
working with older people with mental health problems, such as
social workers, occupational therapists in community reablement Conclusion
teams, and nurses in nursing homes, there is a far larger group of
direct care workers, many of whom do not possess any vocational Good support for older people with mental health problems
or professional qualifications. Temporary workers are overrepre- requires good partnership working between health and social care.
sented in the dementia care workforce (Hussein and Manthorpe, This chapter has shown that the differing origins of health and social
2012). Low levels of pay and poor status are generally cited as key care and different ways in which services have been organized helps
reasons why it can prove to be difficult to retain staff (Moriarty, explain why there have often been barriers to more effective work-
2010; Hussein and Manthorpe, 2011). ing. The government plans for greater integration between health
and social care suggest that there will be opportunities to develop
more ‘joined up’ services. At the same time, such approaches are
Care Homes and Extra Care Housing very dependent upon finding solutions to the long-standing issues
Low levels of training and poor rates of retention are thought about funding for social care and the various and complicated
to contribute to the poor quality of care received by some older pieces of legislation on which it is based.
people with mental health problems in care homes (Mozley et al., Finally, it is important to recognize that much of the care that
2000, 2004; Dening and Milne, 2009). Confidence in responding to older people with mental health problems receive on a daily basis is
behavioural difficulties is a particular issue (Hughes et al., 2008) and delivered by social workers, many of whom have only limited train-
many care home managers would welcome specific advice from old ing about older people’s mental health problems. There are clear
age psychiatrists on this topic (Purandare et al., 2004). In Scotland, opportunities for health professionals to play a part in delivering
the appointment of a nurse with a specific remit to liaise with care training on topics that care workers find relevant and useful for
homes to help them care for people with behavioural problems has their work.
References Department of Health (2010b). A vision for adult social care: capable
communities and active citizens. DH, London. <http://www.dh.gov.uk/
Altmann, R. (2011). Saga press release 13 June 2011: doctors ‘innocently prod_consum_dh/groups/dh_digitalassets/@dh/@en/@ps/documents/
ignorant’ of social care system. <http://www.saga.co.uk/newsroom/ digitalasset/dh_121971.pdf> (accessed 02.04.2012).
press-releases/2011/1–9/doctors-innocently-ignorant.aspx> (accessed
Dickinson, H., et al. (2007). Free personal care in Scotland: a narrative review.
15.03.2012).
British Journal of Social Work, 37, 459–74.
Alzheimer’s Society (2011). Getting personal: making personal budgets work
Eborall, C., Fenton, W., and Woodrow, S. (2010). The state of the adult social
for people with dementia. Alzheimer’s Society, London.
care workforce in England, 2008. The Third Report of Skills for Care’s
Ames, D., et al. (1988). Psychiatric illness in elderly residents of Part III homes in Skills Research and Intelligence Unit. Skills for Care, Leeds. <http://
one London borough: prognosis and review. Age and Ageing, 17, 249–56. www.skillsforcare.org.uk/research/research_reports/state_of_the_adult_
Andrew, T., et al. (2000). Outcome of referral to social services departments social_care_workforce_reports.aspx> (accessed 02.04.2012).
for people with cognitive impairment. International Journal of Geriatric Evans, S., et al. (2007). Supporting independence for people with dementia in
Psychiatry, 15, 401–5. extra care housing. Dementia, 7, 144–50.
Association of Directors of Adult Social Services (2011). ADASS budget Glasby, J. (2007). Understanding health and social care. Policy Press, Bristol.
survey. ADASS, London.
Glasby, J. and Littlechild, R. (2004). The health and social care divide: the
Baldwin, S. and Lunt, N. (1996). Charging ahead: local authority charging experiences of older people, 2nd edition. Policy Press, Bristol.
policies for community care. Policy Press, Bristol.
Glendinning, C., et al. (2004). Funding long-term care for older people: lessons
Balloch, S. (1994). A survey of social services charging policies 1992–1994. from other countries. Joseph Rowntree Foundation, York. <http://www.
Association of Metropolitan Authorities, London. jrf.org.uk/bookshop/eBooks/1859352065.pdf> (accessed 01.04.2012).
Banerjee, S. and Macdonald, A. (1996). Mental disorder in an elderly home Goodchild, C. (2011). Personal budgets for people with dementia. A report on
care population: associations with health and service use. British Journal challenges and solutions to implementation based on interviews with eight
of Psychiatry, 168, 750–6. local authorities in England. Mental Health Foundation, London. <http://
Bergmann, K., et al. (1978). Management of the demented elderly patient in www.mentalhealth.org.uk/content/assets/PDF/publications/Personal_
the community. British Journal of Psychiatry, 132, 441–9. budgets_for_people_living_with_dementia.pdf?view= Standard>
Birrell, D. (2009). The impact of devolution on social policy. Policy Press, Bristol. (accessed 02.04.2012).
Bowes, A. and Bell, D. (2007). Free personal care for older people in Scotland: Hancock, R., et al. (2002). The effect of older people’s economic resources on
issues and implications. Social Policy and Society, 6, 435–45. care home entry under the United Kingdom’s long-term care financing
Care and Support Bill (2013). Available from <http://media.dh.gov.uk.s3. system. Journals of Gerontology Series B: Psychological Sciences and Social
amazonaws.com/network/365/files/2012/07/CARE-AND-SUPPORT- Sciences, 57, S285–93.
BILL-FINAL.pdf> (accessed 06.03.2013). HM Government (2007). Putting people first. A shared vision and
Challis, D. and Hughes, J. (2002). Frail old people at the margins of care: some commitment to the transformation of adult social care. DH, London.
recent research findings. British Journal of Psychiatry, 180, 126–30. <http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/
Challis, D., et al. (2002). Care management, dementia care and specialist PublicationsPolicyAndGuidance/DH_081118> (accessed 02.04.2012).
mental health services: an evaluation. International Journal of Geriatric House of Commons Health Committee (2012). Social care: fourteenth report
Psychiatry, 17, 315–25. of session 2010–12. Volume I, report, together with formal minutes,
Challis, D., et al. (2010). Community support services for people with dementia: HC 1583. The Stationery Office, London. <http://www.publications.
the relative costs and benefits of specialist and generic domiciliary care parliament.uk/pa/cm201012/cmselect/cmhealth/1583/1583.pdf>
services. Personal Social Services Research Unit, Manchester. <http://www. (accessed 07.03.2012).
medicine.manchester.ac.uk/pssru/PSSRUInformation/Expert Briefing Hughes, J., et al. (2008). Care staff working with people with dementia.
CommunitySupport forPeoplewithDementiacompleteFINAL.pdf> Dementia, 7, 227–38.
(accessed 02.04.2012). Humphries, R. (2011). Social care funding and the NHS. An impending crisis?
Clarkson, P. et al. (2012). The identification and detection of dementia and King’s Fund, London.
its correlates in a social services setting: impact of a national policy in Hussein, S. and Manthorpe, J. (2011). Longitudinal changes in care worker
England. Dementia, 11(5), 617–32. turnover and vacancy rates and reasons for job leaving in England (2008–
Commission on Funding of Care and Support (2011).Fairer care funding: the report 2010). King’s College London, Social Care Workforce Research Unit,
of the Commission on Funding of Care and Support. <https://www.wp.dh. London.
gov.uk/carecommission/files/2011/07/Fairer-Care-Funding-Report.pdf> Hussein, S. and Manthorpe, J. (2012). The dementia social care workforce in
(accessed 31.03.2012). England: secondary analysis of a national workforce dataset. Aging and
Commission for Social Care Inspection (2008). Cutting the cake fairly: CSCI Mental Health, 16, 110–18.
review of eligibility criteria for social care. CSCI, London. Innes, A., et al. (2005). Dementia care provision in rural Scotland: service
Daly, M. and Lewis, J. (2000). The concept of social care and the analysis of users’ and carers’ experiences. Health and Social Care in the Community,
contemporary welfare states. British Journal of Sociology, 51, 595–601. 13, 354–65.
Davey, V., et al. (2007). Direct payments: a national survey of direct payments Institute of Public Care (2011). Estimating the number and distribution of
policy and practice. Personal Social Services Research Unit, London. self-funders of care in England. In: People who pay for care: quantitative
<http://www.pssru.ac.uk/archive/pdf/dprla.pdf> (accessed 02.042012). and qualitative analysis of self-funders in the social care market.
Dening, T. and Milne, A. (2009). Depression and mental health in care homes Putting People First Consortium. <http://www.puttingpeoplefirst.
for older people. Quality in Ageing and Older Adults, 10, 40–6. org.uk/_library/Resources/Personalisation/Localmilestones/
Department for Work and Pensions (2011). The Pensioners’ Incomes Series, People_who_pay_for_care_-_report_12_1_11_final.pdf> (accessed
2009–10. Department for Work and Pensions, London. <http:// 31.03.2012).
research.dwp.gov.uk/asd/asd6/2009_10/pi_series_0910.pdf> (accessed Ipsos MORI Social Research Institute (2011). Public opinion research on
15.03.2012). social care funding. A literature review on behalf of the Commission
Department of Health (2010a). Prioritising need in the context of Putting People on the Funding of Care and Support. Ipsos MORI, London. <http://
First: a whole system approach to eligibility for social care guidance on www.dilnotcommission.dh.gov.uk/files/2011/03/MORI.pdf> (accessed
eligibility criteria for adult social care, England 2010. DH, London. <http:// 14.03.2012).
www.dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/@dh/@en/@ Jacobs, S., et al. (2013). The personalization of care services and the early
ps/documents/digitalasset/dh_113155.pdf> (accessed 01.04.2012). impact on staff activity patterns. Journal of Social Work, 13(2), 141–63.
Law Commission (2011). Adult social care. The Stationery Office, London. Platt, D. (2007). The status of social care—a review 2007. DH, London.
<http://lawcommission.justice.gov.uk/docs/lc326_adult_social_care. <http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/
pdf> (accessed 01.04.2012). PublicationsPolicyAndGuidance/DH_074217> (accessed 15.02.2012).
Leece, D. and Leece, J. (2006). Direct payments: creating a two-tiered system Prideaux, S., et al. (2009). Disabled people and self-directed support schemes:
in social care? British Journal of Social Work, 36, 1379–93. reconceptualising work and welfare in the 21st century. Disability and
Lewis, J. (2001). Older people and the health-social care boundary in the UK: Society, 24, 557–69.
half a century of hidden policy conflict. Social Policy and Administration, Purandare, N., et al. (2004). Perceived mental health needs and adequacy of
35, 343–59. service provision to older people in care homes in the UK: a national
Livingston, G., Manela, M., and Katona, C. (1997). Cost of community care survey. International Journal of Geriatric Psychiatry, 19, 549–53.
for older people. British Journal of Psychiatry, 171, 56–69. Rothera, I., et al. (2008). An evaluation of a specialist multiagency home
Lowe, R. (2002) Financing health care in Britain since 1939: history and policy. support service for older people with dementia using qualitative methods.
<http://www.historyandpolicy.org/papers/policy-paper-08.html#about> International Journal of Geriatric Psychiatry, 23, 65–72.
(accessed 15.03.2012). Royal Commission on Long Term Care (1999). With respect to old age: long term
Lyne, K.J., et al. (2006). Analysis of a care planning intervention for reducing care—rights and responsibilities (Cm 4192-I). The Stationery Office, London.
depression in older people in residential care. Aging and Mental Health, Samsi, K. and Manthorpe, J. (2011). ‘I live for today’: a qualitative study
10, 394–403. investigating older people’s attitudes to advance planning. Health and
Manthorpe, J. and Iliffe, S. (2009). Changing the culture of social work Social Care in the Community, 19, 52–9.
support for people with early dementia. Australian Social Work, 62, Samuel, M. (2011) Expert guide to direct payments, personal budgets and
232–44. individual budgets. Community care, 19 August 2011. <http://www.
Manthorpe, J. and Samsi, K. (2012). ‘Inherently risky?’ Personal budgets for communitycare.co.uk/Articles/19/08/2011/102669/direct-payments-per
people with dementia and the risks of financial abuse: findings from sonal-budgets-and-individual-budgets.htm> (accessed 02.04.2012).
an interview-based study with adult safeguarding coordinators. British Schneider, J., et al. (1999). EUROCARE: a cross-national study of co-resident
Journal of Social Work, Epub 27 March. <http://dx.doi.org/10.1093/bjsw/ spouse carers for people with Alzheimer’s disease: I—Factors associated with
bcs023> (accessed 27.03.2012). carer burden. International Journal of Geriatric Psychiatry, 14, 651–61.
Manthorpe, J., et al. (2008). ‘ There are wonderful social workers but it’s a Scottish Government/Alzheimer Scotland—Action on Dementia/Dementia
lottery’: older people’s views about social workers. British Journal of Services Development Centre (2009). Facing dementia together: initial
Social Work, 38, 1132–50. report of the Lothian Dementia Project. Dementia Services Development
Manthorpe, J., et al. (2009). Training for change: early days of individual Centre, Stirling.
budgets and the implications for social work and care management Simon, E. (2010) Cut the cost of nursing home care. Daily Telegraph, 7 June.
practice: a qualitative study of the views of trainers. British Journal of <http://www.telegraph.co.uk/finance/personalfinance/7807967/Cut-the
Social Work, 39, 1291–305. -cost-of-nursing-home-care.html> (accessed 01.04.2012).
Manthorpe, J., et al. (2011). ‘Early days’: knowledge and use of the Mental Snayde, F. and Moriarty, J. (2009). Person-centred home care for people
Capacity Act 2005 by care home managers and staff. Dementia, 10, with dementia: developing a specialist service in an ethnically diverse
283–98. community. Dementia, 8, 148–52.
McCann, M., Grundy, E. and O’Reilly, D. (2012). Why is housing tenure Stevens, M., et al. (2011). Assessing the role of increasing choice in English
associated with a lower risk of admission to a nursing or residential social care services. Journal of Social Policy, 40, 257–74.
home? Wealth, health and the incentive to keep ‘my home’. Journal of Terry, R. and Gibson, R. (2010). Can equity release help older home-owners
Epidemiology and Community Health, 66, 166–9. improve their quality of life? Joseph Rowntree Foundation, York. <http://
Means, R. and Smith, R. (1998). From Poor Law to Community Care. Policy www.jrf.org.uk/sites/files/jrf/equity-release-plans-summary.pdf>
Press, Bristol. (accessed 16.03.2012).
Moran, N., et al. (2011). Joining up government by integrating funding Terry, R. and Gibson, R. (2012). Assessment of equity release pilot schemes.
streams? the experiences of the individual budget pilot projects for Joseph Rowntree Foundation, York. <http://www.jrf.org.uk/sites/files/
older and disabled people in England. International Journal of Public jrf/equity-release-schemes-full.pdf> (accessed 01.04.2012).
Administration, 34, 232–43. Thane, P. (2009) Memorandum submitted to the House of Commons’ Health
Moran, N., et al. (2012). Personalisation and carers: whose rights? Whose Committee Inquiry: Social Care. October 2009: history and policy. <http://
benefits? British Journal of Social Work, 42(3), 461–9. <http://dx.doi. www.historyandpolicy.org/docs/thane_social_care.pdf> (accessed
org/10.1093/bjsw/bcr075> (accessed 02.04.2012). 25.03.2012).
Moriarty, J. (2010) Competing with myths: migrant labour in social care. Vestri, P. (2007). Evaluation of the operation and impact of free personal care.
In: Ruhs, M. and Anderson, B. (eds) A need for migrant labour? Labour Scottish Executive, Edinburgh. <http://www.scotland.gov.uk/Resource/
shortages, immigration and public policy. Oxford University Press, Doc/167844/0046181.pdf> (accessed 31.03.2012).
Oxford. Waine, B., et al. (2004). Developing social care: values and principles. Social
Mozley, C.G., et al. (2000). Psychiatric symptomatology in elderly people Care Institute for Excellence, London. <http://www.scie.org.uk/
admitted to nursing and residential care homes. Aging and Mental publications/positionpapers/pp04/values.pdf> (accessed 15.03.2012).
Health, 4, 136–41. Weinberg, A., et al. (2003). What do care managers do? A study of working
Mozley, C., et al. (2004). Towards quality care: outcomes for people in care practice in older peoples’ services. British Journal of Social Work, 33,
homes. Ashgate, Aldershot. 901–19.
Netten, A., et al. (2012). Personalisation through individual budgets: does Zarit, S.H., Reever, K.E., and Bach-Peterson, J. (1980). Relatives of the
it work and for whom? British Journal of Social Work, 42, 8, 1556–73, impaired elderly: correlates of feelings of burden. Gerontologist, 20,
<http://dx.doi.org/10.1093/bjsw/bcr159> (accessed 02.04.2012). 649–55.
CHAPTER 27
Care homes for older people
Tom Dening and Alisoun Milne
This chapter is in two main parts. In the first, we consider the people Currently, there are relatively few ethnic minority older people
who live in care homes, with particular reference to their physical in long-term care in the UK. In 2001 only 1.2% of those surveyed
and mental health; in the second, we discuss the care home sec- were from an ethnic minority (Bebbington et al., 2001). In the US,
tor in more general terms, looking at the changing nature of care figures indicate that higher rates of residential care occur in the
homes and how care is provided, paid for, and regulated to achieve majority white population. Whilst this may also be the case in the
acceptable standards of quality. To give an idea of the scale and UK, some work suggests that amongst publicly funded residents
importance of this arena, two figures are particularly salient: there they may be overrepresented (Bebbington et al., 2001). In general,
are currently 418,000 older people resident in care homes in the black and minority ethnic (BME) residents are younger, more likely
UK, and the sector has a total current value of around £14 billion. to be male, have higher rates of dementia and incontinence, and
The fact that dementia—or severe memory problems—is estimated are more dependent on admission than their white counterparts
to affect four-fifths of care home residents is also noteworthy; some (Milne and Chryssanthopolou, 2005).
would even argue that dementia care should be considered the A distinction is generally made between residential care homes
main concern of care homes (Quince, 2013). and care homes with nursing (otherwise known as nursing
homes). Residential care homes provide personal care with activi-
Part 1: People Who Live in Care Homes ties of daily living, such as washing, dressing, and giving medica-
tion. Care homes with nursing offer nursing care to at least some of
The proportion of older people living in a care home at any one their residents. They tend to support people with higher depend-
time is relatively small, about 5% of the total over 65 population ency needs, including those who need regular medical attention,
in many higher-income countries (Organization for Economic e.g. stoma care or tube feeding. People admitted to nursing homes
Cooperation and Development, 2005). However, the lifetime risk have higher levels of dependency than those admitted to residen-
of needing residential care is considerable—probably about one tial care (Bebbington et al., 2001). In one study, over 90% of nurs-
person in five will live in a care home towards the end of their life. ing home residents were classified as ‘severely disabled’, compared
As might be expected, the chance of being admitted to a home with 70% in residential care, with the prevalence of severe disabil-
increases with age: for the age group 65–74, 0.8% of the popula- ity significantly higher among women than in men (Bajekal, 2002).
tion resides in care homes, compared with 4% between ages 75 However, there is considerable overlap between the two types of
and 84, and 15% of those aged 85 and over (Laing and Buisson, home. Bowman et al. (2004) found high levels of dementia, physi-
2010). Women residents tend to be older, with an average age of cal disability, and functional impairment in a large care home
85.6 years, compared to 83.2 for men (Milne and Williams, 2000). sample.
People admitted to care homes with nursing are slightly younger
than those admitted to ordinary residential care (Netten et al.,
2001). Moving into a care home
Bebbington et al. (2001) identified several characteristics com- Most people enter a care home because they can no longer live inde-
mon to care home residents. Typically they are over 80 years old, pendently as a result of physical and/or mental illness. Over half
female, have previously lived alone, are poor or dependent solely of the admissions to care homes come from hospital (Bebbington
on benefits, and have multiple disabilities or long-standing illness. et al., 2001); this rises to about two-thirds for nursing home admis-
The great majority of care home residents are single, widowed, sions. Moving into a care home is a major life decision, but it is
or divorced and most are women. The most susceptible group is often taken when people have very little information, when they are
nonmarried women aged 85 or over; they are four times as likely under pressure to make a quick decision, and when they are feeling
to be in a care home as their male counterparts. Factors that pro- ill or frail. Although many older people would rather not move into
tect against admission are: availability of family support, adequate a care home, it is important to recognize that those who have made
income, and living in a housing environment that is adapted to the move can identify advantages with their new home, including
accommodate disability (Milne et al., 2001; Dening and Milne, the safe and secure environment, the support they receive, and the
2005). company of others (Boaz et al., 1999). It is also important to bear in
mind that people do move between care homes too: about a tenth Research and Development Forum, 2006) and the Joseph Rowntree
of admissions are moves from one home to another (Laing and study of Older People’s Vision for Long-term Care (Bowers et al.,
Buisson, 2010). 2009). The former provides several personal testimonies that cover
Most older people wish to remain in their own homes and to live the span of life in a care home (a total of 25 older people), includ-
as independently as possible until death. It is only fairly reluctantly ing the move into care. The latter report involved 84 older people
that most people will accept, first, help provided in the commu- in discussions and focus groups, presenting a synthesis of their
nity and, later, the need to be in full-time residential or nursing findings but also several illustrative quotes from residents. For vari-
care. Residential care is widely regarded as an option of last resort. ous reasons, it may be easier to study families and carers of older
Common triggers that prompt a move into a care home are one people, though caution is needed in drawing inferences from this
or more of the following: bereavement; concerns about health; work about the perspectives of older people themselves. Davies and
poor or unsuitable housing; inadequate or unsatisfactory care or Nolan (2003, 2004, 2006) studied the experience of transition into
a breakdown in care arrangements at home; and/or other people’s residential care from the perspective of relatives, based on 37 inter-
concerns and anxieties for the older person’s wellbeing, safety, and views with 48 people who had assisted a close relative to move in.
protection (Bowers et al., 2009). The ‘other people’ who may be They distinguished three phases embedded in the process of transi-
concerned include relatives, friends, and health and social care per- tion: ‘making the best of it’; ‘making the move’; and ‘making it bet-
sonnel. There is a risk that their wish for a ‘safe’ option may some- ter’, referring to the periods before, during, and after the move.
times override the wishes of the older person. In some cases, for
example if older people have dementia, they may lack capacity to Mental health of residents: dementia
make a competent choice about their own care, in which case a ‘best The prevalence of dementia in care homes is high; a large UK study
interests decision’ can be made. identified 62% of residents as having dementia and a recent survey
The move to a care home often follows some form of crisis, so by the Alzheimer’s Society suggests that this figure may be as high
there is relatively little time to plan the move or to make informed as 80% (Matthews and Dening, 2002; Quince, 2013). Matthews and
choices about which home best suits the older person. The cause Dening (2002) found that the rate of dementia was slightly higher in
of the crisis may be the death of a spouse—or other long-term car- women but it did not increase significantly with age, perhaps reflect-
er—or an episode of serious illness requiring hospital admission, ing the threshold of impairment at which people enter long-term
so, quite apart from the stress of giving up one’s home, there will care. If residents with dementia have greater mortality than those
probably be other losses to contend with. Moreover, once admit- without, then the total burden of dementia may be greater than
ted to the care home, possible opportunities for moving back to this cross-sectional figure suggests. The prevalence of dementia is
one’s own home often rapidly disappear, e.g. if the person’s house fairly consistent across different types of care home, so even homes
or flat is sold to fund the residential care. If a move to a care home that are not registered as ‘dementia care homes’ have a majority of
is necessary, then one of the primary needs older people and their residents with significant cognitive impairment (Macdonald et al.,
families have is for information; some excellent guidance is avail- 2002). Dementia care is undoubtedly one of the principal functions
able, covering many of the practical and financial issues that need of long-term care for older people: e.g. the costs of care for peo-
to be addressed, although independent expert advice is hard to ple with diagnosable dementia are estimated at £8 billion (out of a
find (Alzheimer Scotland, 2007; Counsel and Care, 2009). Publicly total of £13 billion) (see Fig. 27.1). Unfortunately, staff assessments
available information about quality of care, including care home of the presence of dementia may be inaccurate (Sørensen et al.,
standards, is also published by the Care Quality Commission. It is 2001) or influenced by staff attitudes to dementia (Macdonald and
difficult, though, faced with a short timeline, to make a decision— Woods, 2005). There may also be perverse incentives not to identify
which may be driven by concerns such as freeing up a hospital it, as this could result in a more expensive placement being required
bed—for older people and/or their relatives to avail themselves (Macdonald and Dening, 2002).
of all available information and make a balanced and considered Many residents with dementia have behaviour disturbances (or
choice. behavioural and psychological symptoms of dementia; BPSD),
The biggest single health-related predictor of care home admis- especially activity disturbances (agitation), aggression, psychosis,
sion appears to be dementia. For example, Bharucha et al. (2004), and depressed mood, with reported prevalence of such behaviour
in a US study, found that dementia increased the risk of admission problems as high as 80 or 90% (Brodaty et al., 2001; Cheng et al.,
to a nursing home five-fold. In the UK, Banerjee et al. (2003) found 2009). Perhaps of particular concern are self-harming behaviours,
that having a coresident carer at home had a strongly protective which may be active, such as scratching oneself or punching objects
effect—the risk of being institutionalized was 20 times higher in (de Jonghe-Rouleau et al., 2005), or passive, e.g. refusal to take food
people who did not have a carer living with them. This must be or drink (Draper et al., 2003). Overall, behaviour problems are lia-
one of the most striking demonstrations of the importance of car- ble to persist over one or more years (Ballard et al., 2001b; Draper
ers anywhere in the research literature. The market survey of Laing et al., 2001), and whilst some problems resolve, they are often
and Buisson (2010) estimated that around 40% of admissions to replaced by other forms of challenging behaviour. The relationship
care homes were known to be caused by dementia, so the true pro- between dementia severity and types of BPSD is not straightfor-
portion may be even larger. In people without dementia, perhaps ward. It is generally held that activity disturbance (agitation) and
unsurprisingly, physical ill health appears to be the most influential aggressive behaviour are more common amongst residents with
factor (Bharucha et al., 2004). severe dementia, but it is also clear that the underlying mecha-
Most of the evidence about moving into care from older people nisms and predictive factors are incompletely understood (Serra et
themselves is in the form of qualitative accounts; see, for example, al., 2010; Proitsi et al., 2011); the course of BPSD over time is also
the My Home Life programme (Owen and National Care Home highly variable (Garre-Olmo et al., 2010).
CHAPTER 27 care homes for older people 345
All older/YPD
residents:
418,000
£13bn Residents with diagnosable
dementia:
248,000
Residents with
£8bn
dementia Residents in care homes
as a known and units dedicated to
cause of dementia: 81,500; £2.6bn
admission:
142,000; £4.7bn
Fig. 27.1 Numbers and annual costs of older and physically disabled
people with and without dementia in residential settings, public and
independent sectors combined, UK, 2009.
(Source: Laing and Buisson, 2010.)
Although dementia is understood to be a progressive and largely to placebo (Schneider et al., 2005) and conventional antipsychotics
irreversible condition, evidence suggests that care home admission also have similar risks (Wang et al., 2005). Furthermore, there are
does not hasten health-related decline, at least in the short term. In concerns that antipsychotics are used as a convenient substitute for
fact, as a consequence of identifying and treating hitherto unknown good quality care. Banerjee (2009), in a report for the Department
medical conditions and good nutrition, the overall health of some of Health in England, analysed existing data to show that approxi-
demented residents may improve. Buttar et al. (2003) found that, mately 180,000 people with dementia were being prescribed antip-
of those patients with severe cognitive impairment, 14% showed sychotics in 1 year, and this probably resulted in an additional 1800
improvements in the 6 months following admission, apparently deaths and an additional 1620 cerebrovascular adverse events.
due to better and more regular diet, treatment of physical illnesses, Antipsychotics are, however, only modestly effective in treating
and antidepressant treatment. Furthermore, whilst some people behavioural and psychotic symptoms in dementia. Risperidone
with dementia are physically frail, a significant number—who tend and olanzapine are effective in reducing aggression and risperidone
to be admitted as a consequence of behavioural problems such as is effective in reducing psychotic symptoms (Ballard and Waite,
aggression and wandering—are relatively robust. In fact, Magaziner 2006), but, because of the high levels of adverse events, they should
et al. (2005) found that newly admitted nursing home residents only be used if there is marked risk or severe distress. Effectiveness
with dementia were more physically fit and had lower rates of appears to be limited in practice too, as high levels of psychiatric
mortality than residents without dementia. Although other studies symptoms persist in nursing homes despite treatment (Draper
(e.g. Landi et al., 1998) have found an association between severe et al., 2001). Withdrawing patients from antipsychotics is usually
dementia and mortality, this probably results from limited use of well tolerated and adverse effects on behaviour are less common
medical services and medication for this population. It may also than might be expected (Ruths et al., 2008).
reflect a ‘palliative care approach’ with less aggressive treatment of Prescribing of antipsychotics remains controversial. Reducing
people with dementia (Burton et al., 2001). the rates of prescribing has proved difficult, despite the evidence
of harmful side effects and the emphasis placed upon the issue by
Interventions for dementia government policy (Department of Health, 2009). There are vari-
It is usually recommended that nonpharmacological interventions ous reasons for this, although the main one is probably that BPSD
for behaviour problems in dementia are considered ahead of drug in people with severe dementia are difficult to manage and treat
treatments, as they are less likely to have serious side effects. Several (Restifo et al., 2011).
reviews have examined the management of challenging behaviour Other types of medication have been evaluated in treating
in dementia, with some concentrating specifically on care homes behaviour problems in people with dementia, and this literature is
(Allen-Burge et al., 1999; Camp et al., 2002; Snowden et al., 2003; reviewed elsewhere in the book. Relatively few studies have looked
Landreville et al., 2006). This subject is discussed in Chapter 21 of specifically at care home residents. Most of these studies have meth-
this book. odological shortcomings, most often small sample sizes, so the find-
There have rightly been concerns about excessive prescribing of ings should be regarded as provisional. Although trazodone is often
psychotropic drugs, especially antipsychotics, to older people with used for the treatment of agitation in dementia, the evidence for
dementia in care homes. Many studies have shown high propor- its effectiveness is not compelling (Martinon-Torres et al., 2004).
tions of residents receiving antipsychotics, usually between 25 and Drugs licensed for the treatment of Alzheimer’s disease, cholineste-
40%. Most of these patients have dementia, although a proportion rase inhibitors, and memantine do not seem to be especially helpful
have other mental disorders, notably schizophrenia (Snowdon et al., in the treatment of agitation (McShane et al., 2006; Howard et al.,
2005). Antipsychotics have many side effects, e.g. sedation, impaired 2007). It is interesting to note that within 3 months of nursing
mobility, and increased risk of stroke. There is a small but signifi- home admission, about half of residents who were previously on
cant increased risk of death with atypical antipsychotics compared antidementia drugs have had them stopped or the dose reduced
(Parsons et al., 2011). The effects of this, beneficial or otherwise, more about what may be a mixture of transient symptoms, e.g.
have not been studied in any detail to date. Retrospective data have those associated with a move from a person’s own home into a care
suggested some evidence of cognitive and functional decline after home, and more chronic levels of morbidity.
treatment cessation (Daiello et al., 2009), but this may be contami- In treating depression, serotonin reuptake inhibitors are pre-
nated by other factors, so we remain uncertain as to the optimum scribed more commonly than tricyclics. When tricyclic antide-
level of use of antidementia drugs in care homes (Seitz et al., 2009). pressants are used, it is often for indications other than depression,
As mentioned in the section Physical health of care home resi- such as insomnia, pain, or itching (Borson et al., 2002). The treat-
dents, pain is sometimes a potent cause of agitation and behaviour ment offered is often not adequate, e.g. subtherapeutic doses of
change, and clinical trials of effective pain relief have had promising antidepressants or being offered anxiolytics and hypnotics instead
results (Husebo et al., 2011). (Brown et al., 2002). Nonetheless, prescribing of antidepressants
has increased dramatically in several countries, in part due to
Depression concerns about antipsychotics, but probably also in recognition of
The first study of depression in residential homes (Mann et al., 1984) the importance of treating depression in improving quality of life
found that 38% of a residents in a sample of over 400 were depressed. (Arthur et al., 2002; Datto et al., 2002). Unfortunately, prescribing
Depression was particularly associated with visual impairment and may be influenced as much by the characteristics of the home as by
incontinence, increased dependency, problem behaviours such as the clinical needs of residents (Lapane and Hughes, 2004).
wandering and aggression, having been admitted from one’s own Other psychosocial interventions, including care staff training
home, and belonging to a minority religion. A follow-up study (Eisses et al., 2005) and interventions aimed at secondary preven-
after 3.6 years (Ames et al., 1988) found that two-thirds of the resi- tion (Cuijpers and van Lammeren, 2001), often show encouraging
dents had died, but, of those surviving, only 17% had recovered results, including better recognition of depression, but it is difficult
from depression. Interventions for depression were often difficult to demonstrate positive outcomes in terms of lower levels of the
to implement and their outcomes were not particularly successful condition. Multifaceted interventions involving elements of educa-
(Ames, 1990). Since then, numerous studies have reported similar tion, staff training, additional activities, and psychological treat-
findings, with an overall median prevalence for major depression of ment (e.g. Llewellyn-Jones et al., 1999; Hyer et al. 2008) have shown
10% and depressive symptoms of 29% (Seitz et al., 2010). Functional modest improvements in depression scores, but the number of suc-
impairment appears to be an important independent risk factor for cessful studies is limited, and it may be only a minority of residents
depression. Other associations include physical health variables, who are able to benefit (Dozeman et al., 2011).
such as pain, dysphagia, and heart disease; psychological variables, Depression remains a significant problem for care home resi-
such as loneliness and neuroticism; and social variables, such as dents. Improving the situation will require improved detection and
loss and the institutional environment. increased awareness of the importance of depression among care
Assessment of depression in care homes may be difficult for sev- home staff (Dow et al., 2011). More research is needed to evaluate
eral reasons, and depression is often not recognized by care staff those interventions that appear promising, and longitudinal studies
(Bagley et al., 2000; Brühl et al., 2007). A major challenge is in are also needed to inform us about the incidence, persistence, and
diagnosing depression in residents with severe dementia—indeed, outcomes of depression.
prevalence studies often exclude up to a third of residents because it
has been found too difficult to diagnose depression with any accu- Other mental health problems
racy. Further, care staff may lack the knowledge and skills to rec- Although dementia and depression are the commonest mental
ognize depression, especially if the main symptoms are changes in health problems in care home populations, other disorders may be
behaviour, such as aggressive behaviour or disruptive vocalization present too. Sleep disorders are discussed in Chapter 51. Anxiety
(Dwyer and Byrne, 2000; Menon et al., 2001; Bartels et al., 2003a). symptoms are also common, though their severity varies, and anxi-
There may also even be financial disincentives to recognize the ety may be symptomatic of other disorders, such as depression or
presence of mental disorders (Snowdon, 2005). These combined physical illness. Smalbrugge et al. (2005) found that, in a sample
findings have led to the suggestion that residents with dementia of over 300 patients, 5.7% had anxiety disorders, 4.2% had sub-
should be routinely screened for depression (Cohen et al., 2003). threshold anxiety disorders, and nearly 30% had anxiety symptoms.
In a comparison of screening methods (Watson et al., 2009), a Anxiety disorders were associated with health-related variables,
two-item scale, the PHQ-2 (Kroenke et al., 2003), performed best notably depression and stroke. There is a lack of treatment studies
against a gold standard psychiatric interview, with a sensitivity of for anxiety disorders.
0.80 and a specificity of 0.71. The number of older people with severe long-term mental ill-
Identifying depression is important as it has a potent effect nesses, such as schizophrenia and bipolar disorder, has risen in
on wellbeing (Smalbrugge et al., 2006) and is implicated in sui- many developed countries. This is a consequence of the fact that
cides and attempted suicides in both care home residents and they now live into old age due to improved access to life-long medi-
the community-based older population (Scocco et al., 2006). The cal care; they are also more visible, especially since the closure of
course of depression in homes is not well understood. However, large psychiatric hospitals. As well as this, middle-aged and older
it does appear that residents with higher depression scores have people with schizophrenia are much more likely to move into resi-
a higher death rate (Barca et al., 2010), and that almost a half of dential and nursing home care at an earlier age than people without
people depressed on admission to a care home are still depressed 9 previous mental illness. For example, in one study (Andrews et al.,
months later (Sutcliffe et al., 2007). The incidence of new episodes 2009), rates of admission started to diverge significantly between
of depression is about 5% per annum (Smalbrugge et al., 2006). In the ages of 40–65, where nursing home admission risk was 3.9
general, more longitudinal studies are needed: we need to know times greater than for peers with no mental illness. This may have
to do with the loss or burnout of family carers, often ageing parents. Physical health of care home residents
Cognition functioning, and behaviour were important predictors
Comorbidity
of nursing home as opposed to community residence, with cogni-
tive impairment especially important, as it was also a predictor of Care home residents are a frail group, with high levels of physi-
a person’s level of functioning (Bartels et al., 1997). Among older cal and mental health conditions. Chronic illnesses and disabilities
patients with schizophrenia (age range 40–97), residence in assisted are common, frequently with evidence of unmet need or inad-
living facilities was also associated with never having been mar- equate care. For example, although about 80% of residents have
ried, the presence of cognitive impairment, and poorer quality of some degree of hearing impairment, care staff only recognize the
wellbeing (Auslander et al., 2011). Such patients often fit poorly problem in a minority of cases, and provision of hearing aids is
into conventional nursing home environments, and other models consequently low (Cohen-Mansfield and Taylor, 2004). In another
of care, e.g. separate units or supported community living, may study, around 9% of residents had diabetes, but fewer than half were
be preferable (Dencker and Gottfries, 1991; Bartels et al., 2003b). regularly reviewed by a GP or practice nurse (Taylor and Hendra,
Depla et al. (2003) reviewed six Dutch programmes that employed 2000). Even fewer (less than 10%) are reviewed in specialist clinics
different models to cater for older people with chronic mental ill- (Sinclair et al., 1997). Complications of diabetes are common, and
nesses. Overall, the favoured model was deploying psychiatric staff in one study it was found that over half the diabetic residents were
to work with care home staff, rather than having a psychiatric unit in pain on admission (Travis et al., 2004). Cancer is also underdiag-
within the home or the home itself employing its own psychiatri- nosed in some homes, possibly more so in rural areas (Dobalian
cally trained staff. et al., 2003).
Substance misuse remains a neglected area in relation to care Multiple physical problems and/or comorbidity with mental dis-
homes (Joseph, 1995). A study of over 300 residents in north- orders are common. Medical conditions and associated disability
ern France (Leurs et al., 2010) found that two-thirds of residents are the main reasons underlying admission to care homes, rather
drank alcohol on a daily basis and nearly 20% were heavy drink- than nonspecific frailty or social needs. In one UK study of 16,000
ers. In Veterans Administration homes in the US, almost half residents, over half had dementia, stroke, or other neurodegen-
of residents had a lifetime history of significant heavy drinking erative disease, 78% had at least one form of mental impairment,
(abuse or dependence) (Joseph et al., 1995). Later studies showed 71% were incontinent, and 76% needed help with mobility or were
that residents with substance misuse disorders tended to be immobile (Bowman et al., 2004). Overall, 27% were immobile, con-
younger and were more likely to be male and of Afro-American fused, and incontinent. There is considerable overlap between care
origin than other residents. They also were more likely to have homes with and without nursing as to the levels of disability and
psychiatric comorbidity, though they were more independent physical dependency. This may result from changes in residents’
with activities of daily living (Brennan and Greenbaum, 2005). physical health after admission to the home (Rothera et al., 2003),
This distinct profi le leads to challenges in providing care, but but there may also be inconsistencies in the assessments made
also perhaps offers some potential for rehabilitation (Lemke and before admission. It also calls into question the validity of the dis-
Schaefer, 2010). tinction between residential and nursing homes.
Older people with intellectual disabilities (ID) have particular However, there are opportunities for improving health care for
needs in relation to residential care. As is well known, there is an older people when they move into a care home. There should have
increased risk of dementia at younger ages, with certain condi- been a thorough assessment of their needs prior to the move and
tions causing ID, notably Down’s syndrome (Holland, 2000). The this can be repeated, as they will often now be registered with a
diagnosis of dementia can be difficult to make as it relies on func- new GP. This provides the chance to reappraise any long-term con-
tional decline and/or other changes, e.g. in personality. At the same ditions and the objectives of any treatment. This of course should
time, family carers are often themselves becoming old and less able include a review of medication, aiming to simplify it where possible.
to support their adult children with ID. As a result, the move of It is also important to communicate realistically with residents and
a person with ID into residential care is often a result of a social their families about what expectations may exist about the medical
change, such as the death of a parent, rather than a change in the care of the older person. The physical examination of frail older
needs of the person him/herself. This may be associated with the people is discussed in Chapter 11.
double impact of a bereavement as well as loss of previous famil- Pain
iar surroundings. People with ID may be placed in residential care Pain is common, from arthritis and pressure areas especially.
for older people at a relatively young age, which may also cause Although reported prevalence rates vary widely, depending on how
problems for them in terms of social isolation (Winkler et al., 2006) the presence of pain is ascertained and the sample under consid-
and other residents who may not understand or tolerate their needs eration, half or more of care home residents have been identified
or behaviour (Wilkinson, 2004). Furthermore, the staff group may as experiencing one or more painful conditions. Pain can often
be unskilled in meeting their needs. Improving the lives of older make individuals restless, agitated, or even depressed (Cipher and
people with ID who are placed in mainstream older people’s care Clifford, 2004). People who are already depressed are perhaps more
homes requires action on several fronts. These include: ensur- sensitive to pain and are more likely to complain of it (Parmelee
ing that services for people with ID are better equipped to meet et al., 1991). Residents with dementia may be less likely to indicate
age-related needs; preventing people from entering older people’s that they are in pain and thus may not be offered analgesia as fre-
services that are unable to offer them an appropriate level of care quently as those without impaired communication (Nygaard and
and quality of life; and reviewing the appropriateness and quality of Jarland, 2005). There is evidence that improved pain management
placements of all people with ID in residential and nursing homes may benefit the health and quality of life of residents with dementia
for older people (Thompson and Wright, 2001). (Chibnall et al., 2005; Husebo et al., 2011).
Better understanding and improved management of pain in resi- nursing home (Gessert et al., 2000; Mitchell et al., 2003). Feeding
dents with dementia requires wider use of pain rating scales, includ- tubes do not prolong survival in dementia and may be associated
ing those with nonverbal items (such as facial expression) that can with significant complications (Mitchell et al., 1997; Murphy and
be used for patients with language impairment (see Zwakhalen Lipman, 2003), so the persistence of the practice is a matter for con-
et al. (2006) for review of scales). There is also a need for explora- cern, especially as good palliative care can be effective in reducing
tion of the relevance of pain in different forms of dementia since, the use of feeding tubes (Monteleoni and Clark, 2004).
for example, patients with vascular dementia are prescribed analge-
Death and end of life care
sics more frequently than those with Alzheimer’s disease (Scherder
et al., 2005). Given the advanced years and the prevalence of serious health
problems among care home residents, mortality rates are bound to
Mobility and falls be high. The baseline 1-year survival rate for people in UK care
Mobility problems are common in the care home population homes is 66% overall; it is 59% in nursing homes, where the median
and most residents have some limitation of mobility (Bowman survival is about 18 months (Rothera et al., 2002). Care home resi-
et al., 2004; Williams et al., 2005). Thapa et al. (1995) examined dents are often admitted to hospital, notably for medical illnesses
risk factors for falls among 282 residents in 12 homes. Individual such as infections and heart failure (Bowman et al., 2001). Godden
independent risk factors were being aged 75 or over, impaired func- and Pollock (2001) compared emergency hospital admissions
tional ability, balance problems, a fall in the preceding 90 days, and among care home residents with those for older people living in the
behaviour problems. The presence of all five risk factors was associ- community, and found higher risks of admission for all diagnoses
ated with a ten-fold risk of falling. Psychotropic medication also (relative risk = 1.4) and especially for injuries and fractured neck
had an independent effect, approximately doubling the risk—and of femur (relative risk = 4). Care home residents also had a higher
accounting for 36% of the attributable risk of falling—in those resi- risk of dying in hospital, especially within 48hours of admission
dents receiving such medication. Although antipsychotic drugs are (relative risk = 3.6).
probably a key causal factor, selective serotonin reuptake inhibitors Deaths often occur in care homes, with about 20% of deaths in the
also increase the risk of falling and causing injury (Arfken et al., UK occurring in nursing and residential homes (National Council
2001). The presence of dementia approximately doubles the risk of for Palliative Care, 2006). The proportion increases to nearly 30%
sustaining a significant injury (van Doorn et al., 2003). in people over the age of 85 (Fleming et al., 2010). Outcomes of
Various attempts have been made to improve mobility and reduce emergencies, such as cardiac arrest, in care homes are very poor,
the rate of serious falls, including exercise programmes and occu- and it has been argued that there is little point in providing cardi-
pational therapy interventions. The use of hip protectors does not opulmonary resuscitation in most long-term care settings (Conroy
seem to reduce the rate of hip fractures in nursing homes (Parker et al., 2006). Deaths are, however, usually heralded by more grad-
et al., 2006). Two randomized controlled trials of multifactorial falls ual decline, often with increased utilization of healthcare services
prevention programmes both had nonsignificant findings in favour (Bercovitz et al., 2005), and changes in symptoms and mental states
of the interventions (Ray et al., 1997; Dyer et al., 2004). Other stud- of residents are often seen in the last months of life. Depression is
ies have produced mixed outcomes (Ward et al., 2010). often overlooked and undertreated during the last months of life
Nutrition and fluids (Evers et al., 2002), though whether depressive symptoms actually
increase during the last months of life is debatable (Cuijpers and
Concern is often expressed about poor standards of nutrition in care
van Lammeren, 2001). Risk scores calculated from clinical data can
homes. Food may be poorly presented or unappetising, the envi-
predict the risk of mortality within 6 months (Mitchell et al., 2004a)
ronment may be uncongenial, special diets (for religious observ-
and thus may help to reduce the frequency of inappropriate trans-
ance or medical conditions) may be unavailable, and residents often
fers to hospitals (Lamberg et al., 2005).
have conditions that may affect their ability to feed themselves or
Management of dying patients in care homes, especially those
to swallow their food. In a US study of 407 people with demen-
with dementia, varies considerably and is often inadequate (Mitchell
tia in 45 assisted living facilities and nursing homes, 54% had low
et al., 2004b). However, there has been an explosion of interest in
food intake and 51% had low fluid intake (Reed et al., 2005). Staff
palliative and end of life care in dementia (Hughes, 2010) and this
monitoring of residents, having meals in a public dining area, and
topic is discussed in more detail in Chapters 28 and 57.
the presence of noninstitutional features were each associated with
higher food and fluid intake. Some homes have paid considerable
attention to the eating environment, e.g. ensuring that mealtimes The experience of living in a care home: quality of life
are undisturbed by televisions, visitors, and domestic bustle, and and quality of care
that food is served in a family style rather than in an industrial Recent work has begun to pay much more attention to the experi-
canteen (Nijs et al., 2006). ence of living in a care home. This includes systematic projects such
The use of feeding tubes (percutaneous endoscopic gastrostomy as My Home Life (Owen and National Care Home Research and
(PEG) tubes) in care home residents is variable across the world Development Forum, 2006) and a variety of other initiatives and
but especially prevalent in the US, perhaps for defensive reasons. accounts. There are, broadly, four sources of evidence about ‘the
Over one-third of severely cognitively impaired residents in the US, care home experience’. The first relies on research tools or measures
more than 63,000 people, had feeding tubes, according to a study by to evaluate objectively the quality of life (QoL) and/or of care home
Mitchell et al. (2003). Certainly, clinical factors, such as dysphagia residents with dementia. The second is observational approaches,
and history of stroke, increase the likelihood of tube placement, but prominently dementia care mapping (DCM). The third source incor-
other, nonclinical variables appear to be significant also, includ- porates focus groups and interviews, some of which may include a
ing the person’s ethnic origin and the nature and location of the questionnaire and/or rely on third party perspectives—primarily
family carers (Nolan et al., 2003). The fourth, newest, approach an effective way of maximizing the data thus collected (Cook, 2002;
focuses specifically on the subjective experience of the person with Hyden and Orulv, 2009).
dementia (Clare et al., 2008). A large study by the Alzheimer’s Society gathered data from
As regards instruments to measure QoL, a number of scales do 4,084 family carers, care home managers, and workers, using a
exist, some specifically developed for people with dementia, e.g. the mixture of focus groups and interviews; residents were not directly
Quality of Life in Alzheimer’s Disease scale (QoL-AD) (Logsdon included. The study found that people with dementia are not always
et al., 1999; Hoe et al., 2005), the DEMQOL (Smith et al., 2005), and afforded dignity and respect, and expectations of QoL tended to be
the DQol (Brod et al., 1999). These instruments vary considerably in low (Alzheimer’s Society, 2007). Lack of activity and stimulation
nature and content, but common domains include physical function- were also highlighted, with 54% of carers reporting that their rela-
ing, cognitive abilities, ability to participate in meaningful activities, tive was left in his/her room for hours with no attempt from staff to
and mood (Warner et al., 2010). The objectivity of this approach is, engage with the relative. These problems were particularly acute for
however, only relative. There are often differences between observer people with more severe dementia. As the availability of activities
(staff ) ratings and residents, at least partly due to a difference in and opportunities for occupation is a major determinant of QoL,
emphasis: one study found that residents’ QoL scores were most this is a significant concern. Related work has identified the need to
affected by the presence of depression and anxiety, whereas staff rat- ensure residents retain the right to make choices and are accorded
ings were more associated with dependency and behaviour problems some degree of autonomy over decisions that affect their daily life
(Hoe et al., 2006). Caregiver ratings seem to be affected by their own and wellbeing (Train et al., 2005).
attitudes towards dementia care (Winzelberg et al., 2005). Despite Another variation on interviews is the use of ‘talking mats’, a
these measurement difficulties, evidence suggests that QoL in care low-tech framework comprising a textured mat and visual symbols,
homes is largely determined by the existence of mental health prob- allowing residents with limited verbal skills to express their views.
lems and subjective wellbeing (e.g. Samus et al., 2005; Smalbrugge Whilst not a research or interview tool per se, early work suggests
et al., 2006). Systematic assessment of residents’ needs, and consid- it can positively influence the quality of interaction between staff
eration of whether or not they have been met, has been suggested as and residents and enhance staff ’s ability to communicate effectively
a means of improving QoL (Hancock et al., 2006). The Camberwell and to involve residents in decisions about their care (Macer and
Assessment of Need for the Elderly (CANE) (Orrell and Hancock, Murphy, 2009; Murphy et al., 2010).
2004) has been used towards this end. Addressing wellbeing has also Finally, more direct attempts to capture the subjective experience
been highlighted (Brooker, 2011). of people with dementia in care homes have attracted increasing
When people are very dependent and live in a care home, QoL attention. These range from creative projects such as those of John
becomes inextricably linked to quality of care. Quality of care, like Killick, who creates poems from the unstructured conversations
QoL, is also a concept on which there is incomplete agreement. of people with dementia (e.g. Killick and Martin, 2011), to more
For example, there are multiple perspectives to be accommodated directly biographical studies (Surr, 2006). The latter work showed
(Innes and Kelly, 2007). Also, ‘care’ includes not only those ele- how people with varying severity of dementia constructed stories
ments provided within the home but also external services, such about their lives, in terms of themes such as relationships, occupa-
as medical care and social support. Furthermore, good quality care tion/social role, family role, and caring for and being cared by oth-
depends on a range of micro-level (satisfied staff ) and macro-level ers. Clare (2008) applied an integrative phenomenological approach
(financial stability of the provider) factors and their interaction; to the same dataset, exploring the themes that people used to make
this make its assessment complex and multifaceted. sense of their worlds. The themes that emerged were a mixture of
However, aside from these conceptual issues, an important con- isolation and frustration, together with coping, ‘making the best of
tribution has been the development of DCM, an observational oneself ’, and maintaining identity.
approach where items recording residents’ activity can be com- From the above, it is likely that to obtain a full picture of QoL
bined to calculate a dementia care index score (Brooker, 2008). requires a combination of measures, incorporating observations of
Its person-centred basis, relatively widespread use in care set- residents as well as the views of residents, carers, and staff (Brooker,
tings, and validity make it a popular and reliable tool to assess 2011). In relation to quality of care, there is a range of general issues
and enhance quality of care (Brooker, 2011). A brief version (the that are important, but there are also some specific issues that need
Short Observation Framework for Inspection (SOFI)) has also been to be considered as they undoubtedly impact upon the lives of
developed and adopted by the Care Quality Commission’s prede- residents. We know, for example, that poor prescribing patterns,
cessor agency for use in care home inspections (Commission for covert administration of medication, and physical restraint have a
Social Care Inspection, 2008). detrimental effect on residents’ QoL. The abuse of older people is
The third approach, using interviews and focus groups, chal- addressed elsewhere in the book (Chapter 59) and concerns about
lenges the widely held view that people with significant cognitive excessive or inappropriate prescribing are discussed in our dis-
impairment are ‘unsuitable’ interviewees. Mozley et al. (1999), cussion of Interventions for dementia. Physical restraints, includ-
for example, found that a large proportion of their sample with ing clothing, cot sides, and restrictive chairs, are still worryingly
advanced dementia were able to answer questions about their QoL. prevalent and, because of staff attitudes to risk, it can be difficult to
Byrne-Davis et al. (2006), drawing on data from focus groups, reduce even in planned interventions (Sullivan-Marx et al., 1999).
found that people with moderate or severe dementia were able to Although electronic tagging has been advocated in certain circum-
talk about QoL in meaningful ways. The key issues that partici- stances to prevent wandering and to facilitate a more open environ-
pants emphasized were social interaction, psychological wellbeing, ment, it has been criticized for emphasizing technology and control
religion/spirituality, independence, financial security, and health. instead of working towards better, more individualized care (e.g.
Transcriptions of interviews and video or digital recordings can be Hughes and Louw, 2002). Administering medication concealed
in food or beverages appears to be a widespread practice in care The care home market has also become more corporate: the four
homes (Treloar et al., 2000; Kirkevold and Engedal, 2005). As many biggest care home providers supply 23% of the beds, and over half
residents with severe dementia may lack the capacity to consent, or of the independent sector capacity is owned by providers with
withhold their consent, to treatment, this measure may, however, three or more homes, the remainder being small single-home busi-
occasionally be necessary, though it must always be properly dis- nesses. The total value of the care home market in 2010 was esti-
cussed and documented between the care team and the resident’s mated at £14 billion, of which private sector operators accounted
family. for £9.9 billion (Laing and Buisson, 2010). Along with this, homes
are becoming larger, with homes with nursing being on average
twice as large as residential homes (47 compared to 19 beds) (Care
Part 2: The Care Home Sector Quality Commission, 2010). There has been considerable improve-
Care homes in the UK ment in private facilities, e.g. 93% of places in independent sector
care homes now offer single accommodation, whilst 73% of rooms
In the UK, most long-term care for older people is provided in care
have en-suite toilets (Laing and Buisson, 2010).
homes. There are currently nearly 12,000 registered care homes for
Lengths of stay tend to be shorter in nursing care beds than resi-
older people in the UK (Laing and Buisson, 2010). Care homes are
dential care beds; for local authority-supported residents compared
categorized by the type of care they provide, nursing or personal,
with self-funders; and for nonambulant residents compared with
and also by the nature of the provider—local authority (social serv-
ambulant ones. Men also tend to have shorter stays than women.
ices), independent, or voluntary sector. As care homes with nursing
Across all types of home for older people the average length of stay
support residents with higher dependency needs, their fees are con-
is around 2.5 years, with the longest surviving 10% staying for 6
sequently higher than residential care homes. Some homes provide
years or longer. Twenty-seven percent of residents live for more
‘specialist care’ for particular groups, most notably older people with
than 3 years in a care home (Forder and Fernandez, 2011).
dementia. Until recently, some NHS hospitals provided continu-
The availability of residential care varies considerably on a geo-
ing care for a small proportion of very frail older people, including
graphical basis. Inner London, where care is expensive, has the fewest
those with severe dementia. Now the great majority of ‘continuing
places per 1000 of the older population, whilst the highest propor-
care’ patients are placed in nursing homes; a small number of these
tions are in rural and coastal areas (Care Quality Commission,
are NHS run, although in most cases the NHS buys beds in private
2010). Shortages in providing care for people with dementia have
nursing homes (Dening and Milne, 2011). In April 2010 only 3% of
been widely noted in national studies; it has been identified as a par-
all long-term care was ‘continuing care’ in hospital.
ticular challenge in London, the southeast, and southwest of England
The independent care home sector expanded greatly in the
(Wanless et al., 2006; Care Quality Commission, 2010). Patterns of
1980s and early 1990s, with an ageing population and open-ended
care home development and closure are also uneven across the UK.
income support funding. In 1993 funding was transferred from the
For example, between 1991 and 2001 the number of care homes in
state to local authorities, leading to a decline in bed numbers as
the northwest of England fell by over 20% and in Wales by 18%, while
budget-constrained local authorities applied needs assessments for
the number of care homes in the east of England showed little change
the first time. Although the level of dependency of residents has
and the number in Scotland rose by 10% (Banks et al., 2006).
increased, the proportion of residential places funded has remained
steady, probably reflecting the imperative for local authorities to Extra care housing
save money and to place an older person in the least expensive care Extra care housing (ECH) refers to accommodation with
facility. In general, demand has been passed down the continuum self-contained housing units, 24-h staff on site, with domestic
of care, i.e. those who would previously have been admitted to a care and communal facilities available to residents. This model
nursing home are being placed in a residential home and those enables older people to safeguard their capital by purchasing
who would have been admitted to a residential home are remaining or part-purchasing their accommodation; and care services are
in the community (Laing and Buisson, 2010). Despite the policy available on site if needed (Evans, 2009). Despite its popularity in
emphasis on nonresidential alternatives, evidence suggests that theory, it is estimated that the total current supply of ECH is only
demand for care home places has stopped declining, presumably 40,000 places and that a little over 1% of the UK population aged
due to the complex needs profile of most care home residents and 75 years and over occupies an extra care unit (Laing and Buisson,
the limits of nonresidential alternatives (such as extra care housing; 2010). One of the main financial drivers for it—the Department of
see section Extra care housing). Health’s Extra Care Housing Initiative Fund—expired at the end of
Over the past 30 years there has been a significant shift in the bal- 2010 and has not been replaced (Laing and Buisson, 2009).
ance of provision in the UK. Care home services have transformed Although it was hoped that ECH would offer a viable alterna-
from a predominantly public sector activity in the mid-1970s to a tive to care homes, it is unclear to what extent this is possible,
predominantly private sector activity now. In 2010 the private sector particularly for people with complex needs including dementia
accounted for 76% of all long-stay bed capacity; 14% was provided (Bäumker et al., 2010). Recent research suggests that ECH has
by the voluntary sector and just 10% by the public sector (Laing and a positive impact on quality of life and health for most residents
Buisson, 2010). This trend looks likely to accelerate as councils seek and may be cost effective relative to a care home, but most of these
to achieve economies in the face of public expenditure cuts and the outcomes are medium rather than short term (Netten et al., 2011).
NHS continues to reduce its provision of long-stay beds (Office of In a climate of financial austerity, investment in ECH provision
Fair Trading, 2005). In addition, between 1987/88 and 2009/10 there seems unlikely if decision-makers are focused on immediate cost
has been a 60% reduction in the numbers of NHS overnight beds in savings, and partners—housing, health, social services, and ECH
both geriatric and psychogeriatric settings (Lievesley et al., 2011). providers—are not incentivized to plan and work together.
Funding and paying for care the state, protecting those with the fewest resources and capping
The financing of long-term care has, and continues to be, an issue the total amount any individual should be expected to pay: these
of considerable public debate. One of the main issues is the histori- principles have broadly been accepted with the cap being set at
cal distinction between ‘nursing care’—paid for by the NHS—and around £61,000. Additionally, a national minimum eligibility cri-
‘personal care’—means tested and partially or wholly paid for by teria will make access to care more consistent around the country
the individual. This remains one of the most contested aspects of and there will be a new legal right to financial protection from very
the funding terrain and is a source of considerable confusion and high care costs.
geographical inequity. NHS funding eligibility is applicable where
the older person’s needs are regarded as ‘primarily health related’. It Regulation and registration
is noteworthy that such a distinction is not made in Scotland, where All care homes have to be registered with the independent regula-
all long-term care is state funded (Community Care and Health tor of health and social care. In England this is the Care Quality
(Scotland) Act, 2002). Commission (CQC). Corresponding agencies in the rest of the UK
The cost of supporting an individual in long-term care is expen- are: the Care and Social Services Inspectorate Wales, the Social
sive for both the individual and the state; it is also rising as a conse- Care and Social Work Improvement Scotland, and the Regulation
quence of the increasing number of residents with complex needs. and Quality Improvement Authority in Northern Ireland.
Care home fees are estimated to be about £30,000 per annum; in The CQC was established in 2009, introduced by the Health and
2010 the weighted average fees for older care home residents were Social Care Act 2008, and replaces its predecessor, the Commission
£693 per week for nursing care and £498 per week for residential for Social Care Inspection. The new system creates for the first
care (Laing and Buisson, 2010). The highest fee rates, predictably, time a common regulatory framework for both the independent
are in the southeast of England and the lowest are in the north of and public sectors and across both health and social care services.
England and Wales (Featherstone and Whitham, 2010). Just over The National Minimum Standards (NMS) (see below) have been
half the residents (51%) are supported in part or in full by local replaced by generic ‘essential standards’ of safety and quality. There
authorities; 40% of residents are self-funding; and the remainder are considerably fewer requirements and CQC will undertake a
(approximately 8%) are supported by the NHS (Bäumker and much smaller number of on-site inspections, especially for those
Netten, 2011). There has been an increase in the proportion of care homes previously deemed to be ‘excellent’. It will depend more
NHS-funded residents since the eligibility criteria for NHS contin- on ‘paper’ exercises in which providers make available their own
uing care have been widened, as well as outsourcing of continuing evidence of compliance with the essential standards. The star rating
care to nursing homes (Age Concern, 2010). scheme, introduced in 2008 and popular with the public, no longer
An increasing proportion of people find themselves in the exists and has yet to be replaced with an alternative quality rating
self-funding category due to an increase in the rate of home owner- scheme. Critics are concerned that this ‘light touch’ approach to
ship amongst the very old, the population most at risk of entering a regulation will not prove robust enough to protect standards of care
home. Another factor may be more rigorous vetting by councils of or residents from the negative impact of poor practice or abuse.
situations where individuals may have divested themselves of assets They also argue that whatever the flaws of the previous system,
in order to qualify for public support, e.g. transferring the title of their minimum standards had a predominantly positive effect on care
property to family members. To be eligible for local authority fund- home quality and provided a universal benchmark against which
ing, an older person must have capital assets of less than £23,500. homes could be judged.
It is perhaps no surprise that the income and wealth of those over Since the introduction of NMS, care homes for older people have
60 correlates inversely with dependency: in other words, those most made a large improvement, meeting over a quarter (27%) more
in need of care and support tend to have fewer private resources; standards in 2010 (the last year of use of NMS) than in 2003. NMS
they are less likely to be homeowners or have occupational pensions applied to seven key areas: choice of home, health and personal care,
(Wanless et al., 2006). National figures hide wide local variations, daily life and social activities, complaints and protection, environ-
with a higher proportion of private funders in affluent areas of the ment, staffing and management, and administration. Areas in which
country, e.g. 54% in the southern home counties in 2010. Self-funders the greatest improvement has been made since 2003 are: informa-
often pay significantly higher fees—£50–100 extra per week—than tion (58% improvement), quality assurance (51% improvement), and
those funded by the local authority for similar services and ameni- staff supervision (44% improvement). Areas where they consistently
ties. The former group are therefore effectively cross-subsidizing the achieve very high ratings include: community contact (97%), rights
latter group, often within the same care home (Office of Fair Trading, (96%), and autonomy and choice (93%). They perform least well in
2005). Furthermore, self-funding residents are often admitted at relation to record keeping (67%), medication (72%), staff supervision
lower levels of dependency (Bäumker and Netten, 2011). (72%), and service user plans (73%). In 2010, private sector homes
Funding for long-term care is a persistent challenge in the UK met an average of 85% of standards, whilst those run by voluntary
(and internationally) and successive governments have failed to agencies met an average of 90% (Laing and Buisson, 2010). In general,
address it coherently. In February 2013 the English Department nursing homes performed better than residential care homes. Ratings
of Health announced new measures for care funding based on the were significantly lower in the 1500 care homes with a dementia spe-
recommendations of the ‘Dilnot report’ on Fairer Care Funding ciality than in other care homes. Smaller care homes are slightly more
(Commission on Funding of Care and Support, 2011). The report likely to be judged as good or excellent (87%) than medium sized
highlighted the fact that the current funding system is opaque, (84%) or large homes (81%) (Care Quality Commission, 2010). An
inequitable, and unsustainable. It proposed a ‘partnership’ model additional fifth (18%) of care homes were judged to be good or excel-
that shares responsibility for funding care between the citizen and lent in 2010 (85%), compared with 2008 (67%).
UK-wide research suggests that the quality of care older people between the quality of staff training and development and the well-
receive is significantly influenced by where they live; this is par- being, quality of life, and social engagement levels of residents with
ticularly the case for specialist homes. For residents with dementia, dementia (Alzheimer’s Society, 2007; Commission for Social Care
certain factors appear to be particularly important for enhanc- Inspection, 2008). Although initiatives targeting front-line staff
ing wellbeing: the reassurance of daily routine; privacy, dignity, can be effective, evidence strongly indicates that change needs to
and choice; contact with family and friends; and the availability be embedded across the staff team—including managers—and the
of pleasurable activities (Alzheimer’s Society, 2007). It is widely whole organization to have sustained impact (Loveday, 2011).
acknowledged that regulatory systems are especially challenged by Staff quality is regulated by both the NMS for care services and
gathering meaningful data about ‘what matters’ to residents with the Health and Care Professions Council for the training and con-
dementia. This is an issue that has been addressed, at least in part, duct of social care staff. Although the picture is one of gradual
by the routine use of the SOFI tool (see section The experience of improvement, in 2009 10–15% of care homes had still not met the
living in a care home: quality of life and quality of care) during 2002 NMS for ‘qualified staff ’, i.e. that at least 50% of care staff are
care home inspections (Commission for Social Care Inspection, trained (National Vocational Qualification (NVQ) Level 2 or equiv-
2007, 2008). alent) and that all registered managers hold a relevant qualification
(NVQ Level 4 in Management or Care, and a nursing qualification
Care home workforce if appropriate) (Laing and Buisson, 2009). A higher proportion of
Care homes constitute a substantial sector of the health and social private sector homes had not met the standard than their public
care economy. The employer-led agency, Skills for Care, estimated or voluntary sector counterparts. Dementia training is often frag-
that there were over half a million people employed in care homes mented and ad hoc, even in specialist care homes. Indeed, recent
in England (Eborall et al., 2011); 270,000 in residential homes and evidence suggests that one-third of dementia care homes provide
242,000 in nursing homes (though this figure includes all care no dementia training at all (All-Party Parliamentary Group on
homes for adults, not just older people). Of these, approximately Dementia, 2009). The Health and Social Care Act 2012 has removed
75% are front-line care staff: the remainder are managers, domes- the requirement for specific qualification standards in care homes
tic staff, administrative staff, and workers in ancillary roles such as in England, raising concerns about how training and skills will be
maintenance. As might be expected in nursing homes, there is a promoted nationally.
slightly lower proportion of care workers than in residential homes, Care homes are only legally required to provide sufficient num-
with about a sixth (17%) of jobs being registered nurses. bers of suitably trained and capable staff to meet the needs of their
The care home workforce is almost exclusively female, many staff residents, and no ratio of staff to residents is specified in official
are part time, and most are low paid (Eborall et al., 2011). Turnover guidance. Netten et al. (2001) reported that in residential care homes
and vacancy rates in care homes are variable across the country, with there was approximately one full-time member of care staff for
some areas having major difficulties with recruitment and reten- every three residents and one part-time worker for every 2.5 places
tion, e.g. London and southeast England. The vacancy rate for care However, in a study of dementia care homes, the most common
workers was 4.0% in 2010 and the turnover rate was 22.8% (about care staff/resident ratio was just one in four (Cantley and Wilson,
one in four workers leaving per year). A recent survey by the Royal 2002). The RCN survey (2010) found that staffing ratios were not
College of Nursing (RCN, 2010) estimated that care home staff are always maintained at sufficient levels to provide personalized care,
in post for an average of just under 5 years. In 2004, only about a especially if the residents’ needs were complex. It specifically identi-
third of nursing posts in nursing homes were filled by registered fied insufficient numbers of qualified nurses to deliver good qual-
nurses; this is a common reason for nursing home closures (RCN, ity nursing care (RCN, 2010). Early evidence from reviews of care
2004). It is widely recognized that quality of care is more likely to homes’ compliance with the new ‘essential standards’ highlights
be provided by a stable workforce, an issue that is profoundly chal- the importance of having experienced and well-trained staff in care
lenged by high vacancy rates and limited retention. Many consider homes for people with dementia (Care Quality Commission, 2011).
that this problem will only be coherently addressed if wages for care
home staff are significantly increased; a shift that can be ill-afforded Improving Care for Residents
by the care home sector or by local authority commissioners at the
present time. Free movement of labour with the European Union Research
has led to a high proportion of care home staff being from overseas Although care home residents constitute a significant segment of
(Lievesley et al., 2011). Homes frequently employ staff whose first the UK’s older population, they have been relatively neglected by
language is not the same as the residents. researchers and funding bodies. The frail and dependent nature
There is a strong correlation between high quality care and a of most residents makes their needs easy to overlook. Although it
well-trained workforce, particularly in dementia care (Care Quality is important to keep track of the changing profile of the resident
Commission, 2010). This was recognized by the 2009 National population, we need to know much more about what constitutes
Dementia Strategy: ‘workforce development’ is one of its key objec- good and effective care, how to promote positive health amongst
tives and a part of the initial funding of £150 million is targeted in residents despite the high prevalence of chronic conditions, and
this area (Department of Health, 2009). Research suggests that train- how to prepare staff to use best practice all of the time, for every
ing can impact positively on staff attitudes, skills, knowledge, and resident. The positive role that can be played by increasing under-
performance; it can also result in increased confidence and higher standing of the lived experiences of residents and what constitutes a
levels of job satisfaction and retention (National Collaborating ‘good quality of life’ in a care home setting cannot be overstated. As
Centre for Mental Health, 2007; Wild et al., 2010; Skills for Care it is very likely that the proportion of residents with dementia will
and Skills for Health, 2011). A direct correlation has been observed increase in the future, research into high quality dementia care is
of pivotal importance in enhancing the overarching quality of care 2011). Mental health services have a specific role to play in sup-
provided in care homes. port of care home staff and residents, though the most effective and
efficient way of using specialist mental health input has yet to be
Training adequately demonstrated (Thompsell, 2011).
There is no doubt that staff training is the key to improving quality
of care in care homes, especially in relation to supporting residents Conclusion
with dementia. Whilst evidence strongly suggests that training
Although capacity in the care home sector fell for 12 years up to
makes a more sustainable difference when embedded across the
2009, it is predicted that it will begin to significantly increase over
whole care home and/or organization (including management and
the next decade. It is estimated that places will have to—at least—
ancillary staff ), much of the focus of training initiatives is front-line
double by 2043 simply to maintain the current ratio of institutional
care staff. Greater national and local investment in systemic training
to community-based care for people with dementia, the majority
programmes will not only improve care but will increase staff reten-
users of care homes (Macdonald and Cooper, 2007). By 2081 it is
tion rates, improve staff satisfaction levels, and reduce turnover.
likely that around 1.5 million places will be required (Laing and
Promoting good health and wellbeing Buisson, 2010). The debate about how such an increase will be
funded remains unresolved.
Unfortunately little emphasis is placed on promoting the health of The need for expansion reflects a steady growth in the size of the
care home residents, despite evidence that even amongst people very old population and a concomitant increase in the number of
with advanced dementia, improvements in health and wellbeing people with chronic disability and ill health. Complex comorbid
are possible (Livingston et al., 2008). The work that has been done needs, dementia, high levels of challenging behaviour, limited com-
suggests that multifaceted interventions to reduce depression in munication skills, and incontinence are likely to be key character-
care home residents have the capacity not only to reduce or resolve istics of the future care home population. Future demand for care
symptoms but also to improve overall quality of life and wellbe- home places will be influenced by a range of factors, including the
ing (Lyne et al., 2006). Interventions to promote physical health changing but vital role that families play in supporting older rela-
amongst residents also show positive results in terms of impact on tives at home, the costs and quality of community-based care, the
both physical and mental health (Milne, 2009). Other areas where role of alternative forms of long-term care (such as extra care hous-
good practice can improve mental health outcomes include: the ing), and the rising number of older people with substantial assets.
careful management of transition into the care home; knowledge The growth of emerging older populations, such as older people with
about the nature and needs of the older person before admission; learning disabilities and older people from minority ethnic groups,
regular assessment of the mental health status of residents; and an will also play a role, as will end of life care. In terms of provider pat-
understanding that poor health is (often) amenable to medical and/ terns, it is likely that the sector will be increasingly defined by a small
or psychosocial intervention (Ray et al., 2009). A commitment to number of large independent providers; the need for more nursing
person-centred care in the home with a particular focus on a bio- homes, as opposed to residential care homes, is also predicted.
graphical approach and staff spending time on developing relation- Despite the fact that care homes often have a ‘bad press’—and
ships with residents is strongly linked to the promotion of health there are prominent instances of neglect and mistreatment—over-
and wellbeing (Brooker, 2011). This is an arena that would benefit all the standard of both environment and care has improved con-
from additional research investment (Dening and Milne, 2009). siderably over the last 20 years. A combination of better knowledge
Access to services about good practice, a higher level of medical and psychosocial
support to care homes, greater access to training, and enlightened
How well health services work with care homes and provide sup-
regulation should support continued improvements. We might
port to residents lies at the heart of ‘good healthcare’ for the care
also hope to see a change of perception about care homes, so that
home population. Research demonstrates a long history of erratic
they are no longer the option of last refuge but valued facilities for
and inequitable approaches to healthcare delivery by both primary
providing kindness, care, and comfort at the end of life. Greater
and secondary healthcare agencies and practitioners (Goodman
recognition of the complexity of the work done in care homes,
and Davies, 2011). Although there are examples of innovation and
wider awareness of the needs of care home residents, the potential
good practice, these tend to be time limited, discretionary, and
to make an active informed choice to live in a care home, and the
locally determined; often they depend on an individual practition-
inclusion of care homes in community life would go some way to
er’s interest. Care home residents tend not to be seen as a priority
addressing their often invisible, negative, and stigmatized status.
by either primary care or specialist services. This is in part a conse-
We hope that future work will emphasize the perspectives of resi-
quence of residents receiving 24-h care in the home; it is also a fea-
dents themselves (Milne, 2011) and will invest in interventions that
ture of the ‘off the radar’ status of care homes. Recent policy, such
improve the mental health and quality of residents’ lives, the qual-
as the National Dementia Strategy (Department of Health, 2009),
ity and capacity of care staff, and the quality of care and support
has rightly started to place more emphasis on the important con-
residents receive.
tribution services external to care homes can and should make in
supporting residents. Consistent regular primary healthcare input References
to care homes is widely regarded as important by residents and care Age Concern (2010). Paying for permanent residential care: factsheet 10. Age
home staff. Where investment is made to develop this kind of sup- Concern, London.
port, outcomes can include earlier identification of preventable ill- Allen-Burge, R., Stevens, A.B., and Burgio, L.D. (1999). Effective behavioral
ness amongst residents, earlier referral to specialist services, and interventions for decreasing dementia-related challenging behavior in
delayed or preventable hospital admission (Goodman and Davies, nursing homes. International Journal of Geriatric Psychiatry, 14, 213–32.
All-Party Parliamentary Group on Dementia (2009). Prepared to care: Bharucha, A.J., et al. (2004). Predictors of nursing facility admission: a 12-year
challenging the dementia skills gap. Alzheimer’s Society, London. epidemiological study in the United States. Journal of the American
Alzheimer Scotland (2007). A positive choice: choosing long-stay care for a Geriatrics Society, 52, 434–9.
person with dementia. Alzheimer Scotland, Edinburgh. Boaz, A., Hayden, C., and Bernard, M. (1999). Attitudes and aspirations of older
Alzheimer’s Society (2007). Home from home: a report highlighting people: a review of the literature. Department of Social Security, London.
opportunities for improving standards of dementia care in care homes. Borson, S., et al. (2002). Antidepressant prescribing in nursing homes: is
Alzheimer’s Society, London. there a place for tricyclics? International Journal of Geriatric Psychiatry,
Ames, D. (1990). Depression among elderly residents of local-authority 17, 1140–5.
residential homes: its nature and the efficacy of intervention. British Bowers, H., et al. (2009). Older people’s vision for long-term care. Joseph
Journal of Psychiatry, 156, 667–75. Rowntree Foundation, York.
Ames, D., et al. (1988). Psychiatric illness in elderly residents of Part III homes in Bowman, C.E., et al. (2001). Acute hospital admissions from nursing homes:
one London borough: prognosis and review. Age and Ageing, 17, 249–56. some may be avoidable. Postgraduate Medical Journal, 77, 40–2.
Andrews, A.O., et al. (2009). Increased risk of nursing home admission Bowman, C., Whistler, J., and Ellerby M. (2004). A national census of care
among middle aged and older adults with schizophrenia. American home residents. Age and Ageing, 33, 561–6.
Journal of Geriatric Psychiatry, 17, 697–705. Brennan, P.L. and Greenbaum, M.A. (2005). Functioning, problem behavior
Arfken, C.L., Wilson, J.G., and Aronson, S.M. (2001). Retrospective review of and health services use among nursing home residents with alcohol-use
selective serotonin reuptake inhibitors and falling in older nursing home disorders: nationwide data from the VA minimum data set. Journal of
residents. International Psychogeriatrics, 13, 85–91. Studies on Alcohol, 66, 395–400.
Arthur, A., et al. (2002). Factors associated with antidepressant treatment in Brod, M., et al. (1999). Conceptualisation and measurement of quality
residential care: changes between 1990 and 1997. International Journal of of life in dementia: the dementia quality of life instrument (DQoL).
Geriatric Psychiatry, 21, 54–60. Gerontologist, 39, 25–35.
Auslander, L.A., et al. (2001). A comparison of community-dwelling Brodaty, H., et al. (2001). Psychosis, depression and behavioural disturbances
older schizophrenia patients by residential status. Acta Psychiatrica in Sydney nursing home residents: prevalence and predictors.
Scandinavica, 103, 380–6. International Journal of Geriatric Psychiatry, 16, 504–12.
Bagley, H., et al. (2000). Recognition of depression by staff in nursing and Brooker, D. (2008). Quality: the perspective of the person with dementia. In:
residential homes. Journal of Clinical Nursing, 9, 445–50. M. Downs and B. Bowers (eds) Excellence in dementia care: research into
Bajekal, M. (2002). Health survey for England 2000: care homes and their practice, pp. 476–91. Open University Press, Bletchley.
residents. The Stationery Office, London. Brooker, D. (2011). Promoting health and well-being: good practice inside
Ballard, C. and Waite, J. (2006). The effectiveness of atypical antipsychotics the care homes. In: T. Dening and A. Milne (ed.) Mental health and care
for the treatment of aggression and psychosis in Alzheimer’s disease. homes. Oxford University Press, Oxford.
Cochrane Database of Systematic Reviews, CD003476. Brown, M.N., Laplace, K.L., and Luisi, A.F. (2002). The management of
Ballard, C.G., et al. (2001b). A 1-year follow-up study of behavioral and depression in older nursing home residents. Journal of the American
psychological symptoms in dementia among people in care environments. Geriatrics Society, 50, 69–76.
Journal of Clinical Psychiatry, 62, 631–6. Brühl, K.G., Hendrika, J.L., and Martien, T.M. (2007). Nurses’ and nursing
Banerjee, S. (2009). The use of antipsychotic medication for people with assistants’ recognition of depression in elderly who depend on long-term
dementia: time for action. Department of Health, London. care. Journal of the American Medical Directors Association, 8, 441–5.
Banerjee, S., et al. (2003). Predictors of institutionalisation in people with Burton, L.C., et al. (2001). Medical care for nursing home residents:
dementia. Journal of Neurology Neurosurgery and Psychiatry, 74, 1315–16. differences by dementia status. Journal of the American Geriatrics Society,
Banks, L., Haynes, P., and Balloch, S. (2006). Changes in communal provision 49, 142–7.
for adult social care: 1991–2001. Joseph Rowntree Foundation, York. Buttar, A.B., et al. (2003). Six-month cognitive improvement in nursing
Barca, M.L., et al. (2010). A 12 months follow-up study of depression among home residents with severe cognitive impairment. Journal of Geriatric
nursing-home patients in Norway. Journal of Affective Disorders, 120, Psychiatry and Neurology, 16, 100–8.
141–8. Byrne-Davis, L., Bennett, P., and Wilcock, G. (2006). How are quality of life
Bartels, S.J., Mueser, K.T., and Miles, K.M. (1997). A comparative study of ratings made? Towards a model of quality of life in people with dementia.
elderly patients with schizophrenia and bipolar disorder in nursing Quality of Life Research, 15, 855–65.
homes and the community. Schizophrenia Research, 30, 181–90. Camp, C.J., Cohen-Mansfield, J., and Capezuti, E.A. (2002). Use of
Bartels, S.J., et al. (2003a). Agitation and depression in frail nursing home non-pharmacologic interventions among nursing home residents with
elderly patients with dementia: treatment characteristics and service use. dementia. Psychiatric Services, 53, 1397–401.
American Journal of Geriatric Psychiatry, 11, 231–8. Cantley, C. and Wilson, R.C. (2002). Designing and managing care homes for
Bartels, S.J., et al. (2003b). Are nursing homes appropriate for older adults people with dementia. Joseph Rowntree Foundation, York.
with severe mental illness? Conflicting consumer and clinician views Care Quality Commission (2010). The adult social care market and the quality
and implications for the Olmstead decision. Journal of the American of services. CQC, London.
Geriatrics Society, 51, 1571–9. Care Quality Commission (2011). The state of health care and adult social care
Bäumker, T. and Netten, A. (2011). Funding: paying for residential care for in England 2010. CQC, London.
older people. In: T. Dening and A. Milne (eds) Mental health and care Cheng T.W., et al. (2009). Comparison of behavioral and psychological
homes. Oxford University Press, Oxford. symptoms of Alzheimer’s disease among institution residents and
Bäumker, T., Netten, A., and Darton, R. (2010). Costs and outcomes of an memory clinic outpatients. International Psychogeriatrics, 21, 1134–41.
extra care housing scheme in England. Journal of Housing for the Elderly, Chibnall, J.T., et al. (2005). Effect of acetaminophen on behavior, well-being,
24, 151–70. and psychotropic medication use in nursing home residents with
Bebbington, A., Darton, R., and Netten, A. (2001). Care homes for older moderate-to-severe dementia. Journal of the American Geriatrics Society,
people, volume 2: admissions, needs and outcomes. Personal Social 53, 1921–9.
Services Research Unit, University of Kent, Canterbury. Cipher, D.J. and Clifford, P.A. (2004). Dementia, pain, depression, behavioral
Bercovitz, A., et al. (2005). Healthcare utilization of nursing home residents: disturbances, and ADLs: toward a comprehensive conceptualization
comparison between decedents and survivors. Journal of the American of quality of life in long-term care. International Journal of Geriatric
Geriatrics Society, 53, 2069–75. Psychiatry, 19, 741–8.
Clare, L., et al., (2008). The experience of living with dementia in residential people with severe and persistent mental illness. Journal of the American
care: an interpretive phenomenological analysis. Gerontologist, 48(6), Geriatrics Society, 51, 1275–9.
711–20. Dobalian, A., Tsao, J.C., and Radcliff, T.A. (2003). Diagnosed mental and
Cohen, C.I., Hyland, K., and Kimhy, D. (2003). The utility of mandatory physical health conditions in the United States nursing home population:
depression screening of dementia patients in nursing homes. American differences between urban and rural facilities. Journal of Rural Health,
Journal of Psychiatry, 160, 2012–17. 19, 477–83.
Cohen-Mansfield, J. and Taylor, J.W. (2004). Hearing aid use in nursing Dow, B., et al. (2011). Depression in older people living in residential homes.
homes. Part 1: prevalence rates of hearing impairment and hearing aid International Psychogeriatrics, 23, 681–99.
use. Journal of the American Medical Directors Association, 5, 283–8. Dozeman, E., et al. (2011). Feasibility and effectiveness of activity-scheduling
Commission for Social Care Inspection (2007). Guidance for inspectors: short as a guided self-help intervention for the prevention of depression and
observational framework for inspection. Commission for Social Care anxiety in residents in homes for the elderly: a pragmatic randomized
Inspection, London. controlled trial. International Psychogeriatrics, 22, 1–10.
Commission for Social Care Inspection (2008). See me not just my dementia: Draper, B., et al. (2001). Use of psychotropics in Sydney nursing homes:
understanding people’s experiences of living in a care home. Commission associations with depression, psychosis, and behavioral disturbances.
for Social Care Inspection, London. International Psychogeriatrics, 13, 107–20.
Commission on Funding of Care and Support (2011). Fairer care funding: the Draper, B., et al. (2003). Prediction of mortality in nursing home residents:
report of the commission on funding of care and support. Commission on impact of passive self-harm behaviors. International Psychogeriatrics, 15,
Funding of Care and Support, London. 187–96.
Community Care and Health (Scotland) Act (2002). Community Care and Dwyer, M. and Byrne, G.J. (2000). Disruptive vocalization and depression
Health (Scotland) Act, 2002. HMSO, Edinburgh. in older nursing home residents. International Psychogeriatrics, 12,
Conroy, S.P., et al. (2006). Cardiopulmonary resuscitation in continuing care 463–71.
settings: time for a rethink? British Medical Journal, 332, 479–82. Dyer, C.A., et al. (2004). Falls prevention in residential care homes: a
Cook, A. (2002). Using video observation to include the experiences of people randomised controlled trial. Age and Ageing, 33, 596–602.
with dementia in research. In: H. Wilkinson (ed.) The perspectives of Eborall, C., Fenton, W., and Woodrow, S. (2011). The size and structure of the
people with dementia: research methods and motivations. Jessica Kingsley, adult social care sector and workforce in England, 2010. Skills for Care,
London. Leeds.
Counsel and Care (2009). Care home handbook, 3rd edition. Counsel and Eisses, A.M., et al. (2005). Care staff training in detection of depression in
Care, London. residential homes for the elderly: randomised trial. British Journal of
Cuijpers, P. and van Lammeren, P. (2001). Secondary prevention of depressive Psychiatry, 186, 404–9.
symptoms in elderly inhabitants of residential homes. International Evans, S. (2009). Community and ageing: maintaining quality of life in housing
Journal of Geriatric Psychiatry, 16, 702–8. with care settings. Policy Press, Bristol.
Daiello, L.A., et al. (2009). Effect of discontinuing cholinesterase inhibitor Evers, M.M., et al. (2002). The prevalence, diagnosis and treatment of
therapy on behavioral and mood symptoms in nursing home patients with depression in dementia patients in chronic care facilities in the last six
dementia. American Journal of Geriatric Pharmacotherapy, 7, 74–83. months of life. International Journal of Geriatric Psychiatry, 17, 464–72.
Datto, C.J., et al. (2002). Pharmacologic treatment of depression in nursing Featherstone, H. and Whitham, L. (2010). Careless: funding long-term care for
home residents: a mental health services perspective. Journal of Geriatric the elderly. Policy Exchange, London.
Psychiatry and Neurology, 15, 141–6. Fleming, J., et al. (2010). Place of death for the ‘oldest old’: ≥85-year-olds in
Davies, S. and Nolan, M. (2003). ‘Making the best of things’: relatives’ the CC75C population-based cohort. British Journal of General Practice,
experiences of decisions about care-home entry. Ageing and Society, 23, 60, 171–9.
429–50. Forder, J. and Fernandez, J. (2011). Length of stay in care homes. Discussion
Davies, S. and Nolan, M. (2004). ‘Making the move’: relatives’ experiences of Paper 2769. Personal Social Services Research Unit, University of Kent,
the transition to a care home. Health and Social Care in the Community, Canterbury.
12, 517–26. Garre-Olmo, J., et al. (2010). Grouping and trajectories of neuropsychiatric
Davies, S. and Nolan, M. (2006). ‘Making it better’: self-perceived roles of symptoms in patients with Alzheimer’s disease. Part II: two-year patient
family caregivers of older people living in care homes: a qualitative study. trajectories. Journal of Alzheimer’s Disease, 22, 1169–80.
International Journal of Nursing Studies, 43, 281–91. Gessert, C.E., et al. (2000). Tube feeding in nursing home residents with
De Jonghe-Rouleau, A.P., Pot, A.M., and de Jonghe, J.F. (2005). Self-injurious severe and irreversible cognitive impairment. Journal of the American
behaviour in nursing home residents. International Journal of Geriatric Geriatrics Society, 48, 1593–600.
Psychiatry, 20, 651–7. Godden, S. and Pollock, A.M. (2001). The use of acute hospital services by
Dencker, K. and Gottfries, C. (1991). The closure of a major psychiatric elderly residents of nursing and residential care homes. Health and Social
hospital: can psychiatric patients in long-term care be integrated into Care in the Community, 9, 367–74.
existing nursing homes? Journal of Geriatric Psychiatry and Neurology, Goodman, C. and Davies, S. (2011). Good practice outside the care home.
4, 149–56. In: T. Dening and A. Milne (eds) Mental health and care homes. Oxford
Dening, T.R. and Milne, A. (2005). Providing residential and nursing home University Press, Oxford.
care. In: L. Keith (ed.) Menopause, postmenopause and ageing, pp. 98–105. Hancock, G.A., et al. (2006). The needs of older people with dementia in
Royal Society of Medicine, London. residential care. International Journal of Geriatric Psychiatry, 21, 43–9.
Dening, T. and Milne, A. (2009). Depression and mental health in care Hoe, J., et al. (2005). Use of the QOL-AD for measuring quality of life in
homes. Quality in Ageing Special Issue: Depression, Suicide and Self-Harm people with severe dementia—the LASER-AD study. Age and Ageing, 34,
in Older Adults, 10(2), 40–6. 130–5.
Dening, T. and Milne, A. (eds) (2011). Mental health and care homes. Oxford Hoe, J., et al. (2006). Quality of life of people with dementia in residential care
University Press, Oxford. homes. British Journal of Psychiatry, 188, 460–4.
Department of Health (2009). Living well with dementia: a National Dementia Holland, A.J. (2000). Ageing and learning disability. British Journal of
Strategy. DH, London. Psychiatry, 176, 26–31.
Depla, M.F., et al. (2003). Integrating mental health care into residential Howard, R., et al. (2007). Donepezil for the treatment of agitation in
homes for the elderly: an analysis of six Dutch programs for older Alzheimer’s disease. New England Journal of Medicine, 357, 1382–92.
Hughes, J.C. (2010). Promoting palliative care in dementia. Lancet Neurology, Macdonald, A. and Cooper, B. (2007). Long-term care and dementia services:
9, 25–7. an impending crisis. Age and Ageing, 36, 16–22.
Hughes, J.C. and Louw, S. (2002). Electronic tagging of people who wander. Macdonald, A.J. and Dening, T.R. (2002). Dementia is being ignored in NHS
British Medical Journal, 325, 847–8. and social care. British Medical Journal, 324, 548.
Husebo, B.S., et al. (2011). Efficacy of treating pain to reduce behavioural Macdonald, A.J. and Woods, R.T. (2005). Attitudes to dementia and dementia
disturbances in residents of nursing homes with dementia: cluster care held by nursing staff in UK ‘non-EMI’ care homes: what difference
randomised clinical trial. British Medical Journal, 343, d4065. do they make? International Psychogeriatrics, 17, 383–91.
Hyden, L.-C. and Orulv, L. (2009). Narrative and identity in Alzheimer’s Macdonald, A.J., et al. (2002). Dementia and the use of psychotropic
disease: a case study. Journal of Aging Studies 23, 205–14. medication in non-‘elderly mentally infirm’ nursing homes in south east
Hyer, L., et al. (2008). Group, individual, and staff therapy: an efficient and England. Age and Ageing, 31, 58–64.
effective cognitive behavioral therapy in long-term care. American Macer, J. and Murphy, J. (2009). Training care home staff to use talking mats
Journal of Alzheimer’s Disease and Other Dementias, 23, 528–39. with people who have dementia. Joseph Rowntree Foundation, York.
Innes, A. and Kelly, F. (2007). Evaluating long stay care settings: reflections Magaziner, J., et al. (2005). Mortality and adverse health events in newly
on the process with particular reference to DCM. In: A. Innes and L. admitted nursing home residents with and without dementia. Journal of
McCabe (eds) Evaluation in dementia care. Jessica Kingsley, London. the American Geriatrics Society, 53, 1858–66.
Joseph, C.L. (1995). Alcohol and drug misuse in the nursing home. Mann, A.H., Graham, N., and Ashby, D. (1984). Psychiatric illness in
International Journal of the Addictions, 30, 1953–84. residential homes for the elderly: a survey in one London borough. Age
Joseph, C.L., Ganzini, L., and Atkinson, R.M. (1995). Screening for alcohol and Ageing, 13, 257–65.
use disorders in the nursing home. Journal of the American Geriatrics Martinon-Torres, G., Fioravanti, M., and Grimley, E.J. (2004). Trazodone for
Society, 43, 368–73. agitation in dementia. Cochrane Database of Systematic Reviews, 18 Oct,
Killick, J. and Martin L. (2011). Creative work with residents. In: T. Dening (4), CD004990.
and A. Milne (eds) Mental health and care homes. Oxford University Matthews, F.E and Dening, T.R. (2002). Prevalence of dementia in institutional
Press, Oxford. care. Lancet, 360, 225–6.
Kirkevold, Ø. and Engedal, K. (2005). Concealment of drugs in food and McShane, R., Areosa Sastre, A., and Minakaran, N. (2004). Memantine
beverages in nursing homes: cross sectional study. British Medical for dementia. Cochrane Database of Systematic Reviews, 9 Apr, (2),
Journal, 330, 20–2. CD003154.
Kroenke, K., Spitzer, R.L., and Williams, J.B. (2003). The Patient Health Menon, A.S., et al. (2001). Relationship between aggressive behaviors and
Questionnaire–2: validity of a two-item depression screener. Medical depression among nursing home residents with dementia. International
Care, 41, 1284–92. Journal of Geriatric Psychiatry, 16, 139–46.
Laing and Buisson (2009). Laing’s healthcare market review 2009–2010. Laing Milne, A. (2009). Promoting mental health in later life. In: T. Williamson
and Buisson, London. (ed) Older people’s mental health today: a handbook. Mental Health
Laing and Buisson (2010). Care of elderly people: market survey 2010–11. Foundation and Pavilion Publishing, Brighton.
Laing and Buisson, London. Milne, A. (2011). Living with dementia in a care home: capturing the
Lamberg, J.L., et al. (2005). Decisions to hospitalize nursing home residents experiences of residents. Quality in Ageing and Older Adults Special Issue,
dying with advanced dementia. Journal of the American Geriatrics Dementia Care: a positive future, 12 (2), 76–85.
Society, 53, 1396–401. Milne, A. and Chryssanthopoulou, C. (2005). Dementia caregiving in Black
Landi, F., et al. (1998). Comorbidity and drug use in cognitively impaired and Asian populations: reviewing and refining the research agenda.
elderly living in long-term care. Dementia and Geriatric Cognitive Journal of Community and Applied Social Psychology, 15, 319–37.
Disorders, 9, 347–56. Milne, A. and Williams, J. (2000). Meeting the mental health needs of older
Landreville, P., et al. (2006). Non-pharmacological interventions for aggressive women: taking social inequality into account. Ageing and Society, 20,
behavior in older adults living in long-term care facilities. International 699–723.
Psychogeriatrics, 18, 47–73. Milne, A., et al. (2001). Caring in later life: reviewing the role of older carers.
Lapane, K.L. and Hughes, C.M. (2004). Which organizational characteristics Help the Aged, London.
are associated with increased management of depression in US nursing Mitchell, S.L., Kiely, D.K., and Lipsitz, L.A. (1997). The risk factors and impact
homes? Medical Care, 42, 992–1000. on survival of feeding tube placement in nursing home residents with
Lemke, S. and Schaefer, J.A. (2010). VA nursing home residents with severe cognitive impairment. Archives of Internal Medicine, 157, 327–2.
substance use disorders: Mental health comorbidities, functioning, and Mitchell, S.L., et al. (2003). Clinical and organizational factors associated with
problem behaviors. Aging and Mental Health, 14, 593–602. feeding tube use among nursing home residents with advanced cognitive
Leurs, P., et al. (2010). Alcohol consumption patterns in older people living in impairment. Journal of the American Medical Association, 290, 73–80.
nursing homes. Presse Médicale, 39, e280–8. Mitchell, S.L., et al. (2004a). Estimating prognosis for nursing home residents
Lievesley, N., Crosby, G., and Bowman, C. (2011). The changing role of care with advanced dementia. Journal of the American Medical Association,
homes. BUPA and Centre for Policy on Ageing, London. 291, 2734–40.
Livingston, G., et al. (2008). Successful ageing in adversity—the LASER-AD Mitchell, S.L., Kiely, D.K., and Hamel, M.B. (2004b). Dying with advanced
longitudinal study. Journal of Neurology, Neurosurgery and Psychiatry, dementia in the nursing home. Archives of Internal Medicine, 164,
79, 641–5. 321–6.
Llewellyn-Jones, R.H., et al. (1999). Multifaceted shared care intervention Monteleoni, C. and Clark, E. (2004). Using rapid-cycle quality improvement
for late life depression in residential care: randomised controlled trial. methodology to reduce feeding tubes in patients with advanced dementia:
British Medical Journal, 319, 676–82. before and after study. British Medical Journal, 329, 491–4.
Logsdon, R.G., et al. (1999). Quality of life in Alzheimer’s disease: patient and Mozley, C., et al. (1999). ‘Not knowing where I am does not mean I don’t
caregiver reports. Journal of Mental Health and Aging, 5, 21–32. know what I like’: cognitive impairment and quality of life responses
Loveday, B. (2011). Dementia training in care homes. In: T. Dening and A. Milne in elderly people. International Journal of Geriatric Psychiatry, 14,
(eds) Mental health and care homes. Oxford University Press, Oxford. 776–83.
Lyne, K.J., et al. (2006). Analysis of a care planning intervention for reducing Murphy, J., Oliver, T.M., and Cox, S. (2010). Talking mats help involve people
depression in older people in residential care. Aging and Mental Health, with dementia and their carers in decision-making. Joseph Rowntree
10, 394–403. Foundation, York.
Murphy, L.M. and Lipman, T.O. (2003). Percutaneous endoscopic gastrostomy Ruths, S., et al. (2008). Stopping antipsychotic drug therapy in demented
does not prolong survival in patients with dementia. Archives of Internal nursing home patients: a randomized, placebo-controlled study—the
Medicine, 163, 1351–3. Bergen District Nursing Home Study (BEDNURS). International Journal
National Collaborating Centre for Mental Health (2007). Dementia: the of Geriatric Psychiatry, 23, 889–995.
NICE–SCIE guideline on supporting people with dementia and their carers Samus, Q.M., et al. (2005). The association of neuropsychiatric symptoms
in health and social care. British Psychological Society/Royal College of and environment with quality of life in assisted living residents with
Psychiatrists, Leicester/London. dementia. Gerontologist, 45 (1), 19–26.
National Council for Palliative Care (2006). Introductory guide to end of life Scherder, E., et al. (2005). Recent developments in pain in dementia. British
care in care homes. Department of Health, London. Medical Journal, 330, 461–646.
Netten, A., et al. (2001). Care homes for older people. Volume 1: facilities, Schneider, L.S., Dagerman, K.S., and Insel, P. (2005). Risk of death with
residents and costs. Personal Social Services Research Unit, University of atypical antipsychotic drug treatment for dementia: meta-analysis of
Kent, Canterbury. randomized placebo-controlled trials. Journal of the American Medical
Netten, A., et al. (2011). Improving housing with care choices for older people: Association, 294, 1934–43.
an evaluation of extra care housing. Personal Social Services Research Scocco, P., et al. (2006). Suicidal behaviour in nursing homes: a survey in a
Unit, University of Kent, Canterbury. region of north-east Italy. International Journal of Geriatric Psychiatry,
Nijs, A.N., et al. (2006). Effect of family style mealtimes on quality of life, 21, 307–11.
physical performance, and body weight of nursing home residents: cluster Seitz, D.P., et al. (2009). Cholinesterase inhibitor use in U.S. nursing homes:
randomised controlled trial. British Medical Journal, 332, 1180–3. results from the national nursing home survey. Journal of the American
Nolan, M., et al. (2003). Partnerships in family care. Open University Press, Geriatrics Society, 57, 2269–74.
Maidenhead. Seitz, D., Purandare, N., and Conn, D. (2010). Prevalence of psychiatric
Nygaard, H.A. and Jarland, M. (2005). Are nursing home patients with disorders among older adults in long-term care homes: a systematic
dementia diagnosis at increased risk of inadequate pain treatment? review. International Psychogeriatrics, 22, 1025–9.
International Journal of Geriatric Psychiatry, 20, 730–7. Serra, L., et al. (2010). Are the behavioral symptoms of Alzheimer’s disease
Office of Fair Trading (2005). Care homes for older people in the UK: a market directly associated with neurodegeneration? Journal of Alzheimer’s
study. Office of Fair Trading, London. Disease, 21, 627–39.
Organization for Economic Cooperation and Development (2005). Long-term Sinclair, A.J., Allard, I., and Bayer, A. (1997). Observations of diabetes care in
care for older people. OECD, Paris. long-term institutional settings with measures of cognitive function and
Orrell, M. and Hancock, G. (2004) CANE: Camberwell Assessment of Need for dependency. Diabetes Care, 20, 778–84.
the Elderly. Gaskell, London. Skills for Care and Skills for Health (2011). Common core principles for
Owen, T. and the National Care Home Research and Development Forum supporting people with dementia. Skills for Care, Leeds.
(2006). My home life: quality of life in care homes. Help the Aged, London. Smalbrugge, M., et al. (2006). The impact of depression and anxiety on well
Parker, M.J., Gillespie, W.J., and Gillespie, L.D. (2006). Effectiveness of hip being, disability and use of health care services in nursing home patients.
protectors for preventing hip fractures in elderly people: systematic International Journal of Geriatric Psychiatry, 21, 325–32.
review. British Medical Journal, 332, 571–4. Smith, S.C., et al. (2002). Measurement of health-related quality of life for people
Parmelee, P.A., Katz, I.R., and Lawton, M.P. (1991). The relation of pain to with dementia: development of a new instrument (DEMQOL) and an
depression among institutionalized aged. Journal of Gerontology, 46, evaluation of current methodology. Health Technology Assessment, 9, 1–93.
15–21. Snowdon, J. (2005). The role of psychogeriatric services in long-term
Parsons, C., et al. (2011). Cholinesterase inhibitor and memantine use in residential care settings. In: B. Draper, P. Melding, and H. Brodaty (eds)
newly admitted nursing home residents with dementia. Journal of the Psychogeriatric service delivery: an international perspective, pp. 213–33.
American Geriatrics Society, 59, 1253–9. Oxford University Press, Oxford.
Proitsi, P., et al. (2011). A Multiple Indicators Multiple Causes (MIMIC) Snowdon, J., Day, S., and Baker, W. (2005). Why and how antipsychotic drugs
model of behavioural and psychological symptoms in dementia (BPSD). are used in 40 Sydney nursing homes. International Journal of Geriatric
Neurobiology of Aging, 32, 434–42. Psychiatry, 20, 1146–52.
Quince, C. (2013). Low expectations: Attitudes on choice, care and community Snowden, M., Sato, K., and Roy-Byrne, P. (2003). Assessment and treatment
for people with dementia in care homes, Alzheimer’s Society, London. of nursing home residents with depression or behavioral symptoms
Ray, M., Bernard, M., and Phillips, J. (2009). Critical issues in social work with associated with dementia: a review of the literature. Journal of the
older people. Palgrave Macmillan, Basingstoke. American Geriatrics Society, 51, 1305–17.
Ray, W.A., et al. (1997). A randomized trial of a consultation service to reduce Sørensen, L., et al. (2001). Assessment of dementia in nursing home residents
falls in nursing homes. Journal of the American Medical Association, 278, by nurses and assistants: criteria validity and determinants. International
557–62. Journal of Geriatric Psychiatry, 16, 615–21.
Reed, P.S., et al. (2005). Characteristics associated with low food and fluid intake Sullivan-Marx, E.M., et al. (1999). Predictors of continuing physical restraint
in long-term care residents with dementia. Gerontologist, 45 (1), 74–80. use in nursing home residents following restraint reduction efforts.
Restifo, S., Lemon, V., and Waters, F. (2011). Pharmacological treatment of Journal of the American Geriatrics Society, 47, 342–8.
behavioural and psychological symptoms of dementia in psychogeriatric Surr, C.A. (2006). Preservation of self in people with dementia living in
inpatient units. Australasian Psychiatry, 19, 59–63. residential care: A socio-biographical approach. Social Science and
Rothera, I., et al. (2002). Survival in a cohort of social services placements in Medicine, 62, 1720–30.
nursing and residential homes: factors associated with life expectancy Sutcliffe, C., et al. (2007). Depressed mood, cognitive impairment, and
and mortality. Public Health, 116, 160–5. survival in older people admitted to care homes in England. American
Rothera, I., et al. (2003). Health status and assessed need for a cohort of older Journal of Geriatric Psychiatry 15, 708–15.
people admitted to nursing and residential homes. Age and Ageing, 32, Taylor, C. and Hendra, T. (2000). The prevalence of diabetes mellitus and
303–9. quality of diabetic care in residential and nursing homes: a postal survey.
Royal College of Nursing (2004). Nursing assessment and older people. RCN, Age and Ageing, 29, 447–50.
London. Thapa, P.B., et al. (1995). Psychotropic drugs and risk of recurrent falls in
Royal College of Nursing (2010). Care homes under pressure. RCN, ambulatory nursing home residents. American Journal of Epidemiology,
London. 142, 202–11.
Thompsell, A. (2011). Support to care homes. In: T. Dening and A. Milne Warner, J., Milne, A., and Peet, J. (2010). ‘My name is not dementia’: literature
(eds) Mental health and care homes. Oxford University Press, Oxford. review. Alzheimer’s Society, London.
Thompson, D. and Wright, S. (2001). Misplaced and forgotten: people with Watson, L.C., et al. (2009). Practical depression screening in residential care/
learning disabilities in residential services for older people. Mental Health assisted living: five methods compared with gold standard diagnoses.
Foundation, London. American Journal of Geriatric Psychiatry, 17, 556–64.
Train, G.H., et al. (2005). A qualitative study of the experiences of long-term Wild, D., Nelson, S., and Szczepura, A. (2010). Residential care home workforce
care for residents with dementia, their relatives and staff. Aging and development: the rhetoric and reality of meeting older residents’ future care
Mental Health, 9, 119–28. needs. Joseph Rowntree Foundation, York.
Travis, S.S., et al. (2004). Analyses of nursing home residents with diabetes at Wilkinson, H., et al. (2004). Home for good? Preparing to support people with
admission. Journal of the American Medical Directors Association, 5, 320–7. learning difficulties in residential settings when they develop dementia.
Treloar, A., Beats, B., and Philpot, M. (2000). A pill in the sandwich: covert Pavilion, Brighton, and Joseph Rowntree Foundation, York.
medication in food and drink. Journal of the Royal Society of Medicine, Williams, S.W., et al. (2005). Characteristics associated with mobility
93, 408–11. limitation in long-term care residents with dementia. Gerontologist, 45
Van Doorn, C., et al. (2003). Dementia as a risk factor for falls and fall injuries (1), 62–7.
among nursing home residents. Journal of the American Geriatrics Winkler, D., Farnworth, L., and Sloan, S. (2006). People under 60 living
Society, 51, 1213–18. in aged care facilities in Victoria. Australian Health Review, 30,
Wang, P.S., et al. (2005). Risk of death in elderly users of conventional vs. 100–8.
atypical antipsychotic medications. New England Journal of Medicine, Winzelberg, G.S., et al. (2005). Factors associated with nursing assistant
353, 2335–41. quality-of-life ratings for residents with dementia in long-term care
Wanless, D., et al. (2006). The Wanless social care review: securing good care facilities. Gerontologist, 45 (1), 106–14.
for older people—taking a long-term view. King’s Fund, London. Zwakhalen, S.M., et al. (2006). Pain in elderly people with severe dementia:
Ward, J.A., et al. (2010). A cluster randomised controlled trial to prevent a systematic review of behavioural pain assessment tools. BMC
injury due to falls in a residential aged care population. Medical Journal Geriatrics, 6, 3.
of Australia, 192, 319–22.
CHAPTER 28
Palliative care and
end of life care
Elizabeth Sampson and Karen Harrison Dening
The population of Europe and other developed nations is ageing. For in three people over the age of 65 will die whilst affected by demen-
example, by 2020, in the UK and the US, one in four people will be tia (Brayne et al., 2006).
aged over 60, and in Japan this rises to one third. This has occurred Frailty has been defined as:
secondary to an ‘epidemiological transition’: the number of deaths an aggregate expression of risk resulting from age- or disease-associated
in early life from infectious disease has decreased (Omran, 1971), physiologic accumulation of subthreshold decrements affecting multi-
primary prevention of other pathologies such as cardiovascular dis- ple physiologic systems resulting in adverse health outcomes.
ease is becoming more successful, the treatment and survival rates (Abellan van Kan et al., 2008)
of some cancers is improving, and thus a higher proportion of the It also becomes more common with increasing age; by the age of
world’s population is living to old age. This has led to an increase 90 years, 32% of people will be frail (Gavrilov and Gavrilova, 2001).
in the numbers of people living long enough to develop illnesses The older population has a higher proportion of women and both
associated with old age, in particular neurodegenerative diseases dementia and frailty are more common in females (Walston and
such as dementia. These multiple comorbidities will have cumula- Fried, 1999).
tive effects on function, quality of life, and care needs. Frailty and dementia are multifactorial in origin and share some
In the UK and other developed countries there will be a steady common aetiological pathways. Smoking, obesity, lack of physical
increase in the number of deaths, and in the proportion of those activity, and depression have been linked to the development of
who die over the age of 85 years (Gomes and Higginson, 2008). both frailty and dementia ( Woods et al., 2005; Ownby et al., 2006;
Despite the fact that the vast majority of deaths occur in adults over Peters et al., 2008; Hamer and Chida, 2009; Kerwin et al., 2010).
the age of 65 years, there is widespread evidence that older people Thus frailty and dementia have been conceptualized as final com-
have inequitable access to good end of life care, and for people with mon pathways resulting from cumulative exposures over a number
mental health problems such as chronic schizophrenia or with neu- of years (Neale et al., 2001).
rodegenerative diseases such as dementia, there are further barri- Both dementia and frailty have an adverse impact on a range of
ers to adequately meeting their needs (Davies and Higginson, 2004; outcomes. People with dementia are at higher risk of acute hospi-
Sampson et al., 2006). tal admission (Mukadam and Sampson, 2011), falls (Myers et al.,
1991), and being placed in care homes—crisis or transition points
The Epidemiology of Advanced Dementia at which the diagnosis is often made for the first time. Frailty also
increases the risk of emergency hospital admission, overnight
and Frailty hospital stays (Wagner et al., 2006), falls (Fried et al., 2001), and
Worldwide prevalence figures for dementia will rise steeply; cur- institutionalization. These outcomes are independent of physical
rently an estimated 24.3 million people worldwide have demen- comorbidity and other potential confounders such as age and gen-
tia, but the number affected will double every 20 years, to 81.1 der (Avila-Funes et al., 2009).
million in 2040 (Ferri et al., 2005). The prevalence of dementia
increases with age, affecting 6% of those aged 75–79 years and a Dementia and frailty are life-limiting conditions
third of people over the age of 95 (Knapp and Privette, 2007). Thus Despite the impact that dementia and frailty have on older people
increasing numbers of people will die whilst suffering from demen- and their families, they have not traditionally been conceptualized
tia. Estimates from the UK Medical Research Council–Cognitive as ‘terminal’ or ‘life-limiting’ syndromes. For example, care home
Function and Ageing Studies (MRC–CFAS) project, a large multi- medical and nursing staff consistently overestimate prognosis in
centre study looking at the health and cognitive function of 13,000 advanced dementia. In one study of nursing home carers and phy-
older people, suggest that people who died between the ages of 65 sicians, at nursing home admission only 1.1% of residents were per-
and 69 years had a 6% risk of dying with dementia, rising to a 58% ceived to have life-expectancy of less than 6 months; however, 71%
risk of dying with dementia in those over 95 years. In the UK, one died within that period (Mitchell et al., 2004). The median length of
stay in UK nursing homes, where the majority of residents will have of practice. All definitions share a common philosophy of care,
moderate or severe dementia, is 18 months. With a median survival which is to take a holistic approach, valuing autonomy of patients
time of 1.3 years, advanced dementia (as defined by a Functional and their families, with a focus on dignity, a collaborative relation-
Assessment Staging Scale (FAST) score of 7c and above) is associ- ship between healthcare professionals, patients and their families,
ated with a life-expectancy similar to that of well-recognized termi- good communication, and a central goal to maintain the quality of
nal diseases such as metastatic breast cancer (Mitchell et al., 2009). life. These are, of course, all characteristics required by profession-
A recent UK population study gave a median survival time from als working in any mental health setting. It is also important to note
diagnosis of dementia to death of 4.1 years (Xie et al., 2008); this is that end of life care ‘seeks neither to hasten death, nor to postpone
strongly influenced by age, with those aged 65–69 years surviving it’ (Radbruch et al., 2009).
for a median 10.7 years, and those aged 90 years and above surviv- For the purposes of this chapter we will be focusing mainly on
ing for 3.8 years (Xie, et al., 2008). As would be expected, older end of life care of older people with advanced dementia; this is
people defined as being frail also have a significantly increased risk where the focus of policy, research, and clinical provision has most
of death (Klein et al., 2005). recently been directed.
Identifying when frail older people with dementia are reaching
the end of their life can be challenging. Numerous studies have The Clinical Picture of Advanced Dementia
attempted to identify prognostic indicators or indices that may
guide physicians to adopting a more palliative approach to care, but There are no operationalized clinical criteria for advanced demen-
these tools are more reliable at identifying people with dementia at tia, but it has been suggested that it can be defined as a FASTlevel
low risk of dying rather than those at higher risk of death (van der of 7 and above (Reisberg, 1988): the person with dementia cannot
Steen et al., 2005). Clinical judgement, discussion with families and dress him- or herself, is doubly incontinent, and speaks at most
carers, and taking the opportunity to reassess or shift the goals of only a few words.
management towards palliative care at times of intercurrent illness In contrast with the early and middle stages of dementia, the clin-
or transition may be a more practical and reliable approach (van ical features of advanced dementia are less well characterized and
der Steen, 2010). researched. People with advanced dementia suffer a range of symp-
For example, acute physical illness requiring emergency hospital toms, similar to those found in the terminal stages of cancer, e.g.
admission, such as pneumonia and urinary tract infections, may be pain and dyspnoea. Pressure sores, agitation, and eating problems
an indicator of imminent death in people with advanced demen- (i.e. difficulty with swallowing or loss of appetite) are very common
tia (Morrison and Siu, 2000a; Mitchell et al., 2009; Sampson et al., as the end of life approaches (see Table 28.1).
2009a). This population also has markedly increased 6-month mor- A retrospective study comparing symptoms experienced in the
tality after hip fracture (55%) and pneumonia (53%), compared last year of life by 170 people with dementia compared to 1513
to 12% and 13% in cognitively intact patients (Morrison and Siu, cancer patients showed that the symptom burden between the two
2000a). Shorter-term mortality is also very poor, with 24% of those groups was comparable; in particular, 64% of dementia patients
with moderate or severe dementia dying after acute unplanned experienced pain and 57% had loss of appetite. The healthcare
medical admissions, compared to 7.5% of those who did not have needs of both groups were also similar (McCarthy et al., 1997). The
dementia (Sampson et al., 2009a). careful management of these specific symptoms is vital in providing
However, there is little evidence that many active medical inter- a holistic approach to end of life care for people with dementia.
ventions, such as artificial hydration and nutrition or hospital
admission, prolong or improve the quality of life in people with
Pain
dementia (Sampson et al., 2009b; van der Steen, 2010), and there Pain is common in people with advanced dementia and is
has been recent interest in how a palliative approach may be benefi- often underdetected and undertreated (Scherder et al., 2009).
cial for frail older people with advanced dementia. Retrospective interviews with relatives and carers indicate that sig-
nificantly more dementia patients are reported to experience pain
in the last 6 months of life compared to those with cancer (75%
What Is Palliative Care? vs. 60%) (McCarthy et al., 1997), but pain control is often inad-
The World Health Organization defines palliative care as: equate. In patients with fractured neck of femur, those with cog-
an approach that improves the quality of life of patients and their nitive impairment were prescribed a third as much analgesia as
families facing the problems associated with life threatening illness, cognitively intact controls (Morrison and Siu, 2000b). There is no
through the prevention and relief of suffering by means of early iden- consistent evidence to suggest that pain experience is less intense,
tification and impeccable assessment and treatment of pain and other but people with advanced dementia will have difficulties in com-
problems, physical, psychosocial and spiritual. municating that they are in pain and interpreting pain signals. This
(Gomez-Batiste et al., 2007) often manifests as behavioural change such as agitation, distress,
‘End of life care’ may be used synonymously with ‘palliative care’, social withdrawal, depression, or resistive behaviour (Scherder
particularly in the US. Its use is favoured because palliative care et al., 2009).
is sometimes associated only with cancer, whereas end of life care The ‘gold standard’ for pain assessment is self-reporting. Despite
refers to all patients with life-limiting illnesses. In European coun- widely held beliefs to the contrary, many patients with moderate to
tries, end of life care is often used to describe the care given during severe dementia can report pain reliably (Zwakhalen et al., 2006).
the last few hours or days of life, In those with communication difficulties it may be necessary to
The European Association of Palliative Care (EAPC) has recently use direct observation or validated observational pain scales such
published a consensus statement attempting to define the principles as the Abbey Tool (Abbey et al., 2004). The assessment of pain in
CHAPTER 28 palliative care and end of life care 361
Table 28.1 Prevalence of symptoms (percentage) in nursing home residents with advanced dementia
Study details Mitchell et al. (2009) Black et al. (2006) DiGiulio et al. (2008) Aminoff and Adunsky (2005)
Proportion of patients with Proportion of patients with Proportion of patients Proportion of patients in the
symptoms 18 months prior life expectancy of 6 months in the last 30 days of life last week of life (prospective
to death (prospective study) (prospective study) (retrospective study) study)
n = 323 n = 126 n = 141 n = 71
Dyspnoea 46 29 39 –
Pain 40 63 26 18
Pressure ulcers 39 61 47 70
Agitation/restlessness 54 50 20 72
Aspiration 41 15 – –
Eating problems 86 85 – 95
advanced dementia is, however, complex and it has been suggested strong predictor of institutionalization to nursing or residential
that there is no evidence that pain produces any unique features homes (Steele et al., 1990). Different BPSD tend to emerge at dif-
that may not be caused by other sources of distress (Regnard et al., ferent stages of the illness, e.g. mood disorders and depression are
2003); and thus tools such as the Disability Distress Assessment noted earlier in the disease course and psychotic symptoms (hallu-
Tool (DiSDAT) may be more appropriate (Regnard et al., 2007). cinations and delusions) are more common in the moderate stages.
Managing pain in a person with advanced dementia requires Wandering and agitation tend to be the most enduring BPSD. In
careful assessment for possible underlying causes. This should the advanced stages of dementia, over half of patients remain agi-
include clinical examination, observation of behaviours (both at tated and distressed (Mitchell et al., 2009). The management of
rest and during movement), and discussion with family members these symptoms is complex and requires a structured approach,
or staff who know the person well. In some cases, response to a trial using a range of therapeutic approaches. Difficult behaviours such
of analgesic may in itself be a diagnostic tool (Herr et al., 2006). The as aggression and resistiveness to care may be indicators of unmet
aetiology of pain may be acute and may, for example, be associated needs, such as undetected or untreated pain, delirium, or infection.
with urinary retention, constipation (Kovach et al., 2006), myocar- Patients therefore require a full assessment, which can be challeng-
dial infarction, deep vein thrombosis, or acute infection. Common ing in advanced dementia when verbal communication is limited.
chronic causes of pain in people with advanced dementia include Basic aspects of personal care, such as providing glasses or hear-
pressure sores, undetected fractures, poor dentition, awkward posi- ing aids, should be considered first. Environmental modifications,
tioning in chair or bed, and arthritis (Davis and Srivastava, 2003). such as limiting noise or providing outside space or gardens where
Treating pain in this population requires a stepped approach people can walk, can reduce agitation. Psychological interventions,
(American Geriatric Society Panel, 1998). Depending on the cause, e.g. simple ‘ABC’ analysis (documentation of the Antecedents of
nonpharmacological interventions may be helpful, particularly the behaviour, the Behavioural disturbance, and the Consequences
where there is a musculoskeletal aetiology. This may include reposi- of the behaviour) can reveal patterns and triggers for a particular
tioning, provision of optimal seating or specialist beds, physiother- problem, and are also very effective (Kovach et al., 2006; Testad
apy, massage, heat, or cold. Simple analgesia such as paracetamol et al., 2010), and drugs should be used as a last resort. Antipsychotic
and nonsteroidal inflammatory drugs can be very effective. The key medications both typical and atypical can cause parkinsonism, pro-
to achieving a good result is regular prescribing; as-required medi- longation of the QT interval, and increase the risk of stroke and
cation is often not given as people with advanced dementia may not death (Schneider et al., 2005).
report they are in pain or request painkillers. If simple analgesics
are ineffective then stronger analgesia such as opioids should be Eating and swallowing
considered. Side effects such as delirium, constipation, and seda- People with advanced dementia often develop swallowing difficul-
tion may occur, but in practice these can be managed. An important ties. Two methods of enteral tube feeding are used: a nasogastric tube
recent study (Husebo et al., 2011) demonstrated how this approach (a tube that is passed through the nose and into the stomach) or via
to the management of pain significantly reduced agitation in resi- a percutaneous endoscopic gastrostomy (PEG) (directly through the
dents of nursing homes with moderate to severe dementia. Regular abdominal wall into the stomach). There have been no randomized
paracetamol was effective in 63% of participants, and of the 25% controlled trials of these interventions in people with advanced
who received opioids only 2% had these discontinued because of dementia. A recent Cochrane Review (Sampson et al., 2009b) found
side effects. inconclusive evidence that enteral tube feeding provides any ben-
efit in dementia patients in terms of survival time, mortality risk,
Behavioural and psychological symptoms of dementia nutritional parameters, and improvement or reduced incidence of
Behavioural and psychological symptoms of dementia (BPSD) are pressure ulcers. There have been no studies on the effect of these
common in dementia, affecting 90% of people at some time dur- interventions on quality of life. Enteral tube feeding may increase
ing the course of their illness. They are extremely distressing both pulmonary secretions, incontinence, mortality, and morbidity. PEG
for people with dementia and their family and carers, and also a is an invasive surgical procedure with significant postoperative risks,
including aspiration pneumonia, oesophageal perforation, migra- women, and people dying from noncancer related causes (Gomes
tion of the tube, haemorrhage, and wound infection. The decision to and Higginson, 2008). Concurrently, the proportion of older peo-
use enteral tube feeding is emotive and influenced by complex ethi- ple dying in long-term care facilities such as residential and nurs-
cal issues, clinical need, local practice, physician and carer prefer- ing homes has increased across Europe and the US, where 40% of
ence, and whether there is an advance directive or care plan in place. deaths occur in long-term care facilities.
Clinicians may feel pressurized by institutional (Lopez et al., 2010),
societal, or local laws to intervene. For example, in the US, Illinois Care homes
requires a physician to seek a court order to withhold or withdraw When frail older people with dementia are moved to a care home
tube feeding but no other states do. Ethical considerations include they frequently experience multiple losses: physical, mental, social,
whether life in advanced dementia should be artificially prolonged and spiritual. Although death may not necessarily be imminent,
and what is considered to constitute ‘euthanasia’, i.e. the difficult residents of care homes are highly likely to die there, making
issue that by withholding food and fluids death will be hastened. these settings where palliative care is needed (Cartwright, 2002;
It has been argued that the absence of evidence demonstrating a Parker-Oliver et al., 2004; Froggatt et al., 2006). A number of bar-
benefit of tube feeding does not mean that it is inappropriate in all riers have been identified that may challenge the provision of good
patients with advanced dementia and that each individual deserves quality palliative care in these settings. These include high rates
a holistic assessment by a specialist in swallowing; all appropriate of staff turnover, lack of access to primary care medical support
interventions should be considered including the option of a gas- and specialist palliative care resources when required, and fear of
trostomy (Regnard et al., 2010). Some ethicists suggest that artificial censure by regulatory authorities if deaths occur in the care home
nutrition and hydration are medical treatments that can legitimately (Harrison Dening et al., 2012). In the UK, medical support to care
be withheld if their risks, judged according to the patient’s values, out- homes is variable and concerns have been raised when acute illness
weigh their benefits (Gillick, 2000); however, this implies that values occurs at night and on-call doctors may not have the information
are known or documented and this is often not the case. necessary to make the decision that a resident is for palliative care
and not active intervention. This can lead to the distressing situ-
Infections and pneumonia ation where a resident is admitted to the acute hospital and dies
Pneumonia and other infections, especially of the urinary tract, are there. A related issue is that of anticipatory prescribing. For exam-
common in advanced dementia. Over 18 months, 53% of nursing ple, care homes may not hold drugs, particularly opiates, which are
home residents with advanced dementia will have a febrile episode, commonly held by palliative care teams and hospices. The funding
and 41% will have pneumonia (Mitchell et al., 2009). These will be the of long-term care and palliative care varies widely between different
immediate cause of death in up to 71% of cases (Burns et al., 1990). countries. In the US, there is a focus on ‘hospice appropriateness’,
People with advanced dementia are often immobile, bed bound, at whereby Health Management Organizations will fund palliative
increased risk of aspiration, and may have impaired immunological care for people with dementia, once it has been agreed that they
function. The use of antibiotics to treat fevers and recurrent infec- have a likely survival time of less than 6 months. In the Netherlands,
tions is one of the most controversial issues in this field. Fabiszewski care homes are served by specialist nursing home physicians, and
et al. (1990) demonstrated no difference in mortality between peo- therefore much palliative care is given ‘in house’.
ple receiving antibiotics and those receiving only ‘palliative’ care. In
a cohort of Dutch nursing home patients, van der Steen et al. (2002) Hospices
found that patients treated with antibiotics survived for longer (27% It is often stated that people with dementia are ‘denied’ access to
died compared to 90% of those who did not have antibiotics); how- hospices; less than 1% of hospice patients in Europe have a neuro-
ever, this may have been related to the fact that antibiotics were logical diagnosis (Davies and Higginson, 2004), and in the US, less
withheld from patients whom physicians believed had more severe than 7% of hospice patients have a primary diagnosis of dementia
dementia. Some research has suggested that withholding antibiotics (Jennings, 2003). The reasons for this are complex and related to
increases the level of discomfort (van der Steen et al., 2009), but it many of the issues described above, e.g. the fact that dementia is
has also been argued that antibiotics might delay death, leaving the not perceived to be a terminal illness and concerns that hospice
patient exposed to the risk of further pain and suffering and prolong- staff will not be able to manage BPSD (McCarty and Volicer, 2009;
ing the dying phase. In this situation, adequate symptomatic control Sanders and Swails, 2009). Patients with dementia are also less likely
through the use of analgesia may be more appropriate. to be referred to palliative care services, and in many health systems
access to hospices is dependent on recommendations from these
services. It has been argued that hospices provide ‘specialist’ pal-
Care Settings liative care and that they should rightly focus on cases of complex
Many recent national and international palliative care policies have need—most of the symptoms experienced by people with dementia
focused on how best to provide people with the opportunity to die in at the end of life, such as pain and difficulties with swallowing, do
a place of their choice (Department of Health, 2008). International not require specialist intervention but good generalist care.
population surveys have demonstrated how over two-thirds of peo-
ple say that they would prefer to die at home (Gomes et al., 2012). The Role of Person-Centred Care and
However, in western populations over the last hundred years there
has been a large shift away from deaths at home to deaths in hospi- Spirituality at the End of Life
tals and care facilities. For example, in the UK, between 1974 and What it is to be a person is central to our understanding of person-
2003 the proportion of home deaths fell from 31% to 18% overall. centred care. Person-centred care is now a widely acknowledged basis
This occurred at an even higher rate for people aged 65 and over, for dementia care delivery and is a term that is synonymous with
considered. Dementia care policy and strategy has largely been

Box 28.1 Key features of person-centred care
driven by the principles of living well with dementia; however, the
growing recognition and concern of the often poor end of life expe-
◆ Looks at care from the perspective of the individual
riences for people with dementia has led to a policy shift (National
◆ Acknowledges each person as a unique individual with a rich End of Life Care Programme, 2010; National Institute for Health
history and own memories, preferences, wishes, and needs and Clinical Excellence and National End of Life Care Programme,
◆ Respects the dignity, autonomy, and independence of the 2010) that recognizes the need for the principles of high quality
person with dementia care to encompass dying with dementia.
◆ Focuses on the positive rather than the negative
◆ Focuses on strengths and abilities rather than weaknesses and
Advance Care Planning
disabilities The fear of not being seen as a ‘whole person’ or cared for in a
person-centred way has contributed to the anxieties that people have
◆ Promotes wellbeing
about their future health. For example, a recent BUPA survey found
◆ Ensures care is planned around the individual, not around the that, when asked what health worries they were most concerned
care system about when they got older, more people worried about developing
◆ Acknowledges there is usually a reason for a behaviour, and dementia (58%) than were concerned about getting cancer (47%)
views behaviours that challenge others as an expression of feel- or having a heart attack (47%) (Fernández and Forder, 2010). The
ings and/or a means of communication experiences of people with dementia are starting to have a positive
influence on our understanding of person-centred care from the
◆ Accepts the reality of the person with dementia and does not perspective of the person with dementia. Bryden (2005), based on
insist on bringing the person into another reality that can cause her personal experience as a person with dementia, maintains that
distress people with dementia should take the opportunities to explore with
(Kitwood, 1997) other people more existential issues about life and death, while they
still retain the capacity to plan their future care.
Advance care planning (ACP) is understood in a variety of differ-
treating people as individuals, respecting their rights as a person, and ent ways. The term is often used without definition or explanation
(from a professional carer perspective) in building therapeutic rela- and different emphases are placed upon the different elements of the
tionships (McCormack and McCance, 2010). care planning (Henry and Seymour, 2007). ACP usually involves a
Kitwood, a noted leader in the movement for person-centred process of discussing and recording priorities and wishes for future
care in dementia, extended the concept of person-centred care (see care and treatment between individuals and their health and/or
Box 28.1) to consider personhood and the status of being a person, social care providers. It attempts to anticipate the future deteriora-
and developed a definition of personhood: ‘A standing or status tion of a person’s condition so that when a person is no longer able
that is bestowed upon one human being by others, in the context of to communicate, for whatever reason, wishes and preferences that
relationship and social being’ (Kitwood, 1997). This established the have been recorded earlier can be met (Froggatt et al., 2008).
concept of person-centred care within relationships with others. In contrast to those with cancer and other advanced chronic dis-
People with dementia, particularly those in the advanced stages, ease, people with dementia may have profound cognitive impair-
are reportedly often denied care that is ‘person-centred’, with the ment and lack the capacity required to make decisions about their
suggestion they may be classified as already ‘dead’ (Post, 2006). care and treatment. People with dementia are significantly less likely
Post states that, due to the lack of empowering capacities, people to have an ACP compared to those with cancer, although uptake
with advanced dementia may be seen as ‘nonpersons’. The ethical of the process in cancer remains variable (Mitchell et al., 2004).
and philosophical debate on the concepts of self and personhood Ideally, ACP should be attempted in the earlier stages of demen-
draws together the central issues of both dignity and spirituality. tia when a person is still competent to make decisions. An MMSE
There is growing interest in spirituality and spiritual care, which score of 18–20 appears to be a consistent threshold score required
are seen increasingly as essential components in person-centred to make an ACP (Fazel et al., 1999; Gregory et al., 2007). Below this
care and quality of life of people with dementia, especially towards threshold, individuals appear more likely to opt for life-sustaining
the end of life when capacity may be lost. The drive to understand treatments (Fazel et al., 2000), but as dementia progresses there is
better elements of spirituality is also seen in the delivery of pallia- increased caregiver involvement in decision-making (Hirschman
tive and end of life care. A review of spiritual care (Holloway et al., et al., 2004). In current UK clinical practice, it remains rare for
2011) commissioned to support the implementation of the End of a person with advanced dementia to have an ACP, while little is
Life Care Strategy (DH, 2008) found spirituality of particular con- known about how ACP may benefit people with dementia and their
cern within the nursing literature, though the largest body of mate- carers (Harrison Dening et al., 2011).
rial was more concerned with its assessment and much less so on
specific interventions. In dementia care, spirituality is bound up Advance care planning and decisions about
in the concept of person-centred care and embodies what it is to life-sustaining treatment
be an individual with a life history, personality, likes, dislikes, etc. Professionals and family carers often anticipate an adverse reaction
However, acknowledgement and assessment of spiritual needs may to pursuing ACP with the person with dementia, but this is not nec-
be neglected in people dying with advanced dementia (Sampson essarily the case. Finucane et al. (1991) found that such concerns in
et al., 2006) before delivery of any related intervention is even family carers were unfounded, and that the person with dementia
showed no distress either before or after ACP discussions. Carer their general beliefs and personal values about the sort of care they
attitudes may influence the likelihood of people with dementia would like to receive in the future.
being exposed to aggressive treatments at end of life. Mezey et al. Lasting Powers of Attorney (LPA) were also introduced through
(2000) found that carers with a greater sense of burden were more the UK Mental Capacity Act 2005 and enable the appointment of
likely to consent to life-sustaining treatment. Fazel et al. (2000) an attorney to make proxy decisions when the person with demen-
found that people with dementia were more likely to make deci- tia no longer has the capacity to do so. A ‘Personal Welfare’ LPA
sions that were impulsive or opt for life-sustaining interventions allows the attorney to give or refuse consent to medical treatment,
than controls, but Finucane et al. (1991) found that people with if such preferences have been expressed in the document.
dementia were more likely to refuse life-sustaining therapy as they Due to the advancing cognitive difficulties of the person with
became more cognitively impaired. dementia, nurses and allied healthcare professionals must ensure
they communicate with a main carer(s) in addition to the per-
Family carers and decision-making son with dementia in order that appropriate care and support
Decision-making for people with dementia at the end of life may is given. This can give rise to ethical and legal challenges, such
be fraught with problems, with prompts for discussions about ACP as receipt of third party information, disclosure of confidential
perhaps triggered by a medical event, transitions and changes to the health information, consideration of the person with demen-
living situation, or a hospital admission for the person with dementia’s capacity for decision-making, and potential disagreements
tia. In the absence of any defined ACP, decision-making often falls to among family members on the best interests of the person with
family carers, with increasing caregiver ‘burden’ being a significant dementia, for example. This demands that dementia care profes-
predictor of caregiver-dominated decision-making (Hirschman sionals should be aware of the legal and ethical issues involved in
et al., 2004). Feelings such as guilt and a sense of failure when the ACP, as well as developing the skills required to embark on such
person with dementia goes into long-term care, together with a lack difficult conversations.
of information on the disease and its prognosis, leave family carers
unprepared to make effective decisions about end of life care for The Role of Nursing Care and Allied Health
their relative with dementia (Forbes et al., 2000; Harrison Dening
et al., 2012). Professionals
Dementia affects the whole family, and people close to the person
Professional attitudes with dementia find themselves providing increasing amounts of
A priority of UK government policy is to improve the diagnosis care and support as the disease progresses. Hence, care delivered is
rates of dementia (Department of Health, 2009). However, of equal required to be mindful of the whole family unit and the relationships
importance at this time must be supporting people with demen- therein. Various health and social care professionals have a role in
tia to consider their end of life care preferences and choices whilst supporting palliative and end of life care for people with demen-
they still have the cognition, capacity, and language to do so. tia. For example, occupational therapists may provide assessment
Professionals often find this a difficult area to address. Studies show of the home (or usual place of care) environment for equipment
that there is inadequate preparation of both people with dementia and adaptation needs; social care workers may provide assessment
and their carers for future care issues and related decisions, with and provision of personal care and domestic support needs; district
some not engaging in ACP discussions at all (Cavalieri et al., 2002). nurses may help in accessing vital equipment and meeting physical
Despite agreement on many aspects of end of life decision-making, healthcare needs; physiotherapists may provide care and support,
physicians, nurses, and relatives hold different views arising from and advice on mobility and falls prevention, and chest care in cases
their religious beliefs, perspectives of the patient, and care respon- of pneumonia. Many health and social care professionals are keen
sibilities (Rurup et al., 2006). Where ACP do exist, relatives often to provide better quality end of life care to people with dementia.
attach great importance to them and believe they have to be fol- A recent study, using a rapid participatory appraisal methodology,
lowed, whereas with the widespread use of the Mental Capacity Act found that all professionals involved in palliative and end of life
2005, professionals know that only certain types of ACP documents care in dementia were acutely aware of the limitations both in their
have any legal standing (Rurup et al., 2006). own knowledge and skills and in the health and social care sys-
tem within which they were working. Many of the barriers to good
Advance care planning process and tools quality end of life care in dementia highlighted in the studies were
ACP tools are emerging, with little evidence as yet for any improve- associated with lack of interagency communication, not having a
ment in associated outcomes for people with dementia (Harrison clear dementia care pathway, and a sense of helplessness (Harrison
Dening et al., 2011). As part of a wider study in Australia, Caplan Dening et al., 2012; Sampson et al., 2012).
et al., (2006) used an ACP tool (Molloy, 2005) as part of a pro- There are two very clear elements to good quality dementia
gramme to reduce acute hospital interventions. Education and care: the medical model/approach and the social model/approach.
ACP led to a reduction in emergency calls, decreased hospital Both are inextricably linked and both of distinct value to the care
admissions, and a decrease in mortality compared with homes in delivered to people with dementia and their families. Early on in
the control group. the illness it will be necessary to discuss the purpose of diagnostic
In the UK, the Mental Capacity Act 2005 provides a legal frame- investigations, what they entail, and how results will be given. This
work for the ACP process. The Act gives competent adults in is an important time to establish a relationship with people with
England and Wales the legal right to refuse treatment (i.e. artifi- dementia and their family, understanding the main problems from
cial feeding and resuscitation) through the writing of an ‘advance their perspective, and liaising with other health and social care
decision’. People can also make an ‘advance statement’ that reflects providers.
Interventions in End of Life Care for care requires a more rounded approach, and a number of multi-
component complex interventions and pathways have been devel-
Older People oped that may improve the quality and outcomes of care for older
Care management people with dementia in the acute hospital and in the community.
There is a growing interest in a ‘care management’ approach for fam- Care pathways and policies
ilies affected by dementia; this would entail detailed assessment of
need and a coordination of care and support from health, social, and The English National Dementia Strategy (Department of Health,
voluntary care professionals. Such an approach is considered essen- 2009) contained little regarding end of life care, and the English
tial in meeting the palliative and end of life care needs of people with End of Life Care Strategy (Department of Health, 2008) made little
dementia and their family carers (Judd et al., 2011). A care manage- specific reference to dementia. Perhaps this illustrates how patients
ment approach is identified as a positive determinant of how well with dementia fall through gaps in the health and social care sys-
a carer is supported in his or her caring role (Hibberd, 2011), and tem? The National Institute for Health and Clinical Excellence
is an approach that is best applied throughout the illness, from the (NICE) guidelines for dementia (National Collaborating Centre for
diagnosis of dementia, living with dementia, to death and beyond in Mental Health, 2007) do include some recommendations, in par-
supporting families in their loss (Harrison Dening, 2010). ticular the use of the Gold Standards Framework and the Liverpool
Admiral Nursing is one approach in providing care manage- Care Pathway for the care of the dying.
ment throughout the course of the illness (Harrison Dening, 2010). The Gold Standards Framework
Following diagnosis, appropriate follow-up is essential to discuss The UK Gold Standards Framework (GSF) (<http://www.gold-
the prognosis and issues of legal and financial planning, give advice standardsframework.org.uk/>) is a multidimensional programme
on the next steps, commence the discussion of ACP, and continue that supports and trains staff to identify patients requiring pallia-
liaison with local services. Later in the illness, nursing and allied tive or supportive care towards the end of life. It uses a structured
healthcare input will depend largely on symptoms and their man- approach to recognize when the last year of life may have begun, to
agement. Information and guidance provided within the context of assess patients’ needs, symptoms, and preferences, and to plan care
continuing professional support, as in the case of Admiral Nursing around these, in particular supporting people to live and die where
(specialist dementia care nurses), is more successful in supporting they choose. Although not developed specifically for patients with
family carers and reducing anxiety levels (Wills and Woods, 1998). dementia, the GSF developed for care homes attempts to enhance
A distinct advantage of the Admiral Nurse service is the long-term communication between GPs and other specialists, particularly out
nature of the work with families of people with dementia and the of hours care. Introduction of the GSF has been shown to increase
support offered throughout the trajectory of the illness; this sup- the proportion of residents with ACP and to reduce the numbers
port is of particular value during transitions from community to admitted to acute hospitals (Badger et al., 2009).
care home (Harrison Dening, 2010).
The Liverpool Care Pathway
The Liverpool Care Pathway (LCP) was originally developed to
Inappropriate interventions and treatment improve care in hospital for cancer patients during the last 48 h of
at the end of life life (Ellershaw, 2007). It has been modified for use in people dying
Despite the high mortality in advanced dementia, particularly for with other diagnoses and different settings including hospices and
those who are admitted to the acute hospital, people with dementia nursing homes. The pathway has three phases: initial assessment
receive as many painful investigations and procedures (e.g. arterial of the patient, ongoing assessment, and care after death. As well as
blood gas sampling) and are more likely to be physically restrained attending to medical needs, such as the discontinuation of inappro-
compared with patients who are cognitively intact (Morrison and priate interventions and medications and the provision of comfort
Siu, 2000a). This suggests that clinicians fail to adopt a palliative or measures (e.g. mouth care), the pathway enhances person-centred
supportive approach to patient care, possibly because people with care by assessing patients’ insight into their situation and their psy-
advanced dementia are not perceived to be suffering from a ‘terminal chological and spiritual needs and those of their family. There are,
illness’. This may be due to poor understanding of the pathophysiol- however, a number of issues pertinent to people with dementia that
ogy of dementia amongst healthcare professionals (Thuné-Boyle et al., may challenge implementation for these patients. The pathway is
2010). Other vital components of good end of life and person-centred only of use if it is recognized that a patient is moving into the ‘dying
care are also neglected with little acknowledgement of spiritual needs, phase’, and this can be difficult to identify in people with advanced
failure to withdraw inappropriate medications, and nonreferral to dementia. Concerns have been raised that in dementia, ‘dying’ can
hospital palliative care teams (Sampson et al., 2006). Patients dying take many years and that food and fluids may be withdrawn too early
with dementia in nursing homes often receive burdensome interven- (Treloar, 2008). However, the pathway is flexible and it is entirely
tions. In the American study of Mitchell et al., (2009), 29% received appropriate that patients are sometimes moved from the pathway
enteral tube nutrition and 12.4% were hospitalized. back to more active management should their clinical condition
The evidence base on palliative care in dementia is somewhat lim- improve. Future adaptation and evaluation of the pathway for use in
ited, particularly when compared to research on palliative care for people with dementia could significantly enhance quality of care.
people dying from cancer. Systematic reviews have identified how
care mainly focuses on specific interventions such as fever man-
Complex interventions and care programmes
agement policies, pain control, or the withdrawal of care, i.e. not Interventions in hospital settings
prescribing antibiotics, rather than a more active palliative model Ahronheim et al. (2000) conducted a randomized controlled
(Sampson et al., 2005; Goodman et al., 2009). Good person-centred trial of an intervention in which people with dementia received
assessment and care plan recommendations from a palliative care people with dementia require ongoing care is generally longer
team with the goal of enhancing patient comfort in an acute hospi- compared to other illnesses, this means that carers of people with
tal. The intervention did not decrease rates of readmission to hos- dementia have a journey ahead of them, one that may be long and
pital, average length of stay, or mortality, but there was a significant tortuous. Many carers are spouses and therefore likely to be old and
increase in written palliative care plans and patients received fewer to have physical or mental health needs of their own, to the extent
intravenous drugs. The authors highlighted the ‘unique barriers’ to that these may hamper their ability to provide care to the person
providing good end of life care, including prognostic uncertainty, with dementia (Harrison Dening, 2011).
and suggest that care planning may be more effective when done However, the relationship that the carer had with the individual
outside the busy acute hospital environment. prior to his or her illness still remains and ‘carer’ is an additional
Lloyd-Williams and colleagues used multidisciplinary prescrib- but not a replacement identity for previous relationships and roles,
ing guidelines on wards for older people with dementia. There was e.g. husband or son. The changing nature of the relationship often
a significant decrease in the prescribing of antibiotics in the last 2 requires physical and emotional adjustments throughout its course
weeks of life and patients were much more likely to be prescribed and it can bring a number of predeath losses that cause caregivers
analgesia, including opiates (Lloyd-Williams and Payne, 2002). to grieve (Liken and Collins, 1993). Liken and Collins state that
even when family carers are able to work through their predeath
Interventions in nursing homes and specialist care units
grief, they cannot fully re-establish their lives until after the death
A specific approach in the US has been that of the Dementia Special of their relative.
Care Unit (DSCU). Volicer et al. (1994) compared this approach to It is often assumed that when a person with dementia is admitted
nursing home settings and found that the DSCU, which provided a into a long-term care environment this largely resolves the carer
palliative care approach focusing on ‘maintenance of patient’s com- burden and stress. This is often not the case and may leave the carer
fort rather than maximal survival’, led to less discomfort, fewer anti- with feelings of guilt at ‘giving up’. They may feel they have gone
biotic prescriptions, less use of intravenous drugs, and decreased against the expressed wishes of their loved one and this may leave
transfers to the acute hospital. Mortality in the DSCU was higher them with a sense of failure in that they were unable to continue
but costs were significantly lower. (Givens et al., 2011a, 2011b). This can manifest itself in the carer
Training and educational programmes on end of life care for at times being overly critical of the care provided, so it is impor-
nursing home staff also appear to be effective in improving knowl- tant to consider that the previous levels of burden and separation
edge and increasing satisfaction with end of life care in bereaved stress (Kiely et al., 2008) may not be resolved and are in danger
family members (Arcand et al., 2009). of going undetected and unsupported. Providing support to car-
Interventions in the community ers, recognizing the circumstances in which they find themselves,
The Palliative Excellence in Alzheimer Care Effort (PEACE) and allowing them the opportunity to talk about their situation can
Programme is an American programme set up to improve end enable some to provide at-home care for longer periods prior to
of life care for people with dementia. It takes a long-term dis- institutionalization (Etters et al., 2008).
ease management model, integrating ‘palliative care into ongoing
comprehensive primary care of persons with dementia over the Loss, grief, and bereavement
disease course, from initial diagnosis to death’ (Shega, 2003). The Caring for a person with dementia is considered more burdensome
patient-centred principles of the programme include ACP, educa- and ‘unrelenting’ when compared with other long-term conditions
tion on the disease process, improved care coordination, and fam- (Sachs et al., 2004), with increased vulnerability to complex or
ily support. This is implemented by clinical nurse specialists who abnormal grief reactions (Schulz et al., 1997). Some carers find that
coordinate care between families, physicians, and other health and they have grieved so much during the course of the illness that they
social care professionals. Those involved in the programme were have no strong feelings left when the person dies, yet others feel a
more likely to die in hospice or a place of their choosing, less likely further acute sense of loss when the person dies.
to die in the acute hospital, and their carers were more satisfied A sense of loss is one of the most powerful feelings that carers
with the quality of care received. of people with dementia experience. This sense of loss and associ-
In the UK, the ‘Hope for Home’ service has supported patients ated anticipatory grief can be felt whilst the person with dementia
with severe dementia and their families, providing a multidisci- is still alive, and depending on the relationship with the person with
plinary and holistic model coordinated by a specialist in old age dementia, the carer may experience grief for many forms of loss
psychiatry. Preliminary results have shown that total cost savings (see Box 28.2). The manner in which family carers experience and
of home care compared to nursing home care for 14 patients was manage their grief reactions to the predeath losses can influence
£696,930 and that 57% of participants died in their own home not only caregiving outcomes but also subsequent adjustment once
(Treloar et al., 2009)—a significantly higher proportion than would those with dementia have died. It is therefore crucial that their grief
be expected in the UK general population (22%) (Gomes and is addressed (Collins et al., 1993; Almberg et al., 2000).
Higginson, 2008). In a systematic review, Chan et al. (2013) described grief in demen-
tia carers as a complex reaction to losses occurring before and after
Carers’ and Relatives’ Burden, Grief, death. Chan et al. argue that anticipatory grief is greatest in moder-
ate to severe stage dementia and spouse carers, especially when the
Bereavement, and Support person with dementia is institutionalized. Evidence about the preva-
As the number of people with dementia increases, so also will the lence of grief is less robust, with studies reporting anticipatory grief
number of family carers, with current estimations in the UK of between 47% and 71% and complicated grief after death in around
670,000 (Lakey et al., 2012). Given that the period during which 20% of dementia carers. Carer depression increases with anticipatory
Abellan van Kan, G., et al. (2008). The I.A.N.A Task Force on frailty assessment
Box 28.2 Losses that may be experienced by a carer of a person of older people in clinical practice. Journal of Nutrition Health and Aging,
with dementia 12, 29–37.
Ahronheim, J.C., et al. (2000). Palliative care in advanced dementia: a
◆ The future they may have planned together randomized controlled trial and descriptive analysis. Journal of Palliative
◆ The relationship enjoyed prior to dementia Medicine, 3, 265–73.
Almberg, B.E., Grafstrom, M., and Winblad, B. (2000). Caregivers of
◆ The loss of companionship, support, or the special nature of relatives with dementia: experiences encompassing social support and
the relationship bereavement. Aging and Mental Health, 4, 82–9.
American Geriatric Society Panel (1998). The management of chronic pain
◆ Loss of freedom to work or to pursue other activities
in older persons. Journal of the American Geriatrics Society, 46, 635–51.
◆ Loss of finances Aminoff, B.Z. and Adunsky, A. (2005). Dying dementia patients: too much
suffering, too little palliation. American Journal of Hospice and Palliative
◆ Loss of lifestyle Medicine, 22, 344–8.
◆ Loss of childhood when a young person or carer Arcand, M., et al. (2009). Educating nursing home staff about the progression
of dementia and the comfort care option: impact on family satisfaction
◆ Bereaved female spouses are at higher risk of depression with end-of-life care. Journal of the American Medical Directors
following death of their husbands/partners Association, 10, 50–5.
◆ Separation bereavement when the person with dementia is Avila-Funes, J.A., et al. (2009). Cognitive impairment improves the predictive
admitted into care validity of the phenotype of frailty for adverse health outcomes: the
three-city study. Journal of the American Geriatrics Society, 57, 453–61.
◆ Regret and guilt Badger, F., et al. (2009). An evaluation of the implementation of a programme to
◆ Loss or lack of social support networks improve end-of-life care in nursing homes. Palliative Medicine, 23, 502–11.
Black, B.S., et al. (2006). Health problems and correlates of pain in nursing
◆ The person they once knew home residents with advanced dementia. Alzheimer Disease and
(Harrison Dening, 2011) Associated Disorders, 20, 283–90.
Brayne, C., et al. (2006). Dementia before death in ageing societies—the
promise of prevention and the reality. PLoS Medicine, 3(10), e397.
Bryden, C. (2005). Dancing with dementia. Jessica Kingsley, London.
grief. Being a spouse carer and being depressed are the strongest pre-
Burns, A., et al. (1990). Cause of death in Alzheimer’s disease. Age and Ageing,
dictors of complicated grief after death (Chan et al., 2013).
19, 341–4.
It is important that bereaved carers of people with dementia receive Caplan, G.A., et al. (2006). Advance care planning and hospital in the nursing
follow-up care in recognition of their loss and bereavement. Many home. Age and Ageing, 35, 581–6.
carers feel a second, and huge, sense of loss and they may also feel that Cartwright, J.C. 2002. Nursing homes and assisted living facilities as places
their life no longer has the meaning that it has had for many years. for dying. Annual Review of Nursing Research, 20, 231–64.
Often the care services that were present in meeting the person with Cavalieri, T.A., et al. (2002). How physicians approach advance care planning
dementia’s needs instantly withdraw, leaving a void in the carer’s life. in patients with mild to moderate Alzheimer’s disease. Journal of the
Carers are often discharged from services that were related to the per- American Osteopathic Association, 102, 541–7.
son with dementia so their expressions of grief are often not observed Chan, D., et al. (2013). Grief reactions in dementia carers: a systematic review.
or addressed adequately. Care, such as Admiral Nursing, that is deliv- International Journal of Geriatric Psychiatry, 28(1), 1–17.
Collins, C., Liken, M., and Kokinakis, C. (1993). Loss and grief among family
ered with a focus on the whole family affected by dementia will take
caregivers of relatives with dementia. Qualitative Health Research, 3,
into account carer issues and provide much needed support after the 236–53.
death of the person with dementia (Harrison Dening, 2010). Davies, E. and Higginson, I. J. (eds) (2004). Better palliative care for older
people. World Health Organization, Copenhagen.
Conclusion Davis, M.P. and Srivastava, M. (2003). Demographics, assessment and
management of pain in the elderly. Drugs and Aging, 20, 23–57.
Older people dying with dementia often have complex physical and Department of Health (2008). End of life care strategy: promoting high quality
mental health needs. These require a multidisciplinary approach care for all adults at the end of life. DH, London.
that pays attention not just to physical symptoms but also to their Department of Health (2009.) Living well with dementia: a National Dementia
spiritual and social needs, supporting their family carers and the Strategy. DH, London.
staff that work with them, in a range of settings from community to Di Giulio, P., et al. (2008). Dying with advanced dementia in long-term care
the acute hospital. A number of promising care initiatives and path- geriatric institutions: a retrospective study. Journal of Palliative Medicine,
ways have been set up, most of which take such a holistic approach, 11, 1023–8.
in keeping with the core values of palliative care and the philosophy Ellershaw, J. (2007). Care of the dying: what a difference an LCP makes!
Palliative Medicine, 21, 365–8.
of those working in mental healthcare of older people. The recent
Etters, L., Goodall, D., and Harrison, B.E. (2008). Caregiver burden among
research and policy interest in this field underlines its importance dementia patient caregivers: a review of the literature. Journal of the
and provides hope that the care of many frail older people dying American Academy of Nurse Practitioners, 20, 423–8.
with dementia can be improved. Fabiszewski, K.J., Volicer, B., and Volicer, L. (1990). Effect of antibiotic
treatment on outcome of fevers in institutionalized Alzheimer patients.
References Journal of the American Medical Association, 263, 3168–72.
Abbey, J., et al. (2004). The Abbey pain scale: a 1-minute numerical indicator Fazel, S., Hope, T., and Jacoby, R. (1999). Assessment of competence to
for people with end-stage dementia. International Journal of Palliative complete advance directives: validation of a patient centred approach.
Nursing, 10, 6–13. British Medical Journal, 318, 493–8.
Fazel, S., Hope, T., and Jacoby, R. (2000). Effect of cognitive impairment Herr, K., et al. (2006). Pain assessment in the nonverbal patient: position
and premorbid intelligence on treatment preferences for life-sustaining statement with clinical practice recommendations. Pain Management
medical therapy. American Journal of Psychiatry, 157(6), 1009–11. Nursing, 7, 44–52.
Fernández J.-L. and Forder, J. (2010). Ageing societies: challenges and Hibberd, P. (2011). What is the meaning of family-centred Admiral Nursing
opportunities: evidence from the BUPA Health Pulse 2010 International to carers? Unpublished PhD thesis, Northumbria University, Newcastle
Healthcare Survey. British United Provident Association, London. upon Tyne.
Ferri, C.P., et al. (2005). Global prevalence of dementia: a Delphi consensus Hirschman, K.B., et al. (2004). How does an Alzheimer’s disease patient’s
study. Lancet, 366, 2112–17. role in medical decision making change over time? Journal of Geriatric
Finucane, T.E., et al. (1991). Establishing advance medical directives with Psychiatry and Neurology, 17, 55–60.
demented patients: a pilot study. Journal of Clinical Ethics, 4, 51–4. Holloway, M., et al. (2011). Spritual care at the end of life: a systematic review
Forbes, S., Bern-Klug, M., and Gessert, C. (2000). End-of-life decision of the literature. Department of Health, London.
making for nursing home residents with dementia. Journal of Nursing Husebo, B.S., et al. (2011). Efficacy of treating pain to reduce behavioural
Scholarship, 32, 251–9. disturbances in residents of nursing homes with dementia: cluster
Fried, L. P., et al. (2001). Frailty in older adults: evidence for a phenotype. randomised clinical trial. British Medical Journal, 343, d4065.
Journals of Gerontology Series A: Biological Sciences and Medical Sciences, Jennings, B. (2003). Hospice and Alzheimer’s disease: a study in access and
56, M146–56. simple justice. Hastings Centre Report Special Supplement, S24–26.
Froggatt, K.A., et al. (2006). End-of-life care in long-term care settings for Judd, K., Harrison Dening, K., and Weatherhead, I. (2011). Management
older people: a literature review. International Journal of Older People of patients with dementia. In: S. Woodward and A.-M. Mestecky
Nursing, 1, 45–50. (eds) Neuroscience nursing: evidence-based practice, pp. 488–500.
Froggatt, K., et al. (2008). Advance planning in care homes for older people: a Wiley-Blackwell, Chichester.
survey of current practice. International Observatory on End of Life Care, Kerwin, D. R., et al. (2010). The cross-sectional relationship between
Lancaster. body mass index, waist-hip ratio, and cognitive performance in
Gavrilov, L.A. and Gavrilova, N.S. (2001). The reliability theory of aging and postmenopausal women enrolled in the Women’s Health Initiative.
longevity. Journal of Theoretical Biology, 213, 527–45. Journal of the American Geriatrics Society, 58, 1427–32.
Gillick, M.R. (2000). Rethinking the role of tube feeding in patients with Kiely, D.K., Prigerson, H.G., and Mitchell, S.L. (2008). Health care proxy grief
advanced dementia. New England Journal of Medicine, 342, 206–10. symptoms before the death of nursing home residents with advanced
Givens, J.L., et al. (2011a). Mental health and exposure to patient distress dementia. American Journal of Geriatric Psychiatry, 16, 664–73.
among families of nursing home residents with advanced dementia. Kitwood, T. (1997). Dementia reconsidered: the person comes first. Open
Journal of Pain and Symptom Management, 42, 183–91. University Press, Buckingham.
Givens, J.L., et al. (2011b). Grief among family members of nursing home Klein, B.E., et al. (2005). Frailty, morbidity and survival. Archives of
residents with advanced dementia. American Journal of Geriatric Gerontology and Geriatrics, 41, 141–9.
Psychiatry, 19, 543–50. Knapp, M. and Privette, A. (2007). Dementia UK. Alzheimer’s Society,
Gomes, B. and Higginson, I.J. (2008). Where people die (1974—2030): past London.
trends, future projections and implications for care. Palliative Medicine, Kovach, C.R., et al. (2006). The Serial Trial Intervention: an innovative
22, 33–41. approach to meeting needs of individuals with dementia. Journal of
Gomes, B., et al. (2012). Preferences for place of death if faced with Gerontological Nursing, 32, 18–25.
advanced cancer: a population survey in England, Flanders, Germany, Lakey, L., et al. (2012). Dementia 2012: a national challenge. Alzheimer’s
Italy, the Netherlands, Portugal and Spain. Annals of Oncology, 23(8), Society, London.
2006–15. Liken, M.A. and Collins, C.E. (1993). Grieving: facilitating the process for
Gomez-Batiste, X., et al. (2007). Catalonia WHO palliative care demonstration dementia caregivers. Journal of Psychosocial Nursing and Mental Health
project at 15 years (2005). Journal of Pain and Symptom Management, 33, Services, 31, 21–6.
584–90. Lloyd-Williams, M. and Payne, S. (2002). Can multidisciplinary guidelines
Goodman, C., et al. (2009). End of life care for community dwelling older improve the palliation of symptoms in the terminal phase of dementia?
people with dementia: an integrated review. International Journal of International Journal of Palliative Nursing, 8, 370–5.
Geriatric Psychiatry, 25, 329–337. Lopez, R.P., et al. (2010). The influence of nursing home culture on the use of
Gregory, R., et al. (2007). Is the degree of cognitive impairment in patients feeding tubes. Archives of Internal Medicine, 170, 83–8.
with Alzheimer’s disease related to their capacity to appoint an enduring McCarty, C.E. and Volicer, L. (2009). Hospice access for individuals with
power of attorney? Age and Ageing, 36, 527–32. dementia. American Journal of Alzheimer’s Disease and Other Dementias,
Hamer, M. and Chida, Y. (2009). Physical activity and risk of neurodegenerative 24, 476–85.
disease: a systematic review of prospective evidence. Psychological McCarthy, M., Addington-Hall, J., and Altmann, D. (1997). The experience
Medicine, 39, 3–11. of dying with dementia: a retrospective study. International Journal of
Harrison Dening, K. (2010). Admiral Nursing: offering a specialist nursing Geriatric Psychiatry, 12, 404–9.
approach. Dementia Europe, 1, 10–11. McCormack, B. and McCance, T. (2010). Person-centred nursing: theory and
Harrison Dening, K. (2011). Caring for the carers. In: V. Pace, A. Treloar, practice. Wiley-Blackwell, Oxford.
and S. Scott (eds) Dementia: from advanced disease to bereavement, Mezey, M., et al. (2000). Decision-making capacity to execute a health care
pp. 373–82. Oxford University Press, Oxford. proxy: development and testing of guidelines. Journal of the American
Harrison Dening, K., Jones, L., and Sampson, E.L. (2011). Advance care Geriatrics Society, 48, 179–88.
planning for people with dementia: a review. International Psychogeriatrics, Mitchell, S.L., Kiely, D.K., and Hamel, M.B. (2004). Dying with advanced
23, 1535–51. dementia in the nursing home. Archives of Internal Medicine, 164,
Harrison Dening, K., et al. (2012). Barriers to providing end-of-life care for 321–6.
people with dementia: a whole-system qualitative study. BMJ Supportive Mitchell, S.L., et al. (2009). The clinical course of advanced dementia. New
and Palliative Care, doi: 10.1136/bmjspcare-2011–000178 (Epub). England Journal of Medicine, 361, 1529–38.
Henry, C. and Seymour, J. (2007). Advance care planning: a guide for health Molloy, W. (2005). Let me decide. Penguin Canada, Toronto.
and social care staff. NHS End of Life Care Programme, Department of Morrison, R.S. and Siu, A.L. (2000a). Survival in end-stage dementia following
Health, London. acute illness. Journal of the American Medical Association, 284, 47–52.
Morrison, R.S. and Siu, A.L. (2000b). A comparison of pain and its treatment Sampson, E.L., et al. (2009a). Dementia in the acute hospital: prospective cohort
in advanced dementia and cognitively intact patients with hip fracture. study of prevalence and mortality. British Journal of Psychiatry, 195, 61–6.
Journal of Pain and Symptom Management, 19, 240–8. Sampson, E.L., Candy, B., and Jones, L. (2009b). Enteral tube feeding for
Mukadam, N. and Sampson, E.L. (2011). A systematic review of the older people with advanced dementia. Cochrane Database of Systematic
prevalence, associations and outcomes of dementia in older general Reviews, (2) CD007209.
hospital inpatients. International Psychogeriatrics, 23, 344–55. Sampson, E., et al. (2012). Improving end of life care for people with dementia:
Myers, A.H., et al. (1991). Hip fractures among the elderly: factors associated with a rapid participatory appraisal. BMJ Supportive and Palliative Care, doi:
in-hospital mortality. American Journal of Epidemiology, 134, 1128–37. 10.1136/bmjspcare-2011–000177 (Epub).
National Collaborating Centre for Mental Health (2007). Dementia: the Sanders, S. and Swails, P. (2009). Caring for individuals with end-stage
NICE–SCIE guideline on supporting people with dementia and their carers dementia at the end of life. Dementia, 8, 117–38.
in health and social care. British Psychological Society, Leicester, and Scherder, E., et al. (2009). Pain in dementia. Pain, 145, 276–8.
Royal College of Psychiatrists, London. Schneider, L.S., Dagerman, K.S., and Insel, P. (2005). Risk of death with
National End of Life Care Programme (2010). Care towards the end of life atypical antipsychotic drug treatment for dementia: meta-analysis of
for people with dementia: an online resource guide. NHS National End randomized placebo-controlled trials. Journal of the American Medical
of Life Care Programme, London. <http://www.endoflifecareforadults. Association, 294, 1934–43.
nhs.uk/assets/downloads/Dementia_resource___final___20101025.pdf> Schulz, R., Newsom, J.T., and Fleissner, K. (1997). The effects of bereavement
(accessed 06.06.2012). after family caregiving. Aging and Mental Health, 1, 269–82.
National Institute for Health and Clinical Excellence and National End of Shega, J.W.L. (2003). Palliative excellence in Alzheimer care efforts (PEACE):
Life Care Programme (2010). End of life care for people with dementia a program description. Journal of Palliative Medicine, 6, 315–20.
commissioning guide. NICE, London. <http://www.nice.org.uk/ Steele, C., et al. (1990). Psychiatric symptoms and nursing home placement
media/8E6/C4/CommissioningGuideEoLDementia.pdf> (accessed of patients with Alzheimer’s disease. American Journal of Psychiatry, 147,
06.06.2012). 1049–51.
Neale, R., Brayne, C., and Johnson, A.L. (2001). Cognition and survival: an Testad, I., et al. (2010). The effect of staff training on agitation and use of
exploration in a large multicentre study of the population aged 65 years restraint in nursing home residents with dementia: a single-blind,
and over. International Houranl of Epidemiology, 30, 1383–8. randomized controlled trial. Journal of Clinical Psychiatry, 71, 80–6.
Omran, A.R. (1971). The epidemiological transition: a theory of the epidemiology Thuné-Boyle, I.C.V., et al. (2010). Challenges to improving end of life care of
of population change. Milbank Memorial Fund Quarterly, 49, 509–38. people with advanced dementia in the UK. Dementia, 9, 285–98.
Ownby, R.L., et al. (2006). Depression and risk for Alzheimer disease: Treloar, A.J. (2008). Continuous deep sedation: Dutch research reflects problems
systematic review, meta-analysis, and metaregression analysis. Archives with the Liverpool care pathway. British Medical Journal, 336, 905.
of General Psychiatry, 63, 530–8. Treloar, A., Crugel, M., and Adamis, D. (2009). Palliative and end of life care
Parker-Oliver, D., Porock, D., and Zweig, S. (2004). End-of-life care in U.S. of dementia at home is feasible and rewarding: results from the ‘Hope for
nursing homes: a review of the evidence. Journal of American Medical Home’ study. Dementia, 8, 335–47.
Directors Association, 5, 147–55. Van der Steen, J.T. (2010). Dying with dementia: what we know after more
Peters, R., et al. (2008). Smoking, dementia and cognitive decline in the than a decade of research. Journal of Alzheimer’s Disease, 22, 37–55.
elderly: a systematic review. BMC Geriatrics, 8, 36. Van der Steen, J.T., et al. (2002). Withholding antibiotic treatment in
Post, S. (2006). Respectare: moral respect for the lives of the deeply forgetful. pneumonia patients with dementia: a quantitative observational study.
In: J.C. Hughes, S J. Louw, and S.R. Sabat (eds) Dementia: mind, meaning Archives of Internal Medicine, 162, 1753–60.
and the person. Oxford University Press, Oxford. Van der Steen, J.T., et al. (2005). Withholding or starting antibiotic treatment
Radbruch, L., et al. (2009). White Paper on standards and norms for hospice in patients with dementia and pneumonia: prediction of mortality with
and palliative care in Europe—Part 1: Recommendations from the physicians’ judgment of illness severity and with specific prognostic
European Association for Palliative Care. European Journal of Palliative models. Medical Decision Making, 25, 210–21.
Care, 16, 278–89. Van der Steen, J.T., et al. (2009). Discomfort in dementia patients dying from
Regnard, C., et al. (2003). Difficulties in identifying distress and its causes pneumonia and its relief by antibiotics. Scandinavian Journal of Infectious
in people with severe communication problems. International Journal of Diseases, 41, 143–51.
Palliative Nursing, 9, 173–6. Volicer, L., et al. (1994). Impact of special care unit for patients with advanced
Regnard, C., et al. (2007). Understanding distress in people with severe Alzheimer’s disease on patients’ discomfort and costs. Journal of the
communication difficulties: developing and assessing the Disability American Geriatrics Society, 42, 597–603.
Distress Assessment Tool (DisDAT). Journal of Intellectual Disability Wagner, J. T., et al. (2006). Predicting the risk of hospital admission in older
Research, 51, 277–92. persons—validation of a brief self-administered questionnaire in three
Regnard, C., et al. (2010). Gastrostomies in dementia: bad practice or bad European countries. Journal of the American Geriatrics Society, 54,
evidence? Age and Ageing, 39, 282–4. 1271–6.
Reisberg, B. (1988). Functional assessment staging (FAST). Psychopharmacology Walston, J. and Fried, L.P. (1999). Frailty and the older man. Medical Clinics
Bulletin, 24, 653–9. of North America, 83, 1173–94.
Rurup, M.L., et al. (2006). Attitudes of physicians, nurses and relatives Wills, W. and Woods, B. (1998). Report for the NHS Executive North Thames
towards end-of-life decisions concerning nursing home patients with R&D Directorate: an evaluation of the Admiral Nursing Service—an
dementia. Patient Education and Counseling, 61, 372–80. innovative service for the carers of people with dementia. HMSO, London.
Sachs, G.A., Shega, J.W., and Cox-Hayley, D. (2004). Barriers to excellent Woods, N.F., et al. (2005). Frailty: emergence and consequences in women
end-of-life care for patients with dementia. Journal of General Internal aged 65 and older in the Women’s Health Initiative Observational Study.
Medicine, 19, 1057–63. Journal of the American Geriatrics Society, 53, 1321–30.
Sampson, E.L., et al. (2005). A systematic review of the scientific evidence Xie, J., Brayne, C., and Matthews, F.E. (2008). Survival times in people with
for the efficacy of a palliative care approach in advanced dementia. dementia: analysis from population based cohort study with 14 year
International Psychogeriatrics, 17, 31–40. follow-up. British Medical Journal, 336, 258.
Sampson, E.L., et al. (2006). Differences in care received by patients with Zwakhalen, S.M., et al. (2006). Pain in elderly people with severe dementia: a
and without dementia who died during acute hospital admission: a systematic review of behavioural pain assessment tools. BMC Geriatrics,
retrospective case note study. Age and Ageing, 35, 187–9. 60, 3.
CHAPTER 29
The concept of dementia
Introduction from ageing? Should we be trying to identify subtypes of dementia

at all, especially in the oldest old? Is the use of biomarkers appro-
Alan Thomas and Tom Dening priate, given their high cost and arguably their limited diagnostic
accuracy?
After years of stability, dementia and its major causes have recently When preparing the new edition of this book the editors were
been the subjects of much debate. This has led to a surge of newly keenly aware that such debate was occurring around the world.
proposed diagnostic criteria, including in 2005—revised criteria for Consequently, rather than provide a chapter as previously on the
dementia with Lewy bodies (DLB) (McKeith et al., 2005); 2007— diagnostic criteria for dementia, the editors invited authors from
new research criteria for Alzheimer’s disease (AD) (Dubois et al., several perspectives to discuss aspects of this complex melee. These
2007) and criteria for Parkinson’s disease dementia (PDD) (Emre short contributions are not intended to provide a comprehensive
et al., 2007); and 2011—revised criteria for frontotemporal demen- coverage of these interrelated issues, but it is hoped they will stimu-
tia (FTD) (Rascovsky et al., 2011) and three new sets of criteria for late the thinking and inform the understanding of readers. They
AD (Albert et al., 2011; McKhann et al., 2011; Sperling et al., 2011). represent work in progress, and any attempt at the present time to
Ongoing and due out in 2013/14 are revised DSM and ICD criteria give a didactic account will rapidly become obsolete. Therefore we
for dementia. This eruption of interest is due to the pressure from think the reader will find these thoughtful contributions of more
several interacting developments: value. Many of the subjects they touch upon are dealt with in more
detail in the chapters that follow.
◆ The ageing world: the massive increase in older people worldwide
has inevitably been accompanied by a concomitant explosion in
the number developing dementia, doubling every 20 years dur- References
ing the first half of this century from 36 million in 2010. This has Albert, M. S., et al. (2011). The diagnosis of Mild Cognitive Impairment due
forced dementia onto the public agenda and led to an increased to Alzheimer’s disease: recommendations from the National Institute on
demand for earlier diagnosis and improved treatments. Aging–Alzheimer’s Association workgroups on diagnostic guidelines for
Alzheimer’s disease. Alzheimer’s and Dementia, 7, 270–9.
◆ Antidementia treatments: the availability of specific drug treat- Dubois, B., et al. (2007). Research criteria for the diagnosis of Alzheimer’s disease:
ments for AD has increased the awareness of AD in particular revising the NINCDS-ADRDA criteria. Lancet Neurology, 6, 734–46.
and encouraged more and more people to seek diagnosis and Emre, M., et al. (2007). Clinical diagnostic criteria for dementia associated
help earlier. with Parkinson’s disease. Movement Disorders, 22, 1689–707.
McKeith, I. G., et al. (2005). Diagnosis and management of dementia with Lewy
◆ Mild Cognitive Impairment: although it has long been recog-
bodies: third report of the DLB Consortium. Neurology, 65, 1863–72.
nized that the boundaries between normal ageing, milder forms
McKhann, G. M., et al. (2011). The diagnosis of dementia due to Alzheimer’s
of cognitive impairment, and dementia are difficult to deline- disease: Recommendations from the National Institute on Aging–
ate, these pressures have driven a focus on the transition zone Alzheimer’s Association workgroups on diagnostic guidelines for
between normality at one end and dementia at the other end, and Alzheimer’s disease. Alzheimer’s and Dementia, 7, 263–9.
the diagnosis ‘Mild Cognitive Impairment’ has become much Rascovsky, K., et al. (2011). Sensitivity of revised diagnostic criteria for the
more widely used. behavioural variant of frontotemporal dementia. Brain, 134, 2456–77.
Sperling, R. A., et al. (2011). Toward defining the preclinical stages of
◆ Biomarkers: alongside these changes is the increasing study and
Alzheimer’s disease: recommendations from the National Institute on
use of more technological investigations to try to improve the Aging–Alzheimer’s Association workgroups on diagnostic guidelines for
accuracy of the diagnosis of different subtypes of dementia by Alzheimer’s disease. Alzheimer’s and Dementia, 7, 280–92.
relating it more closely to its pathology.
The upshot is that the concept of dementia itself and its rela-
tionship to its causal pathologies have been challenged in several
Part 1
ways including: Where is its ceiling in terms of both the severity Dementia: What Is It and Can We
of cognitive impairment and the amount of impact on everyday
living needed to define dementia? Should dementia continue to
Diagnose It?
include memory impairment as a mandatory criterion or should a This section is written by Peter Whitehouse and Danny George.
broader definition be allowed? Should dementia be discarded as a While the phenomenology of dementia was first recorded in
term because of negative connotations? Is AD really distinguishable ancient Egyptian writing, the ancient Greeks are often credited
with having recognized and formulated a concept akin to demen- younger individuals created a condition in the category of so-called
tia. Memory loss and general intellectual decline, as early symp- presenile dementia occurring in people under the age of 65.
toms of the aging process, were recognized through the ages in the While our attempts to understand the brain changes in a variety
writings of Aristotle, Galen, Hippocrates, Lucretius, Cicero, and of dementias have expanded dramatically in modern times, we have
both the Elder Seneca and the Younger Seneca in the early Roman become increasingly confused about the term itself. When it was first
Empire. The concept of dementia—a term said to have been coined developed for clinical usage, ‘dementia’ was used to describe cognitive
by Celsus in the first century AD—has long carried social implica- problems in schizophrenia—so-called dementia praecox. Currently,
tions for those so diagnosed, and has been associated with reduced the so-called negative symptoms or cognitive problems in schizophre-
civilian and legal competence, as well as with entitlement to sup- nia have resurfaced as an issue. However, it is progressive neurode-
port and protection. According to the writings of Solon and Plato, generative conditions that have placed dementia on the map, partly as
mentally impaired older people were incapable of making a will, a result of the growing number of older people at risk for these con-
were not eligible for official civilian positions, but also could not ditions in the developed and developing world in the late twentieth
be charged with unlawful acts (Kurz and Lautenschlager, 2010). and early twenty-first centuries. Formally speaking, dementia can be
Similarly, in the Roman era, the concept of dementia was used to defined in an adult of any age and also in somebody who has a static
reduce the abilities of patients to enter contracts, handle their own global cognitive impairment rather than a progressive one. Head inju-
affairs, hold public office, and be criminally responsible. Modern ries can, for example, be the cause of generalized cognitive impair-
legal systems contain similar provisions for those diagnosed with ment but do not progress over time after the initial insult.
dementia and judged to have lost capacity and competence. The concept of Alzheimer’s disease often dominates public and
Literally meaning ‘away’ or ‘out’ of ‘mind’ or ‘reason’ in Latin, even scientific and clinical discourse about dementia. Clinicians
the term ‘dementia’ entered the English language from the French are frequently asked what the difference is between dementia and
démence via the French psychiatrist Philippe Pinel, who made nota- Alzheimer’s. Classically, the answer is that dementia is the broad
ble contributions to the categorization of mental disorders in the superordinate category and Alzheimer’s is one specific cause.
late-eighteenth and early nineteenth centuries. Over the centuries, However, the frequency with which the question is asked about
the phenomenology of dementia has been causally associated with differentiating dementia and Alzheimer’s, despite much effort to
witchcraft, moral degeneracy, bad blood, and a dissipation of vital educate people about the classic view, may represent some com-
energy from the brain, amongst other factors. mon wisdom that these terms are more confused and confusing
The term ‘dementia’ was first used in the English language medi- that experts are willing to admit.
cal literature in the middle 1800s. One of the earliest usages was by The clinical and scientific worlds are increasingly being influ-
James C. Pritchard in his textbook A Treatise of Insanity and Other enced by this confusion and classification in terminology. New
Disorders Affecting the Mind first published in 1835. He makes refer- guidelines have been issued that relate to the diagnosis of so-called
ence to Etienne Dominique Esquirol, a psychiatric contemporary of Alzheimer’s disease in clinical and research settings (Dubois et al.,
Pinel in the influential French school, as offering the best conception 2010; McKhann et al., 2011; Hyman et al., 2012). While Alzheimer’s
of dementia. Pritchard refers to dementia as a form of ‘incoherence’ disease is said to be the most common form of dementia, it is
that involves impairments of memory, reasoning, and other cog- increasingly recognized that it is not a single entity but rather het-
nitive abilities. At that time, there was no clear-cut differentiation erogeneous sets of syndromes likely caused by many biological fac-
between the fields of psychiatry and neurology; both were evolving tors that affect the brain as it ages. Even Dr Alzheimer was not sure
and influencing each other and emerging as separate fields. he described a separate disease, and wrote that there was ‘no tenable
In the nineteenth century, the understanding of the brain changes reason to consider [his initial observed cases] as caused by a specific
in dementia was severely limited by the inability to section and stain disease process’ (Alzheimer, 1911). In fact, in these first two reported
tissues histologically. Dementia was differentiated from ‘amentia’, a cases of so-called Alzheimer’s there were arguments about the rela-
term that referred to individuals born with mental retardation who tive importance of plaques and tangles—arguments that continue
never had normal intellectual abilities. Those with dementia devel- to this day. More controversial in the new guidelines is the claimed
oped an acquired form of cognitive incapacity wherein more than relationship between Alzheimer’s disease and aging. Despite billions
one intellectual domain was affected. Dementia was differentiated of dollars being spent on the condition over the past several decades,
from acute confusional states or delirium, as well as conditions like there are no diagnostic tests or even neuropathological changes that
stroke with more focal symptoms such as aphasia or amnesia in can precisely define the difference between somebody whose brain is
which language and memory were affected in relative isolation. aging and somebody who has Alzheimer’s (Whitehouse and George,
At the turn of the twentieth century, brain psychiatry was devel- 2008). In fact, the most recent neuropathological guidelines—rather
oping as a field, and clinical and pathological techniques were than making neuropathology ‘definite’ as biomedicine has been say-
developed to try to discover the biological substrate of psychiatric ing for years—now ‘disentangle’ pathology from clinical features
symptomatology. For example, the renowned psychiatrist and neuro- (Hyman et al., 2012). Several large-scale epidemiological studies
anatomist Theodore Meynert wrote a seminal book in 1874 entitled with autopsies demonstrate that plaques and tangles occur in peo-
Psychiatry: Diseases of the Forebrain. Many competing schools, par- ple without significant dementia. Moreover, the neuropathology
ticularly in Germany, attempted to define syndromes that were based guidelines also point out that much overlap exists in actual cases of
on clinical pathological correlation. Most notably perhaps, Dr Alois so-called Lewy body, vascular, and frontal types of dementias (see
Alzheimer, a German psychiatrist who worked first in Frankfurt and Chapter 7). Hence, once again, the discrete nature of a singular form
then in Munich, described what his department chair Emil Kraepelin of dementia called Alzheimer’s is being challenged.
eventually called ‘Alzheimer’s disease’ in his influential book of psy- More controversial is the proliferation of new labels for people
chiatry of 1910. The combination of plaques and tangles occurring in that do not have functional impairment but still have some cognitive
CHAPTER 29 the concept of dementia 373
difficulties such as ‘Mild Cognitive Impairment’ and now ‘preclini- the negotiation between physicians and society for the appropri-
cal’ or ‘asymptomatic Alzheimer’s disease’. Two new proposed but ate labelling of individuals with diffuse cognitive impairment. In
inconsistent research guidelines on asymptomatic Alzheimer’s light of the changing historical circumstance, it is hoped that a
disease referenced above suggest that we need to use a variety of more careful examination of this word will lead to a greater under-
cerebral spinal fluid and neuroimaging tests to identify those who standing and better social response to this huge epidemiological
are likely to come down with Alzheimer’s disease (see Chapter 28 challenge. The lessons of Alzheimer’s disease over the past century
and Part 3 of this chapter for details). However, these tests are not teach us that medicalizing brain ageing can be dangerous because
well standardized, sometimes not reliable, and now difficult to vali- it guides us towards a fixation on biological approaches rather than
date, given that we do not really know what Alzheimer’s disease is, community/public health-based approaches that might help socie-
even at autopsy. Even the clinical observations of these two sets of ties adapt to the needs and remaining capacities of cognitively chal-
guidelines are in some opposition. The Dubois et al. criteria claim lenged individuals. A prudent approach might focus on developing
that a specific episodic memory problem characterizes the demen- community and policy responses that allow individuals with vary-
tia of Alzheimer’s, whereas the National Institute on Aging and the ing degrees and kinds of cognitive impairment to remain integrated
Alzheimer’s Association workgroups (NIA/AA) guidelines specifi- in society as much as possible.
cally address what they perceive to be an overemphasis on memory Indeed, there is a growing international focus on intergenera-
in the diagnosis. tional relationships, and community organizations, assisted liv-
It is actually becoming increasingly clear that older people with ing homes, and other institutions are demonstrating the benefits
progressive cognitive impairment should be identified as having a of young–old partnerships. One example is the Intergenerational
mixed dementia. MRI scans on practically all people over the age School in Cleveland where individuals with dementia go to school
of 70 show a combination of atrophy and nonspecific white mat- alongside children, which is now being replicated elsewhere. Mutual
ter changes consistent with ischaemia (see Chapters 6 and 12). support and learning through storytelling and other classroom
Hence, it is fair to say that we can no longer clearly differentiate and community-based activities leads to better outcomes for both
degenerative dementias from vascular dementias. Many other stud- kids and seniors (George and Singer, 2011). This is the kind of
ies suggest a considerable overlap between these two conditions, cost-effective, integrated solution that can help us make progress on
particularly with regards to risk factors. Moreover, separation of the challenges of dementia, and is far more powerful than the prom-
frontal lobe dementias and dementias with parkinsonism also is ise of medical panaceas for what are proving to be a family of hetero-
proving increasingly difficult, despite the use of new genetic and geneous, age-related conditions wrongly labelled a single disease.
neuroimaging approaches.
The word ‘dementia’ itself is also under scrutiny because in a
number of languages—including English, French, German, Japanese, References
Korean, and Chinese—‘dementia’ carries a pejorative and often stig- Alzheimer, A. (1911). Über eigenartige Krankheitsfälle des späteren Alters.
matizing connotation. It can be used in common parlance to refer Zeitschrift für die gesamte Neurologie und Psychiatrie, 4, 356–85.
to something or someone behaving badly or stupidly. Consequently, (Translated and with an introduction by H. Förstl and R. Levy, 1991.)
some professional and lay organizations across multiple cultures are American Psychiatric Association DSM-5 Development (2010). Delirium,
attempting to retire the word dementia or adapt its meaning. For dementia, amnestic, and other cognitive disorders. <http://www.dsm5.
instance, the American Psychological Association’s Neurocognitive org/ProposedRevisions/Pages/Delirium,Dementia,Amnestic,OtherCog
nitive.aspx> (accessed 22.08.2010).
Disorders Working Group (2010) announced that it proposed
Dubois, B., et al. (2010). Revising the definition of Alzheimer’s disease: a new
replacing the diagnostic category ‘delirium, dementia, amnestic, and
lexicon. Lancet Neurology, 9, 118–27.
other cognitive disorders’ with ‘delirium, Major neurocognitive dis- George, D.R. and Singer, M. (2011). Intergenerational volunteering and
order, and minor neurocognitive disorder’. This move would effec- quality of life for persons with mild to moderate dementia: results from
tively replace ‘dementia’ with ‘neurocognitive disorders’ of the Major a 5-month intervention study in the United States. American Journal of
and Minor types, the latter being distinguished as a state of prede- Geriatric Psychiatry, 19(4), 392–6.
mentia when a subject is without significant functional impairment Hyman, B.T., et al. (2012). National Institute on Aging—Alzheimer’s
in activities of daily living. The boundary between Minor and Major Association guidelines for the neuropathologic assessment of Alzheimer’s
may hinge on assessment of function in daily life which is influenced disease. Alzheimer’s and Dementia: The Journal of the Alzheimer’s
Association, 8(1), 1–13.
by previous (premorbid) skills, the healthcare professional’s exper-
Kurz, A.F. and Lautenschlager, N.T. (2010). The concept of dementia: retain,
tise and biases, gender roles, and cultural expectations.
reframe, rename, or replace? International Psychogeriatrics, 22(1),
A more comprehensive societal approach has been used to 37–42.
address this issue in Japan where—through a process of negotiation McKhann, G.M., et al. (2011) The diagnosis of dementia due to Alzheimer’s
between the government, professionals, and the lay public—the disease: Recommendations from the National Institute on Aging–
word ‘dementia’ (chiho) was changed to ninchisho (Miyamoto et al., Alzheimer’s Association workgroups on diagnostic guidelines for
2011), a concept whose meaning approximates ‘cognitive condition Alzheimer’s disease. Alzheimer’s and Dementia: The Journal of the
or syndrome’. As chiho carries the meaning of a ‘disease of cognition Alzheimer’s Association, 7(3), 263–9.
associated with idiocy’, it is not difficult to see why an alternative Miyamoto, M., George, D.R., and Whitehouse, P.J. (2011). Government,
professional and public efforts in Japan to change the designation of
conceptual framework would be beneficial to an increasingly aging
dementia (chihō). Dementia, 10, 475–86.
Japanese society.
Whitehouse, P.J. and George, D.R. (2008). The myth of Alzheimer’s: what you
Ultimately, it is clear that the word ‘dementia’ has evolved and aren’t being told about today’s most dreaded diagnosis. St. Martin’s Press,
continues to evolve in medical parlance. It is a concept that reflects New York.
Dominique Esquirol (1772–1840) listed dementia as one basic

Part 2 form of mental illness, others being melancholia, mania, and idi-
A Brief History of the Concept of Dementia ocy. His description of the clinical features comes very close to
present-day diagnostic criteria. It includes impairment of atten-
This section is written by Alexander F. Kurz and Nicola T.
tion, abstract thinking, memory (particularly for recent events),
Lautenschlager.
motivation, and emotional lability. The debate about the precise
psychopathological contour of the syndrome continued during the
The premedical era nineteenth century.
Decline of intellectual ability associated with behavioural change The search for anatomical causes
and reduced psychosocial competence is a pattern of mental
alteration that has been recognized with remarkable consistency In the middle of the nineteenth century, the discipline of neu-
throughout the ages and across cultures. For Solon (600 bc) it was ropathology began to prosper. Using light microscopy and novel
a cause of incapacity for making a will, for Plato (424–348 bc) a staining techniques, researchers set out to uncover the anatomical
reason to excuse unlawful behaviour, and for Cicero (106–43 bc) substrates of mental disorders. In people who developed demen-
an aftermath of a frivolous lifestyle (Torack, 1983). The condition tia late in life but did not show psychological problems previously
has been given different names including amentia, imbecillity, the syndrome was typically associated with degenerative changes
morosis, and anoea, and has been attributed to various causes of brain parenchyma or arteriosclerosis. The British physician
including diseases and old age (Berrios, 2010). Decimus Junius Samuel Wilks was the first to describe brain atrophy as an ana-
Juvenalis (60–127 ac) provided a satirical description: ‘But a tomical feature of senile dementia (Wilks, 1859). Contrary to
greater unhappiness than the loss of limbs is that he does not the belief that gradual decay of cortical neurons was the cause
know so much as the names of his own servants nor the face of atrophy, the Viennese neurologist Emil Redlich demonstrated
of a friend with whom he supped the night before; he forgets plaques as a histopathological correlate (Redlich, 1898). The evi-
his children whom he got and brought up’ (Juvenalis, 1776). In dence provided by neuropathological investigations served to
medical literature, intellectual decline in old age was not a prom- distinguish between mental disorders that lacked an anatomical
inent topic before 1650. This neglect may have several reasons. cause and those that were due to identifiable brain disease. For the
The prevalence of the condition was low due to short average latter, Emanuel Mendel (1839–1907) proposed the term ‘organic’
life-expectancy, knowledge about the brain and its functions was (Mendel, 1902).
poor, and the metaphysical position of Christian theology that
The origins of the modern concept
the soul was ageless stood against scientific discovery (Schäfer
and Karenberg, 2005). Only after René Descartes (1596–1650) The best-known association between Emil Kraepelin (1865–
had separated the immaterial mind from the physical brain, 1926) and dementia is the baptism of the presenile disease that
changes in mental functioning associated with ageing or disease Alois Alzheimer (1964–1915) had discovered in his department
became an object of medical reasoning and research. The link (Alzheimer, 1907). However, Kraepelin contributed more to the
between this condition and the label ‘dementia’ formed in the field than creating this icon. In his writing he was concerned with
middle of the nineteenth century. Originally the term had been the brain localization, dimensional nature, and psychopathological
coined by the Roman encyclopaedist Cornelius Aulus Celsus heterogeneity of dementia. In anatomical respect, Kraepelin noted
(25 bc–50 ac) for prolonged hallucinatory states. that senile dementia is due to widespread brain lesions. With regard
to psychopathology, he postulated a continuum of age-associated
mental weakness featuring impaired comprehension and mental
Early outlines flexibility, memory lapses particularly for recent events, reduced
The English physician Thomas Willis (1621–1675) considered intake and processing of new information, and disturbed atten-
decline of intellect and judgement to be a common outcome of tion (Kraepelin, 1899). On this continuum, senile dementia was an
various medical causes, including disease of the body as well as extreme form that had a significant impact on everyday living and
inherited and age-related diseases of the brain. William Cullen independence (Kraepelin, 1908). In the eighth edition of his influ-
(1710–1790), a Scottish physician and chemist, highlighted impair- ential book he argued that dementia was not a uniform syndrome
ments of memory and judgement as features of ‘amentia’ and listed but showed considerable variability, reflecting different underlying
inheritable conditions, trauma, infection, and old age as possible diseases (Kraepelin, 1910). The Swiss psychiatrist Eugen Bleuler
causes. A more fine-graded psychopathological description of the (1857–1939) shared Kraepelin’s views on the brain localization of
condition was achieved in French psychiatry in the middle of the dementia and on the dimensional nature of the syndrome. Bleuler’s
eighteenth century. It was defined as a typical cluster of symptoms important observation was that diffuse cortical atrophy or general
that distinguished it from other major forms of mental illness such impairmant of the cerebral cortex result in a uniform psychopatho-
as imbecillity, delirium, and mania. The social and legal impli- logical pattern which he initially called ‘organic symptom complex’
cations were the same as in Roman times; those affected by the (Bleuler, 1916) and later ‘psycho-organic syndrome’. Dementia was
condition were considered incapable of entering contracts, sign- a particularly severe expression of this syndrome. According to
ing wills, managing their own affairs, being members of a jury, or Bleuler, changes of memory or personality were still subtle at its
holding public positions (Berrios, 2010). The influential physician onset but gradually became manifested in reduced functional abil-
Philippe Pinel (1745–1826) replaced Cullen‘s term ‘amentia’ by the ity. Since it was unclear at which degree of memory impairment
French word démence and thus introduced the term of dementia or emotional lability dementia begins, Bleuler proposed prag-
into modern medical nomenclature. Pinel’s student Jean Étienne matic criteria for defining the boundary. Kraepelin’s and Bleuler’s
construction designed the concept along two axes: the qualitative alarming (Werner, 1995a). Post-war field studies, however, showed
axis was represented by the multiplicity of cognitive and behav- much higher prevalence rates which increased exponentially with
ioural symptoms, and the quantitative axis was defined as the age, exceeding 30% in the oldest old (Werner, 1995b). In parallel,
degree of disability in everyday living. researchers resumed the debate on the commonalities between the
presenile disease and senile dementia that had been initiated at the
Psychopathological heterogeneity time of Alois Alzheimer (Alzheimer, 1911; Simchowitz, 1911), con-
A peculiar variant of dementia showing prominent forgetfulness, cluding that there were no principal histopathological or clinical
slowing of thought, and personality change was observed in asso- differences (Albert, 1963; Lauter and Meyer, 1968). The unitarian
ciation with several brain diseases that affected subcortical brain position and the epidemiological perspective were brought together
areas (Wilson, 1912; van Bogaert and Bertrand, 1929; Stern, 1939). by Robert Katzman (1925–2008) in his famous editorial where he
The term ‘subcortical dementia’ was first used by Franz Günther called Alzheimer’s disease a ‘major killer’ (Katzman, 1976). The
Ritter von Stockert (1899–1967) for mental disorders in encepha- merger alerted governments and medical bodies to the demographic
litis lethargica (von Stockert, 1932). This symptom pattern was shift and the dementia epidemic it entails (Henderson, 1983) and
later also found in progressive supranuclear palsy (Albert et al., released increasing amounts of research money. Additional pres-
1974). The clinical appearance of subcortical dementia was refor- sure came from Alzheimer’s Associations that formed in many
mulated in modern terms by Jeffrey Cummings, who highlighted countries to raise public awareness and to promote the develop-
deficits in motivation, mood, timing, and arousal (Cummings, ment of specific services. Presently, dementia is recognized as a
1986). Due to aetiological and clinical overlap between cortical public health priority, and national dementia plans have been and
and subcortial dementia, however, the distinction has remained are being developed. The main objectives of these strategies are to
controversial (Turner et al., 2002). Other atypical forms of demen- improve knowledge about dementia, to provide better education
tia associated with circumscribed brain atrophy were described by and training for professionals, to ensure that cognitively impaired
Arnold Pick (1851–1924) at Prague University, including variants older adults are properly diagnosed, and to improve treatment
with dysproportional aphasia (Pick, 1901), apraxia (Pick, 1906), and counselling for people with dementia and their carers (Burns,
and behavioural abnormalities (Pick, 1904). The characteristic his- 2009).
topathology of frontal lobar degeneration was first described by
Erwin Stransky (1877–1962) (Stransky, 1903) and Alois Alzheimer Contemporary psychiatric classifications
(Alzheimer, 1911). A full account of the symptomatology and In the second edition of the Diagnostic and Statistical Manual
course of frontal lobe dementia was contributed by Carl Schneider of Mental Disorders (DSM-II) Bleuler’s organic brain syndrome
(1891–1946) (Schneider, 1927). In the following decades, the was adopted as a uniform symptom cluster that was believed to
frontal-lobe type of dementia was an uncommon theme in neu- result from diffuse impairment of brain tissue and was character-
ropsychiatric research until its rediscovery by Lars Gustafson in ized by global cognitive impairment, including memory, intellec-
Lund (Gustafson, 1987) and David Neary in Manchester (Neary tual functions, and judgement, as well as lability or shallowness
et al., 1988). of affect. Dementia was defined as a particularly severe degree of
the syndrome that grossly compromised the ordinary demands of
The continuum of cognitive impairment life (American Psychiatric Association, 1968). In the eighth and
In accordance with Kraepelin’s and Bleuler’s concept of a contin- ninth revisions of the World Health Organization’s International
uum between age-related intellectual decline and senile dementia, Classification of Diseases (ICD), dementia was also classified as
Viktor Kral (1903—1988) described the condition of a ‘malignant’ a severe organic brain condition (World Health Organization,
form of senile forgetfulness. It was associated with greater func- 1978). In DSM-III (American Psychiatric Association, 1980), the
tional impairment, more behavioural problems, higher mortality, formerly uniform organic brain syndrome was split into several
and greater likelihood of institutionalization than ‘benign’ mem- clusters of psychopathological symptoms, one being dementia.
ory disorder in old age. As underlying causes of the progressive Similarly, ICD-10 distinguishes several organic mental disorders
variant of forgetfulness, he assumed pathological brain changes including dementia, delirium, amnesia, and others (World Health
exceeding involutional alterations, particularly Alzheimer’s dis- Organization, 2007). In both classification systems the qualita-
ease (Kral, 1978). Kral’s postulate of a prodromal stage of the tive and quantitative essentials of Kraepelin’s and Bleuler’s con-
disease was echoed in the development of the Clinical Dementia struction were preserved, since dementia was defined by decline
Rating (Hughes et al., 1982), in the detailed description of the from a previous level in more than one cognitive domain, includ-
clinical evolution of Alzheimer’s disease (Reisberg et al., 1988), ing memory, that interferes with everyday activities. Apart from
and in the conceptualization of ‘Mild Cognitive Impairment’ dropping the term ‘organic’, the DSM-IV introduced an important
(Petersen et al., 1999). conceptual change by lowering the pragmatic threshold for diag-
nosis. Limitations of complex activities, such as going to school,
Discovering the epidemic working, and handling finances, were considered as sufficient
For several decades after Kraepelin’s powerful statement, evidence for an impact on day-to-day living. The definition of
Alzheimer’s disease was looked at as a rare neurological disease dementia included in the revised criteria of the National Institute
that occurred in the presenium, while senile dementia was largely on Aging and the Alzheimer’s Association workgroups (NIA–AA)
ignored as an ill-defined condition between senility and insuffi- for Alzheimer’s disease is analogous (McKhann et al., 2011). Due
cient cerebral blood supply. Due to poor methodology, epidemio- to the inclusion of states with subtle effects on the ability to man-
logical data obtained prior to World War II on the prevalence of age day-to-day problems, the diagnosis of dementia acquires an
‘senile psychosis’ that encompassed dementia were limited and not individualistic note, since the degree of impairment in everyday
living caused by cognitive dysfunction is variable and depends on attempts are currently being made to compensate for the limited
the particular social setting of the person, including the amount of specificity of the concept, particularly when used for subtle degrees
support received. of cognitive and functional deterioration, by laboratory and imag-
ing examinations that directly identify the underlying neurodegen-
Clinical importance erative, cerebrovascular, or metabolic disease (Dubois et al., 2007).
The concept of dementia implies that identifying this particular Whilst some of these techniques have high sensitivity for detecting
symptom pattern is an important part of diagnostic reasoning. the pathology of the Alzheimer’s type, their ability to discriminate
It guides the clinician to chronic, often progressive, and usually between different brain diseases is unsatisfactory. Even if improve-
nonfocal neurodegenerative, cerebrovascular, or metabolic brain ments may be expected from future research, the application of
disease. Although the symptom pattern varies according to the these diagnostic tools will probably remain limited to specialized
localization of the underlying pathology, it is sufficiently specific to institutions and to certain healthcare systems and will not replace
allow a differentiation from purely ageing-related cognitive changes the concept of dementia in diagnosis globally. Since biomarkers
(O’Connor et al., 1996) and other symptom clusters such as delir- indicate brain pathology and not its clinical consequences, they can
ium (Laurila et al., 2004), amnesia, aphasia, or apraxia, which arise also not substitute the concept as a signal for medical and social
from acute or focal pathologies. Additional physical, laboratory, or intervention, unless the aim is prevention of emergence of clini-
neuroimaging examinations are required to identify the underly- cal symptoms. As a major drawback, the current formulation of
ing cause. The concept of dementia also implies that the underly- dementia is modelled on Alzheimer’s disease. Therefore it overem-
ing pathology has reached a level of clinical severity where affected phasizes impairment of memory and inadequately covers variants of
individuals and their proxies require immediate and long-standing the syndrome that arise from frontotemporal lobar degenerations,
treatment and support (Hogan et al., 2008). Currently, dementia is cerebrovascular diseases, and subcortical pathologies (Erkinjuntti,
also a prerequisite for the prescription of specific pharmacologi- 2003; Lautenschlager and Förstl, 2007). In order to accommodate
cal treatments (Christensen and White, 2007) and, in some coun- psychopathological heterogeneity, the concept of dementia needs
tries, a gatekeeper for government subsidy schemes. In addition, to be enlarged along its qualitative axis. To enable the distinction
limitations on functional abilities, reduced personal autonomy, and of subtypes, the definition of a general dementia syndrome could
behavioural changes can provide an indication for and justify the be complemented by several sets of criteria for the major psycho-
reimbursement of nonpharmacological interventions such as occu- pathological variants, e.g. dementia of the Alzheimer’s or fronto-
pational therapy. temporal type (Rascovsky et al., 2011).
Scientific progress Conclusion

The concept of dementia has proven valuable over decades as a The medical concept of dementia formed in the seventeenth cen-
diagnostic landmark of major brain pathologies in old age and tury and attained its classic formulation in the early days of neu-
as a trigger of healthcare interventions for patients and carers. Its ropathological brain research around 1900. Over many decades it
usability is challenged, however, by scientific advances regarding has proven to be clinically useful as a flag for certain chronic brain
diagnosis, treatment, and aetiological diversity. Growing public diseases and as an indicator of the need for medical and social inter-
awareness of cognitive decline and its personal as well as social vention. During the evolution of psychiatric classification systems,
consequences, improvement of diagnostic techniques, and the the concept has been gradually expanded to cover subtle degrees
hope for novel disease-modifying treatments drive the call for an of cognitive and functional impairment, but has retained a focus
early identification of progressive neurodegenerative and cerebrov- on memory impairment. Currently, the concept is challenged by
ascular diseases. The classic symptom cluster of cognitive impair- scientific progress regarding diagnosis, treatment, and aetiological
ment on multiple domains accompanied by behavioural change heterogeneity of neurodegenerative and cerebrovascular diseases.
and interfering with the ability to function independently on eve- It should be complemented by measures that allow an early identi-
ryday tasks is not suitable as a flag for early developmental stages fication of the underlying pathologies and it should be modified to
of the underlying pathologies, because deficits may only be present cover the full spectrum of their clinical expression.
in one domain and limitation of functional ability may be minimal
or absent. It appears possible to expand the concept on its quan- References
titative axis to an extent that these states are also covered. Then, Albert, E. (1963). Senile Demenz und Alzheimer. Die gleiche Krankheit?
however, the barriers would be pulled down that separate dementia Zeitschrift für die gesamte Neurologie und Psychiatrie, 172, 264.
from adjacent concepts such as amnesia, aphasia, apraxia, and per- Albert, M. L., Feldman, R. G. and Willis, A. L. (1974). The ‘subcortical
sonality change that imply entirely different aetiologies, treatments, dementia’ of progressive supranuclear palsy. Journal of Neurology and
and outcomes. Thus, the concept would forfeit much of its role as a Neurosurgery Psychiatry, 37, 121–30.
diagnostic and therapeutic guide. The category of ‘neurocognitive Alzheimer, A. (1907). Über eine eigenartige Erkrankung der Hirnrinde. Allg
disorder’ in the draft version of the DSM-V which includes ‘Major’ Z Psychiatr , 64, 146–8.
and ‘Minor’ degrees of severity may be viewed as such an inflation. Alzheimer, A. (1911). Über eigenartige Krankheitsfälle des späteren Alters.
Zeitschrift für die gesamte Neurologie und Psychiatrie, 4.
In addition to losing specificity, the concept would become applica-
American Psychiatric Association (1968). Diagnostic and statistical manual
ble to individuals who still have available much of their intellectual of mental disorders, 2nd edition. American Psychiatric Association,
capacity and functional competence (Sachdev, 2000). For them, the Washington, DC.
label ‘dementia’ is particularly pejorative and frightening, so that the American Psychiatric Association (1980). Diagnostic and statistical manual
change in terminology appears to be appropriate although it may of mental disorders, 3rd edition. American Psychiatric Association,
eventually not reduce stigma (George et al., 2011). Considerable Washington, DC.
Berrios, G. (2010). Dementia: historical overview. In: Ames, D., Burns, A., Rascovsky, K., et al. (2011). Sensitivity of revised diagnostic criteria for
and O’Brien, J. (eds) Dementia. Hodder Arnold, London. the behavioural variant of frontotemporal dementia. Brain, 134 ,
Bleuler, E. (1916). Lehrbuch der Psychiatrie. Springer, Berlin. 2456–77.
Burns, A. (2009). Another nail in the coffin of the cognitive paradigm of Redlich, E. (1898). Über miliare Sklerose der Hirnrinde bei seniler Atrophie.
dementia. British Journal of Psychiatry, 194, 199–200. Jahrbücher für Psychiatrie und Neurologie, 17, 208–16.
Christensen, M. D. and White, H. K. (2007). Dementia assessment and Reisberg, B., et al. (1988). Stage-specific behaviora, cognitive, and in vivo
management. Journal of the American Medical Directors Association, 8, changes in community residing subjectswith age-associated memory
e89–98. impairment and primary degenerative dementia of the Alzheimer type.
Cummings, J. L. (1986). Subcortical dementia. Neuropsychology, neuro- Drug Development Research, 15, 101–14.
psychiatry, and pathophysiology. British Journal of Psychiatry, 149, 682–97. Sachdev, P. (2000). Is it time to retire the term ‘dementia’? Journal of
Dubois, B., et al. (2007). Research criteria for the diagnosis of Alzheimer’s disease: Neuropsychiatry and Clinical Neuroscience, 12, 276–9.
revising the NINCDS-ADRDA criteria. Lancet Neurology, 6, 734–46. Schäfer, D. and Karenberg, A. (2005). Alter, Krankheit und Demenz:
Erkinjuntti, T. (2003). Subcortical ischemic vascular disease and dementia. Historische Anmerkungen zu einem aktuellen Thema. Z Med Ethik , 51,
International Psychogeriatrics, 15, 23–6. 13–25.
George, D. R., Whitehouse, P. J., and Ballenger, J. (2011). The evolving Schneider, C. (1927). Über Picksche Krankheit. Mschr Psychiat Neurol , 65,
classification of dementia: placing the DSM-V in a meaningful historical 230–75.
and cultural context and pondering the future of ‘Alzheimer’s’. Culture of Simchowitz, T. (1911). Histologische Studien über die senile Demenz. In:
Medicine and Psychiatry, 35, 417–35. Nissl, F. and Alzheimer, A. (eds) Histologische und histopathologische
Gustafson, L. (1987). Frontal lobe degeneration of non-Alzheimer type. II. Arbeiten. Fischer, Jena.
Clinical picture and differential diagnosis. Archives of Gerontology and Stern, K. (1939). Severe dementia associated with bilateral symmetrical
Geriatrics, 6, 209–23. degeneration of the thalamus. Brain, 62, 157–71.
Henderson, A. S. (1983). The coming epidemic of dementia. Australian and Stransky, E. (1903). Zur Lehre von den aphasischen, asymbolischen und
New Zealand Journal of Psychiatry, 17, 117–27. katatonen Störungen bei Atrophie des Gehirns. Mschr Psychiat Neurol,
Hogan, D. B., et al. (2008) Diagnosis and treatment of dementia: 4. Approach 13, 464–87.
to management of mild to moderate dementia. Canadian Medical Torack, R. M. (1983). The early history of senile dementia. In: Reisberg,
Association Journal, 179, 787–93. B. (ed.) Alzheimer’s disease. The standard reference. The Free Press,
Hughes, C. P., et al. (1982). A new clinical scale for the staging of dementia. New York.
British Journal of Psychiatry, 140, 566–72. Turner, M. A., Moran, N. F., and Kopelman, M. D. (2002). Subcortical
Juvenalis (1776). The satires of Juvenal (translated). Faulkner, Dublin. dementia. British Journal of Psychiatry, 180, 148–51.
Katzman, R. (1976). The prevalence and malignancy of Alzheimer disease. A Van Bogaert, L., and Bertrand, I. (1929). Une variété d’atrophie olivo-pointe
major killer. Archives of Neurology, 33, 217–18. à évolution subaigue avec troubles démentiels. Reviews in Neurology, 36,
Krapelin, E. (1899). Psychiatrie. Ein Lehrbuch für Studirende und Ärzte. 6. 165–78.
Auflage, Band II: Klinische Psychiatrie. Barth, Leipzig. Von Stockert, F. G. (1932). Subcorticale Demenz. Ein Beitrag zur
Krapelin, E. (1908). Psychiatrie. Ein Lehrbuch für Studirende und Ärzte, 8. encephalitischen Denkstörung. Archive für Psychiatrie, 97, 77–100.
Aufl.. Barth, Leipzig. Werner, B. (1995a). On the epidemiology of dementia in the 20th century.
Krapelin, E. (1910). Lehrbuch der Psychiatrie, 8. Auflage. Barth, Leipzig. Part 1: epidemiology as part of general psychiatric epidemiology. Journal
Kral, V. A. (1978). Benign senescent forgetfulness. I: Katzman, R., Terry, R. D., of Public Health, 3, 37–50.
and Bick, K. L. (eds) Alzheimer’s disease: senile dementia and related Werner, B. (1995b). On the epidemiology of dementia in the 20th century.
disorders. Raven Press, New York. Part 2: epidemiology of dementia: the process of specialization after
Laurila, J. V., et al. (2004). Detection and documentation of dementia and 1945. Journal of Public Health, 3, 156–85.
delirium in acute geriatric wards. General Hospital Psychiatry, 26, 31–5. Wilks, S. (1859). Lectures on pathological anatomy. Longman, London.
Lautenschlager, N. T. and Frstl, H. (2007). Personality change in old age. Wilson, S. A. K. (1912). Progressive lenticular degeneration: a familial nervous
Current Opinions in Psychiatry, 20, 62–6. disease associated with cirrhosis of the liver. Brain, 34, 295–509.
Lauter, H. and Meyer, J. E. (1968). Clinical and nosological concepts of senile World Health Organization (1978). Mental disorders: glossary and guide
dementia. In: Müller, C. and Ciompi, L. (eds) Senile dementia. Huber, to their classification in accordance with the Ninth Revision of the
Bern. International Classification of Diseases. World Health Organization,
McKhann, G. M., et al. (2011). The diagnosis of dementia due to Geneva.
Alzheimer’sdisease: recommendations from the National Institute on World Health Organization (2007). International statistical classification
Aging and the Alzheimer’s Association Workgroup. Alzheimer’s and of diseases and related health problems, 10th Revision. World Health
Dementia, 7(3), 263–9. Organization, Geneva.
Mendel, E. (1902). Leitfaden der Psychiatrie. Enke, Stuttgart.
Neary, D., et al. (1988) Dementia of frontal lobe type. Journal of Neurology
and Neurosurgery, 51, 353–61. Part 3
O’Connor, D. W., et al. (1996). Cross-national interrater reliability of dementia
diagnosis in the elderly and factors associated with disagreement. The DSM-5 Approach to Dementia
Neurology, 47, 1194–9. This section is written by Perminder S. Sachdev.
Petersen, R. C., et al . (1999). Mild Cognitive Impairment: clinical The Diagnostic and Statistical Manual of Mental Disorders, pop-
characterization and outcome. Archives of Neurology, 56, 303–8.
ularly referred to as DSM, is the classification of mental disorders as
Pick, A. (1901). Senile Hirnatrophie als Grundlage von Herderscheinungen.
proposed by the American Psychiatric Association (APA). In addi-
Wien klin Wschr , 14.
Pick, A. (1904). Zur Symptomatologie der linksseitigen Schläfenlappenatropzie.
tion to providing diagnostic criteria for various psychiatric disor-
Mschr Psychiat Neurol , 16, 376–88. ders, it has brief descriptions of each disorder, which include the
Pick, A. (1906). Über einen weiteren Symptomenkomplex im Rahmen der salient diagnostic features, epidemiological characteristics, associ-
Dementia senilis, bedingt durch umschriebene stärkere Hirnatrophie ated features, and differential diagnosis. The 4th edition of DSM
(gemischte Apraxie). Monatsschr Psychiatr , 19, 97–108. (DSM-IV) was published in 1994, and its descriptions became
internationally influential in defining dementia and its subtypes. important type of dementia, does not receive a mention in DSM-IV.
The APA is currently in the process of revising the criteria toward Pick’s disease is now subsumed under the broader category of FTD.
the publication of DSM-5 in 2013. The Neurocognitive Disorders Many other examples of this can be pointed out, and all of these
Work Group of the DSM-5 Task Force began work in April 2008 have provided an impetus to re-examine the terminology and the
and a working draft was posted on the APA’s website <www.dsm5. criteria.
org> in February 2010. The main recommendations from this
proposal and the rationale for the proposed changes will now be The DSM-5 proposal
discussed. The DSM-5 proposal is an attempt to deal with some of the deficien-
cies described in the section “A critique of the DSM-IV approach
A critique of the DSM-IV approach to dementia
to dementia”, and present a uniform framework for defining all
Dementia is defined in DSM-IV as a syndrome characterized by dementing disorders (Ganguli et al., 2011). The Neurocognitive
deficits in multiple cognitive domains, including memory, which Disorders Work Group recognized that several expert groups have
represent a decline from a previously higher level of function- recently made attempts to describe criteria for various dement-
ing, and which cause significant impairment in social or occupa- ing disorders, including AD (Albert et al., 2011; McKhann et al.,
tional functioning. The criteria for various subtypes of dementia, 2011), DLB (McKeith et al., 2006), and FTD (Cairns et al., 2007;
such as dementia of the Alzheimer’s type and vascular dementia, Gorno-Tempini et al., 2011; Rascovsky et al., 2011). In the Work
are described. There are a number of limitations to the DSM-IV Group’s proposal, an attempt was made to ensure that the DSM-5
approach. criteria were not inconsistent with these criteria sets. This involved
While the term dementia has a historical legacy and wide rec- the inclusion of several experts as advisors in the development of
ognition, it has increasingly become equated with Alzheimer’s the criteria (Ganguli et al., 2011). However, this had to be done
disease (AD) in the minds of lay people and policymakers alike. within the broader framework of DSM-5 which includes all mental
In the case of cognitive impairment due to another cause, such as disorders. While there was a concerted effort to take an interna-
head injury, human immunodeficiency virus (HIV) infection, and tional perspective, the primary constituency for the use of these
multiple sclerosis, among others, it is not customary to use the term criteria is the general psychiatrist practising in the US, and this fact
dementia as it is considered pejorative and stigmatizing (Sachdev, had some influence on the final presentation of the criteria.
2000). Because of the severity and disability associated with the
term dementia, its wide usage prejudices against the early recogni- Mild and Major neurocognitive disorders
tion of cognitive impairment. While it is generally accepted that Dementia in DSM-5 is classified within the broad category of neu-
such impairment lies on a continuum, the category of dementia rocognitive disorders (NCD). The choice of ‘neurocognitive’ in pref-
uses a cut-off based on severity, whereby the disorder is identified at erence to ‘cognitive’ was based on subtle differences in the usage of
a late stage even though potentially disease-modifying treatments the two terms. The latter is used in psychology for all mental repre-
may be more effective before this stage is reached. In the case of sentations of information processing, and indeed for all conscious
vascular dementia (VaD), which is the second most common type activity. Cognitive disturbances are therefore common in men-
of dementia, there have been repeated calls to abandon the term tal disorders such as depression, schizophrenia, bipolar disorder,
dementia in favour of an overarching description of vascular cogni- autism, and obsessive compulsive disorder. The term ‘neurocogni-
tive impairment (Hachinski and Bowler, 1993; O’Brien et al., 2003) tive’ describes cognitive functions closely linked to the function of
to overcome this limitation. particular brain regions, neural pathways, or cortical/subcortical
The inclusion of memory impairment as a necessary criterion networks in the brain, and therefore more accurately describes the
in the DSM-IV definition of dementia has been criticized as an disorders of interest.
‘alzheimerization’ of the term. Memory impairment is not neces- Since cognitive impairment is on a continuum, and the diagnostic
sarily a feature of the early stages of VaD (Looi and Sachdev, 1999), process entails the imposition of a categorical system on this con-
frontotemporal dementia (FTD) (Rabinovici and Miller, 2010), and tinuum, it is recognized that many individuals lie in-between nor-
many other types of dementia. Furthermore, DSM-IV does not mal cognitive functioning and dementia. Various terms have been
provide a systematic description of the cognitive domains that are used to describe this condition, with Mild Cognitive Impairment
likely to be impaired in dementia. Other than memory, it includes (MCI) being the most commonly used in the last decade (Petersen
aphasia, apraxia, agnosia, and executive functioning as the cogni- et al., 2009). MCI therefore represents a condition that is below the
tive domains of interest, but omits complex attention, information threshold for dementia. When followed up over time, some indi-
processing speed, visuoconstructive function, and other cognitive viduals with MCI will progress to dementia, and this is particularly
functions that are frequently impaired in dementia. It does not pro- true if MCI is a manifestation of early AD. However, if MCI is due
vide any guidelines for defining impairment, which is left to the to a nonprogressive aetiology, such as brain injury or encephali-
judgement of the clinician. The consequence of this is a relatively tis, its progression to dementia is not expected. To acknowledge
poor correspondence of DSM-IV dementia with dementia by other continuity in cognitive dysfunction, and yet impose the categori-
criteria (Erkinjuntti et al., 1997). cal reality of a clinical diagnosis, DSM-5 proposes to subcategorize
There are a significant number of other limitations to the DSM-IV NCD into Mild and Major categories, with the latter being roughly
criteria. The last two decades have seen numerous advances in equivalent to dementia, with minor differences. It is true that brain
understanding the aetiopathogenesis of the dementing disorders damage can exist without any evident cognitive impairment. Such
and thereby their conceptualization. Their role as biomarkers in the individuals may be at an increased risk of future decline and are
diagnosis has also received much attention in AD and other dis- worthy of medical attention to prevent such decline. This has been
orders. Dementia with Lewy bodies (DLB), now recognized as an referred to as the preMCI or the ‘brain-at-risk’ stage of cognitive
dysfunction. However, it would be inappropriate to regard this as a normal, Mild, and Major NCD, albeit with the caveat that these
clinical disorder, and the current criteria restrict themselves to the are only guidelines, and clinical judgement is usually necessary
Mild and Major NCD categories. in interpreting the results of whatever tests are applied. For Mild
The asymmetrical use of ‘Mild’ and ‘Major’, instead of the ‘Mild NCD, deficits would typically fall between one and two standard
and Severe’ or ‘Minor and Major’ wording, was the accepted com- deviations below the mean (or between the 3rd to 16th percentiles
promise in line with the DSM-5 proposal (Ganguli et al., 2011). for test scores not normally distributed) of people of similar age,
‘Severe’ is strongly associated with the degree of disability associ- sex, education, and sociocultural background. For Major NCD,
ated with late-stage dementia, and ‘minor’ runs the risk of trivi- deficits would typically fall two or more standard deviations below
alizing what is an important disorder worthy of attention and the mean (or below the 3rd percentile). If appropriate normative
intervention. The term ‘dementia’, despite its many limitations, data for the measure(s) are not readily available (e.g. for those with
was considered historically too important to be abandoned. It has very high or very low education or premorbid intellectual capacity
been subsumed under the concept of Major NCD, except that the or different sociocultural-linguistic background), the clinician may
DSM-IV and ICD-10 definitions of dementia warranted impair- infer from the combination of history and cognitive performance
ment in at least two cognitive domains, of which one was necessar- that there has been a real but modest level of decline in the former,
ily memory. The distinction between Mild and Major (or dementia) and a clear and significant level of decline in the latter. Alternatively,
is based on the severity of cognitive deficits, and more importantly when serial measurements are available, decline from an individu-
on the impairment secondary to them. The traditional approach al’s own previous level would serve as more definitive evidence of
has been to diagnose dementia (or Major) when the cognitive defi- decline.
cits are severe enough to impair social or occupational functioning, The distinction between Mild and Major NCD is primarily based
such that instrumental activities of daily living (IADL) are affected. on the individual’s level of independence. Individuals with Major
This threshold has been retained in the criteria for Major, except NCD will have impairment sufficient to interfere with independ-
that the descriptive emphasis has shifted to ‘interference with inde- ence, such that others will have to take over tasks that they were
pendence’ as distinguishing Major from Mild. It is also noted that previously able to complete on their own. Individuals with Mild
impairment should be attributable to cognitive and not motor or NCD will have preserved independence, although there may be
speech impairment. subtle interference with function, or a report during the cogni-
It is true that Mild disorder can also affect functioning, especially tive history that tasks require more effort or take more time than
in individuals with intellectually demanding occupations, but, in previously. The proposed diagnostic criteria are summarized in
general, such individuals can function close to their previous lev- Table 29.1, with the caveat that these are subject to further revisions
els by instituting compensatory strategies, and their IADLs remain before their publication in mid-2013.
essentially intact. This criterion does raise the dilemma, however,
that the degree of impairment produced by a disorder is being used Identifying the cognitive domains
to diagnose the disorder. This is against the recommendations of A large number of cognitive tests are used to evaluate neuropsy-
the World Health Organization (2001) that the classification of chological function, and there is no consensus on how the cogni-
functioning and disability be kept separate from the classification tive domains should be described. DSM-5 is not designed to be
of diseases. In view of this, thresholds based on cognitive test per- prescriptive about the tests to be used, as these will differ with
formance norms are also recommended as guidelines, to be used in the situation and an individual examiner’s preference. However, it
the clinical context based on the clinician’s judgement. The primary proposes six broad domains of cognition, one of which should be
distinction between Mild and Major, however, relies on the impair- impaired for an NCD to be diagnosed. These domains are: com-
ment in functioning criterion. plex attention, learning and memory, executive ability, language,
visuoconstructional-perceptual motor ability, and social cognition.
Diagnostic thresholds The document provides working definitions of the neurocognitive
The diagnosis of an NCD requires the presence of concern about domains and the corresponding impairments in everyday functions
cognitive function and a determination by the examiner that per- that the clinician may elicit or observe (see Table 29.2), recogniz-
formance falls below the expected level or has been observed to ing that consensus across disciplines may not be achieved on such
decline over time. The concern may be expressed by the patient, or a proposal. A point of difference from the convention for dementia
a knowledgeable informant, or the clinician. A cognitive concern must be noted. A Major NCD can be diagnosed if only one cogni-
differs from a complaint in that it may or may not be voiced spon- tive domain is affected and when the impairment is sufficient to
taneously and may need to be elicited by careful questioning; the meet the other criteria. The concept of Major NCD is therefore
specific symptoms may relate to any cognitive domain and are not broader than dementia and includes the DSM-IV ‘amnestic disor-
restricted to memory. der’, and some patients with severe aphasia would also receive this
The DSM-5 proposal requires that the presence of cognitive diagnosis.
impairment should be objectively verified, preferably by a formal
neuropsychological assessment. However, such testing is not always Subtypes of neurocognitive disorders
available, even in resource-rich countries, and neuropsychological Since NCD often have identifiable aetiology, the diagnostic process
thresholds are sensitive to the specific test used, the conditions of must proceed from recognition of the syndrome of cognitive impair-
testing, and the norms employed. The clinician may therefore form ment to identification of one or more specific aetiologies. DSM-5
an opinion on objective impairment based on a bedside assessment proposes specific criteria for the following aetiologies: Alzheimer’s
and interpret it in light of an individual’s prior abilities. Thresholds disease, cerebrovascular disease, frontotemporal lobar degen-
of impairment are recommended for the distinction between eration, Lewy body disease, Huntington’s disease, traumatic brain
Table 29.1 Proposed diagnostic criteria for Major and Mild neurocognitive disorders in DSM-5 (<www.dsm5.org updated April 2012>)*
Major neurocognitive disorder
A. Evidence of significant cognitive decline from a previous level of performance in one or more of the domains outlined in Table 29.2 based on:
1. Concerns of the individual, a knowledgeable informant, or the clinician that there has been a significant decline in cognitive function; and
2. A decline in neurocognitive performance, typically involving test performance in the range of two or more standard deviations below appropriate norms
(i.e. below the 3rd percentile) on formal testing, or equivalent clinical evaluation.
B. The cognitive deficits are sufficient to interfere with independence (i.e. requiring assistance at a minimum with instrumental activities of daily living (e.g. more
complex tasks such as paying bills or managing medications)).
C. The cognitive deficits do not occur exclusively in the context of a delirium.
D. The cognitive deficits are not primarily attributable to another mental disorder (e.g. Major depressive disorder, schizophrenia).
Mild neurocognitive disorder
A. Evidence of minor cognitive decline from a previous level of performance in one or more of the domains outlined in Table 29.2 based on:
1. Concerns of the individual, a knowledgeable informant, or the clinician that there has been a mild decline in cognitive function; and
2. A decline in neurocognitive performance, typically involving test performance in the range of one and two standard deviations below appropriate norms
(i.e. between the 3rd and 16th percentile) on formal testing, or equivalent clinical evaluation.
B. The cognitive deficits are insufficient to interfere with independence (i.e. instrumental activities of daily living (i.e. more complex tasks such as paying bills or
managing medications) are preserved), but greater effort, compensatory strategies, or accommodation may be required to maintain independence.
C. The cognitive deficits do not occur exclusively in the context of a delirium.
D. The cognitive deficits are not primarily attributable to another mental disorder (e.g. Major depressive disorder, schizophrenia).
*The final criteria will be published in mid-2013.
Table 29.2 Proposed cognitive domains for DSM-5 diagnosis of be more stringent. For instance, Mild NCD should not be attrib-
neurocognitive disorders uted to AD without strong genetic or biomarker evidence.
Domain Cognitive functions affected Behavioural manifestations of neurocognitive disorders
NCD are frequently associated with psychiatric and behav-
Complex attention Sustained attention, divided attention,
selective attention, processing speed
ioural features such as depression, agitation, apathy, delusions,
and hallucinations, and these are often the primary reason for
Executive ability Planning, decision-making, working memory, clinical presentation. There has been some debate as to whether
responding to feedback/error correction,
these should be listed as associated features or be given the sta-
overriding habits, mental flexibility
tus of separate syndromes, such as the psychosis or depression
Learning and memory Immediate memory, recent memory of Alzheimer’s disease (Jeste et al., 2000; Olin et al., 2002). The
(including free recall, cued recall), and proposed DSM-5 approach continues to regard them as special
recognition memory manifestations of NCD that are too nonspecific to be regarded as
Language Expressive language (including naming, distinct syndromes.
fluency, grammar, and syntax) and receptive
language Conclusion
Visuoconstructional and Construction, visual perception, integration of The DSM-5 proposal for diagnostic criteria for neurocognitive dis-
perceptual motor ability perception and motor output; includes praxis orders is a major effort to bring coherence to the field and provide
and gnosis. consistent criteria across the various subtypes of neurocognitive
Social cognition Recognition of emotions, theory of mind, disorders. It recognizes the status of Mild NCD as a valid clinical
behavioural regulation and research entity and proposes that the category of Major NCD
subsumes the widely used term dementia while highlighting the
limitations of the latter. It further proposes diagnostic criteria for
specific NCDs due to the Major aetiologies. It is hoped that the pro-
injury, HIV disease, prion disease, and substance-use-associated
posals will bring uniformity to the criteria used for NCDs interna-
disease. The intention is to make the DSM-5 criteria compatible
tionally. For this to occur, it is important for these proposals to be
with criteria proposed by expert groups within each subtype. This
systematically examined for reliability, validity, and usefulness in
includes the use of relevant biomarkers. There is provision for
clinical practice.
listing more than a single aetiology, as this is not uncommon in
clinical practice and population studies. Aetiologic diagnoses are
possible for both Major and Mild NCD, with examples of how the References
disorder might look different at each level. It is recognized that in Albert, M.S., et al. (2011). The diagnosis of Mild Cognitive Impairment
some cases, the criteria for ascribing aetiology in Mild NCD must due to Alzheimer’s disease: Recommendations from the National
Institute on Aging–Alzheimer’s Association Workgroups on ‘Dementia’ or ‘Major neurocognitive disorder’

diagnostic guidelines for Alzheimer’s disease . Alzheimer’s and
Dementia, 7, 270–9. What then is dementia? To begin with, the term itself may fall into
Cairns, N.J., et al. (2007). Consortium for Frontotemporal Lobar Degeneration. disuse, since the proposed revisions for DSM-V suggest ‘Major
Neuropathologic diagnostic and nosologic criteria for frontotemporal neurocognitive disorder’ as a replacement. The reasons cited for
lobar degeneration: consensus of the Consortium for Frontotemporal this change are that ‘dementia’ has ‘acquired a pejorative or stigma-
Lobar Degeneration. Acta Neuropathologica, 114, 5–22. tizing connotation’ for older adults and that it is less well accepted
Erkinjuntti, T., et al. (1997). The effect of differential diagnostic criteria on among younger adults (e.g. in the context of HIV or head injury).
the prevalence of dementia. New England Journal of Medicine, 337, Changes in nomenclature have of course occurred at regular inter-
1667–74.
vals in the history of mental disorder classifications and have rarely
Ganguli, M., et al. (2011). Classification of neurocognitive disorders in
DSM-5: a work in progress. American Journal of Geriatric Psychiatry, 19,
been universally welcomed. At a first glance, this particular change
205–10. does not look hopeful for the purpose of clinical communication,
Gorno-Tempini, M.L., et al. (2011). Classification of primary progressive being a somewhat convoluted and technical term, as well as failing
aphasia and its variants. Neurology, 76, 1006–14. to possess any obvious convenient abbreviation (‘MND’ overlap-
Hachinski, V.C. and Bowler, J.V. (1993). Vascular dementia. Neurology, 43, ping with both motor neuron disease and the ‘Mild neurocogni-
2159–60. tive disorder’ term which is suggested as a replacement for Mild
Jeste, D. and Finkel, S. (2000). Psychosis of Alzheimer’s disease and related Cognitive Impairment and related constructs). Clinicians, pre-
dementias. Diagnostic criteria for a distinct syndrome. American Journal sumably obliged to use the term in order to justify their practice
of Geriatric Psychiatry, 8, 29–34.
as evidence-based (i.e. linked to research findings), may therefore
Looi, J.C.L. and Sachdev, P. (1999). Differentiation of vascular dementia from
have a lot of ‘do you mean dementia?’ questions before the new
AD on neuropsychological tests. Neurology, 53, 670–78.
McKeith, I.G. (2006). Consensus guidelines for the clinical and pathological
terminology is accepted. More fundamentally, it is unclear to what
diagnosis of dementia with Lewy bodies (DLB): report of the Consortium extent a name change can really influence stigma beyond a tem-
on DLB International Workshop. Journal of Alzheimer’s Disease, 9(S3), porary reprieve. While underlying disorders remain untreatable
417–23. and a cause of significant impairment and loss of dignity, stigma
McKhann, G.M., et al. (2011). The diagnosis of dementia due to Alzheimer’s will simply return gradually as a new label moves from research
disease: Recommendations from the National Institute on Aging– and clinical settings into popular use. A better time for a change
Alzheimer’s Association workgroups on diagnostic guidelines for in terminology would be at a point where significant differences
Alzheimer’s disease. Alzheimer’s and Dementia, 7, 263–9.
are being made to course and outcome; otherwise, such changes
O’Brien, J.T., et al. (2003). Vascular cognitive impairment. Lancet Neurology,
amount to no more than wishful thinking. On the other hand,
2, 89–98.
Olin, J., et al. (2002). Provisional diagnostic criteria for depression of
‘Major neurocognitive disorder’ does shift the terminology more
Alzheimer disease. American Journal of Geriatric Psychiatry, 10, explicitly towards a syndrome description that may have some
125–8. benefits. Ultimately, predictions of how language will change and
Petersen, R., et al. (2009). Mild Cognitive Impairment: ten years later. Archives evolve are invariably inaccurate and probably not worth further
of Neurology, 66, 1546–54. consideration. ‘Dementia’ will be used as the principal terminol-
Rabinovici, G.D. and Miller, B.L. (2010). Frontotemporal lobar degeneration: ogy for the remainder of this chapter.
epidemiology, pathophysiology, diagnosis and management. GNS Drugs,
24, 375–98.
Rascovsky, K., et al. (2011). Sensitivity of revised diagnostic criteria for Defining dementia
the behavioural variant of frontotemporal dementia. Brain , 134, Dementia itself is a relatively straightforward construct, which
2456–77. probably explains its continued and long-standing application
Sachdev, P. (2000). Is it time to retire the term ‘dementia’? Journal of across specialties. Broadly speaking, to have dementia a person
Neuropsychiatry and Clinical Neurosciences, 12, 276–9. must have cognitive impairment, the impairment must reflect a
World Health Organization (WHO) (2001). International classification
decline from previous abilities, the impairment must be affect-
of functioning, disability and health (ICF).<http://www.who.int/
classification/icf/en/> (accessed 19.12.2012 ). ing the person adversely to some extent, and the syndrome as a
whole must be differentiated from similar presentations with dif-
ferent underlying causes such as depression and delirium. These
are essentially the elements of the clinical diagnosis and clinicians
are relied upon to exercise their skill and judgement in interpret-
Part 4 ing the components. However, application of the diagnosis in
research settings requires more standardization. Principally, this
Should We Diagnose Subtypes of Dementia? concerns the definition of ‘cognitive impairment’, specifying the
This section is written by Robert Stewart. domains of function that may be affected and the number of these
Dementia is a ‘syndrome’ rather than a ‘diagnosis’—that is, a impairments required for case definition. Apart from the nomen-
collection of symptoms and associated features that represent clature, one of the key changes in the proposed DSM-V criteria
the common outcome of a variety of underlying disorders. The is a move away from memory impairment as a core and required
argument presented here is that we are safest referring to what feature of the syndrome as a whole, allowing a broader range of
we can actually see and describe, i.e. the syndrome, rather than domains to contribute. This reflects strongly held views that the
implying we know for certain what has caused it. The situation of requirement for memory impairment had been driven by a focus
course may change with advancing knowledge, but it is premature on Alzheimer’s disease and was problematic for dementia associ-
to anticipate this. ated with nonAlzheimer disorders.
Defining the underlying symptom—what is cognitive Dementia subclassification

impairment? For most of its history as a clinical diagnosis, dementia has been
Inevitably there are remaining areas of ambiguity in the research subclassified according to its principal underlying disorders.
diagnosis, with a tension between strict standardization and allow- Subclassification in other mental disorders (e.g. schizophrenia,
ing an element of judgement by a skilled assessor (the former depression) has tended to show more marked historical fluctuations
required in large epidemiological studies using lay interviewers, the and it is reasonable to assume that the more long-standing approach
latter often more appropriate for clinical settings or where reflec- in dementia reflects its ‘organic’ basis. However, the nomenclature
tion of normal practice is desired). So for cognitive impairment as implies knowledge that has not yet been gained, and that may turn
a criterion, research criteria do not tend to specify the severity at out to be misleading. As described in the section The problem of
which ‘impairment’ should be defined within a given domain of mixed disease, there are a number of instances where subclassifica-
function, or how a person’s age and educational background are tion of dementia remains valid; however, it will be argued that this
incorporated in this decision. can rarely be justified in late-onset disorders.
Defining the effects of the syndrome—what is functional The problem of mixed disease
impairment?
Defining dementia according to its underlying cause is certainly
A more important challenge for research diagnostic criteria is appropriate where that cause can be accurately ascertained, e.g. by
the level of interference with functional independence required a genetic test in familial disorders or where other diagnostic tests
to differentiate dementia from Mild Cognitive Impairment (or are available, such as in prion diseases. It may also be appropri-
Major from Mild neurocognitive disorder in the proposed DSM-V ate where the underlying pathology can be reasonably inferred,
nomenclature). This necessarily involves considerable subjective e.g. in the context of a strategic infarct dementia or in late-stage
judgement on the part of the assessor and introduces potentially Parkinson’s disease. The problem with subclassification is that it
wide scope for differences in practice. In clinical settings, this may emerged (and was incorporated into standard diagnostic prac-
not matter too much since the principal purpose of communi- tice) at times when a large proportion of dementia cases were
cating ‘dementia’ as a diagnosis is a prediction that any current what would now be considered as relatively early onset (e.g. below
impairment is likely to deteriorate in future rather than improve age 80). Disorders occurring in these age ranges are indeed likely
(in the absence of disease-modifying treatments) and, to do this, to have single causes. However, because of the remarkable changes
the clinician is often relying considerably (and reasonably) on past in life-expectancy experienced in high-income nations over the last
decline to predict prognosis. However, in a research context it half century, most people with dementia in these settings are much
can lead to substantial variation in dementia and Mild Cognitive older, and underlying causes are much more mixed and subtle, pos-
Impairment prevalence between centres, as well as misleading sibly including pathological processes as yet unknown. Dementia
and somewhat circular arguments around rates of Mild Cognitive in these age groups is therefore poorly represented by mutually
Impairment ‘conversion’. The degree to which a given level of cog- exclusive subcategories.
nitive impairment results in reduced functional capacity clearly
will depend substantially on a person’s individual circumstances The limitations of neuropathology and biological research
and expectations, in addition to those of his or her immediate sup- Neuropathological findings have been key drivers of dementia
port network and wider culture. For example, an individual liv- research, providing important clues concerning causal processes
ing in circumstances or a culture of high social support and low and underlying disorders. Understandably, therefore, dementia
expectations of function in older people may ‘convert’ to dementia subclassification has been driven by neuropathological observa-
at a much more advanced level of impairment, compared to some- tions. However, it is important to bear in mind that neuropatho-
one who is less supported or whose lifestyle is more cognitively logical and other biological research cannot actually demonstrate
demanding. Discussions around the Mild Cognitive Impairment– causation conclusively. Observing brain changes after someone has
dementia interface rarely acknowledge this fundamental flaw in died is limited by the frequently long duration of clinical deteriora-
the constructs, perhaps because clinicians feel confident about tion that has occurred between disorder onset and death (i.e. it is
distinguishing them in their own practice but are less aware of predominantly providing a picture of advanced disease rather than
how much their approach differs from others. Certainly, attempts processes around disease onset). Furthermore, having observed that
to apply Mild Cognitive Impairment criteria in epidemiological people with dementia have more of a particular change in the brain
research have resulted in substantially lower dementia conversion compared to those not affected, it is still not possible to infer that
rates and very variable prevalences. the particular brain change was the cause of the dementia rather
than some biproduct of another process. In vitro research can be
Rethinking Mild Cognitive Impairment
used to refine this hypothesis through demonstrating effects in cell
Difficulties with the Mild Cognitive Impairment construct, how-
cultures, and this can be extended into animal models. However,
ever, should be balanced by the likely benefits of early dementia
to date there is no animal model of Alzheimer’s disease, i.e. one
detection and the clear need to develop a way to achieve and evalu-
displaying the full neuropathological features of the disease as it
ate this process. One solution would be to adopt study designs where
occurs in humans.
broader groups are defined on the basis of cognitive impairment
and where outcomes are cognitive or functional decline rather than The importance of experimental evidence
incident dementia, i.e. refocusing on what can be directly meas- One way of demonstrating causation conclusively would be to gener-
ured rather than what relies on subjective judgement. The growing ate the disease artificially—this is clearly not ethical in humans and,
appreciation of subtle functional impairments in Mild Cognitive regardless of ethics, has not been found to be feasible in animals.
Impairment will probably assist this transition. The only alternative method is to modify a hypothesized disease
process and influence the clinical course. Acetylcholinesterase that pathology is often unknown, associated with a gradually pro-
inhibitors, while clearly having demonstrable effects on clinical gressive dementia not explained by repeated vascular events). The
outcome, have not lived up to some of the more optimistic predic- problem in clinical assessment is that cerebrovascular disease can
tions voiced when they were originally introduced. However, the be measured in vivo and at increasingly subtle levels with routine
observed clinical effects have provided helpful clarification of the use of magnetic resonance imaging. ‘Vascular dementia’ as a diag-
likely role of cholinergic transmission in modifying cognitive func- nosis implies dementia caused by vascular processes. However,
tion in dementia, an effect that could not have been inferred con- the co-occurrence of cerebrovascular disease and dementia, par-
clusively from observational research. The amyloid hypothesis has ticularly in older age groups where some degree of cerebrovascular
predominated in Alzheimer’s disease for many years on the basis of disease is almost inevitable (Fernando et al., 2004), cannot be taken
observation and majority opinion. An important milestone will be to imply that one is the sole cause of the other. In theory, the text-
the definitive testing of that hypothesis through randomized con- book criteria for diagnosing vascular dementia (or vascular cog-
trolled trials, and the answer is likely to emerge in the near future nitive impairment) are intended to facilitate this decision-making.
now that cerebral amyloid can be visualized in vivo, and now that However, the criteria with best performance and reliability tend to
agents are being developed with demonstrable modifying effects. be those that simply detect evidence of cerebrovascular disease (e.g.
If pathogenic forms of amyloid can be removed or modified, but the original Hachinski Ischaemic Scale), while symptoms that are
if drugs with these actions do not substantially affect the clinical used to infer causation (step-wise course, profile of impairment,
course of dementia, the implications for Alzheimer’s research will etc.) perform much more poorly or with poor agreement between
be far wider than those for the pharmaceutical industry. assessors, as described in Chapter 34. Furthermore, many relatives
Limitations of Alzheimer’s disease as a diagnosis of people with apparent Alzheimer’s disease will describe variations
in its rate of progression with episodes of sudden deteriorations.
Even in the absence of experimental evidence, there are substantial
The same problems of rarity of the pathology as a cause of dementia
limitations in our current understanding of Alzheimer’s disease as
in isolation apply to Lewy body dementia (Matthews et al., 2009).
a cause of dementia, which have been demonstrated in neuropatho-
logical research—particularly from the relatively few studies that Why is diagnosis important?
have drawn their data from community samples rather than spe-
The primary purpose of a diagnosis is to facilitate communication
cialist services. These have allowed the likelihood of dementia (or
between a clinician and a patient. In this respect, the term used
cognitive impairment) prior to death to be investigated in rela-
is providing patients with an explanation for their symptoms and,
tion to post-mortem findings, providing at least some clarification
depending on the disorder, some indication about what treatment
of how much the degree of a given pathology is related to clinical
is indicated and what outlook or prognosis can be expected. As
manifestations and how common these pathological changes are
already argued, the subclassification of late-onset (age 80+) demen-
in unaffected members of the general population. An important
tia is in effect covering up guesswork, i.e. implying an underlying
sea change in understanding has arisen from observations that sig-
cause when we know that we are not even very good at estimating
nificant Alzheimer pathology is common in older people with no
the neuropathological features, let alone what processes are actually
evidence of even Mild Cognitive Impairment in life (Bennett et al.,
causing neurodegeneration. Whether this constitutes unreason-
2006; Erten-Lyons et al., 2009). Poor correlations have been found
able dishonesty is for others to judge. The process is most clearly
between clinical diagnoses and neuropathology, with much higher
unhelpful, however, when individuals have seen several clinicians
levels of mixed pathology after death than implied by the diagno-
and have received what is perceived to be conflicting information,
sis in life (White et al., 2005). Neuropathological follow-up from
e.g. told they have Alzheimer’s disease on one consultation and vas-
the large Cognitive Function and Aging study cohort in the UK
cular dementia on another. A more serious problem arises when
importantly revealed substantial overlap in burden of Alzheimer
diagnoses determine treatment eligibility, the most important situ-
pathology between people with and without dementia in life, par-
ation being the interface between Alzheimer’s disease and vascu-
ticularly in older age groups; cortical atrophy, on the other hand,
lar dementia. If the presence of cerebrovascular disease is taken to
was strongly associated with dementia in all age groups (Savva et al.,
indicate ‘vascular dementia’ as a diagnosis (despite ample evidence
2009). In further analyses of attributable risks of dementia at death,
that most late-onset dementia associated with cerebrovascular dis-
only modest contributions were found for individual pathological
ease has mixed aetiology) and if a given treatment is seen as only
features, e.g. 8% of dementia at death accounted for by neocortical
indicated in Alzheimer’s disease, then people with mixed disease
neuritic plaques, 11% by neurofibrillary tangles, 12% by small ves-
will be less likely to receive that treatment for no good reason apart
sel disease, and 3% by Lewy bodies (Matthews et al., 2009). Once
from a diagnostic convention that poorly reflects reality. Since the
again, these findings emphasize that we are on safer ground talking
likelihood of comorbid cerebrovascular disease will be determined
about what we can be sure of (i.e. that an individual’s dementia is
to some extent by age, gender, ethnicity, and social class, there are
caused by loss of neurons and cortical atrophy), rather than what
reasonable grounds for claiming that the application of diagnostic
we are still guessing (i.e. what is causing the loss of neurons in that
subcategories fosters discrimination and inequality.
individual and what their brain will look like after they die).
Limitations with other dementia categories Potential solutions for clinical practice
If causation in Alzheimer’s disease has yet to be conclusively dem- The most obvious clinical solution is to apply commonsense to
onstrated, progress in vascular dementia is even less advanced. As the diagnosis of late-onset dementia, simply recognizing the fact
is described in more detail in Chapter 34, dementia is commonly that these people are likely to have more than one thing wrong
associated with cerebrovascular disease but not often in isolation, with them. In these situations, it is surely better to communicate
i.e. it is frequently mixed with Alzheimer’s disease (or, at least, given ‘dementia’ as a diagnosis rather than to imply a single aetiology
through subcategorizing it, and surely simple enough and more (accepting the problem of heterogeneity in causal mechanisms
honest to inform patients and/or informants that there may be and clinical presentations), or within a number of subcategories
more than one process that has given rise to their symptoms. The (accepting a degree of overlap between these). Essentially, neither
diagnostic formulation is common enough in other areas of mental position reflects the truth and a considerable amount of effort has
health and we feel no pressure to assign depression, for example, to probably been wasted in debate between the two camps. Physicists
a single cause. It may also be helpful to move away from language have been happy for a long time to accept that light exists as both a
that implies causation. In this respect, the proposed DSM-V ter- wave and a particle, so there is no particular reason why dementia
minology has advantages—describing, for example, ‘Major cogni- cannot be viewed, at least within the research arena, as both a single
tive disorder associated with vascular disease’ rather than ‘vascular condition and multiple conditions.
dementia’, or ‘Major cognitive disorder associated with Lewy body
disease’ rather than ‘Lewy body dementia’. However, it is difficult
References
to see this lengthy wording becoming commonly used in clinical
practice. Furthermore, the diagnostic system as currently proposed Bennett, D.A., et al. (2006). Neuropathology of older persons without
cognitive impairment from two community-based studies. Neurology,
in DSM-V continues to foster the myth of single causes (with no
66, 1837–44.
mixed category at the time of writing). Erten-Lyons, D., et al. (2009). Factors associated with resistance to dementia
despite high Alzheimer disease pathology. Neurology, 72, 354–60.
Potential solutions for research—lumping or splitting Fernando, M.S., Ince, P.G., and MRC Cognitive Function and Ageing
The difficulty of diagnosis is less easily resolved in research where Neuropathology Study Group (2004). Vascular pathologies and
subdivision of dementia into underlying syndromes continues to cognition in a population-based cohort of elderly people. Journal of the
be necessary in order to investigate causation. In a wider mental Neurological Sciences, 226, 13–17.
health context, because psychiatric symptoms co-occur in a variety Matthews, F.E., et al. (2009). Epidemiological pathology of dementia:
attributable-risks at death in the Medical Research Council Cognitive
of combinations, with few that can be truly said to be pathogno-
Function and Ageing Study. PLoS Medicine, 6, e1000180.
monic of a particular disorder, there is fertile ground for disagree-
Savva, G.M., et al. (2009). Age, neuropathology, and dementia. New England
ment about how to group symptoms into diagnoses. Dementia Journal of Medicine, 360, 2302–9.
researchers, like those in many other fields of research, are often White, L., et al. (2005). Recent clinical-pathologic research on the causes
themselves categorized into ‘lumpers’ or ‘splitters’ based on whether of dementia in late life: update from the Honolulu-Asia Aging Study.
they prefer, respectively, to consider cases within a single category Journal of Geriatric Psychiatry and Neurology, 18, 224–7.
CHAPTER 30
Hello, I’m me! Living
well with dementia
June and Brian Hennell
Brian Hennell was born in London in June 1938 to Marie, a tal- chose to do six other things which put pressure on us both and made
ented tailoress and Frederick Hennell, a specialist toolmaker. him cross because he didn’t do the shopping. I’m fed up with being a
Military requirements for Frederick’s job relocated the family to verbal punch bag. Brian is the only person I know who can make the
word ‘dear’ sound like a Sergeant Major’s criticism. What is happen-
the Somerset Levels in 1942 where Brian grew up as an only child.
ing to us? Instead of the kind, fun loving and considerate man I knew,
This rural upbringing was to prove definitive in Brian’s life, for he Brian seems to have a split personality . . . When he is nice he is very
became more at home with ‘wild’ animals and nature than other nice but when he is not he is horrid. Can this be the normal ageing
children. process for a man? If so, heaven help us.
Poor health in his early years gave rise to only average educa-
Sensing that her presence was needed at home now if they were
tional results, but a great love of the countryside developed, as did
to have any quality retirement years together, June retired at the age
social skills in meeting with others, especially in ballroom dancing
of 61 in 2006. Her concerns grew—slowly, but it was a very confus-
lessons from the age of 11. He became an avid reader and, as an
ing situation. Was it her fault? What was she doing wrong? Was
adult, enjoyed competing with colleagues to be the first to complete
this old age irascibility? Was their marriage truly at an end as Brian
the Daily Telegraph crossword. During National Service his previ-
occasionally said?
ously spectator-only interest in sport became participative. His first
Life went on. Brian would lose his spectacles three times in a
marriage ended in divorce in 1968, when he married June and had
morning, and his bad temper caused by mislaying such things was
three children between 1968 and 1971.
becoming more frequent. He also forgot things, but he was oblivi-
Brian’s career with the Civil Service was brought to an end fol-
ous of the impact on June. A good friend occasionally commented,
lowing a serious car accident in 1990 when he lost the confidence to
but he bit his head off, telling him to mind his own business.
return to work. Instead he became a house husband and June took
June suspected that the decline in Brian’s feel-good factor was
over the role of primary earner. Brian received appropriate medical
linked to his failing memory. She read in September 2007 that local
help to come to terms with the crash and he succeeded in ceasing
university research was being carried out into the effectiveness of
Prozac around 1995.
omega 3 on memory loss.
In their leisure time June loved to cook, paint, write, and explore
Brian got a place on that study and went through psychometric
creative opportunities, which Brian got swept along with very will-
testing every month for four months whilst taking either omega 3
ingly. Grandchildren shared their time and brought great joy. Fond
or a placebo daily. After four months, both felt that there was no
of travelling, June and Brian visited most of the world and had a
improvement and in due course we learned that he had taken a
wide network of friends throughout. Their home was dubbed
placebo, not the active treatment. What was to prove fundamental,
‘International House’ due to the high number of nonUK visitors in
however, was the written report he received at the end of February
the form of students and others who visited. More than 500 new
2008. The report showed that:
friends from 42 countries stayed at their home from 1992 until 2009.
So, it was a very varied and interesting life that filled Brian’s time. ◆ at initial assessment Brian’s immediate, delayed, and verbal rec-
It was early in 2005 that Brian seemed to be changing, developing ognition memory were well below expected limits for his age
feelings of lesser wellbeing than he had had before. Mood swings ◆ his verbal recognition memory had deteriorated over the
and irrationality were noticed. At times he was irascible, unreason-
4 months, and
able, and unkind, but not so frequently that it became a major issue
for them. It was tolerable. ◆ his visual memory had declined and now fell below expectations
June’s diary entry for 24 November 2005 read: for his age.
What have I genuinely done wrong today? According to Brian There was no real steer for them to take the results forward and
everything is wrong and I am push, push, push yet all I asked him to the report was definitely not alarmist. The summer of 2008 was busy
do before I left for work was to shop at the supermarket. Instead he yet good for them with lots of visitors and interesting activities.
However, Brian’s problems seemed to be worsening, so they started Minutes later Brian inserted a credit card which incurred an
to keep diary entries of concerns. additional charge of £51. Explaining the difference, June made sure
that Brian understood the usage of each card.
Examples The next day, June asked Brian to give her a debit card. He handed
her a credit card. It was as though the previous day’s conversation
1. Super glue had not occurred. Brian had always been financially astute, but now
On 1 August, despite a really good day, everything changed in the confusion seemed to have taken over.
early evening. As a result of June being unable to say ‘yes, that is
right’ prior to Brian adding super glue to a project, he flared up say- 5. Tiles in bathroom
ing, ‘You don’t trust my opinions. If we are together in five years, I’d Discussions took place regarding refurbishing the family bath-
be surprised. We might as well call it a day now as we’ve got nothing room. Sons and bathroom fitters all advised against tiling over
left in our marriage.’ existing tiles, so June and Brian agreed with the tiler that he would
The evening was full of Brian making accusations interspersed remove all the tiles at extra cost and make good the walls before
with bouts of hurtful silence. The only way out of the situation was retiling.
for June to apologize, which she did. On the morning that the tiling was going to start, Brian said to
Brian’s reaction had been unfair, mean, and totally out of pro- June ‘I hope he doesn’t tile over the existing tiles’. Brian had totally
portion. Had June said ‘yes, this is right’ when she didn’t know forgotten the exhaustive discussions the previous day.
whether it was or not, he would have blamed her if it had turned out
wrongly. She was being truthful, but Brian saw it as unsupportive. 6. Directions
With hindsight she should have agreed with him to keep the peace
Their youngest son described the route: ‘Take next right, right
and manage the consequences.
again, and right again’.
Two days later, Brian could not remember the incident and said
Seconds later, Brian said, ‘I have to take the next three lefts, don’t
that he couldn’t believe he had said such hurtful things and was
I?’ It was as though Brian’s brain swapped choices for the wrong
sorry. Clearly he had been frustrated about the placement of the
alternative. This happened very regularly. Another example was
super glue and, in seeking an end to the confusion, he lashed out
shopping in a supermarket, where he would choose the opposite,
verbally.
e.g. buying just onions when they had lots of onions but no other
2. Renewing car tax vegetables.
Only by recording these instances could June and Brian evaluate
Brian told June that he didn’t know what a log book for a car looked what was happening to them. It had been easy to dismiss them in
like, yet he had owned cars for 45 years and had often dealt with log the beginning, but they began to impact so significantly that their
books. He became upset over the inability to tax the car six weeks in children started to notice too.
advance, even though it was unnecessary to be so premature. The final deciding factor came when, planning to visit friends in
This was such a surprise to June because she had always joked Colombia, the home of salsa, they enrolled for classes to learn the
that Brian was born in the driving seat of a car! But it was clear dance. Both had enjoyed many years dancing together, especially
that he had confused needs with timing. Afterwards he said that the cha cha cha, the jive, the quick step, and other ballroom dances,
he found detail too much, even though historically he had dealt but had never tried the salsa. Brian, an ex silver medallist, couldn’t
with many such issues. He couldn’t remember June renewing the learn the steps which June found easy. This was incredible and a
previous year’s car tax by mobile phone from Inverness dockside, shock to both of them.
although it was a unique occasion. He couldn’t even remember They couldn’t go on this way and, having run out of self-help
being in Inverness, yet they had taken a Wedding Anniversary options, realized that the time had come to involve their GP.
cruise down the Caledonian Canal. Having booked an appointment with their GP in August 2008
3. Forgotten telephone call they sent him a letter setting out:
One evening, June suggested that Brian phone the dog-sitter to ◆ their concerns with actual examples of Brian’s behaviour
arrange cover for the next day. He went off to do so whilst June ◆ the results of the omega 3 research, and
cooked dinner for guests.
◆ a request that he take every reasonable step to assist them.
The next day at breakfast June asked whether cover was OK.
Brian said that he hadn’t called the sitter. Their guests were able to Appreciating the detailed information, the GP executed simple
say that Brian did make the call because last evening during dinner tests on Brian which he failed miserably. Taking their concerns seri-
he had told them it was OK. Brian accepted this but had absolutely ously, he:
no recollection of making the call. The sitter confirmed that she was ◆ arranged blood tests which resulted in him recommending regu-
expecting the dog. This incident was surprising because all his life lar vitamin B12 injections just in case they might help, and
Brian had been a stickler for detail.
◆ made a specialist referral to check Brian’s complete physical
4. Difference between a credit card and a debit card health in case the road crash in 1990 could have had residual
effects.
Approaching the airport car park, June checked with Brian which
card he planned to insert into the machine to match up with their By late September 2008, a consultant old age physician was able
booking. Brian replied, ‘A debit card’. She replied ‘Well done, that’s to report, following a thorough examination, that in his opinion
right’. there was no underlying medical reason for Brian’s deteriorating
CHAPTER 30 hello, i’m me! living well with dementia 387
condition and that he would benefit from referral to a memory The psychogeriatrician’s early diagnosis gave Brian the tools to
clinic and the help of a consultant psychogeriatrician. He also put the quality back into their lives. Neither could believe that two
ordered a CT head scan which showed nothing abnormal. tablets of Citalopram could effect such a dramatic change in his
An initial assessment by a memory service followed a month later, feeling of wellbeing. Moreover, of course, it wasn’t the effect of the
but real progress was made at the beginning of February 2009 when medication that was felt immediately, it was the euphoria that they
Brian was seen by the consultant psychogeriatrician. It turned out could see a way forward. Life was worth living again.
to be somewhat of a coincidence that the NHS National Dementia Next steps for them included:
Strategy was released the same day.
◆ selling their large home of 26 years and downsizing to live near their
Again, June and Brian sent in advance of the meeting a list of
two sons and their families in Gloucestershire, at their request, so
instances.
that they would be on hand to provide help and support
June’s diary reflects that this was a massive turning point for
them and that the specialist was wonderful. After a three-hour ◆ making Enduring Powers of Attorney
consultation he gave: ◆ informing the Driver and Vehicle Licensing Agency (DVLA) of
◆ a cautious probable diagnosis of frontotemporal dementia Brian’s difficulties and receiving permission from them to keep
driving for another year.
◆ a prescription for Citalopram
◆ practical help such as advising about an Enduring Power of The consultant met with them again in March 2009 and advised
Attorney, considering their home arrangements, and whether remaining on Citalopram for the rest of Brian’s life. He also ordered
downsizing to live nearer family may be wise a SPECT and MRI scan and repeat psychometric testing.
In June 2009, Memantine (Ebixa) was prescribed and by the next
◆ advice about DVLA notification. month they were able to agree that this was very successful. Brian
June and Brian walked out of the hospital on cloud nine, moving was relaxed. He started to find his own solutions to problems—and
from: it worked. For example, he managed to visualize where he had put
his glasses and lost them less often.
◆ doubting their own judgement
June and Brian have spoken publicly about their gratitude for
◆ feeling vulnerable and fearing the worse having met that particular psychogeriatrician and to the National
◆ balancing on a knife-edge of aggression and frustration, and Health Service for giving them back their lives:
Not every day is wonderful and there certainly are some difficult
◆ afraid of each other’s feelings
times—but we are going forward, taking risks and living life to the
to feeling: full. There is still confusion and the need to make adjustments, to be
flexible and to learn.
◆ relief that a healthcare professional had taken them seriously
Of the future, we both agree that this is an unknown quantity. We
◆ relaxed with one another in total honesty and so happy! work closely as volunteers with the NHS and many other organisa-
tions. Both of us have given and received positive help from support
◆ grateful that the diagnosis was not more serious groups. We believe that:
◆ as though a weight had been lifted from their shoulders
◆ each of us must, as far as we are able, take responsibility for our own
◆ thankful that they were given such fine specialist consultancy whole health and wellbeing and that of those close to us. This means
◆ neither naïve nor complacent but confident that they could meet eating and drinking wisely, taking exercise, and looking after our
challenges head on. teeth, eyes, feet, and other things which are so important to us.
Thanks to receiving a diagnosis, they could evaluate how to go ◆ as individuals in possession of a diagnosis it is important for us
forward. Having a diagnosis is really important for many reasons. It to help ourselves by recording important, individual, and timely
stops you worrying that something even more serious, like a brain facts so that they may be passed on to professionals and used
tumour, is causing the problem. Also it is reassuring that there is a as signposts for cost-efficient individual care plans. An exam-
medical reason for the difficulty, and that divorce is not the way out. ple is the Living Well Handbook which has been developed by
It provides some light at the end of the tunnel. NHS Gloucestershire. (available at <http://www.nhsglos.nhs.uk/
Brian found the diagnosis helpful because it helped explain some your-services/older-peoples-services/dementia/>)
of the strange feelings he had been experiencing, a sense of uneasi- ◆ every pound spent on managing dementia must be made to mat-
ness that something serious was going wrong. These were not ter and, if authorities are to treat those with dementia and carers as
things he had been able to talk about before, as he would have said individuals, a ‘one size fits all ’ brand must be buried once and for all.
he felt ‘fine’. That means listening to and involving everyone in planning whilst
Examples of changes in Brian: it is still possible. What is needed is help at the right time when it is
◆ One day soon after diagnosis, Brian took the sole decision to pre- needed, be it respite, support at home, residential, or palliative care.
pare a salad to go with dinner, preparing eleven salad items from ◆ adopting a ‘can do’ attitude, socializing and supporting others
scratch. After dinner he laid the table for breakfast. He hadn’t and not being afraid to take risks, may maintain and improve
even thought of making such decisions for a long time. feelings of wellbeing for as long as possible. Most things one does
◆ The next day, asked to do a small task immediately, he replied in life come with a risk, and living with dementia brings new
‘yes, no problem’. The week before he would have bitten June’s ones. When empathy is short, we need to be patient and seek
head off! greater understanding through effective communication.
◆ carers must be listened to and supported, for they hold the key to ‘Nothing ventured, nothing gained’ was a favourite phrase of our
cost-efficient and loving care for the ones they hold dear. Failure mothers all those years ago. It is a philosophy worth remembering
to consult carers can waste time and resources and causes sheer now when living with dementia. New memories cannot be made and
frustration on occasions when medical intervention is necessary. the cumulative effect of joy experienced by those with a short-term
Providing breaks must never be forgotten, for an ailing battery memory is denied to those living with dementia. On the other hand,
will eventually cease to function. similarly, upsets are quickly forgotten, which is indeed a blessing.
CHAPTER 31
Epidemiology of dementia
Laura Fratiglioni and Chengxuan Qiu
Dementia is defined as a clinical syndrome characterized by multi- Occurrence

ple cognitive deficits that are severe enough to interfere with daily
functioning, including social and professional activities (American The occurrence of a disease in a specified population is commonly
Psychiatry Association, 1987). Cognitive deficits, as a central com- described by prevalence and incidence. Prevalence refers to the
ponent of the dementing process, involve impairment in memory proportion of people affected by the disease in a defined popula-
and at least one other cognitive domain, such as aphasia, apraxia, or tion at a given point in time or a short period of time, whereas
disturbances in processing speed and executive function. Clinically, incidence is defined as the number of new cases that occur during
Alzheimer’s disease (AD) is the most common cause of demen- a specified period of time in a population at risk for developing
tia, accounting for 60–70% of all dementia cases (Fratiglioni and that disease. Prevalence is determined by both incidence and dura-
Rocca, 2001). tion of the disease. The disease-specific mortality rate can be used,
Dementia has become increasingly relevant from scientific and too, as a measure of incidence, provided that the case-fatality rate
public health perspectives, owing to the worldwide universal phe- is high and that the survival duration of the disease is short and
nomenon of population ageing. In 1990, 26 nations in the world constant (Gordis, 2000). Case-fatality rate refers to the percentage
had more than two million citizens aged 65+ years; it was projected of persons with a specified disease who die as a result of that ill-
that 34 more nations would join the list by 2030 (Kinsella and ness within a given time period. The disease-specific mortality rate
Velkoff, 2002). High-income countries, which have already expe- refers to the probability of dying from a certain disease in a defined
rienced a dramatic increase in absolute number and proportion of population and time period, which is determined by incidence and
old people, will see a continuous increase in their ageing population case-fatality rate.
(Christensen et al., 2009). Due to exponential increase in the risk
of dementia occurrence with increasing age, dementia has become
Prevalence
one of the most common diseases among older people, as well as Since the 1980s, numerous community-based studies across the
the major cause of disability, poor quality of life, institutionaliza- world have been initiated to determine prevalence of dementia and
tion, and death, especially among the oldest old (Agüero-Torres major subtypes of dementia. Meta-analyses or pooling analyses
et al., 1998, 1999, 2001; Börjesson-Hanson et al., 2007). provide more precise estimates of dementia prevalence (Fratiglioni
Since the 1980s, the scientific community has increasingly et al., 1999; Lobo et al., 2000). Despite varying inclusion criteria, the
focused its research on all-cause dementia and its main subtypes of meta-analyses result in rather similar patterns of age-specific prev-
AD and vascular dementia (VaD). The epidemiology of dementia alence of dementia. The prevalence is very low before 60 years of
is one of the leading areas in research of ageing and health, aiming age, and then increases exponentially with advanced age (Fig. 31.1).
to develop preventive and therapeutic strategies against dementing The age-specific prevalence rates of dementia are estimated to be
disorders. Traditionally, prevention of a disease is distinguished as approximately 1% in people aged 60–64 years, 1.5% in 65–69 years,
primary, secondary, and tertiary prevention. Primary prevention 3% in 70–74 years, 6% in 75–79 years, 13% in 80–84 years, 24% in
aims to reduce incidence of the disease by interventions target- 85–89 years, 34% in 90–94 years, and 45% in the most advanced
ing specific risk factors that may decrease risk or delay the onset age of 95+ years (Fratiglioni et al., 2008). After 90 years of age, the
of the disease. Secondary prevention is to reduce incidence and prevalence of dementia may continue to double every 5 years in
prevalence of clinical disease by detecting the disease at an ear- women but not in men (Corrada et al., 2008). A study based on
lier stage, preventing its progression from preclinical phase to a the Delphi consensus approach estimated that global prevalence of
full clinical picture, or shortening its duration. Tertiary preven- dementia among people aged 60+ years was 3.9%, with the regional
tion seeks to reduce impact of complications of long-term disease prevalence being 1.6% in Africa, 3.9% in Eastern Europe, 4.0% in
and disability, which consists of measures to improve or maintain China, 4.6% in Latin America, 5.4 in Western Europe, and 6.4% in
quality of life, minimize patients’ suffering, and maximize poten- North America (Ferri et al., 2005).
tial years of useful life. This chapter is devoted to the epidemiology With regard to dementia subtypes, pooled data of 11 stud-
of dementia concerning its occurrence, determinants, and primary ies in Europe show that, among people aged 65+ years, the
prevention. age-standardized prevalence is 6.4% for all-cause dementia, 4.4%
70 100
90
60
Incidence, per 1000 person-years

80
Prevalence, per 100 population
50 70
60
40
50
30 40
20 30
20
10
10
0 0
65–69 70–74 75–79 80–84 85–89 90–94 95+Age 65–69 70–74 75–79 80–84 85–89 95+Age
Lopes et al. 2007; Worldwide Nitrini et al. 2009; Latin America Kukull et al. 2002; USA Jorm & Jolley 1998; East Asia
Brayne 2006; UK Lobo et al. 2000; Europe Fratiglioni et al. 2000; Europe The CSHA Group 2000; Canada
Dong et al. 2007; China Anstey et al. 2010; Australia Matthews & Brayne 2005; UK Plassman et al; 2011; USA
Fig. 31.1 Age-specific prevalence rates of dementia (per 100 population) across Fig. 31.2 Age-specific incidence rates of dementia (per 1000 person-years) across
the world (Lobo et al., 2000; Brayne 2006; Dong et al., 2007; Lopes et al., 2007; the world (Jorm and Jolley, 1998; Fratiglioni et al., 2000; Canadian Study of Health
Nitrini et al., 2009; Anstey et al., 2010). and Aging Working Group, 2000; Kukull, et al., 2002; Matthews and Brayne, 2005;
Plassman, et al., 2011).
for AD, and 1.6% for VaD (Lobo et al., 2000). Similar to all-cause approximately 1 in people aged 60–64 years, 2–3 in 65–69 years, 6–8
dementia, prevalences of AD and VaD also increase exponentially in 70–74 years, 11–18 in 75–79 years, 20–30 in 80–84 years, 30–50
with advancing age up to 90 years old. Of all dementia cases, AD in 85–89 years, and 70–75 in people 90+ years of age. The lifetime
and VaD account for 54% and 16%, respectively. Earlier studies risk for dementia was estimated to be approximately one in six for
showed a higher proportion of VaD cases in Asia, but later studies middle-aged men and one in five for middle-aged women (Seshadri
did not confirm this pattern. A multicentre survey in China showed and Wolf, 2007; Lobo et al., 2011). Pooled data from Europe suggest
that the prevalence was 3.5% for AD and 1.1% for VaD in people that, on the basis of clinical diagnosis, AD and VaD account for
aged 65+ years (Zhang et al., 2005), suggesting that the proportions 60–70% and 15–20% of all incident dementia cases, respectively.
of dementia subtypes in China are comparable with those reported However, population-based neuroimaging and neuropathological
from Europe and North America. studies reveal that the clinical syndrome of dementia often occurs
As population ages, the number of patients with dementia with concomitant Alzheimer pathologies (i.e. amyloid plaques and
is anticipated to double every 20 years (Wimo et al., 2003; Ferri neurofibrillary tangles) and cerebrovascular disease, suggesting
et al., 2005). In Europe, the number of dementia cases in 2010 was that mixed dementia may be the most common form of demen-
estimated to be more than 6 million; this number is projected to tia (Schneider et al., 2007a; Viswanathan et al., 2009; White, 2009;
reach 14 million in 2050 (Mura et al., 2010). In the US, the total Stephan et al., 2012).
numbers of patients with AD were 4.5 million in year 2000 and There are inconsistent findings with respect to whether the inci-
5.4 million in 2010; the number is projected to reach 13.2 million dence of dementia continues to increase until very advanced ages or
by 2050 (Hebert et al., 2003; Alzheimer’s Association, 2011). In the it will reach a plateau at a certain age. This is relevant for projecting
Asia-Pacific region, the number of dementia cases is anticipated to burden of the disease and for understanding its aetiology. A consist-
increase from 13.7 million in 2005 to 64.6 million by 2050 (Access ent increase with advanced age in dementia incidence suggests that
Economics, 2006). Alzheimer’s Disease International (2010) esti- dementia is an inevitable consequence of ageing, whereas a conver-
mated that in 2010 there were 35.6 million people living with gence to or a decline at certain age suggests that very old people may
dementia worldwide; the number will increase to 65.7 million by be less vulnerable to dementia, owing perhaps to less susceptibility
2030 and 115.4 million by 2050. Nearly two-thirds of all demen- to genetic or environmental factors or their interaction (Miech et al.,
tia patients in the world live in low- and middle-income countries, 2002). Pooled data of population-based studies in Europe suggested
where the sharpest increases in numbers of patients are anticipated a decline in dementia incidence mainly in very old men (Andersen
in the coming decades. et al., 1999). The US Cache County Study found that incidence of
dementia increased with age, peaked, and then started to decline at
Incidence extremely old ages for both men (over 93 years) and women (over
In the last two decades, numerous incidence studies of dementia 97 years) (Miech et al., 2002). Meta-analyses and large-scale stud-
have become available. Meta-analyses have provided rather stable ies in Europe, however, provide no evidence for potential decline in
estimates of incidence of dementia across all regions of the world. dementia incidence among the oldest old (Fratiglioni et al., 2000a;
These analyses show a similar pattern of the age-specific incidence Matthews and Brayne, 2005; Ravaglia et al., 2005a). The apparent
of dementia, i.e. the incidence rate increases almost exponentially decline found in some studies may be an artefact of poor response
with age across all regions (Fig. 31.2). The age-specific incidence rates, selective survival, and the nature of study population sampled
rates of dementia (per 1000 person-years) are estimated to be in very old people (Matthews and Brayne, 2005).
CHAPTER 31 epidemiology of dementia 391
The age-specific incidence rates of dementia are quite similar senile and multi-infarct dementia during the periods 1947–1957
in younger-old age groups (e.g. <75 years) across different con- and 1957–1972 (Rorsman et al., 1986). Since then, findings
tinents, but substantial variations are reported among older ages across numerous studies, however, have been mixed. Studies that
(Fratiglioni et al., 1999). A lower incidence of dementia has been rely on medical records and death certificates have shown an
reported in North America than in Europe and Asia; these varia- increased prevalence of dementia over time (Taylor et al., 2004;
tions are most likely to be due to differences in the study design and Mathillas et al., 2011), but this is likely owing to growing aware-
case-ascertainment procedure. In addition, pooled data of eight ness of dementia among the public and health professionals. The
studies in Europe suggest geographical dissociations across coun- relatively stable prevalence of dementia from the late 1980s to
tries, with higher incidence rates being found among the oldest-old early 2000s has been reported in a few studies in Europe and the
of northwestern than southern countries (Fratiglioni et al., 2000a). US (Lobo et al., 2007; Hall et al., 2009; Rocca et al., 2011). Some
Methodological variations, rather than different regional distribu- studies from Asia showed a nonsignificant increase in dementia
tions of possible risk factors for dementia, may be responsible for prevalence from the 1980s through the 2000s (S. Li et al., 2007;
this pattern. To determine whether there are geographical variations Sekita et al., 2010). Few population-based studies have examined
in dementia incidence among five areas in England and Wales that the secular trends of survival of patients with dementia. The death
are characterized by different vascular risk factor profiles, the study certificate-based studies revealed trends of an increased mortality
used identical methods but found no evidence of variation across of AD from the 1980s through the 2000s (Hoyert and Rosenberg,
areas in incidence of dementia (Matthews and Brayne, 2005). 1997; Griffiths and Rooney, 2006; Steenland et al., 2009), but this,
again, could be due to the increasing awareness of the disease over
Mortality time. Population-based studies from the US revealed a nonsignifi-
The malignant nature of Alzheimer’s dementia was first suggested cant decline (annually approximately 3%) in the risk of dementia
in the Shanghai study of the community residents aged 65+ years, from 1997 through 2008 (Hebert et al., 2010; Rocca et al., 2011).
in which AD was found to increase the risk of death by three-fold, The inconsistent findings until the early 1990s are largely due to
compared to a four-fold risk of death owing to cancer (Katzman methodological issues such as the diversities in terminology and
et al., 1994). This is reinforced by population-based follow-up varying diagnostic procedures and criteria for dementia. While a
studies and pooled data, which show that people with dementia few population studies have recently reported stable prevalence of
have 2–5 times higher risk of dying than those without dementia dementia over time, evidence has emerged that the incidence or
(Baldereschi et al., 1999; Jagger et al., 2000; Helmer et al., 2001; risk of dementia may decrease, as seen for cardiovascular disease
Noale et al., 2003; Gühne et al., 2006). Indeed, follow-up studies (Schmidt et al., 2012; Whalley, 2012).
have consistently shown that dementia shortens life-expectancy,
with the median survival time of patients with dementia ranging Determinants
from 2–4 years after the diagnosis, depending on the age of demen-
tia onset and other demographic features (Agüero-Torres et al., A life-course approach
1999; Witthaus et al., 1999; Wolfson et al., 2001; Helmer et al., 2001; The life-course approach considers biological and environmental
Börjesson-Hanson et al., 2007; Helzner et al., 2008; Mehta et al., factors acting during early life, middle age, and late life as relevant
2008; Xie et al., 2008; Wang et al., 2010). The proportion of deaths for determining occurrence of the disease in late life. This approach
attributable to dementia increases with age; in persons aged 85+ seeks to identify time windows when exposures have the greatest
years, dementia accounts for approximately one-third of deaths effect on health outcome and to determine whether accumulative
(Aevarsson et al., 1998; Tschanz et al., 2004). Although dementia exposures over the life-course have integrative or additive effects on
has been widely regarded as a leading cause of death, death certifi- health (Kuh and Ben-Shlomo, 2004; Whalley et al., 2006; Richards
cates usually grossly underreport dementia (Jin et al., 2004), even and Hatch, 2011). Thus, from the life-course perspective, the risk
when multiple causes of death are taken into account (Beard et al., of dementia occurring in late life is not determined in any single
1996; Ganguli and Rodriguez, 1999). The US national study suggest time period of the lifespan; rather, it reflects the result of complex
that mortality rates of dementia heavily depend on the methods interactions of genetic susceptibility, biological factors, and envi-
used to ascertain the causes of death; when dementia was diagnosed ronmental exposures experienced over the lifespan. For instance,
according to informant interview, the mortality of dementia was the reserve hypothesis, which is proposed to interpret the dispar-
doubled compared with dementia that was diagnosed from review ity between clinical phenotype of cognitive function and load of
of medical records, whereas this rate was 5–9 times greater than the neuropathologies in the brain (Katzman, 1993; Stern, 2002), can
frequency in which dementia was even mentioned on death certifi- be conceived as the sum of its lifetime input (Richards and Deary,
cates, and 15–17 times greater than the rate in which dementia was 2005). As an example in understanding the aetiology of a disease
certified as the underlying cause of death (Lanska, 1998). In the US, from a life-course perspective, evidence has emerged that genetic
AD was the seventh leading cause of death across all ages, and the and nongenetic factors over the lifespan determine the risk of
fifth leading cause of death for people aged 65+ years (Alzheimer’s late-life dementia (Fig. 31.3). Furthermore, the life-course model
Association, 2011). The numbers of death attributed to AD and dia- introduces the concept of ‘time window’ at exposure that is highly
betes are almost identical. relevant for chronic disease with a long-term latent period such
as dementia. A factor might increase the risk of a disease if expo-
Trends over time sure occurs during a certain time period, whereas the same factor
The secular trends in dementia occurrence have been a focus of may show a differential effect on the risk of the disease in another
research since the 1980s, when the Lundby study in south Sweden period over the lifespan, due to various mechanisms, differential
showed no substantial changes in prevalence and incidence of interactions with other factors, or selective survival. For instance,
Risk factors Smoking, depression, inflammatory markers, hyperhomocysteinaemia,

diabetes, occupational exposures and traumatic brain injury, etc.
APOE ε4 allele, familial Hypertension, hyperlipidaemia, obesity, heavy alcohol Low BP, frailty,
aggregation, etc. intake and microvascular disease, etc. weight loss, etc.
Birth Childhood 2nd decade Adulthood-Middle age Transitional age Old age
0 20 60 75 AGE
IQ, high education, ■ Psychosocial factors: Physical exercise, mentally-stimulating activity,
high SES, etc. rich social network and social engagment, etc.
■ Nutritional factors: Mediterranean diets, vegetables, fruits, fish (ω-3 or
Protective factors ω-6 PUFAs), B-Vitamins and antioxidants (e.g. vitamins C and E), etc.
Fig. 31.3 Determinants of late-life dementia from a life-course perspective. APOE, apolipoprotein E gene; BP, blood pressure; IQ, intelligence quotient; SES,
socioeconomic status; PUFAs, polyunsaturated fatty acids.
midlife high blood pressure is a risk factor for late-life dementia, Major risk and protective factors
whereas low blood pressure in late life also is related to an elevated
Demographic features
risk of dementia, indicating that high and low blood pressure may
Age and sex
be involved in the development or expression of dementia through
Age is the most powerful determinant of dementia. The preva-
different mechanisms at different ‘time windows’ over the lifespan
lence and incidence of dementia rise almost exponentially with
(Qiu et al., 2005; Ruitenberg et al., 2005).
advanced age, with the risk of dementia being almost doubled every
Aetiological hypotheses 5 years from age 65–85 (Ziegler-Graham et al., 2008) (see Figs 31.1
and 31.2). Higher prevalence and incidence rates of dementia have
The dementias are multifactorial disorders, in which genetic and
been reported in women than in men in numerous cross-sectional
nongenetic factors as well as their interactions contribute to their
and longitudinal studies. The pooled follow-up data from Europe
initiation, clinical expression, and progression. Identifying the
confirm the gender-specific pattern (Launer et al., 1999; Fratiglioni
influential factors has relevant implications for understanding its
et al., 2000a). However, gender difference in the occurrence of
pathological mechanisms and for the development of therapeutic
dementia was not found in some studies from the US (Bachman
and preventive strategies against dementia. Aetiological studies
et al., 1993; Rocca et al., 1998; Kawas et al., 2000; Katz et al., 2012),
have focused mostly on all-cause dementia and specific AD; while
except among the oldest old (Corrada et al., 2008). The higher inci-
the aetiological profile for VaD is largely similar to that for stroke,
dence of dementia in women than in men, especially among the old-
studies concerning the aetiology of other forms of dementia are
est old, may be due to different mechanisms (Fratiglioni et al., 1997;
limited.
Fratiglioni and Rocca, 2001): (1) men who survive to advanced ages
Several aetiological hypotheses for late-onset dementia have
may become more resistant to dementia; (2) men are more frequently
been proposed (Fratiglioni et al., 2008; Querfurth and LaFerla,
exposed to multiple risk factors such as occupational toxic agents
2010; Kuller and Lopez, 2011; Reitz et al., 2011) (Table 31.1). It
and head trauma, which may lead to earlier occurrence of dementia;
is known that the genetic susceptibility plays a part in not only
and (3) older men have a higher oestrogen level than older women,
familial but also sporadic late-onset AD. The vast majority of ear-
which may partially protect men from dementia.
ly-onset familial AD cases are caused by mutations in a few known
genes, whereas apolipoprotein E gene (APOE) ε4 allele is the main Education, occupation, and socioeconomic status (SES)
susceptibility gene for sporadic AD (Blennow et al., 2006). Of the The Shanghai study was the first report to link low education to an
nongenetic factors, increasing evidence from multidisciplinary increased prevalence of dementia (Zhang et al., 1990). This find-
research supports a few aetiological hypotheses for late-onset ing has been replicated in numerous prevalence studies (Schmand
dementia: (1) the vascular hypothesis implies that cardiovascular et al., 1997; De Ronchi et al., 1998) and incidence studies (Evans
risk factors and related disorders occurring over a life-course are et al., 1997b; Qiu et al., 2001; Di Carlo et al., 2002), and confirmed
involved in the pathogenesis and clinical expression of dementia by systematic reviews and meta-analyses (Caamaño-Isorna et al.,
(Qiu et al., 2005; Chui, 2006); (2) the psychosocial hypothesis 2006; Sharp and Gatz, 2011). Childhood cognitive ability, premor-
states that maintaining a mentally and socially integrated lifestyle bid intelligence quotient (IQ), and bilingualism have protective
in mid- or late life may protect against or delay onset of demen- effects against the onset of dementia (McGurn et al., 2008; Craik
tia by providing the brain with functional and structural reserve et al., 2010; Yeo et al., 2011). People with high socioeconomic posi-
or reducing psychological stress; (3) the inflammatory hypothesis tion from early life onwards had a lower risk of dementia compared
is based essentially on observation that brain neuritic plaques are with those with consistent low position, suggesting that early-life
associated with inflammatory proteins, indicating involvement of exposure to socioeconomic disadvantage may increase the risk of
inflammation in causing neurodegeneration; (4) other biological late-life dementia (Scazufca et al., 2008; Zeki et al., 2011). Because
mechanisms such as oxidative stress and exposure to neurotoxic education and socioeconomic position are both highly corre-
agents also may contribute to the pathogenetic processes of demen- lated, some studies have examined their independent association
tia and AD. It should be noted that these aetiological hypotheses with dementia by relating simultaneously both factors to demen-
are not mutually exclusive. tia risk, in which an independent association was detected only
Table 31.1 Putative risk and protective (in italic) factors for late-onset dementia by aetiological hypotheses from a life-course perspective
Aetiological hypothesis Time periods (‘windows’) acting over the lifespan
At birth Childhood and early Middle age and Old age

adulthood transitional age
Genetic APOE ε4 allele, Familial aggregation
familial aggregation
Vascular Hypertension, obesity, Smoking, very high BP, low BP, diabetes,
dyslipidaemia, smoking, dyslipidaemia, atherosclerosis, cerebral
diabetes, alcohol abuse; microvascular disease, heart disease, high
Antihypertensive therapy homocysteine;
Antihypertensive therapy, limited alcohol
consumption, diets (e.g. ω-3 PUFAs)
Psychosocial Low SES; Depression or psychological Depression or depressive symptoms;
High education, stress; Social engagement, Social network, social engagement, mental
premorbid intelligence physical and mental activity activity, physical activity
Inflammatory Inflammatory markers Inflammatory markers, use of NSAIDs,

hormone replacement therapy
Others: oxidative stress and Aluminium, mercury, Head trauma, deficiency in nutrients
neurotoxic occupational toxic agents (e.g. vitamins A, E, C, B12 and folate)
(e.g. solvents)
AD, Alzheimer’s disease; BP, blood pressure; NSAIDs, nonsteroidal anti-inflammatory drugs; SES, socioeconomic status; PUFAs, polyunsaturated fatty acids.
with education (Evans et al., 1997b; Ravaglia et al., 2002; Karp neuroimaging features, and clinical phenotypes, owing to genetic
et al., 2004). Neuroimaging research also indicated that education susceptibility and gene–environment interactions (van der Flier
played a more important role in providing reserve compared to et al., 2011). First-degree relatives of AD patients have a higher
occupational position and leisure activities (Foubert-Samier et al., lifetime risk of developing AD than the general population or rela-
2012). The effect of high education may be due to multiple mecha- tives of nondemented subjects (Devi et al., 2000; Green et al., 2002;
nisms, such as increased cognitive reserve, high SES in early life, Seshadri and Wolf, 2007). Familial aggregation has been reported
and intelligence (as measured by a surrogate of IQ) (Moceri et al., for both early- and late-onset AD, in which both genetic and envi-
2000; Whalley et al., 2000; Karp et al., 2004; McDowell et al., 2007). ronmental factors as well as their interactions are likely to contrib-
Alternatively, people with high education are more likely to have ute to the phenomenon. Twin studies provide an opportunity to
been diagnosed with dementia at a later stage of neuropathologies address this issue by comparing concordance rates among monozy-
than less educated persons (Qiu et al., 2001). gotic twins, who share both genes and early-life environment, to
Lifetime occupational history may be involved in the devel- those among dizygotic twins, who share only 50% of the genes and
opment of late-life dementia. Having manual work as the life- early-life environment. In the Swedish Twin Registry, the herit-
time principal occupation, especially in manufacturing industry, ability of AD is estimated to be 58–74%, whereas other variances
has been related to dementia in some studies (Stern et al., 1994; are attributable to nongenetic environmental factors (Gatz et al.,
Bonaiuto et al., 1995; Qiu et al., 2003a), suggesting possible involve- 2006). However, the familial aggregation of dementia and sporadic
ments of occupational exposure in dementia. Indeed, occupational AD can only be partially explained by known genetic factors such
exposures to solvents and heavy metals (e.g. aluminium and mer- as APOE ε4 allele, indicating that other susceptibility genes may
cury) have been suggested, but not confirmed, as risk factors for be involved in Alzheimer’s dementia (Huang et al., 2004; Hayden
AD (Mutter et al., 2004; Perl and Moalem, 2006). In addition, et al., 2009).
occupational exposure to extremely low-frequency magnetic fields
APOE genotypes
has been related to dementia and AD in several follow-up studies
(Sobel et al., 1995, 1996; Feychting et al., 2003; Hakansson et al., The APOE ε4 allele is the only established genetic factor for late-on-
2003; Qiu et al., 2004a; Röösli et al., 2007; Andel et al., 2010) and a set sporadic AD and dementia, although it has also been linked
positive association has been concluded by a meta-analysis (García to familial dementia (Verghese et al., 2011). APOE ε4 allele, as a
et al., 2008). susceptibility gene, confers an increased risk for sporadic AD in a
dose-dependent manner (Qiu et al., 2004b); carrying one copy of
Genetic susceptibility the ε4 allele increases risk of AD by approximately three times and
Gene mutations and familial aggregation two copies by about 12 times. Furthermore, carrying one or two
Mutations in amyloid precursor protein, presenilin-1, and copies of the ε4 allele will lead to an earlier onset of AD by about
presenilin-2 genes cause early-onset familial AD that accounts for 10–20 years compared with noncarriers (Verghese et al., 2011).
approximately 1–3% of all AD cases (Blennow et al., 2006). The However, the APOE ε4 allele is neither necessary nor sufficient for
vast majority of Alzheimer patients are sporadic, with consider- the development of AD; that is, even carriers of homozygotic ε4
able heterogeneity in their risk profiles, neuropathological and allele do not necessarily develop AD, whereas persons without the
ε4 allele may have the disease. Overall, approximately 15–20% of 2008; Weyerer et al., 2011), leading to the hypothesis that limited
dementia cases are attributable to the APOE ε4 allele (Evans et al., alcohol intake may protect against dementia. Indeed, a systematic
1997a; Qiu et al., 2004b), but the risk effect of the ε4 allele decreases review found that limited alcohol intake in earlier adult life might
with increasing age (Farrer et al., 1997). The age-varying effect of be protective against dementia, although there is significant het-
APOE ε4 allele may be a result of its interaction with age on brain erogeneity among reviewed studies (Peters et al., 2008c). However,
structure (Filippini et al., 2011). the protective effect of limited alcohol intake against dementia
The suggested association of AD and subtype demen- remains uncertain because the inverse association may be due to
tia with a few other candidate genes or genetic loci, such as information bias, confounding of healthy lifestyle or high SES, dif-
angiotensin-converting enzyme gene, cholesterol 24-hydroxylase ferent approaches in exposure assessments, or misclassification of
gene, fat or obesity-associated FTO gene, and insulin-degrading outcomes. In support of this cautiousness, the deleterious effect of
enzyme gene that are often related to vascular system, remains to alcohol consumption also emerges from two Finnish studies where
be established (Yip et al., 2002; Keller et al., 2011; Guerreiro et al., heavy or binge drinkers at middle age had a more than three-fold
2012; Schrijvers et al., 2012). increased risk for late-life dementia (Järvenpää et al., 2005), espe-
cially among carriers of the APOE ε4 allele (Anttila et al., 2004).
Cardiovascular risk factors
Smoking Overweight or obesity
Experimental and epidemiological studies have suggested that nico- Numerous epidemiological studies have investigated the associa-
tine may exert a protective effect against Parkinson’s disease (Ross tion of overweight or obesity from middle age to late life with risk
and Patrovitch, 2001), which may be true for other neurodegen- of dementia. A systematic review has revealed a lifespan-dependent
erative diseases such as AD. Earlier studies seemed to support this association between body mass index (BMI) and risk of demen-
hypothesis as a lower odds ratio of AD was often reported in smok- tia such that a higher BMI in midlife is a risk factor for dementia,
ers (Fratiglioni and Wang, 2000). However, the apparent protective whereas a decline in BMI during late life may anticipate occur-
effect is due probably to selective survival bias because smokers are rence of dementia (Gustafson, 2006). Indeed, several studies have
proportionally less numerous among prevalent AD cases (Wang linked a higher BMI or central obesity (e.g. a high waist-to-hip
et al., 1999; Hill et al., 2003; Tyas et al., 2003). When incident cases ratio or sagittal abdominal diameter) at around 30–50 years of
are examined, the protective effect of smoking no longer exists age to an increased risk of dementia occurring more than 20 years
(Wang et al., 1999; Doll et al., 2000). Indeed, numerous follow-up later (Kivipelto et al., 2005; Rosengren et al., 2005; Whitmer et al.,
studies found an increased risk of dementia associated with smok- 2005a, 2008; Gustafson et al., 2009; Hassing et al., 2009; Xu et al.,
ing, even with secondhand smoking (Tyas et al., 2003; Aggarwal et 2011). In the cohort of Japanese-American men, a greater decline
al., 2006; Barnes et al., 2010); the association may vary by the APOE in BMI approximately 10 years prior to dementia was detected
ε4 allele (Ott et al., 1998; Merchant et al., 1999; Reitz et al., 2007). (Stewart et al., 2005). A lower BMI approaching to or during
Long-term observational studies revealed that midlife smoking, dementia onset was observed in the long-term cohort study of
especially heavy smoking (≥ 2 packs per day), doubles the risk of Swedish women (Gustafson et al., 2012). In line with this finding,
dementia (Rusanen et al., 2010, 2011) and increases the mortality some follow-up studies of older people suggest that weight loss or
due to dementia (Alonso et al., 2009a), even though the association accelerated decline in BMI is associated with subsequent develop-
of smoking with dementia is affected by selective survival or differ- ment of dementia (Buchman et al., 2005; Johnson et al., 2006; Gao
ential attrition (Tyas et al., 2003; Hernán et al., 2008; Chang et al., et al., 2011; Ogunniyi et al., 2011). Although some studies suggest
2012; Weuve et al., 2012). Meta-analyses of prospective studies con- an association of late-life central obesity with AD (Luchsinger et al.,
firm that current smoking as compared with never smoking is asso- 2011), more studies indicate that low BMI in old ages is related to a
ciated with 50–80% increased risk for AD and dementia (Anstey higher risk for subsequent occurrence of dementia (Nourhashemi
et al., 2007; Peters et al., 2008d; Cataldo et al., 2010). A systematic et al., 2003; Dahl et al., 2008). Meta-analyses of prospective cohort
review suggested that the impact of smoking on dementia might studies suggest an approximately 40–60% increased risk of demen-
vary by age such that smoking was more harmful at younger than tia associated with overweight or obesity; such an association was
older ages (Hernán et al., 2008). Thus, in contrast to the protective stronger in studies with a longer follow-up period (more than 10
effect initially suggested by cross-sectional and case-control studies, years) and younger age (e.g. middle age) of having obesity (Beydoun
population-based prospective studies have provided rather convinc- et al., 2008; Profenno et al., 2010; Anstey et al., 2011). Thus, while
ing evidence that smoking is a risk factor for dementia and AD. midlife overweight or obesity is a risk factor for late-life dementia, a
low BMI and weight loss can be interpreted as markers for preclini-
Alcohol consumption
cal dementia, particularly when measured 6–10 years prior to the
The relationship between alcohol consumption and risk of demen-
clinical diagnosis of dementia.
tia largely depends on amount consumed. Long-term excessive
use of alcohol causes ‘alcoholic dementia’, while frequent alcohol High total cholesterol and therapy with statins
intake may increase risk of VaD (Yoshitake et al., 1995). In the High total serum cholesterol as a major risk factor for atheroscle-
Kungsholmen project, while alcohol abuse was related to prevalent rosis has been investigated in association with AD and dementia
dementia (Fratiglioni et al., 1993), light-to-moderate alcohol con- from a life-course perspective. An association of midlife elevated
sumption was associated with a reduced risk of dementia (Huang cholesterol with an increased risk of late-life AD and dementia
et al., 2002). Several other studies also reported an association is reported in some studies (Notkola et al., 1998; Kivipelto et al.,
of light-to-moderate consumption of wine with a lower risk of 2002; Whitmer et al., 2005b), but not in others (Kalmijn et al., 2000;
dementia, even among very old people (e.g. 75+ years) (Orgogozo Tan et al., 2003; Mielke et al., 2010). Conflicting findings also are
et al., 1997; Elias et al., 1999; Ruitenberg et al., 2002; Mehlig et al., reported when total cholesterol is examined in late life. The French
Three-City study found that hyperlipidaemia was related to an hypertension was associated with dementia among relatively
increased odds ratio of dementia, and nonAlzheimer’s dementia young-old people (Bermejo-Pareja et al., 2010). In very old people
in particular (Dufouil et al., 2005). However, several prospective (e.g. 75+ years), however, the risk effect of high blood pressure on
studies with a short period of follow-up found no association (Li cognitive function is weak and the deleterious effect may be present
et al., 2004; Hayden et al., 2006) or even an inverse association only for very high systolic pressure (e.g. >180 mm Hg) (Guo et al.,
(Kuusisto et al., 1997; Romas et al., 1999; Reitz et al., 2004; Mielke 1999a; Qiu et al., 2003b). By contrast, at this advanced ages, low
et al., 2005) between late-life total cholesterol and risk of dementia blood pressure may predict risk of dementia (Ruitenberg et al.,
and AD. Systematic reviews and meta-analyses of epidemiologi- 2001; Qiu et al., 2003b; Verghese et al., 2003a). As dementia has
cal studies conclude from a life-course perspective that there is an long latent period, it is suggested that low blood pressure can be a
age-dependent association of total cholesterol with risk of demen- sign of impending dementia illness rather than a cause of the dis-
tia, such that high total cholesterol at midlife is more evidently ease. However, several studies with more than 6 years of follow-up
associated with an increased risk of late-life dementia, whereas no confirm such an association (Morris et al., 2001; Verghese et al.,
or an inverse association between total cholesterol and dementia is 2003a; Qiu et al., 2009), suggesting that late-life low blood pressure
often reported in cohort studies of older people (Anstey et al., 2008; may also be involved in the development or clinical expression of
Solomon and Kivipelto, 2009). Long-term observational studies dementia, possibly by affecting cerebral blood perfusion (Qiu et al.,
reveal that total cholesterol begins to decline more than a decade 2005; Ruitenberg et al., 2005). The age-dependent association of
before dementia is clinically manifest (Stewart et al., 2007; Beydoun blood pressure with dementia and cognitive decline concluded in
et al., 2011), indicating that decreasing total cholesterol after midlife an earlier systematic review (Qiu et al., 2005) has been supported in
and low cholesterol in late life may be markers for future develop- recent reviews (Kennelly et al., 2009; Novak and Hajjar, 2010).
ment of dementia and AD (Mielke et al., 2010). Population-based prospective studies often show a reduced risk
Several cross-sectional and case-control studies have reported of dementia associated with use of antihypertensive drugs (Guo
that statin therapy is associated with low prevalence of AD and et al., 1999b; in’t Veld et al., 2001b; Murray et al., 2002; Hajjar et al.,
dementia (Jick et al., 2000; Hajjar et al., 2002; Rockwood et al., 2005; Yasar et al., 2005; Khachaturian et al., 2006; Li et al., 2010);
2002; Rodriguez et al., 2002). However, prospective studies do not the effect may depend on drug types, clinical features of patients,
show any cognitive benefits of statin therapy for patients with AD as well as age and duration of therapy, such that protective effect
or dementia (Heart Protection Study Collaborative Group, 2002; Li is more evident for middle-aged people, for angiotensin receptor
et al., 2004; Zandi et al., 2005; Rea et al., 2005; Feldman et al., 2010; blockers or calcium-channel blockers, for patients with a history of
Sano et al., 2011). Neuropathological studies also are inconsistent cerebrovascular disease, and for long-term treatment (Qiu, 2011).
concerning whether use of statins is associated with less Alzheimer Furthermore, neuropathological and neuroimaging data suggest
pathology and fewer brain infarcts (G. Li et al., 2007a; Arvanitakis that antihypertensive therapy is associated with less Alzheimer
et al., 2008a). Although preclinical research and observational epi- pathologies and slower progression of cerebral white matter lesions
demiological studies have shown apparent promise for statins as (Hoffman et al., 2009; Godin et al., 2011). Despite all these rather
potential preventive and therapeutic agents for AD and dementia, consistent findings from observational studies, large-scale rand-
large-scale clinical trials and Cochrane reviews of clinical trials omized clinical trials have yielded mixed results. These include the
fail to demonstrate that statins given in late life to individuals at following: (1) the Medical Research Council’s trial found that treat-
risk of vascular disease have any therapeutic or preventive effect ing moderate hypertension with diuretics did not influence cogni-
in dementia (McGuinness et al., 2009a, 2010; Shepardson et al., tive function (Prince et al., 1996). (2) The Systolic Hypertension in
2011a, 2011b). the Elderly Programme (SHEP) trial found that active treatment
with thiazide diuretics reduced risk of cardiovascular events, but
Hypertension and antihypertensive therapy not cognitive impairment or dementia (SHEP Research Group,
Hypertension, as the most powerful risk factor for cerebrovascular 1991). However, differential dropout rates between treatment and
disease, can be a risk factor for VaD (Posner et al., 2002; Ninomiya placebo groups might have obscured the ability to assess an effect
et al., 2011; Sharp et al., 2011). Indeed, since 1996 when the lon- of antihypertensive therapy against dementia in the SHEP (Di Bari
gitudinal Gothenburg study suggested that high blood pressure et al., 2001). (3) The Systolic Hypertension in Europe (Syst-Eur)
increased risk of dementia (Skoog et al., 1996), evidence has been trial with a calcium-channel blocker showed that active therapy
accumulating that hypertension or high blood pressure may be a reduced dementia risk by 50% over a 2-year period (Forette et al.,
risk factor even for AD (Launer, 2002; Qiu et al., 2005). The rela- 1998); the extended follow-up data following termination of the
tionship between high blood pressure and risk of dementia and initial trial confirmed the initial finding (Forette et al., 2002).
AD seems to be age-dependent (Petitti et al., 2005; Qiu et al., 2005; (4) The Perindopril Protection against Recurrent Stroke Study
G. Li et al., 2007b; Johnson et al., 2008; Euser et al., 2009; Power (PROGRESS) trial of people with cerebrovascular disease found
et al., 2011). Several population-based studies have provided sub- that blood pressure-lowering regimen did not affect the overall risk
stantial evidence to support the association of midlife high blood of dementia, but did reduce risk of ‘dementia with recurrent stroke’
pressure with an increased risk of late-life dementia (Launer et al., or ‘cognitive decline with recurrent stroke’ (Tzourio et al., 2003).
2000; Kivipelto et al., 2001; Whitmer et al., 2005b; Joas et al., 2012); (5) The Study on COgnition and Prognosis in the Elderly (SCOPE)
such association is independent of APOE genotypes and major car- trial sought to compare the role of candesartan and usual antihy-
diovascular risk factors (Kivipelto et al., 2002). Thus, active con- pertensive drugs in reducing risk of cardiovascular events, cogni-
trol of midlife high blood pressure, especially with regard to high tive decline, and dementia in older patients with mild-to-moderate
systolic pressure, has been proposed to be an effective strategy in hypertension. The trial found no difference in incidence of dementia
reducing risk of late-life dementia (Launer et al., 2010). Untreated and cognitive decline between the two groups (Lithell et al., 2003).
(6) The Hypertension in the Very Elderly Trial–cognitive function may reflect a consequence of complex mechanisms, such as effects
assessment (HYVET-COG) of people 80+ years of age found a non- of hyperglycaemia (Peila et al., 2002; Korf et al., 2006), hyperin-
significant reduction in dementia risk associated with antihyperten- sulinaemia (Luchsinger et al., 2004a), and advanced glycation end
sive therapy; when HYVET-COG data were pooled with data from products (Yaffe et al., 2011), on brain degenerative pathologies or
other trials, antihypertensive therapy marginally reduced dementia effects of the diabetes-related comorbidities, such as hypertension,
risk by 13% (Peters et al., 2008a). (7) Pooled data of the Ongoing obesity, and dyslipidaemia (Sims-Robinson et al., 2010).
Telmisartan Alone and in Combination With Ramipril Global Systemic atherosclerosis
Endpoint Trial (ONTARGET) and the Telmisartan Randomized
Population-based follow-up studies have suggested that markers of
Assessment Study in Angiotensin-Converting-Enzyme-Inhibitor
more severe atherosclerosis, such as carotid atherosclerosis assessed
Intolerant Subjects with Cardiovascular Disease (TRANSCEND)
with B-mode ultrasonography and peripheral arterial atherosclero-
trials on patients with diabetes or cardiovascular disease did not
sis measured with ankle-to-brachial index (ABI), are associated with
prove any cognitive benefits by blocking renin-angiotensin system
an increased risk for AD and dementia (Hofman et al., 1997; Laurin
(Anderson et al., 2011).
et al., 2007; van Oijen et al., 2007). In addition, a greater carotid
The Cochrane review of the SHEP, the Syst-Eur, the SCOPE, and
intimal medial thickness and increased carotid-femoral pulse wave
the HYVET-COG trials found no convincing evidence that late-life
velocity (a marker of central arterial stiffness) are associated with
blood pressure-lowering therapy in hypertensive patients with-
accelerated cognitive decline and poor cognitive performance (Elias
out a history of cerebrovascular disease could prevent dementia
et al., 2009; Wendell et al., 2009; Sander et al., 2010). Furthermore,
(McGuinness et al., 2009b). However, it should be noted that the
the AGES-Reykjavik study found that coronary artery calcification
SCOPE trial, which gave a heavy weighting towards negative results,
(a marker of atherosclerotic burden) was associated with lower cog-
was not a placebo-controlled trial due to ethical concerns. In a recent
nitive performance and an increased likelihood of dementia (Vidal
meta-analysis, overall, antihypertensive therapy did not reduce risk
et al., 2010). Finally, a systematic review supports a low ABI (<0.90)
of dementia, but a reduced risk of dementia was indicated in trials
as a marker for cognitive impairment and dementia (Guerchet
involving diuretics or dihydropyridine calcium-channel blockers,
et al., 2011). These studies suggest that systemic atherosclerosis is
though not in those involving renin system inhibitors (Staessen
associated with an increased risk of dementia.
et al., 2011). The general negative results may be partly explained by
two reasons. First, dementia is considered the secondary end-point Cerebrovascular disease
in most of these trials, in which the trials have to be terminated Cerebral infarcts and recurrent and strategic stroke are the main
when therapeutic benefits are clearly shown for primary end-points risk factors for poststroke dementia (Leys et al., 2005). Systematic
(e.g. cardiovascular disease and stroke). Thus, if the beneficial effect reviews of prospective studies reveal a two- to four-fold increased
of lowering blood pressure on dementia is not as strong as for cardi- risk of dementia associated with clinical stroke (Pendlebury et al.,
ovascular disease and stroke, which is usually the case, a significant 2009; Savva et al., 2010). Numerous studies have consistently
beneficial effect for dementia is unlikely to be proven. Moreover, all shown that clinically silent brain infarcts and cerebral microv-
clinical trials have been conducted among people aged 60+ years, ascular diseases (e.g. lacunar infarcts, white matter lesions, and
when high blood pressure may no longer act as a strong risk fac- microbleeds) detected on brain magnetic resonance imaging
tor for dementia. Thus, it is suggested that therapeutic or preven- (MRI) or indicated by retinal microvascular lesions are associated
tive intervention targeting high blood pressure may be effective in with an increased risk of dementia and cognitive decline (Vermeer
reducing risk of dementia when implemented in middle-aged or et al., 2003; Liebetrau et al., 2004; Prins et al., 2004; Debette and
younger-old people (Qiu, 2011). Markus, 2010; Debette et al., 2010; Pantoni, 2010; Qiu et al., 2010a;
de Jong et al., 2011; Poels et al., 2012). The APOE ε4 allele may
Vascular diseases
have an additive or synergistic interaction with stroke on the risk
Diabetes mellitus
of dementia and cognitive decline (Zhu et al., 2000; Qiu et al.,
An increased risk of dementia among persons with diabetes has
2006a; Jin et al., 2008). However, the role of stroke in AD remains
been reported in numerous prospective studies (Leibson et al.,
debatable, although some reports suggest an association of clini-
1997; Brayne et al., 1998; Ott et al., 1999; Luchsinger et al., 2001;
cal stroke with AD (Honig et al., 2003; Hayden et al., 2004; Reitz
Arvanitakis et al., 2004; Xu et al., 2004; Akomolafe et al., 2006;
et al., 2006).
Ritchie et al., 2010; Ohara et al., 2011). Follow-up studies also sug-
gested that diabetes contributes to progression from mild cognitive Cardiovascular disease
impairment to full dementia (Velayudhan et al., 2010; Xu et al., 2010). Cardiovascular disease is associated with an increased incidence of
Observational studies showed that midlife diabetes as compared dementia in the cohort of Cardiovascular Health Study, with the
to late-life diabetes was more strongly associated with an elevated highest risk being seen in people with peripheral arterial disease
risk of dementia (Alonso et al., 2009b; Xu et al., 2009), suggesting (Newman et al., 2005), suggesting that extensive peripheral athero-
that long-term diabetes plays a crucial role in dementia. Systematic sclerosis is a risk factor for dementia (Hofman et al., 1997; Beeri
reviews and meta-analyses of prospective studies have confirmed et al., 2006). In addition, other heart diseases such as unrecognized
that diabetes increases risk of dementia, and even of Alzheimer’s myocardial infarction, atrial fibrillation, and heart failure may con-
dementia, independent of vascular comorbidities (Biessels et al., tribute to an increased risk of dementia and AD (Ott et al., 1997;
2006; Kopf and Frölich, 2009; Lu et al., 2009; Profenno et al., 2010). Polidori et al., 2006; Qiu et al., 2006b; Lavery et al., 2007; Ikram
In addition, borderline diabetes or impaired glucose tolerance was et al., 2008; Dublin et al., 2011), although findings for myocardial
linked to an increased risk of dementia and AD in very old people infarction and atrial fibrillation have been inconsistent (Bursi et al.,
(Xu et al., 2007). The association of diabetes with AD or dementia 2006; Kwok et al., 2011; Marengoni et al., 2011).
Diets, dietary patterns, and nutrients (Dangour et al., 2010; Ford and Almeida, 2012). It remains unclear
Consumption of unsaturated fat and fish whether prolonged treatment with B-vitamins can reduce the risk
An association of diets rich in saturated fats with an elevated risk of of dementia.
dementia and cognitive decline is suggested by some, but not all, pro- Hyperhomocysteinaemia is a risk factor for cardiovascular dis-
spective studies (Kalmijn et al., 1997; Barberger-Gateau et al., 2002; eases. Thus, increased serum homocysteine may contribute to
Morris et al., 2003a; Laitinen et al., 2006). By contrast, diets rich in dementia through vascular mechanism or neurotoxic effect. An
high polyunsaturated fatty acids (PUFAs) (e.g. fish) may benefit cog- association between high serum homocysteine and an increased
nitive function by maintaining brain vascular health. In addition, risk of dementia is suggested in several cohort studies (Seshadri
a dietary pattern of mixed foods characterized by higher intakes of et al., 2002; Ravaglia et al., 2005b; Schafer et al., 2005), but not in
salad dressing, nuts, fish, tomatoes, poultry, cruciferous vegetables, others (Morris et al., 2003b; Luchsinger et al., 2004b). Meta-analyses
fruits, and green leafy vegetables and a lower intake of high-fat dairy support a positive association between high homocysteine and
products, red meat, organ meat, and butter (i.e. a diet rich in ω-3 dementia, but a causal relationship is uncertain (Ho et al., 2011;
PUFAs, ω-6 PUFAs, vitamin E, and folate, but with low saturated fatty Wald et al., 2011). In addition, high-dose B-vitamins may reduce
acids) was protective against AD and dementia (Barberger-Gateau homocysteine level and slow AD progression (Aisen et al., 2003).
et al., 2007; Gu et al., 2010). One systematic review of follow-up However, a Cochrane review and randomized clinical trials con-
studies found that some studies suggested a reduced dementia risk cluded that supplemental folic acid and vitamin B12 had no benefits
associated with more consumption of fish and ω-3 PUFAs, whereas on cognitive function, although they might be effective in reduc-
others found no independent association of high consumption of ing serum homocysteine (Malouf et al., 2003; Eussen et al., 2006;
fish and PUFAs with dementia risk (Devore et al., 2009; Fotuhi et al., McMahon et al., 2006).
2009; Kröger et al., 2009). Finally, evidence from intervention stud- Clustering of cardiovascular risk factors and related disorders
ies is inconsistent concerning effect of fatty acid supplementation on Cardiovascular risk factors and related disorders often coexist
cognitive decline or dementia (Dangour et al., 2010). among older people, in which they may act additively or syner-
Dietary or supplemental antioxidant vitamins gistically through common pathological pathways (e.g. leading to
Diets rich in fruits, vegetables, antioxidants, and flavonoids benefit atherosclerosis and cerebral hypoperfusion) to affect occurrence
health in general, but it is uncertain whether such a dietary pattern of dementia (Qiu, 2011). Several studies have consistently shown
is helpful for the prevention of dementia (Dai et al., 2006). Some that the risk of dementia increases with increasing number of vas-
follow-up studies have shown a decreased risk of dementia associ- cular or lifestyle factors (Kalmijn et al., 2000; Luchsinger et al.,
ated with use of dietary or supplemental vitamins E and C (Engelhart 2005; Kivipelto et al., 2006; Mitnitski et al., 2006; Qiu et al., 2010b;
et al., 2002; Morris et al., 2002; Zandi et al., 2004), but not in others Gelber et al., 2012). Different risk indices in middle age and late
(Maxwell et al., 2005; Gray et al., 2008). In addition, a higher adher- life have been developed to quantify risk of dementia associated
ence to a ‘Mediterranean diet’ (a dietary pattern with high intake of with clustering of multiple factors including vascular risk factors,
fish, olive oil, fruits, and vegetables rich in antioxidants) has been which could provide reasonable estimation for the probability of
associated with a reduced risk of dementia independent of major dementia occurrence years or even decades later (Kivipelto et al.,
vascular risk factors in the US studies (Scarmeas et al., 2006a, 2006b, 2006; Barnes et al., 2009; Reitz et al., 2010). The risk indices may be
2009), but not in the Three-City study (Féart et al., 2009). Dietary useful for identifying individuals at risk for developing dementia
or supplemental vitamin E intake in either middle age or late life that can be targeted for early intervention.
was not associated with dementia (Luchsinger et al., 2003; Laurin Inflammation
et al., 2004). Furthermore, the Rotterdam study found no associa- Inflammatory markers
tion of plasma levels of vitamins A and E with risk of dementia Inflammation plays a pivotal role in pathogenesis of atherosclerosis,
and cognitive decline (Engelhart et al., 2005). Finally, randomized and neuroinflammation has been implicated in the neurodegenera-
clinical trials with relative short-term use of supplemental vitamin tive cascade that leads to Alzheimer pathologies and clinical demen-
E failed to show any effect against cognitive impairment, although tia (Gorelick, 2010). Elevated serum C-reactive protein (CRP) in
long-term use (e.g. ≥ 15 years) could be neuroprotective (Petersen midlife was associated with an increased risk of both AD and VaD,
et al., 2005; Kang et al., 2006). suggesting that inflammatory markers may be involved in demen-
B-Vitamins and serum homocysteine tia by indicating both peripheral and cerebral vascular mecha-
Experimental research suggests that B-vitamins (i.e. folic acid, B12, nisms, and that inflammatory responses are measurable rather a
and B6) are implicated as protective factors against cognitive deteri- long time before dementia is clinically manifested (Schmidt et al.,
oration and dementia (Morris et al., 2006). However, epidemiologic 2002). Population-based cohort studies also found an association
evidence supporting the association of B-vitamins with dementia of serum inflammatory markers (e.g. CRP and IL-6) measured in
is mixed (Luchsinger and Mayeux, 2004; Fratiglioni et al., 2010). late life (e.g. ≥60 years) with an increased incidence of dementia
Some follow-up studies have suggested an association of low serum (van Exel et al., 2003; Engelhart et al., 2004). Meta-analysis and
vitamin B12 and folate with a high risk of dementia and AD (Wang systematic review strengthen the evidence that AD and dementia
et al., 2001; Maxwell et al., 2002). However, the beneficial effect are accompanied by an inflammatory response, as indicated by
of supplemental B-vitamins against dementia is not supported by increased peripheral concentrations of inflammatory markers (e.g.
follow-up studies (Luchsinger et al., 2007; Nelson et al., 2009). CRP, IL-6, TNF-α, IL-1β, and TGF-β) (Kuo et al., 2005; Swardfager
Furthermore, meta-analyses of randomized clinical trials conclude et al., 2010). Of these serum inflammatory markers, CRP seems to
that supplementation of B-vitamins does not appear to benefit cog- be the most promising one to identify individuals being at risk for
nitive function in individuals with or without cognitive impairment dementia.
Nonsteroidal anti-inflammatory drugs (NSAIDs) Cognitive or mentally stimulating activity

Epidemiological and experimental studies support the hypoth- In 2004, a systematic review found that seven observational studies
esis that use of anti-inflammatory agents such as NSAIDs could reported a lower risk of dementia in subjects with greater engage-
protect against Alzheimer’s pathological process (McGeer and ments in mentally stimulating activities (Fratiglioni et al., 2004).
McGeer, 2007). Indeed, use of NSAIDs has been associated with A subsequent systematic review concluded that complex patterns
a lower risk of AD and dementia in numerous, although not all, of mental activity were associated with a 46% reduction in risk of
observational prospective studies (Stewart et al., 1997; in’t Veld dementia (Valenzuela and Sachdev, 2006). However, due to the
et al., 2001a; Landi et al., 2003; Cornelius et al., 2004; Ancelin long preclinical phase of dementia (Bäckman et al., 2001), reverse
et al., 2011). Systematic review and meta-analysis of prospective causality cannot be ruled out. To eliminate possible reverse causal-
studies confirm that use of NSAIDs for over 2 years has a greater ity, some studies have controlled for cognitive performance at the
benefit against AD and dementia (Etminan et al., 2003; Szekely time when activities are assessed or activities are assessed at least
et al., 2004), suggesting that long-term treatment with NSAIDs 3–5 years before dementia diagnosis (Wang et al., 2002; Verghese
may protect against dementia. However, the beneficial effects of et al., 2003b). Furthermore, data from the Swedish Twin Registry
NSAIDs against AD and dementia may result from various biases indicate that participation in a greater number of leisure activities
such as recall, prescription, and publication biases (de Craen et al., during early life and middle age, especially intellectual-cultural
2005). In addition, neuropathological studies found no evidence activities, is associated with a lower risk of dementia (Crowe et al.,
for an association between use of NSAIDs and less Alzheimer’s 2003). The Three-City study found that late-life cognitively stimulat-
pathologies (Arvanitakis et al., 2008b); instead, use of NSAIDs in ing activities, but not other leisure activities (i.e. social and physical
middle age or late life was associated with more neuritic plaques activities), were associated with a reduced risk of dementia inde-
(Sonnen et al., 2010). Furthermore, the Alzheimer’s Disease pendent of vascular risk factors, depressive symptoms, and func-
Anti-inflammation Prevention Trial (ADAPT) among people aged tional status (Akbaraly et al., 2009). A composite score of cognitive
70+ years with a family history of dementia failed to show any lifestyle that consisted of education, occupational complexity, and
beneficial effect of anti-inflammatory therapy with celecoxib or social engagement was protective against dementia, but not sur-
naproxen against AD and cognition decline; instead, a potentially vival after diagnosis of dementia (Valenzuela et al., 2012). Different
increased risk of AD related to the therapy was indicated (ADAPT types of activities have been examined, such as travelling, knitting,
Group, 2006; Lyketsos et al., 2007; Martin et al., 2008). A new gardening, dancing, playing games and musical instruments, read-
hypothesis has been generated based on the extended results of the ing, cultural activities, and watching specific television programmes
ADAPT that effect of anti-inflammatory therapy may differ at var- (Fabrigoule et al., 1995; Crowe et al., 2003; Verghese et al., 2003b).
ious pathological stages of AD, such that use of NSAIDs confers an Due to individual and cultural differences in choosing specific activ-
adverse effect in later pathological stages, whereas asymptomatic ities, some researchers summarize various mental activities into a
individuals treated with NSAIDs have reduced risk of AD after composite score. For example, a composite cognitive score of lei-
2–3 years (Breitner et al., 2011). In addition, it has been argued sure activities that involve participation in seven common activities
that, for purpose of prevention, a relatively long time period of with information processing as a central component was associated
intervention and observation may be essential to demonstrate any with a reduced risk of dementia after controlling for demograph-
beneficial effect (Meinert and Breitner, 2008). ics, APOE ε4 allele, medical conditions, and depressive symptoms
(Wilson et al., 2002b, 2002c). In the Kungsholmen project, a 4-grade
Psychosocial factors score to characterize mental, social, and physical components of
Social networks and social engagements each activity was developed by both researchers and study partici-
Evidence from prospective studies suggests that poor social net- pants; a high score in two or all of the three components was associ-
work or social disengagement is associated with cognitive decline ated with a reduced risk of subsequent dementia (Karp et al., 2006).
and dementia (Bassuk et al., 1999; Fratiglioni et al., 2000b, 2004). Clearly, different leisure activities require use of different cognitive
The risk for dementia was also increased in people with increas- components, in which the cognitive load of different activities is
ing social isolation and with less frequent or unsatisfactory con- critical for gaining cognitive benefits. Finally, a systematic review of
tacts with relatives or friends (Fratiglioni et al., 2000b), suggesting randomized clinical trials found that cognitive exercise training in
that the quality of social interactions is critical for protective effect healthy older people could produce persistent beneficial effects on
against dementia (Amieva et al., 2010). Low social support at work neuropsychological performance, although it has yet to be proven to
has been associated with an increased risk of dementia, and VaD prevent dementia (Valenzuela and Sachdev, 2009).
in particular (Andel et al., 2012). In addition, late-life low social
engagement and decreased social engagement from middle age to Physical activity
late life were associated with a double risk of dementia (Saczynski A systematic review of prospective studies found that physical
et al., 2006). Finally, being widowed from midlife onwards was activity was associated with a lower risk of dementia in six out of
related to a substantially increased risk of dementia, suggesting that nine studies (Fratiglioni et al., 2004). Several additional follow-up
living with a partner may have a potential protective effect against studies have linked increasing physical exercise to a reduced risk of
late-life dementia (Håkansson et al., 2009). More extensive social cognitive impairment (Etgen et al., 2010), dementia (Larson et al.,
networks imply better social supports leading to better access to 2006; Taaffe et al., 2008), and VaD in particular (Ravaglia et al.,
resources and material goods. Rich social networks may also pro- 2008). In the Kungsholmen project, the physical activity compo-
vide affective and intellectual stimulation that could influence nent (a summary measure of various leisure activities, rather than
health outcomes through social, behavioural, psychological, and any specific physical exercise or sport) was related to a decreased
physiological pathways (Seeman and Crimmins, 2001). risk of dementia (Karp et al., 2006). In addition, low-intensity
activity such as walking may reduce risk of dementia and cognitive et al., 2002; Kang et al., 2004). The Women’s Health Initiative
decline (Abbott et al., 2004; Weuve et al., 2004). A meta-analysis of Memory Study (WHI-MS) found that oestrogen therapy alone
prospective studies yielded a 28% decreased risk of dementia and or in combination with progestin was associated with a two-fold
45% decreased risk of AD for the highest as compared with the low- increased risk for dementia (Espeland et al., 2004; Shumaker et al.,
est level of leisure-time physical activity (Hamer and Chida, 2009). 2004). Furthermore, postmenopausal oestrogen therapy may inter-
In support of this finding, increasing physical activity or a greater act with smoking to substantially increase the risk of AD (Roberts
amount of walking has been associated with larger volumes of brain et al., 2006). It has been argued that, in the WHI-MS, oestrogen
tissue and grey matter (Rovio et al., 2010). A strong protective effect therapy was given 10–15 years after menopause, when the ‘window
of regular physical activities at middle age against late-life dementia of critical time’ for putative benefits of oestrogen therapy on cogni-
was reported (Andel et al., 2008; Chang et al., 2010). The protective tion might have been missed; it may be effective in reducing risk of
effect of midlife physical activities against dementia was particu- dementia when oestrogen therapy is given at a younger age close
larly strong for carriers of the APOE ε4 allele (Rovio et al., 2005), to menopausal time. In support of this hypothesis, a long-term
whereas the protective effect of late-life physical activity was present follow-up study suggests that hormone therapy in midlife protects
only among noncarriers of the ε4 allele (Podewils et al., 2005). The against late-life dementia, whereas hormone therapy initiated in
possible benefits of physical training on cognition also have been late life (postmenopause) has deleterious effect (Whitmer et al.,
evaluated in several small randomized controlled trials. Although 2011). Finally, the therapeutic effect of hormone replacement on
evidence supports the beneficial effects of aerobic physical activi- patients with dementia is not supported by the Cochrane review
ties on cognitive function in healthy older adults (Angevaren et al., (Hogervorst et al., 2009). Thus, many questions on the relation-
2008), the effects of short-term training programmes are equivocal ship between hormonal therapy and risk of dementia remain to
(Churchill et al., 2002). As it takes years to achieve good physical be answered. In particular, it is unclear whether perimenopau-
fitness, brief periods of physical exercise may not have substantial sal initiation of therapy or other dosages or forms of oestrogen
benefits on global cognition, but could still be effective in subsets of (e.g. oestradiol) would have a neuroprotective effect (Craig et al.,
cognitive domains that are more sensitive to the age-related decre- 2005).
ments. This is partly supported by meta-analysis and randomized
Traumatic brain injury (TBI)
controlled intervention studies (Kramer et al., 1999; Colcombe and
TBI has been extensively examined as a possible risk factor for
Kramer, 2003).
AD. Although a link between TBI and AD is biologically plausi-
Depression or depressive symptoms ble (Van Den Heuvel et al., 2007), findings from epidemiological
Several follow-up studies have reported an elevated risk of demen- studies are mixed (Starkstein and Jorge, 2005). In the early 1990s,
tia associated with a history of depression (Wilson et al., 2002a; a meta-analysis of seven case-control studies reported an increased
Green et al., 2003; Modrego and Ferrandez, 2004; Andersen et al., AD risk associated with head injury with consciousness loss in men,
2005; Dal Forno et al., 2005; Dotson et al., 2010; Ritchie et al., 2010; not in women (Mortimer et al., 1991), which was replicated by an
Lenoir et al., 2011). Depression was also associated with a substan- updated meta-analysis more than 10 years later (Fleminger et al.,
tively increased risk for dementia in patients with diabetes (Katon 2003). Some longitudinal studies found no association of head
et al., 2012). Self-reported midlife stress was associated with an trauma with risk of AD (Launer et al., 1999; Mehta et al., 1999),
increased risk of developing dementia a few decades later in women whereas others found an increased risk associated with severe TBI
(Johansson et al., 2010). The meta-analysis of both case-control (Schofield et al., 1997; Plassman et al., 2000). In addition, a syner-
and cohort studies reported a double-increased risk for dementia gistic effect of head injury with APOE ε4 allele on AD risk is sug-
associated with depression (Ownby et al., 2006), where increased gested in some studies (Mayeux et al., 1995), but not confirmed
interval between diagnoses of depression and dementia was asso- by others (Mehta et al., 1999; Guo et al., 2000). TBI may increase
ciated with an increased risk of subsequent dementia, supporting formation of β-amyloid plaques (Nicoll et al., 1995; Emmerling et
the notion that, rather than a prodrome, depression may be a risk al., 2000; Johnson et al., 2010) or reduce brain reserve (Katzman,
factor for dementia. Although recent long-term follow-up studies 2004). Despite biological plausibility, the possible association of
also support a temporal relationship of depression to subsequent TBI with AD needs further investigation, along with their possible
dementia (Saczynski et al., 2010; Li et al., 2011), it remains debat- interaction with APOE genotype.
able regarding whether late-life depression is a preclinical symp-
Frailty and self-rated health condition
tom or a causal risk factor for dementia (Jorm, 2001; Ganguli et al.,
2006; Li et al., 2011). A few studies have explored the association between systemic health
condition and future risk of dementia. A frailty index consisting of
Miscellaneous multiple nontraditional risk factors (e.g. general health condition,
Hormone replacement therapy eye disease, trouble with kidney or stomach, fracture, and polyp-
Numerous observational studies have linked postmenopausal oes- harmacy) has been strongly associated with an increased risk of
trogen therapy to a lower risk of dementia, in which some studies mild cognitive impairment and dementia (Monastero et al., 2007;
even show a dose-response relation of greater protection for longer Song et al., 2011). Anaemia or low haemoglobin was associated
therapy (Tang et al., 1996; Kawas et al., 1997; Waring et al., 1999; with an increased risk of dementia and AD (Atti et al., 2006; Peters
Zandi et al., 2002; Ryan et al., 2009). The biological mechanisms et al., 2008b; Shah et al., 2011). Large-scale population studies also
for possible cognitive benefits include vascular, cholinergic, and reported that self-perception of poor health was associated with an
antioxidant processes. However, other studies and reviews, espe- increased risk of dementia (Yip et al., 2006; Montlahuc et al., 2011).
cially major clinical trials, failed to confirm any protective effect of These studies imply that the general health condition as indicated
oestrogen therapy against dementia (Seshadri et al., 2001; Nelson by a broad set of nontraditional risk factors and geriatric disorders
may anticipate pathophysiologic evolution of dementia among interventions targeting modifiable vascular and psychosocial path-
older people. ways are likely to achieve the goal of primary prevention by pre-
Gene–environment interaction venting or postponing the onset of dementia.
APOE ε4 allele, as the only established genetic factor, may interact Biological plausibility
with various risk factors to affect risk, age of onset, and progression
of dementia (Lahiri et al., 2004; Mielke et al., 2011). The Rotterdam Population-based neuroimaging and neuropathological studies
study showed an interaction of APOE ε4 allele and atherosclerosis have significantly contributed to better understanding of patho-
in AD (Hofman et al., 1997). APOE ε4 allele in combination with logical mechanisms linking major modifiable risk and protective
stroke also substantially increases the risk of sporadic and familial factors, especially with regard to vascular and psychosocial factors,
AD and dementia (Zhu et al., 2000; Rippon et al., 2006). In the to dementia including AD. Increasing evidence from epidemiologi-
Kungsholmen project, APOE ε4 allele interacted with high systolic cal studies supports the hypotheses that cardiovascular risk factors
pressure or low diastolic pressure to greatly increase risk of AD and and related disorders can be linked to dementia by causing cerebral
dementia, whereas use of antihypertensive drugs possibly coun- microvascular damage, possibly by affecting neurodegenerative
teracted the joint effect of the ε4 allele and high systolic pressure process, and by promoting clinical expression of dementia, whereas
on dementia (Qiu et al., 2003b). APOE ε4 allele may also modify psychosocial factors (e.g. high education, mental activity, and social
the relation of alcohol consumption to risk of dementia, such that engagement) may help maintain cognitive function in late life by
midlife heavy alcohol intake was associated with an increased risk providing cognitive reserve.
of dementia only among the ε4 allele carriers (Anttila et al., 2004). Cerebrovascular and neurodegenerative pathologies
Finally, APOE ε4 allele may modify the association of depression Cerebral microvascular diseases (e.g. white matter lesions and
with dementia, such that individuals with both depressive symp- microbleeds) are known to be caused by cardiovascular risk factors
toms and the ε4 allele have a markedly increased risk of dementia such as smoking, hypertension, and diabetes (Vermeer et al., 2007;
(Irie et al., 2008). Although possible interactions of genetic predis- Knopman and Roberts, 2010; Hajjar et al., 2011). Asymptomatic
position with nongenetic factors in modifying risk or age of onset microvascular lesions in the brain have been associated with
of AD and dementia have been suggested, potential biological dementia (Debette et al., 2010; Pantoni, 2010). Population-based
mechanisms for the interaction remain unclear. neuropathological studies support the association of cerebral
atherosclerosis with an increased burden of Alzheimer’s patholo-
Primary Prevention gies; brain infarctions and severe intracranial atherosclerosis
increase the odds of dementia independent of Alzheimer’s patholo-
Epidemiologic evidence gies (Esiri et al., 1999; Schneider et al., 2007b; Dolan et al., 2010).
Following the initiative of the Swedish Council on Technology Atherosclerosis and AD may share common mechanisms such as
Assessment in Health Care (SBU Report, 2006), specific criteria to oxidative stress, inflammation, and toxic β-amyloid. Severe intrac-
summarize epidemiologic evidence concerning risk and protective ranial atherosclerosis or arteriosclerosis could also induce cerebral
factors for dementia were proposed and implemented. Similar to hypoperfusion and trigger accelerated deposition of amyloid-β,
criteria adopted for other chronic diseases, these criteria integrate which in turn contribute to cognitive deterioration and expression
the internal validity with basic causal criteria to assess the qual- of dementia (Snowdon et al., 1997; Esiri et al., 1999; Garcia-Alloza
ity of individual studies reviewed. A 4-grade scale score of quality et al., 2011). Finally, cerebrovascular and Alzheimer’s patholo-
of evidence is developed by integrating the number and propor- gies often coexist in patients with dementia, suggesting that these
tion of studies reporting a specific association with a quality index. lesions may be coinciding processes converging to cause additive
Scientific evidence is considered strong when several articles with brain damage, and thus to promote clinical manifestation of the
a high quality index have consistently reported the same finding; dementia syndrome (Iadecola and Gorelick, 2003; Casserly and
moderately strong evidence includes also high quality reports, but Topol, 2004; Strozyk et al., 2010).
the finding is supported by a limited number of studies, or the qual- Post-mortem imaging studies have directly linked smoking
ity of studies is moderate, but is compensated by several positive and hypertension to neuritic plaques, neurofibrillary tangles, and
reports and is supported by systematic review or meta-analysis; atrophy in the medial temporal lobe (MTL) (Sparks et al., 1995;
limited evidence is defined by a limited number of medium qual- Petrovitch et al., 2000; Roher et al., 2003; Tyas et al., 2003; Korf
ity studies; and insufficient evidence consists of reports with some et al., 2004; Debette et al., 2011). Overweight or obesity and psy-
methodological uncertainty independent of number of published chological stress throughout adult life may contribute to late-life
studies. Based on these criteria, Table 31.2 summarizes the current MTL atrophy and more white matter lesions in women (Gustafson
epidemiologic evidence that is assessed from a life-course perspec- et al., 2004; Johansson et al., 2012). Follow-up data also show that
tive, in which individual putative risk or protective factors are listed patients with diabetes have accelerated progression in brain atrophy,
according to the underlying aetiological hypotheses. Overall, strong which has significant cognitive consequences (Biessels et al., 2006;
evidence from randomized, placebo-controlled trials that supports van Elderen et al., 2010). Impaired cardiac function and an increas-
primary prevention of dementia, especially with regard to AD, is ing burden of cardiovascular risk factors were associated with glo-
generally lacking. However, it is fair to conclude that moderately bal and regional atrophy even among older people (Jefferson et al.,
strong evidence, mostly from prospective observational studies of 2010; Qiu et al., 2012). Finally, global and regional brain atrophy
the general population, supports the hypotheses that vascular and prior to stroke was associated with poststroke dementia, suggesting
psychosocial factors throughout the lifespan are involved in the that occurrence of dementia following a stroke is likely due to con-
development and expression of dementia. Thus, implementing comitant neurodegenerative and cerebrovascular pathologies that
Table 31.2 Epidemiologic evidence supporting risk and protective (in italic) factors for late-onset dementia by aetiological hypothesis
Aetiological hypotheses Epidemiologic evidence
Insufficient Limited Moderate or strong*

Genetic ACE, IDE, and FTO genes, etc. APOE ε4 allele, familial aggregation
Vascular Midlife high cholesterol; Late-life very high andlow blood Smoking, diabetes, midlife hypertension, obesity,
Dietary factors (e.g. fish and pressure, cardiovascular disease stroke, atherosclerosis, cerebral microvascular
vegetables) disease;
Antihypertensive drugs, limited alcohol intake
Psychosocial Low socioeconomic status Physical activity, rich social network, social Depression or depressive symptoms;
engagement High education, mental activity, physical activity,
social engagement
Inflammatory Inflammatory markers Nonsteroidal anti-inflammatory drugs, hormone
replacement therapy
Others: oxidative stress and Head trauma, deficiency in B-vitamins
neurotoxic and antioxidants (vitamins A, E, and C);
exposure to neurotoxic agents
ACE, angiotensin converting enzyme gene; FTO, fat mass and obesity associated gene; IDE, insulin-degrading enzyme gene.
* Strong evidence exists only for APOE ε4 allele as the aetiological factor for Alzheimer’s disease.
exist prior to stroke onset (Pendlebury and Rothwell, 2009; Savva Intervention strategies
et al., 2010).
Identifying modifiable risk and protective factors for demen-
Brain reserve and cognitive reserve tia provides great potential for primary prevention of the disease
Disparity between clinical phenotype of cognitive ageing and the (Fratiglioni et al., 2008; Middleton and Yaffe, 2009; Barnes and
burden of brain pathologies has been often reported (Katzman, Yaffe, 2011). As summarized in Tables 31.1 and 31.2, current evi-
1993; Stern, 2006). The reserve hypothesis, in which the brain dence from epidemiologic research seems to support the notion
can buffer the impact of pathologies on its functional perform- that, from a life-course perspective, preventive strategies aiming at
ance, has been increasingly employed to explain such a dispar- postponing the onset of dementia are likely to be effective if imple-
ity (Tucker and Stern, 2011). Terminologically, ‘brain reserve’ mented in middle-aged and older people living in the community
(passive) is referred to structural resources, whereas ‘cognitive (Fratiglioni and Qiu, 2011). It was estimated that delaying the onset
reserve’ (active) represents the increase in brain functional capac- of dementia by 5 years would halve the prevalence of dementia and
ity (Katzman, 1993; Stern, 2002; Wilson et al., 2003). The concept substantially decrease the number of dementia cases in the com-
of cognitive reserve is proposed to explain why people with high munity; delaying the onset of dementia by even 2 years also would
educational attainments or more frequent participation in cog- have significant economic and societal benefits (Brookmeyer et al.,
nitive activities express no or less severe cognitive symptoms in 1998; Brodaty et al., 2011). At the moment, two intervention strate-
the presence of similar brain pathologies, that is, individuals with gies can be considered for possible primary prevention.
higher cognitive reserve need more pathology in the brain than
those with lower reserve to express dementia. This hypothesis Optimal control of cardiovascular risk factors from a
is supported by several studies, in which the effect of cognitive life-course perspective
reserve on the relationship between brain pathology and cogni- There appears to have the biological plausibility for cardiovascu-
tive function was investigated in vivo using neuroimaging mark- lar risk factors and disorders, such as smoking, hypertension, dia-
ers (Kemppainen et al., 2008; Roe et al., 2008; Solé-Padullés et al., betes, obesity, cardiovascular disease, and cerebral microvascular
2009; Brayne et al., 2010; Reed et al., 2010; Murray et al., 2011; diseases, being involved in the pathogenesis and clinical expression
Vemuri et al., 2011). Similarly, neuropathological data also have of the dementia syndrome. Although the mechanisms are not fully
shown that education and social networks could modify the asso- understood, successful intervention within the vascular pathway
ciation of Alzheimer’s pathologies to cognitive function such that seems possible because most vascular risk factors and disorders are
cognitive function remains higher in subjects with a heavier bur- modifiable or amenable to prevention and treatment. For primary
den of global neuropathologies if they also have high education prevention, control of smoking, midlife high blood pressure, high
or larger social networks (Bennet et al., 2006; Roe et al., 2007). cholesterol, midlife obesity, and diabetes are the major intervention
These studies illustrate how measures of cognitive reserve (e.g. measures. For instance, intervention studies have demonstrated
education, occupational attainment, and social networks) could that diabetes, hypertension, and obesity can be prevented by modi-
counteract or compensate the deleterious effects of cerebrovas- fying lifestyles (e.g. dietary habits and physical exercise) (Knowler
cular and Alzheimer’s pathologies on cognitive performance. The et al., 2002; Lindström et al., 2006). Thus, intervention measures
reserve hypothesis has enormous implications for public health targeting these factors from middle age are likely to reduce the risk
and for dementia prevention in particular. of dementia. Furthermore, to postpone clinical expression of the
dementia syndrome in late life, preventing recurrent cerebrovas- ADAPT Research Group (2006). Cardiovascular and cerebrovascular events
cular disease, and maintaining sufficient cerebral perfusion by ade- in the randomized, controlled Alzheimer’s Disease Anti-inflammatory
quately managing heart failure and very low blood pressure in very Prevention Trial (ADAPT). PLoS Clinical Trials, 1, e33.
Aevarsson, O., Svanborg, A., and Skoog, I. (1998). Seven-year survival rate
old people seem to be critical.
after age 85 years: relation to Alzheimer disease and vascular dementia.
Adoption of mentally stimulating and socially integrated Archives of Neurology, 55, 1226–32.
lifestyles Aggarwal, N.T., et al. (2006). The relation of cigarette smoking to incident
Extensive social networks, active social engagement, and regular par- Alzheimer’s disease in a biracial urban community population.
Neuroepidemiology, 26, 140–6.
ticipation in intellectually stimulating activities significantly lower
Agüero-Torres, H., et al. (1998). Dementia is the major cause of functional
the risk of dementia. The protective effect may be due primarily to dependence in the elderly: 3-year follow-up data from a population-based
increased capacity of cognitive reserve, although other mechanisms study. American Journal of Public Health, 88, 1452–6.
such as reduced psychosocial stress may also be involved (Fratiglioni Agüero-Torres, H., et al. (1999). Mortality from dementia in advanced age:
et al., 2004). Physical exercise may protect the brain from vascular a 5-year follow-up study of incident dementia cases. Journal of Clinical
damage, but the relevance of physical activity itself remains in debate Epidemiology, 52, 737–43.
because most physical activities include also social and mental com- Agüero-Torres, H., et al. (2001). Institutionalization in the elderly: the role
ponents. Complexity or diversity of leisure activities with all physi- of chronic diseases and dementia: cross-sectional and longitudinal data
cal, mental, and social components seems to have the most beneficial from a population-based study. Journal of Clinical Epidemiology, 54,
795–801.
effect on cognitive function (Karp et al., 2006). Mentally stimulat-
Aisen, P.S., et al. (2003). A pilot study of vitamins to lower plasma
ing and socially integrated lifestyles are likely to postpone the onset homocysteine levels in Alzheimer disease. American Journal of Geriatric
of dementia, because epidemiologic evidence tends to support a Psychiatry, 11, 246–9.
causal link between mental activity and a reduced risk of subsequent Akbaraly, T.N., et al. (2009). Leisure activities and the risk of dementia in the
dementia (Valenzuela, 2008). Furthermore, neuroimaging and path- elderly: results from the Three-City Study. Neurology, 73, 854–61.
ological data support the biological plausibility that lifespan complex Akomolafe, A., et al. (2006). Diabetes mellitus and risk of developing
cognitive activity may protect against dementia through multiple Alzheimer disease: results from the Framingham Study. Archives of
pathways, including cognitive reserve, as well as protection from Neurology, 63, 1551–5.
cerebrovascular disease (Valenzuela et al., 2011). Thus, delaying the Alonso, A., et al. (2009a). Cardiovascular risk factors and dementia mortality:
onset of dementia via community-based intervention programmes 40 years of follow-up in the Seven Countries Study. Journal of Neurological
Science, 280, 79–83.
of adopting cognitive lifestyle and engaging mental activity has been
Alonso, A., et al. (2009b). Risk of dementia hospitalisation associated with
emerging as a practical preventive strategy for future test in rand- cardiovascular risk factors in midlife and older age: the Atherosclerosis
omized trials or intervention studies (Valenzuela, 2008). Risk in Communities (ARIC) study. Journal of Neurology, Neurosurgery
and Psychiatry, 80, 1194–201.
Conclusion Alzheimer’s Association, Thies, W., and Bleiler, L. (2011). 2011 Alzheimer’s
disease facts and figures. Alzheimer’s and Dementia, 7, 208–44.
In summary, dementia has become one of the major causes of func- Alzheimer’s Disease International (2010). World Alzheimer report 2010: global
tional dependence, poor quality of life, institutionalization, and mor- economic impact of dementia. <www.alz.co.uk/research/world-report>
tality among older people. As the population ages across the world, (accessed 10.02.2012).
dementia is predicted to reach an epidemic level in a few decades, American Psychiatric Association (APA) (1987). Diagnostic and statistical
a scenario that poses a serious threat not only to public health but manual of mental disorders, 3rd edition, revised (DSM-III-R), pp. 97–163.
also to the social and economic development of our modern society. APA, Washington, DC.
There is accumulating evidence to suggest that cardiovascular risk Amieva, H., et al. (2010). What aspects of social network are protective
for dementia? Not the quantity but the quality of social interactions is
factors and related vascular disorders significantly contribute to the
protective up to 15 years later. Psychosomatic Medicine, 72, 905–11.
development and expression of cognitive ageing and dementia. Thus, Ancelin, M.L., et al. (2012). Steroid and nonsteroidal anti-inflammatory
adequate management of vascular risk factors and related diseases drugs, cognitive decline, and dementia. Neurobiology of Aging, 33,
may be among the most promising preventive measures against 2082–90.
dementia and cognitive decline. In addition, a more extensive social Andel, R., et al. (2008). Physical exercise at midlife and risk of dementia three
network, social engagement, physical activity, and mentally stimu- decades later: a population-based study of Swedish twins. Journals of
lating activity are shown to be critical for preserving late-life cog- Gerontology Series A Biological Sciences and Medical Sciences, 63, 62–6.
nitive function, and thus for delaying the onset of dementia. Taken Andel, R., et al. (2010). Work-related exposure to extremely low-frequency
together, maintaining vascular health and adopting a healthy cogni- magnetic fields and dementia: results from the population-based study
of dementia in Swedish twins. Journals of Gerontology Series A Biological
tive lifestyle from a life-course perspective can be the most promising
Sciences and Medical Sciences, 65, 1220–7.
strategy to achieve late-life cognitive health. Implementation of this Andel, R., et al. (2012). Work-related stress may increase the risk of vascular
intervention strategy may help maintain cognitive ability in late life, dementia. Journal of the American Geriatrics Society, 60, 60–7.
despite a considerable burden of pathology within the brain itself. Andersen, K., et al. (1999). Gender differences in the incidence of AD and
vascular dementia: the EURODEM Studies. Neurology, 53, 1992–7.
References Andersen, K., et al. (2005). Depression and the risk of Alzheimer disease.
Abbott, R.D., et al. (2004). Walking and dementia in physically capable elderly Epidemiology, 16, 233–8.
men. Journal of the American Medical Association, 292, 1447–53. Anderson, C., et al. (2011). Renin-angiotensin system blockade and cognitive
Access Economics (2006). Report for Asia Pacific Members of Alzheimer’s function in patients at high risk of cardiovascular disease: analysis of data
Disease International. Dementia in the Asia Pacific region: the epidemic from the ONTARGET and TRANSCEND studies. Lancet Neurology, 10,
is here. ADI, London. 43–53.
Angevaren, M., et al. (2008). Physical activity and enhanced fitness to improve Biessels, G.J., et al. (2006). Risk of dementia in diabetes mellitus: a systematic
cognitive function in older people without known cognitive impairment. review. Lancet Neurology, 5, 64–74.
Cochrane Database of Systematic Reviews, 3, CD005381. Blennow, K., de Leon, M.J., and Zetterberg H (2006). Alzheimer’s disease.
Anstey, K.J., et al. (2010). Estimates of probable dementia prevalence from Lancet, 368, 387–403.
population-based surveys compared with dementia prevalence estimates Bonaiuto, S., et al. (1995). Education and occupation as risk factors for
based on meta-analyses. BMC Neurology, 10, 62. dementia: a population-based case-control study. Neuroepidemiology,
Anstey, K.J., et al. (2011). Body mass index in midlife and late-life as a risk 14, 101–9.
factor for dementia: a meta-analysis of prospective studies. Obesity B örjesson-Hanson, A., Gustafson, D., and Skoog, I. (2007). Five-year
Reviews, 12, e426–37. mortality in relation to dementia and cognitive function in 95-year-olds.
Anstey, K.J., et al. (2007). Smoking as a risk factor for dementia and cognitive Neurology, 69, 2069–75.
decline: a meta-analysis of prospective studies. American Journal of Brayne, C., et al. (1998). Vascular risks and incident dementia: results from a
Epidemiology, 166, 367–78. cohort study of the very old. Dementia and Geriatric Cognitive Disorders,
Anstey, K.J., Lipnicki, D.M., and Low, L.F. (2008). Cholesterol as a risk factor 9, 175–80.
for dementia and cognitive decline: a systematic review of prospective Brayne, C., et al. (2010). Education, the brain and dementia: neuroprotection
studies with meta-analysis. American Journal of Geriatric Psychiatry, 16, or compensation? Brain, 133, 2210–16.
343–54. Breitner, J.C., et al. (2011). Extended results of the Alzheimer’s disease
Anttila, T., et al. (2004). Alcohol drinking in middle age and subsequent risk anti-inflammatory prevention trial. Alzheimer’s and Dementia, 7,
of mild cognitive impairment and dementia in old age: a prospective 402–11.
population based study. British Medical Journal, 329, 539. Brodaty, H., et al. (2011). The world of dementia beyond 2020. Journal of the
Arvanitakis, Z., (2004). Diabetes mellitus and risk of Alzheimer disease and American Geriatrics Society, 59, 923–7.
decline in cognitive function. Archives of Neurology, 61, 661–6. Brookmeyer, R., Gray, S., and Kawas, C. (1998). Projections of Alzheimer’s
Arvanitakis, Z., et al. (2008a). Statins, incident Alzheimer disease, change in disease in the United States and the public health impact of delaying
cognitive function, and neuropathology. Neurology, 70, 1795–802. disease onset. American Journal of Public Health, 88, 1337–42.
Arvanitakis, Z., et al. (2008b). Relation of NSAIDs to incident AD, change in Buchman, A.S., et al. (2005). Change in body mass index and risk of incident
cognitive function, and AD pathology. Neurology, 70, 2219–25. Alzheimer disease. Neurology, 65, 892–7.
Atti, A.R., et al (2006). Anaemia increases the risk of dementia in cognitively Bursi, F., et al. (2006). Heart disease and dementia: a population-based study.
intact elderly. Neurobiology of Aging, 27, 278–84. American Journal of Epidemiology, 163, 135–41.
Bachman, D.L., et al. (1993). Incidence of dementia and probable Alzheimer’s Caamaño-Isorna, F., et al. (2006). Education and dementia: a meta-analytic
disease in a general population: the Framingham Study. Neurology, 43, study. Neuroepidemiology, 26, 226–32.
515–9. Canadian Study of Health and Aging Working Group (2000). The incidence
Bäckman, L., Small, B.J., and Fratiglioni, L. (2001). Stability of the preclinical of dementia in Canada. Neurology, 55, 66–73.
episodic memory deficit in Alzheimer’s disease. Brain, 124, 96–102. Casserly, I., and Topol, E. (2004). Convergence of atherosclerosis and
Baldereschi, M., et al. (1999). Dementia is a major predictor of death among Alzheimer’s disease: inflammation, cholesterol, and misfolded proteins.
the Italian elderly: the Italian Longitudinal Study on Aging. Neurology, Lancet, 363, 1139–46.
52, 709–13. Cataldo, J.K., Prochaska, J.J., and Glantz, S.A. (2010). Cigarette smoking is
Barberger-Gateau, P., et al. (2002). Fish, meat, and risk of dementia: cohort a risk factor for Alzheimer’s disease: an analysis controlling for tobacco
study. British Medical Journal, 325, 932–3. industry affiliation. Journal of Alzheimer’s Disease, 19, 465–80.
Barberger-Gateau, P., et al. (2007). Dietary patterns and risk of dementia: the Chang, C.C., et al. (2012). Smoking, death, and Alzheimer disease: a case of
Three-City cohort study. Neurology, 69, 1921–30. competing risks. Alzheimer Disease and Associated Disorders, 26, 300–6.
Barnes, D.E. and Yaffe, K. (2011). The projected effect of risk factor reduction Chang, M., et al. (2010). The effect of midlife physical activity on cognitive
on Alzheimer’s disease prevalence. Lancet Neurology, 10, 819–28. function among older adults: AGES-Reykjavik Study. Journals of
Barnes, D.E., et al. (2009). Predicting risk of dementia in older adults: the Gerontology Series A Biological Sciences and Medical Sciences, 65,
late-life dementia risk index. Neurology, 73, 173–9. 1369–74.
Barnes, D.E., et al (2010). Secondhand smoke, vascular disease, and dementia Christensen, K., et al. (2009). Ageing populations: the challenges ahead.
incidence: findings from the Cardiovascular Health Cognition Study. Lancet, 374, 1196–208.
American Journal of Epidemiology, 171, 292–302. Chui, H.C. (2006). Vascular cognitive impairment: today and tomorrow.
Bassuk, S.S., Glass, T.A., and Berkman, L.F. (1999). Social disengagement Alzheimer’s and Dementia, 2, 185–94.
and incident cognitive decline in community-dwelling elderly persons. Churchill, J.D., et al. (2002). Exercise, experience and the aging brain.
Annals of Internal Medicine, 131, 165–73. Neurobiology of Aging, 23, 941–55.
Beard, C.M., et al. (1996). Cause of death in Alzheimer’s disease. Annals of Colcombe, S., and Kramer, A.F. (2003). Fitness effects on the cognitive
Epidemiology, 6, 195–200. function of older adults: a meta-analytic study. Psychological Science, 14,
Beeri, M.S., et al. (2006). Coronary artery disease is associated with Alzheimer 125–30.
disease neuropathology in APOE4 carriers. Neurology, 66, 1399–404. Cornelius, C., et al. (2004). Aspirin, NSAIDs, risk of dementia, and
Bennett, D.A., et al. (2006). The effect of social networks on the relation influence of the apolipoprotein E ε4 allele in an elderly population.
between Alzheimer’s disease pathology and level of cognitive function in Neuroepidemiology, 23, 135–43.
old people: a longitudinal cohort study. Lancet Neurology, 5, 406–12. Corrada, M.M., et al. (2008). Prevalence of dementia after age 90: results
Bermejo-Pareja, F., et al. (2010). Risk of incident dementia in drug-untreated from the 90 + study. Neurology, 71, 337–43.
arterial hypertension: a population-based study. Journal of Alzheimer’s Craig, M.C., Maki, P.M., and Murphy, D.G. (2005). The Women’s Health
Disease, 22, 949–58. Initiative Memory Study: findings and implications for treatment. Lancet
Beydoun, M.A., Beydoun, H.A., and Wang, Y. (2008). Obesity and central Neurology, 4, 190–4.
obesity as risk factors for incident dementia and its subtypes: a systematic Craik, F.I., Bialystok, E., and Freedman, M. (2010). Delaying the onset of
review and meta-analysis. Obesity Reviews, 9, 204–18. Alzheimer disease: bilingualism as a form of cognitive reserve. Neurology,
Beydoun, M.A., et al. (2011). Statins and serum cholesterol’s associations 75, 1726–9.
with incident dementia and mild cognitive impairment. Journal of Crowe, M., et al. (2003). Does participation in leisure activities lead to
Epidemiology and Community Health, 65, 949–57. reduced risk of Alzheimer’s disease? A prospective study of Swedish
twins. Journals of Gerontology Series B Psychological Sciences and Social Engelhart, M.J., et al. (2005). Plasma levels of antioxidants are not associated
Sciences, 58, P249–55. with Alzheimer’s disease or cognitive decline. Dementia and Geriatric
Dahl, A.K., et al. (2008). Overweight and obesity in old age are not associated Cognitive Disorders, 19, 134–9.
with greater dementia risk. Journal of the American Geriatrics Society, Esiri, M.M., et al. (1999). Cerebrovascular disease and threshold for dementia
56, 2261–6. in the early stages of Alzheimer’s disease. Lancet, 354, 919–20.
Dai, Q., et al. (2006). Fruit and vegetable juices and Alzheimer’s disease: the Espeland, M.A., et al. (2004). Conjugated equine estrogens and global
Kame Project. American Journal of Medicine, 119, 751–9. cognitive function in postmenopausal women: Women’s Health
Dal Forno, G., et al. (2005). Depressive symptoms, sex, and risk for Alzheimer’s Initiative Memory Study. Journal of the American Medical Association,
disease. Annals of Neurology, 57, 381–7. 291, 2959–68.
Dangour, A.D., et al. (2010). B-vitamins and fatty acids in the prevention Etgen, T., et al. (2010). Physical activity and incident cognitive impairment
and treatment of Alzheimer’s disease and dementia: a systematic review. in elderly persons: the INVADE study. Archives of Internal Medicine, 170,
Journal of Alzheimer’s Disease, 22, 205–24. 186–93.
Debette, S., and Markus, H.S. (2010). The clinical importance of white matter Etminan, M., Gill, S., and Samii, A. (2003). Effect of non-steroidal
hyperintensities on brain magnetic resonance imaging: systematic review anti-inflammatory drugs on risk of Alzheimer’s disease: systematic
and meta-analysis. British Medical Journal, 341, c3666. review and meta-analysis of observational studies. British Medical
Debette, S., et al. (2010). Association of MRI markers of vascular brain Journal, 327, 128.
injury with incident stroke, mild cognitive impairment, dementia, and Euser, S.M., et al. (2009). The effect of age on the association between blood
mortality: the Framingham Offspring Study. Stroke, 41, 600–6. pressure and cognitive function later in life. Journal of the American
Debette, S., et al. (2011). Midlife vascular risk factor exposure accelerates Geriatrics Society, 57, 1232–7.
structural brain aging and cognitive decline. Neurology, 77, 461–8. Eussen, S.J., et al. (2006). Effect of oral vitamin B-12 with or without folic acid
de Craen, A.J., et al. (2005). Meta-analysis of nonsteroidal antiinflammatory on cognitive function in older people with mild vitamin B-12 deficiency:
drug use and risk of dementia. American Journal of Epidemiology, 161, a randomized, placebo-controlled trial. American Journal of Clinical
114–20. Nutrition, 84, 361–70.
de Jong, F.J., et al. (2011). Retinal vascular caliber and risk of dementia: the Evans, D.A., et al. (1997a). Apolipoprotein E ε4 and incidence of Alzheimer
Rotterdam study. Neurology, 76, 816–21. disease in a community population of older persons. Journal of the
De Ronchi, D., et al. (1998). The effect of education on dementia occurrence American Medical Association, 277, 822–4.
in an Italian population with middle to high socioeconomic status. Evans, D.A., et al. (1997b). Education and other measures of socioeconomic
Neurology, 50, 1231–8. status and risk of incident Alzheimer disease in a defined population of
Devi, G., et al. (2000). Familial aggregation of Alzheimer disease among older persons. Archives of Neurology, 54, 1399–405.
whites, African Americans, and Caribbean Hispanics in northern Fabrigoule, C., et al. (1995). Social and leisure activities and risk of dementia:
Manhattan. Archives of Neurology, 57, 72–7. a prospective longitudinal study. Journal of the American Geriatrics
Devore, E.E., et al. (2009). Dietary intake of fish and omega-3 fatty acids Society, 43, 485–90.
in relation to long-term dementia risk. American Journal of Clinical Farrer, L.A., et al. (1997). Effects of age, sex, and ethnicity on the association
Nutrition, 90, 170–6. between apolipoprotein E genotype and Alzheimer disease: a meta-analysis.
Di Bari, M., et al. (2001). Dementia and disability outcomes in large Journal of the American Medical Association, 278, 1349–56.
hypertension trials: lessons learned from the Systolic Hypertension in Féart, C., et al. (2009). Adherence to a Mediterranean diet, cognitive decline,
the Elderly Program (SHEP) trial. American Journal of Epidemiology, and risk of dementia. Journal of the American Medical Association, 302,
153, 72–8. 638–48.
Di Carlo, A., et al. (2002). Incidence of dementia, Alzheimer’s disease, and Feldman, H.H., et al. (2010). Randomized controlled trial of atorvastatin in
vascular dementia in Italy: the ILSA Study. Journal of the American mild to moderate Alzheimer disease: LEADe. Neurology, 74, 956–64.
Geriatrics Society, 50, 41–8. Ferri, C.P., et al. (2005). Global prevalence of dementia: a Delphi consensus
Dolan, H., et al. (2010). Atherosclerosis, dementia, and Alzheimer’s disease study. Lancet, 366, 2112–17.
in the BLSA cohort. Annals of Neurology, 68, 231–40. Feychting, M., et al. (2003). Occupational magnetic field exposure and
Doll, R., et al. (2000). Smoking and dementia in male British doctors: neurodegenerative disease. Epidemiology, 14, 413–9.
prospective study. British Medical Journal, 320, 1097–102. Filippini, N., et al. (2011). Differential effects of the APOE genotype on brain
Dotson, V.M., Beydoun, M.A., and Zonderman, A.B. (2010). Recurrent function across the lifespan. NeuroImage, 54, 602–10.
depressive symptoms and the incidence of dementia and mild cognitive Fleminger, S., et al. (2003). Head injury as a risk factor for Alzheimer’s disease:
impairment. Neurology, 75, 27–34. the evidence 10 years on; a partial replication. Journal of Neurology,
Dublin, S., et al. (2011). Atrial fibrillation and risk of dementia: a prospective Neurosurgery and Psychiatry, 74, 857–62.
cohort study. Journal of the American Geriatrics Society, 59, 1369–75. Ford, A.H., and Almeida, O.P. (2012). Effect of homocysteine lowering
Dufouil, C., et al. (2005). APOE genotype, cholesterol level, lipid-lowering treatment on cognitive function: a systematic review and meta-analysis
treatment, and dementia: the Three-City Study. Neurology, 64, 1531–8. of randomized controlled trials. Journal of Alzheimer’s Disease, 29,
Elias, M.F., et al. (2009). Arterial pulse wave velocity and cognition with 133–49.
advancing age. Hypertension, 53, 668–73. Forette, F., et al. (1998). Prevention of dementia in randomised double-blind
Elias, P.K., et al. (1999). Alcohol consumption and cognitive performance in placebo-controlled Systolic Hypertension in Europe (Syst-Eur) trial.
the Framingham Heart Study. American Journal of Epidemiology, 150, Lancet, 352, 1347–51.
580–9. Forette, F., et al. (2002). The prevention of dementia with antihypertensive
Emmerling, M.R., et al. (2000). Traumatic brain injury elevates the Alzheimer’s treatment: new evidence from the Systolic Hypertension in Europe
amyloid peptide A beta 42 in human CSF: a possible role for nerve cell (Syst-Eur) study. Archives of Internal Medicine, 162, 2046–52.
injury. Annals of the New York Academy of Sciences, 903, 118–22. Fotuhi, M., Mohassel, P., and Yaffe, K. (2009). Fish consumption, long-chain
Engelhart, M.J., et al. (2002). Dietary intake of antioxidants and risk of omega-3 fatty acids and risk of cognitive decline or Alzheimer disease: a
Alzheimer disease. Journal of the American Medical Association, 287, complex association. Nature Clinical Practice Neurology, 5, 140–52.
3223–9. Foubert-Samier, A., et al. (2012). Education, occupation, leisure activities,
Engelhart, M.J., et al. (2004). Inflammatory proteins in plasma and the risk of and brain reserve: a population-based study. Neurobiology of Aging, 33,
dementia: the Rotterdam study. Archives of Neurology, 61, 668–72. 423.e15–25.
Fratiglioni, L. and Qiu, C. (2011). Possibilities of prevention in cognitive Green, R.C., et al. (2003). Depression as a risk factor for Alzheimer disease:
ageing: dementia as target, delay onset as goal. Lancet Neurology, 10, the MIRAGE Study. Archives of Neurology, 60, 753–9.
778–9. Griffiths, C., and Rooney, C. (2006). Trends in mortality from Alzheimer’s
Fratiglioni, L. and Rocca, W. (2001). Epidemiology of dementia. In: Boller, F. disease, Parkinson’s disease and dementia, England and Wales, 1979–
and Cappa, S.F. (eds) Handbook of neuropsychology: aging and dementia, 2004. Health Statistics Quarterly, (30), 6–14.
pp. 193–215. Elsevier Science, Amsterdam. Gu, Y., et al. (2010). Food combination and Alzheimer disease risk: a
Fratiglioni, L. and Wang, H.X. (2000). Smoking and Parkinson’s and protective diet. Archives of Neurology, 67, 699–706.
Alzheimer’s disease: review of the epidemiological studies. Behavioural Guerchet, M., et al. (2011). Ankle-brachial index as a marker of cognitive
Brain Research, 113, 117–20. impairment and dementia in general population: a systematic review.
Fratiglioni, L., et al. (1993). Risk factors for late-onset alzheimer’s disease: a Atherosclerosis, 216, 251–7.
population-based, case-control study. Annals of Neurology, 33, 258–66. Guerreiro, R.J., Gustafson, D.R., and Hardy, J. (2012). The genetic architecture
Fratiglioni, L., et al. (1997). Very old women at highest risk of dementia and of Alzheimer’s disease: beyond APP, PSENs and APOE. Neurobiology of
Alzheimer’s disease: incidence data from the Kungsholmen Project, Aging, 33, 437–56.
Stockholm. Neurology, 48, 132–8. Gühne, U., et al. (2006). Incident dementia cases and mortality: results of
Fratiglioni, L., et al. (1999). Worldwide prevalence and incidence of dementia. the Leipzig Longitudinal Study of the Aged (LEILA75 +). Dementia and
Drugs and Aging, 15, 365–75. Geriatric Cognitive Disorders, 22, 185–93.
Fratiglioni, L., et al. (2000a). Incidence of dementia and major subtypes in Guo, Z., et al. (1999a). Occurrence and progression of dementia in a
Europe: a collaborative study of population-based cohorts. Neurology, community population aged 75 years and older: relationship of
54, S10–15. antihypertensive medication use. Archives of Neurology, 56, 991–6.
Fratiglioni, L., et al. (2000b). Influence of social network on occurrence of Guo, Z., et al. (1999b). Low blood pressure and incidence of dementia in a
dementia: a community-based longitudinal study. Lancet, 355, 1315–19. very old sample: dependent on initial cognition. Journal of the American
Fratiglioni, L., Paillard-Borg, S., and Winblad, B. (2004). An active and Geriatrics Society, 47, 723–6.
socially integrated lifestyle in late life might protect against dementia. Guo, Z., et al. (2000). Head injury and the risk of AD in the MIRAGE study.
Lancet Neurology, 3, 343–53. Neurology, 54, 1316–23.
Fratiglioni, L., von Strauss, E., and Qiu, C. (2008). Epidemiology of the Gustafson, D. (2006). Adiposity indices and dementia. Lancet Neurology, 5,
dementias of old age. In: Dening, T., et al. (eds) The Oxford textbook 713–20.
of old age psychiatry, 4th edition. Oxford University Press, London, Gustafson, D., et al. (2004). A 24-year follow-up of body mass index and
pp. 391–406. cerebral atrophy. Neurology, 63, 1876–81.
Fratiglioni, L., Mangialasche, F., and Qiu, C. (2010). Brain aging—lessons Gustafson, D.R., et al. (2009). Adiposity indicators and dementia over 32
from community studies. Nutrition Reviews, 68, S119–27. years in Sweden. Neurology, 73, 1559–66.
Ganguli, M. and Rodriguez, E.G. (1999). Reporting of dementia on death Gustafson, D.R., et al. (2012). 37 years of body mass index and dementia:
certificates: a community study. Journal of the American Geriatric Society, observations from the prospective population study of women in
47, 842–9. Gothenburg, Sweden. Journal of Alzheimer’s Disease, 28, 163–71.
Ganguli, M., et al. (2006). Depressive symptoms and cognitive decline in late Hajjar, I., et al. (2002). The impact of the use of statins on the prevalence of
life: a prospective epidemiological study. Archives of General Psychiatry, dementia and the progression of cognitive impairment. Journals of
63, 153–60. Gerontology Series A Biological Sciences and Medical Sciences, 57, M414–8.
Gao, S., et al. (2011). Accelerated weight loss and incident dementia in an Hajjar, I., et al. (2005). Cross-sectional and longitudinal association between
elderly African-American cohort. Journal of the American Geriatrics antihypertensive medications and cognitive impairment in an elderly
Society, 59, 18–25. population. Journals of Gerontology Series A Biological Sciences and
García, A.M., Sisternas, A., and Hoyos, S.P. (2008). Occupational exposure Medical Sciences, 60, 67–73.
to extremely low frequency electric and magnetic fields and Alzheimer Hajjar, I., et al. (2011). Hypertension, white matter hyperintensities,
disease: a meta-analysis. International Journal of Epidemiology, 37, and concurrent impairments in mobility, cognition, and mood: the
329–40. Cardiovascular Health Study. Circulation, 123, 858–65.
Garcia-Alloza, M., et al. (2011). Cerebrovascular lesions induce transient Håkansson, K.,. et al. (2009). Association between mid-life marital status and
β-amyloid deposition. Brain, 134, 3697–707. cognitive function in later life: population based cohort study. British
Gatz, M., et al. (2006). Role of genes and environments for explaining Medical Journal, 339, b2462.
Alzheimer disease. Archives of General Psychiatry, 63, 168–74. Hakansson, N., et al. (2003). Neurodegenerative diseases in welders and
Gelber, R.P., et al. (2012). Lifestyle and the risk of dementia in other workers exposed to high levels of magnetic fields. Epidemiology,
Japanese-American men. Journal of the American Geriatrics Society, 60, 14, 420–6.
118–23. Hall, K.S., et al. (2009). Prevalence rates for dementia and Alzheimer’s disease
Godin, O., et al. (2011). Antihypertensive treatment and change in blood in African Americans: 1992 versus 2001. Alzheimer’s and Dementia, 5,
pressure are associated with the progression of white matter lesion 227–33.
volumes: the Three-City (3C)-Dijon Magnetic Resonance Imaging Study. Hamer, M., and Chida, Y. (2009). Physical activity and risk of neurodegenerative
Circulation, 123, 266–73. disease: a systematic review of prospective evidence. Psychological
Gordis, L. (2000). Epidemiology, 2nd edition, pp. 42–4. W.B. Saunders, Medicine, 39, 3–11.
Philadelphia. Hassing, L.B., et al. (2009). Overweight in midlife and risk of dementia: a 40-year
Gorelick, P.B. (2010). Role of inflammation in cognitive impairment: results follow-up study. International Journal of Obesity (London), 33, 893–8.
of observational epidemiological studies and clinical trials. Annals of the Hayden, K.M., et al. (2004). Self- or proxy-reported stroke and the risk of
New York Academy of Science, 1207, 155–62. Alzheimer disease. Archives of Neurology, 61, 982.
Gray, S.L., et al. (2008). Antioxidant vitamin supplement use and risk of Hayden, K.M., et al. (2006). Vascular risk factors for incident Alzheimer
dementia or Alzheimer’s disease in older adults. Journal of the American disease and vascular dementia: the Cache County study. Alzheimer’s
Geriatrics Society, 56, 291–5. Disease and Associated Disorders, 20, 93–100.
Green, R.C., et al. (2002). Risk of dementia among white and African Hayden, K.M., et al. (2009). Effects of family history and apolipoprotein E
American relatives of patients with Alzheimer disease. Journal of the ε4 status on cognitive decline in the absence of Alzheimer dementia: the
American Medical Association, 287, 329–36. Cache County Study. Archives of Neurology, 66, 1378–83.
Heart Protection Study Collaborative Group (2002). MRC/BHF Heart Johansson, L., et al. (2010). Midlife psychological stress and risk of dementia:
Protection Study of cholesterol lowering with simvastatin in 20,536 a 35-year longitudinal population study. Brain, 133, 2217–24.
high-risk individuals: a randomised placebo-controlled trial. Lancet, Johansson, L., et al. (2012). Midlife psychological distress associated with
360, 7–22. late-life brain atrophy and white matter lesions: a 32-year population
Hebert, L.E., et al. (2003). Alzheimer disease in the US population: prevalence study of women. Psychosomatic Medicine, 74, 120–5.
estimates using the 2000 census. Archives of Neurology, 60, 1119–22. Johnson, D,K,, Wilkins, C.H., and Morris, J.C. (2006). Accelerated weight
Hebert, L.E., et al. (2010). Change in risk of Alzheimer disease over time. loss may precede diagnosis in Alzheimer disease. Archives of Neurology,
Neurology, 75, 786–91. 63, 1312–17.
Helmer, C., et al. (2001). Mortality with dementia: results from a French Johnson, K.C., et al. (2008). A prospective study of the effect of hypertension
prospective community-based cohort. American Journal of Epidemiology, and baseline blood pressure on cognitive decline and dementia in
154, 642–8. postmenopausal women: the Women’s Health Initiative Memory Study.
Helzner, E.P., et al. (2008). Survival in Alzheimer disease: a multiethnic, Journal of the American Geriatrics Society, 56, 1449–58.
population-based study of incident cases. Neurology, 71, 1489–95. Johnson, V.E., Stewart, W., and Smith, D.H. (2010). Traumatic brain injury
Hernán, M.A., Alonso, A., and Logroscino, G. (2008). Cigarette smoking and amyloid-β pathology: a link to Alzheimer’s disease? Nature Review
and dementia: potential selection bias in the elderly. Epidemiology, 19, Neuroscience, 11, 361–70.
448–50. Jorm, A.F. (2001). History of depression as a risk factor for dementia: an updated
Hill, G., et al. (2003). Neyman’s bias re-visited. Journal of Clinical Epidemiology, review. Australian and New Zealand Journal of Psychiatry, 35, 776–81.
56, 293–6. Jorm, A.F. and Jolley, D. (1998). The incidence of dementia: a meta-analysis.
Hoffman, L.B., et al. (2009). Less Alzheimer disease neuropathology in Neurology, 51, 728–33.
medicated hypertensive than nonhypertensive persons. Neurology, 72, Kalmijn, S., et al. (1997). Dietary fat intake and the risk of incident dementia
1720–6. in the Rotterdam Study. Annals of Neurology, 42, 776–82
Hofman, A., et al. (1997). Atherosclerosis, apolipoprotein E, and prevalence Kalmijn, S., et al. (2000). Metabolic cardiovascular syndrome and risk of
of dementia and Alzheimer’s disease in the Rotterdam Study. Lancet, dementia in Japanese-American elderly men: the Honolulu-Asia aging
349, 151–4. study. Arteriosclerosis, Thrombosis, and Vascular Biology, 20, 2255–60.
Hogervorst, E., et al. (2009). Hormone replacement therapy to maintain Kang, J.H., Weuve, J., and Grodstein, F. (2004). Postmenopausal hormone
cognitive function in women with dementia. Cochrane Database of therapy and risk of cognitive decline in community-dwelling aging
Systematic Reviews, 1, CD003799. women. Neurology, 63, 101–7.
Honig, L.S., et al. (2003). Stroke and the risk of Alzheimer disease. Archives Kang, J.H., et al. (2006). A randomized trial of vitamin E supplementation and
of Neurology, 60, 1707–12. cognitive function in women. Archives of Internal Medicine, 166, 2462–8.
Ho, R.C., et al. (2011). Is high homocysteine level a risk factor for Karp, A., et al. (2004). Relation of education and occupation-based
cognitive decline in elderly? A systematic review, meta-analysis, and socioeconomic status to incident Alzheimer’s disease. American Journal
meta-regression. American Journal of Geriatric Psychiatry, 19, 607–17. of Epidemiology, 159, 175–83.
Hoyert, D.L., and Rosenberg, H.M. (1997). Alzheimer’s disease as a cause of Karp, A., et al. (2006). Mental, physical and social components in leisure
death in the United States. Public Health Report, 112, 497–505. activities equally contribute to decrease dementia risk. Dementia and
Huang, W., et al. (2002). Alcohol consumption and incidence of dementia Geriatric Cognitive Disorders, 21, 65–73.
in a community sample aged 75 years and older. Journal of Clinical Katon, W., et al. (2012). Association of depression with increased risk of
Epidemiology, 55, 959–64. dementia in patients with type 2 diabetes: the Diabetes and Aging Study.
Huang, W., et al. (2004). APOE genotype, family history of dementia, and Archives of General Psychiatry, 69(4), 410–17.
Alzheimer disease risk: a 6-year follow-up study. Archives of Neurology, Katz, M.J., et al. (2012). Age-specific and sex-specific prevalence and incidence
61, 1930–4. of mild cognitive impairment, dementia, and Alzheimer dementia in
Iadecola, C., and Gorelick, P.B. (2003). Converging pathogenic mechanisms Blacks and Whites: a report from the Einstein Aging Study. Alzheimer
in vascular and neurodegenerative dementia. Stroke, 34, 335–7. Disease and Associated Disorders, 26, 335–43.
Ikram, M.A., et al. (2008). Unrecognized myocardial infarction in relation to Katzman, R. (1993). Education and the prevalence of dementia and
risk of dementia and cerebral small vessel disease. Stroke, 39, 1421–6. Alzheimer’s disease. Neurology, 43, 13–20.
in’t Veld, B.A., et al. (2001a). Nonsteroidal anti-inflammatory drugs and the risk Katzman, R. (2004). Luigi Amaducci memorial award winner’s paper 2003.
of Alzheimer’s disease. New England Journal of Medicine, 345, 1515–21. A neurologist’s view of Alzheimer’s disease and dementia. International
in’t Veld, B.A., et al. (2001b). Antihypertensive drugs and incidence of Psychogeriatrics, 16, 259–73.
dementia: the Rotterdam Study. Neurobiology of Aging, 22, 407–12. Katzman, R., et al. (1994). The malignancy of dementia: predictors of
Irie, F., et al. (2008). Apolipoprotein E ε4 allele genotype and the effect of mortality in clinically diagnosed dementia in a population survey of
depressive symptoms on the risk of dementia in men: the Honolulu-Asia Shanghai, China. Archives of Neurology, 51, 1220–5.
Aging Study. Archives of General Psychiatry, 65, 906–12. Kawas, C., et al. (1997). A prospective study of estrogen replacement
Jagger, C., et al. (2000). Prognosis with dementia in Europe: A collaborative therapy and the risk of developing Alzheimer’s disease: the Baltimore
study of population-based cohorts. Neurology, 54, S16–20. Longitudinal Study of Aging. Neurology, 48, 1517–21.
Järvenpää, T., et al. (2005). Binge drinking in midlife and dementia risk. Kawas, C., et al. (2000). Age-specific incidence rates of Alzheimer’s disease:
Epidemiology, 16, 766–71. the Baltimore Longitudinal Study of Aging. Neurology, 54, 2072–7.
Jefferson, A.L., et al. (2010). Cardiac index is associated with brain aging: the Keller, L., et al. (2011). The obesity related gene, FTO, interacts with APOE,
Framingham Heart Study. Circulation, 122, 690–7. and is associated with Alzheimer’s disease risk: a prospective cohort
Jick, H., et al. (2000). Statins and the risk of dementia. Lancet, 356, 1627–31. study. Journal of Alzheimer’s Disease, 23, 461–9.
Jin, Y.P., et al. (2004). Sensitivity and specificity of dementia coding in two Kemppainen, N.M., et al. (2008). Cognitive reserve hypothesis: Pittsburgh
Swedish disease registries. Neurology, 63, 739–41. Compound B and fluorodeoxyglucose positron emission tomography in
Jin, Y.P., et al. (2008). Joint effect of stroke and APOE 4 on dementia risk: the relation to education in mild Alzheimer’s disease. Annals of Neurology,
Canadian Study of Health and Aging. Neurology, 70, 9–16. 63, 112–18.
Joas, E., et al. (2012). Blood pressure trajectories from midlife to late life in Kennelly, S.P., Lawlor, B.A., and Kenny, R.A. (2009). Blood pressure and the
relation to dementia in women followed for 37 years. Hypertension, 59, risk for dementia: a double edged sword. Ageing Research Reviews, 8,
796–801. 61–70.
Khachaturian, A.S., et al. (2006). Antihypertensive medication use and Launer, L.J. (2002). Demonstrating the case that AD is a vascular disease:
incident Alzheimer disease: the Cache County Study. Archives of epidemiologic evidence. Ageing Research Reviews, 1, 61–77.
Neurology, 63, 686–92. Launer, L.J., et al. (1999). Rates and risk factors for dementia and Alzheimer’s
Kinsella, K. and Velkoff, V.A. (2002). The demographics of aging. Aging disease: results from EURODEM pooled analyses. Neurology, 52, 78–84.
Clinical and Experimental Research, 14, 159–69. Launer, L.J., et al. (2000). Midlife blood pressure and dementia: the
Kivipelto, M., et al. (2001). Midlife vascular risk factors and Alzheimer’s Honolulu-Asia aging study. Neurobiology of Aging, 21, 49–55.
disease in later life: longitudinal, population based study. British Medical Launer, L.J., et al. (2010). Lowering midlife levels of systolic blood pressure
Journal, 322, 1447–51. as a public health strategy to reduce late-life dementia: perspective from
Kivipelto, M., et al. (2002). Apolipoprotein E ε4 allele, elevated midlife total the Honolulu Heart Program/Honolulu Asia Aging Study. Hypertension,
cholesterol level, and high midlife systolic blood pressure are independent 55, 1352–9.
risk factors for late-life Alzheimer disease. Annals of Internal Medicine, Laurin, D., et al. (2004). Midlife dietary intake of antioxidants and risk of
137, 149–55. late-life incident dementia: the Honolulu-Asia Aging Study. American
Kivipelto, M., et al. (2005). Obesity and vascular risk factors at midlife and Journal of Epidemiology, 159, 959–67.
the risk of dementia and Alzheimer disease. Archives of Neurology, 62, Laurin, D., et al. (2007). Ankle-to-brachial index and dementia: the
1556–60. Honolulu-Asia Aging Study. Circulation, 116, 2269–74.
Kivipelto, M., et al. (2006). Risk score for the prediction of dementia risk in Lavery, L., et al. (2007). The association between congestive heart failure and
20 years among middle aged people: a longitudinal, population-based cognitive performance in a primary care population of elderly adults: the
study. Lancet Neurology, 5, 735–41. Steel Valley Seniors Survey. International Psychogeriatrics, 19, 215–25.
Knopman, D.S. and Roberts, R. (2010). Vascular risk factors: imaging Leibson, C.L., et al. (1997). Risk of dementia among persons with diabetes
and neuropathologic correlates. Journal of Alzheimer’s Disease, 20, mellitus: a population-based cohort study. American Journal of
699–709. Epidemiology, 145, 301–8.
Knowler, W.C., et al. (2002). Reduction in the incidence of type 2 diabetes Lenoir, H., et al. (2011). Depression history, depressive symptoms, and
with lifestyle intervention or metformin. New England Journal of incident dementia: the 3C Study. Journal of Alzheimer’s Disease, 26,
Medicine, 346, 393–403. 27–38.
Kopf, D., and Frölich, L. (2009). Risk of incident Alzheimer’s disease in Leys, D., et al. (2005). Poststroke dementia. Lancet Neurology, 4, 752–9.
diabetic patients: a systematic review of prospective trials. Journal of Li, G., et al. (2004). Statin therapy and risk of dementia in the elderly: a
Alzheimer’s Disease, 16, 677–85. community-based prospective cohort study. Neurology, 63, 1624–8.
Korf, ES, et al. (2004). Midlife blood pressure and the risk of hippocampal Li, G., et al. (2007a). Statin therapy is associated with reduced neuropathologic
atrophy: the Honolulu Asia Aging Study. Hypertension, 44, 29–34. changes of Alzheimer disease. Neurology, 69, 878–85.
Korf, E.S., et al. (2006). Brain aging in very old men with type 2 diabetes: the Li, G., et al. (2007b). Age-varying association between blood pressure and risk
Honolulu-Asia Aging Study. Diabetes Care, 29, 2268–74. of dementia in those aged 65 and older: a community-based prospective
Kramer, A.F., et al. (1999). Ageing, fitness and neurocognitive function. cohort study. Journal of the American Geriatrics Society, 55, 1161–7.
Nature, 400, 418–19. Li, G., et al. (2011). Temporal relationship between depression and dementia:
Kröger, E., et al. (2009). Omega-3 fatty acids and risk of dementia: the findings from a large community-based 15-year follow-up study. Archives
Canadian Study of Health and Aging. American Journal of Clinical of General Psychiatry, 68, 970–7.
Nutrition, 90, 184–92. Li, N.C., et al. (2010). Use of angiotensin receptor blockers and risk of
Kuh, D. and Ben-Shlomo, Y. (2004). A life course approach to chronic disease dementia in a predominantly male population: prospective cohort
epidemiology, 2nd edition, pp. 1–14. Oxford University Press, New York. analysis. British Medical Journal, 340, b5465.
Kuller, L.H., and Lopez, O.L. (2011). Dementia and Alzheimer’s disease: a Li, S., et al. (2007). Is the dementia rate increasing in Beijing? Prevalence and
new direction. The 2010 Jay L. Foster Memorial lecture. Alzheimer’s and incidence of dementia 10 years later in an urban elderly population. Acta
Dementia, 7, 540–50. Psychiatrica Scandinavica, 115, 73–9.
Kuo, H.K., et al. (2005), Sorond F. Relation of C-reactive protein to stroke, Liebetrau, M., et al. (2004). Silent and symptomatic infarcts on cranial
cognitive disorders, and depression in the general population: systematic computerized tomography in relation to dementia and mortality: a
review and meta-analysis. Lancet Neurology, 4, 371–80. population-based study in 85-year-old subjects. Stroke, 35, 1816–20.
Kuusisto, J., et al. (1997). Association between features of the insulin resistance Lindström, J., et al. (2006). Sustained reduction in the incidence of type 2
syndrome and Alzheimer’s disease independently of apolipoprotein diabetes by lifestyle intervention: follow-up of the Finnish Diabetes
E4 phenotype: cross sectional population based study. British Medical Prevention Study. Lancet, 368, 1673–9.
Journal, 315, 1045–9. Lithell, H., et al. (2003). The Study on Cognition and Prognosis in the Elderly
Kwok, C.S., et al. (2011). Atrial fibrillation and incidence of dementia: a (SCOPE): principal results of a randomized double-blind intervention
systematic review and meta-analysis. Neurology, 76, 914–22. trial. Journal of Hypertension, 21, 875–86.
Lahiri, D.K., Sambamurti, K., and Bennett, D.A. (2004). Apolipoprotein Lobo, A., et al. (2000). Prevalence of dementia and major subtypes in Europe:
gene and its interaction with the environmentally driven risk factors: a collaborative study of population-based cohorts. Neurology, 54, S4–9.
molecular, genetic and epidemiological studies of Alzheimer’s disease. Lobo, A., et al. (2007). Prevalence of dementia in a southern European
Neurobiology of Aging, 25, 651–60. population in two different time periods: the ZARADEMP Project. Acta
Laitinen, M.H., et al. (2006). Fat intake at midlife and risk of dementia and Psychiatrica Scandinavica, 116, 299–307.
Alzheimer’s disease: a population-based study. Dementia and Geriatric Lobo, A., et al. (2011). Incidence and lifetime risk of dementia and
Cognitive Disorders, 22, 99–107. Alzheimer’s disease in a Southern European population. Acta Psychiatrica
Landi, F., et al. (2003). Non-steroidal anti-inflammatory drug (NSAID) use Scandinavica, 124, 372–83.
and Alzheimer disease in community-dwelling elderly patients. American Luchsinger, J.A. and Mayeux, R. (2004). Dietary factors and Alzheimer’s
Journal of Geriatric Psychiatry, 11, 179–85. disease. Lancet Neurology, 3, 579–87.
Lanska, D.J. (1998). Dementia mortality in the United States. Results of the Luchsinger, J.A., et al. (2001). Diabetes mellitus and risk of Alzheimer’s
1986 National Mortality Followback Survey. Neurology, 50, 362–7. disease and dementia with stroke in a multiethnic cohort. American
Larson, E.B., et al. (2006). Exercise is associated with reduced risk for incident Journal of Epidemiology, 154, 635–41.
dementia among persons 65 years of age and older. Annals of Internal Luchsinger, J.A., et al. (2003). Antioxidant vitamin intake and risk of
Medicine, 144, 73–81. Alzheimer disease. Archives of Neurology, 60, 203–8.
Luchsinger, J.A., et al. (2004a). Hyperinsulinemia and risk of Alzheimer Meinert, C.L. and Breitner, J.C. (2008). Chronic disease long-term drug
disease. Neurology, 63, 1187–92. prevention trials: lessons from the Alzheimer’s Disease Anti-inflammatory
Luchsinger, J.A., et al. (2004b). Plasma homocysteine levels and risk of Prevention Trial (ADAPT). Alzheimer’s and Dementia, 4(Suppl 1),
Alzheimer disease. Neurology, 62, 1972–6. S7–14.
Luchsinger, J.A., et al. (2005). Aggregation of vascular risk factors and risk of Merchant, C., et al. (1999). The influence of smoking on the risk of Alzheimer’s
incident Alzheimer disease. Neurology, 65, 545–51. disease. Neurology, 52, 1408–12.
Luchsinger, J.A., et al. (2007). Relation of higher folate intake to lower risk of Middleton, L.E. and Yaffe, K. (2009). Promising strategies for the prevention
Alzheimer disease in the elderly. Archives of Neurology, 64, 86–92. of dementia. Archives of Neurology, 66, 1210–15.
Luchsinger, J.A., et al. (2012). Central obesity in the elderly is related to Miech, R.A., et al. (2002). Incidence of AD may decline in the early 90s for
late-onsetAlzheimer disease. Alzheimer Disease and Associated Disorders, men, later for women: the Cache County study. Neurology, 58, 209–18.
26, 101–05. Mielke, M.M., et al. (2005). High total cholesterol levels in late life associated
Lu, F.P., Lin, K.P., and Kuo, H.K. (2009). Diabetes and the risk of multi-system with a reduced risk of dementia. Neurology, 64, 1689–95.
aging phenotypes: a systematic review and meta-analysis. PLoS One, 4, Mielke, M.M., et al. (2010). The 32-year relationship between cholesterol and
e4144. dementia from midlife to late life. Neurology, 75, 1888–95.
Lyketsos, C.G., et al. (2007). Naproxen and celecoxib do not prevent AD Mielke, M.M., et al. (2011). Interaction between vascular factors and the
in early results from a randomized controlled trial. Neurology, 68, APOE ε4 allele in predicting rate of progression in Alzheimer’s disease.
1800–8. Journal of Alzheimer’s Disease, 26, 127–34.
Malouf, M., Grimley, E.J., and Areosa, S.A. (2003). Folic acid with or without Mitnitski, A., et al. (2006). A vascular risk factor index in relation to mortality
vitamin B12 for cognition and dementia. Cochrane Database of Systematic and incident dementia. European Journal of Neurology, 13, 514–21.
Reviews, (4), CD004514. Moceri, V.M., et al. (2000). Early-life risk factors and the development of
Marengoni, A., et al. (2011). Atrial fibrillation, stroke and dementia in the Alzheimer’s disease. Neurology, 54, 415–20.
very old: a population-based study. Neurobiology of Aging, 32, 1336–7. Modrego, P.J. and Ferrandez, J. (2004). Depression in patients with mild
Martin, B.K., et al. (2008). Cognitive function over time in the Alzheimer’s cognitive impairment increases the risk of developing dementia of
disease anti-inflammatory prevention trial (ADAPT): results of a Alzheimer type: a prospective cohort study. Archives of Neurology, 61,
randomized, controlled trial of naproxen and celecoxib. Archives of 1290–3.
Neurology, 65, 896–905. Monastero, R., et al. (2007). Heterogeneity in risk factors for cognitive
Mathillas, J., Lövheim, H., and Gustafson, Y. (2011). Increasing prevalence of impairment, no dementia: population-based longitudinal study from
dementia among very old people. Age and Ageing, 40, 243–9. the Kungsholmen Project. American Journal of Geriatric Psychiatry, 15,
Matthews, F. and Brayne, C. (2005). The incidence of dementia in England 60–9.
and Wales: findings from the five identical sites of the MRC CFA Study. Montlahuc, C., et al. (2011). Self-rated health and risk of incident dementia: a
PLoS Medicine, 2, e193. community-based elderly cohort, the 3C study. Neurology, 77, 1457–64.
Maxwell, C.J., Hogan, D.B., and Ebly, E.M. (2002). Serum folate levels and Morris, M.C., et al. (2001). Association of incident Alzheimer disease and
subsequent adverse cerebrovascular outcomes in elderly persons. blood pressure measured from 13 years before to 2 years after diagnosis
Dementia and Geriatric Cognitive Disorders, 13, 225–34. in a large community study. Archives of Neurology, 58, 1640–6.
Maxwell, C.J., et al. (2005). Supplemental use of antioxidant vitamins and Morris, M.C., et al. (2002). Dietary intake of antioxidant nutrients and the
subsequent risk of cognitive decline and dementia. Dementia and risk of incident Alzheimer disease in a biracial community study. Journal
Geriatric Cognitive Disorders, 20, 45–51. of the American Medical Association, 287, 3230–7.
Mayeux, R., et al. (1995). Synergistic effects of traumatic head injury and Morris, M.C., et al. (2003a). Dietary fats and the risk of incident Alzheimer
apolipoprotein-ε4 in patients with Alzheimer’s disease. Neurology, 45, disease. Archives of Neurology, 60, 194–200.
555–7. Morris, M.C., et al. (2003b). Consumption of fish and n-3 fatty acids and risk
McDowell, I., et al. (2007). Mapping the connections between education and of incident Alzheimer disease. Archives of Neurology, 60, 940–6.
dementia. Journal of Clinical and Experimental Neuropsychology, 29, Morris, M.C., Schneider, J.A., and Tangney, C.C. (2006). Thoughts on
127–41. B-vitamins and dementia. Journal of Alzheimer’s Disease, 9, 429–33.
McGeer, P.L. and McGeer, E.G. (2007). NSAIDs and Alzheimer disease: Mortimer, J.A., et al. (1991). Head trauma as a risk factor for Alzheimer’s
epidemiological, animal model and clinical studies. Neurobiology of disease: a collaborative re-analysis of case-control studies. International
Aging, 28, 639–47. Journal of Epidemiology, 20 Suppl 2, S28–35.
McGuinness, B., et al. (2009a). Statins for the prevention of dementia. Mura, T., Dartigues, J.F., and Berr, C. (2010). How many dementia cases in
Cochrane Database of Systematic Reviews, (2), CD003160. France and Europe? Alternative projections and scenarios 2010–2050.
McGuinness, B., et al. (2009b). Blood pressure lowering in patients without European Journal of Neurology, 17, 252–9.
prior cerebrovascular disease for prevention of cognitive impairment and Murray, A.D., et al. (2011). The balance between cognitive reserve and brain
dementia. Cochrane Database of Systematic Reviews, (4), CD004034. imaging biomarkers of cerebrovascular and Alzheimer’s diseases. Brain,
McGuinness, B., et al. (2010). Statins for the treatment of dementia. Cochrane 134, 3687–96.
Database of Systematic Reviews, (8), CD007514. Murray, M.D., et al. (2002). Preservation of cognitive function
McGurn, B., Deary, I.J., and Starr, J.M. (2008). Childhood cognitive ability with antihypertensive medications: a longitudinal analysis of a
and risk of late-onset alzheimer and vascular dementia. Neurology, 71, community-based sample of African Americans. Archives of Internal
1051–6. Medicine, 162, 2090–6.
McMahon, J.A., et al. (2006). A controlled trial of homocysteine lowering and Mutter, J., et al. (2004). Alzheimer disease: mercury as pathogenetic factor
cognitive performance. New England Journal of Medicine, 354, 2764–72. and apolipoprotein E as a moderator. Neuroendocrinology Letters, 25,
Mehlig, K., et al. (2008). Alcoholic beverages and incidence of dementia: 331–9.
34-year follow-up of the prospective population study of women in Nelson, C., et al. (2009). Dietary folate, vitamin B-12, vitamin B-6 and
Goteborg. American Journal of Epidemiology, 167, 684–91. incident Alzheimer’s disease: the cache county memory, health and aging
Mehta, K.M., et al. (1999). Head trauma and risk of dementia and Alzheimer’s study. Journal of Nutrition, Health and Aging, 13, 899–905.
disease: the Rotterdam Study. Neurology, 53, 1959–62. Nelson, H.D., et al. (2002). Postmenopausal hormone replacement therapy:
Mehta, K.M., et al. (2008). Race/ethnic differences in AD survival in US scientific review. Journal of the American Medical Association, 288,
Alzheimer’s Disease Centers. Neurology, 70, 1163–70. 872–81.
Newman, A.B., et al. (2005). Dementia and Alzheimer’s disease incidence Plassman, B.L., et al. (2000). Documented head injury in early adulthood
in relationship to cardiovascular disease in the Cardiovascular Health and risk of Alzheimer’s disease and other dementias. Neurology, 55,
Study cohort. Journal of the American Geriatrics Society, 53, 1101–7. 1158–66.
Nicoll, J.A., Roberts, G.W., and Graham, D.I. (1995). Apolipoprotein E ε4 Plassman, B.L., et al. (2011). Incidence of dementia and cognitive impairment,
allele is associated with deposition of amyloid beta-protein following not dementia in the United States. Annals of Neurology, 70, 418–26.
head injury. Nature Medicine, 1, 135–7. Podewils, L.J., et al. (2005). Physical activity, APOE genotype, and dementia
Ninomiya, T., et al. (2011). Midlife and late-life blood pressure and dementia risk: findings from the Cardiovascular Health Cognition Study. American
in Japanese elderly: the Hisayama study. Hypertension, 58, 22–8. Journal of Epidemiology, 161, 639– 51.
Noale, M., et al. (2003). Dementia and disability: impact on mortality. The Poels, M.M., et al. (2012). Cerebral microbleeds are associated with worse
Italian Longitudinal Study on Aging. Dementia and Geriatric Cognitive cognitive function: the Rotterdam Scan Study. Neurology, 78, 326–33.
Disorders, 16, 7–14. Polidori, M.C., et al. (2006). Congestive heart failure and Alzheimer’s disease.
Notkola, I.L., et al. (1998). Serum total cholesterol, apolipoprotein E ε4 allele, Neurological Research, 28, 588–94.
and Alzheimer’s disease. Neuroepidemiology, 17, 14–20. Posner, H.B., et al. (2002). The relationship of hypertension in the elderly to
Nourhashemi, F., et al. (2003). Body mass index and incidence of dementia: AD, vascular dementia, and cognitive function. Neurology, 58, 1175–81.
the PAQUID study. Neurology, 60, 117–19. Power, M.C., et al. (2011). The association between blood pressure and
Novak, V. and Hajjar, I. (2010). The relationship between blood pressure and incident Alzheimer disease: a systematic review and meta-analysis.
cognitive function. Nature Review Cardiology, 7, 686–98. Epidemiology, 22, 646–59.
Ogunniyi, A., et al. (2011). Weight loss and incident dementia in elderly Yoruba Prince, M.J., et al. (1996). Is the cognitive function of older patients affected
Nigerians: a 10-year follow-up study. International Psychogeriatrics, 23, by antihypertensive treatment? Results from 54 months of the Medical
387–94. Research Council’s trial of hypertension in older adults. British Medical
Ohara, T., et al. (2011). Glucose tolerance status and risk of dementia in the Journal, 312, 801–5.
community: the Hisayama study. Neurology, 77, 1126–34. Prins, N.D., et al. (2004). Cerebral white matter lesions and the risk of
Orgogozo, J.M., et al. (1997). Wine consumption and dementia in the dementia. Archives of Neurology, 61, 1531–4.
elderly: a prospective community study in the Bordeaux area. Revue Profenno, L.A., Porsteinsson, A.P., and Faraone, S.V. (2010). Meta-analysis
Neurologique, 153, 185–92. of Alzheimer’s disease risk with obesity, diabetes, and related disorders.
Ott, A., et al. (1997). Atrial fibrillation and dementia in a population-based Biological Psychiatry, 67, 505–12.
study: the Rotterdam Study. Stroke, 28, 316–21. Qiu, C. (2011). Epidemiological findings of vascular risk factors in Alzheimer’s
Ott, A., et al. (1998). Smoking and risk of dementia and Alzheimer’s disease disease: implications for therapeutic and preventive intervention. Expert
in a population-based cohort study: the Rotterdam Study. Lancet, 351, Review Neurotherapeutics, 11, 1593–607.
1840–3. Qiu, C., et al. (2001). The influence of education on clinically diagnosed
Ott, A., et al. (1999). Diabetes mellitus and the risk of dementia: the Rotterdam dementia incidence and mortality data from the Kungsholmen Project.
Study. Neurology, 53, 1937–42 Archives of Neurology, 58, 2034–9.
Ownby, R.L., et al. (2006). Depression and risk for Alzheimer disease: Qiu, C., et al. (2003a). Lifetime principal occupation and risk of Alzheimer’s
systematic review, meta-analysis, and metaregression analysis. Archives disease in the Kungsholmen project. American Journal of Industrial
of General Psychiatry, 63, 530–8. Medicine, 43, 204–11.
Pantoni, L. (2010). Cerebral small vessel disease: from pathogenesis and Qiu, C., et al. (2003b). Low blood pressure and risk of dementia in the
clinical characteristics to therapeutic challenges. Lancet Neurology, 9, Kungsholmen project: a 6-year follow-up study. Archives of Neurology,
689–701. 60, 223–8.
Peila, R., Rodriguez, B.L., and Launer, L.J. (2002). Type 2 diabetes, APOE gene, Qiu, C., et al. (2003c). Combined effects of APOE genotype, blood pressure,
and the risk for dementia and related pathologies: the Honolulu-Asia and antihypertensive drug use on incident AD. Neurology, 61, 655–60.
Aging Study. Diabetes, 51, 1256–62. Qiu, C., et al. (2004a). Occupational exposure to electromagnetic fields and
Pendlebury, S.T. and Rothwell, P.M. (2009). Prevalence, incidence, and factors risk of Alzheimer’s disease. Epidemiology, 15, 687–94.
associated with pre-stroke and post-stroke dementia: A systematic review Qiu, C., et al. (2004b). Risk and protective effects of the APOE gene towards
and meta-analysis. Lancet Neurology, 8, 1006–18. Alzheimer’s disease in the Kungsholmen project: variation by age and
Perl, D.P. and Moalem, S. (2006). Aluminum and Alzheimer’s disease, a sex. Journal of Neurology, Neurosurgery and Psychiatry, 75, 828–33.
personal perspective after 25 years. Journal of Alzheimer’s Disease, 9, Qiu, C., Winblad B, and Fratiglioni L (2005). The age-dependent relation of
291–300. blood pressure to cognitive function and dementia. Lancet Neurology,
Peters, R., et al. (2008a). Incident dementia and blood pressure lowering in 4, 487–99.
the Hypertension in the Very Elderly Trial cognitive function assessment Qiu, C., Winblad B, and Fratiglioni L (2006a). Cerebrovascular disease,
(HYVET-COG): a double-blind, placebo controlled trial. Lancet APOE ε4 allele and cognitive decline in a cognitively normal population.
Neurology, 7, 683–9. Neurological Research, 28, 650–6.
Peters, R., et al. (2008b). Haemoglobin, anaemia, dementia and cognitive Qiu, C., et al. (2006b). Heart failure and risk of dementia and Alzheimer
decline in the elderly, a systematic review. BMC Geriatrics, 8, 18. disease: a population-based cohort study. Archives of Internal Medicine,
Peters, R., et al. (2008c). Alcohol, dementia and cognitive decline in the 166, 1003–8.
elderly: a systematic review. Age and Ageing, 37, 505–12. Qiu, C., Winblad, B., and Fratiglioni, L. (2009). Low diastolic pressure and
Peters, R., et al. (2008d). Smoking, dementia and cognitive decline in the risk of dementia in very old people: a longitudinal study. Dementia and
elderly: a systematic review. BMC Geriatrics, 8, 36. Geriatric Cognitive Disorders, 28, 213–19.
Petersen, R.C., et al. (2005). Vitamin E and donepezil for the treatment Qiu, C., et al. (2010a). Cerebral microbleeds, retinopathy, and dementia: the
of mild cognitive impairment. New England Journal of Medicine, 352, AGES-Reykjavik Study. Neurology, 75, 2221–8.
2379–88. Qiu, C., et al. (2010b). Vascular risk profiles for dementia and Alzheimer’s
Petitti, D.B., et al. (2005). Blood pressure levels before dementia. Archives of disease in very old people: a population-based longitudinal study. Journal
Neurology, 62, 112–16. of Alzheimer’s Disease, 20, 293–300.
Petrovitch, H., et al. (2000). Midlife blood pressure and neuritic plaques, Qiu, C., et al. (2012). Medial temporal lobe is vulnerable to vascular risk
neurofibrillary tangles, and brain weight at death: the HAAS. Neurobiology factors in men: a population-based study. European Journal of Neurology,
of Aging, 21, 57–62. 17 Jan.
Querfurth, H.W. and LaFerla, F.M. (2010). Alzheimer’s disease. New England Rorsman, B., Hagnell, O., and Lanke, J. (1986). Prevalence and incidence of
Journal of Medicine, 362, 329–44. senile and multi-infarct dementia in the Lundby Study: a comparison
Ravaglia, G., et al. (2002). Education, occupation, and prevalence of dementia: between the time periods 1947–1957 and 1957–1972. Neuropsychobiology,
findings from the Conselice study. Dementia and Geriatric Cognitive 15, 122–9.
Disorders, 14, 90–100. Rosengren, A., et al. (2005). Body mass index, other cardiovascular risk
Ravaglia, G., et al. (2005a). Incidence and etiology of dementia in a large factors, and hospitalization for dementia. Archives of Internal Medicine,
elderly Italian population. Neurology, 64, 1525–30. 165, 321–6.
Ravaglia, G., et al. (2005b). Homocysteine and folate as risk factors for Ross, G.W. and Petrovitch, H. (2001). Current evidence for neuroprotective
dementia and Alzheimer disease. American Journal of Clinical Nutrition, effects of nicotine and caffeine against Parkinson’s disease. Drugs and
82, 636–43. Aging, 18, 797–806.
Ravaglia, G., et al. (2008). Physical activity and dementia risk in the elderly: Rovio, S., et al. (2005). Leisure-time physical activity at midlife and the risk of
findings from a prospective Italian study. Neurology, 70, 1786–94. dementia and Alzheimer’s disease. Lancet Neurology, 4, 705–11.
Rea, T.D., et al. (2005). Statin use and the risk of incident dementia: the Rovio, S., et al. (2010). The effect of midlife physical activity on structural
Cardiovascular Health Study. Archives of Neurology, 62, 1047–51. brain changes in the elderly. Neurobiology of Aging, 31, 1927–36.
Reed, B.R., et al. (2010). Measuring cognitive reserve based on the Ruitenberg, A., et al. (2001). Blood pressure and risk of dementia: results
decomposition of episodic memory variance. Brain, 133, 2196–209. from the Rotterdam study and the Gothenburg H-70 Study. Dementia
Reitz, C., et al. (2004). Relation of plasma lipids to Alzheimer disease and and Geriatric Cognitive Disorders, 12, 33–9.
vascular dementia. Archives of Neurology, 61, 705–14. Ruitenberg, A., et al. (2002). Alcohol consumption and risk of dementia: the
Reitz, C., et al. (2006). Stroke and memory performance in elderly persons Rotterdam Study. Lancet, 359, 281–6.
without dementia. Archives of Neurology, 63, 571–6. Ruitenberg, A., et al. (2005). Cerebral hypoperfusion and clinical onset of
Reitz, C., et al. (2007). Relation between smoking and risk of dementia and dementia: the Rotterdam Study. Annals of Neurology, 57, 789–94.
Alzheimer disease: the Rotterdam Study. Neurology, 69, 998–1005. Rusanen, M., et al. (2010). Midlife smoking, apolipoprotein E and risk of
Reitz, C., et al. (2010). A summary risk score for the prediction of Alzheimer dementia and Alzheimer’s disease: a population-based cardiovascular
disease in elderly persons. Archives of Neurology, 67, 835–41. risk factors, aging and dementia study. Dementia and Geriatric Cognitive
Reitz, C., Brayne, C., and Mayeux, R. (2011). Epidemiology of Alzheimer Disorders, 30, 277–84.
disease. Nature Review Neurology, 7, 137–52. Rusanen, M., et al. (2011). Heavy smoking in midlife and long-term risk of
Richards, M. and Deary, I.J. (2005). A life course approach to cognitive Alzheimer disease and vascular dementia. Archives of Internal Medicine,
reserve: a model for cognitive aging and development? Annals of 171, 333–9.
Neurology, 58, 617–22. Ryan, J., et al. (2009). Characteristics of hormone therapy, cognitive function,
Richards, M. and Hatch, S.L. (2011). A life course approach to the development and dementia: the prospective 3C Study. Neurology, 73, 1729–37.
of mental skills. Journals of Gerontology Series B-Psychological Sciences Saczynski, J.S., et al. (2006). The effect of social engagement on incident
and Social Sciences, 66 (Suppl 1), i26–35. dementia: the Honolulu-Asia Aging Study. American Journal of
Rippon, G.A., et al. (2006). Familial Alzheimer disease in Latinos: interaction Epidemiology, 163, 433–40.
between APOE, stroke, and estrogen replacement. Neurology, 66, 35–40. Saczynski, J.S., et al. (2010). Depressive symptoms and risk of dementia: the
Ritchie, K., et al. (2010). Designing prevention programmes to reduce Framingham Heart Study. Neurology, 75, 35–41.
incidence of dementia: prospective cohort study of modifiable risk Sander, K., et al. (2010). Carotid- intima media thickness is independently
factors. British Medical Journal, 341, c3885. associated with cognitive decline: the INVADE study. International
Roberts, R.O., et al. (2006). Postmenopausal estrogen therapy and Alzheimer Journal of Geriatric Psychiatry, 25, 389–94.
disease: overall negative findings. Alzheimer Disease and Associated Sano, M., et al. (2011). A randomized, double-blind, placebo-controlled trial
Disorders, 20, 141–6. of simvastatin to treat Alzheimer disease. Neurology, 77, 556–63.
Rocca, W.A., et al. (1998). Incidence of dementia and Alzheimer’s disease: Savva, G.M., and Stephan BCM (2010). Epidemiological studies of the effect
a reanalysis of data from Rochester, Minnesota, 1975–1984. American of stroke on incident dementia: A systematic review. Stroke, 41, E41–6.
Journal of Epidemiology, 148, 51–62. SBU (Statens beredning för medicinsk utvärdering) rapport (2006).
Rocca, W.A., et al. (2011). Trends in the incidence and prevalence of Demenssjukdomar: En systematisk litteraturöversikt (in Swedish), pp.
Alzheimer’s disease, dementia, and cognitive impairment in the United 22–9. SBU, Stockholm.
States. Alzheimer’s and Dementia, 7, 80–93. Scazufca, M., et al. (2008). Risk factors across the life course and dementia
Rockwood, K., et al. (2002). Use of lipid-lowering agents, indication bias, and in a Brazilian population: results from the Sao Paulo Ageing and Health
the risk of dementia in community-dwelling elderly people. Archives of Study (SPAH). International Journal of Epidemiology, 37, 879–90.
Neurology, 59, 223–7. Scarmeas, N., et al. (2006a). Mediterranean diet, Alzheimer disease, and
Rodriguez, E.G., (2002). Use of lipid-lowering drugs in older adults with and vascular mediation. Archives of Neurology, 63, 1709–17
without dementia: a community-based epidemiological study. Journal of Scarmeas, N., et al. (2006b). Mediterranean diet and risk for Alzheimer’s
the American Geriatrics Society, 50, 1852–6. disease. Annals of Neurology, 59, 912–21.
Roe, C.M., et al. (2007). Education and Alzheimer disease without dementia: Scarmeas, N., et al. (2009). Physical activity, diet, and risk of Alzheimer
support for the cognitive reserve hypothesis. Neurology, 68, 223–8. disease. Journal of the American Medical Association, 302, 627–37.
Roe, C.M., et al. (2008). Alzheimer disease and cognitive reserve: variation of Schaefer, E.J., et al. (2006). Plasma phosphatidylcholine docosahexaenoic acid
education effect with carbon 11-labeled Pittsburgh Compound B uptake. content and risk of dementia and Alzheimer disease: the Framingham
Archives of Neurology, 65, 1467–71. Heart Study. Archives of Neurology, 63, 1545–50.
Roher, A.E., et al. (2003). Circle of Willis atherosclerosis is a risk factor for Schafer, J.H., et al. (2005). Homocysteine and cognitive function in a
sporadic Alzheimer’s disease. Arteriosclerosis, Thrombosis, and Vascular population-based study of older adults. Journal of the American Geriatrics
Biology, 23, 2055–62. Society, 53, 381–8.
Romas, S.N., et al. (1999). APOE genotype, plasma lipids, lipoproteins, and Schmand, B., et al. (1997). Low education is a genuine risk factor for accelerated
AD in community elderly. Neurology, 53, 517–21. memory decline and dementia. Journal of Clinical Epidemiology, 50,
Röösli, M., et al. (2007). Mortality from neurodegenerative disease and exposure 1025–33.
to extremely low-frequency magnetic fields: 31 years of observations on Schmidt, M., et al. (2012). 25 year trends in first time hospitalisation for acute
Swiss railway employees. Neuroepidemiology, 28, 197–206. myocardial infarction, subsequent short and long term mortality, and the
prognostic impact of sex and comorbidity: a Danish nationwide cohort Solomon, A. and Kivipelto, M. (2009). Cholesterol-modifying strategies for
study. British Medical Journal, 344, e356. Alzheimer’s disease. Expert Review Neurotherapeutics 9, 695–709.
Schmidt, R., et al. (2002). Early inflammation and dementia: a 25-year Song, X., Mitnitski, A., and Rockwood, K. (2011). Nontraditional risk factors
follow-up of the Honolulu-Asia Aging Study. Annals of Neurology, 52, combine to predict Alzheimer disease and dementia. Neurology, 77,
168–74. 227–34.
Schneider, J.A., et al. (2007a). Mixed brain pathologies account for most dementia Sonnen, J.A., et al. (2010). Nonsteroidal anti-inflammatory drugs are
cases in community-dwelling older persons. Neurology, 69, 2197–204. associated with increased neuritic plaques. Neurology, 75, 1203–10.
Schneider, J.A., et al. (2007b). Subcortical infarcts, Alzheimer’s disease Sparks, D.L., et al. (1995). Increased incidence of neurofibrillary tangles
pathology, and memory function in older persons. Annals of Neurology, (NFT) in non-demented individuals with hypertension. Journal of the
62, 59–66. Neurological Sciences, 131, 162–9.
Schofield, P.W., et al. (1997). Alzheimer’s disease after remote head injury: Staessen, J.A., et al. (2011). Placebo-controlled trials of blood pressure-lowering
an incidence study. Journal of Neurology, Neurosurgery and Psychiatry, therapies for primary prevention of dementia. Hypertension, 57, e6–7.
62, 119–24. Starkstein, S.E. and Jorge, R. (2005). Dementia after traumatic brain injury.
Schrijvers, E.M., et al. (2012). Genome-wide association study of vascular International Psychogeriatrics, 17 Suppl 1, S93–107.
dementia. Stroke, 43, 315–19. Steenland, K., et al. (2009). Recent trends in Alzheimer disease mortality
Seeman, T.E. and Crimmins, E. (2001). Social environment effects on health in the United States, 1999 to 2004. Alzheimer Disease and Associated
and aging: integrating epidemiologic and demographic approaches and Disorders, 23, 165–70.
perspectives. Annals of the New York Academy of Sciences, 954, 88–117. Stephan, B.C., et al. (2012). Neuropathological profile of mild cognitive
Sekita, A., et al. (2010). Trends in prevalence of Alzheimer’s disease and impairment from a population perspective. Alzheimer Disease and
vascular dementia in a Japanese community: the Hisayama Study. Acta Associated Disorders, 26, 205–12.
Psychiatrica Scandinavica, 122, 319–25. Stern, Y. (2002). What is cognitive reserve? Theory and research application
Seshadri, S. and Wolf, P.A. (2007). Lifetime risk of stroke and dementia: of the reserve concept. Journal of International Neuropsychological
current concepts, and estimates from the Framingham Study. Lancet Society, 8, 448–60.
Neurology, 6, 1106–14. Stern, Y. (2006). Cognitive reserve and Alzheimer disease. Alzheimer Disease
Seshadri, S., et al. (2001). Postmenopausal estrogen replacement therapy and and Associated Disorders, 20, S69–74.
the risk of Alzheimer disease. Archives of Neurology, 58, 435–40. Stern, Y., et al. (1994). Influence of education and occupation on the incidence
Seshadri, S., et al. (2002). Plasma homocysteine as a risk factor for dementia of Alzheimer’s disease. Journal of the American Medical Association, 271,
and Alzheimer’s disease. New England Journal of Medicine, 346, 476–83. 1004–10.
Shah, R.C., et al. (2011). Hemoglobin level in older persons and incident Stewart, R., et al. (2005). A 32-year prospective study of change in body
Alzheimer disease: prospective cohort analysis. Neurology, 77, 219–26. weight and incident dementia: the Honolulu-Asia Aging Study. Archives
Sharp, E.S. and Gatz, M. (2011). Relationship between education and of Neurology, 62, 55–60.
dementia: an updated systematic review. Alzheimer Disease and Stewart, R., et al. (2007). Twenty-six-year change in total cholesterol levels
Associated Disorders, 25, 289–304. and incident dementia: the Honolulu-Asia Aging Study. Archives of
Sharp, S.I., et al. (2011). Hypertension is a potential risk factor for vascular Neurology, 64, 103–7.
dementia: systematic review. International Journal of Geriatric Psychiatry, Stewart, W.F., et al. (1997). Risk of Alzheimer’s disease and duration of
26, 661–9. NSAID use. Neurology, 48, 626–32.
SHEP Cooperative Research Group (1991). Prevention of stroke by Strozyk, D., et al. (2010). Contribution of vascular pathology to the clinical
antihypertensive drug treatment in older persons with isolated systolic expression of dementia. Neurobiology of Aging, 31, 1710–20.
hypertension: final results of the Systolic Hypertension in the Elderly Program Swardfager, W., et al. (2010). A meta-analysis of cytokines in Alzheimer’s
(SHEP). Journal of the American Medical Association, 265, 3255–64. disease. Biological Psychiatry, 15(68), 930–41.
Shepardson, N.E., Shankar, G.M., and Selkoe, D.J. (2011a). Cholesterol level Szekely, C.A., et al. (2004). Nonsteroidal anti-inflammatory drugs for the
and statin use in Alzheimer disease: I. Review of epidemiological and prevention of Alzheimer’s disease: a systematic review. Neuroepidemiology,
preclinical studies. Archives of Neurology, 68, 1239–44. 23, 159–69.
Shepardson, N.E., Shankar, G.M., and Selkoe, D.J. (2011b). Cholesterol level Taaffe, D.R., et al. (2008). Physical activity, physical function, and incident
and statin use in Alzheimer disease: II. Review of human trials and dementia in elderly men: the Honolulu-Asia Aging Study. Journals
recommendations. Archives of Neurology, 68, 1385–92. of Gerontology Series A-Biological Sciences and Medical Sciences, 63,
Shumaker, S.A., et al. (2004). Conjugated equine estrogens and incidence of 529–35.
probable dementia and mild cognitive impairment in postmenopausal Tan, Z.S., et al. (2003). Plasma total cholesterol level as a risk factor for
women: Women’s Health Initiative Memory Study. Journal of the Alzheimer disease: the Framingham Study. Archives of Internal Medicine,
American Medical Association, 291, 2947–58. 163, 1053–7.
Sims-Robinson, C., et al. (2010). How does diabetes accelerate Alzheimer Tang, M.X. et al. (1996). Effect of oestrogen during menopause on risk and
disease pathology? Nature Review Neurology, 6, 551–9. age at onset of Alzheimer’s disease. Lancet, 348, 429–32.
Skoog, I., et al. (1996). 15-year longitudinal study of blood pressure and Taylor, D.H. Jr, Sloan, F.A., and Doraiswamy, P.M. (2004). Marked increase
dementia. Lancet, 347, 1141–5. in Alzheimer’s disease identified in medicare claims records between
Snowdon, D.A., et al. (1997). Brain infarction and the clinical expression 1991 and 1999. Journals of Gerontology Series A-Biological Sciences and
of Alzheimer disease: the Nun Study. Journal of the American Medical Medical Sciences, 59, 762–6.
Association, 277, 813–7. Tschanz, J.T., et al. (2004). Dementia: the leading predictor of death in a
Sobel, E., et al. (1995). Occupations with exposure to electromagnetic defined elderly population: the Cache County Study. Neurology, 62,
fields: a possible risk factor for Alzheimer’s disease. American Journal of 1156–62.
Epidemiology, 142, 515–24. Tucker, A.M. and Stern, Y. (2011). Cognitive reserve in aging. Current
Sobel, E., et al. (1996). Elevated risk of Alzheimer’s disease among workers Alzheimer Research, 8, 354–60.
with likely electromagnetic field exposure. Neurology, 47, 1477–81. Tyas, S.L., et al. (2003). Mid-life smoking and late-life dementia: the
Solé-Padullés, C., et al. (2009). Brain structure and function related to Honolulu-Asia Aging Study. Neurobiology of Aging, 24, 589–96.
cognitive reserve variables in normal aging, mild cognitive impairment Tzourio, C., et al. (2003). Effects of blood pressure lowering with perindopril
and Alzheimer’s disease. Neurobiology of Aging, 30, 1114–24. and indapamide therapy on dementia and cognitive decline in patients
with cerebrovascular disease. Archives of Internal Medicine, 163, Wendell, C.R., et al. (2009). Carotid intimal medial thickness predicts
1069–75. cognitive decline among adults without clinical vascular disease. Stroke,
Valenzuela, M.J. (2008). Brain reserve and the prevention of dementia. 40, 3180–5.
Current Opinion in Psychiatry, 21, 296–302. Weuve, J., et al. (2004). Physical activity, including walking, and cognitive
Valenzuela, M.J. and Sachdev, P. (2006). Brain reserve and dementia: a function in older women. Journal of the American Medical Association,
systematic review. Psychological Medicine, 36, 441–54. 292, 1454–61.
Valenzuela, M. and Sachdev, P. (2009). Can cognitive exercise prevent the Weuve, J., et al. (2012). Accounting for bias due to selective attrition: the
onset of dementia? Systematic review of randomized clinical trials with example of smoking and cognitive decline. Epidemiology, 23, 119–28.
longitudinal follow-up. American Journal of Geriatric Psychiatry, 17, Weyerer, S., et al. (2011). Current alcohol consumption and its relationship to
179–87. incident dementia: results from a 3-year follow-up study among primary
Valenzuela, M., et al. (2011). Cognitive lifestyle and long-term risk of care attenders aged 75 years and older. Age and Ageing, 40, 456–63.
dementia and survival after diagnosis in a multicenter population-based Whalley, L.J. (2012). Spatial distribution and secular trends in the
cohort. American Journal of Epidemiology, 173, 1004–12. epidemiology of Alzheimer’s disease. Neuroimaging Clinics of North
Valenzuela, M.J., et al. (2012). Multiple biological pathways link cognitive America, 22, 1–10.
lifestyle to protection from dementia. Biological Psychiatry, 71, 783–91. Whalley, L.J., et al. (2000). Childhood mental ability and dementia. Neurology,
Van Den Heuvel, C., Thornton, E., and Vink, R. (2007). Traumatic brain 55, 1455–9.
injury and Alzheimer’s disease: a review. Progress in Brain Research, 161, Whalley, L.J., Dick, F.D., and McNeill, G. (2006). A life-course approach to
303–16. the aetiology of late-onset dementias. Lancet Neurology, 5, 87–96.
van der Flier, W.M., et al. (2011). Early-onset versus late-onset alzheimer’s White, L. (2009). Brain lesions at autopsy in older Japanese-American men
disease: the case of the missing APOE ɛ4 allele. Lancet Neurology, 10, as related to cognitive impairment and dementia in the final years of
280–8. life: a summary report from the Honolulu-Asia aging study. Journal of
van Elderen, S.G., et al. (2010). Progression of brain atrophy and cognitive Alzheimer’s Disease, 18, 713–25.
decline in diabetes mellitus: a 3-year follow-up. Neurology, 75, Whitmer, R.A., et al. (2005a). Obesity in middle age and future risk of
997–1002. dementia: a 27 year longitudinal population based study. British Medical
van Exel, E., et al. (2003). Interaction of atherosclerosis and inflammation in Journal, 330, 1360.
elderly subjects with poor cognitive function. Neurology, 61, 1695–701. Whitmer, R.A., et al. (2005b). Midlife cardiovascular risk factors and risk of
van Oijen, M., et al. (2007). Atherosclerosis and risk for dementia. Annals of dementia in late life. Neurology, 64, 277–81.
Neurology, 61, 403–10. Whitmer, R.A., et al. (2008). Central obesity and increased risk of dementia
Velayudhan, L., et al. (2010). Risk of developing dementia in people with more than three decades later. Neurology, 71, 1057–64.
diabetes and mild cognitive impairment. British Journal of Psychiatry, Whitmer, R.A., et al. (2011). Timing of hormone therapy and dementia: the
196, 36–40. critical window theory revisited. Annals of Neurology, 69, 163–9.
Vemuri, P., et al. (2011). Cognitive reserve and Alzheimer’s disease biomarkers Wilson, R.S., et al. (2002a). Depressive symptoms, cognitive decline, and risk
are independent determinants of cognition. Brain, 134(Pt 5), 1479–92. of AD in older persons. Neurology, 59, 364–70.
Verghese, J., et al. (2003a). Low blood pressure and the risk of dementia in Wilson, R.S., et al. (2002b). Cognitive activity and incident AD in a
very old individuals. Neurology, 61, 1667–72. population-based sample of older persons. Neurology, 59, 1910–14.
Verghese, J., et al. (2003b). Leisure activities and the risk of dementia in the Wilson, R.S., et al. (2002c). Participation in cognitively stimulating activities
elderly. New England Journal of Medicine, 348, 2508–16. and risk of incident Alzheimer disease. Journal of the American Medical
Verghese, P.B., Castellano, J.M., and Holtzman, D.M. (2011). Apolipoprotein Association, 287, 742–8.
E in Alzheimer’s disease and other neurological disorders. Lancet Wilson, R., Barnes, L., and Bennett, D. (2003). Assessment of lifetime
Neurology, 10, 241–52. participation in cognitively stimulating activities. Journal of Clinical and
Vermeer, S.E., et al. (2003). Silent brain infarcts and the risk of dementia and Experimental Neuropsychology, 25, 634–42.
cognitive decline. New England Journal of Medicine, 348, 1215–22. Wimo, A., et al. (2003). The magnitude of dementia occurrence in the world.
Vermeer, S.E., Longstreth, W.T. Jr, and Koudstaal, P.J. (2007). Silent brain Alzheimer Disease and Associated Disorders, 17, 63–7.
infarcts: a systematic review. Lancet Neurology, 6, 611–19. Witthaus, E., et al. (1999). Burden of mortality and morbidity from dementia.
Vidal, J.S., et al. (2010). Coronary artery calcium, brain function and Alzheimer Disease and Associated Disorders, 13, 176–81.
structure: the AGES-Reykjavik Study. Stroke, 41, 891–7. Wolfson, C., et al. (2001). A reevaluation of the duration of survival after the
Viswanathan, A., Rocca, W.A., and Tzourio, C. (2009). Vascular risk factors onset of dementia. New England Journal of Medicine, 344, 1111–16.
and dementia: how to move forward? Neurology, 72, 368–74. Xie, J., Brayne, C., and Matthews, F.E. (2008). Survival times in people with
Wald, D.S., Kasturiratne, A., and Simmonds, M. (2011). Serum homocysteine dementia: analysis from population based cohort study with 14 year
and dementia: meta-analysis of eight cohort studies including 8669 follow-up. British Medical Journal, 336, 258–62.
participants. Alzheimer’s and Dementia, 7, 412–17. Xu, W., et al. (2004). Diabetes mellitus and risk of dementia in the
Wang, H.X., et al. (1999). Smoking and the occurrence of Alzheimer’s Kungsholmen project: a 6-year follow-up study. Neurology , 63,
disease: cross-sectional and longitudinal data in a population-based 1181–6.
study. American Journal of Epidemiology, 149, 640–4. Xu, W., et al. (2007). The effect of borderline diabetes on the risk of dementia
Wang, H.X., et al. (2001). Vitamin B(12) and folate in relation to the and Alzheimer’s disease. Diabetes, 56, 211–16.
development of Alzheimer’s disease. Neurology, 56, 1188–94. Xu, W., et al. (2009). Mid- and late-life diabetes in relation to the risk of
Wang, H.X., et al. (2002). Late-life engagement in social and leisure activities dementia: a population-based twin study. Diabetes, 58, 71–7.
is associated with a decreased risk of dementia: a longitudinal study Xu, W., et al. (2010). Accelerated progression from mild cognitive impairment
from the Kungsholmen project. American Journal of Epidemiology, 155, to dementia in people with diabetes. Diabetes, 59, 2928–35.
1081–7. Xu, W., et al. (2011). Midlife overweight and obesity increase late-life
Wang, Y., et al. (2010). A five-year community-based longitudinal survival dementia risk: a population-based twin study. Neurology, 76, 1568–74.
study of dementia in Beijing, China: a 10/66 Dementia Research Group Yaffe, K., et al. (2011). Advanced glycation end product level, diabetes, and
population-based study. International Psychogeriatrics, 22, 761–8. accelerated cognitive aging. Neurology, 77, 1351–6.
Waring, S.C., et al. (1999). Postmenopausal estrogen replacement therapy Yasar, S., et al. (2005). Calcium channel blockers and risk of AD: the Baltimore
and risk of AD: a population-based study. Neurology, 52, 965–70. Longitudinal Study of Aging. Neurobiology of Aging, 26, 157–63.
Yeo, R.A., Arden, R., and Jung, R.E. (2011). Alzheimer’s disease and Zandi, P.P., et al. (2005). Do statins reduce risk of incident dementia and
intelligence. Current Alzheimer Research, 8, 345–53. Alzheimer disease? The Cache County Study. Archives of General
Yip, A.G., et al. (2002). An investigation of ACE as a risk factor for dementia Psychiatry, 62, 217–24.
and cognitive decline in the general population. Journal of Medical Zeki Al Hazzouri, A., et al. (2011). Life-course socioeconomic position and
Genetics, 39, 403–6. incidence of dementia and cognitive impairment without dementia in
Yip, A.G., Brayne, C., and Matthews, F.E. (2006). Risk factors for incident older Mexican Americans: results from the Sacramento area Latino study
dementia in England and Wales: the Medical Research Council Cognitive on aging. American Journal of Epidemiology, 173, 1148–58.
Function and Ageing Study: a population-based nested case-control Zhang, M.Y., et al. (1990). The prevalence of dementia and Alzheimer’s
study. Age and Ageing, 35, 154–60. disease in Shanghai, China: impact of age, gender, and education. Annals
Yoshitake, T., et al. (1995). Incidence and risk factors of vascular dementia of Neurology, 27, 428–37.
and Alzheimer’s disease in a defined elderly Japanese population: the Zhang, Z.X., et al. (2005). Dementia subtypes in China: prevalence in Beijing,
Hisayama Study. Neurology, 45, 1161–8. Xian, Shanghai, and Chengdu. Archives of Neurology, 62, 447–53.
Zandi, P.P., et al. (2002). Hormone replacement therapy and incidence of Zhu, L., et al. (2000). Incidence of dementia in relation to stroke and the
Alzheimer disease in older women: the Cache County Study. Journal of apolipoprotein E ε4 allele in the very old: findings from a population-based
the American Medical Association, 288, 2123–9. longitudinal study. Stroke, 31, 53–60.
Zandi, P.P., et al. (2004). Reduced risk of Alzheimer disease in users of Ziegler-Graham, K., et al. (2008). Worldwide variation in the doubling time
antioxidant vitamin supplements: the Cache County Study. Archives of of Alzheimer’s disease incidence rates. Alzheimer’s and Dementia, 4,
Neurology, 61, 82–8. 316–23.
CHAPTER 32
Mild Cognitive Impairment
and predementia syndromes
John T. O’Brien and Louise Grayson
Mild Cognitive Impairment has become a widely used term to of ageing. In particular, age-related changes predominantly affect
describe a condition or conditions where subjects have recognizable certain cognitive domains or components more than others. For
degrees of objective cognitive impairment that fall short of current example, at a global cognitive level, intelligence is often divided into
standardized definitions for either a dementia syndrome in general, crystallized intelligence, which represents the accrual of informa-
or particular disorders such as Alzheimer’s disease (AD), demen- tion over time, and fluid intelligence, which reflects the ability to
tia with Lewy bodies (DLB), or frontotemporal dementia (FTD). acquire or use new information. Crystallized intelligence, which
However, whether Mild Cognitive Impairment is best seen as a dis- is measured by tasks involving vocabulary and knowledge-based
crete syndrome, part of a continuum with normal ageing, or, as is abilities, tends to be relatively preserved with age. In contrast, fluid
increasingly argued, the early expression of various forms of progres- intelligence, which is assessed by tasks such as problem-solving and
sive neurological problems that have yet to fully manifest themselves speed of performance, tends to show an age-related decline (Schaie,
remains an area of ongoing debate. Whilst some widely accepted 1989). More discrete cognitive functions such as memory also show
definitions of Mild Cognitive Impairment syndromes have emerged, specific age-related differences. For example, semantic memory, or
it is important to note that Mild Cognitive Impairment does not memory regarding knowledge of facts about the world, is well pre-
appear in the current international classifications such as DSM-IV served with ageing, whilst declarative memory (representing the
(American Psychiatric Association, 1994) and ICD-10 (WHO, ability to learn new information) tends to decline with age. The rate
1992). Of note, DSM-V is under review and in its current draft form at which cognition declines with age is highly variable, with some
it is proposed that the revised version will include the category of subjects showing little age-related decline, others quite a substan-
Mild Neurocognitive Impairment, which is broadly similar to Mild tial drop, but with most people showing relatively stable cognitive
Cognitive Impairment as discussed here (<www.dsm5.org>). function until around the age of 60 (Hertzog and Schaie, 1988).
This chapter will summarize some of the key issues surrounding There continues to be a debate about whether dementia repre-
the historical development of Mild Cognitive Impairment, consider sents a continuum with normal ageing (a case often championed
the conceptual issues related to the use of the term as a diagnosis, by epidemiologists) or whether it represents a distinct and separate
summarize what is known regarding epidemiology, clinical features, disease process (Brayne and Calloway, 1988). Where those with
pathophysiology, prognosis, and therapeutics, and outline current Mild Cognitive Impairments should be positioned is clearly linked
clinical practice in the area. The chapter will conclude with a review with this debate. Since Mild Cognitive Impairments in general, or
of research developments and look at the recently proposed diag- MCI in particular, represents a state that falls short of dementia, in
nostic criteria, in particular the potential use of biomarkers to allow the continuum model it would simply be viewed as a milder degree
diagnosis of AD at an early, or Mild Cognitive Impairment, stage. of cognitive and functional disturbance than dementia, and would
There are difficulties with terminology in this area, and for the remain on a continuum with both ‘normal’ ageing and dementia.
purposes of this chapter, Mild Cognitive Impairment(s) will be However, if one accepts that dementia is a distinct disease from
used when referring to a more general description of mild prede- normal ageing, this leaves the status of MCI unclear. In this case,
mentia syndromes and Mild Cognitive Impairment (MCI) when it could either represent part of a continuum with ‘normal’ ageing
referring to the more specific syndrome described by Petersen and (as before), or be viewed as a separate ‘subsyndromal’ disorder with
colleagues and detailed in the next section. discrete clinical features and outcome. At the outset, it is important
to be clear that neither model may be right or wrong, and just as
Historical and Conceptual Development light can be viewed as a particle or a wave, both views of MCI may
have validity in certain circumstances. A parallel often used is that
Cognitive impairment and normal ageing of hypertension. Whilst blood pressure is undeniably normally dis-
An age-related decline in cognitive function has long been recog- tributed in a population, there are certain cut-offs (with definitions
nized and has been demonstrated in several longitudinal studies that can change over time in light of new research evidence) that
define a distinct group as ‘hypertensive’ or ‘hypotensive’. Although Similarly, is corroboration by an informant an essential part of diag-
recognizably part of a continuum, the validity for defining such nosis, and, if so, what happens when there is no informant avail-
subgroups at various levels of blood pressure comes from the fact able? To what extent are potential comorbid conditions that may
that such subgroups have a distinct outcome and prognosis (e.g. a be associated with cognitive disturbance excluded? What degree of
certain risk of developing cardiovascular and cerebrovascular dis- objective memory impairment is necessary, and how is this to be
ease) and for which certain interventions have proven efficacy in assessed—on one single test, or several; at one single time point, or
reducing this risk. Similarly, if one adopts the continuum view of at two or more? In the literature, most studies on MCI have adopted
MCI and dementia, there may still be merits in recognizing partic- a cut-off of 1.5 standard deviations or less to define those with MCI.
ular subgroups that are valid in terms of having a particular prog- However, as pointed out by Negash et al. (2005), most investiga-
nosis and/or therapeutic response. tors do not appear to realize that this cut-off was not actually the
one used by the Mayo Clinic group in their original description
Definitions of predementia syndromes to define the concept; it merely represented the mean memory
Those who champion the cause of those with Mild Cognitive impairment that they demonstrated in those defined (on clinical
Impairments as having a separate syndrome refer to its distinct grounds) as having MCI. As such, Negash et al. (2005) emphasize
clinical features, outcome, pathophysiology, and therapeutics, the importance of assessment of the memory complaint and objec-
which are detailed later in this chapter. However, the descriptive measure of dysfunction in the context of the individual—in
tion and characterization of syndromes of cognitive impairment, other words, not just what is statistically less than normal, but what
which fall short of dementia, is not new. Definitions of dementia is less than normal performance for the particular individual. This
have remained reasonably constant over the last decades (though will be particularly important for those of either high or low intel-
this is now changing; see Chapters 29 and 33, Part 1) and generally ligence or education. However, whilst making excellent clinical
require deficits in memory and one or more other higher cortical sense, it could be argued that such a departure from a strict cut-off
functions, together with evidence of decline from premorbid func- will represent a difficulty in reliably applying such criteria in differ-
tioning, which result in impairment in social and/or occupational ent centres, especially in a research context.
functioning. The very nature of defining dementia in this way MCI criteria require general cognitive function to be ‘essentially’
means that there will be those with cognitive difficulties that fall normal. This can allow those with MCI to have minor impair-
short of this definition, either because their cognitive deficit does ments, but if they had significant impairments in a second cogni-
not involve memory, or because they involve one rather than two tive domain then they may well fulfil criteria for dementia. The
cognitive domains, or because cognitive deficits are insufficient to need for clinical judgement rather than the use of cut-off scores is
cause social and/or occupational functioning. There have been sev- again highlighted in this regard by the Mayo Clinic group (Negash,
eral attempts over the last 40 years to categorize and define such 2005). Activity of daily living (ADL) performance is generally well
Mild Cognitive Impairments (O’Brien and Levy, 1992). These are preserved, but in a comparison of subjects with MCI compared to
summarized in Table 32.1. older subjects without MCI, minor functional impairments can be
The concept of MCI developed from the work of Petersen and demonstrated, albeit falling well short of those seen in people with
colleagues at the Mayo Clinic who undertook a series of care- dementia (Petersen, 2004). The issue of not being demented hinges
ful longitudinal studies on older people with cognitive problems on the extent of ADL impairment combined with assessment
(Petersen, 2004; Petersen et al., 1999). Although it is apparent that of whether two or more cognitive domains are affected. It is also
MCI represents yet another in a long line of attempts to define a important to note here that often in the assessment of those with
subclinical cognitive syndrome, it has gained much wider accept- MCI and early dementia, certain clinical dementia rating scales are
ance than previous concepts, to the extent that it has been the focus used, such as the Clinical Dementia Rating Scale (CDR) (Hughes
of large-scale research studies, including therapeutic randomized et al., 1982). Whilst these scales are helpful, they do not directly
controlled studies. MCI has gained such acceptance largely because map to a diagnosis, which can sometimes be a source of confusion.
of the strength of the evidence base, supporting it as a syndrome, For example, a CDR of 0.5 (labelled in the scale as ‘questionable
which had been lacking for previous concepts. For this reason, the dementia’) can be compatible with either the clinical syndrome of
rest of the chapter focuses more specifically on MCI. MCI or early AD.
Recognizing that memory is not the only cognitive domain
affected in those with Mild Cognitive Impairments, the concept of
Diagnostic Criteria MCI has since been broadened to include other subtypes, partic-
The original diagnostic criteria proposed by Petersen and col- ularly ‘non-amnestic’ MCI (Petersen, 2004; Winblad et al., 2004).
leagues (1999) had memory impairment as a core feature, a condi- In this scheme, MCI is divided into amnestic, where memory is
tion that has become known as the amnestic (or memory) form of impaired, and non-amnestic, where it is not, and then each subtype
MCI. These criteria are reproduced in Table 32.2. While these have is further divided into whether a single or more than one cognitive
been widely adopted, there is still no consensus in the field regard- domain is affected, giving four main types of MCI (amnestic sin-
ing a single set of criteria for MCI, though a broader set of crite- gle domain, amnestic multi-domain, non-amnestic single domain,
ria (encompassing non-amnestic problems and multiple domains) non-amnestic multi-domain). This is illustrated in Fig. 32.1. The
were recommended for use by an international consensus group arguments for such subdivision are: (1) this reflects the clinical real-
(Winblad et al., 2004). The necessity for a subjective memory com- ity of how patients present; and (2) the clinical syndromes will or
plaint to be present has been questioned, for as with dementia there may have different outcomes in terms of disease progression. As an
are likely to be subjects who have definite cognitive impairments approach to diagnosis this is illustrated in Fig. 32.2(a), which illus-
but do not complain of these. Do such subjects have MCI or not? trates a situation where different clinical syndromes are illustrated
CHAPTER 32 mild cognitive impairment and predementia syndromes 417
Table 32.1 Historical perspective on definitions of predementia syndromes

Terminology First described by Diagnostic criteria Difficulties encountered
Benign senescent Kral (1962) Referred to patients in a nursing home setting who Lack of standardized criteria to define
forgetfulness complained of difficulty with recall without clear the condition
objective evidence of impairment
Age-associated memory Crook et al. (1987) Changes in older people aged 50+ comprising of Normative data were for younger
impairment (AAMI) subjective complaints of memory loss verified by cohort
decrement of one standard deviation below means Overinclusive, e.g. captures people
established for younger adults with normal age-related decline and
classifies them as ‘abnormal’
Cognitive decline was confined to
memory domain
Late-life forgetfulness Blackford and La Rue Memory decline and deficits on four or more cognitive
(1989) tests
Ageing-associated WHO (1992); Levy (1994) Age-related cognitive decline for any task (memory, Targets a more severe state of
cognitive decline (ACCD) attention, learning, thinking, language, and visuospatial impairment than AAMI
function), allowing for age- and education-adjusted
normative values
Mild cognitive disorder ICD-10 (1993) Disorder of memory, learning, and concentration that Attributable to a general medical
is accompanied by mental fatigue, evidenced on formal condition
neurological testing, that is attributable to cerebral or
systemic disease
Mild neurocognitive DSM-IV (1994) Decline in two or more domains: memory, learning, Attributable to a general medical
impairment attention or speed of processing, language, or executive condition
function. Objective evidence of a neurological or general
medical disorder that is judged to be causal
Cognitive impairment no Canadian Study of Health Impairment in cognitive function with no established Derived from epidemiological sample
dementia and Ageing (Graham et al. dementia Includes people with lifelong
(1997) cognitive impairment (e.g. learning
disability, static encephalopathy)
Mild Cognitive Impairment Reisberg et al. (1988) Stage 3 on the Global Deterioration Scale et al., Criteria depend upon value on a
rating scale, which was not designed
as a diagnostic instrument
Mild Cognitive Impairment Mayo Clinic (Petersen Subjective memory impairment associated with Confined to memory domain
(MCI) et al. 1999) objective memory deficit, adjusted for age and
educational background, no dementia, preserved ADLs
MCI subtypes: De Kosky and Chertow Likely to progress to AD Subtypes require validation
Amnestic (2001) May represent normal ageing or progress to vascular
Multiple domain cognitive impairment or neurodegenerative disorder
Single domain May progress to FTD, DLB, or AD
non-amnestic
MCI—revised International Working The individual is neither normal nor demented,
Group on MCI (Winblad subjective and objective cognitive deficits over time,
et al., 2004) ADLs preserved, and complex instrumental functions
are either intact or minimally impaired
The future
Preclinical AD and Dubois et al. (2010) See text Research and clinical criteria
prodromal AD Biomarkers require validation
MCI due to AD Albert et al. (2011) See text Research and clinical criteria
Biomarkers require validation
Mild neurocognitive Proposed for DSM-V See text Defines cut-off scores on
disorder neurocognitive testing
AD, Alzheimer’s disease; ADLs, activities of daily living; DLB, dementia with Lewy bodies; FTD, frontotemporal dementia.
as having distinct courses. As can be seen, whilst amnestic MCI studies involving non-amnestic MCI. However it is possible to con-
may be particularly helpful in predicting AD, non-amnestic and sider other ways to classify MCI, such as those based on aetiology.
noncognitive types might have less in the way of positive predictive For example, one might consider degenerative MCI as being largely
power in that they will potentially represent early stages of a multi- due to early Alzheimer-type pathology and map in most cases to
tude of different underlying pathologies. The extent to which this the clinical type of amnestic MCI. Vascular MCI, which has been
is true remains uncertain, due to the lack of detailed longitudinal the subject of relatively little research attention, would reflect a
pattern of largely attentional and executive impairment with rela-
tively well-preserved memory in the presence of associated corti-
Table 32.2 Criteria for amnestic Mild Cognitive Impairment (a-MCI) cal and/or subcortical vascular changes on imaging (O’Brien et al.,
◆ Memory complaint usually corroborated by an informant 2003). As such, it would be expected to predominantly manifest as
◆ Objective memory impairment relative to age and educationally matched non-amnestic MCI. This different approach to early diagnosis is
healthy people shown in Fig. 32.2(b). Here, different ‘prodromal’ syndromes repre-
◆ Preserved general cognitive function
sent the early stages of the different underlying pathologies that may
be defined on a much broader basis than just cognition. So, for exam-
◆ Intact activities of daily living
ple, one might have a clinical syndrome (or syndromes) suggestive
◆ Not clinically demented
Cognitive complaint
Not normal for age

Not demented
Cognitive decline
Essentially normal functional activities
MCI
Memory impaired?
Yes No
Amnestic MCI Non-Amnestic MCI
Memory impairment Single non-memory

only? cognitive domain impaired?
Yes No Yes No
Amnestic MCI Amnestic MCI Non-Amnestic MCI Non-Amnestic MCI

Single Domain Multiple Domain Single Domain Multiple Domain
Fig. 32.1 Mild Cognitive Impairment. (Reproduced from Petersen, J Int Med, 2004.)
(B)
Amnestic MCI AD
(A)
Executive MCI, stroke, WMLs VaD
Amnestic MCI AD (mostly)

Language MCI, executive MCI,
behavioural changes FTD
Non-amnestic MCI AD, VaD, FTD, DLB
Attentional MCI, visuo-spatial MCI,

Non-cognitive AD, VaD, FTD, DLB psychosis, movement disorder,
presentations Depression, Psychosis falls/syncope confusional states DLB
Fig. 32.2 MCI subtypes and progression. WML, White matter lesions.
of early cerebrovascular cognitive impairment, early frontotemporal common in those with MCI than controls. Feldman et al. (2004)
disease, or early Lewy body disease. These would then represent the studied over 1000 subjects with MCI using the Neuropsychiatric
clinical entities that would be of interest for further assessment and Inventory, finding some neuropsychiatric symptoms in the major-
follow-up, rather than cognitive-based paradigms. Which of these ity (59%). In a more detailed analysis of a much smaller group of
approaches will ultimately have the greatest clinical utility remains 28 MCI subjects, Hwang et al. (2004) found that the most com-
to be seen. Progress has been made over the last 12 years since mon symptoms in the MCI group were dysphoria (39%), apathy
Petersen et al. (1999) first defined the concept of MCI, but there are (39%), irritability (29%), and anxiety (25%). There were significant
many challenges that remain. MCI is an evolving concept, which is differences in apathy, dysphoria, irritability, anxiety, agitation, and
the subject of ongoing research and debate. aberrant motor behaviour between the MCI and control groups.
In contrast, only delusions were significantly less common in MCI
Epidemiology compared with mild AD subjects. Similarly, Geda and colleagues
(2008) reported on the prevalence of neuropsychiatric symptoms in
There is ongoing debate regarding the reliability and validity of 319 individuals with MCI compared to 1590 subjects with normal
the syndrome of MCI, largely based on findings from longitudinal cognition in a population-based study. They found that approxi-
population-based epidemiological studies. One difficulty is that the mately half of the subjects with MCI and up to 25% of those with
definitions for MCI were developed in a clinic setting and based normal cognition reported at least one neuropsychiatric symp-
on a full and detailed psychiatric/neurological examination of indi- tom. The most distinguishing features between subjects with MCI
vidual subjects. As such, direct translation to epidemiological sam- and normal cognition were apathy (18.5%), depression (27.0%),
pling studies is inherently difficult, if not impossible. Most studies agitation (9.1%), anxiety (14.1%), and irritability (19.4%). They
have been forced to use a variety of adaptations of the MCI crite- observed that apathy, agitation, and irritability were higher in sub-
ria to fit with already existing datasets, and have shown prevalence jects with amnestic MCI than non-amnestic MCI, whereas anxi-
amongst older people to be between 3% and 19%, with incidence ety, depression, delusions, and disinhibition were more common in
of 8–58/1000 per year (Gauthier et al., 2006). The Mayo Clinic non-amnestic MCI. They therefore suggested that neuropsychiatric
study of ageing was designed as a population-based study involv- markers might predict progression to different types of dementia,
ing around 3000 participants aged 70–89 who were nondemented with apathy and irritability being more predictive of amnestic MCI
and cognitively normal or who had MCI at entry. The prevalence of that is likely to progress to AD, and delusions and disinhibition
MCI in this study was estimated at 15% of nondemented popula- more predictive of non-amnestic MCI and progression to nonAD
tion with amnestic MCI being twice as common as non-amnestic dementia. It is clear that MCI, even though defined cognitively,
MCI (Roberts et al., 2008). Risk of developing dementia varied involves more than just cognitive symptoms.
between 11 and 33% over 2 years, but more importantly findings
have demonstrated that between 20 and 50% of those with MCI at a
baseline assessment appear to return to normal a year later (Ritchie Pathophysiology
et al., 2001; Ganguli et al., 2004). Whilst some consider that such Neuroimaging changes
studies argue against the validity of the syndrome, others feel this
Neuroimaging has frequently been used as a tool to investigate brain
represents the fact that many factors affect cognitive performance
changes in different dementias, and so has naturally been applied
in older people apart from underlying brain dysfunction, includ-
to those with MCI to try to seek and understand what, if any, neu-
ing education, psychiatric comorbidity, physical illness, genetics,
robiological changes can be determined. Structural neuroimaging
hormonal changes, and the use of drugs, including anticholinergic
changes are well characterized in AD and include generalized brain
agents. Since many of these factors, and their influence, can change
atrophy with specific involvement of the entorhinal cortex and
over time, this might explain why many cases of MCI are reversible
hippocampus, with increased rates of atrophy on serial magnetic
in community settings (Gauthier et al., 2006). In contrast, it can
resonance (MR) imaging. Reduced hippocampal and/or entorhinal
be argued that those referred to memory clinics and other special-
cortex volume, together with increased rates of atrophy on serial
ized centres are unlike the general population in that they are actu-
MR, have also been shown to be predictive of the subsequent devel-
ally seeking services for a perceived memory problem that is likely
opment of dementia (Rusinek et al., 2003). Since the amnestic form
to have been persistent enough to have prompted referral and, as
of MCI bears a close relationship to AD, it might be expected that
such, incidence and prevalence rates in this setting are higher and
imaging changes of those with MCI may be similar to those with
more stable over time.
AD. This has indeed been shown to be the case, with those with
amnestic MCI showing evidence of mild hippocampal atrophy
Clinical Features compared to older controls, though not with such severe changes as
The primary feature of those with MCI is cognitive impairment, those with established AD. Killiany et al. (2002) used MRI to meas-
most usually but not always in memory. When the subjects with ure the volumes of the entorhinal cortex and hippocampus in 137
MCI were compared with those with very mild AD, memory per- individuals with MCI and found that the volume of the entorhinal
formance was similar, but patients with AD were more impaired in cortex distinguished subjects who developed dementia from those
other cognitive domains as well. ADL skills and functional abili- who did not, with considerable accuracy, whereas hippocampal
ties are largely intact in MCI subjects—part of the reason that they volume loss did not. DeToledo-Morrell et al. (2004) reported that
do not meet criteria for dementia—yet detailed study shows that right hemisphere entorhinal volume showed good predictive val-
those with MCI do have significant but mild impairments, espe- ues in conversion of MCI to AD with a concordance rate of 93.5%.
cially in instrumental (rather than basic) ADLs (Petersen, 2004). Some studies have found that the entorhinal cortex is a better dis-
Neuropsychiatric features have been investigated and are more criminator between those with MCI and normal controls than the
hippocampus (Du et al., 2001; Killiany et al., 2002), whilst others individuals meeting the criteria for amnestic MCI more likely to be
have found both structures to be equally helpful (Xu et al., 2000). PIB positive than those with non-amnestic MCI. Forsberg and col-
Studies investigating other imaging changes, such as MR spec- leagues (2008), in a longitudinal study of 21 individuals with MCI,
troscopy, have also shown changes. Kantarci et al. (2000) found that showed that 33% of the MCI subjects with positive PIB binding
myoinositol/creatinine ratios were increased in MCI subjects rela- converted to AD in the follow-up period, suggesting that amyloid
tive to normal controls, and in those with AD relative to those with imaging might provide prognostic information in MCI. Okello and
MCI. However, N-acetylaspartate (NAA)/creatinine ratios did not colleagues (2009) assessed amyloid burden using PIB in subjects
differ between MCI and controls, but were reduced in AD. Since with amnestic MCI. They followed up 31 subjects with MCI for 1–3
myoinositol reflects glial activity, it was suggested that this might be years to assess conversion rates to AD and compared baseline amy-
a more useful marker earlier in the disease process than NAA, which loid burden between converters and nonconverters. Seventeen of
is usually thought to reflect neuronal integrity. Ackl et al. (2005) their 31 subjects with MCI had increased (11) C-PIB retention at
showed regionally specific spectroscopy changes between MCI and baseline and 14 (82%) of these patients clinically converted to AD
AD. MCI subjects showed reduced NAA in the hippocampus of both during follow-up. Only one (11) C-PIB negative MCI subject (7%)
MCI and AD, but only parietal reductions in the AD group. Using converted to AD. Of the PIB-positive subjects with MCI, half had
a technique allowing whole-brain quantification of NAA, Falini converted to AD within 1 year of baseline PIB-PET; these subjects
et al. (2005) showed that both structural and metabolic findings of had higher levels of tracer retention than slower converters in the
those with MCI were midway between those of healthy volunteers anterior cingulate and frontal cortex. In vivo detection of amyloid
and those of established AD. As such, results of structural and spec- deposition in MCI with PIB-PET might therefore prove useful in
troscopic MR studies to date support the view that MCI has neu- identifying a subgroup of MCI patients who are at higher risk of
robiological changes consistent with the transitional state between progressing rapidly to dementia.
normal ageing and dementia, in this case AD. Similarly, both single To date, limited research has been performed on the neuroim-
photon emission computed tomography (SPECT) and photon emis- aging findings of non-amnestic forms of MCI. Several researchers
sion tomography (PET) studies have generally shown hypoperfusion have emphasized that vascular changes in the white matter may
in those with MCI in areas such as the posterior parietal cortex and be an important component of MCI, although further research
posterior cingulate areas, which are shown to be affected in early is required to determine their role and possible interaction with
AD (Chetelat et al., 2005; Pakrasi and O’Brien, 2005). Chelelat et al. degenerative pathology in the clinical expression of MCI (DeCarli
(2005) reported data from a longitudinal fluorodeoxyglucose PET et al., 2001; Medina et al., 2006). Certainly, the presence of white
study suggesting a high predictive rate of reduced uptake in the pari- matter lesions on MRI in older people, especially when severe (con-
etal association areas and a lower uptake in the posterior cingulate fluent), carries an adverse prognosis in terms of cognitive and func-
in patients with MCI who converted to AD. Hirao et al. (2005) used tional decline, and mortality (Inzitari et al., 2009).
SPECT imaging to compare 52 patients with MCI who converted to
AD (converters) with 24 nonconverters over a 3-year period, and Pathological studies
showed that converters had reductions in regional cerebral blood Detailed pathological studies of large numbers of subjects with MCI
flow (rCBF) in the bilateral parietal areas and the precunei when prospectively assessed during life are not available; indeed, given the
compared to nonconverters. The logistic regression model revealed modest mortality rate of those with MCI, such studies will prove
that reduced rCBF in the inferior parietal lobule, angular gyrus, and difficult to undertake. Initial studies on small numbers of subjects
precunei had high predictive values and discriminative ability of have suggested that individuals with MCI show an increased bur-
converters to nonconverters. In keeping with the structural imaging den of tau pathology in the form of neurofibrillary tangles in the
findings, functional imaging changes are somewhere between those mesial temporal structures, and this correlates well with progres-
seen in normal ageing and AD. sion of MCI (Mitchell et al., 2002). Other studies have shown that
Recent research has focused on the use of amyloid ligands amyloid load in the entorhinal cortex is intermediate between nor-
such as Pittsburgh Compound B ((11) C-PIB), which was the mal subjects and those with AD (Mufson et al., 1999; Bennett et al.,
first to be developed, to study amyloid burden using PET imag- 2002, 2005). Jicha et al. (2006) observed that approximately 70% of
ing in individuals with MCI as a potential early marker of AD the participants with amnestic MCI who later developed dementia
pathology. Amyloid imaging provides an in-vivo measure of one had AD pathology at post mortem, but 20–30% developed another
of the key pathologic hallmarks of AD, fibrillar amyloid β (Aβ) dementing disorder, indicating that while the clinical criteria for
plaques. Research has shown that amyloid deposition assessed amnestic MCI likely predict AD, they are not absolutely specific.
using PIB-PET is significantly elevated in AD when compared to Vascular pathology can also be an important cause of cognitive
controls (Klunk et al., 2004; Jack et al., 2008). Numerous studies impairments (O’Brien et al., 2003). Bennett et al. (2005) showed that
in MCI have shown that PIB uptake is intermediate between AD cerebrovascular disease, in addition to neurodegenerative pathol-
and controls (Lopresti et al., 2005; Jack et al., 2008; Mormino et al., ogy, was associated with MCI in the Religious Orders Study. Lewy
2009). However, in most studies, the distribution of PIB binding for body pathology was rarer, but still present in 8% of MCI cases.
the majority of cases with MCI shows AD-like uptake levels, with
a minority showing low control-level binding and a small number Neurochemistry
falling in between. Overall, 52–87% of MCI patients show elevated A central feature of many dementias, especially AD and dementia
PIB binding, depending upon the criteria used to diagnose MCI with Lewy bodies, is a substantial cholinergic deficit. Mufson et al.
and the threshold for defining PIB-positivity. Wolk et al. (2009) (2002) reported a 38% loss of cholinergic neurons from the nucleus
used PIB to determine the presence of AD pathology in different basalis of Meynert in those with MCI, similar to the reduction
MCI subgroups: 56% of their 26 subjects were PIB positive, with seen in early AD (43%). In contrast, Dekosky and colleagues found
evidence of increased cholinergic activity, with preserved choline 137 (out of 180) consecutively referred patients with MCI who had
acetyltransferase activity in the hippocampus and frontal cortex of lumbar puncture and CSF biomarkers (Aβ42, total tau, and p-tau)
older people with MCI (DeKosky et al., 2002). They suggested that at baseline and were subsequently followed up for 4–6 years to
this up-regulation may reflect a compensatory mechanism in the monitor progression to dementia. Thirty-nine healthy cognitively
face of declining cognitive ability. This is supported by other work stable individuals acted as control subjects. During follow-up, 57
suggesting that cells remaining in the nucleus basalis have enhanced (42%) of the patients with MCI developed AD, 21(15%) developed
metabolic activity in MCI, but not AD (Dubelaar et al., 2006), and another form of dementia, and 56 (41%) remained cognitively sta-
such a compensation has also been seen in some functional MRI ble. A combination of CSF Aβ42 and total tau at baseline yielded
(fMRI) studies that have paradoxically shown increased areas of a sensitivity of 95% and specificity of 83% for detection of AD
activation in those with MCI compared to both normal controls dementia in those with MCI. MCI cases that had pathological CSF
and those with AD (Dickerson et al., 2005). Studies investigating (low levels of a-beta and raised levels of total and p-tau) were at
synaptic proteins have shown some changes, which may reflect loss significantly greater risk of progression to AD. Rates were 27% per
of dendritic plasticity in MCI (Counts et al., 2006). Several stud- year if CSF was abnormal compared to 1% per year if CSF was nor-
ies have looked at α4β2 nicotinic acetylcholine receptor (nACHR) mal. The association between pathological CSF and progression to
function in MCI with conflicting results. Ellis et al. (2008) observed AD was much stronger than and independent of established risk
no significant reduction in α4β2 nACHR binding, as measured with factors, including age, sex, education, ApoEgenotype, and plasma
218F-FA ligand and PET in subjects with early AD compared to homocysteine. The combination of total tau and Aβ42/p-tau ratio
controls. Similarly, Mitsis et al. (2009) reported on 10 individu- yielded similar results. These biomarkers are strongly associated
als with MCI compared to 12 individuals with AD and 10 healthy with the development of AD in individuals with MCI.
controls using 123 I-5-IA-85380 SPECT imaging and found pres- Mattsson and colleagues (2009) investigated the diagnostic accu-
ervation of nACHRs during the prodromal and early stages of AD. racy of CSF biomarkers in a large multicentre study in patients
In contrast, Terriere et al. (2010) reported a pilot study, using 123 with MCI. The study consisted of two parts: a cross-sectional study
I-5-IA-85380 SPECT imaging, in individuals with amnestic MCI involving patients with AD and controls to identify cut-off points
and showed that individuals with MCI had discrete reductions in for biomarkers, followed by a prospective cohort study involving
tracer uptake in the medial temporal cortex when compared to patients with MCI. They followed up 750 individuals with MCI for
healthy controls. In summary, neurochemical changes suggestive at least 2 years. During the follow-up period, 271 (36%) individu-
of transmitter and synaptic alterations have been demonstrated in als with MCI converted to AD and 59 to other dementias (28 with
MCI, though unlike the pathological findings in which MCI is a vascular dementia, 14 with DLB, 7 with FTD, and 10 with other
‘halfway’ point between normal ageing and AD, they suggest some neurological conditions). Patients with MCI who had incipient
degree of compensatory up-regulation, presumably in an attempt AD had higher CSF levels of total tau and p-tau and lower levels
to overcome the decline in neuronal functioning consequent on the of Aβ42 compared to healthy controls, stable MCI cases, and MCI
increased burden of pathology. More research is needed in this area cases with other dementias. Aβ42 levels in MCI patients with Lewy
before definitive conclusions can be drawn. body disease did not differ significantly from those with MCI due
to AD. There was considerable intersite variability in biomarkers.
Genetic and other biomarkers The authors concluded that Aβ42, total tau, and p-tau could be
Genetic associations of MCI appear similar to AD, with increased used to predict with good accuracy which MCI patients were likely
rates of apolipoprotein ε4 (ApoE4) reported which may predict to progress to AD.
future cognitive decline (Farlow et al., 2004). In 2009, Mitchell undertook a meta-analysis of 51 studies that
A range of biomarkers have been developed for use in AD. These had looked at CSF p-tau in the diagnosis and prognosis of MCI
biomarkers reflect the underlying pathophysiological progress. and AD. P-tau was found to be moderately successful in predict-
The first category of biomarkers reflect amyloid beta (Aβ) protein ing progression to dementia in MCI (sensitivity 81.1%, specificity
deposition and include low cerebrospinal fluid (CSF) Aβ42 and 65.3%, positive predictive value 63%, and negative predictive value
positive PET amyloid imaging. The second category of biomark- 83%), showing higher predictive value for absence of progression
ers reflects downstream neuronal degeneration or injury (and, as rather than conversion to AD. They concluded that p-tau would be
such, are not specific for AD). These biomarkers are elevated CSF expected to facilitate 72 correct diagnoses for every 100 individuals
tau (both total tau and phosphorylated tau (p-tau)); decreased with MCI tested and as such would be useful to aid differentiation
18-fluorodeoxyglucose (FDG) uptake on PET in temporoparietal of MCI from healthy controls.
cortex; and disproportionate atrophy on structural MRI in medial, In summary, CSF Aβ42 and tau measures, the ratio of CSF tau/
basal, and lateral temporal lobe, and medial parietal cortex. The Aβ42, PET amyloid measures, and other biomarkers of neuronal
latter two are sometimes referred to as topographical markers. injury such as hippocampal atrophy and temporoparietal hypome-
These biomarkers have also been investigated for their prognos- tabolism have all been shown to predict progression of MCI to AD
tic value in MCI. Numerous studies have shown that AD patients dementia.
display characteristic CSF changes with elevated levels of total tau
protein and p-tau and decreased levels of β amyloid (Aβ42). Some
studies have shown that individuals with amnestic MCI show sim-
Limitations of Biomarkers
ilar changes (Andreasen et al., 2003; Hampel et al., 2004; Li et al., Biomarkers are currently being used in research settings to aid
2007). diagnosis of AD dementia. They are not as yet used in routine clini-
Hansson et al. (2006) investigated the association between CSF cal practice, although this may change in the future when there is a
biomarkers and incipient AD in patients with MCI. They assessed greater understanding of the validity and usefulness of individual
biomarkers. At present, there has been limited research compar- most susceptible to returning to normal cognition; in particular,
ing individual biomarkers and combinations of biomarkers with single domain non-amnestic MCI appears to show the most insta-
one another in multivariate studies and there have been few stud- bility as a diagnostic category over time (Bickel et al., 2006; Jak et al.,
ies directly comparing biomarkers in life with post-mortem data. 2007). However, other studies have shown more stability over time,
Therefore, it is currently difficult to understand the relative impor- e.g. Zanetti and colleagues (2006) followed up patients with MCI
tance of different biomarkers when used together and to interpret over 3 years and found that all MCI subgroups either converted to
results when biomarker data conflict with other biomarker data. dementia (approximately 25%) or retained their MCI status over
The absence of truly predictive studies at the individual subject the period of follow-up.
level also limits the clinical usefulness of biomarkers. Few studies Petersen et al. (1999) described the first and still the largest study
define a specific cut-off value for biomarkers and then prospectively of approximately 220 people (mean age 79 years) who fulfilled
test the predictive value. At present, there is limited standardization Mayo Clinic criteria for amnestic MCI and were then followed for
of biomarker between laboratories and between studies; therefore 3–6 years. Subjects progress to dementia at a rate of approximately
more work is required in this area before biomarkers can be used in 12% per year, in contrast to a normal older cohort who develop
routine clinical practice. Also, it is not yet known whether the best dementia at a rate of 1–2% per year. Moreover, over the course of 6
predictions of the actual rate of progression depend on the degree years, 80% had converted to dementia, suggesting the original MCI
to which an individual expresses biomarkers of neuronal injury. group did indeed represent a population at very high risk of subse-
Mattsson et al. (2010) reviewed the possible aetiology of intercentre quent dementia (Negash, 2005). Most subjects developed AD. In a
and interlaboratory variations in CSF biomarkers and concluded comprehensive review, Bruscoli and Lovestone (2004) identified 19
that there were a number of factors (preanalytical sources of vari- longitudinal studies of MCI and found an overall rate of conversion
ance such as subject selection factors, factors relating to lumbar of 10% per annum, but with large differences between studies. The
puncture, factors relating to storage and handling of biomarkers, main factor accounting for the variability was source of subjects,
analytical courses of variance, and postanalytical factors) contrib- with self-selected clinic attendees having the highest conversion
uting to the large variation in biomarker outcomes between studies. rate. Within studies they found that poor performance on cogni-
They suggested that the use of standardized operating procedures tive testing at baseline predicted conversion with a high degree of
and implementing a comprehensive international quality control accuracy, including that subjective and objective evidence of cogni-
programme could overcome some of these difficulties. This process tive decline did predict conversion to dementia, with such subjects
will take time and therefore it is unlikely that CSF biomarkers will forming a good group to select for disease modification studies.
be used in routine clinical practice in the near future. As well as a poorer cognitive outcome, MCI is also associated with
increasing mortality (Bennett et al., 2002). Overall, therefore, a sub-
stantial body of evidence supports MCI as representing a high-risk
Prognosis group of those destined to develop dementia, in particular AD.
Outcome, i.e. progression to dementia, varies across studies, largely The outcome of non-amnestic forms of MCI has not been well
dependent upon source of referral. In memory clinics, clinical trials, investigated. Within the Canadian Study for Health and Ageing,
and specialist centres, the rate of progression to dementia, particu- Wentzel et al. (2001) followed those with presumed vascular cog-
larly AD, ranges from 10–15% per year, whereas in epidemiological nitive impairment without dementia (CIND) and found that of
studies in which participants are approached prospectively and fol- 149 participants, 46% developed dementia after 5 years, suggest-
lowed up, the rates are lower at between 6 and 10% per year. This ing that vascular CIND was not a benign condition. Boeve et al.
difference in outcome is likely to reflect several factors, including a (2004) studied 21 cases of DLB who had initially presented as MCI.
priori probability of having an early dementia. For example, patients Perhaps surprisingly, 10 had originally presented as amnestic MCI,
who actively seek referral to a memory clinic have a higher chance six as single nonmemory domain MCI, four with multiple domain
of having significant memory problems at the start and therefore amnestic MCI, and one with multiple domain non-amnestic MCI.
a higher chance of the memory problems progressing (selection REM sleep behaviour disorder (RBD) has been described as a pre-
bias). These patients are diagnosed after detailed systematic clinical dictor of Parkinson’s disease and has a strong association with DLB.
and neuropsychological assessments and thus are a more selected In their sample, 11 MCI cases also had RBD (loss of normal muscle
and severe group. Thorough assessment will rule out other poten- atonia during REM sleep, with patients ‘acting out’ their dreams, and
tial causes of mild cognitive problems such as comorbid medical often reported by carers). Of these, 10 (91%) subsequently devel-
or psychiatric conditions, medication, or neurological conditions, oped DLB. Molano et al. (2010) studied eight autopsy-proven cases
which may be more likely to be static or reversible in long-term of DLB and observed that all eight individuals evolved through an
follow-up. In epidemiological studies there is a broader spectrum MCI phase. The MCI subtypes included two with single-domain
of MCI severity and a more heterogeneous group in terms of aetiol- non-amnestic MCI, three with multi-domain non-amnestic MCI,
ogy and therefore a lower chance of converting to dementia in the and three with multi-domain amnestic MCI. The cognitive domains
follow-up period. most frequently affected were attention-executive function and/or
Several studies have shown that not all individuals with MCI visuospatial functioning. Seven of the individuals had a history of
progress to a dementia diagnosis. Indeed, a proportion of individu- dream re-enactment (RBD). The onset of RBD preceded the onset
als diagnosed with MCI improve over time such that at follow-up of cognitive decline in six patients by a median of 10 years. The
they no longer meet the criteria for MCI, especially in epidemio- results from this small study suggest that DLB can pass through an
logical studies. Between 20% (Fischer et al., 2007) and 40% (Bickel MCI stage and can present with any subtype of MCI. All cases with
et al., 2006) of those with MCI appear to revert to normal range RBD and MCI were shown to have autopsy-proven Lewy body dis-
upon repeat testing. Single domain classifications appear to be the ease, suggesting that this is highly predictive of DLB. Patients with
Mild Cognitive Impairments after stroke are also at increased risk Individual clinicians will then vary considerably as to what they
of subsequently developing dementia, with rates of 25% at 2 years tell patients, with some suggesting such cases should be viewed as
reported (Altieri et al., 2004). However, improvements are also seen part of the general population and treated as if they were normal,
in some subjects (Ballard et al., 2003). The syndrome ofMCI of others that they should be informed they have a mild cognitive
frontotemporal type has been described in a small case series by de problem and that they are at an increased risk of developing demen-
Mendonca et al. (2004), who found that a combination of clinical tia in the future. The latter is now becoming standard practice in
features (apathy, disinhibition, irritability), cognitive impairments many specialist services, and information sheets for those with
(in attention, initiation, and conceptual thinking), and neuroimag- MCI have been developed. Often such cases are managed with a
ing (frontal atrophy) predicted six out of seven patients presenting combination of information and support and annual review, which
who subsequently developed FTD within a 2-year period. Further has been suggested as good practice by the American Academy of
work is necessary on long-term outcome of non-amnestic forms of Neurology (Knopman et al., 2001). Support and memory remedia-
MCI, though evidence to date suggests high rates of conversion to tion and training groups for those with MCI and their carers are
dementia. being developed and are likely to be an increasingly important part
of the future management of MCI. Carers of those with MCI may
demonstrate caregiver burden (Garand et al., 2005) similar to car-
Predictors of Conversion to Dementia ers of individuals with dementia, and this might require particular
Several predictors of conversion to dementia in those with MCI intervention.
have been described, including severity and nature of baseline
cognitive performance (particularly impaired episodic recall), Drug treatment
older age, possession of particular genotype (ApoE4), presence of Specific pharmacological therapeutic strategies for MCI are currently
functional impairments (Peres et al., 2006), and particular imaging lacking, as the first wave of clinical studies aimed at symptomatic
changes, including volume loss in the hippocampus and entorhinal drug treatment for amnestic MCI were largely negative. Salloway
cortex (Stoub et al., 2005), increased rates of whole-brain atrophy et al. (2004) reported a study of 270 MCI subjects randomized to
on serial scanning (Jack et al., 2005), and hypometabolism/hypop- donepezil or placebo for 42 days. There was no significant differ-
erfusion on functional imaging (de Leon et al., 2001), as well as the ent between groups on primary outcome measures, though some of
CSF biomarkers compatible with AD and positive amyloid imaging. the secondary outcomes tended to favour donepezil. Petersen et al.
It has been reported that multi-domain amnestic MCI has a higher (2005) reported results from the Memory Impairment Study, which
conversion rate (50% over 3 years) than pure amnestic MCI (20%) showed no significant difference in progression from amnestic
(Tabert et al., 2006). Also, noncognitive features such as depression MCI to AD in those allocated to vitamin E or donepezil compared
and anxiety have been shown to predict progression to dementia in to placebo over a 3-year period. Differences in favour of donepezil
some studies. However, all of these predictors, whilst highly signifi- were reported at 12 months but were nonsignificant at the main
cant at a group level, cannot predict with much certainty whether primary endpoint of 3 years. Intriguingly, there was an interaction
or not an individual with MCI will develop dementia. Until more with genotype in the secondary analysis in that ApoE4 carriers, a
definitive biological or other predictors can be defined, they are subgroup, did have a reduced rate of conversion to dementia when
unlikely to be clinically applicable to individual patients, though treated with donepezil. Gold et al. (2003) reported the results from
they will remain a keen focus of research attention. two international randomized controlled trials (RCTs) comparing
galantamine and placebo in 2048 individuals with amnestic MCI.
Therapeutics and Management The overall rate of progression to AD was lower than anticipated in
the trial, leading to an extension of the study. Overall, the results
Diagnosis were negative; there was a trend towards a reduction in progression
Clinically, patients with MCI will often have been referred from in the galantamine group, but this did not reach statistical signifi-
primary care to specialist services, usually old age psychiatry or cance. Furthermore, there was an unexpected increased mortality
neurology, with cognitive problems, with the main question being in the galantamine treated MCI group (though mortality in the
whether they have an early dementia. Clearly, the first step is a full placebo group was unusually low), leading to the Cochrane group
and thorough assessment of the nature of the problem, its duration, advising against prescription of galantamine for those with MCI
extent, and effects on everyday functioning. As with the assessment because of no clear benefit but risk of harm (Loy and Schneider,
of those with dementia, a full history from patient and informant 2006). Feldman et al. (2007) reported on a multicentre study of
as well as mental state, physical, and cognitive examination will be 1018 subjects with amnestic MCI randomized to rivastigmine or
needed, together, in most cases, with screening laboratory inves- placebo. Again, they failed to show the anticipated rate of progres-
tigations, including blood tests and neuroimaging. Care should sion to AD, resulting in an extension of the study. There were no
be taken to rule out other comorbid conditions (anxiety, physical significant differences in primary outcome measures between the
health conditions, or medication which could be contributing to two groups. Of note, the two groups were not well matched with
the clinical presentation) and other functional psychiatric condi- regard to the frequency of ApoE4 allele, which was 46% in the pla-
tions such as depression, which can present with both subjective cebo arm versus 37% in the rivastigmine arm.
memory complaints and objective evidence of impairment (O’Brien Raschetti et al. (2007) performed a systematic review of the
et al., 2004). Once these other problems have been excluded and evidence base (published and unpublished data) for the use of
assuming the person does not meet criteria for dementia, if MCI cholinesterase inhibitors in MCI. They noted that the enrolment
criteria are met then the most clinically pragmatic action is to use criteria differed between trials and so the study populations were
this ‘label’ as a working diagnosis. not homogenous. Also, cognitive function and ADLs were assessed
using a range of neuropsychological tests and questionnaires with Exercise

cut-offs varying between studies. The duration of the trials ranged In contrast to the negative drug studies, many observational stud-
from 24 weeks to 3 years. No significant differences emerged in the ies have shown that physical activity decreases the risk of cogni-
probability of conversion from MCI to AD or dementia between the tive decline. For example, Abbott et al. (2008) reported that men
treated groups and placebo groups. The only surrogate end-point to who walked for 2 miles/day were 1.8 times less likely than seden-
attain statistical significance over the various studies was on MRI tary men to develop dementia over a follow-up period of 2 years.
whole-brain atrophy, showing a significant difference in the rate of Lautenschlager et al. (2008) investigated this further in an RCT
brain atrophy in favour of the treatment group (galantamine); the comparing structured physical activity to education and care as
clinical significance of this remains unknown at present. They con- usual in subjects with subjective and objective memory impair-
cluded that there was no evidence to support the use of cholineste- ment. They showed that individuals randomized to the exercise
rase inhibitors in patients with MCI. The British Association of group had statistically significant improvements in their cognitive
Psychopharmacology guidelines for dementia do not support the function (as measured by the cognitive section of the Alzheimer’s
use of cholinesterase inhibitors or memantine in individuals with Assessment Scale (ADAS-Cog)) when compared to the control
MCI and conclude that there is no clear evidence that any intergroup at 6 months, which was sustained for the follow-up period
vention can prevent or delay the onset of dementia (O’Brien et al., of 18 months (Lautenschlageret al., 2008.) Although this was a
2011), with European Federation of the Neurological Societies small-scale study, it is potentially important because, in contrast to
guidelines reaching similar conclusions (Hort et al., 2010). other interventions, the results were significant, the level of activity
involved was modest (= 20 min/day), and exercise has other health
Limitations of pharmacological trials
benefits and few risks or side effects and as such would be a simple
Studies investigating pharmacological treatment in MCI are dif- intervention to offer in clinical practice.
ficult to undertake for several reasons. (1) MCI is a heterogeneous Other management strategies for those with MCI are largely
concept. In drug trials with follow-up periods of 3–4 years, around based on knowledge of risk factors and consist of treatment of con-
50–80% of patients with MCI do not develop AD. Inclusion of large current disorders, control of vascular risk factors, and advice and
numbers of patients with MCI who are unlikely to progress to AD support regarding cognitive difficulties.
in treatment trials of drugs designed to modify AD pathophysi-
ology dilutes out the potential benefits of the treatment and may The future
limit the clinical usefulness of the trial. (2) There are recruitment
The status of MCI as an entity remains controversial. A recent study
difficulties; studies often have high drop-out rates and low conver-
by Roberts et al. (2010) showed that 90% of American neurologists
sion rates because they recruit individuals at the mild end of the
recognized MCI as a clinical diagnosis and 70% used it as a diag-
MCI spectrum and these individuals often improve over time. (3)
nostic category. The fact that MCI is a commonly used and accepted
Research has shown that the cholinergic system may actually be
term in clinical practice is reflected in the current draft version of
up-regulated in MCI as a compensatory mechanism for reduced
DSM-V (summarized in Table 32.3), which proposes to include
neurotransmitter levels and, as such, drugs that boost cholinergic
mild neurocognitive disorder as a diagnostic category. However,
activity may interfere with this compensatory mechanism, lead-
the term encompasses people with heterogeneous clinical profiles
ing to reduced therapeutic benefits and possible adverse effects.
and defines subtypes that still require validating. On the one hand,
(4) Also, there is significant variability between trials in terms of
recruitment criteria and outcome measures, which limits compari-
sons among studies. Table 32.3 DSM-V criteria for mild neurocognitive disorder (draft
criteria)
Cognitive training
A Evidence of minor cognitive decline from a previous level of
There has been much interest in cognitive training, since low edu-
performance in one or more of the cognitive domains based on:
cational ability and reduced engagement in mentally stimulating
activities has been shown in population studies to be a predictor 1 Concerns of the patient, a knowledgeable informant, or the clinician
of subsequent dementia. Cognitive training strategies have been that there has been a mild decline in cognitive function
shown to have some benefits with normal older people (Ball et al., and
2002) and effects have also been reported in those with MCI. 2 Mild decline in neurocognitive performance, typically between one
Several small RCTs have shown some benefit of memory remedia- and two standard deviations below appropriate norms (i.e. between
tion groups and cognitive training in patients with MCI (Stott and the 3rd and 16th percentile) on formal testing, or equivalent clinical
Spector, 2011). For example, Rapp et al. (2002) compared cognitive evaluation
training (education about memory loss, relaxation training, mem- B The cognitive deficits are insufficient to interfere with independence
ory skills training, and cognitive restructuring for memory-related (i.e. instrumental activities of daily living (more complex tasks such
beliefs) to no treatment in a group of individuals with MCI and as paying bills or managing medications) are preserved), but greater
showed that the treatment group had significantly better memory effort, compensatory strategies, or accommodation may be required
appraisals than controls at the end of treatment and at a 6-month to maintain independence
follow-up. There were no differences between groups on memory C The cognitive deficits do not occur exclusively in the context of a
performance at post-test, but at follow-up the trained individu- delirium
als showed a trend toward better word list recall than controls,
D The cognitive deficits are not wholly or primarily attributable to
suggesting that individuals with MCI can benefit from cognitive
another axis I disorder (e.g. major depressive disorder, schizophrenia)
training.
it can be argued that careful selection of cases at high risk of devel- include structural MRI, molecular neuroimaging (FDG-PET),
oping a dementia means that it is a valid target, with the goal being and CSF analysis. The advent of these biomarkers and the clearer
prevention of dementia. This views the process as a linear progres- understanding of the pathophysiological mechanisms underlying
sion that can be arrested, but studies thus far have not shown this AD have led to the recognition that AD is a spectrum of illness that
to be true. On the other hand, it could be argued that patients who can be identified both clinically and pathologically earlier, before
progressed did not develop dementia but actually had an early form the full-blown dementia criteria are reached. This has led to the
of it. According to this view, people without the progressive form development of new criteria for the diagnosis of early AD.
will be needlessly exposed to antidementia treatments and their Additionally, the failure of the therapeutic studies on MCI to date
potential side effects for little benefit (Rockwood et al., 2007). may be due to the heterogeneity of subjects included; the identifi-
There is ongoing debate in the literature as to whether MCI is a cation of a pathologically more homogeneous group would allow
discrete syndrome or represents an early stage in an as yet unde- new clinical trials to be undertaken at a stage of illness when there
fined progressive dementia. This is a fundamental issue because remains greater potential to intervene sooner and maintain func-
how one views the concept of MCI ultimately guides research and tional independence before dementia becomes apparent.
therapeutics. For example, if MCI is seen as a discrete syndrome,
then there need to be rigorous attempts to define the syndrome, Dubois Criteria
validate diagnostic criteria for MCI, investigate and determine its
neurobiology, and develop appropriate therapeutic strategies to An International Consensus Group (Dubois et al., 2007, 2010) has
prevent subsequent ‘progression’ to different forms of dementia. developed new research criteria to take into account the advances
However, if it is felt to be the early stage in a progressive degen- in scientific knowledge and biomarkers and to aid the earlier diag-
erative process, then the focus should be on trying to establish the nosis of AD. While requiring validation, the essence of the new
nature of that underlying neurological process at the earliest pos- criteria is a move to diagnose AD before dementia is present and
sible opportunity. Indeed, new guidelines have recently been devel- incorporation of biomarkers to assist with making the diagnosis.
oped by Dubois and colleagues (2010) and the National Institute Dubois et al. proposed ‘preclinical stages of AD’, ‘prodromal stages
on Ageing and Alzheimer’s Association (2011, as described in the of AD’, and ‘AD dementia’, which are summarized in Table 32.4. This
following sections) that recognize that AD is a spectrum and aims section will focus on the preclinical and prodromal stages of AD. A
to identify individuals in the early stages of AD with a view to inter- two-stage preclinical model of AD is proposed in which individu-
vening earlier in the process. These guidelines are now discussed. als are free of cognitive or behavioural symptoms yet have either
biomarker evidence of AD pathology or carry a genetic mutation,
which is autosomal dominant and will develop AD. These have
Background to the New Criteria been termed ‘asymptomatic at-risk’ and ‘presymptomatic AD’,
The NINCDS-ADRDA (National Institute of Neurological and respectively. The second stage is a symptomatic phase where indi-
Communicative Disorders and Stroke–Alzheimer’s Disease and viduals have mild cognitive deficits, which do not reach criteria for
Related Disorders Association) (McKhann et al., 1984). and AD dementia, and biomarker evidence supportive of AD pathol-
DSM-IV criteria are the most commonly used criteria for the diag- ogy. This stage is referred to as ‘prodromal AD’. These terms are
nosis of AD, but these criteria are outdated and lag behind the sci- now explained in more detail.
entific advances that have been made in the 20+ years since their Asymptomatic at-risk state for AD individuals can be identified
original conception. Scientific understanding of the pathophysi- in vivo by evidence of amyloidosis in the brain (with retention of
ological mechanisms underlying different types of dementia, and specific PET amyloid tracers) or in the CSF (with changes in Aβ,
AD in particular, has advanced considerably over the last two dec- tau, and phospho-tau concentrations). At present, it is not known
ades. As highlighted earlier, recent research has led to the devel- whether isolated amyloidosis is sufficient to cause AD, and the
opment of distinctive and reliable biomarkers for the AD, which value of these biological changes to predict the further development
Table 32.4 New criteria (Dubois et al., 2010)

Entity AD diagnosis Presence of impairment on Evidence of biomarkers Additional requirements
specified memory tests in vivo
Typical AD Yes Required Required None
Atypical AD Yes Not required Required Specific clinical presentation
Prodromal AD Yes Required Required Absence of dementia
AD dementia Yes Required Required Presence of dementia
Mixed AD Yes Required Required Evidence of comorbid disorders
Pre-clinical AD: Absence of symptoms of AD
Asymptomatic at risk for AD No Not present Required Absence of symptoms of AD and
Presymptomatic No Not present Not required presence of monogenic AD mutation
MCI No Not required Not required Absence of symptoms or biomarkers

specific for AD
of the disease is unclear; as such, individuals are referred to as an to AD’ (Albert et al., 2011), and ‘Dementia due to AD’ (McKhann
‘at-risk state for AD’. et al., 2011). A full discussion of dementia due to AD is outside the
◆ Presymptomatic AD refers to individuals who have rare autosomal scope of this chapter.
dominant monogenic AD mutations who will develop AD. Sperling et al. (2011) have developed guidelines for what they
term ‘preclinical AD’. At present, these are viewed as research
◆ Prodromal AD describes a symptomatic disease phase where guidelines rather than guidelines for clinical practice, and again it
individuals have AD on the basis of their clinical presentation is recognized that as knowledge advances these guidelines will be
and supportive evidence of Alzheimer pathology from biomar- reviewed and amended. Figure 32.3 shows a graphical representa-
kers, but who do not yet reach criteria for dementia (prede- tion of the spectrum of AD. Figure 32.4 shows a hypothetical model
mentia). Previously, individuals with the features of prodromal of the progression of biomarkers in different stages along the con-
AD were described as having MCI, with an increased risk of tinuum of AD; this is based on the current view of the pathophysi-
developing AD, but not as having identifiable AD. The new ological sequence of events in the development of AD. Figure 32.5
criteria propose that a patient previously diagnosed as having is a graphical representation of the current view regarding preclini-
MCI (i.e. with an amnestic syndrome of the hippocampal type cal AD.
and with biomarker evidence positive for brain amyloidosis) is The NIA-AA guidelines include ‘MCI due to AD’ to describe
no longer to be thought of as at risk for developing AD demen- the predementia phase of AD (Alberts et al., 2011). These cri-
tia, but to be recognized as already having AD at a prodomal teria are shown in Table 32.5. This clinical syndrome is almost
stage. identical to that proposed by Petersen et al. (1987) as MCI. The
Dubois et al. (2010) state that AD starts at the first episode of working group then proceeded to develop this concept further
memory loss and must be accompanied by at least one positive to include the use of biomarkers alongside the clinical concept
biomarker. A ‘prodromal phase’ is recognized but MCI is not con- of MCI due to AD to give further levels of certainty that the MCI
sidered to be part of the AD spectrum. Dubois et al. retain the con- syndrome is due to AD pathology (see Table 32.6). These criteria
cept of MCI for individuals where there is no disease process to assume a stable temporal relationship of biomarkers, with amy-
which MCI can be attributed. It thus effectively becomes a diagno- loid changes appearing before those of ‘neurodegeneration’ (CSF
sis of exclusion for individuals who are suspected to have but do not tau, structural MR, and metabolic PET changes), which occur
meet the proposed new research criteria for AD, in that they devi- before clinical symptoms. The likelihood of AD as the cause of
ate from the clinicobiological phenotype of prodromal AD because
they have memory symptoms that are not characteristic of AD or
because they are biomarker negative. The continuum so Alzheimer’s disease
Aging
National Institute on Ageing–Alzheimer’s Cognitive
function
Preclinical
Association Workgroup (NIA-AA) on
MCI
diagnostic guidelines for Alzheimer’s
disease (2011) Dementia
In parallel with the International Working Group criteria, the
NIA-AA has developed three sets of guidelines based upon current
research: these include a guideline on ‘Preclinical AD—research cri- Years
teria’ (Sperling et al., 2011), ‘Mild Cognitive Impairment secondary Fig. 32.3 Hypothetical framework for clinical continuum in AD.
Abnormal
Amyloid-β accumulation (CSF/PET)
Synaptic dysfunction (FDG-PET/fMRI)
Tau-medicated neuronal injury (CSF)
Brain structure (volumetric MRI)
Cognition
Clinical function
Normal
Preclinical MCI Dementia Fig. 32.4 Hypothetical model of biomarkers of the AD process
expanded to explicate the preclinical phase (based on original work by
Clinical Disease Stage Jack et al., 2005, 2008).
Stage 1
Asymptomatic amyloidosis
-High PET amyloid tracer retention
-Low CSF Aβ1–42
Stage 2
Amyloidosis + Neurodegeneration
-Neuronal dysfunction on FDG-PET/fMRI
-High CSF tau/p-tau
-Cortical thinning/Hippocampal atrophy on sMRI
Stage 3
Amyloidosis + Neurodegeneration + Subtle Cognitive Decline
-Evidence of subtle change from baseline level of cognition MCIAD dementia
-Poor performance on more challenging cognitive tests
-Does not yet meet criteria for MCI
Fig. 32.5 Stages of preclinical AD.
Table 32.5 MCI criteria due to AD address what to do when biomarker evidence conflicts, whereas
Albert et al.’s criteria address this by risk-stratifying individuals.
◆ Cognitive concern reflecting a change in cognition reported by patient, Conversely, the Albert et al. criteria adopt a very amyloid-centric
informant or clinician (i.e. historical or observed evidence of decline over view of AD, with strict temporal ordering of biomarkers (Fig. 32.4),
time) whereas the Dubois et al. criteria permit some flexibility, allowing
◆ Objective evidence of impairment in one or more cognitive domains, for the fact that biomarkers may become positive at different stages
typically including memory (i.e. formal or bedside testing to establish level in different people.
of cognitive function in multiple domains)
◆ Preservation of independence in functional abilities
◆ Not demented
Conclusion
◆ Examine aetiology of MCI consistent with AD pathophysiological process The concept of MCI has now moved from an area of research inter-
◆ Rule out vascular, traumatic, or medical causes of cognitive decline, where est to one widely used in the clinic, and soon to enter major clinical
possible diagnostic classification systems. However, definitions continue to
◆ Provide evidence of longitudinal decline in cognition, when feasible evolve and most emphasis now is on trying to utilize biomarkers
to identify the pathophysiological processes underlying the clini-
◆ Report history consistent with AD genetic factors, where relevant
cal syndrome, both to assist with early diagnosis for the individ-
ual and to identify more homogenous clinical trial populations.
The importance of early detection, especially of prodromal AD, is
underscored by the many drugs in development that are directed
MCI is increased by the presence of both amyloid and neurode- at changing the disease pathogenesis through amyloid immuno-
generative markers. therapy, gamma or beta secretase inhibitors and modulators, alpha
The criteria proposed by Dubois et al. and Albert et al. take dif- secretase activators, tau kinase inhibitors, and nerve growth fac-
fering viewpoints on the nosology of MCI. Dubois criteria move tors. There is a neurobiological imperative to identify patients
away from MCI as a concept, whereas NIA-AA criteria broaden this early before the point of disease where irreversible pathological
concept; indeed, MCI is central to the criteria which use biomark- injury would prevent effective intervention—something that will
ers to risk-stratify patients with MCI into low, intermediate, and gain increasing momentum once disease-modifying therapies are
high probability that MCI is due to AD. The Dubois criteria do not developed.
Table 32.6 New MCI criteria incorporating biomarkers (NIA-AA criteria)

Diagnostic category Biomarker probability of AD aetiology Aβ (PET or CSF) Neuronal injury (tau, FDG, sMRI)
MCI: core clinical criteria Uninformative Conflicting Conflicting
Indeterminant untested Indeterminant untested
MCI due to AD: intermediate likelihood Intermediate Positive Untested
Untested Positive
MCI due to AD: high likelihood Highest Positive Positive
MCI: unlikely due to AD Lowest Negative Negative
FDG, 18-fluorodeoxyglucose; PET, positron emission tomography; sMRI, serial magnetic resonance imaging.
References DeToledo-Morrell, L., et al. (2004). MRI-derived entorhinal volume is a good

predictor of conversion from MCI to AD. Neurobiology of Aging, 25,
Abbott, R.D., et al. (2008). Walking and dementia in physically capable men. 1197–203.
Journal of the American Medical Association, 292, 1447–53.
Dickerson, B.C., et al. (2005). Increased hippocampal activation in mild
Ackl, N., et al. (2005). Hippocampal metabolic abnormalities in mild cognitive cognitive impairment compared to normal aging and AD. Neurology, 65,
impairment and Alzheimer’s disease. Neuroscience Letters, 384, 23–8. 404–11.
Albert, M.S., et al. (2011). The diagnosis of mild cognitive impairment due Du, A.T., et al. (2001). Magnetic resonance imaging of the entorhinal cortex
to Alzheimer’s disease: recommendations from the National Institute on and hippocampus in mild cognitive impairment and Alzheimer’s disease.
Aging–Alzheimer’s Association workgroups on diagnostic guidelines for Journal of Neurology, Neurosurgery and Psychiatry, 71, 441–7.
Alzheimer’s disease. Alzheimer’s and Dementia, 7, 270–9. Dubelaar, E.J., et al. (2006). Increased metabolic activity in nucleus basalis of
Altieri, M., et al. (2004). Delayed poststroke dementia: a 4-year follow-up Meynert neurons in elderly individuals with mild cognitive impairment
study. Neurology, 62, 2193–7. as indicated by the size of the Golgi apparatus. Journal of Neuropathology
American Psychiatric Association (1994). Diagnostic and statistical manual of and Experimental Neurology, 65, 257–66.
mental disorders, 4th edition. APA, Washington, DC. Dubois, B., et al. (2007). Research criteria for the diagnosis of Alzheimer’s
Andreasen, N., et al. (2003). Cerebrospinal fluid levels of total-tau, disease: revising the NINCDS-ADRDA criteria. Lancet Neurology, 6,
phospho-tau and A beta 42 predicts development of Alzheimer’s 734–46.
disease in patients with mild cognitive impairment. Acta Neurologica Dubois, B., et al. (2010). Revising the definition of Alzheimer’s disease: a new
Scandinavica Supplementum, 179, 47–51. lexicon. Lancet Neurology, 9, 1118–27.
Ball, K., et al. (2002). Effects of cognitive training interventions with older Ellis, J.R., et al. (2008). Relationship between nicotinic receptors and cognitive
adults: a randomized controlled trial. (See comment.) Journal of the function in early Alzheimer’s disease: a 2-[18F]fluoro-A-85380 PET
American Medical Association, 288, 2271–81. study. Neurobiology of Learning and Memory, 90, 404–12.
Ballard, C., et al. (2003). Prospective follow-up study between 3 and 15 Falini, A., et al. (2005). A whole brain MR spectroscopy study from patients
months after stroke: improvements and decline in cognitive function with Alzheimer’s disease and mild cognitive impairment. Neuroimage,
among dementia-free stroke survivors > 75 years of age. (See comment.) 26, 1159–63.
Stroke, 34, 2440–4. Farlow, M.R., et al. (2004). Impact of APOE in mild cognitive impairment.
Bennett, D.A., et al. (2002). Natural history of mild cognitive impairment in Neurology, 63, 1898–901.
older persons. Neurology, 59, 198–205. Feldman, H., et al. (2004). Behavioral symptoms in mild cognitive
Bennett, D.A., et al. (2005). Mild cognitive impairment is related to Alzheimer impairment. (Erratum appears in Neurology, 63(4), 764.) Neurology, 62,
disease pathology and cerebral infarctions. Neurology, 64, 834–41. 1199–201.
Bickel, H., et al. (2006). Prevalence and persistence of mild cognitive Feldman, H.H., et al. (2007). Effect of rivastigmine on delay to diagnosis of
impairment among elderly patients in general hospitals. Dementia and Alzheimer’s disease from mild cognitive impairment: the InDDEx study.
Geriatric Cognitive Disorders, 21, 242–50. (Erratum appears in Lancet Neurology, 6(10), 849.) Lancet Neurology, 6,
Blackford, R.C. and la Rue, A. (1989). Criteria for diagnosing age-associated 501–12.
memory impairment: Proposed improvements from the field. Fischer, P., et al. (2007). Conversion from subtypes of mild cognitive
Developmental Neuropsychology, 5, 295–306. impairment to Alzheimer dementia. Neurology, 68, 288–91.
Boeve, B.F., et al. (2004). Mild cognitive impairment preceding dementia Forsberg, A., et al. (2008). PET imaging of amyloid deposition in patients
with Lewy bodies. Neurology, 62, A86–7. with mild cognitive impairment. Neurobiology of Aging, 29, 1456–65.
Brayne, C. and Calloway, P. (1988). Normal ageing, impaired cognitive Ganguli, M., et al. (2004). Mild cognitive impairment, amnestic type: an
function, and senile dementia of the Alzheimer’s type: a continuum? epidemiologic study. Neurology, 63, 115–21.
Lancet, 1, 1265–7. Garand, L., et al. (2005). Caregiving burden and psychiatric morbidity in
Bruscoli, M. and Lovestone, S. (2004). Is MCI really just early dementia? A spouses of persons with mild cognitive impairment. International Journal
systematic review of conversion studies. International Psychogeriatrics, of Geriatric Psychiatry, 20, 512–22.
16, 129–40. Gauthier, S., et al. (2006). Mild cognitive impairment. Lancet, 367, 1262–70.
Chetelat, G., et al. (2005). FDG-PET measurement is more accurate than Geda, Y.E., et al. (2008). Prevalence of neuropsychiatric symptoms in mild
neuropsychological assessments to predict global cognitive deterioration cognitive impairment and normal cognitive aging: population-based
in patients with mild cognitive impairment. Neurocase, 11, 14–25. study. Archives of General Psychiatry, 65, 1193–8.
Counts, S.E., et al. (2006). Differential expression of synaptic proteins in Gold, M., Wong, D., and Truyen, L. (2003). Galantamine for the treatment of
the frontal and temporal cortex of elderly subjects with mild cognitive mild cognitive impairment: Characteristics of patients enrolled in two
impairment. Journal of Neuropathology and Experimental Neurology, 65, double-blind, placebo-controlled studies. European Journal of Neurology
592–601. Supplement, 10 Suppl, 235.
Crook, T., Bahar, H., and Sudilovsky, A. (1987). Age-associated memory Graham, J.E., et al. (1997). Prevalence and severity of cognitive impairment
impairment: diagnostic criteria and treatment strategies. International with and without dementia in an elderly population. Lancet, 349,
Journal of Neurology, 21–22, 73–82. 1793–6.
DeCarli, C., et al. (2001). Cerebrovascular and brain morphologic correlates Hampel, H., et al. (2004). Value of CSF beta-amyloid1–42 and tau as predictors
of mild cognitive impairment in the National Heart, Lung, and Blood of Alzheimer’s disease in patients with mild cognitive impairment.
Institute Twin Study. Archives of Neurology, 58, 643–7. Molecular Psychiatry, 9, 705–10.
DeKosky, S.T., et al. (2002). Upregulation of choline acetyltransferase activity Hansson, O., et al. (2006). Association between CSF biomarkers and incipient
in hippocampus and frontal cortex of elderly subjects with mild cognitive Alzheimer’s disease in patients with mild cognitive impairment: a
impairment. (See comment.) Annals of Neurology, 51, 145–55. follow-up study. Lancet Neurology, 5, 228–34.
de Leon, M.J., et al. (2001). Prediction of cognitive decline in normal elderly Hertzog, C. and Schaie, K.W. (1988). Stability and change in adult intelligence:
subjects with 2-[(18)F]fluoro-2-deoxy-D-glucose/positron-emission 2. Simultaneous analysis of longitudinal means and covariance structures.
tomography (FDG/PET). Proceedings of the National Academy of Sciences Psychology and Aging, 3, 122–30.
of the United States of America, 98, 10966–71. Hirao, K., et al. (2005). The prediction of rapid conversion to Alzheimer’s
de Mendonca, A., et al. (2004). Frontotemporal mild cognitive impairment. disease in mild cognitive impairment using regional cerebral blood flow
Journal of Alzheimer’s Disease, 6, 1–9. SPECT. Neuroimage, 28, 1014–21.
Hort, J., et al. (2010). EFNS guidelines for the diagnosis and management of Mitchell, T.W., et al. (2002). Parahippocampal tau pathology in healthy aging,
Alzheimer’s disease. European Journal of Neurology, 17, 1236–48. <http:// mild cognitive impairment, and early Alzheimer’s disease. Annals of
www.dsm5.org> (accessed 28/11/2011). Neurology, 51, 182–9.
Hwang, T.J., et al. (2004). Mild cognitive impairment is associated with Mitsis, E.M., et al. (2009). 123I-5-IA-85380 SPECT imaging of nicotinic
characteristic neuropsychiatric symptoms. Alzheimer Disease and receptors in Alzheimer disease and mild cognitive impairment. Journal
Associated Disorders, 18, 17–21. of Nuclear Medicine, 50, 1455–63.
Inzitari, D., et al. (2009). Changes in white matter as determinant of global Molano, J., et al. (2010). Mild cognitive impairment associated with limbic
functional decline in older independent outpatients: three year follow-up and neocortical Lewy body disease: a clinicopathological study. Brain,
of LADIS (leukoaraiosis and disability) study cohort. British Medical 133, 540–56.
Journal, 339, 2477–86. Mormino, E.C., et al. (2009). Episodic memory loss is related to
Jack, C.R., Jr., et al. (2005). Brain atrophy rates predict subsequent clinical hippocampal-mediated beta-amyloid deposition in elderly subjects.
conversion in normal elderly and amnestic MCI. Neurology, 65, 1227–31. Brain, 132, 1310–23.
Jack, C.R., Jr., et al. (2008). 11C PiB and structural MRI provide complementary Mufson, E.J., et al. (1999). Entorhinal cortex beta-amyloid load in individuals
information in imaging of Alzheimer’s disease and amnestic mild with mild cognitive impairment. Experimental Neurology, 158, 469–90.
cognitive impairment. Brain, 131, 665–80. Mufson, E.J., et al. (2002). Loss of basal forebrain P75(NTR) immunoreactivity
Jak, A.J., et al. (2009). Contributions of neuropsychology and neuroimaging in subjects with mild cognitive impairment and Alzheimer’s disease.
to understanding clinical subtypes of mild cognitive impairment. Journal of Comparative Neurology, 443, 136–53.
International Review of Neurobiology, 84, 81–103. Negash, S., Geda, Y.E., and Petersen, R.C. (2005). In: A. Burns, J.T. O’Brien,
Jicha, G.A., et al. (2006). Neuropathologic outcome of mild cognitive and D. Ames (eds) Dementia, pp. 338–46. Hodder, London.
impairment following progression to clinical dementia. Archives of O’Brien, J.T. and Levy, R. (1992). Age associated memory impairment. British
Neurology, 63, 674–81. Medical Journal, 304, 5–6.
Kantarci, K., et al. (2000). Regional metabolic patterns in mild cognitive O’Brien, J.T., et al. (2003). Vascular cognitive impairment. Lancet. Neurology,
impairment and Alzheimer’s disease: A 1H MRS study. Neurology, 55, 2, 89–98.
210–17. O’Brien, J.T., et al. (2004). A longitudinal study of hippocampal volume,
Killiany, R.J., et al. (2002). MRI measures of entorhinal cortex vs hippocampus cortisol levels, and cognition in older depressed subjects. American
in preclinical AD. Neurology, 58, 1188–96. Journal of Psychiatry, 161, 2081–90.
Klunk, W.E., et al. (2004). Imaging brain amyloid in Alzheimer’s disease with O’Brien, J.T., Burns, A., and BAP Dementia Consensus Group (2011).
Pittsburgh Compound-B. Annals of Neurology, 55, 306–19. Clinical practice with anti-dementia drugs: a revised (second) consensus
Knopman, D.S., et al. (2001). Practice parameter: diagnosis of dementia (an statement from the British Association for Psychopharmacology. Journal
evidence-based review). Report of the Quality Standards Subcommittee of Psychopharmacology, 25, 997–1019.
of the American Academy of Neurology. (See comment.) Neurology, 56, Okello, A., et al. (2009). Conversion of amyloid positive and negative MCI to
1143–53. AD over 3 years: an 11C-PIB PET study. Neurology, 73, 754–60.
Kral, V. (1962). Senescent forgetfulness: benign and malignant. Canadian Pakrasi, S. and O’Brien, J.T. (2005). Emission tomography in dementia.
Medical Association Journal, 86, 257–60. Nuclear Medicine Communications, 26, 189–96.
Lautenschlager, N.T., et al. (2008). Effects of physical activity on cognitive Peres, K., et al. (2006). Restriction in complex activities of daily living in
function in older adults at risk for Alzheimer disease. Journal of the MCI: impact on outcome. Neurology, 67, 461–66.
American Medical Association, 300, 1027–37. Petersen, R.C. (2004). Mild cognitive impairment as a diagnostic entity.
Levy, R. (1994). Aging-associated cognitive decline. International Journal of Internal Medicine, 256, 183–94.
Psychogeriatrics, 6, 63–8. Petersen, R.C., et al. (1999). Mild cognitive impairment: clinical
Li, G., et al. (2007). CSF tau/Abeta42 ratio for increased risk of mild cognitive characterization and outcome. Archives of Neurology, 56, 303–8.
impairment: a follow-up study. Neurology, 69, 631–9. Petersen, R.C., et al. (2005). Vitamin E and donepezil for the treatment of mild
Lopresti, B.J., et al. (2005). Simplified quantification of Pittsburgh Compound cognitive impairment. New England Journal of Medicine, 352, 2379–88.
B amyloid imaging PET studies: a comparative analysis. Journal of Rapp, S., Brenes, G., and Marsh, A.P. (2002). Memory enhancement training
Nuclear Medicine, 46, 1959–72. for older adults with mild cognitive impairment: a preliminary study.
Loy, C. and Schneider, L. (2006). Galantamine for Alzheimer’s disease and Aging and Mental Health, 6, 5–11.
mild cognitive impairment. (Update of Cochrane Database Systematic Raschetti, R., et al. (2007). Cholinesterase inhibitors in mild cognitive
Reviews, (4), CD001747; PMID: 15495017.) Cochrane Database of impairment: a systematic review of randomised trials. PLoS Medicine/
Systematic Reviews, CD001747. Public Library of Science, 4, e338.
Mattsson, N., et al. (2009). CSF biomarkers and incipient Alzheimer disease Ritchie, K., Artero, S., and Touchon, J. (2001). Classification criteria for mild
in patients with mild cognitive impairment. Journal of the American cognitive impairment: a population-based validation study. Neurology,
Medical Association, 302, 385–93. 56, 37–42.
Mattsson, N., et al. (2010). Inter-laboratory variation in cerebrospinal fluid Roberts, J.S., et al. (2008). Mild cognitive impairment in clinical care: a survey
biomarkers for Alzheimer’s disease: united we stand, divided we fall. of American Academy of Neurology members. Neurology, 75, 425–31.
Clinical Chemistry and Laboratory Medicine, 48, 603–7. Rockwood, K., et al. (2007). Is mild cognitive impairment a valid target of
McKhann, G.M., et al. (2011). The diagnosis of dementia due to Alzheimer’s therapy. Canadian Journal of Neurological Sciences, 34 Suppl 1, S90–6.
disease: recommendations from the National Institute on Aging– Rusinek, H., et al. (2003). Regional brain atrophy rate predicts future
Alzheimer’s Association workgroups on diagnostic guidelines for cognitive decline: 6-year longitudinal MR imaging study of normal
Alzheimer’s disease. Alzheimer’s and Dementia, 7, 263–9. aging. Radiology, 229, 691–6.
Medina, D., et al. (2006). White matter changes in mild cognitive impairment Salloway, S., et al. (2004). Efficacy of donepezil in mild cognitive impairment:
and AD: A diffusion tensor imaging study. Neurobiology of Aging, 27, a randomized placebo-controlled trial. Neurology, 63, 651–7.
663–72. Schaie, K.W. (1989). Perceptual speed in adulthood: cross-sectional and
Mitchell, A.J. (2009). CSF phosphorylated tau in the diagnosis and prognosis of longitudinal studies. Psychology and Aging, 4, 443–53.
mild cognitive impairment and Alzheimer’s disease: a meta-analysis of 51 Sperling, R.A., et al. (2011). Toward defining the preclinical stages of
studies. Journal of Neurology, Neurosurgery and Psychiatry, 80, 966–75. Alzheimer’s disease: recommendations from the National Institute on
Aging-Alzheimer’s Association workgroups on diagnostic guidelines for WHO (1992). The ICD-10 classification of mental and behavioural disorders.
Alzheimer’s disease. Alzheimer’s and Dementia, 7, 280–92. World Health Organization, Geneva.
Stott, J., et al. (2011). A review of the effectiveness of memory interventions Winblad, B., et al. (2004). Mild cognitive impairment—beyond controversies,
in mild cognitive impairment (MCI). International Psychogeriatrics, 23, towards a consensus: report of the International Working Group on Mild
526–38. Cognitive Impairment. Journal of Internal Medicine, 256, 240–6.
Stoub, T.R., et al. (2005). MRI predictors of risk of incident Alzheimer disease: Wolk, D.A., et al. (2009). Amyloid imaging in mild cognitive impairment
a longitudinal study. Neurology, 64, 1520–4. subtypes. (Erratum appears in Annals of Neurology, 66(1), 123. Note:
Tabert, M.H., et al. (2006). Neuropsychological prediction of conversion to DeKoskym, Steven T. (corrected to De-Kosky, Steven T.)]) Annals of
Alzheimer disease in patients with mild cognitive impairment. Archives Neurology, 65, 557–68.
of General Psychiatry, 63, 916–24. Xu, Y., et al. (2000). Usefulness of MRI measures of entorhinal cortex versus
Terriere, E., et al. (2010). 5-(123)I-A-85380 binding to the alpha4beta2-nicotinic hippocampus in AD. Neurology, 54, 1760–7.
receptor in mild cognitive impairment. Neurobiology of Aging, 31, 1885–93. Zanetti, M., et al. (2006). Mild cognitive impairment subtypes and vascular
Wentzel, C., et al. (2001). Progression of impairment in patients with vascular dementia in community-dwelling elderly people: a 3-year follow-up
cognitive impairment without dementia. Neurology, 57, 714–16. study. Journal of the American Geriatrics Society, 54, 580–6.
CHAPTER 33
Alzheimer’s disease
John-Paul Taylor and Alan Thomas
Alois Alzheimer wrote his original report (Alzheimer, 1907) on a the present chapter, the draft DSM-5 criteria now include a diagno-
51-year-old woman (Auguste D) who had suffered an illness involv- sis of major neurocognitive disorder associated with AD. This diag-
ing cognitive impairment and other symptoms including delusions nosis is supported by the presence of either major neurocognitive
and hallucinations. At post mortem he described the principal disorder or mild neurocognitive disorder criteria (which overlaps
neuropathological features, senile plaques and neurofibrillary tan- with mild cognitive impairment—see Chapter 32 for further dis-
gles, of what we now call Alzheimer’s disease (AD). This epony- cussion), although the latter must be supported by the presence of
mous term was coined by Kraepelin and was initially restricted to a a rigorous biomarker such as for example, an autosomal dominant
presenile degenerative dementia. After several decades, it was rec- family history of AD to allow an AD diagnosis.
ognized that older people with identical symptomatology and path- Other notable changes include the recognition that comorbidity
ological features had the same disease and, in fact, formed the great is the norm and that having cerebrovascular disease does not pre-
majority of cases. AD thus referred to a characteristic clinical syn- clude an individual also having AD or vice versa. In addition, the
drome that usually occurs in later life and which is associated with separation between early and late onset has been removed, given
the neuropathological findings described in detail in Chapter 6. the lack of scientific validity in this arbitrary distinction. Finally,
Currently, AD as a clinicopathological entity is evolving yet again criteria are also being proposed for psychosis and depression in
with the development of new diagnostic criteria and, indeed, up AD, although how these and other behaviour disturbances relate to
until recently the term Alzheimer’s disease had been used inter- the diagnostic coding remains to be clarified.
changeably with Alzheimer’s dementia; however, with the advent of Regardless of whether one applies the ICD-10, DSM-IV, or
early biomarkers and greater awareness of an extended preclinical indeed draft DSM-5 criteria, all require a systematic and thorough
phase, AD is no longer regarded as just a ‘dementia’. assessment: a full history (including from an informant), mental
state examination, cognitive examination, physical assessment
(including neurological examination), and investigations. The lat-
Alzheimer’s Disease and the ter are important because alternative causes include both systemic
Dementia Syndrome diseases, e.g. vitamin B12 deficiency and hypothyroidism, and brain
diseases, e.g. normal pressure hydrocephalus and stroke. A list of
Diagnostic criteria for Alzheimer’s disease: ICD-10, suggested investigations is given in Box 33.1.
DSM-IV, and draft DSM-5 criteria
Having confirmed a patient has the dementia syndrome (see New emerging criteria
Chapter 29 for a discussion on this), the next stage is to try to estab- Both the ICD-10 and DSM-IV criteria for AD are closely related
lish the cause. Since AD causes over 50% of all dementia, it has to the more rigorous NINCDS-ADRDA (National Institute
come to be regarded almost as the model for dementia. As such, of Neurological and Communicative Disorders and Stroke–
there is still a tendency to diagnose AD by default rather than by Alzheimer’s Disease and Related Disorders Association) criteria
the positive identification of the clinical syndrome. ICD-10 and (McKhann et al., 1984). The original 1984 NINCDS-ADRDA crite-
DSM-IV both require that dementia due to AD has an insidious ria have been shown to have a sensitivity and specificity for AD of
onset and a gradual progression. AD cannot be diagnosed if there over 80% (Burns et al., 1990f; McKeith et al., 2000c) and have very
is a sudden onset or if there are neurological signs indicating focal good inter-rater reliability (Farrer et al., 1994), but they have now
brain damage or if the dementia can be explained by other brain or been superseded by the revised 2011 criteria. In the intervening years
systemic diseases. At the time of writing, there is an ongoing revi- since the original criteria, our understanding of the pathophysiol-
sion to both ICD and DSM diagnostic criteria (DSM-5; see <www. ogy and clinical progression has advanced and, as a result, there
dsm5.org> for a full description), although these are not the only have been recent and, indeed, necessary reconsiderations of the
new diagnostic criteria for AD (see New emerging criteria). A dif- diagnostic criteria for AD. Thus, in addition to the upcoming ICD
ference between DSM-IV and DSM-5 has been the renaming of the and DSM revisions are two other sets of AD criteria: the Working
category of the generic ‘dementia’ syndrome used in DSM-IV to that Group for New Research Criteria for the Diagnosis of Alzheimer’s
of ‘major neurocognitive disorder’. More specifically and relevant to Disease led by Bruno Dubois (Dubois et al., 2007, 2010) and the
Box 33.1 Recommended investigations in dementia and Box 33.2 Alzheimer disease biomarkers
Alzheimer’s disease
In the Dubois et al. criteria, biomarkers are divided into patho-
◆ Full blood count physiological and topographical markers. Pathophysiological
markers relate to known degenerative processes that are associ-
◆ ESR or viscosity ated with AD, including amyloid deposition (particularly amy-
◆ Urea, creatinine, and electrolytes (including calcium) loid 1–42 beta) and the presence of abnormal tau. Topographic
markers relate to the subsequent impact of these cellular patho-
◆ Liver function tests
logical changes, such as structural atrophy in the medial tempo-
◆ Thyroid function tests ral lobe and reduced metabolism in temporoparietal regions on
◆ Vitamin B12 and Folate PET imaging.
(Reproduced with permission from Dubois et al., 2010.)
◆ Syphilis serology
◆ Blood Glucose
of the disease, and is followed by or associated with other cognitive
◆ Cholesterol impairments (executive dysfunction, language, praxis, and complex
◆ MSU visual processing impairments) and neuropsychiatric changes. The
emphasis on episodic memory deficits or amnestic syndrome of the
◆ Chest X-ray
hippocampal type as a core feature arises from empirical evidence
◆ ECG that the medial temporal lobes are typically the first structure affected
◆ EEG (mild diffuse slowing) by AD pathology before it cascades to other cortical areas. The crite-
ria also allow for further support for the diagnosis of ‘typical AD’ by
◆ CT/MR scan of head (may be normal or show generalized the presence of one or more AD biomarkers (Box 33.2).
atrophy or focal atrophy in medial temporal lobe)
Atypical AD Dubois et al. use this term to describe less common
◆ SPECT scan—useful in selected cases (bilateral symmetric
variants of AD, including syndromes such as progressive nonflu-
temporoparietal hypoperfusion)
ent aphasia, logopenic aphasia (characterized by slow speech and
word-finding difficulties), frontal variant of AD, and posterior cor-
revised McKhann criteria from the Alzheimer’s Association (AA) tical atrophy; however, as these syndromes also map onto other
and National Institute on Aging (NIA) (McKhann et al., 2011). differential diagnoses (e.g. frontotemporal dementias) the Dubois
While there are notable differences between these two diagnos- et al. criteria demand that the diagnosis of AD is supported by pres-
tic systems and ongoing debate regarding the most appropriate ence of amyloid in the brain (tested using, for example, amyloid
nomenclature, within these new criteria there is now an emphasis labeled radioligands) or CSF changes characteristic of AD (e.g. low
on how one defines Alzheimer’s as a disease process and a focus on amyloid-beta, high tau).
classification of Alzheimer’s before the emergence of the dementia, Mixed AD This is used in the Dubois et al. criteria to include
with increasing prominence given to biomarkers from cerebrospi- individuals who meet the ‘typical AD’ criteria but who additionally
nal fluid (CSF) and neuroimaging. have evidence (e.g. clinical, imaging, or biological) of other condi-
International Working Group for New Research criteria for the tions that could be causing their dementia, such as cerebrovascular
diagnosis of Alzheimer’s disease disease or Lewy body disease. Thus, one might consider mixed AD
The criteria proposed by Dubois et al. (2010) seek to establish a if there was a history of parkinsonism, gait disturbance, halluci-
new nosological lexicon for AD by dividing it into three catego- nations, cognitive fluctuations, recent or past history of stroke, or
ries: ‘preclinical stages of AD’, ‘prodromal stages of AD’, and ‘AD evidence of small vessel ischaemic change or infarcts on neuroim-
dementia’; these categories represent a continuum from individu- aging. However, it is important to note that the presence of these
als who evidence the presence of AD pathology as evaluated using symptoms or signs does not equate to an assumption of mixed aeti-
biomarkers (coined ‘symptomatic at risk’), or who have a genetic ology. Indeed, Dubois et al. propose that, in the case of mixed AD
mutation that makes eventual AD dementia a certainty (defined as and cerebrovascular disease, there must be both a history of stroke
‘presymptomatic AD’) but who do not currently have any cogni- and imaging signs of cerebrovascular disease.
tive or behavioural symptoms, through to individuals with mild NIA/AA diagnostic criteria
symptoms and biomarker evidence of AD pathology but who do
The new NIA/AA guidelines for the diagnosis of AD, envisaged as
not reach the threshold for AD dementia diagnosis, and finally to
a replacement to the 1984 NINCDS-ADRDA criteria (McKhann
those who have frank and symptomatic AD dementia. Chapter 32
et al., 1984), are not dramatically different from the original crite-
describes in detail the ‘preclinical stages of AD’ and ‘prodromal
ria, although there is a recognition in that other forms of demen-
stages of AD’, whereas the different lexical concepts of AD demen-
tia such as dementia with Lewy bodies (DLB), vascular dementia
tia including typical, atypical, and mixed AD dementia proposed by
(VaD), and frontotemporal dementia (FTD) have distinguishing
Dubois et al. (2007) are described here:
and different clinical features that were not defined in the NINCDS-
Typical AD This refers to the most common clinical presentation ADRDA criteria. As such, the NIA/AA guidelines have overarching
of AD which is characterized by an ‘early significant and progressive diagnostic criteria for all dementias (defined as all-cause demen-
episodic memory deficit that remains dominant in the later stages tia) as well as specific criteria for dementia caused by Alzheimer’s.
CHAPTER 33 alzheimer’s disease 433
Finally, there is an awareness in the guidelines that genetics and The guidelines suggest that the cognitive impairments should be
biomarkers have a role in increasing the certainty of the diagnosis. diagnosed by taking a history from both the patient and an inform-
A summary of the classification of dementia due to AD by these ant as well as carrying out an objective cognitive assessment. The
guidelines is shown in Table 33.1. All-cause dementia is diagnosed NIA/AA guidelines demand that the cognitive or behavioural
(a prerequisite before considering the diagnosis of AD) when there impairment includes a minimum of two domains (memory, execu-
are cognitive or behavioural symptoms that: tive, visuospatial, and language function; changes in personality,
behaviour, or comportment).
1. interfere with the ability to function at work or at usual Conclusion
activities, At the time of writing, we are currently in a state of flux with regard
to how AD is diagnosed. This probably represents a ‘paradigm shift’
2. represent a decline from previous levels of functioning and
in how we clinically view the condition and how it will be treated in
performing, and
the future. Overall, there are significant overlaps between the new
3. are not explained by delirium or major psychiatric disorder. emerging diagnostic criteria, which share an emphasis on earlier
Table 33.1 NIA/AA diagnostic criteria

Diagnostic classification Comment
Probable AD dementia Meets criteria for dementia (discussed in text) and, in addition, has the following characteristics:
A. Insidious onset. Symptoms have a gradual onset over months to years, not sudden over hours or days
B. Clear-cut history of worsening of cognition by report or observation and
C. The initial and most prominent cognitive deficits are evident on history and examination in one of the following categories:
a. Amnestic presentation: it is the most common syndromic presentation of AD dementia. The deficits should include
impairment in learning and recall of recently learned information. There should also be evidence of cognitive
dysfunction in at least one other cognitive domain e.g. impaired reasoning and handling of complex tasks or impaired
visuospatial ability)
b. Nonamnestic presentations:
◆ Language presentation: the most prominent deficits are in word-finding.
◆ Visuospatial presentation: the most prominent deficits are in spatial cognition, including object agnosia, impaired face
recognition, simultanagnosia, and alexia.
◆ Executive dysfunction: the most prominent deficits are impaired reasoning, judgement, and problem-solving.
Note: for each presentation, deficits in other cognitive domains should also be present.
D. The diagnosis of probable AD dementia should not be applied when there is evidence of (1) substantial concomitant
cerebrovascular disease, defined by a history of a stroke temporally related to the onset or worsening of cognitive
impairment; or the presence of multiple or extensive infarcts or severe white matter hyperintensity burden; or (2) core
features of DLB other than dementia itself; or (3) prominent features of behavioural variant FTD; or (4) prominent
features of semantic variant primary progressive aphasia or nonfluent/agrammatic variant primary progressive aphasia;
or (5) evidence for another concurrent, active neurological disease, or a nonneurological medical comorbidity, or use of
medication that could have a substantial effect on cognition.
Probable AD dementia with increased level of certainty
Probable AD dementia with Evidence of progressive cognitive decline on subsequent evaluations based on information from informants and cognitive
documented decline testing in the context of either formal neuropsychological evaluation or standardized mental status examinations.
Probable AD dementia in a carrier Evidence of a causative genetic mutation (in APP, PSEN1, or PSEN2) increases the certainty that the condition is caused by AD
of a causative AD genetic mutation pathology.
Possible AD dementia
Possible AD dementia with atypical Atypical course meets the core clinical criteria in terms of the nature of the cognitive deficits for AD dementia, but either has
course a sudden onset of cognitive impairment or demonstrates insufficient historical detail or objective cognitive documentation of
progressive decline.
Possible AD dementia with Meets all core clinical criteria for AD dementia but has evidence of (1) concomitant cerebrovascular disease, defined by
aetiologically mixed presentation a history of stroke temporally related to the onset or worsening of cognitive impairment; or the presence of multiple or
extensive infarcts or severe white matter hyperintensity burden; or (2) features of DLB other than the dementia itself; or (3)
evidence for another neurological disease or a nonneurological medical comorbidity or medication use that could have a
substantial effect on cognition.
Probable AD dementia Diagnosis made in persons who meet the core clinical criteria for probable AD and who have biomarker evidence of AD
with evidence of the AD pathophysiology, which increases the certainty that the basis of the clinical dementia syndrome is the AD pathophysiological
pathophysiological process process. Biomarkers include those described in Box 33.2.
(Continued)

Diagnostic classification Comment
Possible AD dementia Reserved for people with clinical criteria for a nonAD dementia but who either have biomarker evidence of AD
with evidence of the AD pathophysiological process or meet the neuropathological criteria for AD.
pathophysiological process
Pathophysiologically proved AD Meets the clinical and cognitive criteria for AD dementia outlined earlier in the text, and the neuropathological examination,
dementia using widely accepted criteria (Hyman and Trojanowski, 1997).
Dementia unlikely to be due to AD (1) Does not meet clinical criteria for AD dementia.
(2) a. Regardless of meeting clinical criteria for probable or possible AD dementia, there is sufficient evidence for an alternative
diagnosis such as HIV dementia, dementia of Huntington’s disease, or others that rarely, if ever, overlap with AD.
(2) b. Regardless of meeting clinical criteria for possible AD dementia, both amyloid and neuronal injury biomarkers are
negative.
(Adapted from McKhann et al., 2011.)
diagnosis, that is, with a view to the development of disease-modi- fluctuation, and spontaneous parkinsonism), or one core feature
fying treatments which would need to be given as early as possible plus one supportive feature (neuroleptic sensitivity, REM sleep
in AD. However, the practical application of criteria that make use behaviour disorder, and dopaminergic abnormalities in the basal
of Alzheimer’s biomarkers in the clinic is probably still a number of ganglia on functional neuroimaging). However, when only one of
years away and the diagnosis of probable AD by NIA/AA criteria these core or supportive features is found, possible DLB may be
does not require these. diagnosed (McKeith et al., 2005), although subjects may meet cri-
teria for possible AD (McKhann et al., 1984) as well. Cognitive
Differential diagnosis fluctuation can occur in AD (particularly in moderate and severe
In practice, the exclusion criteria for AD regarding sudden onset dementia) and visual hallucinations are also found (see Clinical
and focal signs specified in the criteria usually exclude people Features of Alzheimer’s Disease), so the presence of these features
who have had a stroke, although other causes, e.g. head trauma or does not necessarily preclude a diagnosis of AD. The situation can
tumour, will also be excluded. often become clearer over time: the development of parkinsonism
Nevertheless, the presence of cerebrovascular disease has fre- in the earlier stages favours a DLB diagnosis, but parkinsonism due
quently caused confusion for clinicians seeking to define the under- to Alzheimer pathology in the substantia nigra is common in more
lying aetiology of a person’s dementia. severe dementia, whereas if delusions, hallucinations, apathy, and
ICD-10 and DSM-IV allow for a double diagnosis of AD and agitation are less persistent and improve (McKeith et al., 2000a,
vascular dementia, and new NIA/AA guidelines have a category of 2000b) over time without treatment, this makes AD more likely. In
mixed dementia to include individuals who have cerebrovascular addition, uncertainties in the diagnosis can now be further clari-
disease (stroke related to the onset/worsening of the dementia or fied by the use of dopaminergic imaging (see Chapter 12), which
evidence of severe white matter hyperintensities or infarcts on neu- displays high sensitivity and specificity in differentiating DLB from
roimaging) but who also meet the core criteria for a diagnosis of AD. Making a diagnosis of DLB as opposed to AD has important
AD dementia. clinical implications. For example, the former often responds bet-
Ideally, such a diagnosis should be made when there is clear evi- ter to cholinesterase inhibitors and DLB patients can have serious
dence of each disease process contributing to the dementia, e.g. adverse reactions to neuroleptics.
someone with AD who subsequently has a stroke and suddenly In younger patients, FTD should be considered, especially if
declines cognitively. However, in clinical practice a more pragmatic frontal features are prominent, e.g. primitive reflexes, disinhibi-
decision will often be made based on evidence for the presence of tion, marked apathy, or disproportionate impairments in executive
both AD and VaD and in cases where a single pure diagnosis can- compared to memory function. This is dealt with in Chapter 36.
not confidently be made. However, a note of caution also needs to Sometimes, other rarer causes of dementia, e.g. hypothyroidism or
be given with the advance in neuroimaging methodologies, par- normal pressure hydrocephalus, may be found in someone with
ticularly magnetic resonance imaging: the mere presence of mild to apparent AD. However, it should be remembered that conditions
moderate white matter hyperintensities, lesions that are common such as hypothyroidism and mild vitamin B12/folate deficiency are
even in non-cognitively impaired older people, should not be taken common and when they are detected by routine tests this does not
as evidence for a vascular contribution to the dementia. Indeed, necessarily indicate they are making a causal contribution to the
in the new AD criteria developed by Dubois et al., merely having cognitive problems.
white matter changes on imaging without any obvious neurological
signs (e.g. motor or gait disturbance) cannot be used as a basis for Summary
making a diagnosis of mixed AD. The diagnosis of AD is a two-stage process. First, clinical evi-
The other major differential diagnosis for the dementia syn- dence for the dementia syndrome is gathered, and other ill-
drome in older people is DLB, which is covered in Chapter 35. In nesses such as depression and delirium are excluded, and then
the revised consensus criteria for DLB (McKeith et al., 2005), it the presentation is examined to see if it fits the above AD cri-
can be diagnosed confidently (probable DLB) in the presence of at teria of insidious onset and gradual progression in the absence
least two of the three core features (recurrent visual hallucinations, of other possible causes. The two main other causes to consider
in older people are VaD and DLB. Usually a careful assessment Aphasia (dysphasia)
allows such distinctions to be made, but sometimes the clini- Deficits in cortical language production are usually present if they
cian needs to wait patiently for further evidence as the disease are searched for in the early stages of AD. The earliest manifesta-
progresses. Such distinctions are important because depression tion is in word-finding difficulties, where the patient struggles
and delirium clearly merit a different kind of treatment from from time to time to find the correct words to complete sentences.
any cause of dementia. Attempts to cover this up by using circumlocutions are common
and may be cleverly executed by brighter subjects so the deficits are
Clinical Features of Alzheimer’s Disease hidden. At this time there is also usually impairment in the naming
of objects (nominal aphasia), which will initially be for the names
Cognitive symptoms of less common items rather than the more common objects that
These are central to the concept of dementia and AD and they and are frequently tested, e.g. pen or glasses. Thus, pointing at objects
their treatment are dealt with in detail elsewhere (see Chapters 2, in the room, e.g. a television, or parts of objects, e.g. the winder on
10, and 38). Here we will just describe the key symptoms and their a watch, may show some nominal aphasia that would not otherwise
progression as they are usually encountered in AD in clinical prac- be revealed. As AD progresses, these deficits worsen and problems
tice. (Note: here we will use aphasia and dysphasia synonymously, in the comprehension of language develop (receptive dysphasia)
although strictly speaking the former refers to complete impair- and expression of language shows syntactical errors, circumlocu-
ment of language and the latter to partial impairment. The same tions, and simplification of sentence structure. Verbal persevera-
applies mutatis mutandis to apraxia/dyspraxia, alexia/dyslexia, and tion, in which the patient repeats the same answer to consecutive
agraphia/dysgraphia.) questions, may now occur too. The development of these deficits
Amnesia has serious consequences, as patients now increasingly struggle to
understand what is going on around them and to communicate their
Memory impairment is the classic presenting complaint for AD,
confusion and distress clearly to others. It should not be assumed
although other symptoms, e.g. apathy, subtle personality changes,
that a concurrent loss of all understanding occurs, as patients may
and dysphasia, are often found early in the illness too. The amnesia
still be sensitive to emotional and other nonverbal cues.
typically presents with a history of misplacing and losing objects,
As the dementia becomes severe, marked poverty of the con-
forgetting appointments, or repeatedly asking the same question.
tent of speech develops, with the repetition of short simple sen-
At this early stage, working memory (e.g. the ability to repeat back
tences on favourite themes and, eventually, mutism may occur.
a few words or a list of numbers) is relatively unimpaired. However,
Some patients babble meaningless sounds or repeat words or part
the ability to recall such words or numbers a few minutes later is
of words (palilalia and logoclonia) and phrases they hear (echo-
significantly impaired because the earliest deficits in AD are in the
lalia). Progression in language deterioration is not inevitable and
process of encoding or laying down new information in the mem-
many patients remain superficially fluent. They maintain a social
ory stores. This probably results from damage to the hippocampal
discourse that can deceive the unwary about the true extent of their
formation, which is the site of the earliest pathological damage in
cognitive impairment. Careful attention to their language compre-
AD, hence the importance that is now attached to episodic memory
hension and production, however, will invariably reveal a pattern of
dysfunction as a core feature in the newer diagnostic criteria for
language impairment, including the deficits described.
AD. This failure to lay down new memories can be demonstrated
more clearly by tests in which a patient is read a short story and Apraxia (dyspraxia)
asked to recall it (logical memory tests) or when a long list of words Dyspraxia is the failure to carry out complex motor tasks due to defi-
is read and repeated attempts are made to learn the list, e.g. Rey cits in the higher cortical control of movement; not because of dam-
Auditory Verbal Learning Test. Impairment in new learning also age to the peripheral sensory or motor systems or in coordination.
leads to the characteristic disorientation in time and place that is It typically occurs with cortical deficits in the dominant (usually
prominent even in early AD. left) parietal lobe, but as with other cognitive symptoms in demen-
In the early stage of AD, the recall of previously learned infor- tia, dysfunction in other areas is usually present too. Ideomotor
mation remains good, showing that memory retrieval processes apraxia is the form usually elicited and refers to the inability of the
remain intact. This pattern results in the typical pattern of a fail- patient to carry out a motor task to command. Examples include:
ure of new learning (no memory for recent events) combined with asking patients to wave; demonstrate how they brush their teeth; or
clear memories of events earlier in life. Some individuals particu- show how they comb their hair.
larly with high intellectual reserve can compensate for these defi- Other common ‘dyspraxias’ sought by the clinician are dressing
cits and others make use of memory prompts/cues such as diaries dyspraxia and constructional dyspraxia. Strictly speaking, these are
or making notes, and thus minimize, at least in the early stages, the not dyspraxias as they are not necessarily due primarily to higher
impact of disease. However, as AD progresses the memory impair- motor failure. Dressing dyspraxia refers to the common difficulty
ment broadens to include retrieval of older memories and these patients have in putting their clothes on correctly due to faulty visu-
are typically lost according to Ribot’s law (more recent information ospatial processing, which results in them being unable to orientate
is lost before more remote events), so the patient appears to live their body properly in relation to their clothing. It is usually taken to
in the ever more distant past. As well as these losses in episodic indicate right (nondominant) parietal lobe damage. Constructional
memory (personally related events occurring in a specific situa- dyspraxia refers to the failure to adequately reproduce a two- (or
tion) there are accompanying losses in semantic memory (facts three-) dimensional drawing, e.g. a star or a clock, or to the simi-
and vocabulary) and visuospatial memory (remembering pic- lar failure to assemble blocks into the required model. Like dress-
tures, faces etc.). ing, such tasks actually involve multiple cognitive domains (visual,
spatial, executive, sensory, motor), but failure is usually taken to to another and this reveals perseverative problems. A sequence
indicate right (nondominant) parietal damage. All these dyspraxias of alternating shapes or numbers may be written and the patient
have dire consequences for the patient’s ability to carry out many asked to continue the sequence. Motor perseveration can be dem-
activities of daily living. Cooking, cleaning, and general house- onstrated using the Luria two-step or three-step tests (in which the
work require the ability to carry out motor sequences, as do more subject is asked to copy the examiner in alternately opening and
basic activities such as eating and using the toilet. Failure in such closing the hand or making a fist, then an edge, and then a palm,
domains is not primarily due to memory but to dyspraxic losses. respectively). In someone who shows any of these ‘frontal’ features
Relatives may misunderstand why someone who appears healthy it is particularly important to examine for ‘primitive reflexes’ (see
apart from an apparently mild memory impairment should strug- section Physical symptoms—Neurological).
gle in such areas, and explaining the global nature of dementia can Executive dysfunction can arise early in the course of AD,
help them understand the true situation. although the presence of these symptoms in excess of memory dif-
Agnosia and visuospatial dysfunction ficulties, particularly in younger patients, should raise the suspicion
of nonAD causes such as FTD.
Agnosia is the failure to correctly interpret a sensory input in the
presence of an intact sensory system and it is common in AD. It Other cognitive deficits
is due to cortical damage and can occur in any sensory modality, Corresponding to the above deficits in spoken language are impair-
though visual agnosia is the form usually met in clinical practice. ments in reading (dyslexia) and writing (dysgraphia), which can
In visual agnosia a patient sees an object but does not recognize be often identified with simple tests in AD using pen, paper, and
it. Common agnosias include failure to recognize objects around reading material. Acalculia (more accurately anarithmetria) is the
the house, e.g. mistaking the kettle for the teapot or the dustbin loss of the ability to do simple sums and can similarly be tested for
for the toilet, and these too can have serious consequences for the in a straightforward way.
patient. Perhaps the most well known is prosopagnosia in which
the patient fails to recognize a familiar face, and when the face is Neuropsychiatric and other non-cognitive symptoms
that of a relative it can be very distressing for all concerned. Other In his original report, Alois Alzheimer (1907) described both delu-
variants include the ‘mirror sign’—an inability to recognize one- sions and hallucinations in his patient Auguste D. The emphasis
self in the mirror—and there is considerable overlap between these on the cognitive symptoms of dementia in making the diagnosis
misidentification phenomena and those more typically seen where can distract from the fact that neuropsychiatric symptoms are very
the misinterpretations are elaborated by paranoid delusions (see common; indeed, these symptoms tend to increase over time and
section Delusions). almost all AD patients are affected by at least one neuropsychiatric
Related to visual agnosia is the occurrence of visuospatial dif- symptom during their illness (Jost and Grossberg, 1996; Lyketsos
ficulties ranging from copying difficulties, through to recognizing et al., 2002; Steinberg et al., 2008).
whole objects (agnosia). Right/left disorientation can also occur in Box 33.3 gives summary estimates of the percentage preva-
AD and may be detected by asking the patient to obey commands lence in clinical settings (that is, hospital inpatients, day patients,
such as ‘point to the left ear with the right forefinger’. It is important outpatients, and other clinic patients) of the most important AD
to be aware that disproportionate visuospatial difficulties compared neuropsychiatric symptoms derived from the research presented.
to the global cognitive state may be indicative of a non-AD presen- Lifetime rates are much higher and the Cache County prospective
tation, e.g. posterior cortical atrophy or DLB (see Chapter 35). longitudinal study (Steinberg et al., 2008) has provided up to 5-year
Frontal-executive dysfunction period prevalences for some of the major neuropsychiatric symp-
Damage to the frontal lobes has profound effects on personal- toms (Box 33.4).
ity, mood, and behaviour (discussed later). Cognitive effects
include inflexibility in thinking and difficulties in problem-solv-
ing, abstraction, planning, and correctly sequencing behaviour. Box 33.3 Approximate monthly prevalence of noncognitive
This is a complex area but one that should nonetheless be tested symptoms in AD
during assessment of dementia. It is dealt with in more detail in
Chapter 10. Proverbs have traditionally been used to test how Psychotic
concrete and inflexible someone’s reasoning is, but they are often Delusions 20–30%
very difficult to interpret. ‘Similarities and differences’ tests involve Misidentifications 30%
asking the patient what two objects (e.g. table and chair; shoe and Hallucinations 20%
boot) have in common and how they differ. Again, these can be Mood
difficult to interpret, but some attempt should be made, especially Depression 30%
in someone who appears frontally impaired (e.g. markedly disin- Euphoria < 1%
hibited). Easier to interpret are verbal fluency tests of frontal lobe Anxiety 30–50%
function. These involve asking patients to name as many objects in
a category in 1 min (e.g. fruit or animals) or asking them to gener- Behavioural
ate as many words as they can think of beginning with the same Apathy 70%
letter in 1 min (e.g. F, A, and S in ‘the FAS test’). As a rough guide, Agitation 30%
a normal subject should produce more than 15 words for each in a Wandering 20%
minute, and less than 10 is clearly abnormal. A third kind of fron- Physical aggression 20%
tal test involves testing the subject’s ability to shift from one item Verbal aggression 40%
Box 33.4 Period prevalence of neuropsychiatric symptoms in AD from the Cache County prospective longitudinal study over a 5-year
period (Steinberg et al., 2008)
100 baseline n=408 1.5 years n=236 3.0 years n=105 4.1 years n=51
90 5.3 years n=36
80
70
60
Percentage
50
40
30
20
10
0
on
ria
ty
ur
ns
ns
l
y
n
ia
ta
ce
ilit
sio
xie
or
io
tio
sio
ho
iti
to
en
lab
av
ph
es
ib
An
ina
elu
sp
PI
er
nh
eh
gr
y/
eu
N
dy
iff
uc
D
ag
isi
b
ilit
n/
nd
n/
all
or
n/
b
tio
/i
H
io
ita
ot
tio
hy
ss
Ela
Irr
m
ita
re
at
nt
Ap
ep
Ag
ra
D
er
Ab
However, accurately eliciting these mental state phenomena delusions are often short-lived and can therefore be difficult to dis-
can be difficult in people with dementia and the figures quoted tinguish from confabulations. For hospital patients, the prevalence
in following sections and in Boxes 33.3 and 33.4 are derived from of delusions in published studies varies from about 25% (Cooper
interviews with informants, who may overrate some symptoms et al., 1991) to about 50% (Hirono et al., 1998). Burns et al. (1990d)
(e.g. depression) compared with trained observers (Burns et al., studied a sample of AD patients in a catchment area and found 16%
1990b). The rates of all the neuropsychiatric symptoms vary have had delusions at some point in their illness and 11% have had
widely depending on, amongst other things, how the phenomena them in the previous 12 months. This figure is lower than most stud-
are defined, which patients are sampled, and the timeframe over ies looking at clinically derived samples. Ballard and Walker (1999)
which the survey is conducted. Also it is worth noting that many systematically reviewed the literature and found a mean prevalence
of the studies were conducted before the accurate recognition of of delusions of 31% in all clinical settings and 23% in community
DLB was possible, which has a much higher rate of psychosis, and settings with Burns et al. (1990d) finding that delusions were three
so the figures for AD are likely to be inflated by an admixture of times more prevalent in men.
DLB patients. Misidentification syndromes
Associations between the manifestations of neuropsychiatric
Misidentification syndromes are sometimes called delusional
symptoms in AD with the level of cognitive impairment are not
misidentifications and might be better termed symptoms than syn-
consistent, although some symptoms such as apathy (Onyike et al.,
dromes. They are very common in AD, occurring in about 23–30%
2007) and irritability (Craig et al., 2005) appear to be associated
of patients at some time and in about 19% during a single year
with the degree of cognitive decline. Other symptoms such as psy-
(Rubin et al., 1988; Burns et al., 1990c). They have been classified as
chosis may predict accelerated cognitive decline (see Psychotic
delusions and as hallucinations depending on how the phenomena
symptoms for further discussion).
are understood. For example, the common assertion by demented
Finally, it is important to recognize that other non-cognitive
patients that someone else has been in the house, when their family
symptoms such as physical and neurovegetative symptoms have a
denies such a possibility, could be a paranoid delusional belief or
profound impact upon patients with AD and their care.
a visual hallucination. In practice it is often not possible to clarify
Psychotic symptoms the experience well enough to determine the exact form of the psy-
Delusions chotic experience.
Delusional beliefs are common, can be very distressing to carers, Another common kind of misidentification is the conviction that
and may contribute to early institutionalization. They are often a relative or friend is someone else (which may be a Capgras symp-
understandable as attempts by patients to make sense of their situ- tom if the misidentified person is accepted as looking exactly the
ation and include beliefs that objects have been stolen, a spouse is same as his or her ‘double’, though this may be difficult to tease out
unfaithful, or that intruders have been in the house. These beliefs in AD). Commonly, a spouse is misidentified as a son or daughter
commonly have a paranoid flavour and tend to lack the complex- or vice versa, but sometimes the patient believes the misidentified
ity and systemization of the delusions seen in schizophrenia. The person is a stranger and this may provoke a strong reaction. Other
symptoms include beliefs that people or events on the television are

Box 33.5 Pharmacological management of aggression, agitation,
actually in the living room and the misidentification of the patient’s and psychosis in dementia
own image in the mirror as another person. Wrongly believing
strangers to have been in the house appears to be the common- While this box focuses on the management of aggression, agitation,
est form of misidentification, being found in 12–17% of patients at and psychosis in AD, these points are applicable to other neuropsy-
some point; misidentifying people occurs in about 12%; misiden- chiatric and behavioural disturbances that are found in dementia.
tifying images from the television is found in 6–12%; and misiden- ◆ Consider, first, if nonpharmacological and social care inter-
tifying a mirror image occurs in about 4–7% (Rubin et al., 1988; ventions can solve the problem.
Burns et al., 1990c). Overall, studies show that misidentification
occurs in 30% of people with AD (Ballard and Walker, 1999). Like ◆ Always ask why is this patient presenting with this symptom
delusions, these misidentification syndromes appear to be more now? Is there a treatable cause for his/her behavioural distur-
likely to occur in men (Burns et al., 1990c). bance (e.g. pain, constipation, urinary retention)?
Hallucinations ◆ Before instituting any pharmacological treatment always

explain to carers and relatives the rationale for prescribing any
Hallucinations are less common than delusions and misidentifica-
drugs and discuss any potential risks associated with the medi-
tion syndromes. However, study figures here may be underestimates
cation (e.g. stroke and mortality risk with antipsychotics).
reflecting an unavoidable bias in the way in which the symptoms are
rated. The scores are informant based and informants will probably ◆ Perform baseline assessment of the nature, frequency, and
be unaware of the full extent of someone else’s internal hallucina- severity of the behavioural disturbance to compare against
tory experiences whilst it seems reasonable to believe that delusions future changes in presentation.
and misidentifications are almost certain to reveal themselves. ◆ If the symptoms are mild to moderate, consider the use of a
Visual hallucinations are more common than auditory hallucina- cholinesterase inhibitor or memantine, although be aware that
tions and whilst olfactory, gustatory, and tactile hallucinations also the evidence base for the use of these agents to treat aggres-
occur in AD, they are much rarer. Studies in clinical settings have sion/agitation/psychosis is not strong.
found prevalence rates for all hallucinations of between 10% (Mega
et al., 1996) and 40% (Gilley et al., 1997) in AD. In their systematic ◆ If symptoms persist, or are severe in nature with a potential
review, Ballard and Walker (1999) found mean prevalence rates of risk of harm, consider a cautious trial of an antipsychotic.
21% for all hallucinations (14% for visual hallucinations and 7% for ◆ Benzodiazepines should also be considered if behaviour is
auditory hallucinations) in clinical settings. severely disturbed and/or has an anxiety focus, but be alert for
excess sedation and falls.
Management of psychosis in AD
Pharmacological (Box 33.5) and nonpharmacological interventions ◆ Alternative medications include use of citalopram (possibly
may both be of benefit, although use of the former needs to be con- effective in aggression) or trazodone (possesses sedative quali-
sidered in light of their benefit versus potential risks (particularly in ties in addition to antidepressant effect).
the case of antipsychotics—see section below). ◆ Sodium valproate or carbamazepine may also be considered,
Nonpharmacologically, in addition to dealing with general but these need to be used with a high degree of caution given
health matters, treating the sensory deficits that may be exacerbat- their propensity for serious side effects (indeed, there is little
ing hallucinations, increasing the general level of stimulation, e.g. evidence to support the use of valproate, as side effects appear
by arranging day care or attendance at a day centre, may be of ben- to outweigh any potential benefits).
efit to someone who has hallucinations or misidentifications. Other
◆ For all medications:
interventions can include playing soothing music, using digital
video recordings of familiar figures in conversation, and removing ◆ Start low and go slow
the mirror for someone with the mirror sign. ◆ Review use of the drug regularly, looking for side effects
Mood changes ◆ For antipsychotics, be aware that most neuroleptic reactions
Alterations in mood are also extremely common in dementia, with occur in the first 2 weeks, and if parkinsonism manifests con-
depression being the most common. Here we are concerned with sider a diagnosis of latent DLB and stop the antipsychotic
depression occurring during established AD; the relationship of ◆ Do not prescribe the drug for prolonged periods. Regularly
depression to AD as a risk factor is considered in the section Risk reassess the need for the drug; can alternative interventions
Factors for Alzheimer’s Disease and the concept of ‘depressive be applied now the situation is containable?
pseudodementia’ is dealt with in Chapter 42.
One problem is what is meant by the term ‘depression’. Even sep-
arating studies looking at depressed mood, depressive symptoms, the assessment of depressive symptoms also occur because apa-
and depressive syndromes still leaves enormous variability, with thy can easily be mistaken for depression, whilst many depressive
rates varying from 0% to 87% for depressed mood (Knesevich et al., symptoms, e.g. poor appetite, weight loss, and sleep disturbance,
1983; Merriam et al., 1988); from 0 to 89% for all depressive symp- are also common features of dementia itself. In addition, depres-
toms (Wragg and Jeste, 1989); and from 0% to 30% for depressive sive symptoms in dementia tend to be associated more with psy-
syndromes (Burns et al., 1990b; Teri and Wagner, 1991). These rates chotic symptoms and apathy and less with negative cognitions such
vary because different populations are sampled, different instru- as hopelessness, suicidality, guilt, and low self-esteem (Olin et al.,
ments are used, and different raters make the rating. Problems in 2002; Rosenberg et al., 2005), and thus depression in dementia may
represent a distinct syndrome with pathophysiological underpin- with more severe depression, suicidality, or agitation/aggression,
nings that are different from the ‘common or garden’ depression; although further work is needed to clarify the safety of ECT and its
certainly this may have an impact on whether antidepressant medi- efficacy (Rao and Lyketsos, 2000; Gauthier et al., 2010).
cations are effective. Elevation of mood in dementia is much less common and less
In a review of studies examining depressive syndromes diag- frequently studied than depression. Burns et al. (1990b) found only
nosed using standardized criteria, Ballard et al. (1996) found major one patient (out of 178) who reported manic symptoms and only six
depression to occur in 21% of dementia sufferers in clinical settings who had any observable evidence of mania. Similarly, point preva-
and 13% in the community. Depressive symptoms are also very lence rates of euphoria and elation in the Cache County Study were
common, with the Cache County Study reporting an initial preva- less than 1% (Steinberg et al., 2008). Anxiety has been much less
lence of depressive symptoms of 29% at baseline (Steinberg et al., studied than even euphoria in dementia, but, consistent with clini-
2008). Over a 5-year period, this study also noted that period prev- cal experience, it is a common symptom, with studies finding a rate
alence for depression was very high at 77%. However, no clear asso- ranging from 24% (Steinberg et al., 2008) to 50% (Patterson et al.,
ciation between depression and the severity of dementia has been 1990), and in the Cache County Study the 5-year period prevalence
found, with some studies showing depression to be more common of anxiety symptoms was 62% (Steinberg et al., 2008).
in milder dementia and others showing the opposite. Depression
Behavioural symptoms
decreases the quality of life of people with dementia, may reduce
Apathy
the duration of survival (Burns et al., 1991c), and so it would there-
Apathy is diminished motivation and manifests itself as a listless-
fore appear to be important to treat it if possible.
ness in which the patient has lost the drive to engage in goal-ori-
Earlier small trials of selective serotonin reuptake inhibitors
ented behaviour and cognition (Starkstein and Leentjens, 2008).
(SSRIs) (Nyth et al., 1992; Lyketsos et al., 2003), moclobemide
It is often confused with low mood and anhedonia, i.e. the loss of
(Roth et al., 1996), and tricyclic antidepressants (TCAs) (Petracca
ability to enjoy previously pleasurable activities. Although they all
et al., 1996) reported evidence of benefit for depression in demen-
may be present together they are not the same phenomenon (Levy
tia, but TCAs should be avoided because they worsen cognitive
et al., 1998). The distinction is important because carers often think
impairment and have higher drop-out rates (Taragano et al., 1997).
someone with AD is depressed when he/she has apathy as a part of
However, the quality of most of these studies is poor and several
the dementia. Apathy is probably the most common behavioural
others were negative; indeed, a systematic review for the Cochrane
change in AD (Mega et al., 1996; Steinberg et al., 2008), although
collaboration did not find clear evidence for the efficacy of antide-
it is underdiagnosed. Apathy is common early in AD and it typi-
pressants in dementia (Bains et al., 2002). Reinforcing this evidence
cally becomes more severe as the cognitive impairment worsens.
for lack of efficacy are the results from the UK Health Technology
At presentation, apathy is frequently a complaint given by relatives
Assessment Study of the Use of Antidepressants for Depression in
who often misunderstand it as laziness and become frustrated by
Dementia (HTA-SADD) (Banerjee et al., 2011); this double-blind,
it. Explaining that apathy is a feature of dementia can remove this
placebo-controlled trial in a large cohort of probable or possible
misunderstanding and the tension it brings.
AD patients with depression found no treatment benefit with either
Previously apathy has been a somewhat neglected neuropsychi-
sertraline or mirtazapine and increased adverse events were noted
atric aspect of dementia, but more recently it has become a focus
in those on these medications. The lack of efficacy and potential
of academic and clinical interest. Although no formalized consen-
increased side effects of sertraline have also been confirmed by
sus criteria yet exist for apathy, a taskforce (Robert et al., 2009) has
the second phase of the Depression in Alzheimer’s Disease Study
proposed that apathy be defined as ‘a disorder of motivation that
(DIADS-2) (Rosenberg et al., 2010; Weintraub et al., 2010), which
persists over time and meets the following requirements: (1) the
compared sertraline 100 mg with placebo in depression in AD and
core feature of apathy, diminished motivation, must be present for
found no significant difference in symptom change or response or
at least 4 weeks; (2) two of the three dimensions of apathy (reduced
remission rates at either 12 (Rosenberg et al., 2010) or 24 weeks
goal-directed behaviour, goal-directed cognitive activity, and emo-
(Weintraub et al., 2010). They also reported an increase in adverse
tions) must also be present; (3) there should be identifiable func-
events on sertraline and concluded that the evidence does not sup-
tional impairments attributable to the apathy; and (4) criteria are
port the use of SSRIs in depression in AD.
given to exclude symptoms and states that mimic apathy.’
These findings now leave clinicians with difficult choices in man-
Although there are no licensed pharmacological treatments
aging depression in dementia. However, as with trials of antipsy-
available at the present time for apathy, a recent systematic review
chotics, subjects in these studies were less severely depressed than
suggested that there is evidence that the cholinesterase inhibitors
many encountered in clinical practice. Antidepressants may there-
and memantine may be effective in treating this symptom, but
fore retain a role in the treatment of more severe and prototypical
there is less evidence for the efficacy of stimulants, calcium antago-
depressive syndromes, although the argument that patients with
nists, and antipsychotics, and little or no evidence to support the
more severe depression may differentially respond to antidepres-
use of antidepressants and anticonvulsants (Berman et al., 2012).
sants has not been supported (Drye et al., 2011). Simple social
Nonpharmacological interventions such as exercise, multisensory
interventions, e.g. befriending or a day centre, should be consid-
stimuli, and pet therapy may also help, although rigorous data on
ered and it is important to note that even in the pragmatic HTA-
their efficacy are lacking (Brodaty and Burns, 2012).
SADD trial, there appears to be a natural amelioration in depressive
symptoms with time in the placebo group (who had treatment as Overactivity and aberrant motor activity
usual). Furthermore, other agents such as cholinesterase inhibitors A number of overlapping phenomena involving overactive behav-
may also be useful and electroconvulsive therapy (ECT) remains iour are commonly found in AD. The most well known are agita-
a treatment option for dementia patients, particularly for those tion, which may be defined as painful inner tension associated
with excessive motor activity (American Psychiatric Association,, persists, especially when it is associated with night wandering. To
1994), often goes together with aggression (discussed in section reduce the strain from sleep disturbance the spouse may sleep in
Aggression and agitation) and wandering. However, other ‘aber- another bed or in another room. Respite care can be of great help,
rant motor behaviours’ are recognized and sometimes measured, e.g. although placement in residential care may finally be needed.
rummaging in drawers and repeatedly putting clothes on and then Pharmacologically, psychotropic drugs (hypnotics and antipsy-
taking them off. Aberrant motor behaviour including wandering, chotics) are frequently prescribed, although caution needs to be
purposeless hyperactivity, and rummaging appears to affect about applied as these drugs are associated with significant adverse effects,
a quarter of dementia patients (Gauthier et al., 2010). The baseline e.g. falls and oversedation. The use of chronotropic agents, such as
rate for these behaviours in the Cache County Study was reported melatonin and bright light therapy, to manage circadian alterations
to be 7% and increased up to just under 30% after approximately may be a promising new avenue of treatment for sleep disorders in
5 years, which is consistent with the notion that all such overactive dementia, although more evidence is needed.
behaviours tend to become more common with increasing sever- Eating
ity of the dementia. Several drugs may help to reduce such behav-
Difficulties with feeding are common in dementia. These may
iours; while low-dose antipsychotics are the most widely prescribed,
progress from initial problems using cutlery with associated spill-
cholinesterase inhibitors (Gauthier et al., 2010) and carbamazepine
age of food through to complete dependence on others for feeding.
may also be helpful (Tariot et al., 1998). As with other neuropsy-
In addition, there may be specific problem behaviours related to
chiatric symptoms such as aggression and agitation, nonpharmaco-
eating. Some patients refuse food, whilst others stuff food rapidly
logical environmental strategies may be helpful (see Chapter 21).
and clumsily into their mouths. The latter binge eating is common,
Aggression and agitation occurring in 10% of people with AD (Burns et al., 1990a), and
Aggression and agitation are major problems in dementia. They this figure is constant for all severities of dementia, although it is a
occur commonly and can cause great distress to carers. They are a characteristic feature of FTD and should raise this as a diagnostic
frequent reason for admission to hospital and transfer to residential option (see Chapter 36).
care. Aggression is open to different definitions and even when ver- Sexual disinhibition
bal aggression (shouting, swearing) is distinguished from physical
Sexual disinhibition, along with binge eating, is another prob-
aggression (punching, kicking, biting, pushing, pinching, etc.) there
lem that causes great distress to carers. It occurs in about 7% of
is still much scope for differences of opinion in each category about
people with dementia, but unlike binge eating it is rare in mild
what constitutes aggression. Burns et al. (1990a) defined aggression
dementia and increases markedly with dementia severity (Burns
narrowly as behaviour liable to cause physical injury to others and
et al., 1990a).
still found 20% of their AD subjects to have aggression. Broader
definitions naturally lead to higher figures and verbal aggression is Personality changes
usually found to be two to three times more common than physi- Alterations in personality are ubiquitous in dementia and changes
cal aggression. Burns et al. (1990a) found aggression to be much in personality are inseparable from some of the behavioural changes
more common in men with AD. They also found that it became already discussed, e.g. apathy and sexual disinhibition. However, it
more common with increasing cognitive impairment, rising from is important to note that changes in personality do occur very early
8% in mild AD to 24% in severe AD. Agitation, which overlaps with in AD and 75% of people with mild AD show such changes (Rubin
motor overactivity, has rates in dementia that are similar to aggres- et al., 1987). These may precede the cognitive deficits and are often
sion, with point prevalence rates between 13 and 24% in the Cache subtle in form, but nonetheless they are frequently recognized and
County Study (Steinberg et al., 2008). commented upon by carers. Such information should be cautiously
Treatment of agitation and aggression is often similar to that interpreted, as family and friends are often too closely involved to
used for psychosis (see section Pharmacological treatment be able to comment accurately and dispassionately and may either
of aggression, agitation, and psychosis in AD and summary exaggerate or deny any changes. As the AD progresses these ini-
in Box 33.5) and often treatment trials not unreasonably assess tially subtle personality alterations tend to become more evident.
these problems together (e.g. psychosis may be causing aggressive Personality changes not included in the section Behavioural
behaviour). changes are suspiciousness and disengagement. Suspiciousness is
common and was found in 25% of AD subjects (Patterson and Bolger,
Neurovegetative symptoms
1994). It may or may not eventually consolidate into frank paranoid
Sleep
delusions and can cause great friction, as repeated accusations are
Sleep disturbance is another common feature of dementia, which
made about the motives and behaviour of well-meaning carers and
can have devastating consequences for carers and is dealt with in
relatives. Disengagement refers to the detached state of emotional
more detail in Chapter 51. It may take the form of frequent waking,
indifference seen often in AD which is associated with a loss of rap-
reduced sleep quality, or a disturbance to circadian rhythms. More
port with other people. Patients are no longer concerned about the
recent studies have suggested an increased prevalence of sleep
feelings of others or affected by their attention or lack of it. Such disen-
apnoea in AD, although studies have not consistently found this
gagement can be upsettling to relatives and may overlap with apathy,
(Duthie et al., 2011).
but it needs to be distinguished from the low mood of depression.
The well-recognized pattern of day–night reversal was found to
occur in 28% of patients in one study (Reisberg et al., 1987) and Physical symptoms
all forms of sleep disturbance have been found in between 45 and Neurological
70% of AD patients (Rabins et al., 1982; Merriam et al., 1988). Sleep Early in AD, neurological examination is usually normal. Focal neu-
deprivation can eventually lead to exhaustion for the spouse if it rological signs suggest a vascular dementia or other cause of focal
damage, whilst extrapyramidal signs may indicate DLB. Primitive Pharmacological treatment of aggression, agitation,
reflexes, however, may occur early in AD, although their frequency and psychosis in AD
increases as the dementia worsens. Burns et al. (1991a) found the
Antipsychotics
snout reflex was easily the most common, occurring in 41% of their
clinical sample. The grasp reflex was identified in 7% and the pal- Several studies have evaluated whether drug treatments are effective
momental reflex in 2.5%. As the disease progresses, nonspecific for non-cognitive symptoms in dementia, but there is controversy
changes occur in both gait and balance, and myoclonic jerks and about the risk: benefit ratio of these drugs in AD and in dementia
other seizures may develop. In the late stage, bilateral nonfocal in general. Empirically, antipsychotics have been used to treat agita-
signs (e.g. upgoing plantars) may be seen. tion, aggression, and psychosis in AD. However, the evidence base
for their use in AD psychosis is less clear, as studies have tended
Incontinence to be small and have focused on agitation and general behavioural
Urinary incontinence is a common and well-recognized feature of disturbance as outcomes. Several larger randomized trials of atypi-
AD and has been found in 48% of AD subjects (Burns et al., 1990a), cal antipsychotics in AD have been reported, e.g. the National
though this figure disguises its strong association with disease sever- Institute of Mental Health (NIMH) Clinical Antipsychotic Trials
ity. Incontinence was found in only 8% of those with mild AD but in of Intervention Effectiveness–Alzheimer’s Disease (CATIE-AD)
94% of patients with severe AD (Burns et al., 1990a). This associa- (Schneider et al., 2006b), and these, together with the smaller tri-
tion has important implications in diagnosis as incontinence early in als, have been subject to systematic review and meta-analysis (Sink
a dementia is characteristic of other dementias, especially FTD and et al., 2005; Ballard et al., 2006, 2011a; Schneider et al., 2006a). In
normal pressure hydrocephalus. Thus the presence of urinary incon- summary, these reports suggest there is evidence for the efficacy,
tinence early in a dementia should lead to a more serious search for albeit of modest effect, for the use of certain antipsychotics (ris-
possible causes other than AD. It is also vital to consider and exclude peridone and aripiprazole) short term (less than 12 weeks) in the
other causes of incontinence. Urinary tract infections, especially in treatment of psychosis and aggression compared to placebo. The
women, are a common finding and may be causing the incontinence evidence for olanzapine is less clear, with contradictory findings
rather than the dementia. Also the incontinence may result from reported, but there does appear to be a lack of benefit for quetiapine
poor mobility or failure to find the toilet in a strange environment (Ballard et al., 2011a). However, many trials have tended to use sub-
(e.g. when someone has just been admitted to an inpatient unit). optimal doses of quetiapine, making it difficult to draw definitive
General physical changes conclusions. There is also little evidence that typical antipsychot-
AD is associated with a general physical deterioration in which the ics are efficacious for treating behavioural disturbance in demen-
patient loses weight and develops a stooped posture associated with tia (although haloperidol may have some limited efficacy) and the
instability and gait abnormalities. AD patients are more prone to evidence suggests increases in mortality are associated with these
falls; this, combined with increased rates of osteoporosis, leads to agents too, perhaps at higher rates than for atypicals (see further dis-
enhanced likelihood of fractures compared to nondemented sub- cussion that follows and Wang et al. (2005); Laredo et al. (2011)).
jects (Duthie et al., 2011). Evidence for the benefit of antipsychotics long term is also
Weight loss may be an intrinsic feature of the disease process or equivocal and withdrawal studies (e.g. the Dementia Antipsychotic
secondary to the progressive impairments, which can result in an Withdrawal Trial (DART-AD) (Ballard et al., 2009a)) have failed
inadequate consumption of food and fluids. Cronin-Stubbs et al. to demonstrate any worsening in neuropsychiatric symptoms after
(1997) demonstrated that older people with AD lost weight at over antipsychotic discontinuation, although many of the subjects in
three and a half times the rate of healthy age-matched controls, a these studies were probably started on the drugs for milder degrees
rate of weight loss more severe than in older people with cancer. of symptoms. Most importantly, however, any benefit from antipsy-
Weight loss was found to be more rapid in earlier stages of the chotics in dementia needs to be weighed against the significant side
disease, suggesting it is at least in part due to the disease process. effects associated with antipsychotic use (see Box 33.6), and in par-
Whatever the cause, such marked weight loss is likely to contribute ticular the increase in mortality in dementia patients taking these
to the high mortality of AD. agents. Regulatory bodies in different countries have issued warn-
ings about an increased risk of death and cerebrovascular adverse
Summary events in people with dementia taking atypical antipsychotics, e.g.
Whilst cognitive symptoms are central to the diagnosis of AD, neu- the Committee for the Safety of Medicines in the UK (2004) and
ropsychiatric and non-cognitive symptoms are arguably of more the Food and Drug Administration in the US (2005), with an exten-
importance in management. Psychotic and mood symptoms are sion of the warnings to include typical antipsychotics in 2007 by the
very common, a cause of distress to both patients and carers, and, FDA. These warnings have been based on a number of case control
in many cases, treatable in their own right. The behavioural dis- and case register data. Initially much of the data were not publicly
turbances, accompanying personality changes, and neurovegetative available, and consequently in an early meta-analysis (Schneider
symptoms are also much more common than is usually recognized, et al., 2005) on the risk of death with atypical antipsychotics of
can place great strain on carers, and their presence is a major risk 15 eligible trials, nine were unpublished with data extracted from
factor for subsequent institutionalization. abstracts presented at research meetings. This report noted a sta-
Psychosocial interventions are usually necessary, whilst some tistically significant increase of about 50% in mortality in patients
of these (apathy, agitation, aggression, sleep disturbance) may taking atypical antipsychotics but identified a similar increase in
respond to pharmacological treatment. The physical symptoms those on haloperidol. Overall, there appears to be a 1.5- to 1.8-fold
in AD remind us it is a debilitating condition requiring ongoing increase in mortality with antipsychotic use (Ballard et al., 2009a).
assessment and long-term palliative care. This risk may be substantially higher in the first week of medication
Box 33.6 Potential adverse effects of antipsychotics in dementia Antidepressants and mood stabilizers
Although antidepressants (especially trazodone) and mood stabi-
◆ Worsening of cognitive function—equivalent to up to 1 year of lizers (previously carbamazepine was widely used but more recently
cognitive decline sodium valproate has become commonly prescribed) have been
and still are widely used for a range of behavioural disturbances
◆ Sedation and falls × 2 risk in AD, there is no consistent evidence from randomized trials to
◆ Extrapyramidal side effects support such prescribing practice (Sink et al., 2005; Gauthier et al.,
◆ Respiratory infections × 3 risk 2010; Seitz et al., 2011).
Trials using antidepressants to treat agitation have been hampered
◆ Prolonged QTc and torsades de points × 2 risk by small sample size and design issues; perhaps the best evidence
◆ Cerebrovascular event × 2 risk, although possibly × 9 risk in is from a small RCT comparing citalopram with risperidone which
first week of antipsychotic use found that both agents reduced agitation and psychotic symptoms
to a comparable level (Pollock et al., 2007).
◆ Mortality × 1.7–1.8 risk; may be be higher with typical
There have been a number of small studies examining the use
antipsychotics
of valproate for the treatment of agitation and aggression in AD,
(Data derived from Kleijer et al. (2009);
although a Cochrane review (Lonergan and Luxenberg, 2009) noted
Ballard et al. (2011a); and Vigen et al. (2011).)
that the benefits of valproic acid derivatives were equivocal and
indeed were appeared to be associated with significant adverse side
administration (Kleijer et al., 2009) and it appears to be sustained; effects. The evidence for divalproex sodium is no better; a large
the DART-AD study (Ballard et al., 2009a), a long-term randomized RCT (Tariot et al., 2011) found that divalproex sodium treatment
controlled trial (RCT) study in AD patients on antipsychotics, had no effect on the emergence of agitation or psychosis (or slow-
noted a significant excess of mortality over 24–54 months in peo- ing of cognitive impairment) and indeed was associated with sig-
ple randomized to continue antipsychotic medication compared to nificant toxic side effects.
those randomized to discontinue antipsychotic medication. The potential benefit for carbamazepine remains unclear; two
Comparisons with typical antipsychotics are difficult due to the small reports have suggested benefit from carbamazepine in agita-
lack of randomized evidence, but a retrospective cohort study of tion and aggression (>300 mg/day) (Tariot et al., 1998; Olin et al.,
22,890 older subjects taking typical and atypical antipsychotics 2001), with effect sizes potentially greater than that seen with antip-
found the former group to have a higher mortality (Wang et al., sychotics (Ballard et al., 2009b). Larger studies are needed and the
2005). This finding is supported by a veteran/war widow cohort potential toxicity of carbamazepine in AD may be an issue that lim-
study (Hollis et al., 2007) which found that haloperidol was asso- its the use of this agent.
ciated with an increased risk of death compared with individuals
on olanzapine, although chlorpromazine had comparable levels Cholinesterase inhibitors
of risk to atypicals. Another concern has been whether the use of The fact that cholinesterase inhibitors improve cognition and func-
antipsychotics hastens cognitive decline in dementia (McShane tion in AD is well established. However, their use in the treatment
et al., 1997), and an analysis of the effect of atypical antipsychotic of aggression and agitation is less clear; any improvements have
use (olanzapine, quetiapine, and risperidone) on cognitive func- been noted in secondary analyses of datasets whose primary out-
tion from the CATIE-AD trial (Vigen et al., 2011) found that AD come was an improvement in cognition and/or function (Sink et al.,
patients who received an antipsychotic in the 2 weeks prior to cog- 2005; Gauthier et al., 2010). Specific studies examining cholineste-
nitive testing had an average decline 2.46 points greater on Mini- rase effect on agitation or aggression have been disappointingly
Mental State Examination (MMSE) than placebo patients, although negative. For example, the UK multicentre double-blind placebo
in this study it could not be determined whether this was a perma- controlled trial, CALM-AD (a Randomized Placebo Controlled
nent and enduring effect on cognition or more related to the acute Trial of a Cholinesterase Inhibitor in the Management of Agitation
administration of these agents, e.g. enhancing sedation. in Dementia That is Unresponsive to a Psychological Intervention),
Overall, it is clear there are reasons for concern about the use examined the efficacy of donepezil in treating agitation over a
of antipsychotics in dementia. They are associated with increased 12-week period, but it failed to find any significant improvement in
mortality and risk of cerebrovascular events, as well as possibly its primary outcome measure of agitation (Howard et al., 2007).
worsening cognition. There is also no doubt they are overprescribed Nevertheless, the benefits of cholinesterase inhibitors in amelio-
despite the modest benefits they appear to confer. For example, in rating other neuropsychiatric domains in AD should not be forgot-
the UK, it has been estimated that around 180,000 people with ten; the greatest effects appear to be on apathy, excess motor activity
dementia are on antipsychotic medications in any one year, which (Gauthier et al., 2010), and possibly depression. In addition, judi-
equates to an additional 1800 deaths and an additional 1620 cer- cious use of cholinesterase inhibitors early in the course of AD may
ebrovascular adverse events (Banerjee, 2009). delay or prevent the emergence of behavioural changes that occur
However, an important caveat is that patients entering the clini- with disease progression (Cummings et al., 2004).
cal trials were much less psychotic and disturbed than many seen
in clinical practice, since those who are the most psychotic are Memantine
often too ill to participate. Thus in those with persistent and severe As with cholinesterase inhibitors, on the basis of post hoc analy-
psychosis and associated agitation or aggression their use may be ses of previous RCTs, memantine has been shown to have positive
justified, especially given the even weaker evidence for alternative effects on agitation and aggression (Wilcock et al., 2008); indeed,
treatments (Thomas, 2011). its effects on these neuropsychiatric domains in contrast to the
potential beneficial effects of cholinesterase inhibitors on apa-

Box 33.7 Aetiological factors in Alzheimer’s disease
thy, mood, and motor behaviours have led some to suggest that
memantine combined with a cholinesterase inhibitor may be a suit- Established risk factors
able alternative to antipsychotics, particularly as these agents have Age
a much better tolerability (Lopez et al., 2009; Ballard et al., 2011b; Family history
Herrmann et al., 2011). Down’s syndrome
Conclusion Apolipoprotein E4 allele
Autosomal dominant mutations
Whilst helpful, such summaries of evidence hide many of the dif-
ficulties in interpreting the studies. For example, atypical antipsy- Probable risk factors
chotic studies that were carried out in nursing homes often only Depression
included people with moderate severe dementia (MMSE scores Hypertension
of less than 15), raising the issue of whether these findings can Head injury
be generalized to less cognitively impaired and physically frail Possible risk factors
patients. Conversely patients able to give consent to enter rand- Female gender
omized studies tend to be less psychotic and disturbed than many Low intelligence/education
patients encountered in clinical practice. Finally it is difficult to Diabetes
interpret what changes of a few points on rating scales mean in Smoking
clinical practice. Exposure to aluminium
As a general clinical approach, in patients with psychosis, agi-
tation, or aggression, it is important to ask whether any agitation Possible protective factors
and aggression is being caused by undiagnosed or undertreated Anti-inflammatory medication
pain, constipation, hunger, or dehydration; indeed, the benefits of Apolipoprotein E2 allele
adequate analgesia for agitation in nursing home resident demen- Oestrogen
tia patients have been highlighted in a recent cluster trial (Husebo High intelligence/education
et al., 2011). Environmental factors, caregiver interaction, and
social stimulation (either too much or too little) also need to be
considered. doubling every 5 years. But for how long does such a rise continue?
Nonpharmacological approaches may be helpful, although the It is still not clear whether the increase continues indefinitely (i.e.
evidence for such interventions is very weak (Livingston et al., everyone would develop AD if they lived long enough) or tails off.
2005) (see Chapter 21 for details). For example, the use of Melissa Consistent findings from studies in the oldest old are lacking,
aromatherapy has been advocated in treating agitation/aggression, partly given the limited number of individuals that can be recruited
although a recent rigorous double-blind study failed to show that from these age groups and also problems in early studies with the
aromatherapy was better than a cholinesterase inhibitor or placebo use of appropriately operationalized diagnostic criteria. Early prev-
in treating agitation in AD patients (Burns et al., 2011). alence studies reported that the prevalence of dementia appeared
If a nonpharmacological approach fails, then a cholineste- to plateau for people in their 90s and may even drop in the very old
rase inhibitor or memantine could be considered, particularly if (Fichter et al., 1995; Ritchie and Kildea, 1995). Incidence studies are
the behaviour is mild or moderate, because these drugs appear a better measure of whether age itself is a causal factor in AD and
to have the best risk:benefit ratio. If the behaviour disturbance is dementia, and several studies (Matthews and Brayne, 2005) have
more severe, treatment with an antipsychotic, usually an atypical, countered the above findings, instead suggesting that demen-
should be considered. In view of the risks discussed in the section tia incidence continues to rise exponentially, with one US-based
Antipsychotics, the clinician should explain his reasons for such study reporting annual incidences of dementia of over 20% in 95-
treatment to patients and, if appropriate, their next of kin. to 99-year-olds and greater than 40% in centenarians (Corrada
et al., 2010). However, another prospective incidence study found
that while dementia incidence continues to increase beyond age
Risk Factors for Alzheimer’s Disease 85, the rate of increase appears to slow relative to that of 65- to
Identifying risk factors for AD is important for two main reasons: 85-year-olds (Hall et al., 2005). Overall, further work remains to be
First, identification of risk factors and protective factors ena- done to: (1) clarify the role of ageing in dementia; (2) establish why
bles possible new treatments or public health interventions to be there is a lack of an association between AD and apolipoprotein E
developed. Second, these aetiological factors may throw light on (APOE, see section Apolipoprotein E polymorphism) in the old-
the pathogenesis of AD, stimulating new research that can further est old; and (3) explain the finding that up to half of the oldest old
our understanding of the disease process, leading in turn to new with dementia do not have obvious cerebral pathology to account
treatment opportunities. Box 33.7 summarizes our present state of for their cognitive deficits, suggesting that the pathophysiological
knowledge about the aetiological factors involved in AD. processes which dictate when an individual gets dementia, or more
specifically AD, may vary considerably with age (Juva et al., 2000;
Demographic risk factors Kawas and Corrada, 2006).
Age Sex
It is easily forgotten that old age is the most important risk factor for It is well recognized that AD is more common in women, but is
AD, with AD’s prevalence rising rapidly after 60, and approximately female gender an independent risk factor? It could simply be that
women live longer than men and consequently are at higher risk or environmental, has a role in reducing risk in the Nigerian cohort.
live longer with the illness. As with studies examining age as a risk However, it should be noted that a more recent follow-up study
factor, sex differences in the incidence of AD have been reported in from Nigeria reported a higher incidence rate of dementia than
some but not all studies; one meta-analysis found women to have previously thought (Gureje et al., 2011) and thus the reported dif-
a higher incidence of AD than men but not a higher incidence of ference in the original Hendrie et al. study (2001) may have related
dementia in general (Gao et al., 1998). However, three more recent more to an underestimation of dementia incidence.
large longitudinal studies (Barnes et al., 2003; Corrada et al., 2010; Overall, there remain significant problems with examining AD
Katz et al., 2012) found similar rates between men and women, incidence across different races, cultures, ethnic groups, and coun-
even in the very old. tries, including problems with diagnostic classification, imprecise
Therefore evidence for a sex difference in AD incidence remains diagnostic instruments, and low social expectations in certain
inconclusive; if further research supports such a sex-related risk cultures. In addition, even with better assessment and screening
then it could be related to postmenopausal changes in oestrogens, methodologies, the true prevalence and incidence of AD cannot
genetic polymorphisms of sex-related genes, or an interaction of be established without post-mortem confirmation of diagnosis.
sex with APOE genotype (Gao et al., 1998). Until there are clear biological biomarkers for AD that can be easily
applied in both high and low tech environments, we will remain
Ethnicity, race, country, and culture
uncertain as to the true variability in AD incidence across different
It is estimated that the majority of people with dementia in the racial, ethnic, and cultural groups.
world live in developing countries (60% in 2001, rising to 71% by
2040 as a result of changing demographics, according to Ferri et al. Genetic risk factors
(2005)). However, an outstanding issue is whether the inherent
incidence of AD differs across the world and in different ethnic or Family history
racial groups. It has long been recognized that AD is more common in relatives
A plethora of studies have been published, although the findings of people with AD, and 25–50% of people with AD have an affected
have been contradictory and difficult to interpret. One example of relative. A meta-analysis showed that those with a first-degree rela-
a between country and race comparison was the study by Chandra tive (parent, sibling, child) with AD had a 3.5-fold increase in their
et al. (2001), which found a lower incidence of AD in older per- risk of developing AD (van Duijn et al., 1991). Generally, those
sons living in rural India (4.7/1000 person years) compared to a with an early onset of the disease (before 65 years) show a stronger
predominantly white population in the US (17.5/1000 person familial risk, although early-onset AD only accounts for 1–5% of
years). However, lower social expectations of older people in the all AD cases; conversely, the increased risk is very much reduced
Indian cohort meant that individuals with cognitive decline may for those whose relatives develop AD in very old age. Nevertheless,
not have met functional impairment criteria for dementia, thus the genetic contribution to AD risk is increasingly being revised
underestimating the true level of dementia in the Indian sample. upwards and new risk genes are continually being identified, with
A within country comparison (Manhattan, US) of AD incidence recent estimates suggesting a potential genetic contribution of
between different racial groups by Tang et al. (2001) demonstrated up to 70% in AD (Pedersen et al., 2004; Gatz et al., 2005; Ballard
that African-Americans had a greater risk for AD than white peo- et al., 2011b).
ple (for 65- to 74-year-olds, incidence rate per person-year was Down’s syndrome
1.7% for African-Americans vs 0.4% for Caucasians and 4.4% Virtually all people with Down’s syndrome have the neuropatho-
vs 2.6%, respectively, for 75- to 84-year-olds), even adjusting for logical features of AD by the age of 40 years (Mann et al., 1986) and
confounders such as education and vascular disease. However, the this is probably due to having an extra copy of the amyloid precursor
Cardiovascular Health Cohort (Fitzpatrick et al., 2004) only found protein gene located on chromosome 21. However, the prevalence
marginal differences between African-Americans and white peo- of dementia in people with Down’s is much less than 100%, even
ple, and these authors were cautious about the differences due to by age 50, although it is clearly markedly elevated for age (Zigman
concerns about ascertainment methods applied during the data et al., 1996). The reason for this discrepancy between pathology
collection. and function is not understood and complex interactions between
Ethnicity, which encompasses shared culture, religion, language, maternal age and frequency of the APOE ε4 allele (see next section)
geography, etc., may be a more of powerful contributor to AD risk have been proposed. In addition, the added difficulties of making
than race, given commonalities in life experience and environment an accurate clinical diagnosis in this group are likely to contribute
between people of the same ethnic background; thus epidemiologi- to the variable findings. Dementia in Down’s syndrome is dealt with
cal studies comparing different ethnicity despite similar race may in detail in Chapter 50.
give clues about risk or protective factors, and indeed provide an
insight into the underlying pathophysiology of AD. Apolipoprotein E polymorphism
An important example was the Indianapolis–Ibadan Dementia APOE is a plasma protein involved in lipid transport and its gene
Project which used identical assessment methods in Nigeria and is located on chromosome 19 (19q13.2). It has three common
Indianapolis, and reported an increased incidence of dementia alleles—ε2, ε3, and ε4—which in the normal population (in white
and AD in African-Americans compared to Nigerians (Hendrie people) have allele frequencies of about 7%, 77%, and 16%, respec-
et al., 2001). One possible explanation for this finding was the tively (Zannis et al., 1993). Many studies have shown the ε4 allele
observation that APOE ε4 occurrence in the Nigerian cohort does to be much more common in AD subjects, occurring in 30–50% of
not appear to be associated with increased AD risk (Gureje et al., patients (Roses, 1996), and this allele is thought to enhance amy-
2006), suggesting that some protective mediating factor, potentially loid deposition and impair cholinergic function. The estimated
age-adjusted odds ratios for AD are 2.6 for ε2/ε4 heterozygotes, 3.2 Individuals with higher cognitive reserve as measured by the
for ε3/ε4, and 14.9 for ε4/ε4 homozygotes (Farrer et al., 1997). Thus proxies of education, occupational achievement, or engagement in
APOE ε4 alleles have a dose-dependent effect in increasing the risk leisure activities appeared to have a more rapid decline once the
for AD. Other studies have demonstrated that the increased risk diagnosis of AD was made; this may because people with greater
conferred by APOE ε4 is related to it bringing forward the time of cognitive reserve have higher levels of pathology at the ‘point of
onset of AD (Roses, 1996). However, these striking findings need to inflection’ where memory function begins to decline (see Stern
be put in context. Up to 50% of those who are homozygous for ε4 2009 for discussion). In support of this, a functional neuroimag-
live to beyond 90 without developing AD and about two-thirds of ing study showed regional cerebral blood flow to be lower in bet-
those who develop AD have no ε4 allele (Henderson et al., 1995). ter educated subjects in the parietal and temporal lobes compared
It should also be noted that there is evidence the ε2 allele may be with more poorly educated subjects who had been matched for the
protective, as Farrer et al. (1997) found an odds ratio for AD of 0.6 severity of their dementia, indicating they had more advanced dis-
for the combined ε2/ε2 and ε2/ε3 genotypes. ease (Stern et al., 1992).
What is not clear is how factors such as intelligence and educa-
Autosomal dominant gene mutations and other genes tion affect the risk of incident AD. Is it a case that these factors
Three genetic loci have been identified with mutations conferring lead to more efficient use of brain networks and better synaptic
an autosomal dominant pattern of inheritance for AD with almost transmission, the development of compensatory cognitive strate-
complete penetrance. These are mutations in the amyloid precur- gies, or allow one the ability to recruit alternative brain networks
sor protein gene on chromosome 21, in the presenilin 1 gene on in the face of disease, and so the dementia presents later, or not at
chromosome 14, and in the presenilin 2 gene on chromosome 1. all if death supervenes? Alternatively, does intelligence and educa-
While all are associated with early-onset AD (before 65 years) they tion affect performance on objective cognitive tests and so poorly
account for less than 2% of all AD cases (Farrer et al., 1997) and educated/lower intelligence people cross the thresholds more easily
indeed do not seem to be present in even the majority of early- and better educated/more intelligent people are better able to ‘cover
onset familial AD cases (Cruts et al., 1998). Their importance lies in up’ their deficits on testing?
the clues they may give about the pathogenesis of AD. For example, There is evidence for the former; autopsy studies (Bennett et al.,
these mutations all lead to an excessive production of the patho- 2005; Roe et al., 2008) found that education (and brain volume)
genetic long chain form of the amyloid beta protein (amyloid beta interacts with AD pathology to predict cognitive performance,
42) (Hardy, 1997), and the presenilin 1 protein may be an impor- and a longitudinal study (Roe et al., 2011) found that in individu-
tant enzyme (gamma secretase) involved in producing this amyloid als with normal cognitive function but high levels of CSF tau and
beta 42 protein (De Strooper et al., 1999; Saftig et al., 1999). Recent phosphorylated tau at baseline (biomarkers for AD) the time to
advances have identified a significant number of candidate genes incident cognitive impairment was moderated by education and
including CLU (clusterin), PICALM, and GSK3-beta. These genetic brain volume. Interestingly, one of the earlier religious orders stud-
findings are covered in greater detail in Chapter 8 and interested ies which had followed a group of nuns for many years showed that
readers should consult the online meta-analysis website <http:// not only that those nuns with lower verbal ability in early life (a
www.alzgene.org> for up-to-date information. proxy measure of intelligence) had poorer cognitive functioning
and a higher rate of AD but also at post mortem all of those with
Education, intelligence, mental activities, neuropathological evidence of AD had low verbal ability (Snowdon
and cognitive reserve et al., 1996). Therefore higher intelligence may also possibly pro-
Education, intelligence, and mental activities have all been mooted tect against the pathological process of AD itself. However, a recent
to modify the risk of incident AD. A common element uniting these biomarker study (Vemuri et al., 2011) observed that the association
factors is the notion of cognitive reserve, a hypothetical concept between cognition and the American National Adult Reading Test
where compensatory processes or strategies, related to either brain (a measure of premorbid IQ) was found to be additive rather than
function and structure (e.g. people with larger brains or higher interactive with biomarkers of AD pathology. Clearly, further work
IQ can tolerate more neuronal loss before clinical deficits emerge) is required to clarify the exact relationships between education,
or cognitive adaptations (e.g. higher levels of education enhance intelligence, cognitive reserve, and AD.
MMSE test scores). Beyond this is the concept that mental engage-
ment and activities, such as belonging to a social group, reading, Depression
and playing games, help build upon cognitive reserve and lessen It is generally agreed that in the early stages of the dementia
the risk of dementia (this is discussed in more detail in Physical, process, people often seem to have symptoms of depression,
cognitive, and social activity). and indeed their dementia may present initially with depres-
Valenzuela and Sachdev (2006) carried out a review of pub- sion rather than overt cognitive changes. But is depression an
lished data on the effects of premorbid IQ, education, occupa- independent risk factor for developing dementia or AD? Several
tion, and engagement in mental activities on incident dementia, case-control and cohort studies have examined this in the last
and noted that the majority of studies indicated a protective few years and a thorough meta-analysis (Ownby et al., 2006),
effect of education (10 out 15 studies), occupational attainment identifying studies in which a specific diagnosis of depression
(9 out of 12), high premorbid IQ (2 out 2), and mental activi- was made (rather than depressive symptoms), found 20 good
ties (6 out 6). Combining these as markers of cognitive reserve, quality published studies and reported that any history of
Valenzuela and Sachdev (2006) suggested that individuals with depression doubled the risk of developing AD (odds ratios of
high reserve had a decrease in AD risk of 46% compared to those 2.03 (95% CI: 1.73–2.38) for case-control studies and 1.90 (95%
with low reserve. CI: 1.55–2.33) for cohort studies).
However, as there is often quite a short period between the episode control of their hypertension may be the people with the highest
of depression and the onset of dementia, it is not obvious whether levels of vascular damage (therefore the highest risk of AD accord-
the depression is an intrinsic feature of the dementia or a true risk ing to the angiogenesis hypothesis), thereby actually increasing the
factor for dementia. Nevertheless, a recent prospective study based overall incidence of AD (Kuller and Lopez, 2011).
on the Framingham cohort (Saczynski et al., 2010) found that Therefore, the benefits of blood pressure reduction on dementia
depression was associated with double the risk of dementia and risk remain unclear at the current time. Even considering the treat-
AD in older men and women over a 17-year follow-up period, sup- ment of hypertension earlier on in midlife may not be helpful for
porting the hypothesis that depression is a risk factor for dementia reducing AD. The Uppsala Longitudinal Study of Adult Men, which
rather than the other way around. started in 1970 when subjects were 50 years old and has followed
It is not, however, clear how depression mediates its effect, subjects for up to 40 years, failed to find any association between
although its associations with high cortisol levels, chronic inflam- vascular risk factors and AD incidence (Ronnemaa et al., 2011).
matory markers (e.g. tumour necrosis factor-alpha), and hypoth- However, high systolic blood pressure did appear to be a risk factor
alamic–pituitary–adrenal axis activation have been suggested as for all types of dementia, in particular vascular and mixed demen-
mediating processes. Depression may also have its effect via lifestyle tia, so antihypertensive treatment may be more beneficial for these
factors; reduced physical activity and social engagement occurs in dementias than for AD.
depression and these are both factors associated with an increased Beyond hypertension, other manifestations of vascular disease
risk of dementia. Alternatively, increased vascular pathology associ- including ischaemic heart disease, heart failure, renal failure, atrial
ated with depression (Thomas et al., 2004) could provide the miss- fibrillation, and previous cerebrovascular disease are associated
ing link, as vascular factors are associated with increased AD (see with an increased risk of dementia of all types including AD (Bunch
below). Against this argument is the finding from the Framingham et al., 2010; Dublin et al., 2011).
study (Saczynski et al., 2010) that even controlling for major vas- Elevated total cholesterol in midlife, an established risk factor
cular risk factors did not alter the AD risk level associated with for coronary heart disease and stroke disease, has been associated
depression. with an increased risk of developing dementia in cohort studies
(Kivipelto et al., 2001; Whitmer et al., 2005), although there is a lack
Vascular risk factors of association between high cholesterol in late life and AD (Anstey
Over recent years the debate about the relationship of AD to VaD et al., 2008). Reduction of cholesterol using statins has been pro-
has been complicated or clarified (depending on your view) by evi- posed to modify AD through several mechanisms, in addition to
dence that vascular disease may directly contribute to the pathol- the headline effect of cholesterol lowering, including antioxidative
ogy of AD. The so-called angiogenesis hypothesis proposes that effects, inhibition of butyrylcholinesterase, and increasing the traf-
elevated blood pressure leads to small vessel damage and inflam- ficking of amyloid precursor protein (Darvesh et al., 2004). Two
mation with secondary increases in amyloid and neural loss (Kuller cross-sectional analyses have reported a reduction in dementia
and Lopez, 2011). In fact, this is far from a new idea, as in his origi- in people taking statins (Jick et al., 2000; Wolozin et al., 2000).
nal paper Alois Alzheimer (1907) described vascular pathology at However, concern was expressed about ‘bias by indication’ in these
post mortem in Auguste D. This finding has been confirmed by studies (i.e. healthier, better educated people are more likely to take
more recent pathological data showing that AD patients have a high statins), and this view is supported by the finding in a 5 years pro-
level of white matter lesions and small grey matter infarcts. Both spective study that, whilst on cross-sectional analysis, statin use
APOE and homocysteine (discussed in Homocysteine, vitamin was inversely associated with dementia, statins did not reduce inci-
B12, and folate) are also vascular risk factors and may exert their dent dementia risk (Zandi et al., 2005). Further caution about the
pathogenic effects in AD via vascular disease. It has been proposed potential benefits of statins in reducing AD and dementia arises
that a history of hypertension, as long as 10 and 15 years before the from the findings in two large randomized placebo controlled trials
onset of dementia, increases the rate of AD (Skoog et al., 1996), of two different statins (simvastatin in the Heart Protection Study
and therefore that treatment of systolic hypertension in older peo- (Heart Protection Study Collaborative, 2002) and pravastatin in
ple might reduce the rate of dementia and AD (Forette et al., 1998). the PROSPER study (Shepherd et al., 2002)) which both failed to
However, a recent review of longitudinal studies (Daviglus et al., find any benefits of statins on cognitive function or in reducing
2011) has failed to find any significant difference in the incidence dementia risk.
of AD between subjects with and without hypertension, and pooled Other vascular risk factors that may increase the risk of AD are
analyses examining the treatment effect of antihypertensives on the the metabolic factors obesity and diabetes. Diabetes doubles the
incidence of AD have also found no statistically significant effect risk of vascular dementia and, more specifically, AD (Ahtiluoto
(RR: 0.90, 95% CI: 0.79–1.03) (Guan et al., 2011). et al., 2010), with the risk increased in APOE ε4 carriers. A system-
Unfortunately research into the potential risk reduction effects of atic review by Biessels et al. (2006) noted that a majority of cohort
antihypertensives is hampered by the fact that studies have focused studies have suggested that diabetes is associated with an increased
on older individuals and therefore may have ‘missed the boat’ in risk of incident AD, although limitations in studies such as different
terms of preventing any vascular changes and thus consequent diagnostic criteria for diabetes, lack of information on duration of
amyloid formation. These studies also have several other prob- the diabetes, and the quality of glycaemic control make it difficult
lems, including: (1) inconsistent diagnostic criteria for dementia; to determine what elements of diabetes contribute to increasing the
(2) different definitions of hypertension; (3) focusing on vascular risk of AD.
endpoints rather than on dementia; and (4) probable bias due to Obesity in midlife is linked with an increased incidence of AD
survivor effects, i.e. patients who would normally have died with with risks in the order of 2–5 times in individuals with body mass
cerebrovascular disease but remained alive because of successful indexes more than 30 (Gustafson, 2008; Fitzpatrick et al., 2009);
increased waist to hip ratios may also be a specific risk factor 2004). In addition, early case-control studies suggested that smok-
(Gustafson et al., 2009). However, obesity is often coassociated with ers had a 20% reduction in AD (Graves et al., 1991). However, a
vascular and metabolic disturbances (e.g. hypercholesterolaemia, meta-analysis of 19 prospective studies (Anstey et al., 2007) with at
hyperglycaemia, hypertension, hypertriglyceridaemia, and reduced least 12 months of follow-up has refuted the argument that smok-
levels of high density lipoprotein, i.e. the ‘metabolic’ syndrome) ing may be protective for AD: current smokers had an incident risk
which in themselves have been associated with AD risk, making of 1.79 (95% CI: 1.43, 2.23) for developing AD (with a comparable
it difficult to disentangle the unique contribution of obesity to AD level of risk for incident vascular dementia) compared to nonsmok-
risk. Indeed, these vascular/metabolic risk factors may be additive ers. Incidence risks were similar when comparing current vs former
in their effects. For example, in one study, combining obesity with smokers (pooled RR = 1.70; 95% CI, 1.25–2.31). But the view that
hypertension and high cholesterol led to a much larger combined nicotine may be neuroprotective has received support from rand-
risk (OR 6.2) (Kivipelto et al., 2005), and in another prospective omized trials reporting benefits of nicotine therapy on cognition
study (over 5.5 years) diabetes, smoking, hypertension, and heart (e.g. Newhouse et al., 2012), although whether it reduces the devel-
disease all increased the risk of AD, but their combinative effect was opment of AD or dementia remains to be demonstrated.
much stronger, increasing the risk by over three-fold (Luchsinger
Homocysteine, vitamin B12, and folate
et al., 2005).
Cognitive impairment is associated with deficiencies in folate
Head injury and vitamin B12. Homocysteine is an amino acid intermedi-
ary in methionine metabolism and its blood levels increase with
A meta-analysis of 15 case-control studies of head injury (with loss
deficiencies in folate and B12 because its elimination is depend-
of consciousness) showed it to be associated with an increased risk
ent on these vitamins, and the risk of cognitive impairment and
of AD (OR 1.58, 95% CI 1.21–2.06) (Fleminger et al., 2003) and this
dementia associated with their deficiencies may be related to raised
risk appears to be specific to men. However, one major criticism
homocysteine levels.
is that early studies used retrospective assessments from relatives,
High homocysteine levels appear to occur in AD compared to
raising the risk of recollection bias. Interestingly, however, a study
healthy older people, although a meta-analysis of prospective data
using a prospective historical cohort design of World War II vet-
has not provided conclusive evidence that homocysteine enhances
erans who had documented nonpenetrating head injuries demon-
the risk of developing AD (Ho et al., 2011). In addition, system-
strated that both moderate and severe head injury were associated
atic reviews of intervention studies have found no evidence for any
with a significant risk of AD (Plassman et al., 2000).
benefit on cognition or dementia using B12 with and without folate
Further evidence for head injury effects on AD risk come from
(Malouf and Grimley Evans, 2009; Ford and Almeida, 2012).
pathology in that head injury leads to a disturbed balance between
amyloid beta genesis and catabolism, leading to an increase in amy- Oestrogen
loid beta deposits. Indeed, amyloid deposition has been reported to Several early systematic reviews and meta-analyses of cohort and
occur within hours of the injury (Johnson et al., 2010). case-control studies (Yaffe et al., 1998; LeBlanc et al., 2001; Nelson
et al., 2002) reported oestrogen use to be associated with a reduc-
Other possible risk factors tion in the risk of developing AD of 29–34%. However, randomized
Metals trials, e.g. (Henderson et al., 2000; Mulnard et al., 2000; Wang et al.,
Studies some years ago showed increased aluminium in the brains 2000) have failed to find any clinically meaningful evidence of ben-
of subjects with AD and suggested that aluminium is neurotoxic efit in treating patients with mild to moderate AD with oestrogen.
(Crapper et al., 1973; Trapp et al., 1978). This led to the suggestion A large primary prevention trial, the Women’s Health Initiative
that it may play a role in the causation of AD, supported by some Memory Study (WHIMS), examined the possible benefit of HRT/
epidemiological evidence of higher rates of AD in areas with high ERT in reducing the frequency of, or time of, onset of dementia in
aluminium content in the drinking water (Doll, 1993). However, postmenopausal women, but adverse outcomes led to both arms
other studies have not confirmed these early findings and it is being terminated early (treatment led to increased rates of stroke,
clear aluminium is neither necessary nor sufficient to cause AD coronary heart disease, venous thromboembolism, and breast carci-
(Walton, 1991). noma). The use of unopposed oestrogen over 7 years was associated
More recently, attention has switched to other metals includ- with a nonsignificant increased risk of dementia (hazard ratio (HR)
ing iron, zinc, and copper, whose levels in the brain appear to be = 1.49; 95% CI 0.83–2.66) (Shumaker et al., 2004), and treatment
raised and whose homeostasis is disordered in AD. This has led with combined oestrogen and progestin over 4 years led to a dou-
to the hypothesis that these metals may increase oxidative damage bling of dementia risk (HR = 2.05; 95% CI 1.21–3.48) (Hays et al.,
and the formation of amyloid beta plaques (Maynard et al., 2005). 2003). Combining these two groups, there was a highly clinically
However, the role of metallobiology in the pathogenesis of AD is and statistically significant increase in dementia in women taking
far from resolved, as research findings are often controversial and HRT (HR = 1.76; 95% CI 1.19–2.60) (Shumaker et al., 2004).
contradictory (Schrag et al., 2011). However, it has been suggested that there may be a ‘critical win-
dow’ around the perimenopausal or early postmenopausal period,
Smoking in other words, before significant Alzheimer’s neuropathology has
It has been hypothesized that smoking increases nicotinic recep- developed, during which oestrogen therapy may be beneficial in
tors in the cortex and this could counteract the cholinergic deficit enhancing cognition, whereas later oestrogen treatment has nega-
in AD. The role of nicotine as a potential neuroprotective agent in tive effects. Support for this hypothesis was demonstrated in an
AD has received support from the finding that nicotine treatment observational cohort study (Whitmer et al., 2011) involving 1524
reduces beta-amyloid deposition in mice (Hellstrom-Lindahl et al., women which found that those who took hormone therapy at
midlife had a 26% decreased risk of getting dementia compared to (Kang et al., 2006) found no difference between treatment and con-
those who never took hormone therapy, whereas those taking hor- trol groups on measures of cognitive function after 5.6 years and
mone therapy only in late life had a 48% increased risk and women 9.6 years of treatment. In summary, there is no good evidence to
taking hormone therapy at both midlife and late life had a similar support vitamin E supplementation to either treat or prevent AD.
risk of dementia. However, as with similar studies of AD, risk factors Omega-3 fatty acids, frequently obtained from fish oils, have
confounded by indication (healthier better educated women with been similarly mooted to have a range of health benefits mediated
lower risk for dementia take HRT) might explain these findings. by their modulatory effects on inflammatory pathways. However,
their potential effect on the incidence of AD has not been conclu-
Frailty and weight loss
sively established (see Daviglus et al. (2011) for discussion).
Reduced muscle strength and general frailty have been associ- General dietary changes may, however, offer some benefits
ated with an increased risk of dementia, and while midlife obes- in reducing the risk of AD. Emerging evidence has suggested
ity with high BMI is associated with dementia risk, this situation that medium to high fruit and vegetable intake or adherence
reverses in old age, with low BMI being associated with incident to a Mediterranean diet may reduce the risk of AD occurrence
AD (Fitzpatrick et al., 2009); however, it is possible that these fac- (Daviglus et al., 2011), although caution needs to be applied to the
tors are actually markers of dementia pathology rather than being findings of these dietary studies as individuals who take additional
causative, and thus are not true risk factors. nutritional supplements and/or eat healthily also tend to be bet-
Protective factors ter educated and more physically fit (see Physical, cognitive, and
social activity).
Anti-inflammatory medication
A review of case-control studies has suggested arthritis, nonsteroi- Physical, cognitive, and social activity
dal anti-inflammatory drugs, and steroids to be protective factors The cardiovascular benefits of physical activity are well established
for AD, reducing the risk by about 50% (McGeer et al., 1996). This and, given the putative linkage between vascular risk factors and
fits with current views that AD is associated with an inflammatory AD as well as evidence in transgenic animal studies that physical
response in the brain because senile plaques are associated with activity reduces AD pathology, it is not unreasonable to hypoth-
prominent astrocytosis and microglial activation (Dickson et al., esize that physical activity may reduce AD risk. A meta-analysis of
1988) and pro-inflammatory cytokines are increased in the brain nine cohort studies (Daviglus et al., 2011) found that higher levels
in AD subjects (Wood et al., 1993). A dampener for enthusiasm of physical activity were associated with lower risk of incident AD
about the role of inflammation in AD is that all RCTs using anti- (HR = 0.72; 95% CI, 0.53–0.98). However, such studies often relied
inflammatory drugs, including ibuprofen, prednisolone, naproxen, on self-reports of physical activity rather than objective meas-
and COX inhibitors, in established AD have been negative (Aisen ures, associations were not always statistically significant, and in
et al., 2000; Aisen et al., 2003; Imbimbo, 2004; Reines et al., 2004; some studies physical activity appeared to increase the risk of AD
Lyketsos et al., 2007). (Daviglus et al., 2011). Nevertheless, a recent Australian prospec-
tive study of older individuals (50+ years) randomly allocated to an
Vitamins, antioxidants, and nutritional supplements
education and usual care group or to a 24-week home-based pro-
Oxidative stress is thought to be important in the development of gramme of physical activity demonstrated sustained benefits of the
dementia and neurons are especially vulnerable to oxidative stress exercise intervention up to 18 months, in terms of improvements
because the brain has a high oxygen consumption and a lack of in global cognition and delayed recall (Lautenschlager et al., 2008).
antioxidant enzymes, e.g. superoxide dismutase (SOD-1), com- This benefit may be more marked in those without the APOE ε4
pared with other organs (Esposito et al., 2002), and is more depend- allele(Lautenschlager et al., 2008). Whether this improvement in
ent on dietary antioxidants, including tocopherols, ascorbic acid, cognition is due to a delayed or prevented onset of AD remains to
carotenoids, vitamin A, and flavonoids. Although using vitamins be established. It is not clear what intensity of exercise, its duration,
as antioxidants has been frequently claimed to be protective against and optimal time to do the exercise during the lifespan confer the
AD, cross-sectional and cohort studies have reported mixed find- maximum benefit.
ings on the possibility of nutritional and/or supplementary dietary Cognitive engagement and leisure activity involvement may also
antioxidants reducing the risk of dementia (Engelhart et al., 2002; have protective effects, perhaps by improving cognitive reserve,
Morris et al., 2002; Tabet et al., 2002; Larrieu et al., 2004; Engelhart although the evidence base is still limited. Studies in this area suf-
et al., 2005). fer from the inevitable inclusion of individuals who tend to be bet-
A randomized trial of vitamin E supplementation in AD (in ter educated and healthier, factors as already discussed, which in
moderately impaired subjects) reported that after adjustment for themselves reduce AD risk. Furthermore, it may be that in individ-
baseline MMSE (but not before), vitamin E delayed time to a com- uals with prodromal AD, the direct effects of the disease cause early
posite endpoint by 230 days (Sano et al., 1997). However, there social disengagement and reduced motivation and drive, which
were no differences in a range of secondary outcome measures may accentuate the apparent ‘difference’ between healthy individu-
including ADAS-Cog and MMSE. Another randomized trial of als who are socially and cognitively active compared to those who
769 subjects compared progression to possible or probable AD in go on to develop AD.
subjects with amnestic mild cognitive impairment (Petersen et al.,
2005). Subjects were randomized to vitamin E (2000 IU), donepezil Alcohol
(10 mg), or placebo for 3 years and no difference in outcome was Meta-analyses of alcohol use have suggested there may be a
observed, casting doubt on the benefits of antioxidant treatment U-shaped relationship between alcohol consumption and dementia
in preventing or slowing the development of AD. Similarly, a large risk (Anstey et al., 2009; Weuve et al., 2012), such that individuals
RCT of 6377 older women on 600 IU of vitamin E on alternate days who drink mild to moderate amounts of alcohol have a lower risk
of AD and all types of dementia. Part of this effect is thought to Death

be mediated through the indirect effect on cardiovascular health Time from diagnosis to death is highly variable, with the mean
or as part of antioxidant process (e.g. high levels of antioxidants length of survival from diagnosis ranging from a year up to 20 years.
are present in red wine), although the exact neurobiological under- This is highly dependent on how early the diagnosis is made,
pinnings are not known. Higher levels of alcohol intake are poten- although certain other factors, e.g. age at onset, are also important
tially neurotoxic and hence may increase the risk of dementia. (see Predicting rates of decline). The median life-expectancy after
Controversies still to be resolved on the benefits of alcohol include diagnosis appears to be about 5–10 years for patients diagnosed
the optimum quantity, how frequently the alcohol is consumed, in with AD in their 60s and 70s (equivalent to approximately 70%
which part of the life-course is the risk reduction effect most opti- reduction in expected median lifespan) to only 3–4 years if patients
mal (e.g. midlife vs old age), as well as the type of alcohol consump- are in their 90s (an approximate 40% reduction in expected median
tion (Weuve et al., 2012). lifespan) (Xie et al., 2008; Zanetti et al., 2009). Unsurprisingly, the
mildest patients at outset tend to show the longest times to both
institutionalization and death.
Prognosis of Alzheimer’s Disease The cause of death in AD can be difficult to determine. As with
The natural history of Alzheimer’s disease other chronic diseases in older people, many people will die with
Every old age psychiatrist is familiar with the considerable vari- the disease rather than from it. A Swedish autopsy study based on
ability in the pattern, progress, and outcome of AD. All the stud- data from 1974–2004 found that the most common cause of death
ies following the natural history of the disease bear this out as in AD patients was bronchopneumonia (47.3%) (Brunnstrom and
they show substantial variations in symptomatology and rates of Englund, 2009). Difficulties in the terminal stages with feeding and
decline. The practical consequence of this for the clinician is that poor airway protection may be contributory. Since AD is associated
it is very difficult to give accurate predictions to patients or their with marked weight loss and physical decline (Cronin-Stubbs et al.,
families about the likely progress of the disease after the diagno- 1997), these are likely to contribute in their own right to death in
sis has been made. Following the introduction of cholinesterase many people with advanced disease, although frequently AD is not
inhibitors there has been a trend towards earlier referral and even mentioned on the death certificate (Burns et al., 1990e).
diagnosis. This may introduce new challenges, such as requir- Whether AD directly causes death is less certain, although it
ing more investigations and longer periods of evaluation, along seems plausible as some patients do simply fade away without any
with reducing the certainty of diagnosis. As our knowledge of other obvious cause for death occurring. Many others develop other
the longer-term effects of these drugs is still incomplete, this too complications of immobility and physical debilitation and die of a
makes prognostic predictions more imprecise. However, earlier combination of causes.
diagnosis will of course produce a (spurious?) increase in the
length of time someone has AD, which, in that sense, will make Measuring rates of decline
the prognosis appear better. However, despite these uncertainties, Two kinds of instrument (global rating scales and cognitive scales)
patients and relatives will continue to seek information on prog- have been used to try to measure the rate of decline in AD. Global
nosis, and two of the most common outcomes of interest are time measures, e.g. the Clinical Dementia Rating Scale (CDR) (Hughes
to institutionalization into residential or nursing home care and et al., 1982), tend to change very slowly over a few years and so are
time to death. not useful measures of change clinically. Cognitive scales do show
significant changes over the course of a year or so. Some of these are
Institutionalization in common use and familiarity with the data derived from research
Data compiled by the National Alzheimer Coordinating Centre may help a clinician to see how rapidly any patient is declining com-
(Spackman et al., 2011) have suggested annual probabilities of pared with these norms and therefore aid the prediction of progress
institutionalization of 1.2, 3.4, and 6.6% for mild, moderate, and and outcome, although the substantial variability should always be
severe AD dementia, respectively. A higher overall annual inci- borne in mind. The annual rate of change in the MMSE (Folstein
dence of 11.84% was observed in a 2-year prospective French study et al., 1975) is about 2.5 points a year for moderately affected
(REAL-FR (Gillette-Guyonnet et al., 2011)), thus highlighting the patients, with higher rates for more severely affected patients and
wide range of progression evident even in reasonably powered lower rates for milder cases (Salmon et al., 1990). The REAL-FR
studies. Established predictors of institutionalization include the study (Cortes et al., 2008) noted that after 2 years 11% of patients
presence of behavioural symptoms, severe impairments in activi- lost more than 9 points on the MMSE, 66% lost 3–9 points, but a
ties of daily living and cognition, and living alone, as well as depres- substantial minority (23%) remained stable or improved, highlight-
sion (Knopman et al., 1988). Caregiver factors also probably have ing the heterogeneity in cognitive decline among AD patients as
a significant role in predicting early nursing home placement. For well as evidence of plateauing.
example, a prospective study (Gaugler et al., 2003) identified that Two more comprehensive instruments are the CAMCOG (the
older age of carers (more than 80 years) and high perception of care cognitive component of the Cambridge Mental Disorders of the
burden by the carers themselves predicted higher early institution- Elderly (CAMDEX)) and the ADAS-Cog (the Alzheimer’s Disease
alization. This study and other evidence show that there is a complex Assessment Scale—cognitive subscale). Burns et al. (1991b) used
interaction between patient and caregiver factors that influences the the CAMCOG to follow the progress of 110 AD patients. They
course of the pathway to nursing care. Thus it is always important to found a drop of about 12 points in the first year (maximum score
consider the needs of caregivers in the assessment of management 107 with 80/79 as the cutoff for dementia). This included change in
of patients with dementia (Gaugler et al., 2011). every main subsection, but many of the scores were already near
the floor at baseline and this instrument appears less useful in more 2005; Vigen et al., 2011) and, as discussed earlier, are associated
advanced dementia. A couple of older studies have used the ADAS- with increased mortality (Schneider et al., 2005). They have also
Cog (maximum score 70, higher scores = worse performance) been shown to adversely affect patients with DLB (Dickson et al.,
and have shown an annual rate of increase of 8–9 points (Kramer- 1988; McKeith et al., 1992; Ballard et al., 1998). It appears that whilst
Ginsberg et al., 1988; Yesavage et al., 1988). A more recent finding these widely prescribed drugs do show benefits in the short term
from the REAL-FR study reported a slower rate of progression of for specific symptoms, there is a danger they may worsen patient
an average of a 4.5 point gain per year on the ADAD-Cog over a status overall. The evidence suggests they should only be used with
4-year period (Gillette-Guyonnet et al., 2011), which may reflect caution and they need close monitoring and regular review, with
changes in management of AD (memory clinics and availability of the aim of discontinuing treatment as soon as is feasible to reduce
symptomatic drugs) over the past 15–20 years. the longer-term risks.
What about cholinesterase inhibitors? At the outset here, we
Predicting rates of decline should remind ourselves these drugs clearly improve the progno-
Within the rough framework of the above figures for rates of decline sis of AD in the short term for the 50–60% who respond to them
and time to institutionalization and death, are there any clinical (see Chapter 38). Patients enjoy improved cognitive function, a
features that can help sharpen the prediction in individual cases? reduction in disturbed behaviour, and important gains in func-
Certainly, the more severe the dementia the more rapid is the rate tion in everyday life, and such benefits may continue for several
of decline (Morris et al., 1993). Consistent with this, the decline years on treatment. Memantine similarly may confer some subtle
is nonlinear, being slower at the milder stage but accelerating as symptomatic benefits in moderate dementia on global assessment
the dementia progresses (Teri et al., 1995). Age of onset is also a scales and cognition (Schneider et al., 2011). However, there is lit-
recognized predictor of deterioration, with early-onset dementias tle evidence at present as to whether cholinesterase inhibitors or
showing more rapid cognitive and functional decline (Jacobs et al., memantine slow down the disease process itself, and drug trials to
1994; Teri et al., 1995). demonstrate this will be very hard to carry out.
In some studies, the level of education has been demonstrated to
be a predictor of decline (see discussion in Physical, cognitive, and Conclusion
social activity).
The pattern and progression of AD is highly variable. Predictors of
Extrapyramidal symptoms appear to predict more rapid dete-
decline are greater severity of illness and early age of onset, which
rioration in AD, although results are complicated by the likely
both predict a more rapid course, and extrapyramidal and psy-
admixture in earlier studies of patients with AD and DLB (Miller
chotic symptoms, which probably do so too. Whilst antipsychotic
et al., 1991). However, a more recent study in a purer AD group
drugs are of benefit in the short term in treating some non-cog-
found that extrapyramidal symptoms were associated with greater
nitive symptoms, their long-term use may worsen the prognosis;
functional decline and more rapid institutionalization, although
cholinesterase inhibitors, on the other hand, improve prognosis in
there was no association with cognitive decline (Lopez et al., 1997).
the short term but their effects in the longer term on the disease
Gait apraxia may also be a marker of poor survival (Paradise et al.,
are unclear. There remains a pressing need to develop new sympto-
2009), although whether gait apraxia is a reflection or due to more
matic therapies in AD as well as viable disease-modifying agents.
severe brain disease and thus shorter survival or that it leads to
survival-shortening falls or immobility is not clear. Several studies References
have found psychotic symptoms to predict more rapid deteriora-
Ahtiluoto, S., et al. (2010). Diabetes, Alzheimer disease, and vascular
tion, although when groups have been matched for severity of AD dementia: a population-based neuropathologic study. Neurology, 75,
at outset the findings have been mixed. Some studies have found 1195–202.
that psychosis does still predict rate of decline (Chui et al., 1994; Aisen, P. S., et al. (2000). A randomized controlled trial of prednisone in
Stern et al., 1994), but others have not found this (Reisberg et al., Alzheimer’s disease. Alzheimer’s Disease Cooperative Study. Neurology,
1986; Chen et al., 1991). One longitudinal study has not found 54, 588–93.
depression to predict decline (Lopez et al., 1990), but, unsurpris- Aisen, P. S., et al. (2003). Effects of rofecoxib or naproxen vs placebo on
ingly, it has been shown to be associated with greater impairment Alzheimer disease progression: a randomized controlled trial. Journal of
on activities of daily living (Pearson et al., 1989). Both psychotic the American Medical Association, 289, 2819–26.
and depressive symptoms are clinically important regardless of Alzheimer, A. (1907). Uber eine eigenartig Erkrankung der Hirnrinde.
Allgemeine Zeitschrift für Psychiatrie und Psychisch-Gerichtliche
whether they predict deterioration, because they increase the suf- Medizin, 64, 146–148.
fering of the patients and their carers and, as discussed, they should Anstey, K. J., et al. (2007). Smoking as a risk factor for dementia and cognitive
be treated in their own right. decline: a meta-analysis of prospective studies. American Journal of
Other factors that have been investigated without a clear con- Epidemiology, 166, 367–78.
clusion include sex, APOE genotype, and concomitant cerebrovas- Anstey, K. J., Lipnicki, D. M. and Low, L. F. (2008). Cholesterol as a risk factor
cular disease. Similarly, data (see previous sections) on oestrogen for dementia and cognitive decline: a systematic review of prospective
replacement, lipid-lowering agents, vitamins (B6, B12, E), omega-3 studies with meta-analysis. American Journal of Geriatric Psychiatry,
fatty acid supplementation, and NSAIDs have not indicated that 16, 343–54.
Anstey, K. J., Mack, H. A. and Cherbuin, N. (2009). Alcohol consumption
they alter the rate of decline.
as a risk factor for dementia and cognitive decline: meta-analysis of
prospective studies. American Journal of Geriatric Psychiatry, 17, 542–55.
The effects of drug treatment on prognosis
Bains, J., Birks, J. S. and Dening, T. D. (2002). Antidepressants for treating
Antipsychotic drugs, both typical and atypical, appear to worsen depression in dementia. Cochrane Database Systematic Reviews, 4,
cognitive decline in dementia (McShane et al., 1997; Ballard et al., CD003944.
Ballard, C. G. and Walker, M. (1999). Neuropsychiatric aspects of Alzheimer’s Burns, A., et al. (1991c). Factors affecting survival in Alzheimer’s disease.
disease. Current Psychiatry Reports, 1, 49–60. Psychological Medicine, 21, 363–70.
Ballard, C. G., Bannister, C., and Oyebode, F. (1996). Depression in dementia Burns, A., et al. (2011). A double-blind placebo-controlled randomized trial
sufferers. International Journal of Geriatric Psychiatry, 11, 507–15. of Melissa officinalis oil and donepezil for the treatment of agitation
Ballard, C., et al. (1998). Neuroleptic sensitivity in dementia with Lewy in Alzheimer’s disease. Dementia and Geriatric Cognitive Disorders,
bodies and Alzheimer’s disease. Lancet, 351, 1032–3. 31, 158–64.
Ballard, C., et al. (2005). Quetiapine and rivastigmine and cognitive decline Chandra, V., et al. (2001). Incidence of Alzheimer’s disease in a rural
in Alzheimer’s disease: randomised double blind placebo controlled trial. community in India: the Indo-US study. Neurology, 57, 985–9.
British Medical Journal, 330, 874. Chen, J. Y., et al. (1991). Cumulative risks of developing extrapyramidal signs,
Ballard, C., Waite, J., and Birks, J. (2006). Atypical antipsychotics for psychosis, or myoclonus in the course of Alzheimer’s disease. Archives of
aggression and psychosis in Alzheimer’s disease. Cochrane Database Neurology, 48, 1141–3.
Systematic Reviews, 1, CD003476. Chui, H. C., et al. (1994). Extrapyramidal signs and psychiatric symptoms
Ballard, C., et al. (2009a). The dementia antipsychotic withdrawal trial predict faster cognitive decline in Alzheimer’s disease. Archives of
(DART-AD): long-term follow-up of a randomised placebo-controlled Neurology, 51, 676–81.
trial. Lancet Neurology, 8, 151–7. Cooper, J. K., Mungas, D. and Verma, M. (1991). Psychotic symptoms in
Ballard, C. G., et al. (2009b). Management of agitation and aggression Alzheimer’s disease. International Journal of Geriatric Psychiatry, 6, 721–6.
associated with Alzheimer disease. Nature Review Neurology, Corrada, M. M., et al. (2010). Dementia incidence continues to increase with
5, 245–55. age in the oldest old: the 90 + study. Annals of Neurology, 67, 114–21.
Ballard, C., et al. (2011a). Atypical antipsychotics for the treatment of Cortes, F., et al. (2008). Prognosis of Alzheimer’s disease today: a two-year
behavioral and psychological symptoms in dementia, with a particular prospective study in 686 patients from the REAL-FR Study. Alzheimer’s
focus on longer term outcomes and mortality. Expert Opinion on Drug Dementia, 4, 22–9.
Safety, 10, 35–43. Craig, D., et al. (2005). A cross-sectional study of neuropsychiatric symptoms
Ballard, C., et al. (2011b). Alzheimer’s disease. Lancet, 377, 1019–31. in 435 patients with Alzheimer’s disease. American Journal of Geriatric
Banerjee, S. (2009). The use of antipsychotic medication for people with Psychiatry, 13(6), 460–8.
dementia: time for action. A report for the Minister of State for Care Crapper, D. R., Krishnan, S. S. and Dalton, A. J. (1973). Brain aluminum
Services. Department of Health, London. distribution in Alzheimer’s disease and experimental neurofibrillary
Banerjee, S., et al. (2011). Sertraline or mirtazapine for depression in degeneration. Transactions of the American Neurological Association,
dementia (HTA-SADD): a randomised, multicentre, double-blind, 98, 17–20.
placebo-controlled trial. Lancet, 378, 403–11. Cronin-Stubbs, D., et al. (1997). Weight loss in people with Alzheimer’s
Barnes, L. L., et al. (2003). Gender, cognitive decline, and risk of AD in older disease: a prospective population based analysis. British Medical Journal,
persons. Neurology, 60, 1777–81. 314, 178–9.
Bennett, D. A., et al. (2005). Education modifies the association of amyloid Cruts, M., et al. (1998). Estimation of the genetic contribution of presenilin-1
but not tangles with cognitive function. Neurology, 65, 953–5. and -2 mutations in a population-based study of presenile Alzheimer
Berman, K., et al. (2012). Pharmacologic treatment of apathy in dementia. disease. Human Molecular Genetics, 7, 43–51.
American Journal of Geriatric Psychiatry, 20, 104–22. Cummings, J. L., et al. (2004). Reduction of behavioral disturbances and
Biessels, G. J., et al. (2006). Risk of dementia in diabetes mellitus: a systematic caregiver distress by galantamine in patients with Alzheimer’s disease.
review. Lancet Neurology, 5, 64–74. American Journal of Psychiatry, 161, 532–8.
Brodaty, H. and Burns, K. (2012). Nonpharmacological management of Darvesh, S., et al. (2004). Differential effects of lipid-lowering agents on
apathy in dementia: a systematic review. American Journal of Geriatric human cholinesterases. Clinical Biochemistry, 37, 42–9.
Psychiatry, 20(7), 548–64. Daviglus, M. L., et al. (2011). Risk Factors and Preventive Interventions
Brunnstrom, H. R. and Englund, E. M. (2009). Cause of death in patients for Alzheimer Disease: State of the Science. Archives of Neurology,
with dementia disorders. European Journal of Neurology, 16, 48892. 68, 1185–90.
Bunch, T. J., et al. (2010). Atrial fibrillation is independently associated with De Strooper, B., et al. (1999). A presenilin-1-dependent gamma-secretase-
senile, vascular, and Alzheimer’s dementia. Heart Rhythm, 7, 433–7. like protease mediates release of Notch intracellular domain. (See
Burns, A., Jacoby, R., and Levy, R. (1990a). Psychiatric phenomena in Comment.) Nature, 398, 518–22.
Alzheimer’s disease. IV: Disorders of behaviour. British Journal of Dickson, D. W., et al. (1988). Alzheimer’s disease. A double-labeling
Psychiatry, 157, 86–94. immunohistochemical study of senile plaques. American Journal of
Burns, A., Jacoby, R., and Levy, R. (1990b). Psychiatric phenomena in Pathology, 132, 86–101.
Alzheimer’s disease. III: disorders of mood. British Journal of Psychiatry, Doll, R. (1993). Review: Alzheimer’s disease and environmental aluminium.
157, 81–6, 92–4. (See Comment.) Age and Ageing, 22, 138–53.
Burns, A., Jacoby, R., and Levy, R. (1990c). Psychiatric phenomena in Drye, L. T., et al. (2011). Do treatment effects vary among differing baseline
Alzheimer’s disease. II: disorders of perception. British Journal of depression criteria in depression in Alzheimer’s disease study +/– 2
Psychiatry, 157, 76–81, 92–4. (DIADS-2)? International Journal of Geriatric Psychiatry, 26, 573–83.
Burns, A., Jacoby, R., and Levy, R. (1990d). Psychiatric phenomena in Dublin, S., et al. (2011). Atrial fibrillation and risk of dementia: a prospective
Alzheimer’s disease. I: Disorders of thought content. British Journal of cohort study. Journal of the American Geriatric Society, 59, 1369–75.
Psychiatry, 157, 72–6, 92–4. Dubois, B., et al. (2007). Research criteria for the diagnosis of Alzheimer’s
Burns, A., et al. (1990e). Cause of death in Alzheimer’s disease. Age and disease: revising the NINCDS-ADRDA criteria. Lancet Neurology,
Ageing, 19, 341–4. 6, 734–46.
Burns, A., et al. (1990f). Accuracy of clinical diagnosis of Alzheimer’s disease. Dubois, B., et al. (2010). Revising the definition of Alzheimer’s disease: a new
British Medical Journal, 301, 1026. lexicon. Lancet Neurology, 9, 1118–27.
Burns, A., Jacoby, R., and Levy, R. (1991a). Neurological signs in Alzheimer’s Duthie, A., Chew, D., and Soiza, R. L. (2011). Nonpsychiatric comorbidity
disease. Age and Ageing, 20, 45–51. associated with Alzheimer’s disease. QJM, 104, 913–20.
Burns, A., Jacoby, R., and Levy, R. (1991b). Progression of cognitive Engelhart, M. J., et al. (2002). Dietary intake of antioxidants and risk
impairment in Alzheimer’s disease. Journal of the American Geriatrics of Alzheimer disease. Journal of the American Medical Association,
Society, 39, 39–45. 287, 3223–9.
Engelhart, M. J., et al. (2005). Plasma levels of antioxidants are not associated Gureje, O., et al. (2006). APOE epsilon4 is not associated with Alzheimer’s
with Alzheimer’s disease or cognitive decline. Dementia and Geriatric disease in elderly Nigerians. Annals of Neurology, 59, 182–5.
Cognitive Disorders, 19, 134–9. Gureje, O., et al. (2011). Incidence of and risk factors for dementia in the
Esposito, E., et al. (2002). A review of specific dietary antioxidants and Ibadan study of aging. Journal of the American Geriatrics Society,
the effects on biochemical mechanisms related to neurodegenerative 59, 869–74.
processes. Neurobiology of Aging, 23, 719–35. Gustafson, D. (2008). A life course of adiposity and dementia. European
Farrer, L. A., et al. (1994). Interrater agreement for diagnosis of Alzheimer’s Journal of Pharmacology, 585, 163–75.
disease: the MIRAGE study. Neurology, 44, 652–6. Gustafson, D. R., et al. (2009). Adiposity indicators and dementia over
Farrer, L. A., et al. (1997). Effects of age, sex, and ethnicity on the association 32 years in Sweden. Neurology, 73, 1559–66.
between apolipoprotein E genotype and Alzheimer disease. A meta- Hall, C. B., et al. (2005). Dementia incidence may increase more slowly after
analysis. APOE and Alzheimer Disease Meta Analysis Consortium. (See age 90. Neurology, 65, 882–86.
Comment.) Journal of the American Medical Association, 278, 1349–56. Hardy, J. (1997). Amyloid, the presenilins and Alzheimer’s disease. Trends in
FDA (2005). US Food and Drug Administration, Public Health Advisory: deaths Neurosciences, 20, 154–9.
with antipsychotics in elderly patients with behavioural disturbances. FDA, Hays, J., et al. (2003). Effects of estrogen plus progestin on health-related
Silver Spring. quality of life. New England Journal of Medicine, 348, 1839–54.
Ferri, C. P., et al. (2005). Global prevalence of dementia: a Delphi consensus Heart Protection Study Collaborative (2002). MRC/BHF Heart Protection
study. Lancet, 366, 2112–17. Study of cholesterol lowering with simvastatin in 20,536 high-risk
Fichter, M. M., et al. (1995). Dementia and cognitive impairment in the individuals: a randomised placebo-controlled trial. (See Comment.)
oldest old in the community. Prevalence and comorbidity. British Journal (Summary for patients in Current Cardiology Report, 4(6), 486–7; PMID:
of Psychiatry, 166, 621–9. 12379169.) Lancet, 360, 7–22.
Fitzpatrick, A. L., et al. (2004). Incidence and prevalence of dementia in the Hellstrom-Lindahl, E., et al. (2004). Nicotine reduces A beta in the brain
Cardiovascular Health Study. Journal of the American Geriatrics Society, and cerebral vessels of APPsw mice. European Journal of Neuroscience,
52, 195–204. 19, 2703–10.
Fitzpatrick, A. L., et al. (2009). Midlife and late-life obesity and the risk Henderson, A. S., et al. (1995). Apolipoprotein E allele epsilon 4, dementia,
of dementia: cardiovascular health study. Archives of Neurology, and cognitive decline in a population sample. Lancet, 346, 1387–90.
66, 336–42. Henderson, V. W., et al. (2000). Estrogen for Alzheimer’s disease in women:
Fleminger, S., et al. (2003). Head injury as a risk factor for Alzheimer’s disease: randomized, double-blind, placebo-controlled trial. Neurology,
the evidence 10 years on; a partial replication. Journal of Neurology and 54, 295–301.
Neurosurgery Psychiatry, 74, 857–62. Hendrie, H. C., et al. (2001). Incidence of dementia and Alzheimer disease
Folstein, M. F., et al. (1975). ‘Mini-mental state’. A practical method for in 2 communities: Yoruba residing in Ibadan, Nigeria, and African
grading the cognitive state of patients for the clinician. Journal of Americans residing in Indianapolis, Indiana. (See Comment.) Journal of
Psychiatric Research, 12, 189–98. the American Medical Association, 285, 739–47.
Ford, A. H. and Almeida, O. P. (2012). Effect of homocysteine lowering Herrmann, N., Li, A., and Lanctot, K. (2011). Memantine in dementia: a
treatment on cognitive function: a systematic review and meta-analysis of review of the current evidence. Expert Opinion on Pharmacotherapy,
randomized controlled trials. Journal of Alzheimers Disease, 29, 133–49. 12, 787–800.
Forette, F., et al. (1998). Prevention of dementia in randomised double-blind Hirono, N., et al. (1998). Factors associated with psychotic symptoms in
placebo-controlled Systolic Hypertension in Europe (Syst-Eur) trial. Alzheimer’s disease. Journal of Neurology, Neurosurgery and Psychiatry,
Lancet, 352, 1347–51. 64, 648–52.
Gao, S., et al. (1998). The relationships between age, sex, and the incidence Ho, R. C., et al. (2011). Is high homocysteine level a risk factor for cognitive
of dementia and Alzheimer disease: a meta-analysis. Archives of General decline in elderly? A systematic review, meta-analysis, and meta-
Psychiatry, 55, 809–15. regression. American Journal of Geriatric Psychiatry, 19, 607–17.
Gatz, M., et al. (2005). Complete ascertainment of dementia in the Hollis, J., (2007). Antipsychotic medication dispensing and risk of death
Swedish Twin Registry: the HARMONY study. Neurobiological Aging, in veterans and war widows 65 years and older. American Journal of
26, 439–47. Geriatric Psychiatry, 15, 932–41.
Gaugler, J. E., et al. (2003). Caregiving and institutionalization of cognitively Howard, R. J., et al. (2007). Donepezil for the treatment of agitation in
impaired older people: utilizing dynamic predictors of change. Alzheimer’s disease. (See Comment.) New England Journal of Medicine,
Gerontologist, 43, 219–29. 357, 1382–92.
Gaugler, J. E., et al. (2011). Does caregiver burden mediate the effects of Hughes, C. P., et al. (1982). A new clinical scale for the staging of dementia.
behavioral disturbances on nursing home admission? American Journal British Journal of Psychiatry, 140, 566–72.
of Geriatric Psychiatry, 19, 497–506. Husebo, B. S., et al. (2011). Efficacy of treating pain to reduce behavioural
Gauthier, S., et al. (2010). Management of behavioral problems in Alzheimer’s disturbances in residents of nursing homes with dementia: cluster
disease. International Psychogeriatrics, 22, 346–72. randomised clinical trial. British Medical Journal, 343, d4065.
Gillette-Guyonnet, S., et al. (2011). Long-term progression of Alzheimer’s Hyman, B. T. and Trojanowski, J. Q. (1997). Consensus recommendations
disease in patients under antidementia drugs. Alzheimers Dementia, for the postmortem diagnosis of Alzheimer disease from the National
7, 579–92. Institute on Aging and the Reagan Institute Working Group on
Gilley, D. W., et al. (1997). Psychotic symptoms and physically aggressive diagnostic criteria for the neuropathological assessment of Alzheimer
behavior in Alzheimer’s disease. Journal of the American Geriatrics disease. Journal of Neuropathology and Experimental Neurology,
Society, 45, 1074–9. 56, 1095–7.
Graves, A. B., et al. (1991). Alcohol and tobacco consumption as risk factors Imbimbo, B. P. (2004). The potential role of non-steroidal anti-inflammatory
for Alzheimer’s disease: a collaborative re-analysis of case-control drugs in treating Alzheimer’s disease. Expert Opinion on Investigational
studies. EURODEM Risk Factors Research Group. International Journal Drugs, 13, 1469–81.
of Epidemiology, 20, S48–57. Jacobs, D., et al. (1994). Age at onset of Alzheimer’s disease: relation to pattern
Guan, J. W., et al. (2011). No association between hypertension and risk for of cognitive dysfunction and rate of decline. Neurology, 44, 1215–20.
Alzheimer’s disease: a meta-analysis of longitudinal studies. Journal of Jick, H., et al. (2000). Statins and the risk of dementia. Lancet,
Alzheimers Disease, 27, 799–807. 356, 1627–31.
Johnson, V. E., Stewart, W., and Smith, D. H. (2010). Traumatic brain injury Luchsinger, J. A., et al. (2005). Aggregation of vascular risk factors and risk of
and amyloid-beta pathology: a link to Alzheimer’s disease? Nature incident Alzheimer disease. Neurology, 65, 545–51.
Reviews Neuroscience, 11, 361–70. Lyketsos, C.G., et al. (2002). Prevalence of neuropsychiatric symptoms
Jost, B.C. and Grossberg, G.T. (1996). The evolution of psychiatric symptoms in dementia and mild cognitive impairment: results from the
in Alzheimer’s disease: a natural history study. Journal of the American cardiovascular health study. Journal of the American Medical Association,
Geriatric Society, 44(9), 1078–81. 288(12), 1475–83.
Juva, K., et al. (2000). APOE epsilon4 does not predict mortality, cognitive Lyketsos, C.G., et al. (2003). Treating depression in Alzheimer disease:
decline, or dementia in the oldest old. Neurology, 54, 412–15. efficacy and safety of sertraline therapy, and the benefits of depression
Kang, J. H., et al. (2006). A randomized trial of vitamin E supplementation reduction: the DIADS. Archives of General Psychiatry, 60, 737–46.
and cognitive function in women. Archives of Internal Medicine, Lyketsos, C.G., et al. (2007). Naproxen and celecoxib do not prevent
166, 2462–8. AD in early results from a randomized controlled trial. Neurology,
Katz, M. J., et al. (2012). Age-specific and Sex-specific Prevalence and 68, 1800–8.
Incidence of Mild Cognitive Impairment, Dementia, and Alzheimer Malouf, R. and Grimley Evans, J. (2009). Folic acid with or without vitamin
Dementia in Blacks and Whites: A Report From the Einstein Aging B12 for the prevention and treatment of healthy elderly and demented
Study. Alzheimer’s Disease Association Disorders, 26(4), 335–43. people. Cochrane Database of Systematic Reviews.
Kawas, C. H. and Corrada, M. M. (2006). Alzheimer’s and dementia in Mann, D. M., et al. (1986). The topography of plaques and tangles in Down’s
the oldest-old: a century of challenges. Current Alzheimer’s Research, syndrome patients of different ages. Neuropathology and Applied
3, 411–19. Neurobiology, 12, 447–57.
Kivipelto, M., et al. (2001). Midlife vascular risk factors and Alzheimer’s Matthews, F. and Brayne, C. (2005). The incidence of dementia in England
disease in later life: longitudinal, population based study. British Medical and Wales: findings from the five identical sites of the MRC CFA Study.
Journal, 322, 1447–51. PLoS Medicine, 2, e193.
Kivipelto, M., et al. (2005). Obesity and vascular risk factors at midlife and Maynard, C. J., et al. (2005). Metals and amyloid-beta in Alzheimer’s disease.
the risk of dementia and Alzheimer disease. Archives of Neurology, International Journal of Experimental Pathology, 86, 147–59.
62, 1556–60. McGeer, P. L., Schulzer, M., and McGeer, E. G. (1996). Arthritis and anti-
Kleijer, B. C. (2009). Risk of cerebrovascular events in elderly users of inflammatory agents as possible protective factors for Alzheimer’s
antipsychotics. Journal of Psychopharmacology, 23, 909–14. disease: a review of 17 epidemiologic studies. (See Comments.)
Knesevich, J. W. (1983). Preliminary report on affective symptoms in the Neurology, 47, 425–32.
early stages of senile dementia of the Alzheimer type. American Journal McKeith, I., et al. (1992). Neuroleptic sensitivity in patients with senile
of Psychiatry, 140, 233–5. dementia of Lewy body type. British Medical Journal, 305, 673–8.
Knopman, D. S. (1988). Longitudinal study of death and institutionalization McKeith, I., et al. (2000a). Predictive accuracy of clinical diagnostic criteria for
in patients with primary degenerative dementia. Journal of the American dementia with Lewy bodies: a prospective neuropathological validation
Geriatrics Society, 36, 108–12. study. Neurology, 54, 1050–8.
Kramer-Ginsberg, E., et al. (1988). Clinical predictors of course for Alzheimer McKeith, I., et al. (2000b). Efficacy of rivastigmine in dementia with Lewy
patients in a longitudinal study: a preliminary report. Psychopharmacology bodies: a randomised, double-blind, placebo-controlled international
Bulletin, 24, 458–62. study. Lancet, 356, 2031–6.
Kuller, L. H. and Lopez, O. L. (2011). Dementia and Alzheimer’s disease: McKeith, I. G., et al. (2000c). Prospective validation of consensus criteria for
a new direction. the 2010 Jay L. Foster Memorial lecture. Alzheimers the diagnosis of dementia with Lewy bodies. Neurology, 54, 1050–8.
Dementia, 7, 540–50. McKeith, I. G., et al. (2005). Diagnosis and management of dementia
Laredo, L., et al. (2011). Risk of cerebrovascular accident associated with use with Lewy bodies: third report of the DLB Consortium. Neurology,
of antipsychotics: population-based case-control study. Journal of the 65, 1863–72.
American Geriatrics Society, 59, 1182–7. McKhann, G., et al. (1984). Clinical diagnosis of Alzheimer’s disease: report
Larrieu, S., et al. (2004). Nutritional factors and risk of incident dementia in of the NINCDS–ADRDA Work Group under the auspices of Department
the PAQUID longitudinal cohort. Journal of Nutrition, Health and Aging, of Health and Human Services Task Force on Alzheimer’s Disease.
8, 150–4. Neurology, 34, 939–44.
Lautenschlager, N. T., et al. (2008). Effect of physical activity on cognitive McKhann, G. M., et al. (2011). The diagnosis of dementia due to Alzheimer’s
function in older adults at risk for Alzheimer disease: a randomized trial. disease: recommendations from the National Institute on Aging–
Journal of the American Medical Association, 300, 1027–37. Alzheimer’s Association workgroups on diagnostic guidelines for
Leblanc, E. S., et al. (2001). Hormone replacement therapy and cognition: Alzheimer’s disease. Alzheimers Dementia, 7, 263–9.
systematic review and meta-analysis. Journal of the American Medical McShane, R., et al. (1997). Do neuroleptic drugs hasten cognitive decline in
Association, 285, 1489–99. dementia? Prospective study with necropsy follow up. British Medical
Levy, M. L., et al. (1998). Apathy is not depression. Journal of Neuropsychiatry Journal, 314, 266–70.
and Clinical Neurosciences, 10, 314–19. Mega, M. S., et al. (1996). The spectrum of behavioral changes in Alzheimer’s
Livingston, G., et al. (2005). Systematic review of psychological approaches to disease. Neurology, 46, 130–5.
the management of neuropsychiatric symptoms of dementia. American Merriam, A. E., et al. (1988). The psychiatric symptoms of Alzheimer’s
Journal of Psychiatry, 162, 1996–2021. disease. Journal of the American Geriatrics Society, 36, 7–12.
Lonergan, E. and Luxenberg, J. (2009). Valproate preparations for agitation in Miller, T. P., et al. (1991). Cognitive decline in patients with Alzheimer
dementia. Cochrane Database Systemic Reviews, CD003945. disease: differences in patients with and without extrapyramidal signs.
Lopez, O. L., et al. (1990). Alzheimer’s disease and depression: Alzheimer Disease and Associated Disorders, 5, 251–6.
neuropsychological impairment and progression of the illness. American Morris, J. C., et al. (1993). The consortium to establish a registry for
Journal of Psychiatry, 147, 855–60. Alzheimer’s disease (CERAD). Part IV. Rates of cognitive change in
Lopez, O. L., et al. (1997). Extrapyramidal signs in patients with probable the longitudinal assessment of probable Alzheimer’s disease. Neurology,
Alzheimer disease. Archives of Neurology, 54, 969–75. 43, 2457–65.
Lopez, O. L., et al. (2009). Long-term effects of the concomitant use of Morris, M. C., et al. (2002). Dietary intake of antioxidant nutrients and the
memantine with cholinesterase inhibition in Alzheimer disease. Journal risk of incident Alzheimer disease in a biracial community study. Journal
of Neurology and Neurosurgery Psychiatry, 80, 600–7. of the American Medical Association, 287, 3230–7.
Mulnard, R. A., et al. (2000). Estrogen replacement therapy for treatment Roe, C. M., et al. (2008). Interaction of neuritic plaques and education predicts
of mild to moderate Alzheimer disease: a randomized controlled trial. dementia. Alzheimer’s Disease Association Disorders, 22, 188–93.
Alzheimer’s Disease Cooperative Study. Journal of the American Medical Roe, C. M., et al. (2011). Cerebrospinal fluid biomarkers, education, brain
Association, 283, 1007–15. volume, and future cognition. Archives of Neurology, 68, 1145–51.
Nelson, H. D., et al. (2002). Postmenopausal hormone replacement Ronnemaa, E., et al. (2011). Vascular risk factors and dementia: 40-year
therapy: scientific review. Journal of the American Medical Association, follow-up of a population-based cohort. Dementia and Geriatric
288, 872–81. Cognitive Disorders, 31, 460–6.
Newhouse, P., et al. (2012). Nicotine treatment of mild cognitive impairment: Rosenberg, P. B., et al. (2005). Clinical application of operationalized criteria
a 6-month double-blind pilot clinical trial. Neurology, 78, 91–101. for ’Depression of Alzheimer’s Disease’. International Journal of Geriatric
Nyth, A. L., et al. (1992). A controlled multicenter clinical study of citalopram Psychiatry, 20, 119–27.
and placebo in elderly depressed patients with and without concomitant Rosenberg, P. B., et al. (2010). Sertraline for the treatment of depression in
dementia. Acta Psychiatrica Scandinavica, 86, 138–45. Alzheimer disease. American Journal of Geriatric Psychiatry, 18, 136–45.
Olin, J. T., et al. (2001). A pilot randomized trial of carbamazepine for Roses, A. D. (1996). Apolipoprotein E alleles as risk factors in Alzheimer’s
behavioral symptoms in treatment-resistant outpatients with Alzheimer disease. Annual Review of Medicine, 47, 387–400.
disease. American Journal of Geriatric Psychiatry, 9, 400–5. Roth, M., Mountjoy, C. Q., and Amrein, R. (1996). Moclobemide in elderly
Olin, J. T., et al. (2002). Provisional diagnostic criteria for depression of patients with cognitive decline and depression: an international
Alzheimer disease. American Journal of Geriatric Psychiatry, 10, 125–8. double-blind, placebo-controlled trial. British Journal of Psychiatry,
Onyike, C.U., et al. (2007). Epidemiology of apathy in older adults: the Cache 168, 149–57.
County Study. American Journal of Geriatric Psychiatry, 15(5), 365–75. Rubin, E. H., Morris, J. C., and Berg, L. (1987). The progression of personality
Ownby, R. L., et al. (2006). Depression and risk for Alzheimer disease: changes in senile dementia of the Alzheimer’s type. Journal of the
systematic review, meta-analysis, and metaregression analysis. Archives American Geriatrics Society, 35, 721–5.
of General Psychiatry, 63, 530–8. Rubin, E. H., Drevets, W. C., and Burke, W. J. (1988). The nature of psychotic
Paradise, M., et al. (2009). Prediction of survival in Alzheimer’s disease— symptoms in senile dementia of the Alzheimer type. Journal of Geriatric
the LASER-AD longitudinal study. International Journal of Geriatric Psychiatry and Neurology, 1, 16–20.
Psychiatry, 24, 739–47. Saczynski, J. S., et al. (2010). Depressive symptoms and risk of dementia.
Patterson, M. B. and Bolger, J. P. (1994). Assessment of behavioral symptoms in Neurology, 75, 35–41.
Alzheimer disease. Alzheimer Disease and Associated Disorders, 8, 4–20. Saftig, P., Hartmann, D., and De Strooper, B. (1999). The function of
Patterson, M. B., et al. (1990). Assessment of behavioral and affective presenilin-1 in amyloid beta-peptide generation and brain development.
symptoms in Alzheimer’s disease. Journal of Geriatric Psychiatry and European Archives of Psychiatry and Clinical Neuroscience, 249, 271–9.
Neurology, 3, 21–30. Salmon, D. P., et al. (1990). Longitudinal evaluation of dementia of
Pearson, J. L., et al. (1989). Functional status and cognitive impairment the Alzheimer type: a comparison of 3 standardized mental status
in Alzheimer’s patients with and without depression. Journal of the examinations. Neurology, 40, 1225–30.
American Geriatrics Society, 37, 1117–21. Sano, M., et al. (1997). A controlled trial of selegiline, alpha-tocopherol,
Pedersen, N. L., et al. (2004). How heritable is Alzheimer’s disease late in life? or both as treatment for Alzheimer’s disease. The Alzheimer’s Disease
Findings from Swedish twins. Annals of Neurology, 55, 180–5. Cooperative Study. New England Journal of Medicine, 336, 1216–22.
Petersen, R. C., et al. (2005). Vitamin E and donepezil for the treatment of mild Schneider, L. S., Dagerman, K. S., and Insel, P. (2005). Risk of death with
cognitive impairment. New England Journal of Medicine, 352, 2379–88. atypical antipsychotic drug treatment for dementia: meta-analysis of
Petracca, G., et al. (1996). A double-blind placebo-controlled study of randomized placebo-controlled trials. (See Comment.) Journal of the
clomipramine in depressed patients with Alzheimer’s disease. Journal of American Medical Association, 294, 1934–43.
Neuropsychiatry and Clinical Neurosciences, 8, 270–5. Schneider, L. S., Dagerman, K., and Insel, P. S. (2006a). Efficacy and
Plassman, B. L., et al. (2000). Documented head injury in early adulthood and adverse effects of atypical antipsychotics for dementia: meta-analysis
risk of Alzheimer’s disease and other dementias. Neurology, 55, 1158–66. of randomized, placebo-controlled trials. American Journal of Geriatric
Pollock, B. G., et al. (2007). A double-blind comparison of citalopram and Psychiatry, 14, 191–210.
risperidone for the treatment of behavioral and psychotic symptoms Schneider, L. S., et al. (2006b). Effectiveness of atypical antipsychotic drugs
associated with dementia. (See Comment.) American Journal of Geriatric in patients with Alzheimer’s disease. New England Journal of Medicine,
Psychiatry, 15, 942–52. 355, 1525–38.
Rabins, P. V., Mace, N. L., and Lucas, M. J. (1982). The impact of dementia on Schneider, L. S., et al. (2011). Lack of evidence for the efficacy of memantine
the family. Journal of the American Medical Association, 248, 333–5. in mild Alzheimer disease. Archives of Neurology, 68, 991–8.
Rao, V. and Lyketsos, C. G. (2000). The benefits and risks of ECT for patients Schrag, M., et al. (2011). Iron, zinc and copper in the Alzheimer’s disease brain:
with primary dementia who also suffer from depression. International A quantitative meta-analysis. Some insight on the influence of citation
Journal of Geriatric Psychiatry, 15, 729–35. bias on scientific opinion. Progress in Neurobiology, 94, 296–306.
Reines, S. A., et al. (2004). Rofecoxib: no effect on Alzheimer’s disease in a Seitz, D. P., et al. (2011). Antidepressants for agitation and psychosis in
1-year, randomized, blinded, controlled study. Neurology, 62, 66–71. dementia. Cochrane Database Systematic Reviews, CD008191.
Reisberg, B., et al. (1986). Longitudinal course of normal aging and progressive Shepherd, J., et al. (2002). Pravastatin in elderly individuals at risk of
dementia of the Alzheimer’s type: a prospective study of 106 subjects vascular disease (PROSPER): a randomised controlled trial. Lancet,
over a 3.6 year mean interval. Progress in Neuro-Psychopharmacology and 360, 1623–30.
Biological Psychiatry, 10, 571–8. Shumaker, S. A., et al. (2004). Conjugated equine estrogens and incidence of
Reisberg, B., et al. (1987). Behavioral symptoms in Alzheimer’s disease: probable dementia and mild cognitive impairment in postmenopausal
phenomenology and treatment. Journal of Clinical Psychiatry, 48, 9–15. women: Women’s Health Initiative Memory Study. Journal of the
Ritchie, K. and Kildea, D. (1995). Is senile dementia “age-related” or “ageing- American Medical Association, 291, 2947–58.
related”?--evidence from meta-analysis of dementia prevalence in the Sink, K. M., Holden, K. F., and Yaffe, K. (2005). Pharmacological treatment
oldest old. (See Comments.) Lancet, 346, 931–4. of neuropsychiatric symptoms of dementia: a review of the evidence.
Robert, P., et al. (2009). Proposed diagnostic criteria for apathy in Alzheimer’s Journal of the American Medical Association, 293, 596–608.
disease and other neuropsychiatric disorders. European Psychiatry, Skoog, I., et al. (1996). 15-year longitudinal study of blood pressure and
24, 98–104. dementia. Lancet, 347, 1141–5.
Snowdon, D. A., et al. (1996). Linguistic ability in early life and cognitive Vemuri, P., et al. (2011). Cognitive reserve and Alzheimer’s disease biomarkers
function and Alzheimer’s disease in late life. Findings from the Nun are independent determinants of cognition. Brain, 134, 1479–92.
Study. (See Comments.) Journal of the American Medical Association, Vigen, C. L., et al. (2011). Cognitive effects of atypical antipsychotic
275, 528–32. medications in patients with Alzheimer’s disease: outcomes from
Spackman, D. E., et al. (2011). Measuring Alzheimer disease progression CATIE-AD. American Journal of Psychiatry, 168, 831–9.
with transition probabilities: Estimates from NACC-UDS. Current Walton, L. (1991). Alzheimer’s disease and the environment. Alden
Alzheimer‘s Research, 9(9), 1050–8. Press, Oxford.
Starkstein, S. E. and Leentjens, A. F. (2008). The nosological position of apathy Wang, P. N., et al. (2000). Effects of estrogen on cognition, mood, and cerebral
in clinical practice. Journal of Neurology and Neurosurgery Psychiatry, blood flow in AD: a controlled study. Neurology, 54, 2061–6.
79, 1088–92. Wang, P. S., et al. (2005). Risk of death in elderly users of conventional vs.
Steinberg, M., et al. (2008). Point and 5-year period prevalence of atypical antipsychotic medications. New England Journal of Medicine,
neuropsychiatric symptoms in dementia: the Cache County Study. 353, 2335–41.
International Journal of Geriatric Psychiatry, 23, 170–7. Weintraub, D., et al. (2010). Sertraline for the treatment of depression in
Stern, Y. (2009). Cognitive reserve. Neuropsychologia, 47, 2015–28. Alzheimer disease: week-24 outcomes. American Journal of Geriatric
Stern, Y., et al. (1992). Inverse relationship between education and Psychiatry, 18, 332–40.
parietotemporal perfusion deficit in Alzheimer’s disease. Annals of Weuve, J., McQueen, M., and Blacker, D. (2012). The AlzRisk Database.
Neurology, 32, 371–5. Alzheimer Research Forum. <http://www.alzforum.org> (accessed
Stern, Y., et al. (1994). Utility of extrapyramidal signs and psychosis as 05.02.2012).
predictors of cognitive and functional decline, nursing home admission, Whitmer, R. A., et al. (2005). Midlife cardiovascular risk factors and risk of
and death in Alzheimer’s disease: prospective analyses from the dementia in late life. Neurology, 64, 277–81.
Predictors Study. Neurology, 44, 2300–7. Whitmer, R. A., et al. (2011). Timing of hormone therapy and dementia: the
Tabet, N., et al. (2002). Endogenous antioxidant activities in relation to critical window theory revisited. Ann Neurol, 69, 163–9.
concurrent vitamins A, C, and E intake in dementia. International Wilcock, G. K., et al. (2008). Memantine for agitation/aggression and
Psychogeriatrics, 14, 7–15. psychosis in moderately severe to severe Alzheimer’s disease: a pooled
Tang, M. X., et al. (2001). Incidence of AD in African-Americans, Caribbean analysis of 3 studies. Journal of Clinical Psychiatry, 69, 341–8.
Hispanics, and Caucasians in northern Manhattan. Neurology, Wolozin, B., et al. (2000). Decreased prevalence of Alzheimer disease
56, 49–56. associated with 3-hydroxy-3-methyglutaryl coenzyme A reductase
Taragano, F. E., et al. (1997). A double-blind, randomized, fixed-dose trial inhibitors. (See Comment.) Archives of Neurology, 57, 1439–43.
of fluoxetine vs. amitriptyline in the treatment of major depression Wood, J. A., et al. (1993). Cytokine indices in Alzheimer’s temporal cortex:
complicating Alzheimer’s disease. Psychosomatics, 38, 246–52. no changes in mature IL-1 beta or IL-1RA but increases in the associated
Tariot, P. N., et al. (1998). Efficacy and tolerability of carbamazepine for acute phase proteins IL-6, alpha 2-macroglobulin and C-reactive protein.
agitation and aggression in dementia. American Journal of Psychiatry, Brain Research, 629, 245–52.
155, 54–61. Wragg, R. E. and Jeste, D. V. (1989). Overview of depression and psychosis in
Tariot, P. N., et al. (2011). Chronic divalproex sodium to attenuate agitation Alzheimer’s disease. American Journal of Psychiatry, 146, 577–87.
and clinical progression of Alzheimer disease. Archives of General Xie, J., Brayne, C., and Matthews, F. E. (2008). Survival times in people with
Psychiatry, 68, 853–61. dementia: analysis from population based cohort study with 14 year
Teri, L. and Wagner, A. W. (1991). Assessment of depression in patients with follow-up. British Medical Journal, 336, 258–62.
Alzheimer’s disease: concordance among informants. Psychology and Yaffe, K., et al. (1998). Estrogen therapy in postmenopausal women: effects
Aging, 6, 280–5. on cognitive function and dementia. Journal of the American Medical
Teri, L., et al. (1995). Cognitive decline in Alzheimer’s disease: a longitudinal Association, 279, 688–95.
investigation of risk factors for accelerated decline. Journals of Gerontology. Yesavage, J. A., et al. (1988). Rates of change of common measures of impairment
Series A, Biological Sciences and Medical Sciences, 50A, M49–55. in senile dementia of the Alzheimer’s type. Psychopharmacology Bulletin,
Thomas, A. J. (2011). Antipsychotic drugs in dementia: A necessary evil? 24, 531–4.
Maturitas, 68, 109–10. Zandi, P. P., et al. (2005). Do statins reduce risk of incident dementia and
Thomas, A. J., Kalaria, R. N., and O’Brien, J. T. (2004). Depression and Alzheimer disease? The Cache County Study. Archives of General
vascular disease: what is the relationship? Journal of Affective Disorders, Psychiatry, 62, 217–24.
79, 81–95. Zanetti, O., Solerte, S. B., and Cantoni, F. (2009). Life expectancy in
Trapp, G. A., et al. (1978). Aluminum levels in brain in Alzheimer’s disease. Alzheimer’s disease (AD). Archives of Gerontology and Geriatrics, 49
Biological Psychiatry, 13, 709–18. Suppl 1, 237–43.
Valenzuela, M. J. and Sachdev, P. (2006). Brain reserve and dementia: a Zannis, V. I., Kardassis, D., and Zanni, E. E. (1993). Genetic mutations
systematic review. Psychological Medicine, 36, 441–54. affecting human lipoproteins, their receptors, and their enzymes.
Van Duijn, C. M., et al. (1991). Familial aggregation of Alzheimer’s disease Advances in Human Genetics, 21, 145–319.
and related disorders: a collaborative re-analysis of case-control studies. Zigman, W. B., et al. (1996). Prevalence of dementia in adults with and
EURODEM Risk Factors Research Group. International Journal of without Down syndrome. American Journal of Mental Retardation,
Epidemiology, 20, S13–20. 100, 403–12.
CHAPTER 34
Vascular and mixed
dementias
Robert Stewart
Vascular disorders have been established as the most important One of the first set of criteria for vascular dementia to come
‘environmental’ risk factors for late-life dementia in western set- into widespread use in research settings was the Hachinski
tings because of their high prevalence and potential for modifica- Ischaemic Scale (Hachinski et al., 1975), although this was never
tion. However, historically ‘vascular dementia’ as a construct has intended as a diagnostic instrument and was simply a checklist of
attracted substantially less research than Alzheimer’s disease. This 13 features of ‘arteriosclerotic psychosis’ from a widely read text-
is partly because it falls between traditional clinical specialties, book (Mayer-Gross et al., 1969); some refer to the course of the
being seen as the province of stroke researchers by those interested dementia (e.g. abrupt onset, stepwise deterioration, fluctuation),
in dementia and vice versa, but a more fundamental reason is that some to symptoms (e.g. nocturnal confusion, depression, somatic
its validity as a ‘diagnosis’ has been elusive. Diagnoses in psychia- complaints), and some to the likely presence of or risk for cere-
try, even those of organic disorders, remain derived primarily from brovascular disease (e.g. history of strokes or hypertension, focal
a clinician’s subjective judgement applied to a person’s subjective neurological signs). No attempt was made to validate the instru-
experiences, from his or her behaviour reported by a close other, ment or to investigate its scaling properties or the validity of the
interpretations of the person’s level of function, and the environ- different weights given to some items. Despite this, it was found
mental context in which the disorder has occurred. There is thus to have the highest inter-rater reliability in an independent evalua-
ample room for variability in the application of diagnostic criteria, tion of case identification instruments for vascular dementia (Chui
and a major objective of psychiatric research has been to stand- et al., 2000), probably because it seeks to ascertain whether signifi-
ardize this process so that findings from one research group can cant cerebrovascular disease is present in the context of dementia,
be understood by another. Research diagnoses therefore tend to rather than whether it is causally related.
prioritize reliability over validity (since the former is more easily Considering the standard current classification systems, DSM-IV
established) and, particularly when restrictive, can drift a consider- criteria require the presence of dementia and either focal neurolog-
able distance from clinical practice. Sometimes diagnostic systems ical signs/symptoms, or evidence of cerebrovascular disease judged
fail to keep up with accumulating knowledge about a clinical con- to be aetiologically related (American Psychiatric Association,
dition, or fail to reflect changing presentations. Vascular dementia 1994). A requirement for an early stepwise course of decline was
has been a good example of this failure. dropped between DSM-IIIR and DSM-IV. ICD-10 criteria are
stricter, requiring evidence of both focal brain damage and cer-
ebrovascular disease, judged to be aetiologically related, as well as
Definitions of Vascular Dementia an uneven distribution of cognitive deficits. At the time of writing,
Changing fashions for ‘lumping’ or ‘splitting’ dementia are funda- proposals for major neurocognitive disorder associated with vascu-
mental to the problems underlying vascular dementia as a diagno- lar disease in DSM-V have not been released.
sis. Throughout the first half of the twentieth century, later onset Probably the most widely used criteria in research settings
(‘senile’) dementia was viewed as an inevitable consequence of age- to date have been those developed in a 1991 workshop of the
ing and, since atherosclerosis was a popular substrate for ageing, Neuroepidemiology branch of the National Institute of Neurological
most dementia was assumed to be ‘vascular’. Seminal post-mortem Disorders and Stroke and the Association Internationale pour la
studies in the 1960s and early 1970s challenged this by demonstrat- Recherche et l’Enseignement en Neurosciences (NINDS-AIREN)
ing the importance of Alzheimer pathology—previously assumed (Román et al., 1993). A diagnosis of vascular dementia using
only to underlie rare early-onset (‘presenile’) syndromes. However, NINDS-AIREN criteria rests on evidence of cerebrovascular dis-
it was evident that multiple cerebral infarctions could also be ease (both clinical and from neuroimaging), and an observed rela-
responsible for causing dementia. In consequence, a second diag- tionship between cerebrovascular disease and dementia (such as an
nostic concept of multi-infarct dementia arose, later to be subsumed onset within 3 months of an acute stroke or an abrupt deterioration
under the more broad category of vascular dementia. in cognitive functioning, or a fluctuating or stepwise deterioration).
A ‘probable’ diagnosis using this system requires all three of these was found to be more strongly predicted by medial temporal
criteria to be present, while a ‘possible’ diagnosis can be made in the atrophy and thalamic infarcts rather than NINDS-AIREN neu-
following situations: (1) where dementia is accompanied by focal roimaging criteria (Firbank et al., 2012), or white matter hyper-
neurological signs in the absence of neuroimaging data; (2) in the intensities (Firbank et al., 2007).
absence of a clear temporal relationship between clinical stroke and ◆ Risk factors for stroke have long been recognized to be risk factors
dementia; and (3) ‘in patients with subtle onset and variable course not only for vascular dementia but also for clinical Alzheimer’s
of . . . cognitive deficits and evidence of relevant CVD’. disease (Stewart, 1998).
The system of subdividing dementia based on assumed underly-
ing causation has therefore persisted into current diagnostic sched- ◆ Dementia in community samples is frequently associated with
ules and looks as if it will persist, judging from the proposed DSM-V mixed rather than discrete pathology (Holmes et al., 1999).
chapter titles. However, several changes over the 30–40 years since Cerebrovascular disease was found to be rarely associated with
the first criteria were developed suggest that a reappraisal should at dementia in the absence of Alzheimer pathology (Hulette et al.,
least be considered for vascular dementia: 1997).
◆ Changing cohorts. Older people who died in the 1960s lived ◆ Research diagnostic criteria poorly reflect underlying pathol-
through times when there was little opportunity to prevent or ogy. Although cerebrovascular pathology is likely to be present
control cerebrovascular disease. More severe and florid pathol- in cases with a previous diagnosis of vascular dementia, it is
ogy would have been observed at post mortem than would be also frequently present in people with other clinical diagnoses
expected (in developed nations) today and was more clearly a (Holmes et al., 1999; Fernando et al., 2004; White et al., 2005).
primary cause of a dementia syndrome. People with dementia ◆ Clinical diagnostic criteria for vascular dementia have poor
known to clinical services are now frequently in their ninth and inter-rater reliability (Lopez et al., 1994; Chui et al., 2000). Highest
tenth decades, where underlying pathology is much more likely agreement is found where criteria simply estimate the degree
to be mixed. Most people in these age ranges will have mild levels of cerebrovascular disease in people with dementia (Hachinski
of cerebrovascular disease which may be coincidental, or at least et al., 1975), and unsurprisingly much lower agreement is found
not a single causal factor. if a judgement is required as to whether dementia was actually
◆ Advances in neuropathology. Early studies predominantly focused caused by cerebrovascular disease. Different diagnostic schedules
on multiple cortical infarctions. Other more subtle forms of cer- also show poor agreement with each other (Chui et al., 2000).
ebrovascular pathology such as lacunar infarction, white matter ◆ Operational definitions of dementia have been criticized for
disease, vascular amyloid deposition, and microangiopathy have focusing excessively on memory impairment and for poorly
subsequently been highlighted. These may increase the risk of reflecting cognitive impairment relating to vascular disease,
dementia but may not in isolation be sufficient to cause the clini- which more often affects nonmemory domains such as executive
cal presentation; instead, a combination of pathological processes function (Bowler and Hachinski, 2000), although this is being
may be necessary (see Chapter 6 for details). addressed in DSM-V proposals.
◆ Advances in neuroimaging. Clinical diagnostic criteria for mul- The principal reason that the current system has persisted is
ti-infarct or vascular dementia have traditionally relied on a because of potential utility in Alzheimer’s disease research. If risk
history of vascular risk factors, a history suggestive of recurrent factors for a particular pathological process are to be identified,
strokes, and evidence on clinical examination of neurological there is some rationale for excluding people with evidence of any
deficits indicative of cortical stroke. Subsequent technological other pathology. This can be readily carried out for Alzheimer’s dis-
advances allow in vivo identification of much more subtle changes ease, since levels of vascular pathology can be reasonably approxi-
such as white matter hyperintensities and alterations in patterns mated (and screened out) by clinical examination or neuroimaging
of perfusion and connectivity. However, these are all common in (although neuropathologically ‘pure’ Alzheimer’s disease has been
unaffected people in old age and cannot be assumed to be causal found to be less common than diagnostic criteria would suggest
when they are seen in someone with dementia. (Fernando et al., 2004)). The same is not true for vascular demen-
tia, since quantification of Alzheimer pathology in vivo remains
confined to small sample studies. The only means of defining ‘pure’
Problems with ‘Vascular Dementia’ as a vascular dementia is to establish a temporal relationship between
Diagnosis stroke episodes and cognitive decline, and then to exclude cases with
The validity of vascular dementia as a diagnosis rests on the dem- evidence of gradual decline in between strokes. This process relies
onstration of a discrete syndrome (dementia) with a clear primary heavily on retrospective and potentially inaccurate information.
cause (‘vascular’), and with good clinical and clinicopathological The implications are becoming increasingly problematic. If vas-
reliability. These criteria are challenged in several respects: cular dementia as a diagnosis poorly reflects underlying pathol-
ogy and relies on a differentiation with Alzheimer’s disease which
◆ The clinical course of dementia (i.e. ‘stepwise’ vs ‘gradual deterio- involves a large degree of subjective judgement and ‘guesswork’,
ration) is a poor predictor of pathological findings (Fischer et al., then research that attempts to apply these criteria is fraught with
1990). methodological shortcomings. Prevalence studies will be diffi-
◆ Dementia following clinical stroke frequently shows a gradual cult to interpret if between-site equivalence cannot be assumed.
deterioration and has been found to occur without further infarc- Furthermore, any observed between-site differences in prevalence
tion in many, if not the majority, of cases (Tatemichi et al., 1994; (or in the proportion of dementia that is defined as ‘vascular’) may
Kokmen et al., 1996). Consistent with this, poststroke dementia simply reflect underlying differences in stroke incidence. Case
CHAPTER 34 vascular and mixed dementias 459
control studies become difficult if all cases have evidence of cer- Table 34.1 Examples of specific vascular dementia syndromes
ebrovascular disease by definition (or if it is absent or minimal in
Genetic disorders
the case of Alzheimer’s disease), since they will differ from con-
trols in many respects through selection bias, resulting in spurious Sickle cell disease
positive or negative findings. Implications for clinical trials are also CADASIL
substantial and will be discussed later.
Hereditary cerebral haemorrhage with amyloidosis—Dutch and Icelandic
One of the most important consequences of the ‘diagnosis prob-
types
lem’ has been that the role of vascular factors in dementia has
historically been underestimated. If vascular factors are not often Familial British dementia with amyloid angiopathy
associated with dementia in isolation (i.e. other coexisting pathol- Homocystinuria
ogy is required), this does not mean that they are not important
Fabry’s disease
risk factors. The following sections will review some of the evi-
dence for this and discuss potential causal pathways, before con- Hereditary endotheliopathy with retinopathy, nephropathy, and stroke
sidering potential implications for the treatment and prevention of (HERNS)
dementia. Mitochondrial myopathy, encephalopathy, lactic acidosis, and stroke-like
episodes (MELAS)
Does Vascular Dementia Exist? Specific Strategic infarct dementia
Syndromes Conditions associated with cerebrovascular pathology and secondary
dementia
If vascular dementia and Alzheimer’s disease are not clearly distin-
guishable disorders, the continued usefulness of a subcategorizing Subdural haematoma
diagnostic system is questionable. However, this only applies to age Systemic lupus erythematosus
groups where overlapping pathology is likely. People who develop
Polyarteritis nodosa
dementia relatively early (e.g. in their seventh decade or younger)
are much more likely to have a single underlying pathology. It Buerger’s disease
may therefore still be appropriate to subcategorize dementia in Polycythaemia rubra vera
younger-onset cases. Furthermore, there are particular syndromes
Neurosyphilis
within the ‘vascular dementia’ category that can reasonably be
considered as discrete diagnoses. These include genetic disorders
such as ‘cerebral autosomal dominant arteriopathy with subcorti- prevalences of around 20–30% for dementia 3 months after an acute
cal infarcts and leucoencephalopathy’ (CADASIL), a familial dis- stroke (Tatemichi et al., 1992a). The Framingham Study estimated
order manifesting as recurrent strokes (most often lacunar infarcts) a two-fold increased risk over a 10- year period (Ivan et al., 2004),
usually in the fourth to sixth decades, occurring in the absence of and this was found to remain raised over at least two decades after
vascular risk factors and associated in most cases with a subcortical an index stroke (Kokmen et al., 1996). The two-fold increase was
dementia and pseudobulbar palsy. The principal underlying pathol- confirmed by estimates from a recent systematic review (Savva et
ogy is an abnormality in basal smooth muscle cells, predominantly al., 2010) which also concluded that this was not accounted for by
in the media of small cerebral arteries, and underlying mutations demographic factors or vascular risk profile. One study suggested
have been successfully identified in the Notch3 gene on chromo- a stronger association with cognitive impairment in the absence
some 19 (Joutel et al., 1997). Other syndromes of familial vascular of dementia (Srikanth et al., 2004), but this may reflect differences
dementia have been described (Table 34.1) and it is likely that, as in deciding the point at which mild cognitive impairment is dis-
with Alzheimer’s disease, further discrete disorders exist that may tinguished from dementia (i.e. when it is believed to be affecting
be explained by specific mutations. activities of daily living) in the context of a previous stroke. Selective
As well as genetic disorders, cases of dementia have been mortality (in people with stroke who are at risk of developing demen-
described that appear to have been caused by specific single ‘stra- tia) is likely to be a diluting factor and explain a stronger association
tegic’ infarctions, most often within thalamic structures (Tatemichi in younger compared to older people (Ivan et al., 2004; Savva et al.,
et al., 1992b), and which may also be considered as discrete diag- 2010). High levels of variation between studies in reported preva-
noses. In addition, several noncardiovascular disorders that may lences of poststroke dementia are also influenced by differences in
cause secondary dementia do so through their effects on the vascu- methodology: for example, whether the sample was hospital- or
lature (Table 34.1), although such disorders cause only a very small community-based, and how rigorously prestroke dementia was
proportion of dementia cases. ascertained and excluded. Over 90% of the variation was accounted
for by methodology in a 2009 review (Pendlebury and Rothwell,
2009), although pooled estimates indicated poststroke dementia
How Common Is Dementia After Stroke? prevalences ranging from 7.4% in community samples following
Defining dementia after stroke presents semantic challenges—for first-ever stroke and with rigorous exclusion of prestroke dementia
example, in how to refer to a single major stroke that has caused sub- to 41.3% in hospital samples where recurrent stroke and prestroke
stantial and permanent loss of cognitive function. However, these dementia were included. A more recent study followed 355 older
issues and the difficulties in applying diagnostic criteria for vascu- people who were dementia-free at 3 months after stroke and found
lar dementia should not obscure the importance of vascular disor- an incidence of dementia prior to death in 23.9% over an average
ders as risk factors for dementia. Prospective studies have found follow-up of 3.8 years (Allan et al., 2011).
What Are the Risk Factors for PostStroke be associated with atrophic changes on neuroimaging rather than
early cerebrovascular disease (Pohjasvaara et al., 1999). In addition,
Dementia? not only does stroke predict cognitive impairment but also cog-
Although strategic infarct dementia syndromes have been described nitive impairment is associated with a raised risk of future stroke
(Tatemichi et al., 1992b), population-based studies have not dem- (Ferucci et al., 1996), and dementia following stroke predicts fur-
onstrated a clear association between stroke location and risk of ther stroke episodes (Moroney et al., 1997). Taken together, this evi-
dementia, e.g. with respect to arterial territories of cortical infarc- dence suggests that there is a close (although potentially complex)
tions (Censori et al., 1996; Pohjasvaara et al., 1998; Desmond et al., interrelationship between stroke and primary degenerative changes
2000) and the location (or number) or lacunar infarction (Loeb underlying many cases of apparent vascular dementia, and that the
et al., 1992). Other factors that have been identified as associated stroke itself may be a relatively late event in an ongoing process
with risk of dementia following stroke (albeit not always consist- of insidious cognitive decline. One study found that the clinical
ently) are increased age, lower levels of education, previous stroke picture shifted from one consistent with Alzheimer’s disease over
disease, and vascular risk factors such as diabetes, recent smoking, the first 2 years after a stroke to a more classic ‘vascular dementia’
and atrial fibrillation (Censori et al., 1996; Pohjasvaara et al., 1998; picture over years 2–4 (Altieri et al., 2004). This is consistent with a
Barba et al., 2000; Desmond et al., 2000). The role of blood pres- process in which the stroke episode precipitates dementia in people
sure level is uncertain: most report no association, but one study with pre-existing Alzheimer pathology. Those without this vulner-
found that risk of dementia was associated with lower blood pres- ability maintain their cognitive function initially, but remain at risk
sure and orthostatic changes (Pohjasvaara et al., 1998). Disorders of further strokes and of dementia occurring at a later stage as a
associated with hypoxia and ischaemia (such as seizures, arrhyth- result of these. A nested neuropathological follow-up in the cohort
mias, and pneumonia) have also been found to be associated study reported by Allan et al. (2011) concluded that 75% fulfilled
with higher risk of dementia after stroke (Moroney et al., 1996). current vascular dementia criteria. This is likely also to be consist-
The review by Pendlebury and Rothwell (2009) concluded that ent since the exclusion of dementia at 3 months after stroke may
there was supportive evidence for the following factors predicting have removed most of the ‘precipitated Alzheimer’s disease’ cases
poststroke dementia: (1) older age, female sex, and lower educa- in favour of a more vascular picture. However, it should be borne
tion, among demographic characteristics; (2) prestroke cognitive in mind that a neuropathological diagnosis is simply describing the
decline, disability, diabetes, and atrial fibrillation, among comorbid relative extent of different pathological features and conclusions
health states; and (3) haemorrhagic stroke, left hemisphere stroke, are inevitably limited regarding which one of these actually caused
dysphasia, previous stroke, recurrent stroke, and a range of stroke the ante mortem clinical syndrome. However, the findings do sug-
complications. The authors concluded a central role for the stroke gest at least that Alzheimer pathology was not a major feature in an
itself in dementia prediction rather than underlying vascular risk older cohort developing dementia beyond 3 months after a stroke.
factors; however, subsequent correspondence challenged this, cit- A poststroke cohort is likely to yield high levels of cerebrovascular
ing limitations in the evidence base for the review and highlighting disease, much of which might be present in a similar person who
the potential importance of clinically silent cerebrovascular disease did not develop dementia. Comparisons of neuropathological fea-
(Hennerici, 2009), and the cumulative effect of vascular risk fac- tures between participants who did or did not develop dementia
tors on dementia incidence in the relatively old (age 75+) cohort identified microinfarctions and medial temporal lobe atrophy as
described by Allan et al. (2011) contradicts the review by Savva distinguishing features (Allan et al., 2011).
et al. (2010) which concluded that the effect of stroke on dementia
risk was not accounted for by vascular risk factors.
Although prospective studies following a major stroke have been
Vascular Risk Factors and Dementia
important in understanding dementia associated with cerebrov- Although clinical stroke is strongly associated with dementia, it
ascular disease, they may not be generalizable to situations where does not appear to be the initiating event for cognitive decline in
multiple smaller infarctions have occurred. An early study with a many cases. Prestroke dementia has been estimated to be present
different design investigated factors associated with dementia in in 9–14% of people with stroke, depending on the source of the
a sample of people all of whom had multiple cerebral infarctions sample, and milder prestroke cognitive decline may be more com-
(Gorelick et al., 1993). Independent risk factors were: increased mon still. A large body of complementary evidence now strongly
age, lower educational attainment, previous myocardial infarction, suggests that risk factors for cerebrovascular disease are also risk
recent smoking, and lower systolic blood pressure. CT findings factors for dementia—due to both Alzheimer’s disease and vascular
associated with dementia were: more severe stroke, left cortical inf- dementia, as estimated clinically.
arction, and diffuse enlargement of the lateral ventricle suggesting
cerebral atrophy (Gorelick et al., 1992). Midlife hypertension
Hypertension, a powerful risk factor for cerebrovascular disease,
What Are the Clinical Features of Poststroke has received attention for some time as a risk factor for dementia
(Stewart, 1999). However, considering interest in this relationship
Dementia? since at least the 1960s, it is only relatively recently that consist-
As has been mentioned earlier, many cases of dementia following ent findings began to emerge from prospective studies, although
stroke appear to follow an Alzheimer’s-like clinical course and to even now there remain areas of variation. For example, the fol-
occur without further episodes of infarction (Tatemichi et al., 1994; lowing associations have been found with worse late-life cognitive
Kokmen et al., 1996; Honig et al., 2003). Prestroke cognitive decline function: (1) progressively higher midlife diastolic blood pressure
has been found in many cases (Kase et al., 1998), and appears to (DBP) (Kilander et al., 1998); (2) specifically raised midlife systolic
blood pressure (SBP) (≥160 mmHg compared to < 110 mmHg) in the end of the study was associated with a greater previous increase
men (Launer et al., 1995); (3) progressively higher midlife SBP and in SBP from mid- to late-life and a more pronounced decline in
DBP in people who had not received antihypertensive treatment SBP over a 3- to 6-year period prior to the clinical onset (Stewart
(Elias et al., 1993); (4) persistently high SBP (≥140 mmHg) over a et al., 2009).
38-year period (Swan et al., 1998); and (5) high SBP (≥160 mmHg Taking the above two sections together, there is therefore robust
compared to < 140 mmHg) but not DBP (Kivipelto et al., 2001a). evidence that dementia is predicted by higher blood pressure 10–20
Prospective studies with dementia as an outcome have found years earlier, but that blood pressure undergoes an exaggerated
the following associations: (1) with raised midlife SBP and DBP decline in people with dementia over perhaps 5 years or so before
(≥160/95 mmHg) in men for both dementia and AD, but only in the clinical onset of the condition. These appear to be exaggerations
those not previously receiving antihypertensive agents (Launer et of changes expected with age (a rise in blood pressure up to and
al., 2000); (2) with higher SBP at age 70 and higher DBP at age 70 including midlife followed by a decline in later life) and, although
and 75 for participants with AD aged 85 (Skoog et al., 1996); and evidence is less substantial, tend to be also observed for other vas-
(3) with midlife high SBP (≥160 mmHg compared to < 140 mmHg) cular risk factors as described below.
but not high DBP for AD (Kivipelto et al., 2001b). Although most
have analyzed midlife blood pressure as an exposure, it appears that Dyslipidaemia
it is the interval between measurements that is important rather Raised lipid levels have received less attention than blood pressure
than the age at which blood pressure is measured, since delayed risk as risk factors for dementia and results are also conflicting. Some
associations (over a 15-year follow-up) have also been found for studies, predominantly from Scandinavian populations, have sug-
late-life systolic and diastolic blood pressure (Skoog et al., 1996). gested that raised mid-life total cholesterol is associated with later
dementia (Notkola et al., 1998; Kivipelto et al., 2001b; Kivipelto
Late-life hypotension et al., 2002; Kivipelto et al., 2005). However others have found over
Cross-sectional surveys, on the other hand, have tended to find that shorter periods of follow-up that higher total cholesterol levels are
people with dementia have lower blood pressure than those with- associated with lower risk of dementia (Mielke et al., 2005) or show
out, although this is not universally reported. Two large surveys in no association at all (Reitz et al., 2004). Confirming this, a 2008
Sweden and the US, where blood pressure was compared between review concluded consistent evidence for associations between
participants with and without dementia, found lower SBP and DBP midlife total cholesterol levels and risk of Alzheimer’s disease, but
in the former (Guo et al., 1996; Morris et al., 2000), reporting on no evidence for associations with late-life cholesterol levels (Anstey
both all dementia and AD specifically as outcomes in the first of et al., 2008) and findings from three studies have suggested an exag-
these studies (Guo et al., 1996) and just AD as an outcome in the gerated decline in cholesterol levels from mid- to late-life in people
second (Morris et al., 2000). On the other hand, one large survey with dementia compared to comparison survivors (Notkola et al.,
of AD in Finland did not find any difference in blood pressure by 1998; Stewart et al., 2007; Mielke et al., 2010).
dementia status (Kuusisto et al., 1997) and a survey in Japan found
no difference in hypertensive blood pressure levels between partici- Diet and obesity
pants with and without AD (Ueda et al., 1992). Low blood pressure A large amount of attention has focused on dietary risk or protec-
is common in older people and (for both SBP and DBP) known tive factors for dementia. With many measures to choose from and
to be associated with adverse consequences (Mattila et al., 1988; a sizeable risk of false positive findings from multiple analyses, it
Boshuizen et al., 1998; Satish et al., 2001). Hypotensive orthostatic is difficult to draw firm conclusions; however, there are several
blood pressure changes have also been found to be associated with reports that an atherogenic diet is associated with increased risk
dementia, in particular a reduced systolic blood pressure response (Kalmijn et al., 1997; Luchsinger et al., 2002), and a French study
to standing (Vitiello et al., 1993). In a Swedish cohort without found higher adherence to a Mediterranean diet to be associated
dementia, frontal and parietal atrophy was associated with lower with a lower risk of cognitive decline (Féart et al., 2009). As with
SBP and parietal atrophy with lower DBP; those in the cohort with cholesterol and blood pressure, the association between obesity and
Alzheimer’s disease had lower SBP and DBP, and those with vas- dementia has been complicated by the observation of changes in
cular dementia had lower DBP (Skoog et al., 1998). Findings from the risk factor (in this case, accelerated weight loss) prior to the
several studies suggest that, within samples with dementia, lower clinical onset of dementia (Stewart et al., 2005). Initial studies sug-
blood pressure level is also associated with more advanced stage of gesting that midlife obesity, measured by body mass index, is a risk
the disease (Guo et al., 1996; Hogan et al., 1997; Skoog et al., 1998), factor for dementia (Gustafson et al., 2003) have been confirmed
suggesting that there may be effects of neurodegenerative disorders by a systematic review (Gorospe and Dave, 2007) and subsequent
on blood pressure levels. From those studies investigating change in research (Hassing et al., 2009), including studies measuring cen-
rather than level of blood pressure, the following findings have been tral obesity (Whitmer et al., 2008). However, cohorts with shorter
reported: (1) an association between 30-year decline in SBP and follow-up periods have found the opposite (Nourhashemi et al.,
worse psychomotor speed (Swan et al., 1998); (2) cognitive decline 2003) or more complex U-shaped or age-dependent patterns of
over a 3-year period being associated with a 10-mmHg SBP decline association (Rosengren et al., 2005; Luchsinger et al., 2007).
over the same period (Zhu et al., 1998); (3) dementia in women
associated with a reduced level of SBP increase and a higher level of Diabetes and metabolic syndrome
DBP decline over a 10-year period (Petitti et al., 2005); (4) demen- Type 2 (non-insulin-dependent) diabetes has also been found to be
tia and Alzheimer’s disease associated with SBP and DBP decline a risk factor for dementia, both vascular dementia and Alzheimer’s
over the previous 3 years (Qiu et al., 2004); and (5) a 30-year cohort disease (Ott et al., 1999; Peila et al., 2002). Selective mortality is
study of Japanese-American men finding that incident dementia at likely to be an important factor, reducing the co-occurrence of
the two conditions, which may explain some negative findings are more physically active have a lower risk of dementia (Yoshitake
(MacKnight et al., 2002). Strong interactions were reported between et al., 1995; Laurin et al., 2001; Rovio et al., 2005), which may or
hypertension and diabetes as predictors of cognitive impairment may not be mediated through effects on the vasculature.
in the Framingham Study (Elias et al., 1997). However, much of
the effect of diabetes on risk of dementia appears to be independ-
ent of cerebrovascular disease and may involve other nonvascular
Mechanisms of Association
pathways or common underlying factors (Stewart and Liolitsa, What is becoming increasingly apparent is that people at increased
1999). Several studies have also reported associations between risk for stroke are also at increased risk for dementia, whether this
risk of dementia and the prediabetic state of insulin resistance or is defined clinically as vascular dementia (i.e. with evidence of sig-
‘metabolic syndrome’, either estimated through vascular risk fac- nificant comorbid cerebrovascular disease) or Alzheimer’s disease.
tor clustering (Kalmijn et al., 2000) or directly measured hyper- The apparent associations with Alzheimer’s disease have not been
insulinaemia (Luchsinger et al., 2004). One study, however, found clearly explained by missed infarctions (Ott et al., 1999), or by
that both low and high insulin were associated with later dementia dementia with primary vascular pathology being misclassified as
(Peila et al., 2004). In people with diabetes, there is some evidence Alzheimer’s disease (Hulette et al., 1997), although this is difficult
for profiles associated with higher risk of cognitive decline and/ to exclude absolutely. One neuropathological study of a commu-
or dementia. In one study, insulin treatment was associated with a nity population suggested very high prevalence of mixed vascular
higher risk of dementia (Ott et al., 1999) and mild cognitive impair- and Alzheimer pathology (Fernando et al., 2004). An explanation
ment in another (Roberts et al., 2008), the latter study also finding therefore is more likely to lie in links between vascular factors and
associations of cognitive impairment with longer duration, earlier Alzheimer’s disease. These may involve four possible pathways
onset, and complications. Possibly consistent with the association which need not be mutually exclusive.
between insulin treatment and dementia risk, another study follow-
ing a large cohort of people with type 2 diabetes found a higher risk Interactions at a pathological level
associated with one or more hypoglycaemic episodes (Whitmer Cerebrovascular pathology in dementia is reviewed elsewhere in
et al., 2009). this book (see Chapter 6). There are numerous theoretical ways in
which vascular processes might induce or accelerate Alzheimer
Smoking pathology, including amyloid deposition as a response to ischae-
A meta-analysis of 19 studies of smoking and cognitive outcomes mia, links through inflammatory pathways, blood–brain barrier
with at least 12 months follow-up was published in 2007, containing disturbance secondary to cerebrovascular disease, and for diabetes
data on over 26,000 participants followed from 2–30 years (Anstey abnormal protein glycation secondary to prolonged hyperglycae-
et al., 2007). From pooled data, the risk of incident dementia was mia. These processes predict that people with vascular disease will
concluded to be 27% higher in current smokers than never smok- have higher levels of Alzheimer pathology—a hypothesis that is
ers, with correspondingly raised risks of incident Alzheimer’s dis- difficult to test except in the rare instances where neuropathologi-
ease by 79% and vascular dementia by 78%. A second review and cal follow-up has been carried out in people without dementia. An
meta-analysis published in 2008 concluded a 59% increased risk of early study found increased Alzheimer pathology associated with
Alzheimer’s disease in current smokers (Peters et al., 2008). Of the hypertension or coronary artery disease in people without previous
larger cohort studies to have investigated this issue, two reported dementia (Sparks et al., 1995; Sparks et al., 1996). Elevated midlife
‘dose-response’ associations, i.e. increasing risk from light to heavy blood pressure was also associated with decreased brain volume
smokers (Tyas et al., 2003; Juan et al., 2004). Relevant studies pub- and Alzheimer pathology in a cohort study with a relatively large
lished subsequent to the systematic reviews include two large stud- pathological follow-up (Petrovitch et al., 2000). The same study
ies indicating positive associations between smoking and dementia also found associations between diabetes and increased Alzheimer
hospitalization or healthcare contats (Alonso et al., 2009; Rusanen pathology (Peila et al., 2002), although associations were found
et al., 2011) and strong associations between midlife smoking sta- with a protective rather than atherogenic lipid profile (Launer et al.,
tus and late-life dementia risk in 1449 people aged 65–79 years 2001), a finding that requires further clarification. Associations
(Rusanen et al., 2010). Both meta-analyses cited above found no have been reported between smoking and neuritic plaques in a
association between former smoking and dementia risk (Anstey neuropathological follow-up nested within a large cohort study
et al., 2007; Peters et al., 2008). One recent study has also reported (Tyas et al., 2003), suggesting direct influences on Alzheimer’s dis-
an association between passive smoking and increased dementia ease for this exposure; however, associations with reduced cerebral
risk (Barnes et al., 2010), supported by another finding associations perfusion have also been reported (Siennicki-Lantz et al., 2008) in
between passive smoking and cognitive impairment (Llewellyn addition to the well-recognized adverse effects of smoking on cere-
et al., 2009). Finally, a recent review concluded that studies with brovascular disease. Not all studies have found consistent evidence
tobacco industry affiliation were more likely to report a protective for these types of associations—one community study found that
effect for dementia and less likely to report a risk effect than studies raised blood pressure was associated with microinfarcts but not
without such affiliation (Cataldo et al., 2010). with other pathological changes (Wang et al., 2009) and another
found less Alzheimer pathology with previous medicated hyper-
Other vascular risk factors tensive compared to normotensive status (Hoffman et al., 2009).
Other vascular factors associated with dementia include ECG
ischaemia (Prince et al., 1994) and atrial fibrillation (Ott et al., Clinical/symptomatic interactions
1997), as well as measures of peripheral and carotid atherosclerosis As well as acting directly on the progression of Alzheimer pathol-
(Hofman et al., 1997). Several studies have found that people who ogy, vascular disorders may accelerate the onset of symptomatic
dementia at relatively early stages of comorbid Alzheimer’s disease. Table 34.2 A life-course model of the relationship between vascular
In an early US study of older nuns who were screened in late life factors and dementia
and followed to post mortem, a lower level of Alzheimer pathology
Infancy and Genetic factors determining later vascular risk and
was observed in association with dementia if infarction was also
childhood cognitive impairment
present, suggesting that the infarction had accelerated the onset
Environmental stressors affecting both vascular risk (e.g.
(Snowdon et al., 1997). Similar findings were observed around that
blood pressure, proneness to obesity) and cognitive
time in the Oxford OPTIMA study, where early Alzheimer pathol- function (level of attainment in childhood and/or
ogy was associated with much greater cognitive impairment if cer- vulnerability to later neurodegeneration)
ebrovascular disease was also present (Esiri et al., 1999). In another
Early adulthood Level of ‘attained’ cognitive function set determining risk
study using a community-derived sample, significant interactions
of later impairment (‘reserve’)
were found between the presence of cerebral amyloid angiopathy
Socioeconomic status influencing vascular risk profile,
and Alzheimer pathology in the associations of these changes with
and risk behaviour (smoking, diet, physical activity)
previous level of cognitive impairment (Pfeifer et al., 2002). Other
neuropathological follow-up studies of community samples have Midlife Vascular risk factors becoming manifest (hypertension,
suggested a predominance of mixed pathology (Fernando et al., obesity, dyslipidaemia)
2004), with independent influences of Alzheimer’s and nonAlzhe- Possible subtle early cognitive changes secondary to
imer pathology on cognitive function (White et al., 2005) or on the vascular damage; also very early Alzheimer’s and vascular
pathological changes
likelihood of dementia at death (Matthews et al., 2009).
One way in which vascular factors may influence Alzheimer’s Continuing manifestation of lifestyle-related risk factors
disease is that memory impairment secondary to early Alzheimer (diet, smoking, exercise)
pathology may be more likely to be noticed as ‘dementia’ if other Late life Impact of clinical cerebrovascular disease (stroke, TIA)
cognitive domains are also affected. White matter disease may be on cognitive function
important in this respect, since subtle disruption of frontosubcorti- Co-occurring or consequent Alzheimer’s disease
cal pathways may result in impaired executive function. White mat- Metabolic changes associated with frailty (e.g. weight
ter hyperintensities on magnetic resonance imaging are common loss, decline in blood pressure) possibly exaggerated in
in older age groups and are associated with vascular risk factors, preclinical and clinical dementia
particularly hypertension (Breteler et al., 1994b). Although they are Chronic effects of raised vascular risk (e.g. diminished
more common in dementia and, across a population, are associ- blood pressure reactivity)
ated with relative cognitive impairment (Breteler et al., 1994a), at Selective mortality (survivors with vascular risk factors
an individual level they may be severe without any apparent clinical potentially at lower risk due to other unknown
manifestations (Fein et al., 1990). They do not therefore appear to protective factors)
be sufficient in themselves to cause dementia but may precipitate
this in the presence of other pathology, i.e. in effect, reduce the age
of dementia onset. pathological processes begin one or two decades before clinical
symptoms become manifest. Deposition of amyloid occurs in cer-
Common underlying factors ebral blood vessels as well as the brain parenchyma in Alzheimer’s
An association between vascular disease and Alzheimer’s disease disease. This ‘cerebral amyloid angiopathy’, described in detail in
may be explained by a common underlying risk factor. Both vas- Chapter 7, is associated with both cerebral haemorrhage and small
cular status and dementia have causal processes operating across infarctions (Olichney et al., 2000), and is a common and appar-
the life-course, with ample opportunity for interaction at many ently parallel pathology to parenchymal amyloid deposition (Keage
stages (Table 34.2); most research fails or is unable to take this into et al., 2009). Abnormalities in capillary structure have also been
consideration and is potentially limited by focusing at a very late reported in Alzheimer’s disease (de la Torre and Mussivand, 1993).
stage in long-term evolving processes. Vascular risk factors such It is therefore possible that the presence of Alzheimer’s disease may
as hypertension and diabetes are commonly classified as ‘envi- exacerbate or accelerate vascular pathology, or render the brain
ronmental’ risk factors for dementia. However, they are known to more vulnerable to further insults. Early cognitive decline may also
have a substantial familial aetiology and it is possible that common affect factors such as diet, exercise, and adherence to prescribed
genetic factors explain some of their association with Alzheimer’s medication, which in turn may influence the risk of cerebrovascu-
disease (Lovestone, 1999; Stewart and Liolitsa, 1999). Lifestyle fac- lar events.
tors such as diet, physical activity, and personality could also poten-
tially underlie later associations. Similarly, both cognitive function Vascular Risk Factors and Dementia
and risk of vascular disease are determined by socioeconomic sta-
tus from childhood onwards, and at least part of the overlap in later Treatment—Clinical Implications
life may reflect long-standing social inequalities—a mechanism What implications does current research have for clinicians? In the-
that has tended to be ignored in preference to biological links. ory, these should be numerous since vascular disease is one of few
potentially modifiable risk factors for dementia. However, although
Effects of Alzheimer’s disease on the vasculature research attention is increasing in this area, there is woefully little
It is possible that Alzheimer’s disease induces or exacerbates cer- direct evidence, at present, for any intervention to prevent or treat
ebrovascular pathology. Even long-duration prospective stud- dementia through modifying vascular risk. Despite this, there are
ies cannot conclusively demonstrate the direction of causation if particular issues that can be addressed to some extent.
Treatment of dementia through modifying multiple infarctions and dementia. However, it is important to bear
vascular risk in mind that, while some may consider an acute fatal stroke to be
a preferable alternative to end-stage dementia, a nonfatal episode
If vascular disease were to cause dementia through damag-
may lead to a lengthy period of disability and suffering that might
ing the brain directly, or through accelerating the progression of
have been prevented. It is now accepted that intensive screening for
Alzheimer pathology, then it is likely that these processes would
preventable risk factors should take place in people with single or
continue to contribute to the progression of disease after diagno-
recurrent strokes. There is no good reason why people who happen
sis. Interventions to halt or slow the progression of vascular disease
to have dementia should be excluded from that process, although
could therefore be expected at least to prevent further cognitive
an undoubted problem with vascular dementia has been its ‘falling
decline, and possibly even improve cognitive function. However,
between stools’ and the risk for people with this diagnosis receiving
there has been little research into potential interventions. Even for
suboptimal attention from both stroke and dementia services.
aspirin, only one randomized trial has been published: a pilot study
of 70 patients with multi-infarct dementia randomized to aspi- Treatment of dementia in people who have
rin or no additional treatment over 3 years (Meyer et al., 1989).
cerebrovascular disease
Improvement in cognitive scores and cerebral blood flow were
noted over the first 2 years in the treatment group, although the Although there is no good research evidence to suggest that ‘vas-
protocol did not involve a placebo and participants were not blind cular dementia’ and ‘Alzheimer’s disease’ can be adequately sepa-
to their allocation. Another study found that men at high risk of rated as distinct disorders, the persistence of the two diagnoses in
cardiovascular disease who had received warfarin or aspirin (as clinical parlance has led to their application in clinical trials and,
part of a randomized double-blind trial) had better cognitive func- hence, to determining treatment ‘eligibility’. Although there is some
tion at the end of a 5-year trial period than those receiving placebo trial evidence for a beneficial effect of acetylcholinesterase (ACE)
(Richards et al., 1997). Cognitive assessment was not carried out inhibitors in people with ‘vascular dementia’, this is most evident
at the start of the trial, but, since allocation was randomized and for the subgroup with potentially mixed disease (Erkinjuntti et al.,
groups were similar in many other respects, it is likely that this rep- 2002), and a meta-analysis concluded uncertain benefit in the dis-
resents an effect of the intervention. The association with higher order as a whole (Kavirajan and Schneider, 2007). This may reflect
cognitive scores was principally in those receiving aspirin rather previous findings suggesting that cholinergic deficits in pure vas-
than warfarin. Finally, one trial reported a beneficial effect of ator- cular dementia are not present, or are less strong than those in
vastatin on progression of Alzheimer’s disease (Sparks et al., 2005), Alzheimer’s disease or mixed dementia (Perry et al., 2005; Sharp
although it is not certain whether this effect was due to vascular et al., 2009). However, people with ‘pure’ vascular dementia have
or other effects of this cholesterol-lowering agent. Furthermore, also been very hard to identify, and the lack of detectable cogni-
the findings were equivocal at 12 months follow-up and it should tive decline in the placebo group over the course of a standard trial
be borne in mind that other statin trials have failed to show clear means that treatment effects are hard to demonstrate. The danger is
benefits on cognitive decline prior to dementia (as discussed fur- that difficulties encountered in applying diagnostic criteria (which
ther in Vascular Risk Factors and Dementia—Implications for are predictable given the lack of evidence to support their applica-
Prevention of Dementia). Apart from drug treatments, other life- bility) limit enthusiasm for further trials in people with cerebrovas-
style changes may have an impact. There is randomized controlled cular disease and dementia, resulting in an impoverished evidence
trial evidence, for example, of a benefit of aerobic against anaero- base. Recommended indications for future treatments will be lim-
bic exercise on cognitive function in sedentary men (Kramer et al., ited to Alzheimer’s disease alone because of a lack of evidence in
1999) and a beneficial effect of a 6-month exercise programme over other disorders. If dementia with cerebrovascular disease is classi-
an 18-month follow-up period in participants reporting memory fied as ‘vascular’ in clinical practice, then people with Alzheimer’s
difficulties (Lautenschlager et al., 2008). Important questions, such disease may fail to receive treatment on the basis of comorbidity.
as the benefit or not of improved glycaemic control in people with
comorbid dementia and diabetes, remain unanswered in interven- Vascular Risk Factors and Dementia—
tion studies. Implications for Prevention of Dementia
The implications of current research for the prevention of dementia
Prevention of stroke in dementia are substantial, whether vascular risk factors directly induce dementia
It is therefore uncertain whether modification of vascular risk in pathology or accelerate the onset of the clinical syndrome. However,
people with established dementia has an impact on the course of effects may be difficult to demonstrate in conventional clinical trials
cognitive decline. However, there is also little evidence at present because of the potentially long period over which risk factors exert
that such measures are contraindicated. The prescription of aspirin their action, and the changed relationships between risk factor levels
or the treatment of hypertension in a patient with dementia and and dementia, closer to the clinical onset of the latter. Furthermore,
cerebrovascular disease might be carried out not only in the hope most interventions may be well established as beneficial for other
of preventing further cognitive decline but also to prevent stroke. reasons (e.g. preventing cardiac disease and stroke), so that it may
Complex ethical issues surround the question of how intensively to not be possible ethically to have a placebo group, or to retain a
treat comorbid disease in individuals with clinical dementia, partic- double-blind placebo comparison, for a sufficient time period to test
ularly when the latter is at an advanced stage. In addition, although an impact on dementia incidence. All of these issues may explain the
there may be no upper age limit for stroke prevention (Staessen et largely negative results of antihypertensive agents—most trials find-
al., 2000), the effectiveness of interventions such as blood pressure ing no effect on incidence of dementia or cognitive decline (Prince
control has not been adequately assessed in the context of comorbid et al., 1996; Starr et al., 1996; Lithell et al., 2003; Tzourio et al., 2003)
but all probably of far too short a duration to detect meaningful cog- precipitating, or maintaining factors. This approach is at least in
nitive outcomes. In a subgroup analysis of the Perindopril Protection keeping with the tradition of the diagnostic formulation and with
Against Recurrent Stroke Study (PROGRESS), an effect was found of the reality of multiple, interacting, and overlapping disorders (and
an ACE inhibitor and a diuretic on cognitive decline and dementia causes for disorders) in older age groups.
in people with recurrent stroke over the follow-up period (Tzourio
et al., 2003), and a further analysis showed an effect on slowing the References
progression of white matter lesions (Dufouil et al., 2005). A more
Allan, L.M. et al. (2011). Long term incidence of dementia, predictors of
conclusive effect on dementia incidence was found in the Systolic mortality and pathological diagnosis in older stroke survivors. Brain,
Hypertension in Europe (SYST-EUR) trial on nimodipine (Forette 134, 3713–24.
et al., 1998), which was sustained in open-label follow-up (Forette Alonso, A., et al. (2009). Risk of dementia hospitalisation associated with
et al., 2002) and whose findings have received further support by cardiovascular risk factors in midlife and older age: the Atherosclerosis
other small trials of calcium channel blockers in dementia (Birks Risk in Communities (ARIC) study. Journal of Neurology, Neurosurgery
and López-Arrieta, 2002). and Psychiatry, 80, 1194–201.
With respect to cholesterol lowering, three trials of statins have Altieri, M., et al. (2004). Delayed poststroke dementia: a 4-year follow-up
study. Neurology, 62, 2193–7.
found no effect on cognitive function or decline (Santanello et al.,
American Psychiatric Association (1994). Diagnostic and Statistical Manual
1997; Heart Protection Study Collaborative Group, 2002; Shepherd
of Mental Disorders, 4th edition. APA, Washington, DC.
et al., 2002; McGuiness et al., 2009); however, the trial findings of a Anstey, K.J., et al. (2007). Smoking as a risk factor for dementia and cognitive
possible effect of atorvastatin (which has different lipid solubility) decline: a meta-analysis of prospective studies. American Journal of
on cognitive decline in people with dementia (Sparks et al., 2005) Epidemiology, 166, 367–78.
may indicate that further research is required. Anstey, K.J., Lipnicki, D.M., and Low, L.F. (2008). Cholesterol as a risk factor for
As discussed under Vascular Risk Factors and Dementia, dementia and cognitive decline: a systematic review of prospective studies
although high blood pressure is a risk factor for dementia, blood with meta-analysis. American Journal of Geriatric Psychiatry, 16, 343–54.
pressure is frequently lower than average by the time clinical Barba, R., et al. (2000). Poststroke dementia: clinical features and risk factors.
dementia has developed. A concern obviously arises that over- Stroke, 31, 1494–501.
Barnes, D.E., et al. (2010). Secondhand smoke, vascular disease, and dementia
zealous correction of blood pressure may itself be a risk factor for
incidence: findings from the Cardiovascular Health Cognition Study.
dementia (e.g. through episodes of hypoperfusion, and ischaemia American Journal of Epidemiology, 171, 292–302.
or infarction in critical ‘watershed’ zones). However, currently Birks, J. and López-Arrieta, J. (2002). Nimodipine for primary degenerative,
there is little evidence to support this. Blood pressure is observed to mixed and vascular dementia. Cochrane Database of Systematic Reviews,
be progressively lower at increasingly advanced stages of dementia 3, CD000147. doi: 10.1002/14651858.
(Guo et al., 1996), suggesting that it is a secondary phenomenon Boshuizen, H.C., et al. (1998). Blood pressure and mortality in elderly people
and/or a marker of general physical frailty. One observational study aged 85 and older: community based study. British Medical Journal, 316,
of patients with multi-infarct dementia (Meyer et al., 1986) sug- 1780–4.
gested that a better clinical course occurred if systolic blood pres- Bowler, J.V. and Hachinski, V. (2000). Criteria for vascular dementia: replacing
dogma with data. Archives of Neurology, 57, 170–1.
sure remained within the ‘upper limits of normal’ (that is 135–150
Breteler, M.M.B., et al. (1994a). Cognitive correlates of ventricular
mmHg) compared to lower levels. However, it cannot be concluded
enlargement and cerebral white matter lesions on magnetic resonance
whether this observation was explained by blood pressure treat- imaging. Stroke, 25, 1109–15.
ment, factors associated with the dementia, or comorbid disease. Breteler, M.M.B., et al. (1994b). Cerebral white matter lesions, vascular
risk factors, and cognitive function in a population-based study: the
Rotterdam Study. Neurology, 44, 1246–52.
Conclusion: What Is Vascular Dementia? Cataldo, J.K., Prochaska, J.J., and Glantz, S.A. (2010). Cigarette smoking is
Vascular disease has long been recognized as an important cause a risk factor for Alzheimer’s disease: an analysis controlling for tobacco
of dementia. Current evidence supports this, even though vascular industry affiliation. Journal of Alzheimer’s Disease, 9, 465–80.
disorders are frequently not a single cause in isolation, but instead Censori, B., et al. (1996). Dementia after first stroke. Stroke, 27, 1205–10.
interact with other neuropathological processes. Renewed recog- Chui, H.C., et al. (2000). Clinical criteria for the diagnosis of vascular
dementia. Archives of Neurology, 57, 191–6.
nition of this and of the potential for prevention or treatment of
de la Torre, J.C. and Mussivand, T. (1993). Can disturbed brain microcirculation
dementia has resulted in welcome but long-overdue research inter- cause Alzheimer’s disease? Neurological Research, 15, 146–53.
est and advice for those affected (e.g. <http://www.alzheimers. Desmond, D.W., et al. (2000). Frequency and clinical determinants of
org.uk/VascularDementia/index.htm>). Vascular dementia as a dementia after ischemic stroke. Neurology, 54, 1124–31.
research field is therefore alive and well. As discussed in Chapter 29, Dufouil, C., et al. (2005). Effects of blood pressure lowering on cerebral
part 4, the author’s view is that there remain major drawbacks with white matter hyperintensities in patients with stroke: the PROGRESS
subcategorizing dementia, ‘vascular dementia’ being a particularly (Perindopril Protection Against Recurrent Stroke Study) Magnetic
problematic feature. Its continued usefulness is doubted, although Resonance Imaging Substudy. Circulation, 112, 1644–50.
there is uncertainty as to what should replace it. ‘Vascular cognitive Elias, M.F., et al. (1993). Untreated blood pressure level is inversely related
to cognitive functioning: the Framingham study. American Journal of
impairment’ achieves a shift away from a memory-focused view of
Epidemiology, 138, 353–64.
dementia. However, the name itself still assumes that cerebrovascu-
Elias, P.K., et al. (1997). NIDDM and blood pressure as risk factors for poor
lar disease can be separated out and identified as a single underly- cognitive performance. Diabetes Care, 20, 1388–95.
ing cause of cognitive impairment. For clinical purposes, it may be Erkinjuntti, T., et al. (2002). Efficacy of galantamine in probable vascular
most valid to consider dementia as the principal diagnosis and to dementia and Alzheimer’s disease combined with cerebrovascular
include vascular disease as one of several potential predisposing, disease: a randomised trial. Lancet, 359, 1283–90.
Esiri, M.M., et al. (1999). Cerebrovascular disease and threshold for dementia Ivan, C.S., et al. (2004). Dementia after stroke: the Framingham Study. Stroke,
in the early stages of Alzheimer’s disease. Lancet, 354, 919–20. 35, 1264–68.
Féart, C., et al. (2009). Adherence to a Mediterranean diet, cognitive decline, Joutel, A., et al. (1997). Strong clustering and stereotyped nature of Notch3
and risk of dementia. Journal of the American Medical Association, 302, mutations in CADASIL patients. Lancet, 350, 1511–15.
638–48. Juan, D., et al. (2004). A 2-year follow-up study of cigarette smoking and risk
Fein, G., et al. (1990). Preservation of normal cognitive functioning in elderly of dementia. European Journal of Neurology, 11, 277–82.
subjects with extensive white-matter lesions of long duration. Archives of Kalmijn, S., et al. (1997). Dietary fat intake and the risk of incident dementia
General Psychiatry, 47, 220–3. in the Rotterdam Study. Annals of Neurology, 42, 776–82.
Fernando, M.S., Ince, P.G., and MRC Cognitive Function and Ageing Kalmijn, S., et al. (2000). Metabolic cardiovascular syndrome and risk of
Neuropathology Study Group (2004). Vascular pathologies and dementia in Japanese-American elderly men. Arteriosclerosis, Thrombosis
cognition in a population-based cohort of elderly people. Journal of the and Vascular Biology, 20, 2255–60.
Neurological Sciences, 226, 13–17. Kase, C.S., et al. (1998). Intellectual decline after stroke: the Framingham
Ferucci, L., et al. (1996). Cognitive impairment and risk of stroke in the older Study. Stroke, 29, 805–12.
population. Journal of the American Geriatrics Society, 44, 237–41. Kavirajan, H. and Schneider, L.S. (2007). Efficacy and adverse effects of
Firbank, M.J., et al. (2007). Medial temporal atrophy rather than white cholinesterase inhibitors and memantine in vascular dementia: a
matter hyperintensities predict cognitive decline in stroke survivors. meta-analysis of randomised controlled trials. Lancet Neurology, 6,
Neurobiology of Aging, 28, 1664–9. 782–92.
Firbank, M.J., et al. (2012). Neuroimaging predictors of death and dementia Keage, H.A.D., et al. (2009). Population studies of sporadic cerebral amyloid
in a cohort of older stroke survivors. Journal of Neurology, Neurosurgery angiopathy and dementia: a systematic review. BMC Neurology, 9, 3
and Psychiatry, 83, 263–7. Kilander, L., et al. (1998). Hypertension is related to cognitive impairment. A
Fischer, P., et al. (1990). Course characteristics in the differentiation of 20-year follow-up of 999 men. Hypertension, 31, 780–6.
dementia of the Alzheimer type and multi-infarct dementia. Acta Kivipelto, M., et al. (2001a). Midlife vascular risk factors and late-life mild
Psychiatrica Scandinavica, 81, 551–3. cognitive impairment. Neurology, 56, 1683–9.
Forette, F., et al. (1998). Prevention of dementia in randomised double-blind Kivipelto, M., et al. (2001b). Midlife vascular risk factors and Alzheimer’s
placebo-controlled Systolic Hypertension in Europe (Syst-Eur) trial. disease in later life: longitudinal, population based study. British Medical
Lancet, 352, 1347–51. Journal, 322, 1447–51.
Forette, F., et al. (2002). The prevention of dementia with antihypertensive Kivipelto, M., et al. (2002). Apolipoprotein E e4 allele, elevated midlife total
treatment. New evidence from the Systolic Hypertension in Europe cholesterol level, and high midlife systolic blood pressure are independent
(Syst-Eur) study. Archives of Internal Medicine, 162, 2046–52. risk factors for late-life Alzheimer disease. Annals of Internal Medicine,
Gorelick, P.B., et al. (1992). Cranial computerised tomographic observations 137, 149–55.
in multi-infarct dementia. Stroke, 23, 804–11. Kivipelto, M., et al. (2005). Obesity and vascular risk factors at midlife and
Gorelick, P.B., et al. (1993). Risk factors for dementia associated with multiple the risk of dementia and Alzheimer’s disease. Archives of Neurology, 62,
cerebral infarcts. Archives of Neurology, 50, 714–20. 1556–60.
Gorospe, E.C. and Dave, J.K. (2007). The risk of dementia with increased Kokmen, E., et al. (1996). Dementia after ischemic stroke: a population-based
body mass. Age and Ageing, 36, 23–9. study in Rochester, Minnesota (1960–1984). Neurology, 19, 154–9.
Guo, Z., et al. (1996). Low blood pressure and dementia in elderly people: the Kramer, A.F., et al. (1999). Ageing, fitness and neurocognitive function.
Kungsholmen project. British Medical Journal, 312, 805–8. Nature, 400, 418–19.
Gustafson, D., et al. (2003). An 18-year follow-up of overweight and risk of Kuusisto, J., et al. (1997). Association between features of the insulin resistance
Alzheimer disease. Archives of Internal Medicine, 163, 1524–8. syndrome and Alzheimer’s disease independently of apolipoprotein
Hachinski, V.C., et al. (1975). Cerebral blood flow in dementia. Archives of E4 phenotype: cross sectional population based study. British Medical
Neurology, 32, 632–7. Journal, 315, 1045–9.
Hassing, L.B., et al. (2009). Overweight in midlife and risk of dementia: a Launer, L.J., et al. (1995). The association between midlife blood pressure
40-year follow-up study. International Journal of Obesity, 33, 893–8. levels and late-life cognitive function. Journal of the American Medical
Heart Protection Study Collaborative Group (2002). MRC/BHF Heart Association, 274, 1846–51.
Protection Study of cholesterol lowering with simvastatin in 20,536 Launer, L.J., et al. (2000). Midlife blood pressure and dementia: the
high-risk individuals: a randomised placebo-controlled trial. Lancet, Honolulu-Asia aging study. Neurobiology of Aging, 21, 49–55.
360, 7–22. Launer, L.J., et al. (2001). Cholesterol and neuropathologic markers of AD. A
Hennerici, M.G. (2009). What are the mechanisms for post-stroke dementia? population-based autopsy study. Neurology, 57, 1447–52.
Lancet Neurology, 8, 973–5. Laurin, D., et al. (2001). Physical activity and risk of cognitive impairment
Hoffman, L.B., et al. (2009). Less Alzheimer disease neuropathology in medicated and dementia in elderly persons. Archives of Neurology, 58, 498–504.
hypertensive than nonhypertensive persons. Neurology, 72, 1720–6. Lautenschlager, N.T., et al. (2008). Effect of physical activity on cognitive
Hofman, A., et al. (1997). Atherosclerosis, apolipoprotein E, and prevalence function in older adults at risk for Alzheimer disease. Journal of the
of dementia and Alzheimer’s disease in the Rotterdam Study. Lancet, American Medical Association, 300, 1027–37.
349, 151–4. Lithell, H., et al. (2003). The Study on Cognition and Prognosis in the Elderly
Hogan, D.B., Ebly, E.M., and Rockwood, K. (1997). Weight, blood pressure, (SCOPE): principal results of a randomized double-blind intervention
osmolarity, and glucose levels across various stages of Alzheimer’s disease trial. Journal of Hypertension, 21, 875–86.
and vascular dementia. Dementia and Geriatric Cognitive Disorders, 8, Llewellyn, D.J., et al. (2009). Exposure to secondhand smoke and cognitive
147–51. impairment in non-smokers: national cross sectional study with cotinine
Holmes, C., et al. (1999). Validity of current clinical criteria for Alzheimer’s measurement. British Medical Journal, 338, b462.
disease, vascular dementia and dementia with Lewy bodies. British Loeb, C., et al. (1992). Dementia associated with lacunar infarction. Stroke,
Journal of Psychiatry, 174, 45–50. 23, 1225–9.
Honig, L.S., et al. (2003). Stroke and the risk of Alzheimer disease. Archives Lopez, O.L., et al. (1994). Reliability of NINDS-AIREN clinical criteria for
of Neurology, 60, 1707–12. the diagnosis of vascular dementia. Neurology, 44, 1240–5.
Hulette, C., et al. (1997). Clinical-neuropathological findings in multi-infarct Lovestone, S. (1999). Diabetes and dementia: is the brain another site of
dementia: a report of six autopsied cases. Neurology, 48, 668–72. end-organ damage? Neurology, 53, 1907–9.
Luchsinger, J.A., et al. (2002). Caloric intake and the risk of Alzheimer Pfeifer, L.A., et al. (2002). Cerebral amyloid angiopathy and cognitive
disease. Archives of Neurology, 59, 1258–63. function: the HAAS autopsy study. Neurology, 58, 1629–34.
Luchsinger, J.A., et al. (2004). Hyperinsulinemia and risk of Alzheimer Pohjasvaara, T., et al. (1998). Clinical determinants of poststroke dementia.
disease. Neurology, 63, 1187–92. Stroke, 29, 75–81.
Luchsinger, J.A., et al. (2007). Measures of adiposity and dementia risk in Pohjasvaara, T., et al. (1999). Clinical and radiological determinants of
elderly persons. Archives of Neurology, 64, 392–8. prestroke cognitive decline in a stroke cohort. Journal of Neurology,
MacKnight, C., et al. (2002). Diabetes mellitus and the risk of dementia, Neurosurgery and Psychiatry, 67, 742–8.
Alzheimer’s disease and vascular cognitive impairment in the Canadian Prince, M., Cullen, M., and Mann, A. (1994). Risk factors for Alzheimer’s
Study of Health and Aging. Dementia and Geriatric Cognitive Disorders, disease and dementia: a case-control study based on the MRC elderly
14, 77–83. hypertension trial. Neurology, 44, 97–104.
Matthews, F.E., et al. (2009). Epidemiological pathology of dementia: Prince, M.J., et al. (1996). Is the cognitive function of older patients affected
attributable-risks at death in the Medical Research Council Cognitive by antihypertensive treatment? Results from 54 months of the Medical
Function and Ageing Study. PLoS Medicine, 6, e1000180 Research Council’s treatment trial of hypertension in older adults. British
Mattila, K., et al. (1988). Blood pressure and five year survival in the very old. Medical Journal, 312, 801–4.
British Medical Journal, 296, 887–9. Qiu, C., et al. (2004). Decline in blood pressure over time and risk of dementia:
Mayer-Gross, W., Slater, E., and Roth, M. (1969) Clinical psychiatry, 3rd a longitudinal study from the Kungsholmen Project. Stroke, 35, 1810–15.
edition. Baillière, Tindall and Carsell, London. Reitz, C., et al. (2004). Relation of plasma lipids to Alzheimer disease and
McGuiness, B., et al. (2009). Statins for the prevention of dementia. vascular dementia. Archives of Neurology, 61, 705–14.
Cochrane Database of Systematic Reviews, 2 , CD003160 . doi: Richards, M., et al. (1997). Is there any evidence for a protective effect of
10.1002/14651858. antithrombotic medication on cognitive function in men at risk of
Meyer, J.S., et al. (1986). Improved cognition after control of risk factors for cardiovascular disease? Some preliminary findings. Journal of Neurology,
multi-infarct dementia. Journal of the American Medical Association, Neurosurgery and Psychiatry, 62, 269–72.
256, 2203–9. Roberts, R.O., et al. (2008). Association of duration and severity of diabetes
Meyer, J.S., et al. (1989). Randomized clinical trial of daily aspirin therapy mellitus with mild cognitive impairment. Archives of Neurology, 65,
in multi-infarct dementia. Journal of the American Geriatrics Society, 37, 1066–73.
549–55. Román, G.C., et al. (1993). Vascular dementia: diagnostic criteria for
Mielke, M.M., et al. (2005). High total cholesterol levels in late life associated research studies: Report of the NINDS-AIREN International Workshop.
with a reduced risk of dementia. Neurology, 64, 1689–95. Neurology, 43, 1609–11.
Mielke, M.M., et al. (2010). The 32-year relationship between cholesterol and Rosengren, A., et al. (2005). Body mass index, other cardiovascular risk
dementia from midlife to late life. Neurology, 75, 1888–95. factors, and hospitalization for demetia. Archives of Internal Medicine,
Moroney, J.T., et al. (1996). Risk factors for incident dementia after stroke. 165, 321–6.
Role of hypoxic and ischaemic disorders. Stroke, 27, 1283–9. Rovio, S., et al. (2005). Leisure-time physical activity at midlife and the risk of
Moroney, J.T., et al. (1997). Dementia after stroke increases the risk of dementia and Alzheimer’s disease. Lancet Neurology, 4, 705–11.
long-term stroke recurrence. Neurology, 48, 1317–25. Rusanen, M., et al. (2010). Midlife smoking, apolipoprotein E and risk of
Morris, M.C., et al. (2000). The cross-sectional association between blood dementia and Alzheimer’s disease: a population-based cardiovascular
pressure and Alzheimer’s disease in a biracial community population of risk factors, aging and dementia study. Dementia and Geriatric Cognitive
older persons. Journal of Gerontology, 55A, M130–6. Disorders, 30, 277–84.
Notkola, I.-L., et al. (1998). Serum total cholesterol, apolipoprotein E 4 allele, Rusanen, M., et al. (2011). Heavy smoking in midlife and long-term risk of
and Alzheimer’s disease. Neuroepidemiology, 17, 14–20. Alzheimer disease and vascular dementia. Archives of Internal Medicine,
Nourhashemi, F., et al. (2003). Body mass index and incidence of dementia: 171, 333–9.
the PAQUID study. Neurology, 60, 117–19. Santanello, N.C., et al. (1997). Effect of pharmacologic lipid lowering on
Olichney, J.M., et al. (2000). Relationship between severe amyloid angiopathy, health-related quality of life in older persons: results from the Cholesterol
apolipoprotein E genotype, and vascular lesions in Alzheimer’s disease. Reduction in Seniors Program (CRISP) Pilot Study. Journal of the
Annals of the New York Academy of Sciences, 903, 138–48. American Geriatrics Society, 45, 8–14.
Ott, A., et al. (1997). Atrial fibrillation and dementia in a population-based Satish, S., et al. (2001). The relationship between blood pressure and mortality
study. Stroke, 28, 316–21. in the oldest old. Journal of the American Geriatrics Society, 49, 367–74.
Ott, A., et al. (1999). Diabetes mellitus and the risk of dementia. Neurology, Savva, G.M., et al. (2010). Epidemiological studies of the effect of stroke on
53, 1937–42. incident dementia: a systematic review. Stroke, 41, e41–6.
Peila, R., et al. (2002). Type 2 diabetes, APOE gene, and the risk for dementia Sharp, S.I., et al. (2009). Choline acetyltransferase activity in vascular
and related pathologies: The Honolulu-Asia Aging Study. Diabetes, 51, dementia and stroke. Dementia & Geriatric Cognitive Disorders, 28,
1256–62. 233–8.
Peila, R., et al. (2004). Fasting insulin and incident dementia in an elderly Shepherd, J., et al. (2002). Pravastatin in elderly individuals at risk of vascular
population of Japanese-American men. Neurology, 63, 228–33. disease (PROSPER): a randomized controlled trial. Lancet, 360, 1623–30.
Pendlebury, S.T. and Rothwell, P.M. (2009). Prevalence, incidence, and factors Siennicki-Lantz,A., et al. (2008). Smoking-related changes in cerebral
associated with pre-stroke and post-stroke dementia: a systematic review perfusion in a population of elderly men. Neuroepidemiology, 30, 84–92.
and meta-analysis. Lancet Neurology, 8, 1006–18. Skoog, I., et al. (1996). 15-year longitudinal study of blood pressure and
Perry, E., et al. (2005). Absence of cholinergic deficits in ‘pure’ vascular dementia. Lancet, 347, 1141–5.
dementia. Neurology, 64, 132–3. Skoog, I., et al. (1998). A population-based study on blood pressure and brain
Peters, R., et al. (2008). Smoking, dementia and cognitive decline in the atrophy in 85-year-olds. Hypertension, 32, 404–9.
elderly, a systematic review. BMC Geriatrics, 8, 36 Snowdon, D.A., et al. (1997). Brain infarction and the clinical expression of
Petitti, D.B., et al. (2005). Blood pressure levels before dementia. Archives of Alzheimer disease. Journal of the American Medical Association, 277,
Neurology, 62, 112–16. 813–17.
Petrovitch, H., et al. (2000). Midlife blood pressure and neuritic plaques, Sparks, D.L., et al. (1995). Increased incidence of neurofibrillary tangles
neurofibrillary tangles, and brain weight at death: the Honolulu Asia (NFT) in non-demented individuals with hypertension. Journal of the
Aging Study. Neurobiology of Aging, 21, 57–62. Neurological Sciences, 131, 162–9.
Sparks, D.L., et al. (1996). Increased density of senile plaques (SP), but not Tatemichi, T.K., et al. (1992b). Confusion and memory loss from capsular
neurofibrillary tangles (NFT), in non-demented individuals with the genu infarction: a thalamocortical disconnection syndrome? Neurology,
apolipoprotein E4 allele: comparison to confirmed Alzheimer’s disease 42, 1966–79.
patients. Journal of the Neurological Sciences, 138, 97–104. Tatemichi, T.K., et al. (1994). Risk of dementia after stroke in a hospitalised
Sparks, D.L., et al. (2005). Atorvastatin for the treatment of mild to moderate cohort: results of a longitudinal study. Neurology, 44, 1885–91.
Alzheimer Disease. Archives of Neurology, 62, 753–7. Tyas, S.L., et al. (2003). Mid-life smoking and late-life dementia: the
Srikanth, V.K., et al. (2004). Progressive dementia after first-ever stroke: a Honolulu-Asia Ageing Study. Neurobiology of Aging, 24, 589–96.
community-based follow-up study. Neurology, 63, 785–92. Tzourio, C., et al. (2003). Effects of blood pressure lowering with perindopril
Staessen, J.A., et al. (2000). Risks of untreated and treated isolated systolic and indapamide therapy on dementia and cognitive decline in patients
hypertension in the elderly: meta-analysis of outcome trials. Lancet, 355, with cerebrovascular disease. Archives of Internal Medicine, 163,
865–72. 1069–75.
Starr, J.M., Whalley, L.J., and Deary, I.J. (1996). The effects of antihypertensive Ueda, K., et al. (1992). Prevalence and etiology of dementia in a Japanese
treatment on cognitive function: results from the HOPE study. Journal of community. Stroke, 23, 798–803.
the American Geriatrics Society, 44, 411–15. Vitiello, B., et al. (1993). Autonomic dysfunction in patients with dementia of
Stewart, R. (1998). Cardiovascular factors in Alzheimer’s disease. Journal of the Alzheimer type. Biological Psychiatry, 34, 428–33.
Neurology, Neurosurgery and Psychiatry, 65, 143–7. Wang, L.Y., et al. (2009). Blood pressure and brain injury in older adults:
Stewart, R. (1999). Hypertension and cognitive decline. British Journal of findings from a community-based autopsy study. Journal of the American
Psychiatry, 174, 286–7. Geriatrics Society, 57, 1975–81.
Stewart, R. and Liolitsa, D. (1999). Type 2 diabetes mellitus, cognitive White, L., et al. (2005). Recent clinical-pathologic research on the causes
impairment and dementia. Diabetic Medicine, 16, 93–12. of dementia in late life: update from the Honolulu-Asia Aging Study.
Stewart, R., et al. (2005). A 32-year prospective study of change in body Journal of Geriatric Psychiatry and Neurology, 18, 224–7.
weight and incident dementia: the Honolulu-Asia Aging Study. Archives Whitmer, R.A., et al. (2008). Central obesity and increased risk of dementia
of Neurology, 62, 55–60. more than three decades later. Neurology, 71, 1057–64.
Stewart, R., et al. (2007). 26-year change in total cholesterol levels and incident Whitmer, R.A., et al. (2009). Hypoglycemic episodes and risk of dementia
dementia. The Honolulu-Asia Aging Study. Archives of Neurology, 62, in older patients with type 2 diabetes mellitus. Journal of the American
55–60. Medical Association, 301, 1565–72.
Stewart, R., et al. (2009). Change in blood pressure and incident dementia: a Yoshitake, T., et al. (1995). Incidence and risk factors of vascular dementia
32-year prospective study. Hypertension, 54, 233–40. and Alzheimer’s disease in a defined elderly Japanese population: the
Swan, G.E., Carmelli, D., and LaRue, A. (1998). Systolic blood pressure Hisayama Study. Neurology, 45, 1161–8.
tracking over 25 to 30 years and cognitive performance in older adults. Zhu, L., et al. (1998). Blood pressure reduction, cardiovascular diseases, and
Stroke, 29, 2334–40. cognitive decline in the Mini-Mental State Examination in a community
Tatemichi, T.K., et al. (1992a). Dementia after stroke: baseline frequency, risks population of very old people: a three-year follow-up. Journal of Clinical
and clinical features in a hospitalised cohort. Neurology, 42, 1185–93. Epidemiology, 51, 385–91.
CHAPTER 35
Dementia with Lewy
bodies and Parkinson’s
disease dementia
Arvid Rongve and Dag Aarsland
Dementia with Lewy bodies (DLB) and Parkinson’s disease demen- discovered in 1912 the intraneuronal eosinophilic inclusions, later
tia (PDD) belong to the α-synucleinopathies, pathologically char- named Lewy bodies, during his research on parkinsonism. In 1923
acterized by aggregation of α-synuclein in Lewy bodies in the brain Lewy described 43 patients with parkinsonism, of whom 21 were
(Table 35.1). Dementia in Parkinson’s disease (PD) typically devel- demented (Alafuzoff et al., 2009; Geldsetzer et al., 2010), but DLB
ops several years after the motor symptoms, but in a subgroup of PD was first recognized as a neurodegenerative entity in 1961 when
patients, mild cognitive impairment (MCI) has been found from Okazaki described two clinical cases with dementia, disorienta-
the very start of motor symptoms (Aarsland et al., 2009a) and some tion, hallucinations, and profound motor symptoms with rapid
of these patients develop dementia rapidly (G. Halliday et al., 2008). progression and Lewy-body pathology (Okazaki et al., 1961). The
In DLB the dementia syndrome develops simultaneously or within triad of dementia, parkinsonism, and psychosis was considered the
a year after motor parkinsonism, or dementia develops simultane- core syndrome, and later fluctuating confusion, frequent falls, neu-
ously with the other core symptoms, i.e. well-formed visual hal- roleptic sensitivity, and syncope were added. In 1998 α-synuclein
lucinations and cognitive fluctuations (McKeith et al., 2005). The (αS) was identified as the main constituent of Lewy bodies, des-
cognitive profile of the Lewy body dementias, i.e. DLB and PDD, ignating a new group of neurodegenerative disorders named the
is characterized by attentional, executive, and visuospatial impair- α-synucleinopathies (Spillantini et al., 1998).
ment, but memory impairment is also common (Collerton et al.,
2003). Persons with DLB and PDD have more sleep disturbances Nomenclature and Clinical Diagnostic
(Rongve et al., 2010a), neuropsychiatric symptoms (Aarsland et al.,
2008b), autonomic dysfunction(Allan et al., 2007), a higher ten- Criteria for DLB
dency to fall (Allan et al., 2009), faster admission to a nursing home AD pathologies such as tau-inclusions and in particular amyloid
(Rongve et al., 2010c), more impaired quality of life (Bostrom et al., plaques also occur in DLB, although usually to a lesser degree
2007a), use more resources (Bostrom et al., 2007b), and may have than in AD. Uncertainty regarding the contribution of underlying
a faster progression of cognitive decline and shorter survival com- pathology to the clinical picture resulted in different groups desig-
pared to persons with Alzheimer’s disease (AD) (Williams et al., nating different names for the same clinical phenomenon; diffuse
2006), thus underlining the clinical importance of the condition. Lewy body disease (Kosaka et al., 1984), dementia with cerebral
Some of these patients have particularly good response to treat- Lewy bodies (Eggertson and Sima, 1986), and senile dementia
ment with the cholinesterase inhibitors (ChEIs) (Burn et al., 2006a) of Lewy body type (Perry et al., 1990). The importance of Alzheimer
and some develop neuroleptic hypersensitivity syndrome (Aarsland pathology was stressed by other groups, suggesting names like AD
et al., 2005b), and thus an accurate diagnosis of DLB and PDD is with PD changes (Ditter and Mirra, 1987), AD with incidental
essential in clinical practice. Lewy bodies (Joachim et al., 1988), and Lewy body variant of AD
(Hansen et al., 1990).
In the early 1990s, both the Nottingham group (Byrne et al.,
Historical Background 1991) and the Newcastle group (McKeith et al., 1992) proposed
In 1817, James Parkinson described the clinical features of what clinical diagnostic criteria for what later became DLB. In 1995 an
he designated the shaking palsy, later renamed Parkinson’s dis- international consortium developed the first consensus criteria for
ease by Charcot. He did not, however, describe a dementia syn- a clinical diagnosis of DLB (McKeith et al., 1996) characterized by
drome in his patients, but stated that ‘the intellect and senses are dementia accompanied by the core features—fluctuating cogni-
uninjured’ (Parkinson, 1817). Fritz Heinrich Lewy (1885–1950) tion and consciousness, spontaneous features of parkinsonism, and
Table 35.1 α-Synucleinopathies made. In the absence of any core features, one or more sugges-
tive features is sufficient for possible DLB. Probable DLB should
Name Characteristic features not be diagnosed on the basis of suggestive features alone)
Dementia with Lewy bodies Dementia with parkinsonism, visual ◆ REM sleep behaviour disorder
hallucinations, fluctuations, and REM
sleep behaviour disorder (RBD) ◆ Severe neuroleptic sensitivity
Parkinson’s disease Tremor, rigidity, akinesia, and gait ◆ Low dopamine transporter uptake in basal ganglia demon-
disturbance strated by SPECT or PET imaging.
Parkinson’s disease dementia Dementia developed after more than ◆ Supportive features (commonly present but not proven to have
1 year of parkinsonian motor symptoms diagnostic specificity)
Multiple system atrophy Autonomic dysfunction, parkinsonism, ◆ Repeated falls and syncope
and ataxia
◆ Transient, unexplained loss of consciousness
Idiopathic REM sleep behaviour Acting out dream content during REM
disorder sleep ◆ Severe autonomic dysfunction, e.g. orthostatic hypotension,
urinary incontinence
Pure autonomic failure Orthostatic hypotension, constipation,
sweating, and impotence ◆ Hallucinations in other modalities
◆ Systematized delusions
◆ Depression
recurrent complex visual hallucinations. Additional supporting ◆ Relative preservation of medial temporal lobe structures on
features such as frequent falls, syncope, transient loss of conscious- computed tomography (CT)/magnetic resonance imaging
ness, systematized delusions, and severe sensitivity to treatment (MRI) scan
with antipsychotic drugs were listed. In the revised version of
the criteria (see Clinical Diagnostic Criteria for DLB), sugges- ◆ Generalized low uptake on SPECT/PET perfusion scan with
tive features like REM sleep behaviour disorder (RBD), a positive reduced occipital activity
CIT-SPECT or PET scan, and neuroleptic sensitivity were added as ◆ Abnormal (low uptake) of [123I]metaiodobenzyl guanidine
diagnostic criteria (McKeith et al., 2005). ([123I]MIBG) myocardial scintigraphy
Several preliminary reports suggested that the original consen- ◆ A diagnosis of DLB is less likely
sus criteria had high specificity but low sensitivity (McKeith et al.,
2000b), but the most systematically designed study with prospective ◆ in the presence of cerebrovascular disease evident as focal neu-
validation against a pathological diagnosis found 83% sensitivity and rological signs or on brain imaging
95% specificity (McKeith et al., 2000b). The criteria have been found ◆ in the presence of any other physical illness or brain disorder
to be more sensitive if RBD was included as a core feature (88%) or sufficient to count in part or in total for the clinical picture
if probable DLB could be diagnosed with only RBD plus dementia
◆ if parkinsonism only appears for the first time at a stage of
(90%). The specificity in this study was 73% (Ferman et al., 2011).
severe dementia
The following are clinical diagnostic criteria for DLB:
◆ Temporal sequence of symptoms
◆ Central features (essential for a diagnosis of possible or
probable DLB) ◆ DLB should be diagnosed when dementia occurs before or
concurrently with parkinsonism (see McKeith et al., (2005) for
◆ Dementia defined as progressive cognitive decline of sufficient
full text).
magnitude to interfere with normal social or occupational
function.
◆ Prominent or persistent memory impairment may not nec- Clinical Diagnostic Criteria for PDD
essarily occur in the early stages but is usually evident with A Movement Disorders Society Task Force defined clinical crite-
progression. ria for probable and possible PDD (Emre et al., 2007), and later
◆ Deficits on tests of attention, executive function, and visuospa- operationalization of these criteria for practical administration
tial ability may be especially prominent. in the clinic with cognitive tests and suggested cut-off values has
been published (Dubois et al., 2007; Goetz et al., 2008). The criteria
◆ Core features (two core features are sufficient for a diagnosis of
define core and associated clinical features: The two core features
probable DLB, one for possible DLB)
are (1) a diagnosis of PD according to Queen Square Brain Bank
◆ Fluctuating cognition with pronounced variations in attention Criteria (Hughes et al., 1993), and (2) a dementia syndrome with
and alertness. insidious onset and slow progression with cognitive impairment
◆ Recurrent visual hallucinations that are typically well formed in more than one domain representing a decline from previous
and detailed. level and impairment in activities of daily living (ADL) independ-
ent of motor or autonomic symptoms. Associated clinical features
◆ Spontaneous features of parkinsonism. are (1) cognitive features—impaired attention, executive functions,
◆ Suggestive features (if one or more of these is present in the visuospatial functions, memory, or language; and (2) behavioural
presence of one core feature, a diagnosis of probable DLB can be features—apathy, changes in personality, mood-like depression or
CHAPTER 35 dementia with lewy bodies and parkinson’s disease dementia 471
anxiety, hallucinations, delusions, and excessive daytime sleepi- pathological verification of the diagnosis (Matsui et al., 2009). A
ness. Practical procedures for cognitive testing at different levels community-based survey from Japan, the Hisayama Study, found
of diagnostic accuracy are defined, including simple bedside tests DLB in 10.6% of neuropathologically confirmed cases and pure
such as a score below 26 on the Mini-Mental State Examination DLB neuropathologically in 4.4% (Matsui et al., 2009). The Islington
(MMSE) as a global cognitive measure, reversed months test for Community Study of Dementia from North London found 9.7%
attention, lexical fluency, and clock drawing for executive func- with a clinical diagnosis of probable DLB and 30.5% with possible
tion, MMSE pentagons for visuospatial functions, and three-word or probable DLB combined (Stevens et al., 2002). A health survey
recall from the MMSE as a test for memory. The authors suggest in the Kupio area in Finland of people 75 years and older found a
the Neuropsychiatric Inventory (NPI) to detect behavioural fea- dementia prevalence of 22% and the proportion of DLB was 21.9%
tures and the clinical interview with a caregiver to diagnose exces- (Rahkonen et al., 2003).
sive daytime sleepiness and explore ADL abilities. The MMSE is These three studies provide the best estimates of the prevalence
suboptimal for DLB and PDD due to little focus on executive func- of DLB in the general population and suggest that DLB accounts
tions, and scales such as the Montreal Cognitive Assessment and for 10–22% of the dementias in the 65+ age group, indicating
PD-Cognitive Rating Scale have been validated and recommended that about 1% of the population over 65 years suffer from DLB.
(Kulisevsky and Pagonabarraga, 2009). In a study from western Norway applying the revised consensus
criteria on a referral cohort to old age psychiatry and geriatric
medicine clinics, 15.8% of persons with mild dementia were
Mild Cognitive Impairment (MCI) diagnosed with probable DLB (Aarsland et al., 2008b). Only four
During the last 5 years, researchers have focused on MCI in PD, i.e. incidence studies exist and report the incidence of DLB from
cognitive impairment without significant functional consequences. 0.7–1.4 new cases per 1000 persons per year (de Lau et al., 2004;
The exact time when dementia should be diagnosed can be diffi- Matsui et al., 2009).
cult to determine in PD due to the fact that ADL difficulties can be
caused by motor, sleep, and autonomic problems, in addition to the PDD
cognitive failure. Clinical diagnostic criteria for MCI in PD have Although considered to be relatively rare (Brown and Marsden,
been developed (Litvan et al., 2012). Typically, attentional and exec- 1984), several more recent studies have reported that dementia is
utive deficits occur early, but visuospatial and memory impairment common in PD. A systematic review found that in cross-sectional
are also common, and there is interindividual heterogeneity in the studies, more than 30% of PD patients have dementia (Aarsland
cognitive profile (Barone et al., 2011; Litvan et al., 2011). The clini- et al., 2005a). Subsequent longitudinal studies reported a three to
cal significance of PD-MCI is highlighted by findings that PD-MCI six times higher incidence of dementia in PD compared to nonPD
is associated with shorter time to PDD (Janvin et al., 2005). There is subjects (Aarsland et al., 2003), and in two long-term studies it was
some evidence that patients with different cognitive profiles differ shown that up to 80% of PD patients develop dementia (Buter et al.,
in dementia risk, i.e. that those with attention and executive defi- 2008; Hely et al., 2008). In the CamPaign study, based on an inci-
cits, usually due to frontosubcortical dopaminergic lesions, may dence PD cohort, the incidence of dementia was somewhat lower,
have a lower dementia risk than those with visuoconstructive and with less than 50% of the cohort having developed dementia 8 years
semantic memory impairment, which is more associated with pos- after diagnosis. The mean time to dementia in this study was 6.2
terior and possibly nondopaminergic lesions (Williams-Gray et al., years, with old age, postural and gait disturbance, and early cogni-
2007). tive impairment predicting shorter time to dementia (Evans et al.,
Little is known regarding the predementia stages of DLB. Some 2011). Based on the clinical course and neuropathology, different
patients with MCI, particularly those with nonamnestic MCI, subtypes of PD have been identified with different risk of dementia
develop clinical DLB (Molano et al., 2010). Other predementia DLB during the different stages of the disease (Halliday and McCann,
syndromes include idiopathic RBD (Claassen et al., 2010), primary 2010).
autonomic failure (Larner et al., 2000), and somatoform disorder
(Onofrj et al., 2010). In a retrospective study, impaired memory PD-MCI
was reported to be the most common presenting symptom in DLB, In addition to those with PDD, 20–25% of nondemented PD
followed by hallucinations and depression. Problem-solving diffi- patients have mild cognitive impairment (PD-MCI) (Aarsland
culties and symptoms of parkinsonism were other common pre- et al., 2010) and 15–20% even at time of diagnosis in de-novo PD
senting symptoms (Auning et al., 2011). (Aarsland et al., 2009a; Elgh et al., 2009).
Epidemiology Molecular pathology

DLB The Lewy body diseases share aggregation of α-synuclein and forma-
There are no well-designed systematic studies to inform about tion of Lewy bodies and Lewy neurites as their common hallmarks
the prevalence and incidence of DLB. In clinical dementia cohort of pathology. The normal structure and function of α-synuclein is
studies and population-based epidemiological studies applying the not yet well known, but it is believed to be involved in synaptic plas-
original diagnostic criteria from 1996, the reported proportion ticity. Increased expression of αS will inhibit synaptic reclustering
with DLB ranges from 0% to 30.5% (Zaccai et al., 2005; Rongve after neurotransmitter release and thus inhibit neurotransmitter
et al., 2006; Aarsland et al., 2008b). Epidemiological dementia release in the synaptic cleft (Nemani et al., 2010). Pathological phos-
studies have usually not included standardized assessments of phorylation and aggregation into toxic oligomers and pathological
core and suggestive DLB features, and only two studies present spread of oligomers from one neuron to adjacent neurons has been
suggested as a possible mechanism for spreading the α-synuclein DLB and AD (Aarsland et al., 2008a). CIT-SPECT, the visualization
pathology within the CNS (Danzer et al., 2009). α-Synuclein pro- of the striatal dopamine transporter, a measure of the dopaminergic
duced intracellularly can be excreted in the extracellular space in presynaptic nigrostriatal system, is the first established biomarker
a calcium-dependent way, leading to decreased cell viability and in the Lewy body dementias (McKeith et al., 2007). A meta-analysis
increased Lewy-related pathology (Emmanouilidou et al., 2010). found a sensitivity of 86.5% and specificity of 93.6% (Papathanasiou
et al., 2012). A pathological scan could identify those patients with
Neuropathology possible DLB who progressed to probable DLB after 12 months
(O’Brien et al., 2009). Myocardial scintigraphy (MIBG) has been
Different pathological staging systems have been proposed for DLB, shown to reliably identify DLB even in mild cases (Suzuki et al.,
PDD, and PD. Kosaka in 1980 proposed to differentiate three sub- 2006) and is included as a supportive feature in the diagnostic cri-
types of DLB pathologically: brainstem, transitional, and cortical teria for DLB. Both CIT-SPECT and myocardial scintigraphy can
(Kosaka et al., 1984). The revised pathological consensus criteria reliably differentiate PD from other movement disorders (Hauser
(McKeith et al., 2005) now implement severity and distribution of and Grosset, 2012; King et al., 2011).
both Alzheimer’s and Lewy body pathology in the CNS in these Earlier studies found reduced perfusion in occipital cortical areas
three locations and define the likelihood that the pathological find- in DLB as compared to AD on perfusion SPECT images, but this
ings are correlated with a clinical DLB syndrome as low, intermedi- finding is not useful clinically due to low sensitivity and specifi-
ate, or high. The revised pathological criteria performed reasonably city (Lobotesis et al., 2001), although, more recently, sensitivity and
well in a validation study (Fujishiro et al., 2008). specificity of 73% and 72% were found, respectively (Colloby et al.,
Braak proposed criteria for staging the Lewy body pathology in 2010). Studies using glucose PET have found reduced activities in
PD and others have proposed models for DLB (Muller et al., 2005; neocortex associated with cognitive impairment even early in PD
Leverenz et al., 2008). Three distinctive groups have been described (Pappata et al., 2011).
neuropathologically in dopa-responsive PD patients recruited and Structural imaging using MRI usually shows preservation of the
followed until death; one group with younger onset PD and longer medial temporal lobes in DLB as compared to AD. Most studies of
duration of disease with neuropathology distribution correspond- DLB and AD have reported low accuracy, but high accuracy was
ing to the Braak stages; a second group characterized by an early shown in a prospective study with pathological verification of the
malignant dementia dominant syndrome and severe neocorti- clinical diagnosis (Burton et al., 2009). MRI has also demonstrated
cal disease as described in DLB; and a third group consisting of cortical atrophy in PDD (Burton et al., 2005; Beyer and Aarsland,
persons who had older onset of PD, shorter survival, and a more 2008) and PD-MCI (Apostolova et al., 2011) compared to cogni-
complex disease with additional pathologies and higher Lewy body tively intact PD. In a recent study, atrophy of hippocampus and
loads in the brain (G. Halliday et al., 2008). parietal-temporal cortex predicted future cognitive decline in PD
(Weintraub et al., 2012).
Genetics Promising findings have been reported also using quantitative
Mutations in SNCA, MAPT, LRRK2, and HLA have been found EEG, with relatively more slowing in DLB and PDD than in AD
to increase the risk for PD, and together with new genetic find- (Bonanni et al., 2008).
ings explain about half the genetic risk to develop PD (Hardy, In AD, the concentration pattern of cerebrospinal fluid (CSF)
2010; Nalls et al., 2011). The genetic underpinnings of dementia proteins like beta-amyloid species such as aβ42, total tau, and p-tau
in PD have been much less studied. Increased dementia risk in have been found to distinguish between AD and normal controls
family members of people with PD has been reported (Kurz et al., (Mattsson et al., 2009). However, a change in the same direction has
2006), associations between dementia and mutations in MAPT been shown in DLB, and thus the specificity against DLB is not high
(Seto-Salvia et al., 2011) and PSEN2 (Meeus et al., 2012) have been and CSF cannot presently be used to distinguish between AD and
reported, and conflicting results regarding APOE have been found DLB (Mollenhauer et al., 2006; Skogseth et al., 2011).
(Williams-Gray et al., 2009). Previously known mutations in AD The concentration of αS, the key protein in DLB and PD, can
and PD have been found in both PDD and DLB and suggest genetic be measured in CSF and plasma. Although lower concentrations
overlap in these conditions (Meeus et al., 2012). have been reported in CSF in Lewy-body disorders, this can-
Most DLB cases occur sporadically, although families have been not convincingly differentiate DLB from AD or normal controls
described having many affected members with gene alterations in (Noguchi-Shinohara et al., 2009; Spies et al., 2009; Mollenhauer
different locations, some of which overlap with PD (Nervi et al., et al., 2011).
2011). Traditional genetic studies have identified multiplications Low CSF concentrations of aβ42 in PD have been found to cor-
(Singleton et al., 2003; Chartier-Harlin et al., 2004) and mutations relate with poor cognition in cross-sectional studies, particularly
in the gene encoding α-synuclein (SNCA) (Polymeropoulos et al., regarding memory (Alves et al., 2010). Recently, this finding was
1997; Zarranz et al., 2004; Yamaguchi et al., 2005) and β-synuclein confirmed in a longitudinal study showing that low CSF levels of
(SNCB) (Ohtake et al., 2004). Genome-wide association studies aβ42 predicted significant cognitive decline during the next 2 years
have not yet been presented in DLB and PDD, and more studies are (Siderowf et al., 2010). In contrast, amyloid imaging usually finds
needed to reveal the genetic causes of the Lewy body dementias. low amyloid load in PDD compared to AD and DLB (Jokinen et al.,
2010).
Biomarkers
A variety of imaging, electrophysiological, blood, and CSF biomar- The Clinical Profile of DLB and PDD
kers have been studied as potential diagnostic markers between See Table 35.2 for a list of clinical symptoms.
Table 35.2 Clinical symptoms in DLB and PDD and executive dysfunction, and cortical cognitive impairment in
memory and language and a mixed group are described (Janvin
Cognitive impairment Attentional
et al., 2003, 2006). However, in PDD most cognitive domains are
Visuospatial impaired (Goetz et al., 2008).
Executive
Memory Cognitive fluctuations
Language Fluctuations in DLB commonly occur in both cognition and level
Cognition Fluctuating of arousal and sleepiness, with some patients changing rapidly from
Neuropsychiatric and Depression normal alertness and cognition within seconds or minutes, yet oth-
behavioural symptoms ers have much slower shifts of days or weeks. Cognitive fluctuation
Apathy
is one of the core features of DLB and occurs in 13–85% of cases.
Anxiety
However, inter-rater reliability is low, and standardized instru-
Hallucinations ments should be applied (Walker et al., 2000; Ferman et al., 2004).
Delusions The Mayo Fluctuations Composite Scale consists of four questions
Agitation posed by the clinician to a caregiver regarding daytime drowsiness
Somatoform disorder and lethargy, daytime sleep of 2 or more hours, staring into space
Obsessive compulsive disorder for long periods, and episodes of disorganized speech. The scale has
Sleep disturbances REM sleep behaviour disorder been validated in cognitively normal older people, AD, and DLB,
and fluctuations are present if three or four of the answers are ‘yes’.
Excessive daytime sleepiness
Insomnia
Neuropsychiatric symptom profile
Restless legs syndrome
In addition to the visual hallucinations, delusions, and misidenti-
Nocturnal leg cramps
fication syndromes, apathy, depression, anxiety, and auditory hal-
Obstructive sleep apnoea
lucinations are common in DLB (Ricci et al., 2009). In PD, impulse
Sleep walking
control disorders like pathological gambling, hypersexuality, and
Periodic leg movements during sleep compulsive buying have been described and found to be associated
Sleep attacks with dopamine-agonist treatment (Weintraub et al., 2010), but these
Unintended sleep episodes symptoms have rarely been explored in DLB. Somatoform disorder,
Motor Symmetrical or asymmetrical parkinsonian defined as medically unexplained symptoms, was found to occur
symptoms in 7% of patients with PD and 12% of DLB patients, preceding the
Postural instability DLB diagnosis for 6 months to 10 years in all cases (Onofrj et al.,
Gait disorder 2010). Personality traits like diminished emotional response may
distinguish DLB from AD (Galvin et al., 2007). Neuropsychiatric
Autonomic Orthostatic hypotension and behavioural symptoms are also very common in PDD, and
Syncope were found in 89% in one study; most frequently reported were
Incontinence depression (58%), apathy (54%), anxiety (49%), and hallucinations
Constipation (44%) (Aarsland et al., 2007).
Impotence
Sleep disturbances in DLB and PDD
In DLB, RBD has been included as a suggestive feature in the clini-
cal diagnostic criteria (McKeith et al., 2005). Up to 80% of patients
Cognitive profile with mild DLB or PDD have sleep disturbance, as compared to 56%
Although caregivers report memory problems to be the most fre- of age-matched cognitively normal controls and 64% of patients
quent initial symptom (Auning et al., 2011), in DLB most studies diagnosed with mild AD. The most common symptoms are insom-
describe an initial impairment in visuospatial and executive cogni- nia (47%), sleep-related leg cramps (42%), excessive daytime sleepi-
tive domains. In a review, Collerton et al. (2003) concluded that ness (41%), RBD (39%), restless legs syndrome (31%), obstructive
DLB is a visual-perceptual and attentional-executive dementia. sleep apnoea (26%), and periodic leg movements during sleep
In a cluster analysis to group all types of mild dementia subtypes (21%) (Rongve et al., 2010a).
according to the level of parkinsonism, hallucinations, fluctuations, In PD, a wide range of different sleep disturbances have been
and RBD, a group of patients was identified with high scores for described, including insomnia, excessive daytime sleepiness,
parkinsonism, hallucinations, and fluctuations and consisted of sleep attacks or unintended sleep episodes, RBD, restless legs syn-
exclusively patients diagnosed as having clinically DLB or PDD. drome, sleep-related leg cramps, periodic leg movements during
This group scored significantly lower on tests for visuospatial cog- sleep, obstructive sleep apnoea, and sleep walking/somnambulism
nitive skills as compared to the other three groups (Rongve et al., (Jauregui-Barrutia et al., 2010; Mondragon-Rezola et al., 2010).
2010b). Early visuospatial cognitive impairment has been found Sleep disturbances in dementia have been thought to relate to
to predict visual hallucinations in DLB (Hamilton et al., 2012). In pathology in specific brain areas such as the suprachiasmatic
PDD, different types of cognitive profiles in MCI and dementia have nucleus, hypothalamus, brainstem, and pons area. Neurochemical
been described; in MCI, both a subcortical profile with attentional changes including melatonin and acetylcholine and tend to differ in
different types of dementia related to the specific brain pathology of as the frontotemporal dementias. Corticobasal degeneration has
the type of dementia involved (Claassen et al., 2010). tau pathology and is characterized by more severe motor impair-
ment with apraxia, agnosia, parkinsonism, aphasia, and alien hand
Motor symptoms in DLB syndrome, with poor response to L-dopa. Multiple system atrophy,
The classical Parkinson symptoms such as resting tremor, brady- an α-synucleinopathy, is characterized clinically by the early devel-
kinesia, rigidity, and postural changes also occur in DLB, but are opment of ataxia, autonomic failure, and symmetrical parkinson-
usually bilateral, and postural instability, gait disorder, and rigidity ism with rigidity and bradykinesia without tremor. Patients have
usually dominate (Burn et al., 2006b). Neuroleptic medication can poor response to L-dopa treatment and poor prognosis. MRI can
cause or severely worsen parkinsonism and other motor disability in many cases inform the clinical diagnosis. In frail older people,
in dementia, particularly in DLB and PDD (Aarsland et al., 2005b). symptoms like parkinsonism and impaired cognition can be caused
Of note, in older people, musculoskeletal problems are common by neuroleptic medication, and thus a drug history must be taken
and lead to motor problems as well. during the diagnostic process. A CIT-SPECT can be used to dif-
ferentiate brainstem parkinsonism from other motor impairments
Autonomic failure in DLB and PDD such as essential tremor and vascular parkinsonism (Vlaar et al.,
Autonomic failure can be an early and prominent feature of DLB and 2008; Contrafatto et al., 2012) and other disorders such as PSP and
PDD as a consequence of involvement of the ganglia and peripheral MSA (Goebel et al., 2011).
nervous system, which can be visualized with heart scintigraphy.
Symptoms of autonomic failure are more common in DLB than Pharmacological Treatment of PDD and DLB
in AD and include orthostatic hypotension, cardiac arrhythmias,
syncope, constipation, impotence, urinary retention, and excessive Few randomized controlled treatment trials (RCTs) exist for PDD
sweating (Allan et al., 2007; Sonnesyn et al., 2009). and DLB, but rivastigmine was found to improve cognition and
neuropsychiatric symptoms in DLB (McKeith et al., 2000a, 2004).
In PDD, rivastigmine improved cognition, psychiatric symptoms,
Clinical Differential Diagnosis and function in a multicentre RCT which paved the way for inter-
DLB and AD can be difficult to reliably differentiate clinically. The national approval (Emre et al., 2004), and such treatment is recom-
clinical diagnosis of dementia is based on an interview with both mended in guidelines (O’Brien and Burns, 2011). Those with visual
the person with dementia and the caregiver, and an active approach hallucinations had the strongest response (Burn et al., 2006a).
to the core and suggestive clinical features is important to detect Recent studies have suggested that memantine is a safe and a
DLB. In addition to a clinical examination including physical and potentially effective treatment for DLB and PDD, although findings
neurological examination and cognitive testing, supplemental have been inconsistent, with significant improvement on a global
blood tests, MRI, and CSF analysis to exclude other intracranial measure and in neuropsychiatric symptoms in DLB only (Emre et
pathology is indicated. Although MRI and CSF can support a diag- al., 2010), but both sleep (Aarsland et al., 2009b) and quality of life
nosis of AD, these techniques cannot reliably distinguish between (Larsson et al., 2011) were shown to improve in another study.
AD and DLB. Of note, both β-CIT dopamine transporter SPECT L-dopa is the main treatment of motor symptoms in PD. L-dopa
and cardiac scintigraphy (MIBG) have been shown to have high has been shown to improve parkinsonian symptoms in DLB in
sensitivity and specificity to differentiate between probable DLB open-label studies, but side effects such as hallucinations and
and nonDLB dementia (see section Biomarkers). A cognitive orthostatic hypotension can occur (Molloy et al., 2005). L-dopa has
profile with executive and visuospatial impairment and preserved a complex and state-dependent effect on cognition in PD, and can
memory and language can aid in the differential diagnosis between improve some symptoms (MacDonald et al., 2011).
DLB and AD, although different cognitive profiles can be found in Clozapine is the only drug with convincing effect on psychotic
both conditions. symptoms in PD, including some with cognitive impairment
As per the consensus criteria, DLB is differentiated from PDD (Rongve et al., 2012). Although some open-label reports suggest
based on the 1-year rule; in DLB, parkinsonian motor symptoms that atypical antipsychotics such as quetiapine and clozapine may
can start up to 1 year before the dementia syndrome. If motor par- be useful (Poewe, 2005), the two RCTs in DLB reported either no
kinsonian symptoms started more than 1 year before dementia, the effect of quetiapine on agitation and psychosis (Kurlan et al., 2007)
condition is diagnosed as PDD. Whether this is a biologically valid or worsening of psychiatric symptoms from risperidone and cita-
distinction, or whether PDD and DLB are merely syndromes on a lopram (Culo et al., 2010). Antipsychotic medications have been
continuum of Lewy body disease, is not known. Currently, unlike found to significantly increase the risk of cerebral haemorrhages
the distinction between DLB and AD, there are no major prognos- and infarctions, pneumonia, and death in older people with demen-
tic or therapeutic consequences related to the distinction between tia (Ballard et al., 2009), and in DLB and PDD neuroleptic hyper-
PDD and DLB. sensitivity is more common than in AD (Aarsland et al., 2005b).
Vascular dementia and frontotemporal dementia can in most We therefore advise to administer antipsychotic medication for this
cases be differentiated from DLB based on the clinical interview and group of patients only if psychotic symptoms are severe and persist-
examination and supplemental tests like MRI and perfusion SPECT. ent, and preferably in hospital with close monitoring.
MRI can assist in identifying vascular parkinsonism. Progressive In PD, only nortriptyline and pramipexole have demonstrated
supranuclear palsy (PSP) is a tauopathy and characterized by axial antidepressant effect compared to placebo, but unfortunately no
parkinsonism, early tendency to fall backwards, and impaired RCTs for treating depression have been carried out in DLB or PDD
vertical eye movements. PSP can be complicated by subcortical (Seppi et al., 2011). Autonomic failure can be treated symptomati-
dementia and must be differentiated from other tauopathies such cally, i.e. orthostatic hypotension can be improved by reducing or
omitting drugs with orthostatic hypotension as a known side effect, Allan, L. M., et al. (2009). Incidence and prediction of falls in dementia: a
anti-hypertensive drugs and secure intake of sufficient amounts of prospective study in older people. PLoS One, 4 (5), e5521.
liquids. Standard laxatives can be used to treat constipation; anti- Alves, G., et al. (2010). CSF amyloid-beta and tau proteins, and cognitive
performance, in early and untreated Parkinson’s disease: the Norwegian
cholinergic agents are used to treat bladder dysfunction but should
ParkWest study. Journal of Neurology, Neurosurgery and Psychiatry, 81
be used with caution, particularly in older patients who have cog- (10), 1080–6.
nitive decline and dementia. Phosphodiesterase inhibitors can be Apostolova, L., et al. (2011). Hippocampal and ventricular changes in
used to treat sexual dysfunction (Zesiewicz et al., 2010). RBD can Parkinson’s disease mild cognitive impairment. Neurobiology of Aging,
be treated using clonazepam, melatonin, ChEI, or pramipexole 33 (9), 2113–24.
(Aurora et al., 2010). Pharmacological treatment of DLB and PDD Auning, E., et al. (2011). Early and Presenting Symptoms of Dementia
is, however, particularly challenging, as improving one aspect of the with Lewy Bodies. Dementia and Geriatric Cognitive Disorders, 32 (3),
disease might worsen other aspects; for example, treating the motor 202–8.
symptoms with dopaminergic medication in some persons releases Aurora, R. N., et al. (2010). Best practice guide for the treatment of REM
sleep behavior disorder (RBD). Journal of Clinical Sleep Medicine, 6 (1),
or increases psychosis and delirium. 85–95.
Ballard, C., et al. (2009). The dementia antipsychotic withdrawal trial
Nonpharmacological Treatment (DART-AD): long-term follow-up of a randomised placebo-controlled
trial. Lancet Neurology, 8 (2), 151–7.
of DLB and PDD Barone, P., et al. (2011). Cognitive impairment in nondemented Parkinson’s
No systematic studies exist regarding the nonpharmacological disease. Movement Disorders, 26 (14), 2483–95.
treatment of DLB or PDD, but general principles regarding non- Beyer, M. K. and Aarsland, D. (2008). Grey matter atrophy in early versus
pharmacological interventions in dementia apply. In PD, some late dementia in Parkinson’s disease. Parkinsonism Related Disorders, 14
(8), 620–5.
improvement in cognitive tests has been reported after cognitive
Bonanni, L., et al. (2008). EEG comparisons in early Alzheimer’s disease,
training in a randomized trial (Paris et al., 2011). It is important to dementia with Lewy bodies and Parkinson’s disease with dementia
inform patients and caregivers regarding the characteristic features patients with a 2-year follow-up. Brain, 131 (Pt 3), 690–705.
of DLB, including the potentially harmful effects of RBD and neu- Bostrom, F., et al. (2007a). Patients with dementia with lewy bodies have more
roleptics. Most authors recommend identifying one or only a few impaired quality of life than patients with Alzheimer disease. Alzheimer’s
target symptoms to deal with first. Online support groups for carers Disease and Associated Disorders, 21 (2), 150–4.
and information for carers, patients, and professionals are available Bostrom, F., et al. (2007b). Patients with Lewy body dementia use more
from the internet; see <http://lbda.org/>, <http://www.lewybody. resources than those with Alzheimer’s disease. International Journal of
Geriatric Psychiatry, 22 (8), 713–9.
org/>, and <http://www.parkinsons.org.uk/>.
Brown, R. G. and Marsden, C. D. (1984). How common is dementia in
Parkinson’s disease? Lancet, 2 (8414), 1262–5.
References Burn, D., et al. (2006a). Effects of rivastigmine in patients with and without
Aarsland, D., et al. (2003). Prevalence and characteristics of dementia in visual hallucinations in dementia associated with Parkinson’s disease.
Parkinson disease: an 8-year prospective study. Archives of Neurology, Movement Disorders, 21 (11), 1899–907.
60 (3), 387–92. Burn, D. J., et al. (2006b). Motor subtype and cognitive decline in Parkinson’s
Aarsland, D., Zaccai, J., and Brayne, C. (2005a). A systematic review of disease, Parkinson’s disease with dementia, and dementia with Lewy
prevalence studies of dementia in Parkinson’s disease. Movement bodies. Journal of Neurology, Neurosurgery and Psychiatry, 77 (5), 585–9.
Disorders, 20 (10), 1255–63. Burton, E. J., et al. (2005). Brain atrophy rates in Parkinson’s disease with and
Aarsland, D., et al. (2005b). Neuroleptic sensitivity in Parkinson’s disease and without dementia using serial magnetic resonance imaging. Movement
parkinsonian dementias. Journal of Clinical Psychiatry, 66 (5), 633–7. Disorders, 20 (12), 1571–6.
Aarsland, D., et al. (2007). Neuropsychiatric symptoms in patients with Burton, E. J., et al. (2009). Medial temporal lobe atrophy on MRI differentiates
Parkinson’s disease and dementia: frequency, profile and associated care Alzheimer’s disease from dementia with Lewy bodies and vascular
giver stress. Journal of Neurology, Neurosurgery and Psychiatry, 78 (1), cognitive impairment: a prospective study with pathological verification
36–42. of diagnosis. Brain, 132 (Pt 1), 195–203.
Aarsland, D., et al. (2008a). Early discriminatory diagnosis of dementia Buter, T. C., et al. (2008). Dementia and survival in Parkinson disease: a
with Lewy bodies. The emerging role of CSF and imaging biomarkers. 12-year population study. Neurology, 70 (13), 1017–22.
Dementia and Geriatric Cognitive Disorders, 25 (3), 195–205. Byrne, E. J., et al. (1991). Diagnostic criteria for dementia associated with
Aarsland, D., et al. (2008b). Frequency and case identification of dementia cortical Lewy bodies. Dementia, 2, 283–4.
with Lewy bodies using the revised consensus criteria. Dementia and Chartier-Harlin, M. C., et al. (2004). Alpha-synuclein locus duplication as a
Geriatric Cognitive Disorders, 26 (5), 445–52. cause of familial Parkinson’s disease. Lancet, 364 (9440), 1167–9.
Aarsland, D., et al. (2009a). Cognitive impairment in incident, untreated Claassen, D. O., et al. (2010). REM sleep behavior disorder preceding other
Parkinson disease: the Norwegian ParkWest study. Neurology, 72 (13), aspects of synucleinopathies by up to half a century. Neurology, 75 (6),
1121–6. 494–9.
Aarsland, D., et al. (2009b). Memantine in patients with Parkinson’s Collerton, D., et al. (2003). Systematic review and meta-analysis show that
disease dementia or dementia with Lewy bodies: a double-blind, dementia with Lewy bodies is a visual-perceptual and attentional-executive
placebo-controlled, multicentre trial. Lancet Neurology, 8 (7), 613–18. dementia. Dementia and Geriatric Cognitive Disorders, 16 (4), 229–37.
Aarsland, D., et al. (2010). Mild cognitive impairment in Parkinson disease: a Colloby, S. J., et al. (2010). Covariance 99mTc-exametazime SPECT patterns
multicenter pooled analysis. Neurology, 75 (12), 1062–9. in Alzheimer’s disease and dementia with Lewy bodies: utility in
Alafuzoff, I., et al. (2009). Staging/typing of Lewy body related differential diagnosis. Journal of Geriatric Psychiatry and Neurology, 23
alpha-synuclein pathology: a study of the BrainNet Europe Consortium. (1), 54–62.
Acta Neuropathologica, 117 (6), 635–52. Contrafatto, D., et al. (2012). [(123) I]FP-CIT-SPECT asymmetry index
Allan, L. M., et al. (2007). Autonomic dysfunction in dementia. Journal of to differentiate Parkinson’s disease from vascular parkinsonism. Acta
Neurology, Neurosurgery and Psychiatry, 78 (7), 671–7. Neurologica Scandinavica, 126 (1), 12–16.
Culo, S., et al. (2010). Treating neuropsychiatric symptoms in dementia with Hardy, J. (2010). Genetic analysis of pathways to Parkinson disease. Neuron,
Lewy bodies: a randomized controlled-trial. Alzheimer’s Disease and 68 (2), 201–6.
Associated Disorders, 9 July (Epub). Hauser, R. A. and Grosset, D. G. (2012). [(123) I]FP-CIT (DaTscan) SPECT
Danzer, K. M., et al. (2009). Seeding induced by alpha-synuclein oligomers brain imaging in patients with suspected parkinsonian syndromes.
provides evidence for spreading of alpha-synuclein pathology. Journal of Journal of Neuroimaging, 22 (3), 225–30.
Neurochemistry, 111 (1), 192–203. Hely, M. A., et al. (2008). The Sydney multicenter study of Parkinson’s
de Lau, L. M., et al. (2004). Incidence of parkinsonism and Parkinson disease: the inevitability of dementia at 20 years. Movement Disorders,
disease in a general population: the Rotterdam Study. Neurology, 63 (7), 23 (6), 837–44.
1240–4. Hughes, A. J., et al. (1993). A clinicopathologic study of 100 cases of
Ditter, S. M. and Mirra, S. S. (1987). Neuropathologic and clinical features Parkinson’s disease. Archives of Neurology, 50 (2), 140–8.
of Parkinson’s disease in Alzheimer’s disease patients. Neurology, 37 (5), Janvin, C., et al. (2003). Neuropsychological profile of patients with
754–60. Parkinson’s disease without dementia. Dementia and Geriatric Cognitive
Dubois, B., et al. (2007). Diagnostic procedures for Parkinson’s disease Disorders, 15 (3), 126–31.
dementia: recommendations from the movement disorder society task Janvin, C. C., Aarsland, D., and Larsen, J. P. (2005). Cognitive predictors of
force. Movement Disorders, 22 (16), 2314–24. dementia in Parkinson’s disease: a community-based, 4-year longitudinal
Eggertson, D. E. and Sima, A. A. (1986). Dementia with cerebral Lewy study. Journal of Geriatric Psychiatry and Neurology, 18 (3), 149–54.
bodies. A mesocortical dopaminergic defect?. Archives of Neurology, 43 Janvin, C., et al. (2006). Cognitive profiles of individual patients with
(5), 524–7. Parkinson’s disease and dementia: comparison with dementia with Lewy
Elgh, E., et al. (2009). Cognitive function in early Parkinson’s disease: bodies and Alzheimer’s disease. Movement Disorders, 21 (3), 337–42.
a population-based study. European Journal of Neurology, 16 (12), Jauregui-Barrutia, A., et al. (2010). Sleep disorders in Parkinson’s disease:
1278–84. REM sleep behaviour disorder and restless legs syndrome. Reviews of
Emmanouilidou, E., et al. (2010). Cell-produced alpha-synuclein is secreted Neurology, 50 Suppl 2, S15–19.
in a calcium-dependent manner by exosomes and impacts neuronal Joachim, C. L., Morris, J. H., and Selkoe, D. J. (1988). Clinically diagnosed
survival. Journal of Neuroscience, 30 (20), 6838–51. Alzheimer’s disease: autopsy results in 150 cases. Annals of Neurology,
Emre, M., et al. (2004). Rivastigmine for dementia associated with Parkinson’s 24 (1), 50–6.
disease. New England Journal of Medicine, 351 (24), 2509–18. Jokinen, P., et al. (2010). [(11)C]PIB-, [(18)F]FDG-PET and MRI imaging
Emre, M., et al. (2007). Clinical diagnostic criteria for dementia associated in patients with Parkinson’s disease with and without dementia.
with Parkinson’s disease. Movement Disorders, 22 (12), 1689–707; quiz Parkinsonism Related Disorders, 16 (10), 666–70.
837. King, A. E., Mintz, J., and Royall, D. R. (2011). Meta-analysis of 123I-MIBG
Emre, M., et al. (2010). Memantine for patients with Parkinson’s disease cardiac scintigraphy for the diagnosis of Lewy body-related disorders.
dementia or dementia with Lewy bodies: a randomised, double-blind, Movement Disorders, 26 (7), 1218–24.
placebo-controlled trial. Lancet Neurology, 9 (10), 969–77. Kosaka, K., et al. (1984). Diffuse type of Lewy body disease: progressive
Evans, J. R., et al. (2011). The natural history of treated Parkinson’s disease in dementia with abundant cortical Lewy bodies and senile changes of
an incident, community based cohort. Journal of Neurology, Neurosurgery varying degree – a new disease?. Clinical Neuropathology, 3 (5), 185–92.
and Psychiatry, 82 (10), 1112–18. Kulisevsky, J. and Pagonabarraga, J. (2009). Cognitive impairment in
Ferman, T. J., et al. (2004). DLB fluctuations: specific features that reliably Parkinson’s disease: tools for diagnosis and assessment. Movement
differentiate DLB from AD and normal aging. Neurology, 62 (2), 181–7. Disorders, 24 (8), 1103–10.
Ferman, T. J., et al. (2011). Inclusion of RBD improves the diagnostic Kurlan, R., et al. (2007). Quetiapine for agitation or psychosis in patients with
classification of dementia with Lewy bodies. Neurology, 77 (9), 875–82. dementia and parkinsonism. Neurology, 68 (17), 1356–63.
Fujishiro, H., et al. (2008). Validation of the neuropathologic criteria of Kurz, M. W., et al. (2006). Familial occurrence of dementia and parkinsonism:
the third consortium for dementia with Lewy bodies for prospectively a systematic review. Dementia and Geriatric Cognitive Disorders, 22 (4),
diagnosed cases. Journal of Neuropathology and Experimental Neurology, 288–95.
67 (7), 649–56. Larner, A. J., Mathias, C. J., and Rossor, M. N. (2000). Autonomic failure
Galvin, J. E., et al. (2007). Personality traits distinguishing dementia with preceding dementia with Lewy bodies. Journal of Neurology, 247 (3),
Lewy bodies from Alzheimer disease. Neurology, 68 (22), 1895–901. 229–31.
Geldsetzer, F., et al. (2010). Who was the man who discovered the‘ Lewy Larsson, V., et al. (2011). Quality of life and the effect of memantine in
bodies’? Movement Disorders, 25 (12), 1765. dementia with Lewy bodies and Parkinson’s disease dementia. Dementia
Goebel, G., et al. (2011). A novel computer-assisted image analysis of and Geriatric Cognitive Disorders, 32 (4), 227–34.
[123I]beta-CIT SPECT images improves the diagnostic accuracy of Leverenz, J. B., et al. (2008). Empiric refinement of the pathologic assessment
parkinsonian disorders. European Journal of Nuclear Medicine and of Lewy-related pathology in the dementia patient. Brain Pathology, 18
Molecular Imaging, 38 (4), 702–10. (2), 220–4.
Goetz, C. G., Emre, M., and Dubois, B. (2008). Parkinson’s disease dementia: Litvan, I., et al. (2011). MDS task force on mild cognitive impairment in
definitions, guidelines, and research perspectives in diagnosis. Annals of Parkinson’s disease: Critical review of PD-MCI. Movement Disorders, 26
Neurology, 64 Suppl 2, S81–92. (10), 1814–24.
Halliday, G. M. and McCann, H. (2010). The progression of pathology in Litvan, I., et al. (2012). Diagnostic criteria for mild cognitive impairment in
Parkinson’s disease. Annals of New York Academy of Science, 1184, Parkinson’s disease: Movement Disorder Society Task Force guidelines.
188–95. Movement Disorders, 27 (3), 349–56.
Halliday, G., et al. (2008). The progression of pathology in longitudinally Lobotesis, K., et al. (2001). Occipital hypoperfusion on SPECT in dementia
followed patients with Parkinson’s disease. Acta Neuropathologica, 115 with Lewy bodies but not AD. Neurology, 56 (5), 643–9.
(4), 409–15. MacDonald, P. A., et al. (2011). The effect of dopamine therapy on ventral
Hamilton, J. M., et al. (2012). Early visuospatial deficits predict the occurrence and dorsal striatum-mediated cognition in Parkinson’s disease: support
of visual hallucinations in autopsy-confirmed dementia with Lewy from functional MRI. Brain, 134 (Pt 5), 1447–63.
bodies. American Journal of Geriatric Psychiatry, 20 (9), 773–81. Matsui, Y., et al. (2009). Incidence and survival of dementia in a general
Hansen, L., et al. (1990). The Lewy body variant of Alzheimer’s disease: a population of Japanese elderly: the Hisayama study. Journal of Neurology,
clinical and pathologic entity. Neurology, 40 (1), 1–8. Neurosurgery and Psychiatry, 80 (4), 366–70.
Mattsson, N., et al. (2009). CSF biomarkers and incipient Alzheimer disease and quadriparesis in flexion. Journal of Neuropathology and Experimental
in patients with mild cognitive impairment. Journal of the American Neurology, 20, 237–44.
Medical Association, 302 (4), 385–93. Onofrj, M., et al. (2010). Cohort study on somatoform disorders in Parkinson
McKeith, I. G., et al. (1992). Operational criteria for senile dementia of Lewy disease and dementia with Lewy bodies. Neurology, 74 (20), 1598–606.
body type (SDLT). Psychology and Medicine, 22 (4), 911–22. Papathanasiou, N. D., et al. (2012). Diagnostic accuracy of (123)I-FP-CIT
McKeith, I. G., et al. (1996). Consensus guidelines for the clinical and (DaTSCAN) in dementia with Lewy bodies: a meta-analysis of published
pathologic diagnosis of dementia with Lewy bodies (DLB): report of studies. Parkinsonism and Related Disorders, 18 (3), 225–9.
the consortium on DLB international workshop. Neurology, 47 (5), Pappata, S., et al. (2011). Mild cognitive impairment in drug-naive patients
1113–24. with PD is associated with cerebral hypometabolism. Neurology, 77 (14),
McKeith, I., et al. (2000a). Efficacy of rivastigmine in dementia with Lewy 1357–62.
bodies: a randomised, double-blind, placebo-controlled international Paris, A. P., et al. (2011). Blind randomized controlled study of the efficacy
study. Lancet, 356 (9247), 2031–6. of cognitive training in Parkinson’s disease. Movement Disorders, 26 (7),
McKeith, I. G., et al. (2000b). Prospective validation of consensus criteria for 1251–8.
the diagnosis of dementia with Lewy bodies. Neurology, 54 (5), 1050–8. Parkinson, J. (1817). An essay on the shaking palsy. Whittingham and
McKeith, I. G., et al. (2004). Hallucinations predict attentional improvements Rowland, London.
with rivastigmine in dementia with Lewy bodies. Dementia and Geriatric Perry, R. H., et al. (1990). Senile dementia of Lewy body type. A clinically and
Cognitive Disorders, 18 (1), 94–100. neuropathologically distinct form of Lewy body dementia in the elderly.
McKeith, I. G., et al. (2005). Diagnosis and management of dementia with Journal of Neurological Science, 95 (2), 119–39.
Lewy bodies. Third report of the DLB Consortium. Neurology, 65, Poewe, W. (2005). Treatment of dementia with Lewy bodies and Parkinson’s
1863–72. disease dementia. Movement Disorders, 20 Suppl 12, S77–82.
McKeith, I., et al. (2007). Sensitivity and specificity of dopamine transporter Polymeropoulos, M. H., et al. (1997). Mutation in the alpha-synuclein gene
imaging with 123I-FP-CIT SPECT in dementia with Lewy bodies: a identified in families with Parkinson’s disease. Science, 276 (5321),
phase III, multicentre study. Lancet Neurology, 6 (4), 305–13. 2045–7.
Meeus, B., et al. (2012). DLB and PDD: a role for mutations in dementia and Rahkonen, T., et al. (2003). Dementia with Lewy bodies according to the
Parkinson disease genes? Neurobiology of Aging, 33 (3), 629. consensus criteria in a general population aged 75 years or older. Journal
Molano, J., et al. (2010). Mild cognitive impairment associated with limbic of Neurology, Neurosurgery and Psychiatry, 74 (6), 720–4.
and neocortical Lewy body disease: a clinicopathological study. Brain, Ricci, M., et al. (2009). Clinical findings, functional abilities and caregiver
133 (Pt 2), 540–56. distress in the early stage of dementia with Lewy bodies (DLB) and
Mollenhauer, B., et al. (2006). Total tau protein, phosphorylated tau (181p) Alzheimer’s disease (AD). Archives of Gerontology and Geriatrics, 49 (2),
protein, beta-amyloid(1–42), and beta-amyloid(1–40) in cerebrospinal e101–4.
fluid of patients with dementia with Lewy bodies. Clinical Chemistry Rongve, A., Aarsland, D., and Ballard, C. (2006). Current perspectives in
Laboratory Medicine, 44 (2), 192–5. dementia with Lewy bodies. Aging and Health, 2 (3), 461–72
Mollenhauer, B., et al. (2011). Alpha-synuclein and tau concentrations in Rongve, A., Boeve, B. F., and Aarsland, D. (2010a). Frequency and correlates
cerebrospinal fluid of patients presenting with parkinsonism: a cohort of caregiver-reported sleep disturbances in a sample of persons with early
study. Lancet Neurology, 10 (3), 230–40. dementia. Journal of the American Geriatrics Society, 58 (3), 480–6.
Molloy, S., et al. (2005). The role of levodopa in the management of dementia Rongve, A., et al. (2010b). Core and suggestive symptoms of dementia
with Lewy bodies. Journal of Neurology, Neurosurgery and Psychiatry, 76 with lewy bodies cluster in persons with mild dementia. Dementia and
(9), 1200–3. Geriatric Cognitive Disorders, 29 (4), 317–24.
Mondragon-Rezola, E., et al. (2010). Sleep disorders in Parkinson’s disease: Rongve, A., et al. (2010c). P01–371-Shorter time until nursing home
insomnia and sleep fragmentation, daytime hypersomnia, alterations admission in lewy body dementias as compared to Alzheimer’s dementia.
to the circadian rhythm and sleep apnea syndrome. Rev Neurology, 50 European Psychiatry, 25, 584.
Suppl 2, S21–6. Rongve, A., et al. (2012). Psychosis in Parkinson’s disease. Tidsskrift for den
Muller, C. M., et al. (2005). Staging of sporadic Parkinson disease-related norske Laegeforening, 132 (2), 155–8. Review. Norwegian.
alpha-synuclein pathology: inter- and intra-rater reliability. Journal of Seppi, K., et al. (2011). The Movement Disorder Society evidence-based
Neuropathology and Experimental Neurology, 64 (7), 623–8. medicine review update: treatments for the non-motor symptoms of
Nalls, M. A., et al. (2011). Imputation of sequence variants for identification Parkinson’s disease. Movement Disorders, 26 Suppl 3, S42–80.
of genetic risks for Parkinson’s disease: a meta-analysis of genome-wide Seto-Salvia, N., et al. (2011). Dementia risk in Parkinson disease: disentangling
association studies. Lancet, 377 (9766), 641–9. the role of MAPT haplotypes. Archives of Neurology, 68 (3), 359–64.
Nemani, V. M., et al. (2010). Increased expression of alpha-synuclein reduces Siderowf, A., et al. (2010). CSF amyloid {beta} 1–42 predicts cognitive decline
neurotransmitter release by inhibiting synaptic vesicle reclustering after in Parkinson disease. Neurology, 75 (12), 1055–61.
endocytosis. Neuron, 65 (1), 66–79. Singleton, A. B., et al. (2003). Alpha-synuclein locus triplication causes
Nervi, A., et al. (2011). Familial aggregation of dementia with Lewy bodies. Parkinson’s disease. Science, 302 (5646), 841.
Archives of Neurology, 68 (1), 90–3. Skogseth, R. E., et al. (2011). Biomarkers in spinal fluid of patients with
Noguchi-Shinohara, M., et al. (2009). CSF alpha-synuclein levels in dementia dementia. Tidsskrift for den Norske Laegeforening, 131 (22), 2235–8.
with Lewy bodies and Alzheimer’s disease. Brain Res, 1251, 1–6. Sonnesyn, H., et al. (2009). High prevalence of orthostatic hypotension in
O’Brien, J. T. and Burns, A. (2011). Clinical practice with anti-dementia drugs: mild dementia. Dementia and Geriatric Cognitive Disorders, 28 (4),
a revised (second) consensus statement from the British Association for 307–13.
Psychopharmacology. Journal of Psychopharmacology, 25 (8), 997–1019. Spies, P. E., et al. (2009). Cerebrospinal fluid alpha-synuclein does not
O’Brien, J. T., et al. (2009). Diagnostic accuracy of 123I-FP-CIT SPECT in discriminate between dementia disorders. Journal of Alzheimers Disease,
possible dementia with Lewy bodies. British Journal of Psychiatry, 194 16 (2), 363–9.
(1), 34–9. Spillantini, M. G., et al. (1998). alpha-Synuclein in filamentous inclusions of
Ohtake, H., et al. (2004). Beta-synuclein gene alterations in dementia with Lewy bodies from Parkinson’s disease and dementia with lewy bodies.
Lewy bodies. Neurology, 63 (5), 805–11. Proceedings of the National Academy of Science USA, 95 (11), 6469–73.
Okazaki, H., Lipkin, L. E., and Aronson, S. M. (1961). Diffuse intracytoplasmic Stevens, T., et al. (2002). Islington study of dementia subtypes in the
ganglionic inclusions (Lewy type) associated with progressive dementia community. British Journal of Psychiatry, 180, 270–6.
Suzuki, M., et al. (2006). Impaired myocardial (123)I-metaiodobenzylguanidine Williams, M. M., et al. (2006). Survival and mortality differences between
uptake in Lewy body disease: Comparison between dementia with Lewy dementia with Lewy bodies vs Alzheimer disease. Neurology, 67 (11),
bodies and Parkinson’s disease. Journal of Neurological Science, 240 1935–41.
(1–2), 15–19. Yamaguchi, K., et al. (2005). Abundant neuritic inclusions and microvacuolar
Vlaar, A. M., et al. (2008). Diagnostic value of 123I-ioflupane and changes in a case of diffuse Lewy body disease with the A53T mutation
123I-iodobenzamide SPECT scans in 248 patients with parkinsonian in the alpha-synuclein gene. Acta Neuropathologica, 110 (3), 298–305.
syndromes. European Neurology, 59 (5), 258–66. Zaccai, J., McCracken, C., and Brayne, C. (2005). A systematic review of
Walker, M. P., et al. (2000). The clinician assessment of fluctuation and the prevalence and incidence studies of dementia with Lewy bodies. Age
one day fluctuation assessment scale. Two methods to assess fluctuating Ageing, 34 (6), 561–6.
confusion in dementia. British Journal of Psychiatry, 177, 252–6. Zarranz, J. J., et al. (2004). The new mutation, E46K, of alpha-synuclein
Weintraub, D., et al. (2010). Impulse control disorders in Parkinson disease: causes Parkinson and Lewy body dementia. Annals of Neurology, 55 (2),
a cross-sectional study of 3090 patients. Archives of Neurology, 67 (5), 164–73.
589–95. Zesiewicz, T. A., et al. (2010). Practice parameter: treatment of nonmotor
Weintraub, D., et al. (2012). Alzheimer’s disease pattern of brain atrophy predicts symptoms of Parkinson disease: report of the Quality Standards
cognitive decline in Parkinson’s disease. Brain, 135 (Pt 1), 170–80. Subcommittee of the American Academy of Neurology. Neurology, 74
Williams-Gray, C. H., et al. (2007). Evolution of cognitive dysfunction in an (11), 924–31.
incident Parkinson’s disease cohort. Brain, 130 (Pt 7), 1787–98.
Williams-Gray, C. H., et al. (2009). Apolipoprotein E genotype as a risk factor
for susceptibility to and dementia in Parkinson’s disease. J Neurology, 256
(3), 493–8.
CHAPTER 36
Frontotemporal dementia
Vincent Deramecourt, Florence
Lebert, and Florence Pasquier
Frontotemporal dementia (FTD) is characterised clinically by and praxis preserved), and investigation (neurological, neuropsy-
progressive changes in social, behavioural, and language function. chological, EEG, and imaging), suggestive of an impairment of the
The highest incidence of FTD is between 50 and 60 years of age, anterior part of the brain. An update and extension of the Lund
although it sometimes occurs much earlier and late-onset FTD has and Manchester criteria was published in 1998 (Neary et al., 1998).
been reported. FTD is the second commonest cause of degenera- According to this revision, FTD is one of three clinical presenta-
tive dementia in patients aged 65 years or less, after Alzheimer’s tions of frontotemporal lobar degeneration (FTLD), together with
disease (AD) (Ratnavalli et al., 2002), and it accounts for about progressive nonfluent aphasia (PNFA) and semantic dementia (SD,
5% of late-onset dementia (Pasquier et al., 1999a). Since its initial i.e. fluent aphasia and associative agnosia). The generic term FTLD
description, especially by Arnold Pick, it remains underdiagnosed refers to the common neuropathological feature, i.e. the circum-
and often misdiagnosed as AD (Mendez et al., 1993; Knopman scribed progressive degeneration of the frontotemporal lobes.
et al., 2005). About 40% of FTD cases are not diagnosed (Rosso Thus FTD may refer to all presentations of FTLD or to one of
et al., 2003), and those who are are often diagnosed after some three main variants (the behavioural variant, bvFTD), with the
delay (Pasquier et al., 2004). other two variants (SD and PNFA) being designated as language
variants of FTD. Diagnostic criteria for primary progressive apha-
History of FTD: Evolution of Concepts and sia (PPA) and bvFTD were last revised in 2011 (Gorno-Tempini
et al., 2011; Rascovsky et al., 2011) and now include relevant neu-
Terminology roimaging features.
Arnold Pick described six patients with circumscribed atrophies
(Pick, 1892, 1901a, 1901b, 1904, 1906) more than a century ago. Epidemiology of Frontotemporal Dementia
Two pathological hallmarks of focal brain atrophies were described
by Aloïs Alzheimer in 1911 (Alzheimer, 1910–1911): neuronal bal- Prevalence and incidence
looning (Pick cells) and intraneuronal argyrophilic inclusion bod- The highest prevalence has been reported from two independent
ies (Pick bodies) without senile plaques or tangles. Difficulties with studies in the UK and one Italian study, with an estimated preva-
translation contributed to the long-lasting confusion between Pick’s lence of FTD of 15–22 per 100,000 inhabitants aged 45–64 years
atrophy (nonspecific circumscribed atrophy) and Pick’s disease (Ratnavalli et al., 2002; Harvey et al., 2003; Borroni et al., 2010),
(with specific histological features) (Pasquier and Petit, 1997) and, which was almost half of the prevalence of AD in this age group
to this day, difficulties remain in relating the range of pathological (Harvey et al., 2003). However, a study from the Netherlands esti-
changes in FTD to different patterns of clinical presentation. Lars mated the prevalence of FTD to be significantly lower (9.4 per
Gustafson, a psychiatrist in Lund (Sweden), observed patients with 100,000 in the age group of 60–69 years) (Rosso et al., 2003b).
degenerative dementia whose behaviour differed from that of typical The lower prevalence relative to AD in that series is consistent
patients with AD, and Neary and colleagues from Manchester pub- with some pathological series (Ikeda et al., 2004a). In a Swedish
lished similar cases at the same period. The Lund and Manchester population-based sample of 85-year-olds, the estimated prevalence
groups both emphasized the frequency of these dementias and the was 3.1 per 100 inhabitants and thus was higher than previously
reliability of the clinical distinction from AD. An international con- expected in this age group. However, only a minority of these
ference on ‘frontal lobe degeneration of nonAlzheimer type’ was patients were demented at the time of examination (Gislason et al.,
held in Lund in 1992 and an entire issue of Dementia was dedi- 2003). Two reported incidence studies of FTD were remarkably con-
cated to this update in 1993. In 1994, the Lund and Manchester sistent: 3.5 and 4.1 cases per 100,000 person-years in the age-group
groups published a consensus on ‘clinical and neuropathological of 45–64 years (Knopman et al., 2004; Mercy et al., 2008).
criteria for frontotemporal dementia’ (Brun et al., 1994). The core
diagnostic features included behavioural disorders of insidious Age
onset and slow progression, affective symptoms, speech disorders, FTD are presenile dementias, with a median age of onset between
intact abilities to negotiate the environment (spatial orientation 45 and 60 years, although approximately 10% have an age of onset
of over 70 years (up to 89 years) (Seelaar et al., 2008). Patients with the absence of visuospatial and perceptual symptoms is a reliable
FTD (mean age 57.5 years) and SD (59.3 years) had an earlier age at feature (Hodges et al., 2004).
onset than patients with PNFA (63.0 years) in the large US–German
cohort (Johnson et al., 2005). Behavioural variant FTD (bvFTD)
An international consortium recently published criteria for pos-
Sex sible, probable, and definite bvFTD (Box 36.1) (Rascovsky et al.,
Although there is no overall sex preponderance in FTD (Rosso 2011).
et al., 2003), in the US–German cohort there were significantly The clinical features are so characteristic that the diagnosis may
more men diagnosed as having FTD (63.5%) and SD (66.7%) than be made on the basis of an interview with a close relative even after
PNFA (39%). the patient’s death (Barber et al., 1995). Behavioural and affective
Genetics Box 36.1 Revised diagnostic criteria for the behavioural variant of
This is addressed in detail in Chapter 8 and so will only be summa- frontotemporal dementia (Rascovsky et al., 2011)
rized here. About 30–50% of bvFTD patients have a positive family
history of dementia (at least one first-degree relative with dementia I. Neurodegenerative disease
before the age of 80 years, or identification of a mutation) (Stevens The following symptom must be present to meet criteria for
et al., 1998; Seelaar et al., 2008; Rohrer et al., 2009). Patients with bvFTD:
SD or PNFA have a much lower frequency. An autosomal pattern
A. Shows progressive deterioration of behaviour and/or cog-
of inheritance is found in 10–27% of all FTD patients (Rosso et al.,
nition by observation or history (as provided by a knowl-
2003; Seelaar et al., 2008; Rohrer et al., 2009). Incidence of FTD
edgeable informant)
is increased ten-fold in the first-degree relatives of FTD patients
compared with the incidence of FTD in the population, without II. Possible bvFTD
clustering of other causes of dementia (Grasbeck et al., 2005). Three of the following behavioural/cognitive symptoms (A–F)
The first report of genetic linkage in an FTD pedigree was to must be present to meet criteria. Ascertainment requires that
chromosome 17q21–22 (Wilhelmsen et al., 1994), and the term symptoms be persistent or recurrent, rather than single or rare
‘FTD and parkinsonism linked to chromosome 17’ (FTDP-17) events:
was adopted to describe the clinical and pathological spectrum.
Mutations in the tau gene (microtubule-associated protein tau = A. Early behavioural disinhibition (one of the following
MAPT) have been first identified in some of these families (about 45 symptoms A1–A3 must be present):
different mutations so far), but a number of these families showed A1: Socially inappropriate behaviour
neither tau mutations nor tau deposition on neuropathological A2: Loss of manner or decorum
assessment, suggesting that there was a second gene involved that
is located close to the MAPT gene. This gene was identified in 2006, A3: Impulsive, rash, or careless actions
coding for progranulin (PGRN), a growth factor involved in mul- B. Early apathy or inertia (one of the following symptoms
tiple physiological and pathological processes including develop- B1–B2 must be present):
ment, wound repair, inflammation, and tumorigenesis (Baker et al.,
B1: Apathy
2006; Cruts et al., 2006). More than 65 different mutations of PGRN
gene have been published so far. B2: Inertia
The genetic heterogeneity of FTD is further emphasized by the C. Early loss of sympathy or empathy (one of the following
rare occurrence (less than 5% of familial cases) of mutations in the symptoms C1–C2 must be present):
VCP, CHM2B, TARDBP, and FUS genes.
Hereditary FTD has also been associated with amyotrophic lat- C1: Diminished response to other people’s needs and
eral sclerosis in some families, and this form has shown linkage to feelings
another genetic locus: chromosome 9q21-q22 (Hosler et al., 2000). C2: Diminished social interest, interrelatedness, or per-
The mutation is an expansion of noncoding hexanucleotide repeat sonal warmth
in the gene C9ORF72 (DeJesus-Hernandez et al., 2011). D. Early perseverative, strereotyped, or compulsive/ritualistic
Genetic counselling is difficult in FTD because of the clinical het- behaviour (one of the following symptoms D1–D3 must
erogeneity, misdiagnoses, and loss of family history. In addition, be present):
not all the mutations have been reported in families with clear auto-
somal dominant inheritance. D1: Simple repetitive movements
D2: Complex, compulsive, or ritualistic behaviours
Clinical Features D3: Stereotypy of speech
FTD includes a progressive behavioural disorder with insidi- E. Hyperorality and dietary changes (one of the following
ous onset, affective symptoms, language disorder, frontal execu- symptoms E1–E3 must be present):
tive dysfunction with preserved spatial orientation, and selective
E1: Altered food preferences
frontotemporal atrophy or hypoperfusion/hypometabolism. The
absence of severe amnesia is debatable (Graham et al., 2005), but
(Continued)
CHAPTER 36 frontotemporal dementia 481
changes (change of character and disordered social conduct) are

Box 36.1 (Continued)
the dominant features initially and throughout the disease course.
E2: Binge eating, increased consumption of alcohol or Contrary to AD, behavioural changes precede or are associated
cigarettes with cognitive decline, and memory impairment is rarely the first
symptom reported by the family, who typically consider it of sec-
E3: Oral exploration or consumption of inedible objects ondary importance compared to the behavioural disorder. The
F. Neuropsychological profile: executive/generation deficits most frequently reported first symptom is a loss of interest, often
with relative sparing of memory and visuospatial function attributed to depression, although families usually recognize that
(all of the following symptoms F1–F3 must be present): patients do not show sadness or feelings of worthlessness or guilt
(Pasquier et al., 1999b). Socially disruptive behaviours may lead to
F1: Deficits in executive tasks
arrest and even prosecution (Miller et al., 1997). Sociopathic acts
F2: Relative sparing of episodic memory include unsolicited sexual acts, traffic violations, physical assaults,
F3: Relative sparing of visuospatial skills and other unacceptable behaviours. About 10% of patients with
FTD exhibit a dramatic change in their self as defined by changes in
III. Probable bvFTD political, social, religious (Miller et al., 2001), or musical (Geroldi
All of the following symptoms (A–C) must be present to meet et al., 2000) values, usually correlated with a selective nondominant
criteria: frontal dysfunction. Right-sided FTD is associated with socially
A. Meets criteria for possible bvFTD undesirable behaviour (criminal behaviour, aggression, loss of job,
alienation from family/friends, financial recklessness, sexually devi-
B. Exhibits significant functional decline (by caregiver report ant behaviour, and abnormal response to spousal crisis) (Mychnack
or as evidenced by Clinical Dementia Rating Scale or et al., 2001).
Functional Activities Questionnaire scores)
C. Imaging results consistent with bvFTD (one of the follow- Behaviour
ing symptoms C1–C2 must be present): A behavioural frontotemporal dysfunction assessment scale vali-
dated on patients at the mild stage (MMSE > 18) helps to distin-
C1: Frontal and/or anterior temporal atrophy on MRI
guish between FTD, AD, and vascular dementia (VaD) (Lebert
or CT
et al., 1998). The items of the structured interview are classified into
C2: Frontal and/or anterior temporal hypoperfusion or four classes: self-monitoring dyscontrol, self-neglect, self-centered
hypometabolism on PET or SPECT behaviour, and affective behaviour (Box 36.2). A score of one is
IV. bvFTD with definite FTLD pathology given for the class if at least one symptom is present and it reflects
a substantial change from the patient’s premorbid state and is not a
Criterion A and either criterion B or C must be present to meet
long-standing character trait. A total score of three or more gives a
criteria
sensitivity of 100%, a specificity of 93%, and a diagnostic accuracy
A. Meets criteria for possible or probable bvFTD of FTD of 97% (Lebert et al., 1998). The assessment of neuropsychi-
B. Histopathological evidence of FTLD on biopsy or at post atric symptoms with other standardized scales or inventories can
mortem also be useful to distinguish dementia patients with FTD and AD
C. Presence of a known pathogenic mutation (Levy et al., 1996; Kertesz et al., 1997, 2000; Swartz et al., 1997b;
Mendez et al., 1998; Hirono et al., 1999; Bozeat et al., 2000a). FTD
V. Exclusion criteria for bvFTD patients frequently have an increase in appetite, with a loss of social
Criteria A and B must be answered negatively for any bvFTD graces, and they may eat quickly. They have a frequent altered food
diagnosis. Criterion C can be positive for possible bvFTD but preference for sweet foods, and significant weight gain occurs in
must be negative for probable bvFTD more than 30% of FTD patients. Stereotypic behaviours are more
A. Pattern of deficits is better accounted for by other nonde- complex in FTD than in AD (Nyatsanza et al., 2003) and some
generative nervous system or medical disorders of them are more common in FTD than in AD: counting/clock
watching; consistently choosing the same leisure activity or hobby;
B. Behavioural disturbance is better accounted for by a psy-
repetitively eating the same food; and rigid adherence to routine
chiatric diagnosis
(Nyatsanza et al., 2003). Delusions are reported but hallucinations
C. Biomarkers strongly indicative of Alzheimer’s disease or are rare.
other neurodegenerative process At interview, FTD patients with sociopathic acts are aware of
VI. Additional features their wrong behaviour, but cannot prevent themselves from acting
impulsively (Mendez et al., 2005b). They claim subsequent remorse,
A. Presence of motor neuron findings suggestive of motor but they do not act on it or show concern for the consequences.
neuron disease Loss of insight is a core diagnostic criterion for FTD. FTD
B. Motor symptoms and signs similar to corticobasal degen- patients may recognize behavioural change or cognitive deficits
eration and progressive supranuclear palsy (such as memory difficulties), but do not see any problem with
C. Impaired word and object knowledge these changes, although they may have lost their job, have finan-
cial difficulties, or be separated from their spouse and children. It
D. Motor speech deficits
differs from the unawareness of disease (anosognosia) observed in
E. Substantial grammatical deficits AD. In a study comparing questionnaires completed by patients
Box 36.2 Frontotemporal Behavioural Scale (BFS) (Lebert et al., Cognition

1998) At early stages, brief global ratings, such as the MMSE, are normal,
but more detailed cognitive batteries, such as the Dementia Rating
I. Self-monitoring dyscontrol Scale (Mattis, 1976), are more sensitive to early disease. Before
becoming too apathetic, patients are usually compliant during such
Changes of food taste (e.g. has developed new preference for
testing, although they manifest an ‘economy of effort’ and tend to
sweets)
provide ‘don’t know’-type responses. Responses can be rapid and
Hyperorality (eats excessively, puts inedible things in mouth) impulsive or very delayed, and sometimes logorrhoea may be dif-
Alcohol abuse (new appetite for alcohol) ficult to control. The striking feature, when formally testing the
Verbal disinhibition (makes remarks without social awareness, patients, is the dissociation between severe alteration in personal-
loss of social tact) ity and behaviour and breakdown in social competence, and the
relative preservation of cognitive skills. Most studies comparing
Behavioural disinhibition (behaves without social awareness,
FTD patients with AD patients have matched for overall dementia
without tact)
severity with the MMSE, and found significantly worse perform-
Irritability (becomes easily irritable without reason) ance among AD patients on verbal anterograde memory tests, non-
Inappropriate emotional reacting (laughs or cries without any language measures such as visual construction, nonverbal memory,
change of affective context or mood changes) and calculation. A comparison of the pattern of cognitive deficits in
Restlessness (becomes physically overactive at any time, unable autopsy-confirmed FTD and AD patients showed that patients with
to stay in the same place for a long time) FTD performed significantly worse than patients with AD on letter
and category fluency tests (sensitive to frontal lobe dysfunction),
II. Self-neglect (decline in personal hygiene and grooming) but significantly better on the Mattis Dementia Rating Scale mem-
Not washing, dirtiness, neglect of personal hygiene ory subscale, block design test, and clock drawing test (sensitive
Neglect of clothing, lack of harmonization of clothing to medial temporal and parietal association cortices) (Rascovsky
et al., 2002).
Lack of hair care Most FTLD patients are not obviously amnesic. They are able to
III. Self-centred behaviour provide autobiographical information and behave as if they had
Apathetic (lacks initiative, needs to be stimulated to initiate ‘absent-mindedness’ and ‘faulty-attention’ rather than primary
things, tendency to sleep unless stimulated) amnesia. They do not forget personally relevant events such as
mealtimes. However, they are impaired in formal testing. The pat-
Perseverative, stereotyped behaviour (ritualistic preoccupa- tern of cognitive decline differs from that of AD: memory perform-
tions, becomes anxious about money, food, tobacco, times of ance benefits from cues and from the provision of multiple-choice
meals, etc.) alternative responses; FTD patients have better encoding, and dem-
Perseverative thoughts, inflexibility onstrate a slower forgetting rate than AD patients. This suggests a
Hypochondriasis (somatic complaints) deficit of retrieval strategies more than a storage problem (Pasquier
et al., 2001). However, severe memory impairment does occur in
Social neglect (lack interest in social activities)
FTD: about 10% of patients with FTD from the brain banks of
Selfishness, self-centredness, lack of empathy or concern for Cambridge (UK) and Sydney showed this pattern (Graham et al.,
others, and loss of feelings of embarrassment 2005), with no noticeable behavioural change initially in half the
IV. Affective disorders subjects. All of them developed behavioural features and the initial
diagnosis of AD was revised to FTD in half of them. Deficits in
Elation (elated at any time)
episodic memory are thus more common than previously reported
Apparent sadness (at any time, the face is unexpressive) and sometimes as severe as in AD even after accounting for disease
Flat affect (affective indifference, especially for family severity (Hornberger et al., 2010).
members) A defective performance on the classical ‘frontal lobe’ (execu-
Emotionalism (heightened tendency to cry more frequently, tive) tasks, such as tests of abstraction, planning, mental flexibil-
more easily, or more vigorously because of precipitating cir- ity (Stroop test, Trail making test), sorting tests (Wisconsin Card
cumstances: thoughts (about family, illness, sad events); Sorting test), logical arrangement of images, or verbal fluency, is
expression of sympathy, arrival or departure of visitors, pres- not specific for FTD. Moreover, these tests may be performed in the
ence of strangers, inability to perform a task, watching televi- normal range in the early stage of the disease.
sion (scenes of tragedy, war, etc.); listening to music) Patients with bvFTD have difficulties in recognizing facial emo-
tions, including anger, sadness, disgust, fear, and contempt—a
deficit that may contribute to their impaired social skills (Lavenu
and first-degree relative informants on the current personality of et al., 1999; Lavenu and Pasquier, 2005; Lough et al., 2006; Snowden
the patients, patients with FTD exaggerated positive qualities and et al., 2008). Impairment is most pronounced in the recognition
minimized negative qualities. Informants completed a second ques- of negative emotions (Rosen et al., 2002; Fernandez-Duque and
tionnaire retrospectively, describing the subjects’ personality before Black, 2005). Besides facial recognition, deficits in the recognition
disease onset. The self-reports of patients with FTD most closely of emotion involve other modalities such as vocal emotion (Keane
matched their premorbid personalities, suggesting a failure to et al., 2002). Using the Interpersonal Reactivity Index—a measure
update their self-image after disease onset (Rankin et al., 2005b).
of cognitive and emotional empathy—completed by first-degree Lack of emotional response is pervasive in bvFTD, whereas it
relatives of patients, levels of empathy are lower in FTD than in is more selective in SD, affecting particularly the capacity to show
AD and normal controls (Rankin et al., 2005a), with a disrup- fear. Semantic loss may explain the lack of awareness of danger in
tion of cognitive empathy but not of the emotional component of patients with SD: these patients no longer have conceptual knowl-
empathy. edge about the potentially adverse properties of objects (e.g. boiling
FTD patients have an impaired theory of mind (Gregory et al., water) or situations (e.g. crossing a busy road). Loss of emotional
2002): they have impaired ability to interpret social situation and insight and absence of the feelings of disgust favours bvFTD,
ascribe mental states to others, which is interpreted as one compo- whereas presence of insight and the capability of showing disgust
nent of widespread executive deficits (Snowden et al., 2003). These favours SD (Snowden et al., 2001).
difficulties result in impaired detection of social faux pas (Gregory
et al., 2002), discrimination of sincere from sarcastic exchanges Language—cognition
(Kipps et al., 2009), and understanding of situations requiring Patients with SD often show deficits in formal neuropsychologi-
moral judgement (Mendez and Shapira, 2009). cal testing, especially involving verbal material. Their breakdown
in semantic representations underlying language comprehension
Phenocopies is shown by a breakdown in naming tests such as word-picture
A subset of patients who present with the clinical features of bvFTD matching (single word comprehension), the pyramid and palm
do not progress to frank incapacitating dementia (Kipps et al., tree test, and a category fluency test. Assessment of language shows
2007). Such patients are almost always men and they either remain semantic paraphasias (frank within-category semantic errors: e.g.
stable over many years or improve (Davies et al., 2006; Hornberger ‘dog’ instead of ‘goat’, or use of general, high-level category terms
et al., 2009). Several features distinguish these nonprogressor or when the context requires a specific word: e.g. ‘creature’ instead of
phenocopy cases from those with true FTD, notably normal or ‘goat’), and surface dyslexia or dysgraphia (difficulties with reading
marginal impairment on neuropsychological tests of executive or spelling irregular words).
function, preserved memory and social cognition, a lack of overt They have good recognition memory (if perceptually iden-
atrophy on MRI, and normal metabolic imaging. The aetiology of tical) for objects and faces, but they appear to be affected by
the phenocopy syndrome is still a matter of debate. a peculiar form of autobiographical amnesia, involving older
memories more than recent ones (Snowden et al., 1996; Piolino
Semantic dementia (SD) et al., 2003; Hou et al., 2005). Also in the case nonautobiograph-
Semantic dementia presents with progressive loss of vocabulary ical remote memory, they appear to show a ‘reverse’ temporal
affecting expressive and receptive language in the context of fluent gradient, with better retrieval of recent famous names (Hodges
speech production. The patients show anomia, impairment in sin- and Graham, 1998). This finding has been attributed to the rela-
gle word comprehension, and impoverished semantic knowledge, tive sparing of the hippocampal formation, which according to
with preservation of phonology, syntax, visuospatial abilities, and some models is responsible for not only memory acquisition
day-to-day (episodic) memory (Snowden et al., 1989; Hodges et al., but also the storage of recent memories. On the other hand, the
1992). Aware of their difficulties, patients with SD may complain of severe disruption of temporal neocortex would be responsible
‘loss of words’. Some authors have described a left temporal variant for the loss of semantic memory, as well as of remote autobio-
of FTD with mainly language disorders (Seeley et al., 2005), and a graphical knowledge.
right temporal variant of FTD with mainly prosopagnosia changes Because of the multimodal semantic memory impairment,
(Joubert et al., 2004). patients have impaired identification of sounds (Bozeat et al.,
2000b), objects, famous faces (Snowden et al., 2004), places and
Behaviour monuments, odours, etc. (associative agnosia). They have normal
The only behaviours that differ significantly between bvFTD and SD performances in tests of attention, executive functions (Perry and
on behavioural scales are apathy, greater in bvFTD, and sleep disor- Hodges, 2000), nonverbal problem-solving (progressive matrices),
ders, which are more frequent in SD (Liu et al., 2004). Depression is and perceptual and spatial abilities.
also more frequent in SD than in bvFTD (Bozeat et al., 2000a; Liu et
al., 2004), a finding that may be explained by the better insight that Progressive nonfluent aphasia (PNFA)
patients with SD have into some of their symptoms. PNFA is included in the primary progressive aphasia (PPA) spec-
Both bvFTD and SD groups show significantly lower levels of trum. Subsequently, some behavioural changes may develop. In
empathy than either AD patients or normal controls, but patients PNFA, the predominant impairment is a profound disturbance of
with SD show disruption of both emotional and cognitive empathy, expressive language, accompanied by left perisylvian atrophy and
whereas bvFTD patients show only disruption of cognitive empa- anterior insular hypometabolism leading to mutism, but even those
thy (Rankin et al., 2005a). Social avoidance occurs more often in patients who are mute may have relatively well-preserved memory
bvFTD and social seeking in SD (Snowden et al., 2001). and visuospatial orientation and function surprisingly well in the
Patients with SD characteristically show a narrowing of inter- community in contrast to aphasia in AD, which is usually superim-
ests and often become preoccupied with a single activity, which posed on memory and visuospatial loss.
they pursue assiduously. They are more likely to establish repeti-
tive behavioural routines and to clockwatch, and they show greater Behaviour
emotional insightfulness, as defined by demonstration of distress, There is no behaviour change initially. Later, changes similar to
anxiety, or concern when confronted by their difficulties. They those of bvFTD may occur. However, self-neglect and bland affect
manifest unusual preoccupation for money and parsimony. are unusual.
Language—cognition hypometabolism, whereas the disinhibited syndrome demonstrates

They exhibit nonfluent spontaneous speech which is laborious a selective hypometabolism in interconnected limbic structures (the
and characterized by agrammatism, phonemic paraphasias (sound cingulate cortex, hippocampus/amygdala, and accumbens nucleus)
distortions, distorted sound substitutions), and a lack of words (Franceschi et al., 2005). PET amyloid imaging (see Chapter 12),
(Gorno-Tempini et al., 2011). Confrontation naming may be better which uses specific beta-amyloid tracers, has shown promising
than expected when listening to the spontaneous speech. Numerous results in discriminating AD and FTD cases (Engler et al., 2008).
phonemic paraphasias occur in oral and written language, even if
written production is less impaired than oral output. Reading aloud CSF biomarkers
is affected, but comprehension (in all modalities) is preserved for a Although a low level of tau in the CSF is never seen in AD and
long time. is observed in a third of neuropathologically confirmed FTD
It is difficult to demonstrate the integrity of nonlanguage (Grossman et al., 2005), CSF biomarkers are not yet reliable diag-
domains in PNFA because most neuropsychological tests of mem- nostic markers for FTD, since tau levels vary widely. This is probably
ory, reasoning, and attention require language competence for because of the large spectrum of histopathologies contributing to
their performance. One study tested reasoning and cognitive flex- the disease. Phospho-tau has been shown to differentiate AD from
ibility nonverbally in patients with PNFA using a modified version nonAD groups of dementia, especially from FTD (Hampel and
of the visual verbal test. Patients with PPA and normal controls Teipel, 2004), but not consistently (Verbeek et al., 2005). Decreased
performed similarly, detecting commonalities among objects and levels of progranulin protein are found in plasma, serum, and cer-
shifting from one sorting principle to another. In contrast, both AD ebrospinal fluid (CSF) by ELISA, and may reliably differentiate
and FTD patients were significantly impaired on both measures PGRN mutations carriers from noncarriers (Finch et al., 2009).
(Wicklund et al., 2004).
Differential Diagnosis
Physical examination
In most cases, it is easy to distinguish between AD and FTD
Physical signs are generally limited to the presence of primitive (Pasquier, 2005). Information from close relatives can distinguish
reflexes. FTD can be associated with clinical manifestation of an between the two diseases (Barber et al., 1995): FTD presents with
extrapyramidal disorder or motor neuron disease. Limb rigidity is progressive changes in social conduct in the context of good cog-
reported in bvFTD but not in semantic dementia (Snowden et al., nitive skills, preserved spatial orientation, or progressive language
2001). disorders; AD starts usually with anterograde episodic memory
deficits or, more rarely, with language or visuospatial disorders in
Investigations the context of preserved social skills. The presence of impaired ori-
entation and apraxia increases the likelihood of a patient having
Imaging in FTD AD, while the presence of problem-solving difficulty increases the
Standard neuropsychological tests and conventional brain imag- likelihood of a patient having FTD. Imaging shows only anterior
ing techniques (MRI and SPECT) may not be sensitive to the cerebral abnormalities in FTD more or less spreading posteriorly.
early changes in FTD (Pasquier et al., 1997; Gregory et al., 1999). Changes located in the posterior part of the brain with no continu-
Over time, however, abnormalities in the anterior part of the brain ity with anterior changes exclude the diagnosis of FTD, but pre-
develop in all patients and examples are given in Chapter 12. dominant changes in the anterior part of the brain can be observed
Structural imaging in confirmed AD.
Structural imaging commonly shows atrophy of the frontal lobes, Perhaps the most difficult differential diagnosis of FTD is VaD,
the anterior part of the temporal lobes, anterior cingulate, and insu- mainly the subcortical ischaemic subtype, because both have apathy
lar cortex (Schroeter et al., 2007). Increasing severity of atrophy (up and irritability as key features. In VaD, disinhibition and physical
to 8% per year) occurs with increasing disease duration, but severity neglect are frequent symptoms (just like in FTD), whereas hyper-
of atrophy is not related to pathological subtype (Kril et al., 2005). emotivity is much more common in VaD. In FTD, lack of concern
Patterns of grey matter atrophy might be predictive of the underly- or bland affect are more frequent than hyperemotionalism. MRI is
ing pathology in bvFTD defined by their neuropathological marker, a crucial tool for the differential diagnosis, looking for evidence of
namely tau protein or TDP-43, with bilateral dorsolateral prefron- cerebrovascular disease, and an important issue in the management
tal atrophy in Pick’s disease, and asymmetric left and right tempo- of patients is the control of vascular risk factors in VaD.
ral lobe atrophy being associated with FTLD-TDP and FTLD-tau, Patients with dementia with Lewy bodies may present early exec-
respectively (Whitwell et al., 2005). Brain changes are not limited to utive and language dysfunction or behavioural or affective states
the cortex. Atrophy is also present in many subcortical brain regions, that suggest FTD (Bonner et al., 2003). But cognitive fluctuations,
including the amygdala, hippocampus, caudate, striatum, thalamus, complex visual hallucinations, and REM-sleep disorders are not
and hypothalamus (Piguet et al., 2011), accompanied by reduction symptoms of FTD, although hallucinations have been described in
in connectivity among subcortical structures (Zhou et al., 2010). families with FTD with PGRN mutation.
Other neurodegenerative diseases may also present with symp-
Functional imaging toms suggestive of FTD. However, subcortical dementias, e.g. due
PET or SPECT shows decreased resting glucose metabolism levels to Huntington’s disease, usually have other characteristic features
in frontal and temporal cortices in FTD, contrasting with a decrease (see Chapter 37). Progressive supranuclear palsy (PSP) may present
in parietal, posterior cingulate, and temporal regions in AD (Jeong with frontal lobe symptoms before the parkinsonian features appear
et al., 2005). Functional imaging indicates that an apathetic syn- and the supranuclear palsy occurs. Similarly, overlap exists between
drome is associated with a prevalent dorsolateral and frontal medial FTD and MND, both clinically and pathologically.
It is also important to note that many symptoms in FTD may Progression of physical signs
suggest the presence of a mental illness other than dementia, but Weight gain typically occurs initially, but when apathy prevents
criteria for full psychiatric syndromes are rarely met. Apathy, loss overeating, weight loss occurs. Difficulty with swallowing is fre-
of interest, and change in eating behaviour may be interpreted as quent and may be alarming because of the inappropriate hypero-
depression, but FTD patients do not experience sadness or guilt, do rality. Extrapyramidal features, mainly rigidity, are almost constant,
not show pessimism and suicidal thinking, and in fact show a sur- and hemiparesis is relatively frequent in asymmetrical cases. A
prising lack of concern about themselves. FTD patients may have minority of patients develop motor neuron disease. Low blood
stereotypical movements with compulsive behaviours suggestive pressure, cold extremities, hypotension, and other dysautonomic
of obsessive-compulsive disorders. However, patients have no dis- disorders occur. Seizures are infrequent. Patients with FTD often
tressing feelings of internal or psychic tension, anxiety, or depres- die suddenly, even before the terminal stage of the disease, and the
sion (Mendez et al., 2005a). Disinhibition and hyperactivity may cause of this is often not identified. This may be related to dysau-
be suggestive of a manic episode, but, again, other characteristic tonomia, and choking was the direct cause of premature death in
symptoms of mania, e.g. grandiose thinking and elated mood, tend more than 13% of patients (Pasquier et al., 2004).
not to occur.
Survival
Natural History Although FTD is often thought to have a shorter course than AD,
Progression of behaviour it is not clear this is the case because there is a longer delay before
first symptoms and diagnosis in FTD than in AD (Pasquier et al.,
Behavioural changes progress whatever the presentation (Marczinski
2004). In one study, the mean duration of FTD was estimated at 8
et al., 2004). Some changes, such as restlessness, eating disorders,
years (2–20 years) (Snowden et al., 1996 and in a series of 552 con-
and hyperorality, are long lasting. Verbal disinhibition decreases
secutive patients followed-up for 5.5 years, mean duration of the
together with a reduction of speech, evolving toward mutism. All
disease was 2 years longer in bvFTD (n = 73) than in AD (n = 479)
patients become increasingly apathetic. Recently, the frontotempo-
patients, probably because at entry mean age was 10 years younger
ral dementia rating scale (FRS) was developed specifically for FTD:
and mean MMSE score was 5 points higher in FTD than in AD.
this staging tool incorporates changes in behaviour and activities of
But when adjusted for age, sex, and level of education, survival rate
daily living (Mioshi et al., 2010).
did not differ significantly (Pasquier et al., 2004). Determinants of
Progression of cognitive changes reduced survival in FTD are the association with ALS and language
impairment at diagnosis (Garcin et al., 2009).
At first, FTD typically manifests with behavioural changes and rela-
tively stable global cognition. Cognitive decline later occurs, with
the mean MMSE decreasing by 2.3 points in 2 years in one study Neuropathology
(Pasquier et al., 1999b). However, important individual variations Unlike other dementia syndromes, notably AD, FTLD encompasses
occurred and 21 patients (30%) could not perform the MMSE after considerable pathological heterogeneity. This is dealt with in detail
this period. After a 5-year follow-up the mean annual decline of in Chapter 6 and summarized in Fig. 36.1. About 40% of cases are
the MMSE score was found to be 0.9 + 1.4 in FTD patients with a characterized by tau deposition in neurons or glial cells, or both.
mean MMSE score at first visit of 24.5 + 11.5 (Pasquier et al., 2004). This first group comprises several subvariants, including Pick’s dis-
Other studies have reported a larger annual MMSE score decline, ease (with the classical argyrophilic inclusion bodies named Pick’s
probably because the drop in responses below a score of 18 was not bodies), corticobasal degeneration, progressive supranuclear palsy,
taken into account (Roberson et al., 2005). argyrophilic grains disease, and cases with MAPT gene mutations.
FTLD Neuropathology
~40% ~60% <1%
Fig. 36.1 Neuropathological algorithm for the
Tau-positive Ubiquitin-positive No neuropathological characterization of FTLD and
inclusions inclusions inclusion potential associated gene. 3R/4R, Three or four repeat
tauopathy; PiD, Pick disease; FTDP-17, frontotemporal
90% 10% dementia with parkinsonism linked to chromosome
TDP-43 + FUS + 17 (MAPT mutation); PSP, progressive supranuclear
Pick bodies Neuronal/glial Tangles palsy; CBD, corticobasal degeneration; AGD,
3R tauopathy 4R tauopathy 3R + 4R
argyrophylic grains disease; AD NFTd, neurofibrillary
Tauopathy
tangle dementia; FTLD-TDP, frontotemporal lobar
degeneration with TDP-43 inclusions; FTLD-FUS,
PSP frontotemporal lobar degeneration with FUS
PiD NFTd FTLD-TDP FTLD-FUS FTLD- FTLD- inclusions; aFTLD-U, atypical frontotemporal lobar
DCB Type A (aFTLD-U. Ups ni
FTDP-17 FTDP-17 degeneration with ubiquitin-positive inclusions;
AGD Type B
Type C NIFID
FTDP-17 NIFID, neurofilament intermediate inclusion
Type D BIBD)
disease; BIBD, basophilic inclusion bodies dementia;
FTLD-Ups, frontotemporal lobar degeneration
MAPT gene C90RF72 VCP PGRN FUS CHMP2B with ubiquitine-positive inclusions; FTLD-ni,
gene gene gene gene gene frontotemporal lobar degeneration with no inclusion.
Patients with tau-positive pathology tend to be older than those al., 2004b) and stereotypical movements with sertraline (Mendez
without tau pathology (Hodges et al., 2004). et al., 2005a).
Most of the remaining cases are tau-negative and The second randomized study was of 300 mg/day of trazadone,
ubiquitin-positive, and have inclusions comprising the 43 kDa also in a 6-week crossover trial, in 26 patients (mean MMSE score
TAR-DNA-binding protein FTLD-TDP (Neumann et al., 2006). 20.8 + 8.3, mean NPI score 53 + 17.9 at baseline) (Lebert et al.,
But a minority (5–10%) are negative for both tau and TDP-43. In 2004; Lebert, 2006). Trazodone is an atypical serotonergic agent
these cases, usually younger with an age at onset under 45 years and with moderate serotonin reuptake inhibition and a serotonergic
characterized by severe atrophy of the caudate nucleus but no fam- (5-HT1A, 5-HT1C, and 5-HT2) antagonist effect with an active
ily history, inclusions of the RNA-binding protein FUS have been metabolite (metachlorophenylpiperazine (m-CPP)) which is a
identified (Neumann et al., 2009). direct serotonin receptor agonist. It is also an adrenergic alpha-1
A major topic of investigation has been to establish the associa- (postsynaptic), alpha-2 (presynaptic), and H1 histaminic blocker.
tion between clinical phenotypes and molecular pathology. Unlike There was a large and significant decrease of the NPI score with tra-
the progressive aphasic syndromes, which are generally associated zodone (mean 18.22 versus 1.1 points with placebo). A decrease of
more with one histological form of FTLD than another (especially more than 50% in the NPI score was observed in 10 patients on tra-
SD with FTLD-TDP), in bvFTD any of the histological variants can zodone. The improvement was mainly based on the improvement
be found, with almost as many cases with tau pathology as cases in four items of the NPI scale—eating disorders, agitation, irritabil-
with TDP-43 pathology, and a small proportion of FTLD-FUS ity, and depression/dysphoria. There was no change on the MMSE
cases (Josephs et al., 2011). score and trazadone was well tolerated. The placebo-controlled
trial was followed by an open-label extension study for more than
Neurochemistry 2 years (116 weeks) of 300 mg/day trazadone (Lebert, 2006). The
mean NPI score between baseline and final assessment decreased
The neurochemical pathology of FTD is discussed in Chapter 7, but
by 20.5 + 9.5 points and no patient had an increase in final NPI
the main neurobiological change reported in FTD is a serotonergic
score. This indicates that trazodone may be effective in reducing
deficit and this appears to be more postsynaptic than presynaptic
behavioural symptoms in FTD in the long term, and the magni-
(Sparks and Markesbery, 1991; Procter et al., 1999; Franceschi et al.,
tude of efficacy from trazodone may be higher than that from other
2005). Low brain levels of serotonin are associated with symptoms
serotonergic agents. Delusions, aggression, anxiety, and irritabil-
of anxiety, depression, quarrelsome behaviours, and impulsive
ity decreased significantly with 150 mg/day of trazodone; 300 mg
aggression (aan het Rot et al., 2006; Wrase et al., 2006).
was necessary to decrease depression, disinhibition, and aberrant
There is clinical evidence of basal ganglia dopamine dysfunction
motor behaviour (Lebert and Pasquier, 1999).
in some FTD (extrapyramidal symptoms), and a PET study has
Since trazodone differs from other serotonin agents in two
shown decreased presynaptic dopamine transporter in the puta-
aspects—the 5-HT2a antagonist effect and an agonist effect related
men and caudate of FTD patients.
to its metabolite, the m-CPP, and the 5-HT2a receptors are mainly
In contrast to AD, the cholinergic system is spared, with intact
present in frontal regions and decreased in FTD—drugs with 5-HT2a
acetylcholine levels, cortical choline acetyltransferase, postsynaptic
receptor actions have a good rationale for pharmacologic research in
muscarinic receptor binding, and preservation of neurons in the
FTD (Lebert, 2006). A meta-analysis of antidepressant medication in
nucleus basalis of Meynert (Huey et al., 2006).
FTD patients has been performed (Huey et al., 2006) and consistent
with these data concluded that serotonergic treatments improve the
Treatments behavioural but not the cognitive symptoms in FTD.
The paucity of pharmacologic trials for FTD is probably due to a
combination of the illness only recently being clinically defined Other treatments
and the failure to recognize FTD patients. Currently, the treatments A within-subjects, double-blind, placebo-controlled procedure
used for FTD are based on medications used to treat the behav- investigated the effects of a single dose of methylphenidate (40
ioural symptoms in other conditions. This strategy has limitations, mg) and showed that it ‘normalized’ the decision-making behav-
since different physiological processes may underlie similar clinical iour of patients such as they became less risk taking on medica-
symptoms. tion, although there were no significant effects on other aspects of
cognitive function. Moreover, there were no normal subjective and
Serotonergic treatments autonomic responses to methylphenidate seen in older subjects
Serotonergic treatments have been most studied, but only two (Rahman et al., 2006). A case report showed a significant behav-
double-blind randomized controlled trials are available. The first, a ioural improvement, whereas at the same time slowing on quantita-
crossover study of 10 patients, reported that treatment with 40 mg tive EEG normalized partially (Goforth et al., 2004).
of paroxetine for 6 weeks did not improve behavioural symptoms Selegiline (Moretti et al., 2002) and moclobemide (Adler et al.,
and was associated with a worsening of some memory tasks com- 2003) improved affect, behaviour, and speech in open-label tri-
pared to placebo (Deakin et al., 2004). The worsened cognition may als in a small number of subjects, but patients with FTD treated
reflect the anticholinergic effects of higher dose paroxetine (Huey with cholinesterase inhibitors showed an increase in aggressiveness
et al., 2006). Most open-label studies of selective serotonin reuptake and agitation (Moretti et al., 2003), and hypersensitivity of FTD
inhibitors (SSRIs), however, have showed an improvement of patients to neuroleptics has been reported (Pijnenburg et al., 2003).
behaviour (Swartz et al., 1997a; ; Lebert and Pasquier, 1999; Chow, Memantine was also tested in bvFTD patients but did not show
2003; Ikeda et al., 2004b). SSRIs have also been used in open trials, any significant effects, except on the Frontal Behavioural Inventory
with benefit on stereotyped behaviours with fluvoxamine (Ikeda et score (Vercelletto et al., 2011).
Nonpharmacological treatment Franceschi, M., et al. (2005). Glucose metabolism and serotonin receptors in
the frontotemporal lobe degeneration. Annals of Neurology, 57, 216–25.
Language disorders need intervention from speech therapists, Garcin, B., et al. (2009). Determinants of survival in behavioural variant
and they may also be able to help with difficulties in swallowing. frontotemporal dementia. Neurology, 73 (20), 1656–61
Information to caregivers, support groups, and social support Geroldi, C., et al. (2000). Pop music and frontotemporal dementia. Neurology,
are necessary. Patients should be advised not to drive and efforts 55, 1935–6.
should be made to prevent hyperphagia and excessive alcohol Gislason, T.B., et al. (2003). The prevalence of frontal variant frontotemporal
consumption. dementia and the frontal lobe syndrome in a population based sample
of 85 year olds. Journal of Neurology, Neurosurgery and Psychiatry, 74,
867–71.
References Goforth, H.W., et al. (2004). Quantitative electroencephalography in
aan het Rot, M., et al. (2006). Social behaviour and mood in everyday life: the frontotemporal dementia with methylphenidate response: a case study.
effects of tryptophan in quarrelsome individuals. Journal of Psychiatry Clinical EEG and Neuroscience, 35, 108–11.
and Neuroscience, 31, 253–62. Gorno-Tempini, M.L., et al. (2011). Classification of primary progressive
Adler, G., Teufel, M. and Drach, L.M. (2003). Pharmacological treatment of aphasia and its variant. Neurology, 76 (11), 1006–14.
frontotemporal dementia: treatment response to the MAO-A inhibitor Graham, A., et al. (2005). Pathologically proven frontotemporal dementia
moclobemide. International Journal of Geriatric Psychiatry, 18, 653–5. presenting with severe amnesia. Brain, 138, 597–605.
Alzheimer, A. (1910–1911). Über eigenartige Krankheitsfälle des späteren Grasbeck, A., et al. (2005). Dementia in first-degree relatives of patients with
Alters. Zeitschrift fűr die gesamte Neurologie und Psychiatrie, 4, 356–85. frontotemporal dementia. A family history study. Dementia and Geriatric
Baker, M., et al. (2006). Mutations in proganulin cause tau-negative Cognitive Disorders, 19 (2–3), 145–53.
frontotemporal dementia linked to chromosome 17. Nature, 442, Gregory, C.A., Serra-Mestres, J. and Hodges, J.R. (1999). Early diagnosis
916–19. of the frontal variant of frontotemporal dementia: how sensitive are
Barber, R., Snowden, J.S. and Craufurd, D. (1995). Frontotemporal dementia standard neuroimaging and neuropsychologic tests? Neuropsychiatry,
and Alzheimer’s disease: retrospective differentiation using information Neuropsychology, and Behavioral Neurology, 12, 128–35.
from informants. Journal of Neurology, Neurosurgery and Psychiatry, 59, Gregory, C., et al. (2002). Theory of mind in patients with frontal variant
61–70. frontotemporal dementia and Alzheimer’s disease: theoretical and
Bonner, L.T., et al. (2003). Familial dementia with Lewy bodies with an practical implications. Brain, 125, 752–64.
atypical clinical presentation. Journal of Geriatrics Psychiatry and Grossman, M., et al. (2005). Cerebrospinal fluid profile in frontotemporal
Neurology, 16, 59–64. dementia and Alzheimer’s disease. Annals of Neurology, 57, 721–9.
Borroni, B., et al. (2010). Is frontotemporal lobar degeneration a rare disorder? Hampel, H. and Teipel, S.J. (2004). Total and phosphorylated tau proteins:
Evidence from a preliminary study in Brescia county, Italy. Journal of evaluation as core biomarker candidates in frontotemporal dementia.
Alzheimer’s Disease, 19, 111–16. Dementia and Geriatric Cognitive Disorders, 17, 350–4.
Bozeat, S., et al. (2000a). Which neuropsychiatric and behavioural features Harvey, R.J., Skelton-Robinson, M. and Rossor, M.N. (2003). The prevalence
distinguish frontal and temporal variants of frontotemporal dementia and causes of dementia in people under the age of 65 years. Journal of
from Alzheimer’s disease. Journal of Neurology, Neurosurgery and Neurology, Neurosurgery and Psychiatry, 74, 1206–9.
Psychiatry, 69, 178–86. Hirono, N., et al. (1999). Distinctive neurobehavioral features among
Bozeat, S., et al. (2000b). Non-verbal semantic impairment in semantic neurodegenerative dementias. Journal of Neuropsychiatry and Clinical
dementia. Neuropsychologia, 38, 1207–15. Neurosciences, 11, 498–503.
Brun, A., et al. (1994). Clinical and neuropathological criteria for Hodges, J.R. and Graham, K.S. (1998). A reversal of the temporal gradient
frontotemporal dementia. Journal of Neurology, Neurosurgery and of for famous person knowledge in semantic dementia: implications for
Psychiatry, 57, 416–18. the neural organisation of long-term memory. Neuropsychologia, 36,
Chow, T.W. (2003). Frontotemporal dementias: Clinical features and 803–25.
management. Seminars in Clinical Neuropsychiatry, 8, 58–70. Hodges, J.R., et al. (1992). Semantic dementia: progressive fluent aphasia
Cruts, M., et al. (2006). Null mutations in progranulin cause ubiquitin-positive with temporal lobe atrophy. Brain, 115, 1783–806.
frontotemproal dementia linked to chromosome 17q21. Nature, Hodges, J.R., et al. (2004). Clinicopathological correlates in frontotemporal
442(7105), 916–19. dementia. Annals of Neurology, 56, 399–406.
Davies, R.R., et al. (2006). Progression in frontotemporal dementia: Hornberger, M., et al. (2009). Can progressive and non-progressive
identifying a benign behavioral variant by magnetic resonance imaging. behavioural variant frontotemporal dementia be distinguished at first
Archives of Neurology, 63, 1627–31. presentation? Journal of Neurology, Neurosurgery and Psychiatry, 80,
Deakin, J.B., et al. (2004). Paroxetine does not improve symptoms and impairs 591–3.
cognition in frontotemporal dementia: a double-blind randomized Hornberger, M., et al. (2010). How preserved is episodic memory in
controlled trial. Psychopharmacology (Berlin), 172, 400–8. behavioral variant frontotemporal dementia? Neurology, 74, 472–9.
DeJesus-Hernandez, M., et al. (2011). Expanded GGGCC hexanucleotide Hosler, B.A., et al. (2000). Linkage of familial amyotrophic lateral sclerosis
repeat in non coding region of C9ORF72 causes chromosome 9p-linked with frontotemporal dementia to chromosome 9q21-q22. Journal of the
FTD and ALS. Neuron, 72, 245–56. American Medical Association, 284, 1664–9.
Engler, H., et al. (2008). In vivo amyloid imaging with PET in frontotemporal Hou, C.E., Miller, B.L. and Kramer, J.H. (2005). Patterns of autobiographical
dementia. European Journal of Nuclear Medicine and Molecular Imaging, memory loss in dementia. International Journal of Geriatric Psychiatry,
35, 100–6. 20, 809–15.
Fernandez-Duque, D. and Black, S.E. (2005). Impaired recognition of Huey, E.D., Putman, K. and Grafman, J. (2006). A systematic review of
negative facial emotions in patients with frontotemporal dementia. neurotransmitter deficits and treatments in frontotemporal dementia.
Neuropsychologia, 43, 1673–87. Neurology, 66, 17–22.
Finch, N., et al. (2009). Plasma progranulin levels predict progranulin Ikeda, M., Ishikawa, T. and Tanabe, H. (2004a). Epidemiology of
mutation status in frontotemporal dementia patients and asymptomatic frontotemporal lobar degeneration. Dementia and Geriatric Cognitive
family members. Brain, 132, 583–91. Disorders, 17, 265–8.
Ikeda, M., et al. (2004b). Efficacy of fluvoxamine as a treatment for behavioral Mendez, M.F., et al. (1993). Pick’s disease versus Alzheimer’s disease: A
symptoms in frontotemporal lobar degeneration patients. Dementia and comparison of clinical characteristics. Neurology, 43, 289–92.
Geriatric Cognitive Disorders, 17, 117–21. Mendez, M.F., et al. (1998). Behavioral differences between frontotemporal
Jeong, Y., et al. (2005). Pattern of glucose hypometabolism in frontotemporal dementia and Alzheimer’s disease: a comparison on the BEHAVE-AD
dementia with motor neuron disease. Neurology, 64, 734–6. rating scale. International Psychogeriatrics, 10, 155–62.
Johnson, J.K., et al. (2005). Frontotemporal lobar degeneration: demographic Mendez, M.F., Shapira, J.S., and Miller, B.L. (2005a). Stereotypical movements
characteristics of 353 patients. Archives of Neurology, 62, 925–30. and frontotemporal dementia. Movement Disorders, 20, 742–5.
Josephs, K.A., et al. (2011). Neuropathological background of phenotypical Mendez, M.F., et al. (2005b). Acquired sociopathy and frontotemporal
variability in frontotemporal dementia. Acta Neuropathologica, 122, dementia. Dementia and Geriatric Cognitive Disorders, 20, 99–104.
137–53. Mercy, L., et al. (2008). Incidence of early-onset dementias in Cambridgeshire,
Joubert, S., et al. (2004). Progressive prosopagnosia. Clinical and neuroimaging United Kingdom. Neurology, 71, 1496–9.
results. Neurology, 63, 1962–5. Miller, B.L., et al. (1997). Aggressive, socially disruptive and antisocial
Keane, J., et al. (2002). Face and emotion processing in frontal variant behaviour associated with fronto-temporal dementia. British Journal of
frontotemporal dementia. Neuropsychologia, 40, 655–65. Psychiatry, 170, 150–5.
Kertesz, A., Davidson, W., and Fox, H. (1997). Frontal Behavioral Inventory: Miller, B.L., et al. (2001). Neuroanatomy of the self: evidence from patients
diagnostic criteria for frontal lobe dementia. Canadian Journal of with fronto-temporal dementia. Neurology, 57, 817–21.
Neurological Sciences, 24, 29–36. Mioshi, E., et al. (2010). Clinical staging and disease progression in
Kertesz, A., et al. (2000). The Frontal Behavioral Inventory in the differential frontotemporal dementia. Neurology, 74, 1591–7.
diagnosis of frontotemporal dementia. Journal of the International Moretti, R., et al. (2002). Effects of selegiline on fronto-temporal dementia:
Neuropsychological Society, 6, 460–8. a neuropsychological evaluation. International Journal of Geriatric
Kipps, C.M., et al. (2007). Behavioural variant frontotemporal dementia: not Psychiatry, 17, 391–2.
all it seems? Neurocase, 13, 237–47. Moretti, R., et al. (2003). Frontotemporal dementia: paroxetine as a possible
Kipps, C.M., et al. (2009). Combined magnetic resonance imaging and treatment of behavior symptoms. A randomized, controlled, open
positron emission tomography brain imaging in behavioural variant 14-month study. European Neurology, 49, 13–19.
frontotemporal degeneration: refining the clinical phenotype. Brain, Mychnack, P., et al. (2001). The influence of right frontotemporal dysfunction
132, 2566–78. on social behavior in frontotemporal dementia. Neurology, 56 (Suppl 4),
Knopman, D.S., et al. (2004). The incidence of frontotemporal lobar S11–15.
degeneration in Rochester, Minnesota, 1990 through 1994. Neurology, Neary, D., et al. (1998). Frontotemporal lobar degeneration. A consensus on
62, 506–8. clinical diagnostic criteria. Neurology, 51, 1546–54.
Knopman, D.S., et al. (2005). Antemortem diagnosis of frontotemporal lobar Neumann, M., et al. (2006). Ubiquitinated TDP-43 in frontotemporal lobar
degeneration. Annals of Neurology, 57, 480–8. degeneration and amyotrophic lateral sclerosis. Science, 314, 130–3.
Kril, J.J., et al. (2005). Distribution of brain atrophy in behavioral variant Neumann, M., et al. (2009). A new subtype of frontotemporal lobar
frontotemporal dementia. Journal of the Neurological Sciences, 232, degeneration with FUS pathology. Brain, 132, 1282–93.
83–90. Nyatsanza, S., et al. (2003). A study of stereotypic behaviours in Alzheimer’s
Lavenu, I. and Pasquier, F. (2005). Perception of emotion on faces in disease and frontal and temporal variant frontotemporal dementia.
frontotemporal dementia and Alzheimer’s disease. A longitudinal study. Journal of Neurology, Neurosurgery and Psychiatry, 74, 1398–402.
Dementia and Geriatric Cognitive Disorders, 19, 37–41. Pasquier, F. (2005). Telling the difference between frontotemporal
Lavenu, I., et al. (1999). Perception of emotion in frontotemporal dementia dementia and Alzheimer’s disease. Current Opinion in Psychiatry, 18,
and in Alzheimer’s disease. Alzheimer’s Disease and Associated Disorders, 628–32.
13, 96–101. Pasquier, F. and Petit, H. (1997). Frontotemporal dementia: its rediscovery.
Lebert, F. (2006). Behavioral benefits of trazodone are sustained for the long European Neurology, 38, 1–6.
term in frontotemporal dementia. Therapy, 3, 93–6. Pasquier, F., et al. (1997). The use of SPECT in a multidisciplinary memory
Lebert, F. and Pasquier, F. (1999). Trazodone in the treatment of behaviour clinic. Dementia and Geriatric Cognitive Disorders, 8, 85–91.
in frontotemporal dementia. Human Psychopharmacology—Clinical and Pasquier, F., Lebert, F. and Petit, H. (1999a). Organisation des centres de la
Experimental, 14, 279–81. mémoire et perspectives. Revue Neurologie (Paris), 155, 4S83–4S92.
Levy, M.L., et al. (1996). Alzheimer disease and frontotemporal dementias. Pasquier, F., et al. (1999b). The clinical picture of frontotemporal dementia:
Behavioral distinctions. Archives of Neurology, 53, 687–90. diagnosis and follow-up. Dementia and Geriatric Cognitive Disorders, 10
Lebert, F., et al. (1998). Frontotemporal behavioural scale. Alzheimer’s Disease (Suppl 1), 10–14.
and Associated Disorders, 12, 335–9. Pasquier, F., et al. (2001). Memory impairment differs in frontotemporal
Lebert, F., et al. (2004). Frontotemporal dementia: a randomised, controlled dementia and Alzheimer’s disease. Neurocase, 7, 161–71.
trial with trazodone. Dementia and Geriatric Cognitive Disorders, 17, Pasquier, F., Richard, F. and Lebert, F. (2004). Natural history of frontotemporal
355–9. dementia: comparison with Alzheimer’s disease. Dementia and Geriatric
Liu, W., et al. (2004). Behavioral disorders in the frontal and temporal variants Cognitive Disorders, 17, 253–7.
of frontotemporal dementia. Neurology, 62, 742–8. Perry, R.J. and Hodges, J.R. (2000). Differentiating frontal and temporal
Lough, S., et al. (2006). Social reasoning, emotion and empathy in variant frontotemporal dementia from Alzheimer’s disease. Neurology,
frontotemporal dementia. Neuropsychologia, 44, 950–8. 54, 2277–84.
Marczinski, C.A., Davidson, W. and Kertesz, A. (2004). A longitudinal study Pick, A. (1892). Über die Beziehungen der senilen Hirnatrophie zur Aphasie.
of behavior in frontotemporal dementia and primary progressive aphasia. Prager Medizinische Wochenschrift, 17, 165–7.
Cognitive and Behavioral Neurology, 17, 185–90. Pick, A. (1901a). Uber Symptomenkomplexe bedingt durch die Kombination
Mattis, S. (1976). Mental status examination for organic mental syndrome sub-corticaler Herdaffekionen mit seniler Hirnatrophie. Wiener Klinische
in the elderly patients. In: Bellak, L. and Karasu, T. (eds) Geriatrics Wochenschrift, 1121.
psychiatry: a handbook for psychiatrists and primary care physicians. Pick, A. (1901b). Senile Hirnatrophie als Grundlage von Hederscheinungen.
Grune & Stratton, New York, pp. 77–121. Wiener Klinische Wochenschrift, 14, 403–4.
Mendez, M.F. and Shapira, J.S. (2009). Altered emotional morality in Pick, A. (1904). Zur Symptomatologie der linksseitigen Schlafenlappenatrophie.
frontotemporal dementia. Cognitive Neuropsychiatry, 14, 165–79. Monatschrift für Psychiatrie und Neurologie, 16, 378–88.
Pick, A. (1906). Über einen weiteren Symptomencomplex im Rahmen der Snowden, J.S., Goulding, P.J., and Neary, D. (1989). Semantic dementia:
Demenita Senilis, bedingt durch umschriebene starkere Hirnatrophie a form of circumscribed cerebral atrophy. Behavioural Neurology, 2,
(gemischte Apraxie). Monatschrift für Psychiatrie und Neurologie, 19, 167–82.
97–108. Snowden, J.S., Griffiths, H.L. and Neary, D. (1996). Semantic-episodic
Piguet, O., et al. (2011). Eating and hypothalamus changes in memory interactions in semantic dementia: implications for retrograde
behavioural-variant frontotemporal dementia. Annals of Neurology, 69, memory function. Cognitive Neuropsychology, 13, 1101–37.
312–19. Snowden, J.S., et al. (2001). Distinct behavioural profiles in frontotemporal
Pijnenburg, Y.A., et al. (2003). Vulnerability to neuroleptic side effects in dementia and semantic dementia. Journal of Neurology, Neurosurgery
frontotemporal lobar degeneration. International Journal of Geriatric and Psychiatry, 70, 323–32.
Psychiatry, 18, 67–72. Snowden, J.S., et al. (2003). Social cognition in frontotemporal dementia and
Piolino, P., et al. (2003). Autobiographical memory and autonoetic Huntington’s disease. Neuropsychologia, 41, 688–701.
consciousness: triple dissociation in neurodegenerative diseases. Brain, Snowden, J.S., Thompson, J.C., and Neary, D. (2004). Knowledge of famous
126, 2203–19. faces and names in semantic dementia. Brain, 127, 860–72.
Procter, A.W., Qurne, M. and Francis, P.T. (1999). Neurochemical features of Snowden, J.S., et al. (2008). Emotion recognition in Huntington’s disease and
frontotemporal dementia. Dementia and Geriatric Cognitive Disorders, frontotemporal dementia.
10 Suppl 1, 80–4. Sparks, D.L. and Markesbery, W.R. (1991). Altered serotoninergic and
Rahman, S., et al. (2006). Methylphenidate (‘Ritalin’) can ameliorate abnormal cholinergic synaptic markers in Pick’s disease. Archives of Neurology, 48,
risk-taking behavior in the frontal variant of frontotemporal dementia. 796–9.
Neuropsychopharmacology, 31, 651–8. Stevens, M., et al. (1998). Familial aggregation in frontotemporal dementia.
Rankin, K.P., Kramer, J.H. and Miller, B.L. (2005a). Patterns of cognitive and Neurology, 50, 1541–5.
emotional empathy in frontotemporal lobar degeneration. Cognitive and Swartz, J.R., et al. (1997a). Frontotemporal dementia: treatment response to
Behavioral Neurology, 18, 28–36. serotonin selective reuptake inhibitors. Journal of Clinical Psychiatry, 58,
Rankin, K.P., et al. (2005b). Self awareness and personality change in 212–16.
dementia. Journal of Neurology Neurosurgery and Psychiatry, 76, Swartz, J.R., et al. (1997b). Behavioral phenomenology in Alzheimer’s disease,
632–9. frontotemporal dementia, and late-life depression: a retrospective
Rascovsky, K., et al. (2002). Cognitive profiles differ in autopsy-confirmed analysis. Journal of Geriatrics Psychiatry and Neurology, 10, 67–74.
frontotemporal dementia and AD. Neurology, 58, 1801–8. Verbeek, M.M., et al. (2005). Cerebrospinal fluid tau levels in frontotemporal
Rascovsky, K., et al. (2011). Sensitivity of revised diagnostic criteria for dementia. Annals of Neurology, 58, 657–7.
the bevavioural variant of frontotemporal dementia . Brain , 134, Vercelletto, M., et al. (2011). Memantine in behavioural variant
2456–77. frontotemporal dementia: neagtive results. Journal of Alzheimers
Ratnavalli, E., et al. (2002). The prevalence of frontotemporal dementia. Disease, 23, 749–59.
Neurology, 58, 1615–21. Whitwell, J.L., et al. (2005). Magnetic resonance imaging signatures of
Roberson, E.D., et al. (2005). Frontotemporal dementia progresses to death tissue pathology in frontotemporal dementia. Archives of Neurology, 62,
faster than Alzheimer disease. Neurology, 65, 719–25. 1402–8.
Rohrer, J.D., et al. (2009). The heritability and genetics of frontotemporal Wicklund, A.H., Johnson, N. and Weintraub, S. (2004). Preservation of
lobar degeneration. Neurology, 73, 1451–6. reasoning in primary progressive aphasia: further differentiation from
Rosen, H.J., et al. (2002). Emotion comprehension in the temporal variant of Alzheimer’s disease and the behavioral presentation of frontotemporal
frontotemporal dementia. Brain, 125, 2286–95. dementia. Journal of Clinical and Experimental Neuropsychology, 26,
Rosso, S.M., et al. (2003). Frontotemporal dementia in The Netherlands: 347–55.
patients characteristics and prevalence estimates from a population-based Wilhelmsen, K.C., et al. (1994). Localization of desihibition-dementia-parki
study. Brain, 126, 2016–22. nsonism-amyotrophy complex to 17q21–22. American Journal of Human
Schroeter, M.L., et al. (2007). Towards a nosology for frontotemporal lobar Genetics, 55, 1159–65.
degenerations-a meta-analysis involving 267 subjects. Neuroimage, 36, Wrase, J., et al. (2006). Serotonergic dysfunction: brain imaging and
497–510. behavioral correlates. Cognition and Affective Behavioral Neurosciences,
Seelaar, H., et al. (2008). Distinct genetic forms of frontotemporal dementia. 6, 53–61.
Neurology, 71, 1220–6. Zhou, J., et al. (2010). Divergent network connectivity changes in behavioural
Seeley, W.W., et al. (2005). The natural history of temporal variant variant frontotemporal dementia and Alzheimer’s disease. Brain, 133,
frontotemporal dementia. Neurology, 64, 1384–90. 1352–67.
CHAPTER 37
Neurological dementias
Andrew Graham
Dementia with onset in old age is usually due to neurodegenerative Hydrocephalus is an umbrella term for the abnormal accumu-
disease. Alzheimer pathology is the commonest cause, followed lation of CSF within the cranium, whether due to defective CSF
by Lewy body changes and, more rarely, the range of pathologies production, flow or absorption. The two main types are obstruc-
associated with frontotemporal dementia (FTD). These neurode- tive hydrocephalus, where there is a structural barrier to CSF flow
generative conditions, in which dementia is the dominant feature, through the ventricular system, and communicating hydrocephalus,
are often referred to as the primary dementias and are covered sep- where the defect is inadequate CSF absorption by the arachnoid
arately in Chapters 33, 35, and 36. Unfortunately, despite decades of granulations. Hydrocephalus secondary to overproduction of CSF
research, the primary dementias are currently incurable. is rare. Communicating hydrocephalus may in turn be classified as
Less commonly, dementia can arise as a complication of general either primary or secondary. In secondary communicating hydro-
medical conditions: in such secondary dementias, cognitive impair- cephalus, the defect in CSF absorption is due to a well-established
ment is typically accompanied by symptoms and signs in other cause of damage to the arachnoid granulations, e.g. meningitis,
organ systems. Most commonly, the underlying pathology is vascu- subarachnoid haemorrhage, or significant head injury. In primary
lar disease: vascular risk factors such as smoking, obesity, diabetes, communicating hydrocephalus, no such cause is apparent.
hypertension, and hypercholesterolaemia are also all risk factors In 1965, Adams and colleagues published a description of three
for dementia, and vascular cognitive impairment and vascular patients with ventriculomegaly, gait disturbance, urinary incon-
dementia are covered separately in Chapter 34. General medical tinence, and dementia in whom symptoms were improved by the
conditions such as hypothyroidism and vitamin B12 deficiency are insertion of a ventricular shunt (Adams et al., 1965). All three
often listed as causes of a secondary dementia and are included in patients had CSF opening pressures within the normal range, and
screening because they are treatable. However, dementia due solely in two of the three cases no secondary underlying cause for the
to such conditions is rare and they are probably better regarded as hydrocephalus could be found. Subsequently, this scenario was
treatable comorbidities rather than ‘causes’ as such. termed ‘idiopathic normal pressure hydrocephalus’ (INPH), but
Alternatively, dementia may develop as just one facet of a more the label may be misleading—intracranial pressure (ICP) in INPH
extensive neurological disorder. Such neurological dementias are is in fact on average slightly higher than normal, with frequent
much less common than primary or vascular dementia but impor- superadded pulses of slightly increased pressure known as B waves.
tant to recognize because their management and outcomes may be For this reason, some authors prefer the alternative label ‘idiopathic
very different. Accurate diagnosis and appropriate treatment often adult hydrocephalus syndrome’.
requires referral outside of old age psychiatry services, and this The pathophysiology of INPH is incompletely understood. CSF
chapter focuses on a selection of neurological dementias where a dynamics are disturbed, with a high resistance to CSF outflow,
neurological consultation is likely to be of value: idiopathic normal slightly raised intracranial pressure, increased B-waves, an increased
pressure hydrocephalus; Huntington’s disease; multiple sclerosis; aqueductal stroke volume, and ventriculomegaly. However, the
autoimmune limbic encephalitis; and prion disease. underlying cause for this disturbance is unknown, as is the mecha-
nism by which these CSF pressure abnormalities affect the brain and
Idiopathic Normal Pressure result in clinical symptoms. A leading theory is that the symptoms
of INPH are due to a complex interaction between a disturbance of
Hydrocephalus (INPH) CSF dynamics and cerebrovascular small vessel disease, resulting in
Overview low-grade cerebral ischaemia of periventricular white matter that can
INPH is a disorder in which abnormal accumulation of cerebros- be seen on PET imaging (Momjian, Owler et al., 2004). Microdialysis
pinal fluid (CSF) results in gait disturbance, urinary incontinence, studies show that CSF removal is associated with improvement in
and dementia. Clinicians are anxious not to miss the diagnosis as this periventricular ischaemia (Agren-Wilsson et al., 2005).
it is potentially treatable, and the possibility of INPH is not infre-
quently raised in neuroimaging reports. However, the disorder in Epidemiology
its full form is uncommon, diagnosis is challenging, and treatment There are no definitive figures for the incidence and prevalence
can be associated with significant complications. of INPH. In two prospective community-based surveys in San
Marino and Bavaria, the population prevalence of INPH in people stenosis. Based on the combination of clinical features and imag-
older than 65 years was 0.5% and 0.4%, respectively (Casmiro et al., ing abnormalities, consensus clinical diagnostic criteria for possible
1989; Trenkwalder et al., 1995)—almost on a par with the preva- and probable INPH have been proposed (Relkin et al., 2005).
lence of Parkinson’s disease (PD). However, other studies suggest a If specialist assessment confirms the diagnosis of INPH, the next
rather lower incidence and prevalence. In a survey of six Swedish step is to decide if surgery (the placement of a ventricular shunt)
neurosurgical centres, the number of shunt insertions performed will be of benefit. This usually involves further invasive investiga-
for INPH was only 1–2/100,000 per year (Tisell et al., 2005), while tions, either to measure aspects of CSF dynamics directly or to sim-
in a retrospective survey of Mayo Clinic records the number was ulate the result of a shunt by removing a volume of CSF. The most
1.2/100,000 per year (Klassen and Ahlskog, 2011). Of course, these widely used investigation is probably the CSF tap test, where 40–50
studies exclude patients with INPH who did not proceed to shunt- ml of CSF is withdrawn by lumbar puncture, with assessment of
ing, but even so these figures suggest that INPH is an uncommon gait and cognition before and afterwards. Unfortunately, while this
cause of dementia in old age. test has a reasonable specificity and positive predictive value (i.e. a
positive result is a good predictor of a response to shunting), it has
Clinical features a low sensitivity and negative predictive value (i.e. a negative result
The diagnosis of INPH rests first of all on clinical suspicion. The does not exclude a response to shunting). For this reason, some
cardinal feature is disturbance of balance and gait, which almost neurosurgical centres advocate a prolonged trial of CSF removal,
always precedes the development of other symptoms and wors- usually involving the placement of a lumbar catheter and the steady
ens insidiously over months and years. Patients walk with a mildly drainage of approximately 10 ml of CSF per hour for up to 72 h.
broad-based, small-stepped gait, and locomotion is impaired to There is certainly some evidence that external lumbar CSF drainage
a degree out of all proportion to the examination findings on the is more sensitive in predicting response to shunting than a simple
couch: this is often termed a gait apraxia. Although minor extrapy- tap test, but drainage has the drawback of requiring an inpatient
ramidal features may be found on examination, the presentation is stay and carries a risk of catheter-related complications, including
quite distinct from that of idiopathic PD. Cognitive impairment is infection.
present in the majority of patients with INPH by the time of diag- Alternatively, CSF dynamics may be measured directly.
nosis, generally with a ‘subcortical’ pattern, including pronounced Twenty-four-hour intracranial pressure monitoring often reveals
slowness of thought, difficulty in sustaining, dividing, and switch- abnormalities that indicate poor cerebral compliance, such as the
ing attention, and difficulties in planning. However, simple bedside presence of low amplitude B waves (as mentioned in the section
screening tests are generally unable to discriminate between the Overview). However, studies suggest that the presence or absence
profiles of cognitive impairment associated with INPH and AD. of B waves has a poor correlation with response to shunting. A
Finally, urinary urgency, frequency, and incontinence are common more comprehensive assessment of CSF dynamics can be obtained
but nonspecific features. by CSF infusion testing, in which saline is infused into the sub-
arachnoid space through a lumbar puncture needle, while CSF
Investigations pressure is continuously recorded via a second lumbar puncture
A CT or MRI brain scan should be performed in any patient with needle placed at a different level. In most patients, the pressure rises
clinical features suggestive of INPH. The first aim of neuroimaging constantly and then reaches a plateau. From the baseline pressure,
is to exclude any structural lesion causing obstructive hydrocepha- plateau pressure, and infusion rate, the resistance to CSF absorp-
lus, and the second is to look for supporting evidence of INPH. tion can be calculated. At present, CSF infusion measurements suf-
Enlargement of the ventricles is the cardinal feature of INPH, and fer from the same problem as drainage studies when it comes to
is traditionally assessed by calculation of Evan’s ratio (the ratio of predicting a response to shunting, namely a good specificity but
the greatest width of the frontal horns of the lateral ventricles to the poorer sensitivity. However, sophisticated analysis of the infusion
maximum internal width of the skull). An Evan’s ratio greater than data and refinements of the technique may improve its sensitivity in
0.3 suggests significant ventriculomegaly, but other features are also the future. Clinical factors and investigation results that may help
important: e.g. disproportionate widening of the temporal horns predict the response to shunting are summarized in Table 37.1.
of the lateral ventricles, or flattening of the cortical sulci superi-
orly. In patients where significant ventriculomegaly is accompanied Treatment and outcome
by marked cortical atrophy or signs of extensive cerebrovascular The surgical treatment for INPH is placement of a ventriculo-
disease, ventriculomegaly may simply represent loss of brain tissue peritoneal (VP) shunt. The range of available shunt systems is wide
rather than INPH. and preferences vary from centre to centre. Shunts are coupled to a
If there is supporting evidence for INPH on neuroimaging, the valve, which may be flow-regulated or differential pressure-regulated,
next step is typically specialist referral to either neurology or neu- and current devices generally allow the reading and adjustment of
rosurgical services. Before proceeding to more invasive tests and valve settings after implantation. Data from the UK Shunt Registry
a decision about surgery, it is important to exclude degenerative do not show any significant differences in the shunt revision rates
disorders such as PD; significant vascular disease; and common between the various valve systems, although the management of
neurological disorders of old age such as cervical spondylotic mye- complications such as overdrainage (causing a low-pressure subdural
lopathy, lumbar canal stenosis, and a polyneuropathy. Exclusion of haematoma) is made much easier by an adjustable valve (Zemack
these possibilities rests largely on clinical assessment, but occasion- and Romner, 2002). Typical preoperative and postoperative CT scan
ally spinal imaging or neurophysiology studies may be necessary. appearances in INPH are shown in Fig. 37.1.
Any MRI of the brain should also be carefully examined to rule The nature, magnitude, and duration of the benefit obtained by
out subtle secondary causes of hydrocephalus such as aqueduct shunting is controversial. In nonplacebo controlled studies, up
CHAPTER 37 neurological dementias 493
Table 37.1 Factors associated with response or lack of response to shunting in INPH
Factors associated with response to shunting Factors associated with a lack of response to shunting
◆ Gait disturbance preceding cognitive impairment ◆ Cognitive impairment preceding gait disturbance
◆ Mild cognitive impairment, present for a short duration ◆ Moderate or severe cognitive impairment, present for two or more years
◆ Clinical improvement (usually in gait) following lumbar puncture or continuous ◆ MRI shows significant cerebral atrophy or white matter abnormalities
lumbar CSF drainage
◆ Resistance to CSF outflow of 18 mmHg/ml per min or greater during continuous
lumbar CSF infusion test
◆ Presence of B waves more than 50% of the time during continuous lumbar CSF
monitoring
Fig. 37.1 Preoperative (left) and postoperative

(right) axial CT brain scans in INPH. Preoperatively,
the ventricles are widened (upper left) and the
cortical sulci in the top slices are partially effaced
(lower left). Evan’s ratio, calculated as the ratio of
the greatest width of the frontal horns of the lateral
ventricles (A) to the maximal internal diameter of
the skull (B), is approximately 0.4; a ratio more than
0.3 is significant. Postoperatively, observe the tip of
the shunt catheter in the lateral ventricle (upper
right) and traversing the brain parenchyma (lower
right), with a reduction in ventricular size and
widening of cortical sulci.
(Reproduced with permission from Malm and Eklund 2006.)
to two-thirds of patients demonstrate significant initial improve- et al., 2004; Klassen and Ahlskog, 2011), particularly when cog-
ment after shunting (Malm et al., 2000; Hebb and Cusimano, nitive impairment is severe or long-standing. One reason for this
2001), although only up to a third show sustained benefit at 3 years lack of response may be additional AD pathology, which of course
(Malm et al., 2000; Klassen and Ahlskog, 2011). Unfortunately, out is very common with older age even in the ‘normal’ population.
of the key triad of impairments in INPH, dementia is the feature However, benefits to gait appear to be independent of the pres-
least likely to respond to shunting (Savolainen et al., 2002; Poca ence or absence of AD pathology, and so, provided that patients
and carers appreciate the limitations of surgical treatment, shunt- disorder is hyperkinetic, with chorea or dystonia, later progressing
ing for gait improvement alone may still be perfectly appropriate. to an akinetic-rigid state. However, some patients with late-onset
Improvement also appears to be independent of age (Marmarou HD may present with an akinetic-rigid syndrome from the outset,
et al., 2005). There are no placebo-controlled trials of shunting causing diagnostic confusion (Reuter et al., 2000). Other physi-
for INPH recorded in the UK Cochrane Database (Esmonde and cal clues to the diagnosis include oculomotor abnormalities (e.g.
Cooke, 2002—last updated September 2008). slowed voluntary saccades and impersistence of gaze).
HD is also associated with a wide range of psychiatric and cogni-
Key points: INPH tive symptoms. The commonest psychiatric disturbance is depres-
sion, affecting 40–50% of patients with HD during the course
◆ INPH is an uncommon, but potentially reversible cause of
of their illness. The suicide rate in HD is reported to be four to
dementia in old age.
six times higher than in the general population, and in one early
◆ INPH should be excluded in patients with the classical triad of cohort of HD patients, suicide accounted for over 2% of deaths
gait impairment, dementia, and urinary incontinence. (Schoenfeld et al., 1984). Anxiety and obsessive-compulsive symp-
◆ Assessment comprises CT or MRI brain imaging followed by toms are also frequently reported. Psychosis is less common, but
referral to specialist neurology or neurosurgery services. up to 10% of patients show psychotic symptoms at some stage in
their illness, usually poorly systematized paranoia with behavioural
◆ CSF removal or dynamic CSF testing are usually required before
changes such as aggression and irritability. An early age of onset
making a decision on suitability for shunt insertion.
may be associated with an increased risk of psychosis, and psychiat-
◆ Gait impairment is the feature most likely to improve after ric disturbance can occasionally be the presenting feature of HD in
shunting. younger patients (Lovestone et al., 1996). Frank auditory hallucina-
◆ Dementia is less likely to be reversible where cognitive impair- tions of the type seen in schizophrenia are rare.
ment is severe or of long-standing, perhaps because of additional Cognitively, patients with HD show prominent deficits on tests
AD pathology; however, provided the risks are understood, of attention, semantic verbal fluency, processing speed, and execu-
shunting may still be considered for the benefits to gait alone tive function. Performance on tests of memory can be poor, but
the profile of problems (with recall affected more than recognition)
suggests difficulties with retrieval rather than encoding. Cognitive
Huntington’s Disease (HD) impairment can also be detected prior to diagnosis in gene-positive
individuals at risk for the development of HD (Lawrence et al.,
Overview
1998a, 1998b).
HD is an inherited, autosomal dominant, neurodegenerative dis-
order resulting from an unstable expansion of a CAG trinucleotide Investigations
repeat in the gene coding for the protein huntingtin on chromosome MRI in patients with HD may show generalized cortical atrophy or
4. The normal repeat number is between 6 and 35, and expansions specific caudate atrophy (see Fig. 37.2). However, the only definitive
of 40–44 repeats or greater lead to HD with 100% penetrance. The investigation is genetic testing for the CAG expansion. This is a rou-
clinical features of HD usually emerge in early adulthood, but both tine procedure, but because of the significant implications of a posi-
juvenile and late-onset presentations are well recognized. HD is com- tive test for patients and their families, referral to clinical genetics
monly associated with prominent psychiatric symptoms, and when it services for counselling prior to testing is generally recommended.
presents in old age the clinical syndrome may be atypical. The family
history may also be obscured by factors such as death from other
causes in an affected parent or sibling, before clinical features of HD
are manifest; missing or ‘hidden’ relatives with unrecognized HD,
particularly relatives who committed suicide; and nonpaternity. Up
to 10% of cases of HD may represent new mutations. For old age psy-
chiatrists, an appreciation of the wide range of clinical features in HD
is therefore important to avoid a missed or delayed diagnosis. The
only assessment that definitively excludes HD is genetic testing.
Epidemiology
HD is one of the commonest inherited neurodegenerative illnesses,
with a prevalence in most western European countries of between
4 and 8 per 100,000 (Harper, 1986). Lower prevalences are seen
in certain other countries (e.g. Finland) and ethnic groups (e.g.
Afro-Caribbean populations), but essentially HD is found world-
wide. The peak age of onset is between 35 and 44 years, but HD can
present at any age. Both sexes are affected equally.
Clinical features
HD classically presents with the combination of a movement Fig. 37.2 Coronal FLAIR MRI in HD, showing bilateral caudate atrophy (arrows).
disorder and a frontal dementia. Typically, the initial movement (Reproduced with permission from Wild and Tabrizi, 2007.)
Treatment and outcome 50 years) is well recognized, although there is no consensus as to

There is no cure for HD. Neuropsychiatric symptoms may all be whether this refers to age at first symptoms or at diagnosis, and
treated with conventional medications, but many HD patients have onset (however defined) is still usually before the age of 65 years. In
an increased sensitivity to side effects. Longitudinal studies show a retrospective Israeli study of 640 patients cared for at a specialist
a steady decline in cognitive function, with progressive functional MS centre, 30 (4.6%) had their first symptoms after the age of 50
disability and eventual death after an interval ranging from 10–30 (mean 53.5 years, SD 3.1, range 50–62) (Polliack et al., 2001), while
years. a similar single-centre German study identified 52 patients over
a 10-year period aged over 50 years at diagnosis (mean 57 years,
Key points: HD SD 7, range 50–82) (Kis et al., 2008).
However, while MS presenting de novo to old age psychiatry with
◆ HD is an inherited, autosomal dominant, neurodegenerative predominantly cognitive symptoms and onset after the age of 65
condition that most usually presents with the combination of a years is uncommon, two other scenarios are possible. First, patients
frontal dementia and hyperkinetic movement disorder. with established MS may develop dementia in later life, either due
◆ Late-onset HD is well described and may present atypically, e.g. entirely to their MS, additional degenerative disease, or a combina-
as an akinetic-rigid syndrome. Sometimes the family history is tion of both. In any of these situations, accurate diagnosis and man-
obscured or lacking. agement is likely to be more challenging. Second, the possibility of
demyelinating disease may be raised on imaging investigations of
◆ HD can only be completely excluded by genetic testing. Because dementia, especially with the sensitivity of MRI for white matter
of the implications of a positive test, it is usually wise to involve abnormalities. Over the age of 50 years, white matter changes are
clinical genetics services prior to making a diagnosis. much more likely to be due to vascular disease than demyelina-
◆ There is no cure for HD, although survival may be prolonged in tion, but once the possibility of MS has been raised, careful clinical
some patients. evaluation may be needed to exclude it.
Clinical features
Multiple Sclerosis (MS) Acute inflammatory/demyelinating attacks in MS may affect any
Overview site within the CNS, giving rise to an enormous breadth of pos-
MS is a chronic, relapsing, and remitting or progressive disease of sible clinical manifestations. The commonest symptoms involve
the central nervous system (CNS) that involves both inflamma- motor, sensory, cerebellar, and visual/oculomotor systems, but
tory/demyelinating and neurodegenerative processes. Symptoms neuropsychiatric and cognitive disturbance are also well described.
are typically motor or sensory and intermittent and unpredictable The commonest neuropsychiatric disturbance is mood disorder
to start with, later becoming progressive and disabling, but the dis- (euphoria, mania, or depression); psychosis is rare, but recognized.
ease course in MS is extremely variable. Most commonly, the initial Cognitively, impairment is generally most marked on tests of atten-
pattern is of attacks affecting different sites in the CNS at different tion, processing speed, and executive function, followed by impair-
times, but with good recovery between attacks (relapsing-remitting ment on tests of verbal memory (recall more so than recognition).
MS). However, after a period of years, recovery after attacks tends Language functions are generally better preserved, and frank apha-
to be less complete, leading to a gradual accumulation of fixed dis- sia secondary to MS is rare. In keeping with the presumed fron-
ability, followed later by steady deterioration irrespective of the fre- tal/subcortical basis of this profile of deficits, a good predictor of
quency or severity of acute attacks (secondarily progressive MS). cognitive impairment in MS is the degree of frontal lobe atrophy
Occasionally, patients follow a steadily progressive course from dis- (Benedict et al., 2002). Cognitive impairment may occasionally be
ease onset (primary progressive MS). It is tempting to assume that the presenting or only manifestation of MS (Zarei, Chandran et al.,
reducing the frequency or severity of acute attacks would retard 2003).
the progression of disability in MS, but unfortunately this is not
necessarily the case. Increasingly, evidence suggests that neurode- Investigations
generation is taking place alongside inflammatory/demyelinating The key investigation in suspected MS is an MRI scan, typically
changes from an early stage (Chard et al., 2002). A primary pro- showing multiple high signal white matter lesions. However, as
gressive course, particularly with motor symptoms, may be more mentioned above, high signal white matter lesions are a common
common in late-onset disease. finding in old age and may be entirely nonspecific or reflect vas-
cular disease. Radiological features that might specifically suggest
Epidemiology demyelination include an ovoid shape to the lesions; involvement
The prevalence of MS varies widely according to geography and of the corpus callosum, brainstem, and cerebellum; and enhance-
ethnic group, with the highest rates in populations of north- ment after contrast when lesions are acute (see Fig. 37.3); diagnos-
ern European descent. In a longitudinal study over 15 years in tic criteria specify the profile of expected MRI abnormalities in MS
Olmstead County, Minnesota, the prevalence of MS was 177 per very closely (McDonald et al., 2001; Polman et al., 2005). Other
100,000 of the adult population (Mayr et al., 2003); in a similar supportive investigation results include unilateral or bilateral delay
longitudinal study in Cambridge, UK, the prevalence was 152 per on visual evoked potentials (VEPs) and positive oligoclonal bands
100,000 (Robertson et al., 1996). MS is the commonest neurologi- in CSF (but not serum) at lumbar puncture. Assessment of sus-
cal disorder of young adults in the UK, with disease onset typically pected MS and access to these secondary investigations will usually
between the ages of 15 and 50 years. Later onset MS (age more than require referral to local neurology services.
A B
Fig. 37.3 MRI findings in MS. (Left) Axial

FLAIR sequence showing multiple high signal
periventricular white matter lesions, some
transversely oriented; (Right) axial T1-weighted
sequence showing that some of these lesions
enhance after gadolinium, indicating active disease.
(Reproduced with permission from Trip and Miller, 2005.)
Treatment and outcome be rare, always paraneoplastic, and resistant to treatment (Brierley
et al., 1960; Corsellis et al., 1968), but studies over the past decade
There is no curative treatment for MS. For acute relapses, a short
suggest that LE associated with autoantibodies is more common
course of corticosteroids may speed recovery but has no impact
than previously thought, is not always associated with cancer, and is
on the development of later disability. For patients with frequent
potentially treatable in a substantial number of patients (Honnorat,
disabling relapses, disease-modifying treatments are now available.
2010).
The beta-interferons and glatiramer acetate reduce relapse rates by
Autoantibodies associated with LE are directed against neuronal
approximately a third; agents in development such as alemtuzumab
antigens that may be either intracellular or lie on the cell surface.
may reduce relapse rates by up to two-thirds. However, once again,
Neurological disorder associated with antibodies directed against
the impact of these treatments on the development of eventual dis-
intracellular antigens is more likely to be accompanied by underly-
ability remains to be shown, and all may be associated with signifi-
ing cancer, the evidence that the antibodies themselves are actually
cant side effects. Given the frontal/subcortical profile of deficits in
pathogenic is weak, and the response to immune treatment is gen-
dementia associated with MS, it might be expected that cognitive
erally disappointing (Voltz, 2002; Dalmau et al., 2008). By contrast,
rehabilitation would be of benefit, but systematic studies in this
neurological syndromes associated with antibodies directed against
area are lacking.
cell surface antigens are less likely to be accompanied by cancer, the
Key points: MS pathogenic role of the antibodies is much clearer, and the response
to immune treatment can be very good. The major advance in the
◆ MS may be associated with a variety of cognitive and psychiatric past decade has been the isolation of antivoltage gated potassium
disturbances including dementia. channel (VGKC) antibodies in patients with nonparaneoplastic
◆ MS is an uncommon cause of dementia in old age, but might LE (Thieben et al., 2004; Vincent et al., 2004), although it is now
be suspected in a patient with a clinically isolated syndrome or appreciated that anti-VGKC antibodies are in fact directed against
established MS in middle life who later goes on to develop a a range of channel-associated proteins such as LGI-1 (leucine-rich,
frontal-subcortical pattern of cognitive impairment. glioma-inactivated protein 1) rather than necessarily against the
channel itself (Irani et al., 2010; Lai et al., 2010). Common autoan-
◆ The key test in suspected MS is an MRI scan, but distinguishing tibodies associated with LE are shown in Table 37.2, together with
between demyelination and vascular disease in old age can be other associated neurological syndromes and underlying cancer(s),
difficult, and specialist assessment and/or secondary investiga- where relevant. As further antibody targets are identified this list
tions (evoked potentials and CSF examination) are likely to be will undoubtedly lengthen.
required for a definite diagnosis.
Epidemiology
Limbic Encephalitis (LE) The incidence and prevalence of LE are unknown. Referral data from
the UK national reference laboratory in Oxford suggest the disorder
Overview is rare, with an incidence of 0.5–1 cases/100,000 per year, but it is
LE is an umbrella term for a range of immune disorders character- highly likely that not all cases are currently ascertained. In a recent
ized by relatively selective inflammation or antibody-mediated dys- prospective study of encephalitis in the UK, antibody-mediated
function of the limbic system. Typical manifestations of LE include encephalitis was diagnosed in 16 of 203 patients, or 1 in 12, making
altered mental status, memory impairment, and seizures, often it half as common as herpes simplex encephalitis (36 cases) and
accompanied by high signal changes in the medial temporal lobes significantly more common than varicella zoster encephalitis (nine
on MRI. When first described in the 1960s, LE was considered to cases) (Granerod et al., 2010). Of the 16 patients, nine were positive
Table 37.2 Antibodies associated with LE

Antibody Commonly associated neurological Commonly associated underlying Response to treatment (immune therapy
syndrome(s) cancers or cancer removal)
Against cell-surface antigens
Anti-VGKC: Typical LE Not usually associated with cancer Excellent if immunotherapy instituted early
LGI-1 subtype Faciobrachial dystonic seizures
Anti-VGKC: Typical LE SCLC Limited, especially if associated with SCLC
CASPR2 subtype Thymoma
Anti-NMDA receptor Psychosis, cognitive and behavioural Ovarian teratomas (younger women) Improvement in up to 60%, but recovery
change, distinct from typical LE SCLC (rare) typically prolonged
Against intracellular antigens

Anti-Hu LE SCLC (> 90%) Poor
Cerebellar ataxia
Subacute sensory neuropathy
Anti-Ma2 LE Testicular germ-cell tumours (younger Good in up to 30% (typically younger men
Brainstem encephalitis men) with testicular germ-cell tumours)
Non-SCLC, breast
Anti-CV2/CRMP5 LE SCLC (> 70%) Poor
Cerebellar ataxia Thymoma
Subacute sensory neuropathy
Anti-amphiphysin LE SCLC Poor
Stiff person syndrome Breast
VGKC, voltage gated potassium channel; LGI-1, leucine-rich, glioma-inactivated protein 1; CASPR2, contactin-associated protein 2; NMDA, N-methyl-D-aspartate; SCLC, small cell lung
cancer.
for anti-NMDA (N-methyl-D-aspartate) antibodies and seven for features or early disease may be normal. In anti-NMDA, recep-
anti-VGKC channel antibodies. The median age at presentation tor encephalitis imaging may be normal, or show predominantly
for patients with anti-VGKC encephalitis is in the seventh decade, extratemporal abnormalities; the findings in paraneoplastic LE
making it entirely possible that such patients could present to old are likewise variable. Typical MRI findings in patients with para-
age psychiatry initially. neoplastic and anti-VGKC antibody-associated LE are illustrated
in Fig. 37.4.
Clinical features CSF examination may show a variety of abnormalities in all forms
The core clinical features of LE are behavioural disturbance, sub- of LE, including a lymphocytosis or pleocytosis, raised protein, and
acute memory impairment, and seizures. When all features are (occasionally) unmatched CSF oligoclonal bands. EEG may con-
present, LE is unlikely to be mistaken for a primary degenerative firm focal cerebral dysfunction or epileptiform activity but is not
dementia, but the diagnosis is more challenging when features are specific. Finally, patients with anti-LGI-1 antibodies often show
present in isolation. Patients may occasionally present with isolated serum hyponatraemia. However, a definitive diagnosis can only be
psychiatric or cognitive disturbance (Parthasarathi et al., 2006), or reached with the appropriate antibody tests.
with a very focal presentation of repetitive, brief, dystonic seizures
involving the face and arm (faciobrachial dystonic seizures), pro-
Treatment and outcome
gressing to cognitive or behavioural disturbance later. Anti-NMDA As indicated in Table 37.2, response to treatment in autoimmune
receptor encephalitis tends to present rather differently to classi- LE very much depends on the associated antibody and whether the
cal LE, with more prominent psychiatric disturbance, followed by a disorder is paraneoplastic. Patients with anti-LGI-1 antibodies and
variety of hyperkinetic movement disorders, and eventually obtun- no underlying cancer respond the best to immunotherapy, which
dation. In paraneoplastic LE associated with antibodies to intra- might include steroids, intravenous immunoglobulin, plasma
cellular antigens, such as Hu, Ma2, and CV2/CRMP-5, there may exchange, and rituximab. However, if treatment is delayed, hip-
be other features besides LE, particularly a sensory neuropathy or pocampal atrophy may develop and clinical deficits become per-
cerebellar ataxia. manent. Seizures in LE tend to be rather resistant to conventional
antiepileptic drug treatment.
Investigations
Key points—LE
On neuroimaging, patients with anti-VGKC antibodies and classi-
cal LE tend to show abnormal high signal in the medial temporal ◆ LE is a condition in which relatively selective inflammation or
lobes on MRI. However, imaging in patients with isolated clinical dysfunction of the limbic system leads to behavioural disturbance,
A B
Fig. 37.4 MRI findings in LE. (A) Axial FLAIR

sequence in paraneoplastic LE showing bilateral
mesial temporal high signal change; (B) coronal
FLAIR sequence in anti voltage gated potassium
channel LE showing high signal change in the
hippocampi bilaterally (arrows).
(Reproduced with permission from Schott, 2006; Dalmau
and Rosenfeld, 2008.)
subacute memory impairment, and seizures, classically associ- valine-valine, VV), and the size and ratio of the PrPSc protein
ated with medial temporal lobe abnormalities on MRI. fragments remaining after limited digestion (Brandner, 2011). The
◆ Anti-VGKC antibody-associated LE due to anti-LGI-1 antibod- commonest form of sporadic prion disease is classical sporadic
ies tends to present with a full spectrum of clinical features, Creutzfeldt–Jakob disease (CJD), which presents with a rapidly
together with abnormalities on imaging, sometimes abnormali- progressive dementia and myoclonus; the two other main forms of
ties on EEG and CSF examination, and often abnormalities of sporadic disease are iatrogenic CJD and variant CJD, which may
serum sodium; it may respond favourably to immune treatment present rather differently.
if instituted early enough. Anti-CASPR-2 VGKC disease is more Iatrogenic CJD arises because although PrPSc is not a conven-
likely to be paraneoplastic and resistant to immunotherapy. tional infectious agent, prion diseases are transmissible from human
to human by exposure to affected body tissues or fluids. Iatrogenic
◆ Anti-NMDA receptor encephalitis presents rather differently to CJD can be contracted from PrPSc-contaminated human-derived
typical LE, with more psychiatric disturbance initially, sometimes growth hormone, dura mater extracts used in neurosurgical pro-
relatively normal investigations, and a very variable course. cedures, corneal grafts, and EEG depth electrodes, and typically
◆ Paraneoplastic LE is associated with a variety of antibodies to presents many years after exposure, usually with a rapidly progres-
intracellular antigens, may be accompanied by other neurologi- sive cerebellar syndrome rather than dementia with myoclonus.
cal syndromes such as sensory neuropathy and cerebellar ataxia, Variant CJD arises because transmission may even take place
and tends to be resistant to treatment. across species, leading to one of the most worrying UK public health
disasters of recent decades. Scrapie, an endemic encephalopathic
illness of sheep and goats in the UK (but not other countries such as
Prion Diseases New Zealand), has long been recognized to be a prion disease but
was never previously thought to be transmissible to humans. Then,
Overview during the 1980s and early 1990s the UK saw an epidemic of a novel
The prion diseases are a group of disorders in which fulminant prion disease of cattle, termed bovine spongiform encephalopathy
spongiform neurodegeneration results in a rapidly progressive and (BSE). The initial cause of the BSE epidemic was likely the trans-
ultimately fatal dementia. The pathogenic process involves conver- mission of abnormal prion protein to cattle when nervous tissue
sion of a normal cell surface protein, termed cellular prion protein from scrapie-affected sheep found its way into cattle feed. Once this
(PrPC), into an abnormally folded and protease-resistant isoform ‘species barrier’ had been jumped, matters were compounded when
(PrPSc). This abnormally folded protein is then deposited in the nervous tissue from affected cattle was, astonishingly, incorporated
brain: whether it is the conversion or the deposition of PrPSc that back into cattle feed again. Eventually the total number of cattle
gives rise to the subsequent neurodegeneration is still a matter for definitely diagnosed with BSE in the UK was nearly 200,000, but
debate. the true number is likely to have been much higher; an additional
Approximately 10–15% of cases are familial, secondary to a vari- four million asymptomatic cattle aged more than 30 months were
ety of genetic abnormalities in the PRNP gene, including point also slaughtered in an effort to halt the epidemic. Despite these
mutations in the C-terminal domain, premature STOP codon measures, it is estimated that over 600,000 cattle affected with BSE
mutations, and the insertion of additional octapeptide repeats entered the human food chain (Anderson et al., 1996).
in the N-terminal domain; different mutations can be associated It was not long before patients with a novel type of CJD were
with very different clinical phenotypes. The remainder of cases are recognized, often with purely psychiatric symptoms in the initial
sporadic. The optimum classification of sporadic prion disease is stages (Bateman et al., 1995; Britton et al., 1995). The mean age of
undecided, with between four and twelve so-called strains identi- onset in variant CJD (vCJD) has to date been much earlier than
fied, depending on clinical phenotype, histopathological appear- for classical sporadic CJD, but old age psychiatrists should not be
ance, the pattern of polymorphisms at codon 129 of the PRNP complacent—of the first 100 patients with vCJD, the oldest was
gene (methionine-methionine, MM; methionine-valine, MV; or aged 74 years (Spencer et al., 2002). One of the key differences
between sporadic CJD and vCJD is that tissues outside the nerv- Investigations
ous system can be affected. The BSE prion protein is hypothesized Subtle imaging abnormalities are increasingly recognized in prion
to enter the body through the gut, and PrPSc deposition can be diseases. Some patients with sporadic CJD show high signal in
seen in lymphoid tissue at a variety of sites, including the appen- the caudate nuclei or cortical ribbon on MRI, while vCJD may be
dix (Hilton et al., 1998) and the tonsils. This also means that associated with high signal in the pulvinar nucleus of the thalamus
vCJD can potentially be transmitted between humans by expo- (see Fig. 37.5). This ‘pulvinar sign’ now forms part of the diagnostic
sure to affected non-nervous tissue, including via blood transfu- criteria for vCJD, although it is not entirely specific and has also
sion (Llewelyn et al., 2004; Peden et al., 2004). To date (February been described in other conditions such as Wernicke–Korsakoff
2013), 176 cases of definite or probable vCJD have been reported syndrome and paraneoplastic LE (Doherty et al., 2002; Mihara
in the UK, of whom all have died (with pathological confirmation et al., 2005).
of the diagnosis in 122 patients) (<http://www.cjd.ed.ac.uk/fig- The EEG is often abnormal, classically with periodic sharp wave
ures.htm>, accessed 15/02/2013). However, since 2005, the annual complexes (PSWCs) (seen in sporadic CJD but not vCJD). At lum-
death rate from vCJD has been low (no more than 5 patients per bar puncture, CSF constituents are normal, with negative oligoclonal
year, compared to a peak of 28 in 2000). Hopefully, despite the bands, but a raised CSF 14–3–3 protein may be found. A compre-
mass exposure of the UK population, a major epidemic of vCJD hensive study of investigation results in 2451 patients with patho-
now looks unlikely. logically confirmed prion disease suggests that the most important
determinants of abnormal results are age at onset, disease duration,
Epidemiology and molecular subtype (Collins et al., 2006). For example, the prob-
Classical sporadic CJD, the commonest form of human prion ability that the EEG will show PSWCs increases with patient age
disease, has an approximate annual incidence of 1–2 per million but decreases with disease duration. Similarly, disease duration of
population. There is little or no variation in incidence across geo- more than 12 months reduces the chance of a positive CSF 14–3–3
graphical boundaries or ethnic groups. The mean age at onset is result. Regarding molecular biology, patients presenting with clas-
approximately 65 years (range 42–91), with a mean duration to sical sporadic CJD and predominant cortical involvement (likely
death of 4 months (range 1–18) (Parchi et al., 1999). MM1 and MV1 subtypes) are more likely to show PSWCs on EEG,
while patients presenting with ataxia and predominant subcortical
Clinical features involvement are more likely to show basal ganglia abnormalities on
Classical sporadic CJD is characterized by relentlessly progressive MRI. Overall, a single negative test cannot exclude prion disease
cognitive decline accompanied by a range of other neurological and combinations of investigations are recommended.
features. The most common presentation is a rapidly progressive The key test in familial prion disease is obviously sequencing of
dementia with myoclonus, but alternative presentations include the PRNP gene. The gold standard in sporadic cases is tissue biopsy
progressive cortical blindness (the Heidenhain variant) and a with demonstration of abnormal accumulations of PrPSc; in clas-
progressive cerebellar syndrome (the Brownell–Oppenheimer sical sporadic CJD, this means brain biopsy, but in vCJD (where
variant). lymphoid tissue outside of the CNS is also affected), tonsillar
A B C
Fig. 37.5 MRI findings in sporadic and variant CJD. (A) Coronal FLAIR sequence in sporadic CJD, showing bilateral hyperintensity of caudate and putamen (arrows), but
no signal change within the thalami; (B) coronal FLAIR sequence in variant CJD, showing the opposite pattern with bilateral hyperintensity of the thalami (arrows), but
no signal change within caudate and putamen; (C) coronal FLAIR sequence in sporadic CJD, showing no deep nuclear changes but patchy bilateral hyperintensity of the
cortical ribbon.
(Reproduced with permission from Worrall et al., 1999; Knight and Will, 2004.)
biopsy may be diagnostic. For vCJD a diagnostic blood test is also Dalmau, J. and Rosenfeld, M. R. (2008). Paraneoplastic syndromes of the
now available, with reasonable sensitivity and excellent specificity CNS. Lancet Neurology, 7(4), 327–40.
(Edgeworth et al., 2011). In the UK, advice on the investigation Dalmau, J. A. et al. (2008). Anti-NMDA-receptor encephalitis: case series and
analysis of the effects of antibodies. Lancet Neurology, 7(12), 1091–8.
and management of patients with suspected prion disease can be
Doherty, M. J., et al. (2002). Diffusion abnormalities in patients with Wernicke
obtained through regional neurology services or the National Prion
encephalopathy. Neurology, 58(4), 655–7.
Clinic. Edgeworth, J. A., et al. (2011). Detection of prion infection in variant
Creutzfeldt–Jakob disease: a blood-based assay. Lancet, 377(9764),
Treatment and outcome 487–493.
There is no curative treatment for prion disease and all forms are Esmonde, T. and Cooke, S. (2002). Shunting for normal pressure
ultimately fatal. Life-expectancy in classical sporadic CJD is usually hydrocephalus (NPH). Cochrane Database of Systematic Reviews, doi:
measured in weeks to months, but survival in variant CJD or famil- 10.1002/14651858.CD003157.
ial prion disease may sometimes be prolonged. Symptoms such as Granerod, J., et al. (2010). Causes of encephalitis and differences in their
clinical presentations in England: a multicentre, population-based
myoclonus and sleep disturbance may be palliated with appropriate
prospective study. Lancet Infectious Diseases, 10(12), 835–44.
medication.
Harper, P. S. (1986). The prevention of Huntington’s chorea. Journal of the
Royal College of Physicians of London, 20(1), 7–14.
Key points—Prion disease Hebb, A. O. and Cusimano, M. D. (2001). Idiopathic normal pressure
◆ Prion disease is an uncommon cause of dementia in old age. hydrocephalus: a systematic review of diagnosis and outcome.
Neurosurgery, 49(5), 1166–84; discussion 1184–66.
◆ The form of prion disease most likely to present to old age psy- Hilton, D. A., et al. (1998). Prion immunoreactivity in appendix before clinical
chiatrists is classical sporadic CJD, with the combination of a onset of variant Creutzfeldt–Jakob disease. Lancet, 352(9129), 703–4.
rapidly progressive dementia and myoclonus. Honnorat, J. (2010). Autoimmune limbic encephalitis: an expanding concept.
Lancet Neurology, 9(1), 24–5.
◆ Positive investigations in prion disease include abnormal signal
Irani, S. R., et al. (2010). Antibodies to Kv1 potassium channel-complex proteins
change on MRI, PSWCs on EEG, and 14–3–3 protein in the CSF, leucine-rich, glioma inactivated 1 protein and contactin-associated
but the yield of these tests depends on the age of onset, disease protein-2 in limbic encephalitis, Morvan’s syndrome and acquired
duration, and underlying molecular subtype. neuromyotonia. Brain, 133(9), 2734–48.
Kis, B., et al. (2008). Clinical characteristics of patients with late-onset
◆ Definitive diagnosis can only be achieved by tissue biopsy, or in
multiple sclerosis. Journal of Neurology, 255(5), 697–702.
familial cases, sequencing of the PRNP gene.
Klassen, B. T. and Ahlskog, J. E. (2011). Normal pressure hydrocephalus: how
◆ The prognosis in late-onset classical sporadic CJD is uniformly often does the diagnosis hold water? Neurology, 77(12), 1119–25.
poor, with deterioration and death in months. There is currently Knight, R. S. and Will, R. G. (2004). Prion diseases. Journal of Neurology,
no effective treatment. Neurosurgery and Psychiatry, 75 Suppl 1, i36–42.
Lai, M., et al. (2010). Investigation of LGI1 as the antigen in limbic encephalitis
previously attributed to potassium channels: a case series. Lancet, 9(8),
References 776–85.
Adams, R. D., et al. (1965). Symptomatic occult hydrocephalus with ‘normal’ Lawrence, A. D., et al. (1998a). Evidence for specific cognitive deficits in
cerebrospinal fluid pressure. A treatable syndrome. New England Journal preclinical Huntington’s disease. Brain, 121, 1329–41.
of Medicine, 273, 117–26. Lawrence, A. D., et al. (1998b). The relationship between striatal dopamine
Agren-Wilsson, A., et al. (2005). Brain energy metabolism and intracranial receptor binding and cognitive performance in Huntington’s disease.
pressure in idiopathic adult hydrocephalus syndrome. Journal of Brain, 121 (Pt 7), 1343–55.
Neurology, Neurosurgery and Psychiatry, 76(8), 1088–93. Llewelyn, C. A., et al. (2004). Possible transmission of variant Creutzfeldt–
Anderson, R. M., et al. (1996). Transmission dynamics and epidemiology of Jakob disease by blood transfusion. Lancet, 363(9407), 417–21.
BSE in British cattle. Nature, 382(6594), 779–88. Lovestone, S., et al. (1996). Familial psychiatric presentation of Huntington’s
Bateman, D., et al. (1995). Sporadic Creutzfeldt–Jakob disease in a 18-year-old disease. Journal of Medicine and Genetics, 33(2), 128–31.
in the UK. Lancet, 346(8983), 1155–6. Malm, J. and Eklund, A. (2006). Idiopathic normal pressure hydrocephalus.
Benedict, R. H., et al. (2002). Frontal cortex atrophy predicts cognitive Practical Neurology, 6, 14–27.
impairment in multiple sclerosis. Journal of Neuropsychiatry and Clinical Malm, J., et al. (2000). Three-year survival and functional outcome of
Neuroscience, 14(1), 44–51. patients with idiopathic adult hydrocephalus syndrome. Neurology,
Brandner, S. (2011). Diversity of prion diseases: (no) strains attached? Acta 55(4), 576–8.
Neuropathologica, 121(1), 1–4. Marmarou, A., et al. (2005). The value of supplemental prognostic tests for the
Brierley, J. B., et al. (1960). Subacute encephalitis of later adult life—mainly preoperative assessment of idiopathic normal-pressure hydrocephalus.
affecting the limbic areas. Brain, 83(3), 357–8. Neurosurgery, 57(3 Suppl), S17–28; discussion ii–v.
Britton, T. C., et al. (1995). Sporadic Creutzfeldt–Jakob disease in a 16-year-old Mayr, W. T., et al. (2003). Incidence and prevalence of multiple sclerosis in
in the UK. Lancet, 346(8983), 1155. Olmsted County, Minnesota,, 1985–2000. Neurology, 61(10), 1373–7.
Casmiro, M., et al. (1989). Frequency of idiopathic normal pressure McDonald, W. I., et al. (2001). Recommended diagnostic criteria for multiple
hydrocephalus. Archives of Neurology, 46(6), 608. sclerosis: guidelines from the International Panel on the Diagnosis of
Chard, D. T., et al. (2002). Brain atrophy in clinically early relapsing-remitting Multiple Sclerosis. Annals of Neurology, 50(1), 121–7.
multiple sclerosis. Brain, 125(Pt 2), 327–37. Mihara, M., et al. (2005). The ‘pulvinar sign’ in a case of paraneoplastic
Collins, S. J., et al. (2006). Determinants of diagnostic investigation limbic encephalitis associated with non-Hodgkin’s lymphoma. Journal of
sensitivities across the clinical spectrum of sporadic Creutzfeldt–Jakob Neurology, Neurosurgery and Psychiatry, 76(6), 882–4.
disease. Brain, 129(Pt 9), 2278–87. Momjian, S.B., et al. (2004). Pattern of white matter regional cerebral blood
Corsellis, J. A., et al. (1968). ‘Limbic encephalitis’ and its association with flow and autoregulation in normal pressure hydrocephalus. Brain, 127(Pt
carcinoma. Brain, 91(3), 481–96. 5), 965–72.
Parchi, P., et al. (1999). Classification of sporadic Creutzfeldt–Jakob disease Spencer, M. D., et al. (2002). First hundred cases of variant Creutzfeldt–
based on molecular and phenotypic analysis of 300 subjects. Annals of Jakob disease: retrospective case note review of early psychiatric and
Neurology, 46(2), 224–33. neurological features. British Medical Journal, 324(7352), 1479–82.
Parthasarathi, U. D., et al. (2006). Psychiatric presentation of voltage-gated Thieben, M. J., et al. (2004). Potentially reversible autoimmune limbic
potassium channel antibody-associated encephalopathy. Case report. encephalitis with neuronal potassium channel antibody. Neurology,
British Journal of Psychiatry, 189, 182–3. 62(7), 1177–82.
Peden, A. H., et al. (2004). Preclinical vCJD after blood transfusion in a PRNP Tisell, M., et al. (2005). National and regional incidence of surgery for adult
codon 129 heterozygous patient. Lancet, 364(9433), 527–9. hydrocephalus in Sweden. Acta Neurologica Scandinavica, 112(2),
Poca, M. A., et al. (2004). Is the placement of shunts in patients with 72–5.
idiopathic normal-pressure hydrocephalus worth the risk? Results of a Trenkwalder, C., et al. (1995). Starnberg trial on epidemiology of Parkinsonism
study based on continuous monitoring of intracranial pressure. Journal and hypertension in the elderly. Prevalence of Parkinson’s disease and
of Neurosurgery, 100(5), 855–66. related disorders assessed by a door-to-door survey of inhabitants older
Polliack, M. L., et al. (2001). Late-onset multiple sclerosis. Journal of the than 65 years. Archives of Neurology, 52(10), 1017–22.
American Geriatrics Society, 49(2), 168–71. Trip, S. A. and Miller, D. H. (2005). Imaging in multiple sclerosis. Journal of
Polman, C. H., et al. (2005). Diagnostic criteria for multiple sclerosis: 2005 Neurology, Neurosurgery and Psychiatry, 76 Suppl 3, iii11–18.
revisions to the ‘McDonald Criteria’. Annals of Neurology, 58(6), 840–6. Vincent, A., et al. (2004). Potassium channel antibody-associated
Relkin, N., et al . (2005). Diagnosing idiopathic normal-pressure encephalopathy: a potentially immunotherapy-responsive form of limbic
hydrocephalus. Neurosurgery, 57(3 Suppl), S4–16; discussion ii–v. encephalitis. Brain 127(Pt 3), 701–12.
Reuter, I., et al. (2000). Late onset levodopa responsive Huntington’s disease Voltz, R. (2002). Paraneoplastic neurological syndromes: an update
with minimal chorea masquerading as Parkinson plus syndrome. Journal on diagnosis, pathogenesis, and therapy. Lancet Neurology, 1 (5),
of Neurology, Neurosurgery and Psychiatry, 68(2), 238–41. 294–305.
Robertson, N., et al. (1996). Multiple sclerosis in south Cambridgeshire: Wild, E. and Tabrizi, S. (2007). The differential diagnosis of chorea. Practical
incidence and prevalence based on a district register. Journal of Neurology, 7, 360–73.
Epidemiology and Community Health, 50(3), 274–9. Worrall, B. B., et al. (1999). Type 1 protease resistant prion protein and valine
Savolainen, S., et al. (2002). Five-year outcome of normal pressure homozygosity at codon 129 of PRNP identify a subtype of sporadic
hydrocephalus with or without a shunt: predictive value of the clinical Creutzfeldt–Jakob disease. Journal of Neurology, Neurosurgery and
signs, neuropsychological evaluation and infusion test. Acta Neurochir Psychiatry, 67(5), 671–4.
(Wien), 144(6), 515–23. Zarei, M., et al. (2003). Cognitive presentation of multiple sclerosis: evidence
Schoenfeld, M., et al. (1984). Increased rate of suicide among patients with for a cortical variant. Journal of Neurology, Neurosurgery and Psychiatry,
Huntington’s disease. Journal of Neurology, Neurosurgery and Psychiatry, 74(7), 872–7.
47(12), 1283–7. Zemack, G. and Romner, B. (2002). Adjustable valves in normal-pressure
Schott, J. M. (2006). Limbic encephalitis: a clinician’s guide. Practical hydrocephalus: a retrospective study of 218 patients. Neurosurgery,
Neurology, 6, 143–53. 51(6), 1392–400.
CHAPTER 38
Pharmacological
treatment of dementia
Roy W. Jones
This chapter will summarize the available clinical evidence for may be important in the pathogenesis of dementia for a number
specific pharmacological treatment of dementia, with a particular of reasons, including its activity at the N-methyl-D-aspartate
emphasis on practical considerations and realistic expectations of (NMDA) receptor (Danysz et al., 2000). The NMDA receptor is
currently available antidementia drugs. It will not consider mild capable of long-term potentiation and is probably a prerequisite for
cognitive impairment (MCI) nor will it consider areas such as the memory formation and is therefore a potential target for therapeu-
treatment of motor symptoms for people with Parkinson’s disease tic interventions in dementia. The only other successful treatment
dementia (PDD) or dementia with Lewy bodies (DLB). to date involving a neurotransmitter approach has been targeted
The search for specific treatments for dementia has inevitably at glutamate and led to the licensing of memantine, which acts as
concentrated on Alzheimer’s disease (AD), partly because it is an uncompetitive antagonist at the NMDA receptor (Danysz et al.,
the commonest cause of dementia and partly because scientific 2000).
progress has provided more potential therapeutic targets for AD
than other dementias. Many neurochemical studies have been car- Treatment of Alzheimer’s Disease
ried out in AD, particularly looking at specific neurotransmitters,
and a number of approaches have been, or are being, investigated. Cholinesterase inhibition
Abnormalities have been reported in a number of neurotransmit- Changes in the cholinergic system in AD have been known for over
ters, although the changes in acetylcholine appear to be particularly 25 years. A number of therapeutic approaches have been tried to
consistent and significant. It is likely that these changes are second- stimulate the cholinergic system, including the use of acetylcholine
ary to neuronal damage and death. By analogy with Parkinson’s dis- precursors such as choline and lecithin and the use of cholinergic
ease (PD), it would be expected that replacement therapies could agonists (both muscarinic and nicotinic), but AChEIs are the only
potentially enhance residual synaptic activity. Whilst such therapy compounds to have shown consistent efficacy in both clinical trials
may still be important, it would only be expected to provide symp- and clinical practice. In 1993, tacrine became the first agent spe-
tomatic benefit and not affect the underlying disease process. The cifically approved for treating the cognitive symptoms of AD and
cholinergic hypothesis (Francis et al., 1999) suggests that AD results was licensed in countries such as the US and Sweden but not in
from a selective loss in cholinergic neurons, especially within the the UK. Early trials with tacrine combined its administration with
basal forebrain and neocortex. This loss results in decreased acetyl- lecithin, but this is unnecessary. Whilst the efficacy of tacrine in
choline (ACh) levels, and strategies to enhance these levels provide mild to moderate AD was clear (Arrieta and Artalejo, 1998) and
a logical target for treatment. Acetylcholinesterase (AChE) is the beneficial to a number of patients, there were problems with its tol-
principal enzyme that breaks down ACh and the first drugs licensed erability, especially with a reversible hepatotoxicity, which limited
specifically for use in AD were AChE inhibitors (AChEI). its usefulness. In addition, it needed to be taken frequently, which
In PD patients with cognitive impairment, there is cell loss in is a limitation in predominantly older people who also have mem-
the cholinergic basal nucleus and a cortical deficit in AChE that ory problems. Tacrine was prescribed for around 300,000 patients
correlates with the severity of dementia and is independent of any worldwide, but it has been supplanted by later AChEIs with fewer
coexisting AD. Patients with DLB also show a severe loss of choline tolerability problems and a reduced dosing frequency.
acetyltransferase in the cerebral cortex, which is even greater than Three other AChEIs have now been licensed for use in AD in
that found in AD and which also correlates with the severity of the most countries: donepezil (Aricept), rivastigmine (Exelon), and
dementia. Such deficits therefore suggest a possible therapeutic role galantamine (Reminyl). These three drugs are reaching the end of
for cholinergic therapies in both PDD and DLB. their patent lives, so that generic forms have become or are becom-
Glutamate, the principal fast excitatory neurotransmitter in the ing available. AChEIs block the hydrolysis of ACh at the synapse,
CNS, is the neurotransmitter of the neocortical pyramidal cells, the increasing its availability to muscarinic and nicotinic receptors.
neurons that are selectively lost in AD (Francis et al.1993). Glutamate All three AChEIs are active against AChE, which is the principal
form of the enzyme. Rivastigmine is also active against butyrylcho- is a small molecule and both lipophilic and hydrophilic, it is well
linesterase, whilst galantamine appears to be an allosteric modula- suited for transdermal delivery, and once-daily patches (4.6 mg/24
tor of nicotinic receptors in animal models. It has been suggested h and 9.5 mg/24 h) have now been licensed. The 2008 Cochrane
that these subsidiary actions may lead to different clinical effects, update (Birks et al., 2009) for rivastigmine does include one study
although this has never been convincingly demonstrated. All three (Winblad et al., 2007) of the skin patch where adverse events for
AChEIs are licensed in the UK for use in mild to moderate demen- the 9.5-mg/24 h patch were not significantly different from placebo,
tia in AD. Donepezil is also licensed for severe dementia in some although skin irritation is occasionally seen. A higher 17.4-mg/24
countries including the US but not within Europe. Rivastigmine is h patch did cause more adverse events and a current study is being
also approved in the UK for mild to moderate dementia in PD. completed with a 13.3-mg/24 h patch. The skin patch has overcome
Donepezil some of the tolerability problems with oral rivastigmine and offers
an alternative dosing route that may be preferred by and for some
Donepezil was licensed in the UK in 1997 and is the most widely
patients.
used AChEI and has been studied the most intensively. It is ben-
eficial for people with mild, moderate, and severe AD (Birks and Galantamine
Harvey, 2009), and is associated with improvements in cognitive Galantamine was licensed in the UK in 2000 for the treatment of
function and activities of daily living. The effects on cognition mild to moderately severe AD. The drug is initially given as a tablet
remained measurable and statistically significant at 52 weeks in at a noneffective dose of 4 mg twice daily for 4 weeks, increasing
one placebo-controlled study (Winblad et al., 2001). The Cochrane to 8 mg twice daily for a further 4 weeks, with a maximum dose of
Review suggested that 10 mg was marginally better than 5 mg and 12 mg twice daily. An extended release form of galantamine is now
that the lower dose may be preferable because of improved toler- available as capsules of 8, 16, and 24 mg and these are administered
ability, but clinically most patients are titrated to 10 mg if possible. once daily.
It is the simplest to use, with an effective initial dose, one dose titra- Galantamine improves global and cognitive symptoms at doses
tion, and no interaction with food. Anecdotally, higher doses up of 16 mg/day or greater for at least 6 months in people with mild
to 20 mg have been used by clinicians and, more recently, a 23-mg to moderate AD (Loy and Schneider, 2006). Although there are
dose has been approved by the FDA in the US. This approval has no long-term placebo-controlled trials, a more recent analysis
been criticized (Schwartz and Wolosin, 2012) as unjustified on both suggested that patients demonstrating improvement, stability, or
efficacy and safety grounds and that its purpose is some additional limited cognitive decline after 2–5 months at their maintenance
patent protection for a dose that cannot be achieved with 5- or dose were more likely to experience longer-term continued benefit
10-mg tablets. with galantamine (Kavanagh et al., 2011). Its safety profile in AD
In some countries, it is recommended that AChEI treatment is similar to other AChEIs with respect to cholinergically medi-
should stop when AD becomes severe, and reimbursement may ated gastrointestinal symptoms (Loy and Schneider, 2006). The
depend on this. Severity is often characterized by the Mini-Mental prolonged-release once-daily formulation at 16–24 mg/day was
State Examination (MMSE) (Folstein et al., 1975), with severe AD found to have similar efficacy and side-effect profiles as the equiva-
considered to be an MMSE score of less than 10, despite the known lent twice-daily regime (Loy and Schneider, 2006).
limitations if the MMSE is used on its own for such a purpose. In
practice, there have been little data to guide the clinician about Adverse effects of cholinesterase inhibitors
whether to continue treatment or not when the disease progresses
The principal difference between the three AChEIs is in the fre-
from moderate to severe AD, even though there are potential con-
quency and type of adverse events (O’Brien et al., 2011). The main
cerns that continuing may be associated with an increase in adverse
side effects are gastrointestinal, particularly nausea, vomiting, and
outcomes, including syncope, the need for permanent pacemak-
diarrhoea, as would be expected given the cholinergic enhance-
ers, and hip fractures (Gill et al., 2009). The recent donepezil and
ment provided by the compounds. These side effects are most likely
memantine in moderate to severe AD (DOMINO) study (Howard
soon after treatment is commenced or when the dose is increased,
et al., 2012) suggests that continued treatment with donepezil was
hence the reason the drugs are all titrated up over a number of
associated with cognitive and functional benefits over the course of
weeks. Compliance with therapy is always a potential issue and this
12 months in patients where the clinician was considering a change
may be more so in older patients with a memory problem who are
in drug treatment (i.e. stopping donepezil or introducing meman-
also very likely to be taking a number of other medications. There
tine) on the basis of National Institute for Health and Clinical
are particular issues if rivastigmine capsules have been omitted for
Excellence (NICE) guidance (2007) at the time, discussions with
a few days, since it is necessary to retitrate up from 1.5 mg twice
the patient and caregivers, and the physician’s clinical judgement.
daily, because there has been a case of severe vomiting and a rup-
Rivastigmine tured oesophagus in someone restarted straightaway at a high dose
Rivastigmine was licensed in the UK in 1998 for the treatment of (Babic et al., 2000). The rivastigmine skin patch may be helpful for
mild to moderately severe AD. The drug is administered as cap- people who cannot tolerate an oral dose of an AChEI, as gastroin-
sules given twice daily with food, commencing at a noneffective testinal side effects are reduced, but skin sensitivity may occasion-
dose of 1.5 mg twice daily and increasing gradually and accord- ally be a problem instead.
ing to tolerability at a minimum of 2-weekly intervals to a maxi- Adverse interactions with other drugs have not generally been an
mum of 6 mg twice daily. The efficacy of oral rivastigmine has been issue, although it is important that everyone is aware of the ther-
well demonstrated (Birks et al., 2009), but adverse effects such as apy they are receiving and to mention this when any other medi-
nausea and vomiting especially at higher doses are more frequent, cal issue arises because of their role as cholinergic stimulants. Care
for example, by comparison with donepezil. Since rivastigmine should be exercised in patients with reversible airways disease or
CHAPTER 38 pharmacological treatment of dementia 505
peptic ulcer disease and in those with cardiac conduction defects or detectable effect on cognitive function and functional decline
significant bradycardia, as the compounds increase vagal tone. measurable at 6 months in patients with moderate to severe AD.
The review also noted that memantine was well tolerated and that
Switching from one AChEI to another slightly fewer patients with moderate to severe AD taking meman-
The rationale for switching from one AChEI to another is that they tine develop agitation in comparison with placebo.
are from different chemical classes and will have slightly differ- One of the striking features of clinical trials with memantine has
ent pharmacological properties. The main reason for switching is been that adverse events are often similar to placebo (in contrast to
because of poor tolerability, usually gastrointestinal, and the hope AChEIs) and, in general, gastrointestinal side effects are uncom-
that an alternative compound will avoid this. There is no accept- mon and memantine is extremely well tolerated (Jones, 2010).
able or easy way of identifying subjects in advance who are likely to Memantine therefore may be especially suitable for older and/or
respond to an AChEI, but differences, e.g. in individual metabolism frailer patients.
of a drug, may lead to higher or lower drug levels, which could
improve efficacy but possibly at the risk of side effects. A recent Add-on or combination therapy of AChEIs with
study has suggested that functional polymorphisms in the CYP2D6 memantine
gene can influence the clinical efficacy of donepezil (Seripa et al., Simultaneous initiation of therapy with an AChEI and memantine
2011). A significantly higher frequency of gene variants conferring has not been studied. This may partly be explained by their differ-
decreased or absent enzyme activity was observed in responders in ent licensed indications, being mild to moderately severe AD for
comparison with nonresponders (73.68% vs 36.84%; P = 0.005); it AChEIs and moderate to severe AD for memantine. It would be
might be predicted that ‘nonresponders’ would be able to tolerate a useful to know whether commencing subjects, for example, on a
higher dose of donepezil because of their enhanced metabolism of half dose (5 mg) of donepezil with a half dose (10 mg) of meman-
the compound and that more patients would then show a positive tine would lead to similar or increased efficacy compared with full
response to the compound. doses of either drug alone. In addition, adverse events, particu-
Few studies of switching have been carried out, and all switch- larly gastrointestinal, might be reduced by combination therapy
ing has been from donepezil; no recent or double-blind studies are and this was seen in subjects receiving memantine and donepezil
available. However, it does appear that a proportion of patients may compared with placebo and donepezil in one memantine add-on
benefit from a switch (Burns et al., 2006), and a change of formula- study (Tariot et al., 2004). There are no published studies investi-
tion (particularly from a standard oral one to a liquid or transder- gating the addition of an AChEI to patients already established on
mal patch) does offer additional opportunities to try to find a memantine and this may also be because of the different licensed
treatment that is tolerated satisfactorily. NICE has recommended indications, since patients are usually commenced on an AChEI,
that memantine should be considered as an alternative to an AChEI with memantine being considered as a later alternative or addi-
in moderate AD if there are contraindications or tolerability issues tional treatment; increasingly, in countries like the US, adjunctive
(NICE, 2011). therapy with memantine is being commenced in patients who still
have mild AD.
N-methyl-D-aspartate (NMDA) receptor antagonism The effect of adding memantine to an AChEI is not entirely clear.
Memantine An initial placebo-controlled study of patients with moderate to
Memantine is an uncompetitive, moderate affinity antagonist at severe AD (Tariot et al., 2004), which was one of the pivotal licens-
the NMDA receptor. It shows strong voltage dependency and rapid ing studies for memantine in the US, did show clear benefit in both
blocking and unblocking kinetics. It is believed to restore glutama- cognitive and noncognitive symptoms (behaviour as measured
tergic neuronal transmission to physiological levels, while prevent- with the Neuropsychiatric Inventory (NPI)) over 24 weeks when
ing the effects of tonic, pathologically elevated synaptic glutamate memantine was added to the therapy of patients who were stable on
levels that may lead to neuronal dysfunction and damage (Danysz donepezil. On the other hand, a similar 24-week study published
et al., 2000). Memantine was licensed in the UK in 2002 for the in 2008 (Porsteinsson et al., 2008), where memantine was added to
treatment of moderately severe to severe AD, but this licence was patients with mild to moderate AD who were stable on one of the
subsequently extended to cover moderate dementia in AD. The three AChEIs (although most were receiving donepezil), failed to
drug was initially recommended for 10 mg twice-daily dosage, but show any clear cognitive or noncognitive benefit. Open-label obser-
clinical trials demonstrated that 20 mg once-daily dosing was iden- vational studies have suggested that treatment with antidementia
tical (e.g. Jones et al., 2007); the dosage has now been modified to drugs can slow admission to institutional care and that patients on
once-daily dosing, beginning with a noneffective dose of 5 mg and combination therapy gain the most benefit (Lopez et al., 2009), and
increasing to 20 mg over a 4-week titration period. Either 10 mg or that memantine in combination with an AChEI slows both cog-
20 mg is an effective dose and it is still possible to give 10 mg twice nitive and functional decline with small-to-medium effect sizes in
a day if this is more convenient (e.g. if patients are also receiving comparison with AChEIs alone or no treatment (Atri et al., 2008).
rivastigmine or galantamine twice daily). Both of these studies have potential confounds since the cohorts on
NICE failed to recommend memantine when it first reviewed the different treatments are potentially different and largely from
its cost-effectiveness in AD in 2006 (NICE, 2007), but this advice different time periods. More recently, the DOMINO study in the
was altered in its latest review in 2011 (NICE, 2011). Memantine UK did demonstrate that starting memantine at the time donepezil
is recommended in moderate AD for those for whom AChEIs are was discontinued was better than not adding memantine, but that
contraindicated or where tolerability is poor, and it is the only drug adding memantine while continuing with donepezil did not seem
licensed in the UK for severe AD. A Cochrane Review (McShane, significantly better than donepezil alone (Howard et al., 2012).
2006) confirmed that memantine had a small, beneficial, clinically This study has been criticized (Tariot, 2012) partly because of the
marked placebo decline in subjects randomized to receive neither the patient’s condition, but this is often very subjective and largely
memantine nor donepezil. Tariot also comments that there is no depends on the primary caregiver’s opinion.
evidence for the lack of adjunctive memantine therapy based on In an individual patient, slowing the rate of decline may have
the small sample sizes at the study end and the differential dropout important consequences, such as reduced caregiver stress and
rates between the groups, such that the failure to demonstrate a sta- delayed nursing home placement (Lopez et al., 2009). However, it
tistically significant effect does not mean or even imply that there is is important that the patient, caregiver, and others such as the pri-
no effect. The trial also suffered from a significantly lower recruit- mary care physician have realistic expectations of the treatment.
ment of subjects than planned and a significant dropout rate, even Since patients may improve very little, stabilize, or even decline
in the group that continued with donepezil, where only half of the despite benefiting, people may become disillusioned if they have
patients actually continued the treatment for the entire 1-year study not been warned about this; this may affect compliance or lead to
period (Schneider, 2012). Unfortunately, such a trial is unlikely to premature withdrawal of therapy.
be repeated, yet more information is still required about long-term
benefits, safe discontinuation of AChEIs, and the role of combina- Treatment of Non-Alzheimer Dementia
tion therapy with memantine (Schneider, 2012).
Dementia with Lewy bodies (DLB) and Parkinson’s
disease dementia (PDD)
Practical Considerations and Realistic Patients with DLB or PDD have a severe deficit in cortical ACh
Expectations of Treatment in AD levels, so that AChEIs might be expected to help. Oral (but not
transdermal) rivastigmine is licensed in the UK specifically for
How early should treatment be initiated?
mild to moderately severe dementia in people with idiopathic PD.
Although the AChEIs are only licensed for people with a diagnosis Although no other drugs are licensed for PDD, and no drugs what-
of AD, a number of clinicians do commence treatment for people soever are licensed for DLB, there is evidence that the other AD
with MCI, where many but not all subjects will go on to develop a drugs may be beneficial in some patients.
dementia, usually AD, in the coming years. The evidence to sup- Visual hallucinations and visuoperceptual deficits are common
port this is poor, with a number of negative trials with different to both PDD and DLB, together with fluctuating attention. The
AChEIs and for galantamine an unexplained excess death rate in combination of cognitive, neuropsychiatric, motor, and autonomic
the drug-treated group from two placebo-controlled MCI trials features in DLB causes considerable functional decline. Treating
(Loy and Schneider, 2006). neuropsychiatric features such as hallucinations may exacerbate
parkinsonism, while antiparkinsonian medications may exacer-
Less than expected decline bate psychosis. In general, patients with DLB do not respond that
In a chronic progressive neurodegenerative disease like AD, well to antiparkinsonian medication, although more recent studies
improvements in symptoms are important and the evidence sug- suggest that around a third of subjects with DLB do obtain a good
gests that some 40–75% of patients experience a significant ben- motor response to L-dopa (Goldman et al., 2008). Side effects must
efit from AChEIs, but it is difficult to predict who will respond be monitored carefully and the dose increased slowly, avoiding
in advance. Although there may be an initial, usually relatively high doses. Other antiparkinsonian medications should be used
short-term, improvement above baseline, stabilization or less than extremely cautiously because of the likelihood of inducing confu-
expected decline are more realistic therapeutic targets. It would be sion and psychosis (Goldman et al., 2008).
expected that a patient treated early and persistently with medi- A recent Cochrane review (Rolinski et al., 2012) of AChEIs for
cation for AD will show less evidence of behavioural, functional, DLB and PDD only identified six suitable studies (with either
and cognitive deterioration over a period of time than would be donepezil or rivastigmine but not galantamine) for inclusion
expected in the absence of drug treatment (Geldmacher et al., within the meta-analysis, only one of which included patients with
2006). There is substantial evidence that the available treatments DLB. The authors concluded that there were statistically significant
for AD are effective in reducing deterioration in cognition, activi- improvements in global assessment, cognitive function, behavioural
ties of daily living, and behaviour (Geldmacher et al., 2006). For disturbance, and activities of daily living scales in PDD, but that no
example, using a definition of marked clinical worsening for a statistically significant improvement was observed in DLB, where
patient showing cognitive decline (equivalent to the average natural they suggest further trials are necessary. A previous Cochrane
decline in moderate to severe AD patients over 6 months) plus any review (Wild et al., 2003) suggested that patients with DLB who
decline in both activities of daily living and global function, 30% of suffer from behavioural disturbance or psychiatric problems may
patients on placebo showed clinical worsening at 6 months in com- benefit from rivastigmine if they tolerate it, but that the evidence
parison with only 14% on donepezil (Wilkinson et al., 2009). For was weak and further trials (including with other AChEIs) were
memantine, 21% of patients showed a significantly greater degree needed. The NICE-SCIE dementia guideline (2007) concluded that
of worsening on placebo in comparison with 11% on memantine people with DLB who have noncognitive symptoms causing signifi-
(Wilkinson and Anderson, 2007). The recent DOMINO study pro- cant distress (e.g hallucinations or delusions), or leading to behav-
vides further support that patients who are declining whilst taking iour that challenges, should be offered an AChEI.
a drug for AD may still be benefiting by comparison with patients More recently, there has also been some evidence for the ben-
not on therapy (Howard et al., 2012). This does, however, make it efit of memantine on cognition and clinical global impression of
more difficult to decide whether or not treatment is beneficial for change in subjects with either DLB or PDD, although the evidence
the individual patient. If appropriate, therapy may be withdrawn is sometimes conflicting. A randomized controlled trial in sub-
for a short period or the dose reduced to see if there is a change in jects with either DLB or PDD showed significant cognitive benefit
(1.9 difference in MMSE) for memantine compared with placebo There may be reluctance to use warfarin in people with dementia
and significant benefit on clinical global impression of change, the where compliance is likely to be a problem. However, the under-
primary outcome, with almost 30% of patients having a moderate utilization of aspirin and warfarin in older stroke patients with
or substantial improvement on memantine compared with 0% on dementia may be responsible for their increased risk of recurrence
placebo (Aarsland et al., 2009); a preliminary subgroup analysis and death and is potentially modifiable (Moroney et al., 1999). It is
suggested a more pronounced global response in PDD compared important to take into account a patient’s overall situation in try-
with DLB. There was no benefit on noncognitive symptoms, no evi- ing to decide whether the benefits of anticoagulation outweigh any
dence of either improvement or worsening of parkinsonism, and risks, but age alone should not be the deciding factor. If necessary,
memantine was well tolerated. On the other hand, in the largest it may be safer though less effective to select alternatives such as
study of patients with mild to moderate DLB or PDD to date (Emre aspirin, 75–300 mg/day, or another antiplatelet drug such as clopi-
et al., 2010), there was improvement at 24 weeks on global clinical dogrel, and there may now be a place for newer alternatives to war-
status and in behavioural symptoms (measured using the NPI) in farin such as dabigatran etexilate (recently approved for use in AF
the DLB group compared with placebo but not for patients with by NICE (2012)).
PDD. The authors conclude that memantine might be considered
as a treatment option for people with DLB. Antidementia drug treatment for VaD
Several clinical trials in VaD have been conducted with the drugs
Vascular dementia currently licensed for AD. A meta-analysis in 2007 (Kavirajan and
There are no specific antidementia treatments licensed for vas- Schneider, 2007) reviewed all VaD placebo-controlled clinical tri-
cular dementia, so it is important to focus on managing the als of donepezil, rivastigmine, galantamine, and memantine. There
underlying cardiovascular risk factors, including hypertension, were small benefits of uncertain significance in cognition in patients
diabetes, heart disease, and hypercholesterolaemia; these should with VaD, but there was significant heterogeneity of the patients
be treated whenever possible. This is particularly true for younger that made generalizations about efficacy difficult. The evidence
people with milder dementia, when such conditions should be indicates that neither AChEIs nor memantine should be prescribed
treated vigorously. It is more difficult to decide about treatment, in people with VaD, although those with mixed VaD and AD may
for example, of a high cholesterol in someone much older, but if benefit (O’Brien et al., 2011).
other treatment is being considered then there is no good rea-
son to withhold potentially effective treatment. It is important Other dementias
to consider quality of life versus value of life and it is not easy No specific antidementia treatments are licensed for use in other
to decide at what point in time cardiovascular prevention is no dementias such as frontotemporal dementia (FTD). Classification
longer worthwhile. Lipid lowering therapy is probably inappro- of FTD is becoming more complex and the different clinical syn-
priate for a person with terminal cancer or a life-expectancy of dromes may ultimately reveal a variety of underlying pathology
less than 2 years or where there is an irreversibly poor quality of including AD. AD pathology may be seen in 25% of cases of non-
life, and at this stage end of life care becomes more important; fluent/agrammatic primary progressive aphasia (PPA), 25% of
however, such decisions should be made individually and not on cases of semantic variant PPA, and 50% of the logopenic variant of
the basis of age alone. PPA (Grossman, 2012). It would seem reasonable that the subset
There is good evidence for the effectiveness of treating hyper- of FTD with underlying AD pathology might respond to AChEIs
tension in older people, but clinicians are often reluctant to do or memantine, although there are no formal trials confirming this,
so. There may be concerns about compliance and also the risk of and worsening of behavioural symptoms in FTD by AChEIs has
side effects such as postural hypotension. Successful blood pres- been reported (Mendez et al., 2007). The logopenic variant of PPA
sure control can enhance cognitive performance in patients with and another condition, posterior cortical atrophy, are both thought
multi-infarct dementia (Meyer et al., 1986) and antihypertensive to be atypical variants of AD (Mendez et al., 2002; Gorno-Tempini
treatment is associated with a lower incidence of dementia in older et al., 2008), so it would be worth trying AChEIs or memantine in
people with isolated systolic hypertension (Forette et al., 1998). these conditions.
Aggressive treatment of hypertension may be unreasonable, espe- There is little evidence for the efficacy of AChEIs or memantine
cially in patients with severe dementia. in other important causes of dementia, such as corticobasal degen-
Atrial fibrillation (AF) affects around 5% of people over 65 eration, Huntington’s disease, and prion disease; in progressive
and 10% of those over 75 (Hampton, 1999). Guidelines are clear supranuclear palsy, limited clinical trial data with AChEIs suggests
in recommending anticoagulant and antiplatelet drugs that can that the cognitive syndrome may not be helped and that activities
reduce the risk of stroke for people with permanent AF. There is a of daily living/mobility scores may actually deteriorate (O’Brien
five-fold increase in stroke risk at around 5% per year. This risk is et al., 2011).
reduced by about two-thirds with warfarin and by about one-fifth There have been limited formal trials of drug treatment for
with aspirin. The highest risk is seen in people with a previous people with Down’s syndrome and AD. There were no trials of
stroke, those over the age of 75, and those with hypertension, coro- rivastigmine or galantamine, but one randomized controlled trial
nary artery disease, diabetes, heart failure, or left ventricular dys- has been done with donepezil and reported as part of a Cochrane
function (Hampton, 1999). Older people with AF have the lowest review (Mohan et al., 2009) and showed a modest nonstatisti-
frequency of anticoagulant use (Moroney et al., 1999); dementia cally significant trend in favour of those able to tolerate donepezil
appears to be a significant independent determinant of nontreat- treatment. Memantine was not effective for people with dementia
ment with either aspirin or warfarin for the prevention of recur- in Down’s syndrome in a recently published trial (Hanney et al.,
rent stroke (Moroney et al., 1999). 2012).
Other Putative Antidementia Therapies expressions of the underlying brain disease and not simply a reflec-
tion of the distress and other reactions from a person with demen-
Ginkgo biloba tia (Sultzer et al., 2003). The most common problems are affective
There is still considerable controversy about the use and value of syndromes (depression, anxiety, and irritability), apathy, agitation,
Ginkgo biloba, which is an extract from the leaves of the maiden- aggression, psychosis (delusions and hallucinations), and disorders
hair tree that has been used in China for many years and for many of sleep. BPSD cause distress to the individual with dementia, add
different reasons. The extracts contain many agents including flavo- considerably to the stresses experienced by family and professional
noids, terpenoids (including several ginkgolides that are unique to carers, and can result in serious risks to the person and others. They
the ginkgo tree), and organic acids. An updated Cochrane review are also associated with earlier institutionalization, higher costs, and
(Birks and Grimley Evans, 2009) comments that ginkgo appears to mortality (Herrmann et al., 2006; Habermann et al., 2009).
be safe with no excess adverse effects compared with placebo. Many Good practice recommendations such as the NICE-SCIE
of the early trials used unsatisfactory methods with small numbers Dementia Guideline recommend nonpharmacological interven-
of subjects, and publication bias could not be excluded. Overall, tions as the first-line approach for BPSD and emphasize the impor-
the evidence that it has predictable and clinically significant ben- tance of assessing pain and other medical conditions that can often
efits for people with dementia is inconsistent and unreliable. A sub- precipitate the development of these problems.
group analysis on people with AD also failed to show any consistent Despite the fact that many of these symptoms may improve,
pattern of benefit. Since this review, three other studies have been change, or resolve over a period of a few weeks, pharmacological
reported, one examining people with dementia and two primary interventions have often been used as a first-line treatment. In par-
prevention studies (O’Brien et al., 2011). One study in subjects with ticular, there has been concern about the excessive use of antipsy-
dementia and neuropsychiatric features found no additional bene- chotic drugs for the management of BPSD. This was highlighted in
fit from ginkgo added to donepezil. The primary prevention studies the UK Department of Health Report (2009) which estimated that
found that ginkgo did not prevent the development of dementia or at least 180,000 people with dementia are being prescribed antip-
AD but that there did appear to be an increased risk of stroke and sychotics each year in the UK. A number of regulatory bodies have
transient ischaemic attack in the ginkgo-treated groups (O’Brien et warned about the risks of these drugs in people with dementia. It
al., 2011); there have been concerns expressed previously about pos- is estimated that there are an extra 1800 deaths a year in the UK
sible bleeding complications with ginkgo that are probably uncom- because of antipsychotic drug use, together with an additional 1620
mon but may be linked to antagonism of platelet-activating factor. cerebrovascular adverse events (about half of which are severe).
People taking anticoagulants, antiplatelet agents, or with a bleeding
diathesis should be especially cautious about taking ginkgo. Use of cholinesterase inhibitors or memantine
for BPSD
Folic acid and B vitamins There have been more than 30 randomized controlled trials over
Folate is an essential dietary element and low plasma and red cell 6–12 months with AChEI, but few have specifically looked at peo-
folate levels lead to increases in plasma total homocysteine, which ple with BPSD, particularly clinically relevant agitation. Evidence
is a risk factor for brain atrophy, cognitive impairment, and demen- from these trials suggests an overall effect on neuropsychiatric
tia. Plasma concentrations of homocysteine can be lowered by symptoms over 6 months, but the main benefits are probably for
dietary administration of B vitamins (folic acid and vitamin B12). anxiety and apathy rather than agitation and aggression (Ballard et
A Cochrane review (Malouf and Grimley Evans, 2009) concluded al., 2009). In a 12-week trial to treat agitation in patients with AD
that there was no evidence that folic acid with or without vita- (Howard et al., 2007), donepezil was not shown to be more effective
min B12 improved cognitive function in unselected older people than placebo.
with or without dementia. Since this review, an 18-month study of Memantine appears to be a promising treatment for BPSD,
high-dose folic acid, vitamin B12, and vitamin B6 involving 409 sub- based on the main clinical trials which appear to show a benefit
jects with mild to moderate AD (MMSE 14–26) and normal plasma in irritability, lability, agitation, aggression, and psychosis (Ballard
homocysteine levels has been reported (Aisen et al., 2008). No ben- et al., 2009). Clinically, it does seem that memantine may damp
efits on cognition were seen in the intervention group, but there down agitation and it has been used by clinicians for this purpose.
was an unexpected increase in depression. Another more recent Unfortunately, in a recent trial, memantine did not improve patients
study (Smith et al., 2010) in subjects with mild cognitive impair- with clinically significant agitation from care homes or hospitals
ment showed that plasma concentrations of homocysteine could be (Fox et al., 2012) but it was significantly better than placebo for cog-
lowered by high dose folic acid, vitamin B6, and vitamin B12 given nition. The authors comment that it still remains to be determined
over 24 months and that the accelerated rate of brain atrophy in whether memantine has a role in milder agitation in AD.
such subjects could be slowed. Further research is needed to con- Agitation, aggression, and psychosis
firm this result and to see whether similar benefits could be seen in It is important to look for causes of agitation and this has been
people with AD. confirmed in a study that demonstrated that effective manage-
ment of pain significantly reduced agitation in residents of nurs-
Management of Noncognitive Symptoms ing homes with moderate to severe dementia (Husebo et al., 2011).
Paracetamol 1 g up to 4 times a day is a suitable and safe option.
Almost all people with AD will develop one or more noncognitive
symptom in the form of behavioural and psychological symptoms of Antipsychotics
dementia (BPSD) at some time in their disease (Steinberg et al., 2008). Risperidone is the only antipsychotic specifically licensed for use
It appears that in the vast majority of cases, BPSD are fundamental in dementia, and only for short-term (up to 6 weeks) treatment of
persistent aggression in moderate to severe AD unresponsive to to previous belief, monoamine-based antidepressants are not gen-
nonpharmacological approaches and when there is risk of harm to erally effective in depression in dementia. As is typical of clinical
the patient or others. The starting dose should be 0.25 mg twice trials, the most severely ill patients were not assessed; for severe
daily, increasing gradually to a maximum dose of 1 mg twice daily. depression, pharmacological treatment may be appropriate and a
Alternative antipsychotic drugs include olanzapine, aripiprazole, recent Alzheimer’s Society guide (Alzheimer’s Society, 2012) rec-
and quetiapine, but the evidence relating to these is more limited ommends citalopram 10 mg daily, increasing to a maximum of 20
and quetiapine is best avoided (Alzheimer’s Society, 2012). It is mg per day. Higher doses should not be used in patients older than
worth emphasizing that patients with DLB may show severe sensi- 65 years because of an association of the drug with dose-dependent
tivity to neuroleptic drugs such as the atypical antipsychotics, and QT interval prolongation (MHRA, 2011).
fatal reactions have occurred (McKeith et al., 1992).
Sleep disturbance
Depression If sleep hygiene measures have failed, short-term treatment (4
Until recently, there was an absence of evidence about the potential weeks) with a hypnotic such as zopiclone (maximum 7.5 mg per
benefits of antidepressants for treating depression in the context of day) or zolpidem (5 mg per day) can be helpful, although the evi-
pre-existing dementia. However, two large trials have changed this. dence for this is mainly anecdotal (Alzheimer’s Society, 2012).
The Depression in Alzheimer’s Disease–phase 2 (DIADS-2) study
in the US compared sertraline 100 mg (n = 67) with placebo (n
= 64) in depression in AD and found no significant difference in
Overview of Pharmacological Treatment for
symptom change or response or remission rates, but did find an People with Dementia
increase in adverse events on sertraline (Rosenberg et al., 2010), See Table 38.1 for a summary overview.
and concluded that the evidence does not support the use of SSRIs
in depression in AD. A Study of Antidepressants for Depression in
Dementia (the HTA-SADD study) in the UK has reported similar Conclusion
findings in a larger sample of 326 subjects. In this study, a broader Unfortunately no new drugs have been licensed for the treatment
and more representative definition of dementia was used and sub- of dementia since memantine in 2002; the search for more effective
jects randomized to placebo (n = 111), sertraline (n = 107), or mir- and potentially disease-modifying drugs continues. Many com-
tazapine (n = 108) (Banerjee et al., 2011). Again, sertraline was not pounds have been investigated for AD with numerous unsuccessful
better than placebo and was associated with more adverse events, or failed trials and this has led to discussion about the reasons for
and mirtazapine was also found to be no better than placebo on any this. Most approaches have focused on the amyloid hypothesis and
outcomes. These two large studies now demonstrate that contrary compounds that in some way affect the production, deposition, or
Table 38.1 Summary overview of pharmacological treatment for people with dementia
◆ Assess the patient with a possible dementing disorder carefully, including a formal screening instrument (e.g. MMSE)
◆ Establish that the patient has dementia
◆ Establish the most likely type of dementia
◆ If vascular dementia, consider sources of emboli (e.g. carotid disease, AF)
– if atrial fibrillation, consider anticoagulation
– give low-dose aspirin (unless contraindicated, if so consider clopidogrel)
– ensure other relevant conditions (e.g. hypertension, diabetes) are being managed appropriately
◆ If dementia with Lewy bodies, consider:
– L-dopa for parkinsonian symptoms with careful monitoring for emergent psychosis
– AChEIs (and possibly memantine), especially for visual hallucinations
– do not use neuroleptics
◆ If Alzheimer’s disease, consider oral AChEIs (donepezil, galantamine, or rivastigmine) and explain goals of treatment (some early benefits potentially but
stabilization or less than expected decline more likely). Assess cognition (e.g. MMSE). Titrate to maximum dose if possible according to side effects and benefits
– if problems with first AChEI, consider switching to another AChEI (consider transdermal rivastigmine) or memantine
– if AChEIs poorly tolerated or contraindicated, consider memantine
– as disease progresses, consider switching to memantine or adding memantine to the AChEI, particularly if the patient is developing agitation
– monitor the patient’s progress regularly (approximately 6 monthly)
– if stopping drug treatment, consider dose reduction rather than complete withdrawal, and be aware that the patient may deteriorate and medication may need
to be restarted
AChEI, acetylcholinesterase inhibitor; AF, atrial fibrillation; MMSE, Mini-Mental State Examination.
removal of the toxic 42 amino acid peptide amyloid-beta (Aß). It is Ballard, C. G., et al. (2009). Management of agitation and aggression
known that the deposition of amyloid within the brain begins many associated with Alzheimer disease. Nature Reviews, 5 (5), 245–55.
years before the clinical manifestations of AD are apparent and it Banerjee, S., et al. (2011). Sertraline or mirtazapine for depression in
dementia (HTA-SADD): a randomised, multicentre, double-blind,
may be that such drugs need to be given earlier in the AD process
placebo-controlled trial. Lancet, 378, 403–11.
before the person develops Alzheimer’s dementia. The potential use Birks, J. and Grimley Evans, J. (2009). Ginkgo biloba for cognitive impairment
of biomarkers such as MRI structural imaging, PET amyloid func- and dementia. Cochrane Database of Systematic Reviews, <http://www.
tional imaging, and CSF amyloid and tau are increasingly being mrw.interscience.wiley.com/cochrane/clsysrev/articles/CD003120/
considered so that underlying AD pathology can be picked up at frame.html>.
the MCI stage, and trials in such patients may be more effective. A Birks, J. and Harvey, R.J. (2009). Donepezil for dementia due to Alzheimer’s
new lexicon for AD has been proposed introducing the concept of disease. Cochrane Database of Systematic Reviews, <http://onlinelibrary.
prodromal AD (Dubois, 2010), as well as the National Institute on wiley.com/doi/10.1002/14651858.CD001190.pub2/citedby>.
Aging/Alzheimer’s Association (NIA/AA) criteria that represent an Birks, J., et al. (2009). Rivastigmine for Alzheimer’s disease. Cochrane
Database of Systematic Reviews, <http://www.mrw.interscience.wiley.
updated version of the previous NINCDS-ADRDA criteria, which com/cochrane/clsysrev/articles/CD001191/frame.html.
have been part of the standard entry criteria to AD clinical trials Burns, A., et al. (2006). Clinical practice with anti-dementia drugs: a
but which were developed in the 1980s before the availability of consensus statement from British Association for Psychopharmacology.
more specific biomarkers (Jack et al., 2011; McKhann et al., 2011). Journal of Psychopharmacology, 20, 732–55.
The main licensed drug treatments currently available (AChEIs Danysz, W., et al. (2000). Neuroprotective and symptomatological action of
and memantine) are generally only considered to be symptomatic memantine relevant for Alzheimer’s disease—a unified glutamatergic
therapies with no significant effect on halting or reversing the hypothesis on the mechanism of action. Neurotoxicity Research, 2,
underlying disease process. However, the boundary between symp- 85–97.
Department of Health (2009). The use of anti-psychotic medication for people
tomatic and disease-modifying therapies is blurred, particularly in
with dementia: time for action. <www.dh.gov.uk/prod_consum_dh/
identifying the differences clinically. Current treatments may change groups/dh_digitalassets/document/digitalasset/dh_108302.pdf>.
the disease course by delaying deterioration and, for example, entry Dubois, B., et al. (2010). Revising the definition of Alzheimer’s disease: a new
to nursing homes (Lopez et al., 2009) without changing the disease lexicon. Lancet Neurology, 9, 1118–27.
itself. It will be a significant hurdle for new therapies to be accepted Emre, M., et al. (2010).Memantine for patients with Parkinson’s disease
as disease-modifying until surrogate markers such as neuroimag- dementia or dementia with Lewy bodies: a randomised, double-blind,
ing or CSF biomarkers are accepted by regulators. In addition, such placebo-controlled trial. Lancet Neurology, 9, 969–77
drugs will need to show clinical efficacy over a sustained period of Folstein, M.F., Folstein, S.E., and McHugh, P.R. (1975). ‘Mini-mental state’.
time (12– 18 months minimum), including possibly a change in the A practical method for grading the cognitive state of patients for the
clinician. Journal of Psychiatric Research, 12, 189–98.
slope of decline and delay in reaching particular disease milestones,
Forette, F., et al. (1998). Prevention of dementia in randomised double-blind
such as a change in disease severity (using a measure such as the placebo-controlled Systolic Hypertension in Europe (Syst-Eur) trial.
Clinical Dementia Rating Scale or the transition from prodromal Lancet, 352, 1347–51.
AD to Alzheimer’s dementia) or dependence. There are likely to Fox, C., et al. (2012). Efficacy of memantine for agitation in Alzheimer’s
be considerable challenges with administration of these drugs as dementia: a randomised double-blind placebo controlled trial. Plos One,
they may need initial monitoring for safety with MRI scans and 7 (5), E35185. doi: 10.1371/Journal.PONE.0035185.
may involve regular injections or infusions; cost-effectiveness will Francis, P.T., et al. (1993). Cortical pyramidal neurone loss may cause
undoubtedly also be a consideration if they are to achieve wide- glutamatergic hypoactivity and cognitive impairment in Alzheimer’s
disease: investigative and therapeutic perspectives. Journal of
spread usage. However, such advances in treatment for people with
Neurochemistry, 60, 1589–604.
AD and other dementias are essential given the future predicted Francis, P.T., et al. (1999). The cholinergic hypothesis of Alzheimer’s disease:
increase worldwide in the numbers of older people and people with a review of progress. Journal of Neurology Neurosurgery and Psychiatry,
dementia. 66, 137–47.
Geldmacher, D.S., et al. (2006). Realistic expectations for treatment success in
References Alzheimer’s disease. Journal of Nutrition Health and Aging, 10, 417–29.
Aarsland, D., et al. (2009). Memantine in patients with Parkinson’s Gill, S.S., et al. (2009). Syncope and its consequences in patients with
disease dementia or dementia with Lewy bodies: a double-blind, dementia receiving cholinesterase inhibitors: a population-based cohort
placebo-controlled, multicentre trial. Lancet Neurology, 8, 613–18. study. Archives of Internal Medicine, 169, 867–73.
Aisen, P.S., et al. (2008). High-dose B vitamin supplementation and cognitive Goldman, J.G., et al. (2008). Effects of dopaminergic medications on
decline in Alzheimer disease: a randomized controlled trial. Journal of psychosis and motor function in dementia with Lewy bodies. Movement
the American Medical Association 300(15), 1774–83. Disorders, 23, 2248–50.
Alzheimer’s Society (2012). Optimising treatment and care for people with Gorno-Tempini, M.L., et al. (2008). The logopenic/phonological variant of
behavioural and psychological symptoms of dementia: a best practice guide primary progressive aphasia. Neurology, 71, 1227–34.
for health and social care professionals. Alzheimer’s Society, London, Grossman, M. (2012).The non-fluent/agrammatic variant of primary
<www.alzheimers.org.uk/bpsdguide>. progressive aphasia. Lancet Neurology, 11, 545–55.
Arrieta, J.L. and Artalejo, F.R. (1998). Methodology, results and quality Habermann, S., et al. (2009). Predictors of entering 24-h care for people with
of clinical trials of tacrine in the treatment of Alzheimer’s disease: a Alzheimer’s disease: results from the LASER-AD study. International
systematic review of the literature. Age and Ageing, 27, 161–79. Journal of Geriatric Psychiatry, 24, 1291–8.
Atri, A., et al. (2008).Long-term course and effectiveness of combination Hampton, J.R. (1999). The management of atrial fibrillation in elderly
therapy in Alzheimer disease. Alzheimer Disease and Associated patients. Age and Ageing, 28, 249–50.
Disorders, 22, 209–21. Hanney, M., et al. (2012). Memantine for dementia in adults older than 40
Babic, T., et al. (2000). Spontaneous rupture of oesophagus related to years with Down’s syndrome (MEADOWS): a randomised, double-blind,
rivastigmine. Age and Ageing, 29, 370–1. placebo-controlled trial. Lancet, 379, 528–36.
Herrmann, N., et al. (2006). The contribution of neuropsychiatric symptoms Reviews, <http://www.mrw.interscience.wiley.com/cochrane/clsysrev/
to the cost of dementia care. International Journal of Geriatric Psychiatry, articles/CD007178/frame.html>.
21, 972–6. Moroney, J.T., et al. (1999). Treatment for the secondary prevention of
Howard, R. J., et al. (2007). Donepezil for the treatment of agitation in stroke in older patients: the influence of dementia status. Journal of the
Alzheimer’s disease. New England Journal of Medicine, 357, 1382–92. American Geriatrics Society, 47, 824–9.
Howard, R., et al. (2012). Donepezil and memantine for moderate-to-severe NICE (2007). Alzheimer’s disease—donepezil, galantamine, rivastigmine
Alzheimer’s disease. New England Journal of Medicine, 366, 893–903. (review) and memantine. TA111. National Institute for Health and
Husebo, B. S., et al. (2011). Efficacy of treating pain to reduce behavioural Clinical Excellence, London.
disturbances in residents of nursing homes with dementia: cluster NICE (2011). Donepezil, galantamine, rivastigmine and memantine for the
randomised clinical trial. British Medical Journal, 343, d4065. treatment of Alzheimer’s disease. TA217. National Institute for Health and
Jack, C.R., et al. (2011). Introduction to the recommendations from the Clinical Excellence, London.
National Institute on Aging-Alzheimer’s Association workgroups on NICE (2012). Dabigatranetexilate for the prevention of stroke and systemic
diagnostic guidelines for Alzheimer’s disease. Alzheimer’s and Dementia, embolism in atrial fibrillation. TA249. National Institute for Health and
7, 257–62. Clinical Excellence, London.
Jones, R.W. (2010). A review comparing the safety and tolerability of NICE-SCIE (2007). Supporting people with dementia and their carers in health
memantine with the acetylcholinesterase inhibitors. International Journal and social care. Guideline 42. <http://www.nice.org.uk/nicemedia/pdf/
of Geriatric Psychiatry, 25, 547–53. CG042NICEGuideline.pdf>.
Jones, R. W., et al. (2007). Safety and tolerability of once-daily versus O’Brien, J.T. and Burns, A., on behalf of the BAP Dementia Consensus
twice-daily memantine: a randomised double-blind study in moderate to Group (2011). Clinical practice with anti-dementia drugs: a revised
severe Alzheimer’s disease. International Journal of Geriatric Psychiatry, (second) consensus statement from the British Association for
22, 258–62. Psychopharmacology. Journal of Psychopharmacology, 25, 997–1019.
Kavanagh, S., et al. (2011). Long-term response to galantamine in relation Porsteinsson, A.P., et al. (2008). Memantine treatment in patients with mild
to short-term efficacy data: pooled analysis in patients with mild to to moderate Alzheimer’s disease already receiving a cholinesterase
moderate Alzheimer’s disease. Current Alzheimer Research, 8, 175–86. inhibitor: a randomized, double-blind, placebo-controlled trial. Current
Kavirajan, H. and Schneider, L.S. (2007). Efficacy and adverse effects of Alzheimer Research, 5, 83–9.
cholinesterase inhibitors and memantine in vascular dementia: a Rolinski, M., et al. (2012). Cholinesterase inhibitors for dementia with
meta-analysis of randomised controlled trials. Lancet Neurology, 6, Lewy bodies, Parkinson’s disease dementia and cognitive impairment
782–92. in Parkinson’s disease. Cochrane Database of Systematic Reviews,
Lopez, O.L., Becker, J.T., and Wahed, A.S. (2009). Long-term effects of <http://www.mrw.interscience.wiley.com/cochrane/clsysrev/articles/
the concomitant use of memantine with cholinesterase inhibition in CD006504/frame.html>.
Alzheimer disease. Journal of Neurology Neurosurgery and Psychiatry, 80, Rosenberg, P.B., et al. (2010). Sertraline for the treatment of depression
600–7. in Alzheimer disease. American Journal of Geriatric Psychiatry, 18,
Loy, C. and Schneider, L. (2006). Galantamine for Alzheimer’s disease and 136–45.
mild cognitive impairment. Cochrane Database of Systematic Reviews, Schneider, L.S. (2012). Discontinuing donepezil or starting memantine for
<http://www.mrw.interscience.wiley.com/cochrane/clsysrev/articles/ Alzheimer’s disease. New England Journal of Medicine, 366, 957–9.
CD001747/frame.html>. Schwartz, L.M. and Woloshin, S. (2012). How the FDA forgot the evidence:
Malouf, R. and Grimley Evans, J. (2009). Folic acid with or without vitamin the case of donepezil 23 mg. British Medical Journal, 344, 1756–833.
B12 for the prevention and treatment of healthy elderly and demented Seripa, D., et al. (2011). Role of cytochrome P4502D6 functional
people. Cochrane Database of Systematic Reviews, <http://www.mrw. polymorphisms in the efficacy of donepezil in patients with Alzheimer’s
interscience.wiley.com/cochrane/clsysrev/articles/CD004514/frame. disease. Pharmacogenetic Genomics, 21, 225–30.
html>. Smith, A.D., et al. (2010). Homocysteine-lowering by B vitamins slows
McKeith, I., et al. (1992). Neuroleptic sensitivity in patients with senile the rate of accelerated brain atrophy in Mild Cognitive Impairment: a
dementia of Lewy Body type. British Medical Journal, 305, 673–8. randomized, controlled trial. Plos One, 5(9), e12244.
McKhann, G.M., et al. (2011). The diagnosis of dementia due to Steinberg, M., et al. (2008). Point 5–year period prevalence of neuropsychiatric
Alzheimer’s disease: recommendations from the National Institute on symptoms in dementia: the Cache County Study. International Journal of
Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Geriatric Psychiatry, 23, 170–7.
Alzheimer’s disease. Alzheimer’s and Dementia, 7, 263–9. Sultzer, D.L., et al. (2003). Delusional thoughts and regional frontal/temporal
McShane, R., AreosaSastre, A., and Minakaran, M. (2006). Memantine for cortex metabolism in Alzheimer’s disease. American Journal of Psychiatry,
dementia. Cochrane Database of Systematic Reviews, <http://onlinelibrary. 160, 341–9.
wiley.com/doi/10.1002/14651858.CD003154.pub5/pdf/standard>. Tariot, P. (2012). Comment on Howard, R., et al. (2012). Donepezil and
Mendez, M.F., Ghajarania, M., and Perryman, K.M. (2002). Posterior cortical memantine for moderate-to-severe Alzheimer’s disease. New England
atrophy: clinical characteristics and differences compared to Alzheimer’s Journal of Medicine, 366, 893–903. Alzheimer Research Forum, 9 March
disease. Dementia and Geriatric Cognitive Disorders, 14, 33–40. 2012. <http://www.alzforum.org/pap/annotation.asp?powID = 33967>
Mendez, M.F., et al. (2007). Preliminary findings: behavioral worsening on (accessed 20.02.2013).
donepezil in patients with frontotemporal dementia. American Journal Tariot, P.N., et al. (2004). Memantine treatment in patients with moderate
of Geriatric Psychiatry, 15, 84–7. to severe Alzheimer disease already receiving donepezil: a randomized
Meyer, J.S., et al. (1986). Improved cognition after control of risk factors for controlled trial. Journal of the American Medical Association, 291,
multi-infarct dementia. Journal of the American Medical Association, 317–24.
256, 2203–9. Wild, R., Pettit, T., and Burns, A. (2003). Cholinesterase inhibitors for
MHRA (2011). Citalopram and escitalopram: QT interval prolongation—new dementia with Lewy bodies. Cochrane Database System Reviews, <http://
maximum daily dose restrictions (including in elderly patients), contra onlinelibrary.wiley.com/doi/10.1002/14651858.CD003672/full>.
indications, and warnings. <www.mhra.gov.uk/Safetyinformation/ Wilkinson, D. and Andersen, H.F. (2007). Analysis of the effect of memantine
DrugSafetyUpdate/CON137769>. in reducing the worsening of clinical symptoms in patients with moderate
Mohan, M., Carpenter, P. K., and Bennett, C. (2009). Donepezil for dementia to severe Alzheimer’s disease. Dementia and Geriatric Cognitive Disorders,
in people with Down syndrome. Cochrane Database of Systematic 24, 138–45.
Wilkinson, D., et al. (2009). Effectiveness of donepezil in reducing clinical Winblad, B., et al. (2007). A six-month double-blind, randomized, placebo-
worsening in patients with mild-to-moderate Alzheimer’s disease. controlled study of a transdermal patch in Alzheimer’s disease—
Dementia and Geriatric Cognitive Disorders, 28, 244–51. rivastigmine patch versus capsule. International Journal of Geriatric
Winblad, B., et al. (2001). A 1-year, randomized, placebo-controlled study of Psychiatry, 22, 456–67.
donepezil in patients with mild to moderate AD. Neurology, 57, 489–95.
CHAPTER 39
Management of dementia
Sarah Cullum
Dementia is a global health priority, and the costs of managing for Excellence (NICE-SCIE, 2006), which have emphasized the
dementia are becoming increasingly relevant as the prevalence importance of services being person-centred, relationship-centred,
of dementia continues to rise. Alzheimer’s Disease International autonomy-centred, and providing good quality care, as well as
estimated that there were 36 million people living with demen- being based on robust research evidence. These guidelines have
tia worldwide in 2010 (Wimo and Prince, 2010), and that this helped shape services in the UK over the past 5 years, and, depend-
will increase to over 115 million by 2050. The greatest increase in ing on resources available, the guidelines would apply equally to the
the prevalence of dementia is in low- and middle-income coun- management of dementia in other countries. For that reason, they
tries where life-expectancy is rising rapidly. The worldwide cost of have been used as a reference source to update this chapter on the
dementia in 2010 was estimated at US $604 billion, the majority management of dementia.
of which currently occurs in higher-income countries where the The dementias are progressive disorders and the presentation
recognition of dementia is relatively higher. But even in these coun- changes over time. People with dementia have different needs at dif-
tries, medical care costs account for only 16% of the costs, the prin- ferent stages of the disease, and their management needs (and those
cipal costs being due to social care. Much of the social care cost is of their carers and environments) will vary accordingly. This chap-
borne by families of people with dementia, as they often have diffi- ter will concentrate on the assessment and management of people
culty accessing scarce resources from agencies that are struggling to with dementia in the early and later stages of the disorder. At both
cope with demand (Bourne, 2007). Thus the current management stages we will consider engagement, autonomy, respect, protection,
of dementia ought to focus on supporting and developing the care and ethical issues, as well as describing the clinical and psychoso-
provided by family members and carers in the community, because cial aspects of treatment. The emphasis will be on the broader prin-
their education and empowerment will help to optimize the quality ciples of management, which will apply to people with dementia of
of life for people with dementia. any aetiology, their families, carers, and communities, and will take
Dementia encompasses a group of neurodegenerative disorders into account the different settings in which they may present. More
that are characterized by progressive loss of cognitive function and specific treatments will be covered in other chapters.
ability to perform activities of daily living that can be accompa-
nied by neuropsychiatric symptoms and challenging behaviours of Early Dementia
varying type and severity. Dementia changes the way people with
the disease interact with the world around them, as well as how Presentation
the world responds to them. This includes friends, families, carers, People with early dementia may present in a number of different
communities, and services, as well as the wider environment. The ways. They may notice changes in themselves and self-refer to health
social model of dementia (Gilleard, 1984; Kitwood, 1997) helped professionals in primary care, or their family members may become
to redefine dementia as a disability, produced by social exclusion, concerned and suggest that they have an assessment. Health and
mostly due to fear and lack of understanding. The person-centred social care professionals sometimes suspect cognitive impairment
model of care that emerged in the 1980s promoted the under- in people referred to them for other reasons, but they frequently do
standing of the person behind the dementia, and the context in not know the person well and so may miss cognitive changes until
which he/she lives. Thus, the management of dementia is not lim- later in the illness. Some people may present in the general hospital,
ited to the management of individuals and the disease, but must as delirium may complicate an otherwise commonplace physical
also include their personal and physical environment. This means disorder in early dementia. Irrespective of the location and presen-
that modern dementia services must understand the management tation, all people with suspected dementia should have a full and
issues in a variety of sociocultural settings, and should identify the comprehensive diagnostic assessment and evaluation of their need.
specific needs of people with dementia and their carers, regardless Current UK policy and guidelines (NICE-SCIE, 2006; Department
of their age, class, ethnicity, religion, sexuality, or comorbid dis- of Health, 2009) recommend that assessment and diagnosis should
ability (including learning disability). This is reflected in recent UK be carried out by specialist memory services. These are mostly
guidelines for dementia care produced by the National Institute run by mental health services in the UK (Lindesay et al., 2002;
for Health and Clinical Excellence and the Social Care Institute Department of Health, 2009) and they provide assessment and
diagnosis of uncomplicated early dementia, provision of informa- than a member of the person’s family, so that relatives’ personal
tion, initiation and monitoring of treatment, and postdiagnostic opinions do not overly influence the assessment (Moriarty et al.,
support. If the presentation is complicated by other physical, men- 2011). In the UK, the carers of people with dementia have the
tal, or social care needs, the assessment and initial management right to receive an assessment of their own needs, as set out in
plan may need to be carried out by other services, e.g. by staff in the the Carers and Disabled Children Act 2000 and the Carers (Equal
acute hospital or by the community mental health team. However, Opportunities) Act 2004. The assessment of dementia is discussed
the principles of management described in this chapter will broadly in more detail in other chapters (see Chapter 9).
apply to people with dementia in all services and settings. Appropriate investigations help to confirm the diagnosis and aeti-
Currently, only one-third of dementia is recognized in primary ology of dementia, which enables a more accurate prediction of the
care in the UK (Bourne, 2007). Consequently, it is often families, natural course of the disease, which in turn facilitates information
unskilled workers, or health and social care professionals who try sharing, forward planning, and individually tailored management
to manage the initial problems associated with dementia, with- packages for people with dementia and their families. NICE-SCIE
out having any awareness of the underlying diagnosis. This can guidelines (2006) recommend that a basic dementia screen should
take place in a variety of different settings, including the person’s be performed at the initial presentation, including routine haema-
home, primary care, day services, sheltered housing, residential tology, inflammatory markers, serum B12 and folate, and tests for
care, acute hospitals, and intermediate care. Failing to recognize electrolytes, calcium, glucose, liver, renal and thyroid function. A
dementia can cause difficulties for people with dementia and for urine sample is collected if a urinary tract infection is suspected.
their carers who may have no understanding of the changes taking Testing for syphilis and HIV is not routinely undertaken nowadays.
place, and this can potentially lead to neglect or abuse. Thus one Other physical investigations that help with the diagnosis and the
of the main tasks for specialists in the management of dementia aetiology or subtype of the dementia include neuroradiological
is to improve the recognition of early dementia through educa- imaging, both structural, e.g. magnetic resonance imaging (MRI)
tion of the public and people working in mainstream settings, or computed tomography (CT), and functional, e.g. single-photon
to facilitate referrals for assessment, and to support the carers emission computed tomography (SPECT). Sometimes electroen-
(paid and unpaid) who are providing care for the person with cephalography or cerebrospinal fluid examination can be helpful
dementia (Department of Health and Care Services Improvement in elucidating the diagnosis. Brief tests such as the Mini Mental
Partnership, 2005; Bourne, 2007; Department of Health, 2009). State Examination (MMSE) (Folstein et al., 1975) are often used
In recent years there has been a drive to improve the detection of in the initial assessment of cognitive function. In early dementia,
dementia (Brayne et al., 2007) so that families and carers are bet- these tests can be misleading, as they are affected by educational
ter able to plan care. Some have even advocated screening older and sociocultural background (Tombaugh and McIntyre, 1992), by
people for dementia to improve recognition (Ashford et al., 2006, functional illness, especially depression, and by non-Alzheimer’s
2007). However, there is evidence that up to 50% of older people dementias. In frontotemporal dementia, changes in behaviour
that screen positive for cognitive impairment will refuse further and personality often occur before cognitive decline, and in Lewy
assessment and investigation (Boustani et al., 2006). This is mainly body dementia, memory is often initially spared. Thus, in early
due to concerns about the potentially negative impact of the diag- dementia, more extensive neuropsychological assessment may be
nosis of dementia on psychological heath, ability to keep a driver’s required. This is usually carried out by a clinical psychologist using
licence, and on health, life, holiday, and motor insurance. Due to a neuropsychological test battery that evaluates specific domains of
the potential cost of dementia on society, there is increasing pres- cognitive function, such as attention and concentration, orienta-
sure to identify the disease as early as possible. But, in the early tion, short- and long-term memory, praxis, language, and executive
stages of dementia, people are likely to retain capacity to refuse function, e.g. the Cambridge Cognition Examination (CAMCOG)
further assessment and treatment, in which case little more than (Huppert et al., 1995). From the results, a neuropsychological pro-
watchful waiting can be offered. It is unlikely that attitudes will file is generated that may help to identify a particular subtype of
change until the diagnosis of dementia carries less stigma and we dementia.
are able to offer more effective treatments. The management challenge is to identify the unique circum-
stances and needs for each person and to individually tailor a care
Assessment plan to meet those needs. Once a diagnosis has been made and a
A good assessment of dementia will include a detailed description management plan produced, most dementia will be managed by
of the presenting problems, psychiatric and medical history, cur- the primary care and social care services. Occasionally, input from
rent medication, functional ability, living situation, and the ability more specialist services will be required, but care is handed back
of significant others to provide care. It will also include a physi- to the primary care team as soon as possible in most cases. This
cal examination and mental state examination, and assessment allows the primary care team to provide good continuity of care for
of the risks of harm to self and others. If consent is given, it is the person with dementia. However, throughout the course of the
useful to corroborate the history with a member of the family or illness, there is always a need for effective and confidential sharing
someone who knows the person well, as, even in mild dementia of information across people, agencies, and settings, plus collabora-
there may be subtle changes in memory, behaviour, or personal- tive care that includes and supports carers. The role of the dementia
ity that the person with dementia has not noticed. Although peo- specialist is increasingly to facilitate this process by offering expert
ple with dementia and their carers are typically seen together, it is advice to workers in primary care, secondary care, social care, and
also valuable to see them separately to allow both the opportunity residential care settings, in order to provide optimal care for the
to discuss any confidential matters. If the person does not speak person with dementia (Department of Health and Care Services
English it is good practice to use an approved interpreter, rather Improvement Partnership, 2005).
CHAPTER 39 management of dementia 515
Management of early dementia the diagnosis, responding to patient reactions, focusing on quality
The three main management tasks in early dementia of any aetiol- of life and wellbeing, planning for the future, and communicating
ogy are helping the patient and family come to terms with the diag- effectively (Lecouturier et al., 2008). These are described in more
nosis, optimizing quality of life in the present, and planning for the detail in Box 39.1. Many services in the UK may not be adequately
future. Underpinning all of these is the need to respect the auton- resourced to provide this level of involvement, so it is important
omy of the person with dementia and to provide person-centred that people with dementia and their families are given adequate
and relationship-centred care. direction to alternative sources of information and support too.
Health and social care professionals should provide written mate-
Person-centred care rial about the signs, symptoms, course, prognosis, and treatments
Person-centred care has become the cornerstone of the manage- for the particular subtype of dementia, and signpost people to
ment of dementia. It upholds the idea that, throughout the course local care and support services and voluntary organizations, and
of their illness, people with dementia should remain valued as to sources of financial advice, legal advice, and advocacy (NICE-
individual human beings with their own personal (albeit chang- SCIE, 2006). There are particularly good resources available online;
ing) perspective on the world; they will require authentic and e.g. the Alzheimer’s Society (<www.alzheimers.org.uk>) produces
respectful communication from the people around them at all an extensive range of fact sheets on dementia and provides a chat
times (Kitwood, 1997). In early dementia, this approach requires an room for people with dementia and their carers.
active process of respecting the person with dementia’s autonomy
in decision-making, whilst also maintaining duty of care. In the UK
there is now a legal framework in place that helps guide health and Box 39.1 Summary of behaviours in the disclosure of the
social care professionals, namely the Mental Capacity Act (2005), diagnosis of dementia
which was implemented in 2007. Under the Mental Capacity Act,
people are assumed to have capacity to make decisions about their Category of behaviour Subcategories of behaviours
care until proved otherwise, and even if a person does not have Preparing for disclosure Plan disclosure meeting
capacity, then a decision made in that person’s best interest must Arrange postdiagnosis support
take into account his or her past and current preferences on the Establish rapport
subject. In early dementia, the principle of respecting autonomy Prepare the patient
applies to diagnostic assessment, confidentiality, gaining consent
Elicit preferences for disclosure
to speak with family members, requesting physical investigations,
disclosing the diagnosis, starting treatment with drugs, future Integrating family Identify and involve appropriate family members
planning, collaborative care, and more complex decisions such as members Manage differing information needs of patient
who should take responsibility when a person is no longer able and family
to manage his or her finances. The majority of people with early Avoid collusion with family members
dementia are likely to retain capacity to make decisions in most Exploring the patient’s Explore patient ideas
areas, but service providers can sometimes assume otherwise and perspective Elicit patient expectations
undermine the autonomy of people with dementia. Hence health
and social care providers are advised to support training of their Disclosing the diagnosis Tailor information to patient preferences and
ideas
employees regarding autonomy and capacity to consent in people
with dementia, to ensure that they respect their legal and human Check understanding
rights (Department of Health, 2009). Explore the meaning(s) of the diagnosis
Discuss prognosis
Disclosing the diagnosis
Responding to the Explore the patient’s emotional response
Most people with dementia wish to know their diagnosis (Pinner patient’s reactions Elicit and address patient questions and concerns
and Bouman, 2003); however, the experience of diagnosis disclosure
can be distressing for people with dementia, their family members, Focusing on quality of Foster hope
and for the healthcare professionals involved. Some practitioners life and wellbeing Explore coping strategies
acknowledge that they are reluctant to impart a diagnosis for a pro- Planning for the future Clarify follow-up arrangements
gressive terminal disease with as yet no effective cure and uncertain Discuss support services available
prognosis (Bamford et al., 2004). Others have acknowledged the
Negotiate management plan
emotional difficulty of the task of disclosing the diagnosis and have
Discuss prevention and health promotion
recommended that health professionals themselves be supported
in this role (Arber and Gallagher, 2003). People with dementia and Communicating Develop rapport
their families require adequate time to discuss the diagnosis and effectively Use appropriate verbal and nonverbal
its implications. The disclosure of the diagnosis of dementia is best communication
conducted over a series of contacts, rather than in a single inter- Use active listening skills
view, because the majority of patients and a significant minority Involve the patient
of carers do not retain information about the diagnosis after the Structure and signpost the consultation
first disclosure (Barrett et al., 2006). Eight key behaviours have Consider issues of anti-discriminatory practice
been identified as ‘best practice’: preparing for disclosure, integrat-
(Modified from Lecouturier et al. 2008.)
ing family members, exploring the patient’s perspective, disclosing
The aim is to inform and empower the person with dementia, but tramadol, fentanyl, and buprenorphine), but severe pain can also
the vast amount of information can be daunting at what is a difficult worsen cognitive impairment. Balancing the benefits versus the
time for the person receiving the diagnosis. Engaging the carers of harm of potent analgesics can be a challenging task for even the
the person with dementia in collaborative planning for the present most experienced practitioner. If delirium is present, the severity
and the future is therefore vital, but it is important to remember of dementia can be overdiagnosed in an acute hospital setting, so
that consent should be obtained from the person with dementia it is essential to obtain collateral information from someone who
to have the family member involved (NICE-SCIE, 2006), as people knows the patient well about the level of cognitive function prior to
with early dementia often wish to retain control over disclosure to the current illness. Recent sudden deterioration almost always indi-
carers (Tracy et al., 2004). Furthermore, the wish to have a fam- cates a delirium. In the UK, NICE guidelines recommend that all
ily member or friend involved may change, as might the person’s cases of previously undiagnosed dementia in the acute hospital be
capacity to decide, so requesting consent is a process that should referred to a hospital mental health liaison service for a full assess-
be repeated frequently. The key objective here is to encourage good ment (Department of Health, 2009). If the patient was previously
communication with and between people with dementia, their car- able to live independently at home, then the service will usually
ers, and health and social care professionals, whilst keeping upper- recommend prompt treatment of the underlying physical disorder,
most the importance of the perspective and the rights of the person a swift and safe discharge home, with intermediate care if required,
with dementia. followed by a full assessment by the local memory service once the
condition has stabilized. The aim of this approach is to return the
Optimizing quality of life in early dementia
patient to a familiar environment as quickly as possible, since hos-
Although dementia is a progressive disorder, it usually presents in pital admission in dementia is associated with adverse outcomes,
later life when there are many other factors—biological and nonbi- including increased length of stay, institutionalization, and mortal-
ological—that affect the optimal function of both the brain and the ity (Royal College of Psychiatrists, 2005; Sampson et al., 2009).
body. A specialist assessment of the person with dementia will thus
include assessment of the contribution and potential reversibility of Psychological factors
these other factors, the management of which will help to maximize Psychological disorders such as depression or anxiety have a more
the function and quality of life of the person with dementia. subtle effect upon cognition and functional ability. Depression and
Physical factors anxiety may arise secondary to the patient’s insight and concern
The saying ‘what’s good for the body is good for the brain’ is use- about the changes associated with the early stages of dementia, or
ful guidance to people with dementia. Although there is little in response to the change in other people’s behaviour towards them.
research evidence available, the benefits of good nutrition, regular They may be worried about the future, or aware that people around
sleep, exercise, reducing alcohol and tobacco intake, and keeping them are becoming annoyed with their increasing forgetfulness.
to a weekly and daily routine are likely to reduce the impact of the People with early dementia of all types should be given the oppor-
disease. More specific health interventions may include controlling tunity to have postdiagnostic counselling to explore what the diag-
high blood pressure, dyslipidaemia, diabetes, vascular risk factors, nosis means to them and to develop coping strategies (NICE-SCIE,
and managing chronic illnesses such as osteoporosis and sensory 2006). Until recently, psychotherapy and counselling were seldom
impairment, all of which can also exacerbate the extent of cogni- offered to people with dementia and hence there is little research
tive and functional impairment in people with early dementia. The evidence regarding the effectiveness of these methods (Cheston and
Royal College of Psychiatrists has recently highlighted the fact that Jones, 2009). Other interventions that are not specifically aimed at
alcohol misuse and recreational drug use is increasingly common in people’s response to their diagnosis, such as cognitive stimulation
older people, as baby boomers reach retirement age (Royal College programmes, reminiscence therapy, and validation therapy, have all
of Psychiatrists, 2011). Both alcohol and recreational drugs will been shown to be helpful in reinforcing positive coping strategies,
amplify the cognitive impairment seen in an underlying demen- but most systematic reviews of intervention studies have concluded
tia, due to the physiological and metabolic changes associated with that there are insufficient robust studies to draw conclusive findings
ageing that increase the effect of the substances on the brain. But (Clare et al., 2003; Neal and Briggs, 2003; Woods et al., 2005).
the impairment may be at least partially reversible if it is possible Some people with early dementia will have a comorbid depres-
to persuade the person to abstain from using the implicated sub- sive illness. The prevalence of depression in people with dementia
stances. This can be challenging, as the habit is often long-standing is approximately 20%, with the annual incidence rates of major and
and, by this time, the person may have lost the capacity to under- minor depression being reported as 10% and 30%, respectively (Ballard
stand the condition or its consequences. et al., 1996b, 1996c). People with early dementia who have a coexist-
Physical factors may cause a delirium in a person with early ing depressive illness are likely to present with typical symptoms of
dementia, particularly if the patient presents in an unfamiliar depression including anhedonia, amotivation, tearfulness, insomnia,
environment such as a general hospital setting. Common causes and lack of appetite, but the syndrome may also adversely influence
of delirium are infection (respiratory and urinary tract), heart fail- their cognitive function and activities of daily living. As treatment
ure, electrolyte or metabolic imbalance, and untreated pain. Many may substantially improve cognition and function, a new diagnosis
prescribed drugs can worsen cognitive impairment, particularly of dementia should usually be suspended until after the depressive
those with anticholinergic side effects, e.g. tricyclic antidepressants illness is successfully treated. Many studies have linked depression in
and drugs prescribed for urinary frequency and bladder instabil- mild cognitive impairment with increased conversion to dementia,
ity, as they can exacerbate the symptoms caused by the cholinergic and have explored whether treatment with antidepressants affects the
deficit that is a feature of most dementing illnesses. Other medi- rate of conversion; however, a recent review of the area reported that
cations that commonly cause confusion are opioids (including findings are inconsistent and inconclusive (Enache et al., 2011).
Treatment of depression in early dementia is likely to enhance result of people with dementia trying to make sense of their world:
psychological wellbeing, physical function (through improved if they are unable to find familiar items or money (often misplaced
motivation), and general quality of life. After treating comorbid due to forgetfulness) they may conclude that they have been sto-
physical problems and pain, nonpharmacological interventions len. Less commonly, in early dementia the emergence of delusions
for depression in early dementia include maintaining pleasurable may be part of the syndrome associated with Lewy body dementia
activities, addressing relationship and environmental difficulties, (McKeith et al., 1996). If Lewy body dementia is suspected, it is use-
exercise, and psychological support (Teri et al., 1997, 2003), but ful to enquire about other symptoms such as fluctuating cognition,
there have been insufficient robust studies for a meta-analysis of visual hallucinations, REM sleep disorder, parkinsonian symptoms,
evidence of effectiveness. Moreover, a recent Cochrane review and sensitivity to dopamine antagonists, e.g. neuroleptics, but also
of psychotherapeutic treatments for depression in older people antiemetics such as metoclopramide.
without dementia was also inconclusive due to small sample sizes
(Wilson et al., 2008); the best evidence was for cognitive behav- Environmental and social factors
ioural therapy, which is now also recommended to people with The assessment of environmental and social circumstances will
early dementia (NICE-SCIE, 2006). include an appraisal of the person with dementia’s living situation
Recent reviews of pharmacological interventions for depression and important relationships, in particular who provides any care
have reported good outcomes for efficacy of antidepressant medi- that is required, the person’s preferred hobbies and interests, and an
cation in older people without dementia (Mottram et al., 2006), and assessment of his or her activities of daily living. In early dementia,
this is likely to hold true for people with mild cognitive impair- most activities of daily living remain intact, although there may be
ment or early dementia. As there is frequently an overlap of physi- subtle deterioration in some activities such as managing finances or
cal symptoms, such as apathy and fatigue, between depression and remembering to take medication. The opinion of an informant who
dementia, it is important to identify target nonsomatic symptoms knows the patient well is useful, as people with dementia may not
such as anhedonia and guilt, which can be monitored to ensure always have insight into their difficulties. An assessment carried out
reduction in symptoms over time. Rating scales for depression, in the person’s home creates the opportunity to make a quick evalu-
e.g. the 15-item Geriatric Depression Scale (GDS-15) (Sheikh and ation of any environmental risks (e.g. out-of-date food or burnt
Yesavage, 1986), have been shown to be valid for use in people with saucepans) and the need for any aids, equipment, or adaptations in
early dementia (Burke et al., 1989), and can be used to help assess the house. As part of planning for the future, it is worth evaluating
the severity of depression prior to treatment and to examine treat- current ease of access to rooms inside the house, particularly the
ment response. Usually a selective serotonin reuptake inhibitor bedroom or bathroom if these are upstairs, access to the accommo-
(SSRI) is used first-line, but if the person does not respond, then dation itself, as well as access to local shops and other facilities.
a sequenced treatment plan as recommended in various guidelines Environmental factors can be manipulated to help maintain inde-
(NICE, 2009; Taylor et al., 2009) is essential. If there is no response pendent living. Advice regarding enhancing activities of daily living
to the first class of antidepressant, then another from the same or a can be sought from an occupational therapist and/or physiothera-
different class should be tried; or if there is a partial response, then pist. Environmental modifications in the home may be required,
augmentation with another antidepressant can be used, e.g. adding e.g. the instalment of stair rails or shower facilities. Modifications
mirtazapine to an SSRI. Robust treatment strategies are more likely outside the home may also be of benefit, e.g. a ramp may be required
to lead to successful outcomes and lessen therapeutic nihilism. to access the front door so that the person can leave and return
Thorough history taking will identify if there is a past history of to his or her house more easily, thus increasing the opportunities
depressive illness and previous response to antidepressant medica- for exercise and social interactions. Assistive technology can also
tion. Previous good response usually predicts future response and help promote independence, e.g. an automatic pill dispenser can
it makes sense to try successful treatments again. Older patients remind people with dementia when it is time to take their medica-
on long-term antidepressant therapy may have taken tricyclic anti- tion that might otherwise have been forgotten; this is particularly
depressants for many years. As the anticholinergic side effects are useful for people who live alone and have no other means of being
likely to worsen the confusion seen in dementia, it may be nec- reminded.
essary to switch the patient to either another tricyclic antidepres- Assessments and modifications required for people with early
sant with fewer anticholinergic side effects (e.g. lofepramine) or an dementia could be provided by nonspecialist health and social care
antidepressant from an alternative class such as an SSRI. Risks of services for older people, but many of the staff in these services do
antidepressants in older people include anticholinergic side effects, not have the training or the skills to help people who are mildly
gastrointestinal bleeding, hyponatraemia, and falls and fractures confused, so people with dementia may not receive the same qual-
(Coupland et al., 2011). As with treatment of depression in any ity of services as people without dementia. It is becoming part of
other patient group, the patient with dementia should be kept on the remit of specialist mental health services to educate staff in
a maintenance treatment regime using the drug and dose that mainstream services (particularly in primary care, intermediate
resulted in effective remission of symptoms: ‘the dose that gets you care, and general hospitals), so that they can provide good quality
well, keeps you well’ (Reynolds et al., 2006). care to the increasing number of people that they will see who will
Occasionally, people with early dementia will present with psy- have a dementing illness (Department of Health and Care Services
chotic symptoms. These may be hallucinations, which are usually Improvement Partnership, 2005). Indeed, it has been argued that
visual, well-formed, and detailed, and may be recognized as unreal this training should really start with education of the public about
and even pleasant by the person. Some people may present with the disorder, reducing the stigma attached to it and tackling age
delusions, often persecutory in nature; they may accuse people of discrimination (Department of Health, 2009), which would pro-
trying to harm them or steal from them. The delusions may be a duce a whole-population effect (Rose, 1985), resulting in a positive
shift in the distribution of care provided for people with dementia If the clinician feels they have insufficient information about driv-
and their carers. ing to advise the patient, they can suggest taking a disabled driving
Pharmacological interventions for cognitive symptoms of dementia assessment at a local centre. In those who can continue to drive, it
is sensible to advise that they avoid driving in heavy traffic, at night,
The use of cholinesterase inhibitors in the management of dementia
or in bad weather and to drive accompanied if possible. Decisions
is presented in Chapter 38 and will not be discussed in detail here.
regarding driving must be reviewed regularly, as dementia is a pro-
The three main cholinesterase inhibitors, donepezil, rivastigmine,
gressive disorder and thus driving ability will change. If the person
and galantamine, have similar efficacy in improving symptoms of
with dementia refuses to stop driving, then the person’s family may
cognition, function, and behaviour (Birks, 2006). They may also
need to disable the car, or even involve the police to prevent unsafe
delay the onset of the behavioural and psychological symptoms of
driving becoming a hazard to the public.
later dementia, reduce caregiver burden, and delay institutionaliza-
tion. Cholinesterase inhibitors are increasingly used for treatment of Planning for the future
noncognitive symptoms of dementia, particularly the hallucinations
A major aspect of management in early dementia is planning care
and delusions that sometimes present early in Lewy body demen-
for the future. At this stage people with dementia usually still has
tia. Memantine acts at different receptor sites to the cholinesterase
the capacity to make decisions about their care in the future (Moye
inhibitors; it is thought to block N-methyl-D-aspartate (NDMA)
et al., 2004) and should be completely involved in decision-making.
receptors and to have a neuroprotective effect on the brain. In a
People with early dementia may be aware that they are beginning
2006 Cochrane review, McShane et al.(2006) reported that meman-
to find it difficult to manage their financial affairs, perhaps not pay-
tine was well tolerated and conferred some benefit on functional
ing their bills on time or forgetting how much money they have
and cognitive outcomes in dementia, and in 2011 NICE recom-
withdrawn from the bank. They may become vulnerable to finan-
mended that the drug can be used in people with moderate AD
cial exploitation, whether due to consumer fraud or undue coer-
in whom cholinesterase inhibitors are contraindicated or not toler-
cion by people known to them who aim to persuade them to use
ated due to adverse side effects (NICE, 2011).
their assets against their best interests. For these reasons it is useful
Risk management to discuss whom they would like to manage their financial affairs
Areas of potential risk will become apparent in a comprehensive when they are no longer able. In the UK, people can make a Lasting
multidisciplinary assessment of the person with dementia. In early Power of Attorney for Property and Financial Affairs (LPA(PFA)),
dementia, the risks to self and others are usually less than in peo- which can either be used as soon as it is registered or when peo-
ple with later dementia. When they do occur, typical risks include ple lose capacity. They may also wish to appoint someone to make
driving ability, financial mismanagement and/or exploitation, falls decisions about their future health and social care, in which case
(particularly in Lewy body dementia), aggression (usually verbal), they can make a separate Lasting Power of Attorney for Health and
and in some cases suicidality. Forgetfulness may increase the like- Welfare (LPA(H&W)). There is provision within the LPA(H&W)
lihood of poor medicines management and occasionally alcohol to allow the attorney to make decisions about life-sustaining treat-
misuse. Similarly, out-of-date food may present a hygiene risk and ment. Alternatively, people can make their own advance statement
forgotten burning cigarettes may cause fires, both of which could about life-sustaining treatment (Royal College of Physicians, 2009).
be considered a public health risk. Psychosis in early dementia is These decisions need to be made after considerable thought and
sometimes associated with accusations of theft or other wrong-do- discussions between individuals, their close family members, and,
ing, which may result in retaliatory action from the accused, thus if appropriate, their care providers.
increasing the risk to self and others. These issues are covered in more detail in Chapter 56.
Some of these risks can be easily addressed, for example arrang-
ing for medication to be supplied in a blister pack or dosette box, or Carer support
setting up the delivery of ‘meals on wheels’ or frozen ready meals. The diagnosis of dementia in a loved one can have a huge emotional
Adaptations can be made in the home to reduce risks and promote impact on carers. They will face many difficult challenges, includ-
independent living for as long as possible. Risks around aggression ing supporting the person who has been diagnosed with dementia,
and suicidality can usually be managed by giving support and advice whilst also coming to terms with it themselves. People with early
to family members to reduce risk, but will sometimes require inpa- dementia may become aware of the impact of their diagnosis on the
tient assessment and treatment in a mental health unit. Decisions carer, which may have an effect on their own psychological wellbeing
regarding driving ability can be a more complicated subject. Many as well as the relationship. The carer will gradually need to take over
older drivers are very reluctant to give up driving because, partic- decision-making and will have to weigh up the balance of auton-
ularly in rural areas, it severely limits their independence. People omy versus risk in the process, which may lead to disagreements
with early dementia do not have to automatically give up driving, and distress. This can happen even at early stages of dementia, with
but if the history indicates that driving skills are affected, the patient decisions about driving, for example. Carers may have problems
is advised to inform the appropriate driving authority. A history of looking after their own needs as well as those of the person with
accidents or near misses (either given by the patient or by someone dementia. They may admit to frustration, along with verbal and
who knows them), evidence of executive dysfunction or visuospa- sometimes physical aggression towards the person with dementia.
tial deficits on neuropsychological testing, and the new prescrip- Pre-existing ambivalence in the relationship may make adjustment
tion of psychotropic medication may all affect ability to drive. In more difficult and abuse more likely (Homer and Gilleard, 1990).
the UK, people with dementia are legally obliged (if advised by As the disease progresses it has further psychological effects on the
their doctor) to notify the Driver and Vehicle Licensing Agency carer, in particular on the spouse who often goes through a griev-
(DVLA) and should also inform their motor insurance provider. ing process for aspects of his or her partner that disappear, whilst
having to quickly adjust to new aspects that may emerge. Dementia by others who bring it to the attention of the appropriate agencies.
may also have an impact on a couple’s sexual relationship. Being These behaviours have been described as ‘behaviours that chal-
less sexually active may alter the nature of the previous relationship, lenge’, because they are usually problematic, both to people with
which may further intensify the sense of loss. Less commonly, there the behaviour and to the people around them, which can include
may be an increase in sexual demand or expression of inappropri- caregivers, care workers, hospital patients, care home residents, and
ate sexual behaviours, as sometimes seen in early frontotemporal visitors. They are also referred to as the behavioural and psycho-
dementia, which may put the carer or others at risk. It is essential logical symptoms of dementia (BPSD), and, for the sake of con-
that carers are supported with whatever they are dealing with, and venience, that term will be used here. If people with later dementia
do not feel they have to shoulder the burden alone. still live at home, BPSD may include wandering, leaving the front
Fundamental to the ongoing quality of relationships with the per- door open, poor hygiene, hoarding, incontinence, self-neglect,
son with dementia is the need for good communication. Families weight loss, and false allegations about people stealing from them
and carers are encouraged to use clear, simple conversation, to or harming them in some way. If they live in sheltered accommoda-
reduce high emotional expression, to use memory aids, and to be tion, extra care housing, residential care, or a nursing home, then
nonjudgemental. This is probably the most important area in which their behaviours are likely to be those that staff and other residents
quality of life for people with dementia and their families can be find difficult to tolerate, and may include screaming, shouting, call-
optimized, and provides a solid foundation for care as the disease ing out, wandering, agitation, verbal or physical aggression, inap-
progresses. Specific interventions for carers of people with early propriate incontinence, faecal smearing, and inappropriate sexual
dementia include individual and group psychoeducation, peer sup- behaviour. Other symptoms may include delusions, hallucinations,
port groups, support by telephone and via the internet, training anxiety, depression, apathy, misidentification, and disinhibition. At
courses, problem-solving, communication courses, and cognitive this stage of dementia, individuals are less likely to have capacity to
behavioural therapy (CBT) (Sorensen et al., 2002; Brodaty et al., make decisions about their care and, having little insight into the
2003b). Psychosocial interventions for caregivers of people with extent of their difficulties, may refuse help. The ethics of interven-
dementia help to reduce psychological morbidity in carers (Brodaty ing in someone’s life without their consent, possibly in response to
et al., 2003a) and delay the need for institutional care (Spijker et al., third-party information, often has to be considered in these cir-
2008). Children in the family are also affected by dementia; it can cumstances. Specialist input from community-based mental health
be very confusing to see the changes in a much-loved grandparent. and social care services may be required in the short term, but, as
There are now excellent sources of information about dementia for with early dementia, management should be a collaborative proc-
children too, e.g. The Milk’s in the Oven (Mental Health Foundation, ess, working with families, primary healthcare professionals, and
2011). Systemic therapy and family therapy have also been shown social care services to empower and enable them to provide good
to be helpful in families of people with dementia, providing a space quality care for the person with later dementia.
for stories to be heard and long-standing disputes to be resolved,
so that communication and respect can be maximized (Gilleard, Assessment
1996). In later dementia it is critical to carry out as complete an assess-
ment as possible and to accurately describe the target symptoms
Later Dementia in order to develop a robust management plan. Gathering all the
information required can be difficult as people with dementia may
Presentation not be able to communicate their difficulties easily, so an inform-
Some people with dementia do not present until later in the course ant’s history becomes increasingly important. If the person with
of the disease. This is commonly because the person providing care dementia is living in a care home, it is important to bring together
for the individual with dementia believes that his/her behaviour is the views of both the care home staff and visiting family members
an inevitable part of ageing and is perhaps unaware that help and or friends, as every perspective will be valuable. Care home staff are
support are available. In these situations, presentation is often pre- able to give details of day-to-day behaviours, but friends and family
cipitated by a change in the carer’s circumstances; the carer may die members are able to bring years of knowledge about the person and
or become too psychologically or physically unwell to be able to his or her usual ways of reacting to events. Involving families and
provide care any longer. Other people with fairly advanced demen- carers in the assessment process has the added benefit of involving
tia do not have carers but are able to continue to live independently them in the management process too.
in their own homes without help for many years, particularly if they Physical investigations may be required to help exclude treatable
have well-established routines, having lived in the same house for conditions and to elucidate the aetiology of the dementia, which
a long time (often over 50 years), using the same routes, and going can help explain patterns of behaviour and predict a treatment
to the same shops. Again, presentation is usually triggered by a response to some pharmacological interventions. However, a per-
physical, psychosocial, or environmental change. Similarly, other son with later stage dementia may be unwilling to have blood taken
people live in residential care for many years before their dementia or to undergo neuroradiological imaging, in which case it may be
is identified, often because their gradual cognitive and functional necessary to rely solely on clinical history, mental state examina-
impairments are not recognized by care home staff. In all of these tion, and observations from carers to establish the likely diagnosis.
scenarios, people with later dementia are more likely to present in It is often a change in the patient’s behaviour that precipitates an
crisis or at immediate risk, and are thus more likely to require spe- assessment request. The change is usually caused by factors other
cialist dementia care from health and/or social care services. than the underlying dementia, such as pain, physical illness, change
Reasons for referral in later dementia are usually due to behav- in medication, and mood disorders, but also environmental factors
iours identified as high risk or challenging that have been noticed such as change in routine or surroundings. These factors need to be
actively enquired after, investigated, and, if found, tackled robustly it is often the case that people with dementia are cared for by a
in the management plan, to optimize the function and quality of life rapid turnover of staff, both in their own homes and in institutions
of the person with dementia. Thus the importance of a careful his- (Bourne, 2007). Poor pay and lack of training opportunities reflect
tory, taken from as many sources as possible, plus a sensitive mental the lack of value given to dementia care, and this is one of the fac-
state examination, becomes clear in a situation that often depends tors leading to some staff treating people with dementia as less than
solely on clinical acumen, without which the person with dementia human. Thus, another of the management goals of dementia spe-
may continue to suffer unnecessary distress. cialists is to raise awareness by providing educational programmes
and modelling behaviour to help build a foundation for good care
Management of later dementia and management of dementia. Central to this is the requirement
The progressive nature of dementia means that there are additional for good communication. People with later dementia are also often
management issues in later stages. This section will focus on the unable to express their distress effectively, nor able to comprehend
ongoing requirement for person-centred care, the emergence of any verbal reassurance given to them by caregivers. This may be
behavioural and psychological symptoms of dementia, making due to sensory problems such as deafness or visual impairment as
decisions for people who no longer have capacity to do so for them- well as receptive or expressive dysphasia. In these circumstances,
selves, and end of life care in people who are increasingly frail and active listening for communication (even if no words are spoken),
have limited ability to communicate their needs. speaking clearly in short sentences, good nonverbal communica-
tion from the carer such as a clear and kind facial expression or the
Person-centred care touch of a hand, and the active expression of respect and trustwor-
Person-centred care remains at the heart of the management of thiness are paramount. Even if capacity for understanding language
later dementia. However, it can be challenging to maintain this is severely impaired, courteous communication is noticed by the
approach with individuals who are unable to articulate their opin- person with dementia and helps to maintain the feeling of safety
ions about choices in life, from choosing which food to eat through that should underlie all caring and treatment activities (Kitwood,
to decisions about their treatment, place of residence, and palliative 1997).
care. Hence, the central task in the management of later dementia
is to facilitate respect for the person’s life-long preferences as far Behavioural and psychological symptoms of dementia
as possible in all areas of decision-making, whilst also providing The behavioural and psychological symptoms of dementia are
care that is safe, lawful, and optimizes quality of life for people with strongly associated with carer distress (Donaldson et al., 1997),
dementia and their carers. The concept of ‘personhood’, i.e. respect- entry into institutional care (Brodaty, 1996), and prolonged length
ing and valuing each person regardless of the extent of impairment, of stay in hospitals (Royal College of Psychiatrists, 2005). They
is especially important in the later stages of the disease, and is fun- increase in frequency and severity as the dementia progresses and
damental to maintaining positive social relationships in dementia will affect up to 90% of people with dementia (Robert et al., 2005).
(Kitwood and Bredin, 1992). As people get older, their family and The successful management of BPSD requires understanding the
friends may pass away and with them the biography, life stories, behaviour as an attempt by the person to express an unmet need,
and experiences of the person with advanced dementia. Unless which may be physical or psychological (Stokes, 2000). The behav-
there is a proactive attempt to preserve the people’s identity by iour needs to be understood as a method of communication; the
creating sufficient documentation of their existence, people with challenge is to understand the communication. Failing to under-
dementia can be left with no life context, which makes it much stand the communication, or worse, treating the behaviour as a
easier for other people to treat them as ‘nonpersons’. Thus, one of nuisance, will only result in intensifying the behaviour intended
the most important roles of health and social care professionals is to communicate the unmet need. To understand the behaviour we
to assist in the maintenance of the identity of people with demen- need to first understand the individuals—their personal and fam-
tia, usually by facilitating the recording of their personal history ily background, their premorbid personality, their past medical and
and lifetime achievements. One of the ways in which this can be psychiatric history, their current physical and psychological prob-
achieved is by life story work, which aims to capture memories in lems, their communication difficulties in terms of speech and hear-
imaginative ways that may include words, photographs, and music, ing, and their social situation. Second, we need to be able to make
which can be stored in various formats, e.g. in a scrapbook or on a accurate and nonjudgemental observations of a person’s behaviour.
DVD (Moos and Bjorn, 2006). This helps carers to know the per- The most commonly used approach is the ‘ABC’ method, in which
son behind the dementia, and allows them to respect the person’s the context or antecedent (A) of the behaviour is reported, the
preferences for food, music, hobbies, and people, as well as his/her behaviour (B) itself is described, and then the consequences (C) of
spirituality and religion. the behaviour are observed and documented. The understanding of
If the person with dementia is living in a care home, then familiar BPSD depends upon the cooperation of carers who are often under
items such as furniture and photographs can help to maintain a sense considerable stress themselves. As most carers have no training,
of security, trust, and comfort in an increasingly disorientating and this task can appear overwhelming and often futile. Dementia spe-
confusing world. Equally important is the presence of a calm and cialists undertake to enable and empower carers, family members,
friendly routine with familiar carers. Relating to the person with and significant others in the objective observation of behaviour
dementia rather than to the patient with a disease is particularly without adding opinion or judgement. Once information has been
important in clinical settings such as general hospitals and clinics, gathered about the nature of the behaviours, they can, in the light of
where the unfamiliar environment and routines can be extremely what they know about the person’s background, create a hypothesis
disorientating, confusing, and threatening, sometimes leading to about what the behaviour means in terms of unmet need; then a
fear and anxiety and even paranoia and psychosis. Unfortunately management plan can be initiated and implemented.
The causes of BPSD, and therefore their management, can be (Thakur and Blazer, 2008) and in general hospitals is approximately
broadly categorized into physical, psychological, and environmen- 20% (Royal College of Psychiatrists, 2005), compared to a commu-
tal or social factors. The causes are often multifactorial, and conse- nity prevalence of 5–10% (Copeland et al., 1999). Recent life events
quently it is important to actively seek out possible reasons for the such as bereavement or relocation can also be trigger factors for
behaviour, and thus possible treatments, in each of these areas. depression, even if the person with dementia seems to be unaware
Physical factors of them. Symptoms of affective disorders in later dementia include
diurnal variation of mood or other BPSD (e.g. calling out or aggres-
People with dementia are five times more likely to develop delir-
sion that is worse in the mornings but diminishes by the end of
ium from physical causes than other people of the same age (Cole,
the day), poor appetite and weight loss, disturbed sleep, tearfulness
2004). The physical problems may be quite minor but can still
and depressive content to speech, and an expressed wish to either
cause a delirium in a person with advanced dementia, and this may
die or end their life. Sometimes the depressive symptoms appear
present as challenging behaviour. It can be very difficult to recog-
as recurring but short-lived episodes in people with dementia,
nize delirium in people with advanced dementia and usually the
but if the episodes persist they should be treated as evidence of a
only sign is a sudden change in behaviour, often recognized by
depressive disorder. The Cornell Scale for Depression in Dementia
people who know them well. Physical and pharmacological fac-
(Alexopoulos et al., 1988) is a useful rating scale to assess the sever-
tors can also have a more subtle but nonetheless insidious effect in
ity of depressive disorder and to monitor its response to treatment,
dementia, without causing an acute delirium, and should always
as it utilizes the observations of carers and staff.
be considered in BPSD. Pain, dehydration, and constipation are
Psychosis is a relatively common feature in later dementia but
all common causes of agitation and calling out in people who are
may require some degree of skill to elicit if there are communica-
unable to verbally express their discomfort. Urinary tract infections
tion difficulties. Psychotic symptoms could indicate a delirium or
are also common, and can often be recurrent in this patient group.
affective disorder, but may also be a characteristic of the under-
A physical examination and review of medication is necessary,
lying dementia, of any aetiology but most commonly Lewy body
and a dementia screen including blood tests and a urine sample
dementia. Hallucinations are commonly visual and may be simple,
may be required. The person with dementia may refuse physical
e.g. moving figures in the carpet, or more complex, such as animals
examination and investigations, and sometimes the diagnosis has
or people. In some cases the person with dementia re-experiences
to be made using observation and educated guesswork only. Any
whole periods of their lives as if they were real, e.g. they can relive
physical cause found for BPSD, including iatrogenic factors such as
distressing war-time events or childhood sexual abuse. This may
side effects or interactions of medication, should be treated. Many
result in behaviour aimed at self-defence against the perceived
people with later dementia have painful conditions associated with
threat, including verbal and physical aggression towards others or
ageing that they may not be able to communicate to others, and
barricading themselves away from others. Nonpharmacological
adequate analgesia should be given. A recent trial reported that
interventions should always be considered at the outset; however,
treating behavioural symptoms with a trial of stepped analgesia
if the distress is too great, or the risk to self or others too high,
(most receiving paracetamol) significantly reduced agitation in
then pharmacological interventions may be required. These are dis-
people with moderate to severe dementia (Husebo et al., 2011). If
cussed in the section Pharmacological management of BPSD.
the patient requires admission to general hospital for appropriate
treatment of physical causes, hospital staff should be also aware that Environmental and social factors
simply the change in environment is sufficient to worsen confusion
Assessment of all dementia, but particularly dementia in the later
in people with dementia.
stages, requires an evaluation of the context or system in which
Psychological factors the person with dementia is living. The physical and social envi-
Depressed mood and anxiety in later stages of dementia are often ronment impacts upon people with dementia, and equally people
overlooked or dismissed as an inevitable part of the dementing and their behaviour will have an impact upon their environment.
process. Affective disorders may not be recognized as they may Consequently, assessments of the living situation and significant
present with challenging behaviours such as agitation, anxiety, irri- relationships are paramount. This applies whether the person is still
tability, calling out, crying, and clinging to people for reassurance. living in the family home or is in residential care or the general hos-
People with dementia are more likely to suffer with depression than pital. People with later dementia are easily disorientated, confused,
healthy people and anxiety may be the most prominent symptom. and frightened by new environments and unfamiliar people. A new
Depression and anxiety in dementia may respond well to antide- environment can precipitate behaviours that challenge, particularly
pressant therapies, so there is a need to proactively search for indi- if staff do not understand why patients are fearful and aggressive,
cations in the history or presentation that suggest an underlying shouting for help, or perhaps urinating inappropriately because they
treatable affective disorder. Risk factors for depression in dementia cannot find the toilet. This may be further aggravated if the person
have been reported as family history, past history, younger age at with dementia is in a general hospital where numerous transfers
onset of dementia, and female gender (Ballard et al., 1996a). An between wards often take place. The behaviour of people with later
unhappy childhood may also be relevant, particularly as people dementia also affects the environment around them. Behaviours
with later dementia are often reliving their childhood memories such as aggression, agitation, apathy, shouting, and sexually inap-
by this stage in the disease and no longer have the psychological propriate behaviour may help increase the risk of abuse—either
defence mechanisms to protect themselves against emotional pain. through neglect of the person or by more deliberate acts, such as
A major cause of depression in dementia is environment, and loca- physical aggression or restraint, psychological bullying, or sexual
tion should be kept in mind during assessment; the prevalence of exploitation. This triggers a cycle of events, as the abuse exacerbates
depression amongst older people in long-term care is up to 35% the behaviours that may intensify the abuse. In these situations, it is
important to adopt a nonjudgemental and sympathetic approach, addressing the cause of the behaviour. This is unacceptable and puts
with assessment of the whole system, which will hopefully create people with dementia at risk, as antipsychotics have been reported
opportunities to help all of those involved. to increase the risk of cardiovascular events, extra pyramidal side
Nonpharmacological management of BPSD effects, and falls, as well as increasing mortality (Schneider et al.,
2005; Ballard and Waite, 2006). Furthermore, a Department of
There is increasing evidence that nonpharmacological approaches
Health review (Banerjee, 2009) concluded that only 15–25% of the
are effective as first-line management for BPSD (Ballard et al.,
18,000 people with dementia prescribed antipsychotics every year
2009). Low-intensity nonpharmacological interventions include
actually get any benefit, but that antipsychotic prescribing for peo-
good person-centred care, improving social interactions (such as
ple with dementia causes 1800 additional strokes and 1600 addi-
one-to-one conversations or engaging in an activity based on the
tional deaths each year in the UK. Thus, the only indication for the
person’s interests or hobbies), promoting pleasant activities, and
use of antipsychotic medication is where the patient is severely dis-
gentle exercise (Department of Health, 2009; Alzheimer’s Society,
tressed and agitated and nonpharmacological methods have failed,
2011). Environmental factors that have been identified in the
and/or if the patient or others are at risk.
assessment must also be addressed. This may mean involving the
Antipsychotics have limited use in dementia. They do not
managers of the institution, e.g. requesting that a hospital restrict
improve restlessness or nonaggressive behavioural symptoms but
the transfer of people with dementia between different hospital
can reduce aggression and psychosis, although they are of less ben-
wards, or providing staff training programmes. More specific non-
efit in the longer term (Ballard and Waite, 2006; Schneider et al.,
drug treatments for BPSD recommended in the NICE dementia
2006a, 2006b). The two drugs with the best evidence of effective-
guidelines include aromatherapy, animal-assisted therapy, massage,
ness are risperidone and aripiprazole, but other atypical antipsy-
multisensory stimulation, and therapeutic dance or music therapy.
chotic drugs such as olanzapine and quetiapine are also used in
There is little evidence for the efficacy or effectiveness of these treat-
practice. Risperidone is the only antipsychotic licensed for the
ment approaches (Chung et al., 2002; Vink et al., 2004; Forbes et
treatment of severe and persistent aggression in dementia in the
al., 2008, 2009), but this may be due to the lack of adequately pow-
UK, but a recent survey of UK Old Age Psychiatry Consultants
ered studies. Most of these therapies, if individually tailored to the
reported that the majority used quetiapine as the most appropriate
person with dementia, will increase social interaction and reduce
agent for BPSD (Bishara et al., 2009). Current advice is that treat-
boredom, so it is likely that they will have therapeutic value.
ment should be targeted at specific symptoms, which should be
Pharmacological management of BPSD precisely described at baseline and at frequent follow-up; and that
If behavioural and psychological symptoms have not improved treatment is reduced and discontinued within 3 months (Banerjee,
through nonpharmacological approaches, or if there are significant 2009; Alzheimer’s Society, 2011), as 70% of people have no worsen-
risks to the person with dementia and/or others, then a pharmaco- ing of symptoms when antipsychotics are stopped (Ballard et al.,
logical approach may be appropriate. 2008). Antipsychotic medication should be avoided if at all pos-
If a depressive disorder is suspected, then treatment with antide- sible in Lewy body dementia, as neuroleptic sensitivity is common
pressants might be considered. There is only weak evidence from and antipsychotics can cause sudden death (McKeith et al., 1992;
systematic reviews and meta-analyses that antidepressants are Aarsland et al., 2005). Cholinesterase inhibitors are frequently used
effective in depression in people with dementia (Bains et al., 2002; first-line to treat psychosis in Lewy body dementia, but if these
Nelson and Devanand, 2011), but there are very few studies in these are ineffective then the antipsychotic quetiapine may be tried as it
meta-analyses and they are usually underpowered, compare differ- appears to have the least side effects of the atypical antipsychotics
ent antidepressants, and use different depression rating scales that and is sometimes effective (Zahodne and Fernandez, 2008). There
may not be particularly sensitive to change in people with dementia. is some evidence that the antipsychotic clozapine is useful in treat-
A recent UK multicentre study (Banerjee et al., 2011) compared the ing psychosis in Parkinson’s disease dementia, but no studies have
effectiveness of sertraline and mirtazapine with placebo in the larg- been carried out in Lewy body dementia to date (Weintraub and
est ever randomized trial of depression in dementia, and found that Hurtig, 2007).
the two antidepressants had no greater benefit and increased the Benzodiazepines are usually avoided in older people because
risk of adverse events. However, all three treatment groups showed they can increase confusion and the risk of falls. However, those
a consistent improvement in depression rating, indicating that non- with a shorter half-life, e.g. lorazepam, are sometimes used in the
drug treatments deployed by community mental health teams may short-term treatment of agitation. Antidepressants have also been
be the most helpful intervention for depression in dementia. Such used for the treatment of agitation and psychosis (rather than
an approach would reduce the risk of distressing side effects of anti- depression) in dementia; two Cochrane reviews reported that SSRIs
depressants which the patient cannot communicate easily to carers but not trazodone (Martinon-Torres et al., 2004; Seitz et al., 2011)
and that may simply worsen BPSD. Conversely, most clinicians will were effective in reducing agitation compared to placebo, but that
have personal experience that some people with depression and further trials were needed.
dementia do respond to antidepressant therapies, and SSRIs tend Cholinesterase inhibitors are used to delay the decline in cogni-
to be used first-line as they are considered to have the least adverse tive function in people with earlier dementia, and they may also be
side effects. of benefit in noncognitive symptoms, including depression, apathy,
If symptoms continue to cause distress or risk, then treatment and anxiety, but not agitation or aggression (Howard et al., 2007).
with antipsychotics may be appropriate. In recent years, there has Cholinesterase inhibitors are used for psychosis in Lewy body
been considerable concern about the risks associated with the use of dementia, in which the hallucinations are thought to be correlated
antipsychotic drugs to control challenging behaviour. Antipsychotic with the extent of cholinergic deficit. Rivastigmine has the best evi-
medication has sometimes been used to treat the symptoms without dence (McKeith et al., 2000), although the research evidence base
for the rest of the cholinesterase inhibitors is weak (Wild et al., interests into account) whilst also addressing their concerns too.
2003). Reanalysis of data from previous clinical trials has suggested These matters are often complex, raising emotive issues and very
that memantine may also be effective for the treatment of agitation personal judgements, so a frank discussion and subsequent shared
and aggression in later dementia (Gauthier et al., 2005; Cummings agreement with all involved is a worthwhile undertaking.
et al., 2006; Wilcock et al., 2008) and may be a safer alternative to Risk management
antipsychotic drugs; however, replication of results in new clinical
A good risk assessment will identify the risks to the person with later
trials is required to confirm this possibility.
dementia and possibly also to others. If the person with dementia is
Anticonvulsants, such as sodium valproate, have been reported
living alone in his or her own home then self-neglect, wandering,
as effective in the treatment of aggression and agitation in some
fire risks, falls, and medicines management are the commonest risks
small trials, but a Cochrane review (Lonergan and Luxenberg,
to self, and these should be systematically considered. The person
2009) reported that valproate was ineffective and was associated
with dementia, and particularly BPSD, may also be a risk to oth-
with an unacceptable rate of side effects. The prevalence of epilepsy
ers. For example, there may be physical aggression towards a spouse
in dementia is higher than in healthy individuals and is increased
who is no longer recognized and thought of as a stranger, or to a
in advanced dementia (Amatniek et al., 2006), so there may be a
care home assistant who is trying to provide personal care. People
theoretical basis for the use of anticonvulsants in later dementia,
with dementia may respond to hallucinations, delusions, or irrita-
but further research evidence is needed.
bility associated with depression or pain, by making accusations or
Management of sexually inappropriate behaviour hitting out at those nearest to them. Those nearest could be family
members or carers, but in a care home environment they might also
Sexually inappropriate behaviours have been defined as ‘sexual
include other residents, and even other residents’ visitors, some of
behaviours not suited for their context and which impair the care
whom may strike back. Under these circumstances, in the UK, it is
of the patient in a given environment’ (Black et al., 2005). What
recommended to inform the appropriate local authorities regarding
constitutes appropriateness will depend on the opinions and judge-
the safeguarding of the person with dementia and other vulnerable
ments of staff and relatives who are providing care, but sexually
adults, and to ensure that appropriate action is taken to reduce the
inappropriate behaviours (although present in only 7% of people
risks to the person with dementia and to other adults (Department
with dementia (Burns et al., 1990)) frequently cause concern, par-
of Health and Home Office, 2000; Association of Directors of Social
ticularly in institutional settings. The behaviours include sexual-
Services, 2005). People with later dementia may also live with their
ized remarks, grabbing, exposing, public masturbation, fondling,
families where young children are present. The risks to younger
sexual advances to staff or other residents, implied sexual acts
children should always be considered, including the risk of aggres-
(e.g. requesting unnecessary genital care), and attempts at geni-
sion and inappropriate sexual behaviour, but also the possible risk of
tal contact or penetration. As with other BPSD, it is important to
neglect if children are left in the care of the person with dementia.
search for physical and psychological causes. Sexual contact may
People with dementia are themselves vulnerable to abuse and
be a source of comfort in people with dementia who are anxious,
exploitation. Risk of abuse of people with dementia by others
depressed, or lonely. Dysuria secondary to a urinary tract infec-
increases as the dementia progresses, as they become less able to
tion can cause genital irritation, leading to scratching which can
protect themselves, and to complain. A recent systematic review
be misinterpreted (or turn into) masturbatory activity. The choice
found that one in four vulnerable older people were at risk of abuse
of treatment depends on the urgency of the situation, the type of
(Cooper et al., 2008), both in their own home and in institutional
behaviours, and the underlying causes. Physical disorders, delirium,
settings. The commonest forms of abuse of vulnerable adults are
affective disorders, and psychosis can all cause sexually inappropri-
psychological, physical, neglect, sexual, financial, and discrimina-
ate behaviours and should be treated. Psychological approaches can
tory. Signs that might alert health and social care professionals to
be used, and should include boundary setting: giving clear instruc-
the possibility of abuse include fear or agitation in the presence
tions for appropriate behaviour and provision of private rooms, as
of the abuser, bruising to the face and body, and the appearance
well as reducing the reinforcements of behaviour. If the behaviour
of pressure ulcers. In the UK, the Mental Capacity Act (2005) has
presents a risk to the person or others, then clinicians often resort to
introduced significant new protections from abuse for people who
a variety of pharmacological agents, such as SSRIs, antipsychotics,
lack capacity by creating the criminal offences of ill treatment or
or antiandrogens (e.g. cyproterone). There have been no published
wilful neglect: these offences can be committed by anyone respon-
randomized trials of pharmacotherapy for sexual disinhibition in
sible for the person’s care.
dementia (Ozkan et al., 2008) and successful management usually
requires a trial of several different medications. Legal issues
Sexual behaviour between residents in care homes often causes When people with dementia no longer have capacity to make deci-
concern. Some staff may feel that the behaviour is inappropriate, sions about their health, welfare, or financial matters and they have
particularly if the person with dementia still has an involved partner, not made advance directives, then decisions have to be made in
or that one or both residents are unable to consent to the behaviour. their best interests. Under the Mental Capacity Act (2005), health
Others may feel it is every person’s right to have a sexual relation- and social care professionals can work with families to come to
ship whilst ignoring potential vulnerability and risks. Health and decisions in the person’s best interests. In the case of a person refus-
social care professionals need to be able to establish the capacity ing pharmacological treatment, this may result in the decision to
of the people involved to consent to the sexual activity in the new covertly administer medication to the person with dementia in
relationship. If the person with dementia does not have capacity to order to improve his or her physical and/or psychological wellbe-
consent, then there has to be a sensitive discussion with staff and ing. Best-interest decisions cannot be made for financial matters,
family about the appropriateness of the relationship (taking best so if advanced provision has not been made, then, in the UK, the
family are advised to apply to the Court of Protection to ensure that might not be able to communicate their needs, e.g. that they are
they are legally protected to make financial decisions on behalf of in pain or thirsty. Health professionals providing end of life care
the person with dementia. should be skilled in interpreting change in activity or behaviour,
Most dementia can be managed by primary care, social care, or facial expressions, as well as the usual clinical signs. Diagnosing
and voluntary services. As dementia progresses there is increas- dying is also an important skill, so that the family can be aware
ing likelihood that behavioural and psychological symptoms will that death is imminent and religious or spiritual needs can be met.
emerge and, depending on the severity, secondary care services Drugs used in end of life care are usually to manage symptoms such
may be involved. Community mental health service will provide as terminal agitation and restlessness. Most commonly, haloperi-
support and care for people with dementia and their carers within dol or midazolam is used, although this may not be necessary if
the framework of the Care Plan Approach (Department of Health, the person is receiving opioid analgesia by a syringe driver. Other
2001). A care manager or care coordinator is allocated, who will drugs are usually discontinued, as long as this does not lead to dis-
facilitate the implementation of a care plan aimed at promoting tressing withdrawal effects. Respiratory tract secretions cause noisy
independent living for as long as possible. In relatively rare circum- breathing (the ‘death rattle’), which may be upsetting for the family
stances, people with dementia may require assessment and treat- and carers; these can be managed by repositioning of the patient,
ment in an inpatient mental health unit if they become severely suction if appropriate, and subcutaneous administration of hyos-
disturbed and need to be admitted for their own safety or the safety cine. The most important aspect of management is good communi-
of others. In the UK, they may need to be detained against their will cation with the family and primary care team to ensure people with
under Section 2 or Section 3 of the Mental Health Act (2007). Some dementia and their relatives experience a ‘good death’.
people with dementia do not require psychiatric admission but are
no longer capable of taking care of themselves at home and require Carer support
placement in institutional care. If they refuse but do not have capac- Advanced dementia can have a devastating impact on carers, who
ity to decide where they want to live, then the Mental Health Act are often older and may be frail themselves. Without support, they
(2007) can be used to require people to reside at a specified place frequently end up exhausted, isolated, and unable to cope. It is
in their best interests, under a guardianship order. However, most at this point that people with later dementia may present in cri-
people with dementia do not require care under the Mental Health sis, when their carers can no longer look after them due to their
Act; the move to residential care is usually unproblematic if the own physical or psychological ill health. Specialist dementia care
individuals involved handle it sensitively. On the other hand, some services can support carers by facilitating access to transport, sit-
people with dementia may express a wish to leave after the move ting services, day care, domiciliary care, night sitting services, and
has taken place. In the interest of their safety, care home staff may respite care in the person’s own home or in a care home, as well as
need to prevent the person from leaving the building. If the per- provide psychosocial support for the carers themselves. Recent UK
son with dementia continues to express a wish to leave, then, as the government policy has promoted the identification of dementia at
facility is not a mental health hospital, an application is made to the an earlier stage to prevent these crises and to hopefully delay the
appropriate supervisory body for assessment under Deprivation of permanent placement of the person with dementia into institu-
Liberty Safeguards (DoLs) to make a judgement as to whether the tional care (Department of Health, 2009).
person who lacks capacity can be lawfully detained in the facility.
This is discussed further in Chapter 63. Conclusion
End of life care The general management of dementia includes the understanding
of not only complex medical problems, with interplay of physical
The progressive nature of dementia means that a palliative care
and emotional wellbeing, but also complex social, ethical, and legal
approach to management is employed throughout the course of the
issues. Quality care cannot be achieved by families and unskilled
disease until death. This approach aims to optimize quality of life
workers alone. There is a need for education and support from skilled
for people with dementia and their carers and to ensure that they
dementia specialists with broad-ranging knowledge and a willing-
are supported with both dignity and respect and treated as persons
ness to educate others. Improved recognition of dementia will come
throughout. However, when people with dementia are at the end of
about with improved public education about the disease, and better
their lives, there are issues that are specific to dementia that need to
support for people with dementia will occur when the full impact on
be addressed. As people with dementia are unlikely to have capacity
them, their families, and carers is acknowledged. Better management
to make decisions about their healthcare, it will fall upon the carers
of the disease will happen when society accepts the increasing likeli-
of people with dementia to make choices in their best interests. This
hood that each and every individual will either suffer with dementia
is best done using an end of life care pathway, e.g. the Liverpool
or know someone with dementia, and whether we value human life
Care Pathway (Ellershaw and Ward, 2003), which allows decisions
sufficiently to improve that experience for all involved.
about feeding, nutrition, withdrawing or withholding treatment,
and cardiopulmonary resuscitation to be made calmly, ahead of References
time, taking into account all of the available information. Most
Aarsland, D., et al. (2005). Neuroleptic sensitivity in Parkinson’s disease and
deaths occur in hospitals, but this may not be the ideal environ-
parkinsonian dementias. Journal of Clinical Psychiatry, 66, 633–7.
ment for the person with dementia; advanced care planning allows
Alexopoulos, G. S., et al. (1988). Cornell Scale for Depression in Dementia.
the family to state clearly what their wishes are for their relative Biological Psychiatry, 23, 271–84.
and to ensure that ethical and legal principles are upheld, thus safe- Alzheimer’s Society (2011). Optimising treatment and care for people with
guarding the person with dementia, themselves, and the healthcare behavioural and psychological symptoms of dementia: a best practice guide
team who are providing care. By this stage, people with dementia for health and social care professionals. Alzheimer’s Society, London.
Amatniek, J. C., et al. (2006). Incidence and predictors of seizures in patients Burke, W. J., et al. (1989). Use of the Geriatric Depression Scale in dementia
with Alzheimer’s disease. Epilepsia, 47, 867–72. of the Alzheimer type. Journal of the American Geriatrics Society, 37,
Arber, A. and Gallagher, A. (2003). Breaking bad news revisited: the push for 856–60.
negotiated disclosure and changing practice implications. International Burns, A., Jacoby, R., and Levy, R. (1990). Psychiatric phenomena in
Journal of Palliative Nursing, 9, 166–72. Alzheimer’s disease. IV: disorders of behaviour. British Journal of
Ashford, J. W., et al. (2006). Should older adults be screened for dementia? Psychiatry, 157, 86–94.
Alzheimer’s Dementia, 2, 76–85. Cheston, R. and Jones, R. (2009). A small-scale study comparing the impact of
Ashford, J. W., et al. (2007). Should older adults be screened for dementia? It psycho-education and exploratory psychotherapy groups on newcomers
is important to screen for evidence of dementia! Alzheimer’s Dementia, to a group for people with dementia. Aging and Mental Health, 13,
3, 75–80. 420–5.
Association of Directors of Social Services (2005). Safeguarding adults. A Chung, J. C., et al. (2002). Snoezelen for dementia. Cochrane Database of
National Framework of Standards for good practice and outcomes in adult Systematic Reviews, CD003152.
protection work. ADSS, London. Clare, L., et al. (2003). Cognitive rehabilitation and cognitive training
Bains, J., Birks, J. S., and Dening, T. R. (2002). The efficacy of antidepressants for early-stage Alzheimer’s disease and vascular dementia. Cochrane
in the treatment of depression in dementia. Cochrane Database of Database of Systematic Reviews, CD003260.
Systematic Reviews, CD003944. Cole, M. G. (2004). Delirium in elderly patients. American Journal of Geriatric
Ballard, C. and Waite, J. (2006). The effectiveness of atypical antipsychotics Psychiatry, 12, 7–21.
for the treatment of aggression and psychosis in Alzheimer’s disease. Cooper, C., Selwood, A., and Livingston, G. (2008). The prevalence of elder
Cochrane Database of Systematic Reviews, CD003476. abuse and neglect: a systematic review. Age and Ageing, 37, 151–60.
Ballard, C., et al. (1996a). The prevalence, associations and symptoms of Copeland, J. R., et al. (1999). Depression in Europe. Geographical distribution
depression amongst dementia sufferers. Journal of Affective Disorders, among older people. British Journal of Psychiatry, 174, 312–21.
36, 135–44. Coupland, C., et al. (2011). Antidepressant use and risk of adverse outcomes
Ballard, C. G., et al. (1996b). A one-year follow-up study of depression in in older people: population based cohort study. British Medical Journal,
dementia sufferers. British Journal of Psychiatry, 168, 287–91. 343, d4551.
Ballard, C., Bannister, C., and Oyebode, F. (1996c). Depression in dementia Cummings, J. L., et al. (2006). Behavioral effects of memantine in Alzheimer
sufferers. International Journal of Geriatric Psychiatry, 11, 507–15. disease patients receiving donepezil treatment. Neurology, 67, 57–63.
Ballard, C., et al. (2008). A randomised, blinded, placebo-controlled trial in Department of Health (2001). National Service Framework for Older People.
dementia patients continuing or stopping neuroleptics (the DART-AD Department of Health, London.
trial). PLoS Medicine, 5, e76. Department of Health (2009). Living well with dementia: a National Dementia
Ballard, C., et al. (2009). Management of agitation and aggression associated Strategy. Department of Health, London.
with Alzheimer’s disease: controversies and possible solutions. Current Department of Health and Care Services Improvement Partnership (2005).
Opinions in Psychiatry, 22, 532–40. Everybody’s business. Integrated mental health service for older adults: a
Bamford, C., et al. (2004). Disclosing a diagnosis of dementia: a systematic service development guide. Department of Health, London.
review. International Journal of Geriatric Psychiatry, 19, 151–69. Department of Health and Home Office (2000). No secrets: guidance on
Banerjee, S. (2009). The use of antipsychotic medication for people with developing and implementing multi-agency policies and procedures to
dementia: time for action. Department of Health, London. protect vulnerable adults from abuse. Department of Health, London.
Banerjee, S., et al. (2011). Sertraline or mirtazapine for depression in Donaldson, C., Tarrier, N., and Burns, A. (1997). The impact of the symptoms
dementia (HTA-SADD): a randomised, multicentre, double-blind, of dementia on caregivers. British Journal of Psychiatry, 170, 62–8.
placebo-controlled trial. Lancet, 378, 403–11. Ellershaw, J. and Ward, C. (2003). Care of the dying patient: the last hours or
Barrett, A. M., et al. (2006). Short-term effect of dementia disclosure: how days of life. British Medical Journal, 326, 30–4.
patients and families describe the diagnosis. Journal of the American Enache, D., Winblad, B., and Aarsland, D. (2011). Depression in dementia:
Geriatric Society, 54, 1968–70. epidemiology, mechanisms, and treatment. Current Opinions in
Birks, J. (2006). Cholinesterase inhibitors for Alzheimer’s disease. Cochrane Psychiatry, 24, 461–72.
Database of Systematic Reviews, CD005593. Folstein, M.F., Folstein, S.E., and McHugh, P.R. (1975). ‘Mini-mental state’.
Bishara, D., et al. (2009). Expert opinion on the management of behavioural A practical method for grading the cognitive state of patients for the
and psychological symptoms of dementia (BPSD) and investigation clinician. Journal of Psychiatric Research, 12, 189–98.
into prescribing practices in the UK. International Journal of Geriatric Forbes, D., et al. (2008). Physical activity programs for persons with dementia.
Psychiatry, 24, 944–54. Cochrane Database of Systematic Reviews, CD006489.
Black, B., Muralee, S., and Tampi, R. R. (2005). Inappropriate sexual behaviors Forbes, D., et al. (2009). Light therapy for managing cognitive, sleep,
in dementia. Journal of Geriatric Psychiatry and Neurology, 18, 155–62. functional, behavioural, or psychiatric disturbances in dementia.
Bourne, J. (2007). Improving services and support for people with dementia. Cochrane Database of Systematic Reviews, CD003946.
National Audit Office, London. Gauthier, S., Wirth, Y., and Mobius, H. J. (2005). Effects of memantine on
Boustani, M., et al. (2006). Who refuses the diagnostic assessment for behavioural symptoms in Alzheimer’s disease patients: an analysis of the
dementia in primary care? International Journal of Geriatric Psychiatry, Neuropsychiatric Inventory (NPI) data of two randomised, controlled
21, 556–63. studies. International Journal of Geriatric Psychiatry, 20, 459–64.
Brayne, C., Fox, C., and Boustani, M. (2007). Dementia screening in primary Gilleard, C. (1984). Living with dementia: community care of the elderly
care: is it time? Journal of the American Medical Association, 298, 2409–11. mentally infirm. Croom Helm, Beckenham.
Brodaty, H. (1996). Caregivers and behavioral disturbances: effects and Gilleard, C. J. (1996). Family therapy with older clients. In: R. T. Woods (ed.)
interventions. International Psychogeriatrics, 8 Suppl 3, 455–58. Handbook of the clinical psychology of ageing. Wiley, Chichester.
Brodaty, H., et al. (2003a). Randomized controlled trial of different models of Homer, A. C. and Gilleard, C. (1990). Abuse of elderly people by their carers.
care for nursing home residents with dementia complicated by depression British Medical Journal, 301, 1359–62.
or psychosis. Journal of Clinical Psychiatry, 64, 63–72. Howard, R. J., et al. (2007). Donepezil for the treatment of agitation in
Brodaty, H., Green, A., and Koschera, A. (2003b). Meta-analysis of Alzheimer’s disease. New England Journal of Medicine, 357, 1382–92.
psychosocial interventions for caregivers of people with dementia. Huppert, F. A., et al. (1995). CAMCOG—a concise neuropsychological test
Journal of the American Geriatric Society, 51, 657–64. to assist dementia diagnosis: socio-demographic determinants in an
elderly population sample. British Journal of Clinical Psychology, 34 (Pt Reynolds, C. F., III, et al . (2006). Maintenance treatment of major
4), 529–41. depression in old age . New England Journal of Medicine , 354 ,
Husebo, B. S., et al. (2011). Efficacy of treating pain to reduce behavioural 1130–8.
disturbances in residents of nursing homes with dementia: cluster Robert, P. H., et al. (2005). Grouping for behavioral and psychological
randomised clinical trial. British Medical Journal, 343, d4065. symptoms in dementia: clinical and biological aspects. Consensus paper
Kitwood, T. M. (1997). Dementia reconsidered: the person comes first. Open of the European Alzheimer disease consortium. European Psychiatry, 20,
University Press, Buckingham. 490–6.
Kitwood, T. and Bredin, K. (1992). Towards a theory of dementia care: Rose, G. (1985). Sick individuals and sick populations. International Journal
personhood and well-being. Ageing Society, 12, 269–87. of Epidemiology, 14, 32–8.
Lecouturier, J., et al. (2008). Appropriate disclosure of a diagnosis of Royal College of Psychiatrists (2005). Who cares wins. Improving the
dementia: identifying the key behaviours of ‘best practice’. BMC Health outcome for older people admitted to the general hospital: guidelines for
Services Research, 8, 95. the development of liaison mental health services for older people. RCP,
Lindesay, J., et al. (2002). The second Leicester survey of memory clinics London.
in the British Isles. International Journal of Geriatric Psychiatry, 17, Royal College of Physicians (2009). Advance care planning: national guidelines.
41–7. RCP, London.
Lonergan, E. and Luxenberg, J. (2009). Valproate preparations for agitation in Royal College of Psychiatrists (2011). Our invisible addicts. First report of the
dementia. Cochrane Database of Systematic Reviews, CD003945. Older Persons’ Substance Misuse Working Group of the Royal College of
Martinon-Torres, G., Fioravanti, M., and Grimley, E. J. (2004). Trazodone Psychiatrist. College Report CR16. RCP, London.
for agitation in dementia. Cochrane Database of Systematic Reviews, Sampson, E. L., et al. (2009). Dementia in the acute hospital: prospective
CD004990. cohort study of prevalence and mortality. British Journal of Psychiatry,
McKeith, I., et al. (1992). Neuroleptic sensitivity in patients with senile 195, 61–6.
dementia of Lewy body type. British Medical Journal, 305, 673–8. Schneider, L. S., Dagerman, K. S., and Insel, P. (2005). Risk of death with
McKeith, I. G., et al. (1996). Consensus guidelines for the clinical and atypical antipsychotic drug treatment for dementia: meta-analysis of
pathologic diagnosis of dementia with Lewy bodies (DLB): report of the randomized placebo-controlled trials. Journal of the American Medical
consortium on DLB international workshop. Neurology, 47, 1113–24. Association, 294, 1934–43.
McKeith, I., et al. (2000). Efficacy of rivastigmine in dementia with Lewy Schneider, L. S., Dagerman, K., and Insel, P. S. (2006a). Efficacy and
bodies: a randomised, double-blind, placebo-controlled international adverse effects of atypical antipsychotics for dementia: meta-analysis
study. Lancet, 356, 2031–6. of randomized, placebo-controlled trials. American Journal of Geriatric
McShane, R., Areosa, S. A., and Minakaran, N. (2006). Memantine for Psychiatry, 14, 191–210.
dementia. Cochrane Database of Systematic Reviews, CD003154. Schneider, L. S., et al. (2006b). Effectiveness of atypical antipsychotic drugs
Mental Health Foundation (2011). The milk’s in the oven. A booklet about in patients with Alzheimer’s disease. New England Journal of Medicine,
dementia for children and young people. MHF, London. 355, 1525–38.
Moos, I. and Bjorn, A. (2006). Use of the life story in the institutional care Seitz, D. P., et al. (2011). Antidepressants for agitation and psychosis in
of people with dementia: a review of intervention studies. Ageing and dementia. Cochrane Database of Systematic Reviews, CD008191.
Society, 26, 431–54. Sheikh, J. and Yesavage, J. (1986). Geriatric Depression Scale (GDS): Recent
Moriarty, J., Sharif, N., and Robinson, J. (2011). SCIE research briefing 35: evidence and development of a shorter version. Clinical Gerontology, 5,
black and minority ethnic people with dementia and their access to support 165–73.
and services. Social Care Institute for Excellence, London. Sorensen, S., Pinquart, M., and Duberstein, P. (2002). How effective
Mottram, P., Wilson, K., and Strobl, J. (2006). Antidepressants for depressed are interventions with caregivers? An updated meta-analysis. The
elderly. Cochrane Database of Systematic Reviews, CD003491. Gerontologist, 42, 356–72.
Moye, J., et al. (2004). Capacity to consent to treatment: empirical comparison Spijker, A., et al . (2008). Effectiveness of nonpharmacological
of three instruments in older adults with and without dementia. The interventions in delaying the institutionalization of patients with
Gerontologist, 44, 166–75. dementia: a meta-analysis. Journal of the American Geriatric Society,
NICE (2009). Depression in adults: full guidance. Clinical Guideline 90. 56, 1116–28.
National Institute for Health and Clinical Excellence, London. Stokes, G. (2000). Challenging behaviour in dementia: a person-centred
NICE (2011). NICE technology appraisal guidance 217. Donepezil, galantamine, approach. Winslow Press, Bicester.
rivastigmine and memantine for the treatment of Alzheimer’s disease Taylor, D., Paton, C., and Kapur, S. (2009). The Maudsley prescribing guidelines,
(review of NICE technology appraisal guidance 111). National Institute 10th revised edition. Informa Healthcare Informa Pharmaceuticals,
for Health and Clinical Excellence, London. London.
NICE-SCIE (2006). Dementia: supporting people with dementia and their Teri, L., et al. (1997). Behavioral treatment of depression in dementia patients:
carers in health and social care. Clinical Guideline 42. National Institute for a controlled clinical trial. Journal of Gerontology Series B. Psychological
Health and Clinical Excellence and Social Care Institute for Excellence, Sciences and Social Sciences, 52, 159–66.
London. Teri, L., et al. (2003). Exercise plus behavioral management in patients with
Neal, M. and Briggs, M. (2003). Validation therapy for dementia. Cochrane Alzheimer disease: a randomized controlled trial. Journal of the American
Database of Systematic Reviews, CD001394. Medical Association, 290, 2015–22.
Nelson, J. C. and Devanand, D. P. (2011). A systematic review and Thakur, M. and Blazer, D. G. (2008). Depression in long-term care. Journal of
meta-analysis of placebo-controlled antidepressant studies in people the American Medical Directors Association, 9, 82–7.
with depression and dementia. Journal of the American Geriatric Society, Tombaugh, T. N. and McIntyre, N. J. (1992). The mini-mental state
59, 577–85. examination: a comprehensive review. Journal of the American Geriatrics
Ozkan, B., et al. (2008). Pharmacotherapy for inappropriate sexual behaviors Society, 40, 922–35.
in dementia: a systematic review of literature. American Journal of Tracy, C. S., et al. (2004). To tell or not to tell? Professional and lay perspectives
Alzheimer’s Disease and Other Dementias, 23, 344–54. on the disclosure of personal health information in community-based
Pinner, G. and Bouman, W. P. (2003). Attitudes of patients with mild dementia dementia care. Canadian Journal of Aging, 23, 203–15.
and their carers towards disclosure of the diagnosis. International Vink, A. C., et al. (2004). Music therapy for people with dementia. Cochrane
Psychogeriatrics, 15, 279–88. Database of Systematic Reviews, CD003477.
Weintraub, D. and Hurtig, H. I. (2007). Presentation and management Wilson, K. C., Mottram, P. G., and Vassilas, C. A. (2008). Psychotherapeutic
of psychosis in Parkinson’s disease and dementia with Lewy bodies. treatments for older depressed people. Cochrane Database of Systematic
American Journal of Psychiatry, 164, 1491–8. Reviews, CD004853.
Wilcock, G. K., et al. (2008). Memantine for agitation/aggression and Wimo, A. and Prince, M. (2010). World Alzheimer’s Report 2010: the global
psychosis in moderately severe to severe Alzheimer’s disease: a pooled economic impact of dementia. Alzheimer’s Disease International, London.
analysis of 3 studies. Journal of Clinical Psychiatry, 69, 341–8. Woods, B., et al. (2005). Reminiscence therapy for dementia. Cochrane
Wild, R., Pettit, T., and Burns, A. (2003). Cholinesterase inhibitors for Database of Systematic Reviews, CD001120.
dementia with Lewy bodies. Cochrane Database of Systematic Reviews, Zahodne, L. B. and Fernandez, H. H. (2008). Pathophysiology and treatment of
CD003672. psychosis in Parkinson’s disease: a review. Drugs and Aging, 25, 665–82.
CHAPTER 40
Delirium
Jenny Hogg
Delirium (acute confusional state) is a clinical syndrome that com- Diagnostic Criteria and Assessment Tools
monly affects older people with an acute medical illness. Whilst
rates vary from study to study, it would seem to affect approximately for Delirium
30% of hospital inpatients aged over 65 years, but in the community The main diagnostic criteria for delirium in current use are defined
the prevalence is found to be lower at approximately 1–2%, rising in the fourth edition of the Diagnostic and Statistical Manual of
steeply with age to 14% in those over 85 years, but even this may Mental Disorders (DSM-IV) (American Psychiatric Association,
be an underestimate (Folstein et al., 1991; Rahkonen et al., 2001). 1994), which are similar to the International Classification of
Despite being the most common psychiatric condition affecting Diseases 10 (ICD-10) (WHO, 1993) criteria, and to those in the
hospital inpatients, it is predominantly managed by general physi- recent UK guidelines (Potter and George, 2006). These state that in
cians and geriatricians, many of whom have had no specific train- order to make the diagnosis of delirium, a patient must show each
ing to recognize or deal with it and its consequences. of the features 1–4 listed in Table 40.1. There have been numerous
Delirium is characterized by an abrupt global impairment of cog- bedside assessments for the specific diagnosis of delirium, although
nitive processes, including thinking, remembering, and perceiving. in practice these are rarely used by physicians and junior medical
There are attention abnormalities with a decreased awareness of staff, the diagnosis usually being reached on clinical grounds. The
environment and self. There is a defective ability to extract, process, most commonly used diagnostic tool is the Confusion Assessment
and retain information. The person’s grasp of situations is faulty and Method (CAM) (see Table 40.2). It is designed to be used by
there is a diminished capacity to act in the customary purposeful, nonpsychiatrists and can be used by trained nursing staff. It has
sustained, and goal-directed manner. Symptoms usually fluctuate, been shown to be sensitive, specific, and reliable in the detection of
being more prominent at night. The person may display hyperac- delirium (Inouye et al., 1990). A further, more recent study has sug-
tive or hypoactive behaviour, or a mixture of the two. There must be gested that the CAM has the best evidence for diagnostic accuracy
a precipitating medical or physical cause. of delirium and takes just 5 min to administer (Wong et al., 2010).
Historical Perspective Subsyndromal Delirium

Outside medical circles, the term ‘delirium’ has been used meta- Subsyndromal delirium is a term that can be applied to patients
phorically to denote any form of mental aberration, including possessing one or more, but not all, of the symptoms of delirium.
uncontrollable excitement or emotion. To some extent this wider The risk factors are identical to those for delirium and, as one
meaning persists to the present day. Delirium is derived from the may guess, the outcome for such patients is intermediate between
Latin words ‘de’ meaning ‘out of ’ and ‘lira’ meaning the ‘furrow’; those with no diagnostic features of delirium and those with the
this produced the now obsolete English verb ‘delire’, meaning to go full-blown syndrome. Patients with subsyndromal delirium require
wrong, to go astray from reason, to rave, to wander in mind, or to identification and clinical attention in line with management of
be delirious or mad (Murray, 1897; Hart, 1936). The clinical con- delirium in order to achieve the best outcome (Levkoff et al., 1996;
stellation of a symptomatic acute mental disorder associated with Cole et al., 2002).
fever, that featured cognitive and behavioural disturbance as well
as disruption of sleep, which improved when the fever improved,
was recognized 2500 years ago by Hippocrates and his peers. They Making the Diagnosis of Delirium
made the distinction between this syndrome (delirium) and mental Delirium is underdiagnosed for a variety of reasons and may be
illness of unknown cause (dementia) (Lipowski, 1990). Sadly, some unrecognized by doctors and nurses in up to two-thirds of cases
2000 years later, the syndrome of delirium is often undiagnosed as (Treloar and Macdonald, 1995; Inouye et al., 2001; Foreman and
clinicians fail to appreciate the distinction so eloquently described Milisen, 2004). This is a cause of great concern as delirium is
in ancient times. present in at least 10% of acute unselected medical admissions in
Table 40.1 DSM-IV criteria for delirium specifically questioned, the physician may remain unaware of noc-
turnal problems with the patient or may assume it is the patient’s
1 Disturbance of consciousness (i.e. reduced clarity of awareness of the usual behaviour. The patient may not have the opportunity to be
environment) with reduced ability to focus, sustain, or shift attention.
assessed by a multidisciplinary team and the case may only come to
2 A change in cognition (such as memory deficit, disorientation, language light when there are problems discharging the patient. This is a par-
disturbance) or the development of a perceptual disturbance that is not ticular problem with patients living in residential care, as the carers
better accounted for by a pre-existing or evolving dementia. who know the patient best in terms of activities of daily living rarely
3 The disturbance develops over a short period of time (usually hours to visit during the hospital stay.
days) and tends to fluctuate during the course of the day. In medical units operating an unselected admissions policy, an
4 There is evidence from the history, physical examination, or laboratory old age psychiatry liaison service is vital, especially for dealing with
findings that the disturbance is caused by the direct physiological this type of scenario. In many hospitals, the nursing teams on gen-
consequences of a general medical condition, substance intoxication, or eral medical wards have received training in the area of delirium
substance withdrawal. by their liaison team and have the opportunity to refer to the team
themselves. There are now a few specialist dual-care units around
the country where patients with delirium can be admitted for
Table 40.2 Confusion Assessment Method (CAM). The diagnosis of assessment and management by a multidisciplinary team. Such
delirium requires the presence of features 1 and 2 and either 3 or 4 teams will usually consist of a geriatrician, old age psychiatrist,
nurses with both mental health and physical health training, occu-
1 Acute onset and fluctuating course pational therapists, physiotherapists, and psychologists. Currently,
This feature is usually obtained from a family member or nurse and is however, the majority of hospitals rely solely on a liaison model or
shown by positive responses to the following questions: a consultation model.
Is there any evidence of an acute change in the patient’s mental status from In hospital settings, the diagnosis should be considered in all
baseline? older patients who have cognitive impairment, both at presenta-
Did the abnormal behaviour fluctuate during the day, i.e. come and go, or tion and during their inpatient stay. A key factor in diagnosis is
increase or decease in severity? obtaining a good premorbid history from a close relative or carer.
2 Inattention A baseline assessment of cognitive function in all older patients is
This feature is shown by a positive response to the question: Did the patient useful, and many units now perform a routine Abbreviated Mental
have difficulty focusing attention, e.g. being easily distractible or having Test Score (Jitapunkul et al., 1991) on all admissions to the acute
difficulty keeping track of what was being said? medical unit as part of an admission pro forma, or they may use
3 Disorganized thinking
the Mini-Mental State Examination (MMSE) (Folstein et al., 1975),
or other tests such as clock drawing (Fisher and Flowerdew, 1995).
Was the patient’s thinking disorganized or incoherent, such as rambling or
irrelevant conversation, unclear or illogical flows of ideas, or unpredictable
Repeating such ratings at intervals during the inpatient stay may
switching from subject to subject? highlight the fluctuations in cognition that characterize the deliri-
ous patient O’Keefe et al., 2005). Patients with dementia will usually
4 Altered level of consciousness
have a more fixed cognitive impairment. Serial cognitive assess-
This feature is shown by any answer other than alert to the following ment is particularly useful in detecting the 15% of older admissions
question: Overall, how would you rate this person’s level of consciousness?
who will develop a delirium during their stay, and as delirium is
(Alert [normal]; vigilant [hyperalert]; lethargic [drowsy, easily aroused]; more common in those with dementia, changes in cognitive test
stupor [difficult to arouse]; coma [unrousable]) scores may provide valuable information about a developing delir-
ium in these high-risk patients too. Premorbid baseline informa-
tion is especially helpful in this group, particularly if obtained from
a typical hospital in the UK (George et al., 1997), with a further a carer or relative.
15% of older people developing the delirium syndrome during
their hospital stay (Inouye et al., 1999). There are often problems
with making the diagnosis from the beginning of the patient’s ill- Differential Diagnosis of Delirium
ness. The initial assessment is undertaken either by the accident The main differential diagnoses of delirium are dementia and
and emergency doctors or in the general practice setting. As cases depression. As all three conditions are common in the older hos-
of delirium come to light acutely, and often during the evening or pitalized population, it is hardly surprising to find the patient may
night, the patient’s usual GP is not always available. Many patients have two or three of these conditions to a greater or lesser degree.
already have a diagnosis of a dementia, and the acute confusion Dementia is the main risk factor for developing delirium. It is
observed is often assumed to be the patient’s norm, especially if thought that up to two-thirds of cases of delirium have coexistent
patients are unaccompanied during their assessment. Once admit- dementia (Inouye, 1997; Cole, 2004). To a lesser degree, depression
ted to hospital, the patient is usually under the care of a nongeriatri- (Gustafson et al., 1988; Foreman, 1989) has been identified as a risk
cian. Nongeriatricians have rarely received training in any aspects factor for delirium.
of old age psychiatry and will often only recognize the hyperac- Delirium is distinguished from dementia by several features.
tive phenotype of delirium. There may be an underappreciation of The most important is the mode of onset. Delirium develops over
the importance of making a diagnosis due to the lack of specific hours to days, whereas dementia usually commences gradually
medical treatment and unawareness of the clinical consequences, over months or years. Fluctuations in mental state typically occur
which will be discussed later in this chapter. Unless nursing staff are frequently and rapidly, within minutes or hours, in delirium, but in
CHAPTER 40 delirium 531
dementia changes occur gradually. Although more subtle rapid cog- they are fearful the delirious patient will fall. The hyperactive group
nitive fluctuations do occur in dementia with Lewy bodies (DLB), is at high risk of falling as they will often pace the ward with poor
cognitive fluctuations are a key diagnostic feature (see Chapter appreciation of danger from, for example, wet floors and obstacles.
35). Arousal levels are also abnormal in delirium, the patient being Although they may usually use a walking aid, they will often leave it
either hyperaroused or somnolent, and psychomotor activity is behind and grasp onto walls and pieces of hospital furniture, many
increased or decreased along with these changes. Delusions and of which are on wheels. Such patients require a high level of obser-
hallucinations are more common in delirium than in dementia of vation in order to prevent falls, and are probably best managed on
the Alzheimer’s type and vascular dementia. In DLB, hallucinations specialist units in cubicles. Sadly, there is a lack of such accommo-
are common, and other features of this condition such as parkin- dation in most hospitals.
sonism should be sought to aid the diagnosis. These patients are About a third of patients with delirium have the ‘hypoactive’ type.
prone to delirium-like episodes, and if the frequency and severity Such patients are underdiagnosed as they do not cause the upheaval
of hallucinations suddenly increase this should be suspected. DLB on the ward their hyperactive peers do. Classically patients have
is therefore an important differential diagnosis in unexplained poor oral intake and can be recognized on the ward slumped over
delirium. Vascular dementia of the multi-infarct type is character- their meal tray with food falling out of their mouths. They may fall
ized by stepwise deteriorations of cognitive function. Occasionally, asleep mid-way through a conversation with their relatives. These
one of these stepwise changes can cause an abrupt change in the patients require vigilance to avoid pressure damage, malnutrition,
patient’s mood or personality, particularly if the new lesion is in and dehydration. They often require intravenous fluids, special
the frontal lobe. This can manifest in a similar way to a hyperac- mattresses, and intravenous medication. Often one may see the
tive delirium and can be very difficult to differentiate. Other acute untouched morning medication pot in the afternoon, which con-
lesions causing apathy and poor volition can similarly be difficult to tains the vital antibiotics required to treat the underlying medi-
separate from the hypoactive type of delirium. cal illness. These patients often require feeding and supervision to
Depression may also be mistaken for the hypoactive type of take their medications. Despite often only modest rises in infection
delirium. Patients may present to hospital with a relatively minor and inflammatory markers, it may be wisest to choose intravenous
illness, but because of their failure to engage with staff, poor nutri- antibiotics to ensure adequate delivery, as this may well speed up
tional intake, and lack of volition to attend to their own personal recovery, reduce hospital length of stay, and therefore minimize the
hygiene, delirium is suspected. Often the dilemma can be resolved complications described.
with a good premorbid account, which will usually describe a grad- The most common phenotype of delirium is the mixed type
ual rather than sudden onset of such symptoms. The problem with which affects almost half of all cases. As its name suggests, the
distinguishing the two can arise because of difficulties in establish- patient will fluctuate between hyperactivity and hypoactivity. The
ing this, particularly if the patient lives alone with no close relatives former is often only evident at night, and careful questioning of
or friends. The two diagnoses can usually be separated by perform- the night nursing staff is often needed to make the diagnosis. By
ing cognitive testing and observing for cognitive fluctuations and day, these patients can seem ready to go home, and indeed may
by observing the mood, which is more constantly low in depression often demand discharge as they lack insight and recollection of
but typically fluctuates in delirium. the previous night’s events. Behaviour and sleep charts are useful
Similarly, mania can occasionally be mistaken for a hyperactive in this situation, as they will highlight such problems to both the
delirium. Again, the history is important as this will usually reveal daytime nursing and medical staff as well as the patient and rela-
a gradual acceleration of the person’s heightened mood over days tives. Sensitive handling of such a situation is required, and it is
to weeks, rather than the more precipitous changes in mood seen worth involving close relatives at an early stage so they may help
in delirium. persuade their loved one it is too early for home. Sadly, with cur-
Thus the key to arriving at a diagnosis of delirium is the history. rent bed pressures, such patients are often discharged too early and
The patient is often unable to give a lucid history, so it is impera- require readmission. This is an unhappy consequence for patients
tive that a history is taken from a relative or carer. Careful ques- and their families and care must be taken in planning such patients’
tioning about the patient’s usual cognition, physical abilities, and discharge to avoid this.
continence is vital, as often the acute medical condition causing the
delirium is minor and may need to be looked for carefully if the
diagnosis is suspected. Causes of Delirium
To attribute delirium to a ‘cause’ one must be able to relate the acute
confusion to an observed factor that is known to cause confusion,
Clinical Types of Delirium and then see improvement following treatment or cessation of the
There are three phenotypes of delirium: hyperactive (classical), factor responsible (Francis et al., 1990). We now recognize that
hypoactive, and mixed. The hyperactive or ‘florid’ type, despite delirium not only occurs with fever and infection but also can occur
being the least common type, is the most frequently diagnosed. in many other physical health problems. Several researchers have
These patients often display signs of increased sympathetic activ- analyzed the causes of delirium in hospital admissions. Findings
ity, including tachycardia, sweating, dilated pupils, flushed face, are similar from most studies. One series (George et al., 1997) is
and increased blood pressure. They are restless and repeatedly seek representative of the delirium experience in most district general
reassurance from staff. They will often interfere with other patients’ hospitals in the UK. This found that the major causes were infec-
belongings, causing distress. Fellow patients with acute medical ill- tion (one-third), stroke (one-tenth), and medication (one-tenth).
nesses find these patients difficult to share a room with. They are Other causes are detailed in Table 40.3. Several patients (25%) had
often kept awake all night, and by day find it difficult to relax as more than one contributory factor.
Table 40.3 Causes of delirium in 171 sequential medical admissions delirium (Simpson and Kellett, 1987). The more factors present, the
higher the risk of developing a delirium. Other studies (Schor et al.,
Cause Number of cases Percentage 1992) have identified in addition male sex, fracture on admission,
Infection 73 34 and neuroleptic and narcotic use, and although this study failed to
Chest 40 – identify medications with anticholinergic effects as a risk factor, a
Urinary 25 – further study (Han et al., 2001) showed that exposure to such medi-
cations is associated with a subsequent increase in the severity of
Other 8 –
delirium symptoms.
Stroke 24 11 Iatrogenic factors that may precipitate a delirium in susceptible
Drugs (analgesics, hypnotics, sedatives, 24 11 persons, such as those with baseline factors described in the para-
and anticholinergics) graph above, have also been identified (Inouye and Charpentier,
Myocardial infarction 11 5 1996). These include use of physical restraints, malnutrition, more
than three new medications added, and use of a bladder catheter.
Fractures 10 5
Hip 7 –
Other 3 –
Prevention of Delirium During
Carcinoma 10 5
a Hospital Stay
With this in mind, one can see how attempts can be made to reduce
Fluid and electrolyte imbalance 9 4
the incidence of delirium in hospital patients. Some authors suggest
Heart failure 8 4 that delirium could be prevented in up to a third of patients who
Diabetes (hyper- or hypoglycaemia) 7 3 develop delirium during their hospital stay (Inouye et al., 1999;
Marcantonio et al., 2001).
Peripheral vascular disease/gangrene 6 3
Although several trials into prevention of delirium with thera-
Alcohol eithdrawal 6 3 peutic agents have been conducted, these are largely in the elec-
Gastrointestinal bleed 5 2 tive orthopaedic surgery groups. Trials using haloperidol pre- and
postoperatively showed no efficacy in reducing the incidence of
Respiratory failure 5 2
postoperative delirium but did appear to reduce length of stay and
Pulmonary embolus 4 2 the severity and duration of the delirium (Kalisvaart et al., 2005).
Anaemia 4 2 Although theoretically the acetylcholinesterase inhibitors should be
of use in the prevention of delirium, small trials using such agents
Perforation of duodenal ulcer 2 1
as donepezil have failed to demonstrate any benefit in prevention or
Subdural haematoma 2 1 treatment of delirium postoperatively (Liptzin et al., 2005).
Brian tumour 1 0.5 Groups of patients with acute fractured neck of femur have also
been studied. These patients are often frail and old and frequently
Miscellaneous 6 3
share the baseline characteristics that predispose to delirium. Several
Total 217* studies have shown that delirium develops in between 35% and 65%
* 42 patients (25%) had two or more equally contributory causes. of such cases without any preventative measures (Gustafson et al.,
(From George et al. 1997.) 1988). The interventions shown to be of particular value in the popu-
lation with a fractured neck of the femur included avoidance of peri-
operative hypotension, postoperative hypoxaemia, prompt surgical
Predisposing and Precipitating Factors for intervention (within 24 h where possible), use of spinal anaesthe-
sia (as opposed to general anaesthesia), and more aggressive use of
Delirium blood transfusion (Gustafson et al., 1991). Patients with communica-
Researchers in the field of delirium have looked at the factors that tion difficulties (e.g. deafness, poor vision, dysarthria, and dysphasia,
are common to those patients presenting with delirium. In older as well as pre-existing dementia) may not be able to express their
people, delirium is rarely caused by a single factor. It is usually a pain. A study of hip fracture patients showed that those patients who
composite of the mental and physical vulnerability of the patient developed delirium received significantly lower amounts of analge-
and the severity of the physical insults. As one may expect, the more sia compared with those who did not develop delirium (Robinson
vulnerable the patient, the easier it is for that patient to develop delir- and Vollmer, 2010). It is hypothesized that pain may be an important
ium. Predisposing factors have been shown in many studies to be component in this group and it is recommended that staff should be
increasing age, pre-existing dementia, severity of illness, metabolic aware of this and offer analgesia regularly.
and electrolyte imbalance, the use of psychoactive medications, and Nursing interventions shown to be of use include orientation
a previous episode of delirium (Gustafson et al., 1988; Rockwood, to time, place, and situation, correction of sensory deficits, and
1989; Rogers et al., 1989; Francis et al., 1990; Trzepacz and Francis, increased continuity of care (Williams et al., 1985). Later work
1990; Schor et al., 1992). Similarly, other studies have identified has shown that a geriatrician consultation with optimization of
risk factors for those who develop delirium during their hospital pre-existing medical conditions can help further (Marcantonio
stay. These factors include visual impairment, increasing severity et al., 2001); in this study, not only was the incidence of delirium
of illness, cognitive impairment, and dehydration (Inouye et al., reduced, but also in those who developed delirium the duration
1993). Preoperative anxiety may be a risk factor for postoperative and severity were reduced.
A more difficult group to study are older patients presenting to only where the toilet facilities are but also how to summon assist-
an acute medical unit. The incidence of delirium at presentation to ance if needed. The balance between adequate lighting to ensure
hospital can be even higher than in the population with a fractured patients do not sustain a fall on visiting the bathroom and dark-
neck of the femur, although the two groups share many other fea- ness to assist patients to fall asleep and maintain sleep is a tricky
tures. In those without delirium at presentation, the rates of devel- one. It may be useful to establish which patients regularly use the
oping delirium during a hospital stay can be reduced from 15% to bathroom during the night and position them on the ward accord-
10% with a multicomponent intervention programme (Inouye et ingly. Patients known to be noisy at night may benefit from a cubi-
al., 1999). These interventions included the following protocols: cle, but as cubicle accommodation is a premium in the UK this is
orientation (communication boards, reorientation to environment, not always possible.
getting to know staff ), therapeutic activities (discussion of current
events, structured reminiscence, word games), nonpharmacological Early Mobilization
measures to promote sleep (warm drinks, relaxation tapes, music,
back massage, restoration of day/night pattern), sleep enhancement For patients who are independently ambulant, early mobilization is
(noise reduction at night, schedule adjustments, e.g. of medica- easily achievable. The target population, however, often use a mobil-
tion and procedures), and early mobilization (ambulation or bed/ ity aid such as a walking frame, stick(s), or crutches. Many patients
chair-based exercises three times a day, minimal use of immobi- fail to bring their mobility aid to hospital when acutely admitted,
lizing equipment, e.g. bladder catheters, physical restraints). For which further hampers the aim of early mobilization. Asking fam-
those with specific problems, a visual protocol, hearing protocol, ily members to bring such equipment in at an early stage is useful,
and dehydration protocol were used as appropriate. Although this especially if the admission falls over a weekend, as physiotherapy
research was done in the US, it should be transferable to clinical staff are not generally available for mobility assessments over this
practice elsewhere, and the new UK guidelines (Potter and George, period in the UK. A detailed history of the patient’s usual mobility
2006) embrace many of these interventions. is of great use, and attempts to maintain it will not only assist in
prevention of delirium but also facilitate discharge.
Patients may have been catheterized on admission because of
Orientation continence problems or to monitor fluid balance. This should be
Unfortunately many acute units in the UK seem unable to accom- reviewed and the catheter removed at the earliest possible opportu-
modate patients in the same location for the length of their hospital nity. The initial decision to catheterize should be made carefully by
stay. Patients can be moved around the hospital in excess of four senior staff unless the patient is in urinary retention or is critically
times during their admission, and it is little wonder they become unwell. Cot sides and other forms of physical restraint should be
disorientated. An older patient with a simple urinary tract infec- avoided where possible (Evans and Strumpf, 1989; Lofgren et al.,
tion is often first in line to be boarded from a medical ward to a 1989; Sullivan-Marx, 1994).
surgical ward to make way for a patient perceived to be more medi-
cally unwell, with little regard for the possible sequel of a delirium
(Mattice, 1989).
Management of the Patient with Delirium
Initial clinical management of delirium
Therapeutic Activities As mentioned in the section Making the Diagnosis of Delirium,
the key lies in establishing an acute change in cognition and behav-
The use of cognitive stimulation is helpful in preventing delirium.
iour. Often this information is only available from relatives and car-
Many relatives will unwittingly engage in this activity with their
ers. In parallel with investigating the baseline status of the patient,
loved one, but there are often short visiting hours on wards, with
the clinician must perform appropriate medical investigations. In
patients frequently complaining that the hospital day feels very
light of the information known about the likely causes of delirium
long. Good access to newspapers, reading books, and the television
there are certain investigations required for all patients present-
is of help, as well as more structured activities with therapy and
ing with acute confusion. Often, frail older people fail to display
ward staff. Patients will often enjoy interaction with other patients,
the classical signs of, for example, a chest infection, such as fever,
but it will often take members of staff to introduce patients to each
cough, increased sputum production, and dyspnoea. Instead they
other to ‘break the ice’. Chronic short staffing can be a barrier to
may present with anorexia and somnolence with damage to pres-
activities such as word games, structured reminiscence, and daily
sure areas resulting from their immobility.
discussions of current events. In units where such activities occur
Minimum investigations for cases of delirium should include
there is a higher level of satisfaction and lower degree of boredom
thorough clinical examination and review of medications includ-
described by the patients.
ing those purchased ‘over the counter’. Blood tests should include
full blood count, glucose, urea and electrolytes, corrected calcium,
Sleep Interventions liver function and thyroid function tests, inflammatory mark-
Hospitals are noisy places, especially at night. Many patients are ers (e.g. C-reactive protein), urinary dipstick, and if appropriate
unused to sharing a room with anyone and hence may have dif- a mid-stream urinary specimen. If sepsis is suspected, blood cul-
ficulties falling asleep and maintaining sleep due to background tures should be done as well as arterial blood gases for lactate levels
noises such as a fellow patient snoring. Sleep deprivation has been and hypoxia. A chest X-ray and 12-lead ECG should be performed
shown as a risk factor in the development of delirium, hence the in most cases. Investigations requiring disruption, such as a trip
need to ensure a refreshing night’s sleep. Ward staff should ensure to the X-ray department, should be performed at an appropriate
that patients are as relaxed as possible by nightfall and know not time unless clinically urgent. Waking a patient for a ‘routine’ chest
X-ray at 3 a.m. should be avoided where possible. If the cause is still housed in busy, noisy wards, with care performed by complete
unclear, a rectal examination may be warranted, as faecal impaction strangers. They frequently move beds within a ward or move to
is a known cause of delirium, especially in people with advanced other wards around the hospital, with yet more strangers dealing
dementia. with their personal care. Interventions including urinary catheter-
Prompt treatment with appropriate fluids, antibiotics, and oxy- ization and the use of physical restraints have been shown to be
gen, via the route that is the most likely to ensure adequate delivery detrimental to patients at risk of developing delirium. Intuitively it
but minimize distress, is the optimum. If the patient is excessively would follow that those already with delirium would tolerate such
somnolent, adequate fluids and medications via the oral route are interventions badly. Most nondelirious inpatients complain about
unlikely to be achieved and the intravenous route should be used. poor sleep due to excessive night-time noise. Patients with delir-
If the patient is hyperactive and wandersome, then the intrave- ium need their sleep even more and are often in hospital for much
nous route may be hazardous, and prolonged one-to-one attention longer with further sleep deprivation than those without delirium.
to get medication and fluids delivered is probably the best choice. The current delirium guidelines (Potter and George, 2006) encap-
Subcutaneous fluids overnight may also be effective. It is impor- sulate all aspects of good supportive practice. These acknowledge
tant to record what medication is actually being taken rather than that while the most important aspect of management is to treat the
merely what is delivered to the bedside. Similarly, accurate record- underlying cause, other areas need attention in order to facilitate a
ing of fluid and nutritional intake is an important factor in the smooth recovery. The guidelines are summarized in Table 40.4.
recovery of the patient.
Although the hyperactive form of delirium seems the most Medicines management
problematic, it is in fact the hypoactive form that carries a worse It is known that medication often has a central role in the causa-
prognosis. Such patients need very strict attention to nutrition and tion of delirium and may be the only cause in about 10% of all
fluid balance as they are prone to malnutrition and dehydration. cases (George et al., 1997). Patients in the community, particu-
Due to their relative immobility they are at high risk of hypostatic larly those in 24-h care, are often commenced on medication for a
pneumonias, pressure damage, and deep vein thrombosis. The
attending physicians and nursing staff need to be aware of these
potential problems. Prompt use of pressure-relieving mattresses, Table 40.4 Summary of delirium guidelines
prophylactic low-molecular weight heparin, interventions such as
nasogastric feeding tubes, and low threshold for prescribing anti- Ensure:
biotics at the earliest sign of chest infection are vital to promote a ◆ Lighting levels are appropriate for the time of day.
good outcome. ◆ Regular and repeated (at least three times daily) cues to improve personal
Problems in dealing with the hyperactive form of delirium are orientation.
slightly different. Whilst sharing the problems of nutrition and fluid ◆ Use of clocks and calendars to improve orientation.
balance, this group is also very prone to falls. Patients with delirium ◆ Hearing aids and spectacles are available as appropriate and in good
who fall and fracture the neck of their femur do very badly. Judicious working order.
use of hip protectors may help, but often these patients may need ◆ Continuity of care from nursing staff.
one-to-one nursing during their worst phase. These patients are ◆ Encouragement of mobility and engagement in activities and with other
also susceptible to oversedation. Overzealous staff can be too hasty people.
with a second or third dose of sedative, which may cause patients ◆ Patients are approached and handled gently.
to become virtually comatose, putting them at similar risks to their ◆ Elimination of unexpected and irritating noise (e.g. pump alarms).
hypoactive peers.
◆ Regular analgesia, e.g. paracetamol.
◆ Encouragement of visits from family and friends who may be able to help
Supportive and behavioural management of delirium calm the patient.
It is humbling to read an adaptation of Barrough’s account of the ◆ Explanation of the cause of the confusion to relatives. Encourage family
management of delirium, first published in 1583: to bring in familiar objects and pictures from home and participate in
Treatment had to involve attention to the patient’s needs: if he was rehabilitation.
troubled by the light, he had to be placed in a dark room. Moreover: ‘let ◆ Fluid intake to prevent dehydration (use subcutaneous fluids if necessary).
his dearest friends come to him, and let them sometime speake gently ◆ Good diet, fluid intake, and mobility to prevent constipation.
and softly unto him, and sometimes rebuke him sharply’. His diet had ◆ Adequate CNS oxygen delivery (use supplemental oxygen to keep
to be light. The patient had to be left undisturbed, since ‘perturbations saturation above 95%).
of the mind do hurt frenetick [delirious] persons exceedingly’. Sleep
◆ Good sleep pattern (use milky drinks at bedtime, exercise during the day).
needed to be ensured by the use of appropriate medications such as
opium or henbane, taking care not to oversedate the patient, as this Avoid:
could turn the ‘frenesie [hyperactive form of delirium] into a lethargie
◆ Inter- and intraward transfers.
[hypoactive form of delirium] whereby you may cause him to sleepe
◆ Use of physical restraint.
so, that you can awake him no more’.
◆ Constipation.
(Lipowski, 1990) ◆ Anticholinergic drugs where possible and keep drug treatment to a
minimum.
Clearly, there is much sense in the above narrative. In the
◆ Catheters where possible.
twenty-first century, patients with delirium are in some ways man-
aged less well than they were some 500 or so years ago. They are (From Potter and George, 2006.)
particular symptom, and although ideally their medication should mg (sometimes in combination with a benzodiazepine). However,
be reviewed regularly in the light of their current behaviours, this the newer atypical antipsychotics have fewer extrapyramidal side
rarely happens unless the patient presents to the GP with a problem effects and may be safer in this regard. Both risperidone and olan-
that is linked to that medication. zapine are also available as oro-dispersible tablets and in injection
In any older patient presenting to hospital, but especially for form. The dose of risperidone is similar to that for haloperidol; for
those with delirium, it is important to evaluate the patient’s medica- olanzapine, 2.5 mg is a reasonable starting dose for an older per-
tion list. Often patients and their relatives have little recollection of son. Current evidence suggests that haloperidol, risperidone, and
why a particular drug was commenced. Drugs acting on the central olanzapine are of comparable effectiveness (Lonergan et al., 2007;
nervous system should receive particular attention, as studies have Grover et al., 2011), and there are inadequate data regarding the
shown that these drugs are the most incriminated in cases of delir- efficacy of other antipsychotics, such as quetiapine and aripipra-
ium. Patients with dementia presenting with a delirium are often zole. In the UK, the British Geriatrics Society guidance suggests the
on a neuroleptic such as risperidone, olanzapine, or quetiapine; use of haloperidol or lorazepam (Potter and George, 2006), while
particularly if the patient has a hypoactive form of delirium, these the National Institute for Health and Clinical Excellence recom-
should be omitted and reintroduced only if necessary. Conversely, mends haloperidol or risperidone, should medication be required
if the patient is on a cholinesterase inhibitor (donepezil, rivastig- (NICE, 2010).
mine, or galantamine) this should not be stopped, as the effects may As the cholinergic system seems to be involved in the aetiology
be devastating on the patient’s baseline function once the delirium of delirium, it is little wonder that clinicians have pondered the
has settled. The other group of agents to be scrutinized carefully potential benefits of the cholinesterase inhibitors in delirium. One
are the opiates. Patients frequently present on tramadol, for exam- case report (Wengel et al., 1998) suggested that the cholinesterase
ple, which may have originally been started following a fracture or inhibitor donepezil caused a dramatic improvement in symptoms
other specific injury. Careful evaluation of pain is needed and in of delirium in a patient with mild Alzheimer’s disease, but more
many cases such medications can be withdrawn successfully. recently results have been disappointing. A recent review article
that summarized and critically evaluated delirium treatment tri-
Treating delirium with medication als concluded there was a need for well-designed randomized,
Because of the fluctuant nature of delirium, a patient who is somno- double-blind, placebo-controlled trials to investigate the drug
lent by day may be extremely active at night. Often the most junior treatment of various aspects of delirium, including delineating
doctor is called through the night to prescribe something to ‘calm treatment by delirium subtype, dose ranging studies, and optimal
the patient down’. Such doctors are often ill-prepared to handle the duration of therapy (Bourne et al., 2008).
request and either may prescribe a large dose of sedative (partly due
to the high doses suggested in the textbooks and formularies), or Role of the Electroencephalogram (EEG)
when a smaller dose is prescribed they may not put a timeframe on
when the dose can be repeated, so the patient may receive several
in Delirium
doses within a short space of time. The following day the damage is In the literature, there are numerous references to the EEG changes
often evident with the patient being found almost comatose, barely seen in delirium. These are the increase in slow-wave activity and
protecting his or her own airway. Many hospitals now have a junior the slowing and disruption of the normal alpha rhythm (Romano
doctors’ handbook to address this issue and training is delivered and Engel, 1944). Whilst some comment that the EEG is widely
on the subject of night sedation at an early stage. In parallel to the accepted as a valuable ancillary laboratory procedure for diagno-
training of junior doctors on this subject, nursing training, espe- sis and serial evaluation of delirium (Koponen, 1991), this tool is
cially for those on night shift, is of use. If the patient can be sup- rarely used routinely in the UK. The practicalities of routine EEG
ported on the ward through the night with measures such as warm examination for delirious patients seem prohibitive. As discussed
milky drinks, relative quiet, and if possible a single cubicle, then a in the section Supportive and Behavioural Management of
better outcome will be achieved in the long run. Delirium, it is important to avoid disturbing and distressing inter-
Sometimes, despite supportive measures, sedation is required for ventions where possible. As neurology services in the UK tend to
the safety of the patient, other patients, and staff. If there is no evi- be located in tertiary referral centres, this investigation for many
dence of psychotic symptoms, or in patients with DLB or Parkinson’s would require an ambulance trip. The EEG is used as a research
disease, then sedation can be offered with lorazepam. This is given tool in delirium and also when the diagnosis is in doubt (Brenner,
in doses of 0.5–1 mg orally and it can be repeated at 2-hourly inter- 1991). Of special value is its use when the differential includes con-
vals up to a maximum of 3 mg per day. Lorazepam can also be given ditions such as Creutzfeldt–Jakob disease and nonconvulsive status
intramuscularly, but there have been problems with the supply of epilepticus.
parenteral lorazepam, so small doses of midazolam (2.5 mg) are
an alternative. Another nonantipsychotic drug is the sedating anti- Communication in Delirium
histamine perphenazine (25–50 mg orally or 10–25 mg intramus-
cularly). If patients are very distressed, have psychotic symptoms, Communication with patients
or if their behaviour is posing a risk to others, then it is reason- Patients who have had an episode of delirium often recall their
able to use an antipsychotic drug. There is probably most experi- experiences as frightening. They may feel they ‘lost’ the time dur-
ence with haloperidol, in small doses starting at 0.5 mg. This can ing which they had their worst symptoms. Some patients do retain
be given at 2-hourly intervals if the patients fail to settle. A maxi- some insight, especially those without dementia, and it is important
mum of 5 mg per day is recommended. If the oral route is impos- to communicate as clearly and effectively as possible with patients
sible then haloperidol can be given intramuscularly at a dose of 1–2 throughout their time in hospital. It is well recognized that sensory
impairment (especially vision and hearing) is a risk factor for the both a knowledge angle and an interpersonal skills angle. Gentle
development of delirium, and intuitively one would suppose that explanation of what delirium is and what can be expected in terms
the ability to communicate adequately would be important in the of recovery then needs to be delivered. This needs to be carefully
recovery period. Patients may have lucid periods during their delir- considered. No longer is it acceptable to assure relatives that delir-
ium and one should opportunistically use these times to reassure ium is a transient phenomenon with no lasting effects. On the con-
the patient and orientate them to their location and condition. trary, delirium carries a significant morbidity and mortality of its
It seems unproductive and cruel to dwell on the previous evening’s own, when other factors such as severity of illness are controlled
behaviours with the patient. One often overhears relatives doing for. The amount of information given at this time may be brief, as
this very thing and understandably the patient feels anxious, upset, the delirium may have come as a complete shock to relatives.
and scared of the evening ahead. A quiet word with relatives early In cases where the index delirium is the first sign of any abnormal
on in the admission can be helpful. When patients are ready for cognitive behaviour, information needs to be given with an assur-
discharge it can be useful to explain to them that they have had an ance that follow-up will be planned with an appropriate person.
episode of delirium and what the identified cause was (if known). Ideally this should be a psychogeriatrician, but where this profes-
They should be warned it may recur, but the best advice for them is sional is not readily available, a geriatrician with an interest in this
to seek their GP earlier rather than later if they are feeling unwell. field may be a reasonable alternative. If the patient has had mild
Timely antibiotics may avert a further delirium and this informa- problems prior to the index delirium, the diagnosis will usually be
tion should be conveyed to the GP and relatives as well. that of a mild dementia. Again, specialist follow-up with psycho-
geriatric colleagues is recommended (Potter and George, 2006). The
Communication with relatives and carers relatives and carers need to be reassured that the course of delirium
There are three phases of communication with relatives and carers is fluctuant. If the patient has a good day which is then followed
that need to be completed for a satisfactory outcome. The first phase by two bad days, relatives and carers may assume the worst. They
is the fact-finding and tentative diagnostic phase. Information ini- need advice on how best they can help the patient. Suggesting they
tially needs to flow from the relatives and carers to hospital staff. bring a few photographs or perhaps a favourite ornament may be of
This includes details of home situation, usual activities of daily help. If the patient is not eating or drinking well it may be helpful
living, usual medications, and mobility. This information can be for them to visit at mealtimes to assist with feeding. Playing simple
gleaned over the phone if necessary, although a face-to-face meet- card games and talking about yesteryear may help improve distress.
ing is preferable. More sensitive is the information regarding the Where appropriate, advice on conversation can be of use. Delirious
patient’s usual cognitive abilities. Relatives and carers will often patients often believe they are back in the past, and constant correc-
say ‘they were fine’ when asked about memory and awareness of tion is rarely helpful. Steering conversation away from such topics,
personal risk. One often needs to probe more deeply to ensure with gentle reorientation, is of use. Relatives and carers will appre-
they have not been attributing ‘leaving the gas on’ or ‘failing to ciate the time given to this discussion, even if they become upset
lock the door’ to ‘well, it’s their age, isn’t it doctor?’ One may often during it. It is of value to have a member of the nursing staff present,
uncover evidence that there has been a decline in cognitive abil- as often this will help improve relationships on the ward. Relatives
ity and sometimes just asking the questions can enable relatives to often feel they must ‘stick up for their parent’s rights’ and this often
see that they have been failing to acknowledge it. Other evidence comes across as intimidating and aggressive to junior nursing and
of functional decline may come from information about recent medical staff. Even senior staff members may unwittingly display
performance in the so-called instrumental activities of daily living negative body language, which further exacerbates the problem.
(IADL), e.g. using the telephone, use of transport, responsibility for Probing gently into some of the verbal complaints can be done at
medication intake, and handling finances. Enquiring about these this meeting if appropriate.
abilities, which are all relatively independent of age, sex, and educa- The third phase involves a discussion of the future. Families often
tional level, may help clarify the presence of a pre-existing cognitive want to know what to expect in the short, medium, and longer term.
decline (Barberger-Gateau et al., 1992). Clearly, this will depend on many factors, including age, comor-
The second phase involves informing the relatives and carers bidity, and the severity of the acute physical illness that caused the
about delirium. They often find seeing their loved one distressed delirium, which are all patient specific. It should be emphasized
and agitated very upsetting and require appropriate counselling. that the presence of delirium will add a further complication, ham-
Often they will become angry with ward staff as a way of expressing pering the acute recovery, as it has been shown that delirium is an
their feelings, and may appear unduly irate, particularly if belong- independent prognostic determinant of hospital outcomes (Inouye,
ings of the patient have become lost or misplaced. The best way to 1998a).
counsel the relatives and carers is usually to remove them from the
immediate bed area into a quiet room. Asking them what infor-
mation they have been given is a good way to start the conversa- Prognosis of Delirium
tion. Letting them tell their version of events from ‘being aware of The medium-term prognosis used to be thought of as good.
something being wrong’ to the present moment is both informative Providing the patient recovered from the initial illness, the delir-
and therapeutic for the relative. Encouraging junior medical staff ium was thought to be completely reversible. However, studies
and nursing staff to observe this meeting is often very helpful for from the last two decades suggest that this is not the case. It is best
all concerned. Meeting relatives is a cause of much anxiety with to suggest to relatives that the majority of the features of delirium
junior doctors, especially if they are unsure of the facts and have improve with time and that the time span can vary from days to
witnessed the relative being assertive with the nursing staff. Seeing weeks. In many cases, features of the delirium can extend beyond
how their seniors handle the situation is invaluable training, from discharge, lasting up to and beyond 6 months. In the longer term,
delirium may herald the arrival of dementia. Long-term follow-up post-hospital discharge by one group (Murray et al., 1993), whilst
of delirium survivors shows an excess incidence of dementia. Clear other researchers have shown that such differences may take some
transcription of conversations with relatives is very important. In time to be apparent. At 3 months, one group found no difference at
subsequent clinic appointments it is unlikely that the clinician will all (Pompei et al., 1994) and although Francis et al.’s (1990) study
recall exactly what was said and what the specific worries and fears showed no statistical difference in functional decline at 6 months’
of the patient and relatives were. It is useful to inform the GP of the follow-up, at 2 years a statistical difference was observed between
content of such conversations as GPs will often be the first port of the functional abilities of the groups (Francis and Kapoor, 1992).
call if further difficulties arise. If mild and previously unrecognized There are higher readmission rates amongst those with delirium
cognitive impairment is uncovered during the hospital stay this too (George et al., 1997) and significantly increased rates of long-term
needs to be conveyed to the GP. Issues such as noncompliance with institutionalization evident as early as 6 months postdischarge
medications may need to be addressed in different ways by the pri- (Francis and Kapoor, 1992; O’Keeffe and Lavan, 1997; Inouye.,
mary care team in light of this information. 1998b).
With the exception of researchers in the field of delirium, most The majority of studies in the 1990s showed an excess mortality
clinicians thought until recently that delirium was a transient in patients who have had an episode of delirium compared with
reversible syndrome with a good prognosis. Many general physi- those with a similar physical illness but without delirium. This
cians still believe this to be the case, although, as already alluded was found at 6 and 12 months’ follow-up in one study (George
to above, the evidence overwhelmingly points to delirium being a et al., 1997) and in follow-up to 3 years in another (Rockwood
marker for physical and cognitive decline and increased mortality. et al., 1999). The cause for these poor physical outcomes remains
Dementia recognition can be improved by training interventions unclear. Researchers have strived to reduce possible bias by match-
with clinical teams (Rockwood et al., 1994) but, however, to date, ing patients carefully with controls, yet still these differences exist.
the evidence as to whether systematic detection and multidiscipli- For example, when dementia patients with a delirium are matched
nary management of delirium improves its outcome remains con- to dementia patients without delirium the physical differences and
flicting (Cole et al., 2002; Lundström et al., 2005). increased mortality remain. Controversially, a recent study of hip
fracture patients suggested that if pre-existing cognitive impair-
Physical health aspects of the prognosis of delirium ment was accurately controlled for, there was no excess mortality
In the past, relatives and carers would be confidently told that the in those with delirium. However, delirium in patients with prefrac-
cognitive state of the patient with delirium would return to nor- ture dementia was associated with an increased risk of death from
mal almost as quickly as it had become impaired. The attending stroke and dementia (Juliebø et al., 2010).
physician would concentrate chiefly on discussing the prognosis of
the (causative) medical condition and its severity, rather than the Mental health aspects of the prognosis of delirium
accompanying delirium. In some ways this was understandable. The aspects of prognosis that worry relatives the most are behav-
Many physicians have had no postgraduate training in psychiat- iour and cognition. Anecdotally, relatives may describe a previous
ric conditions and feel out of their comfort zone when discussing episode of delirium from which the patient never fully recovered.
such matters. In the 1980s and 1990s, numerous studies showed There is some evidence that a substantial proportion of cases of
that delirium was an independent risk factor in poor outcomes delirium are unresolved at discharge, and in about half of the cases
from hospital admission and that recovery from it is much slower there are persistent symptoms at 1 month. Even by 6 months, not
than had been assumed. This message has been poorly received and all the new symptoms of delirium had disappeared (Levkoff et al.,
acknowledged by general medical physicians. It is astonishing that 1992; George et al., 1997; Marcantonio et al., 2000, 2003; Kiely et al.,
a condition affecting at least 10% of medical admissions should be 2004). Treloar and Macdonald (1997a, and b) found that the clini-
so misunderstood. cal diagnosis of delirium using ICD or DSM criteria was not a very
The differences in prognosis of those with and those without accurate predictor of the degree of cognitive recovery and argued
delirium begin at the onset of the delirium, and persist through for a concept of reversible cognitive dysfunction as an alternative.
hospital stay and discharge until death. Various endpoints have It is now known that delirium does have long-term cogni-
been used by researchers to demonstrate this. Hospital mortality tive implications, even for those deemed to have recovered from
was featured in one of the first studies (Rabins and Folstein, 1982). it. Follow-up of patients in several studies has shown accelerated
Although the authors did show an excess hospital mortality, subse- decline in cognition, compared with controls matched for cognitive
quent studies, which perhaps controlled more carefully for severity function at baseline (Koponen and Riekkinen, 1989; Francis and
of illness and presence of dementia, did not reach statistical signifi- Kapoor, 1992; Rockwood et al., 1999; Dolan et al., 2000; Rahkonen
cance (Francis et al., 1990; Levkoff et al., 1992; George et al., 1997; et al., 2000; McCusker et al., 2001; Jackson et al., 2004). It is not just
O’Keeffe and Lavan, 1997; Inouye, 1998a). Increased length of hos- those with dementia who show an accelerated decline. For example,
pital stay of delirious versus nondelirious patients has not been sta- those with delirium who had cognitive testing scores in the nor-
tistically proven as yet, but there is certainly a trend. This work is mal range at discharge had higher than expected rates of dementia
fraught with statistical difficulties, as many of the controls develop when followed for 2 years.
incidental delirium during their hospital stay. Other factors such as In 1959, Engel and Romano suggested that delirium and demen-
the development of complications of hospitalization, e.g. urinary tia may be different aspects of a similar process in which delirium
incontinence, falls, and pressure damage, were more likely to occur may induce irreversible brain damage. They based their hypoth-
in the delirious (Gustafson et al., 1988; O’Keeffe and Lavan, 1997). esis on the EEG findings of slowing in delirium. They proposed
Excess functional decline in those patients with delirium com- a mechanism of impaired cerebral metabolism, ranging from
pared with those without has been demonstrated at 3 and 6 months decreased ability to synthesize neurotransmitters to eventual cell
death (Engel and Romano, 1959). Other researchers suggest that These statements are not without some contention. A small study
delirium may induce long-term or even permanent cognitive of older nursing home residents showed serum anticholinergic
decline, which is manifested by functional decline (Murray et al., activity was raised during illness and declined following recovery.
1993). Alternatively, delirium may be a marker of dementia, or they This effect was observed in those with and without delirium, and
may share a common pathogenesis (Eikelenboom and Hoogendijk, was not observed to be related to changes in medication (Flacker
1990). Acute physical illness may uncover a brain that is operat- and Lipsitz, 1999). This work followed previous studies that have
ing on low reserves of neurotransmitter activity, with little ability questioned the hypothesis that reduced cholinergic neurotransmis-
to cope with the increased demands. When put under stress, the sion is the main pathophysiological precipitant of delirium. These
brain no longer copes and the concept of ‘acute brain failure’ occurs authors challenge the assumption that serum anticholinergic activ-
(Francis and Kapoor, 1992). ity is a stable patient characteristic, resulting from their medications
Knowledge of the prognosis of delirium has increased over or their metabolites. If this were true, they argue that manipulation
the last two decades. This knowledge now needs to be dissemi- of medication could prevent delirium, which is clearly not the case.
nated amongst acute care physicians and geriatricians in order for Epidemiological studies of delirium have not shown anticholiner-
patients to be best managed. The development of standard guide- gic medications to be statistically associated with delirium in either
lines (Potter and George, 2006; NICE, 2010) and incorporation of medical or postoperative patients (Francis and Kapoor, 1992; Schor
delirium into the undergraduate and postgraduate curricula may et al., 1992; Marcantonio, 1994). These authors suggest that raised
help to address this. serum anticholinergic activity and delirium may arise from the
same endogenous trigger, rather than the raised activity being the
Neuropathophysiology of Delirium causative factor for the other.
Other neurotransmitters and iatrogenic agents that stimulate or
In recent small studies, patients with delirium have a high degree inhibit neurons can contribute to delirium. Dopaminergic activity
of metabolic brain disturbance. This is demonstrated by higher CSF is intimately involved with cholinergic activity, the former appear-
lactate and protein levels, but lower neuron-specific enolase levels ing to have a regulatory control on the latter (Trzepacz and van der
when compared to those with dementia (Caplan et al., 2010). There Mast, 2002). Agents interfering with this balance can precipitate
needs to be further work done in this area to clarify the specific delirium. The majority of agents used to treat Parkinson’s disease
relationship of the various measurable substances in CSF to further can precipitate delirium, e.g. levodopa and the dopamine agonists.
our understanding. Dopamine antagonists, including antipsychotics such as haloperi-
There are thought to be three factors involved in the pathogenesis dol, can treat symptoms of delirium. Serotonin may play an impor-
of delirium at the cellular level. These include neurotransmitters, tant role in the development of delirium. The ‘serotonin syndrome’
glucocorticoids, and cytokines. characterized by confusion, restlessness, tremor, and diaphoresis
(excessive sweating) shares many of the features seen in hyperac-
The role of neurotransmitters in the pathogenesis of tive delirium.
delirium Unfortunately, trying to study a particular neurotransmitter
One major hypothesis is that a deficiency of the neurotransmitter in isolation is a difficult task. In most areas of the brain there are
acetylcholine plays a key role in the pathogenesis of delirium. This numerous pathways that use different transmitters. It may be that
is supported by the following statements from the works of sev- the balance between various neurotransmitters is lost in delirium,
eral researchers (Tune et al., 1981; Golinger et al., 1987; Trzepacz, rather than any particular one being either present in excess or
1994, 1996; Mach et al., 1995; Flacker et al., 1998; Han et al., 2001; deficient.
Trzepacz and van der Mast, 2002; Roche, 2003): Other neurotransmitters may have a role, but the evidence
◆ Delirium can be induced in susceptible individuals by anticholin- is less well developed. These include norepinephrine,
ergic drugs. gamma-aminobutyric acid, glutamate, and melatonin. Some
believe that these other neurotransmitters have an effect on delir-
◆ Patients on anticholinergic drugs have higher rates of delirium. ium by interfering with the cholinergic/dopaminergic balance
◆ Medications used to treat myasthenia gravis, an autoimmune (Shigeta et al., 2001; Trzepacz and van der Mast, 2002; Cole, 2004).
condition with antibodies directed against the neuromuscular A recent small trial has used melatonin versus placebo to prevent
acetylcholine receptors, can reverse delirium caused by anti- delirium and the results were encouraging, despite the smallness
cholinergic drugs. of the trial (Al-Aama et al., 2011).
◆ Cholinesterase inhibitors improve some of the symptoms of The role of the hypothalamo-pituitary axis (HPA) in
delirium even in cases not related to anticholinergic medication. delirium
◆ Serum anticholinesterase activity is known to be raised in patients In physical illness, one of the body’s responses is to produce glu-
with delirium. cocorticoids. This is a protective mechanism designed to promote
return to health. In certain conditions, the way in which the body
◆ Delirium is associated with a reduction in the cerebral oxida-
responds becomes maladaptive. Cortisol excess is a well-described
tive metabolism, with subsequent reduction of neurotransmitter
cause of neuropsychiatric abnormalities. In Cushing’s syndrome,
levels.
where glucocorticoid hormones are produced in excess without
◆ Hypoxia seems to disproportionately affect synthesis of cerebral the usual negative-feedback control, marked changes are observed
acetylcholine, as it would seem this neurotransmitter pathway is in mood and cognition. The hippocampus seems to be the area
particularly sensitive to low oxygen levels. of the brain most affected by excess cortisol, owing to its high
numbers of glucocorticoid receptors. Reduction in hippocampal thiamine, is important. Although the majority of older patients pre-
volume has been observed in both Cushing’s syndrome patients senting with delirium will not have an alcohol withdrawal aetiology,
and those with major depression (Olsson, 1999). Hypercortisolism it is important to establish an alcohol history, as the consequences
has been demonstrated in delirium associated with lower respira- of missing such a case can be devastating, due to the possible devel-
tory infection, postoperative delirium, and in delirium following opment of Wernicke’s encephalopathy or Wernicke–Korsakoff syn-
stroke (McIntosh et al., 1985; Lipowski, 1990; O’Keeffe and Devlin, drome. If in doubt, concentrated B vitamins should be used. As
1994). Abnormalities of the HPA axis similar to those described in the older population changes and alcohol problems, particularly
major depression are present. This can be demonstrated using the amongst women, are destined to increase, a good awareness of
dexamethasone suppression test. Patients with a lower respiratory delirium tremens and the consequences of failing to treat should
tract infection will have high endogenous cortisol levels in response be borne in mind. Pharmacological agents used in other causes of
to the stress of the infection. In nondelirious subjects given dex- delirium, such as haloperidol, should be used with caution.
amethasone, the cortisol levels dropped (normal response) in the
majority of patients (6/7 patients). In the majority of delirious sub- Benzodiazepine withdrawal
jects, however, the cortisol levels were not suppressed (7/9 patients) Abrupt reduction in dosage or cessation of any sedative drug may
(O’Keeffe and Devlin, 1994). result in a withdrawal syndrome similar to that seen in alcohol
withdrawal. This should be avoided with gradual reduction in dos-
The role of cytokines in delirium age over months, and in some cases years. Although thought to be
Cytokines, when used at supraphysiological doses, may lead to safe drugs a few decades ago, the benzodiazepines are highly addic-
delirium. Interleukin-2 (IL-2), which has a central role in both cel- tive in susceptible individuals and there are still many patients on
lular and humoral immunity, is the best studied of the cytokines. such agents decades later. Sometimes abrupt withdrawal can occur
Delirium has been observed in patients receiving IL-2 therapy and inadvertently when patients are admitted to hospital and are unable
the effect seems dose dependent (Denicoff et al., 1987; Rosenberg et to give a clear account of their medications (Moss, 1991; Moss and
al., 1989; Fenner et al., 1993). The mechanism by which IL-2 causes Lanctot, 1998). If the baseline medication list is in doubt, one must
delirium is uncertain. EEG changes pointing to diffuse cerebral contact the GP at the earliest possibility to ensure an iatrogenic
dysfunction have been observed. Despite a considerable amount withdrawal syndrome is avoided.
of research into the neuropathophysiology of delirium, the mecha-
nisms remain poorly understood. The confusion was summed up References
by Trzepacz and van der Mast (2002): ‘Certain neuroanatomical Al-Aama, T., et al. (2011). Melatonin decreases delirium in elderly patients: A
and neurotransmitter systems may represent final common path- randomized, placebo-controlled trial. International Journal of Geriatric
ways for the otherwise diverse aetiologies of delirium, or delirium Psychiatry, 26, 687–94.
may be the final common symptom of multiple neurotransmitter American Psychiatric Association (1994). Diagnostic and statistical manual of
abnormalities’. mental disorders, 4th edition. APA, Washington, DC.
It is likely that all of these potential contributors have a role to Barberger-Gateau, P., et al. (1992). Instrumental activities of daily living as a
screening tool for cognitive impairment in elderly community dwellers.
play, but perhaps to different degrees in different individuals. What
makes one person susceptible to delirium compared with a fellow
Bourne, R.S., et al. (2008). Drug treatment of delirium: Past, present and
patient with similar risk factors remains unclear. future. Journal of Psychomotor Research 65, 273–82.
Brenner, R.P. (1991). Utility of EEG in delirium: past views and current
Other Types of Delirium practice. International Psychogeriatrics, 3, 211–29.
Caplan, G.A., et al. (2010). Cerebrospinal fluid in long-lasting delirium
Delirium tremens compared with Alzheimer’s dementia. Journal of Gerontology: Series A:
This term was introduced by Sutton in 1813 and is still widely used Biological and Med Sciences, 65, 1130–6.
today. It is defined as a delirious state that occurs within a week Cole, M.G. (2004). Delirium in elderly patients. American Journal of
of withdrawal of alcohol. Its diagnostic criteria include marked Psychiatry, 12, 7–21.
autonomic hyperactivity and associated features including: vivid Cole, M.G., et al. (2002). Systematic detection and multidisciplinary care
of delirium in older medical inpatients; a randomized trial. Canadian
hallucinations which can be visual, auditory, and tactile; delusions;
Medical Association Journal, 167, 753–9.
agitation; a coarse irregular tremor. Seizures can occur, especially if Denicoff, K., et al. (1987). The neuropsychiatric effects of treatment with
excess alcohol consumption has taken place for at least 5 years, and interleukin-2 and lymphocyte-activated killer cells. Annals of Internal
often herald a florid delirium. Medicine, 107, 293–300.
Although anecdotally common, the incidence of true delirium Dolan, M.M., et al. (2000). Delirium on hospital admission in aged hip
tremens is considered to be low. Some believe it to be less than 1% fracture patients, prediction of mortality and 2 year functional outcomes.
of all alcoholics undergoing withdrawal, but in specific popula- Journals of Gerontology. Series A, Biological Sciences and Medical Sciences,
tions, such as those hospitalized for acute mental and physical dis- 55A, M527–34.
turbances associated with alcohol, the incidence was as high as 40% Eikelenboom, P. and Hoogendijk, W.J. (1990). Do delirium and Alzheimer’s
dementia share specific pathogenic mechanisms? Dementia and Geriatric
(Lipowski, 1990). Men are affected by delirium tremens more com-
Cognitive Disorders, 10, 319–24.
monly than women by four- or five-fold, and on average it occurs Engel, G.L. and Romano, J. (1959). Delirium, a syndrome of cerebral
in the fifth decade of life. insufficiency. Journal of Chronic Diseases, 9, 260–77.
Management involves the use of benzodiazepines (Trzepacz, Evans, L.K. and Strumpf, N.E. (1989). Tying down the elderly: a review of the
1999), both for symptoms and seizure treatment. Nutritional prob- literature on physical restraint. Journal of the American Geriatrics Society,
lems frequently coexist and treatment with B vitamins, especially 37, 65–74.
Fenner, M., et al. (1993). Neuropsychiatric symptoms during treatment with Inouye, S.K., et al. (1999). A multicomponent intervention to prevent
interleukin-2 and interferon-alpha [letter]. Lancet, 341, 372. delirium in hospitalised older patients. New England Journal of Medicine,
Fisher, B.W. and Flowerdew, G. (1995). A simple model for predicting 340, 669–76.
postoperative delirium in older patients undergoing elective orthopedic Inouye, S.K., et al. (2001). Nurses’ recognition of delirium and its symptoms:
surgery. Journal of the American Geriatrics Society, 43, 175–8. comparison of nurse and researcher ratings. Archives of Internal Medicine,
Flacker, J.M. and Lipsitz, L.A. (1999). Serum anticholinergic activity changes 161, 2467–73.
with acute illness in elderly medical patients. Journals of Gerontology. Jackson, J.C., et al. (2004). The association between delirium and cognitive
Series A, Biological Sciences and Medical Sciences, 54, M12–16. decline: a review of the empirical literature. Neuropsychological Reviews,
Flacker, J.M. and Marcantonio, E.R. (1998). Delirium in the elderly: optimal 14, 87–98.
management. Drugs and Ageing, 13, 119–30. Jitapunkul, S., Pillay, I., and Ebrahim, S. (1991). The Abbreviated Mental Test:
Flacker, J.M., et al. (1998). The association of serum anticholinergic activity its use and validity. Age and Ageing, 20, 332–6.
with delirium in elderly medical patients. American Journal of Geriatric Juliebø, V., et al. (2010). Delirium is not associated with mortality in elderly
Psychiatry, 6, 31–41. hip fracture patients. Dementia and Geriatric Cognitive Disorders 30,
Folstein, M.F., Folstein, S.E., and McHugh, P.R. (1975). Mini-mental state: 112–20.
a practical method for grading the cognitive state of patients for the Kalisvaart, K.J., et al. (2005). Haloperidol prophylaxis for elderly hip-surgery
clinician. Journal of Psychiatric Research, 12, 189–98. patients at risk of delirium: a randomized placebo-controlled study.
Folstein, M.F., et al. (1991). The epidemiology of delirium in the Journal of the American Geriatrics Society, 53, 1658–66.
community: the Eastern Baltimore Mental Health Survey. International Kiely, D.K., et al. (2004). Characteristics associated with delirium persistence
Psychogeriatrics, 3, 169–76. among newly admitted post-acute facility patients. Journals of Gerontology.
Foreman, M.D. (1989). Confusion in the hospitalised elderly, incidence, onset Series A Biological Sciences and Medical Sciences, 59, 344–49.
and associated features. Research in Nursing and Health, 12, 21–9. Koponen, H. (1991). Electroencephalographic indices for diagnosis of
Foreman, M. and Milisen, K. (2004). Improving recognition of delirium in delirium. International Psychogeriatrics, 3, 249–51.
the elderly. Primary Psychiatry, 11, 46–50. Koponen, H., et al. (1989). Delirium among elderly persons admitted to a
Francis, J. and Kapoor, W.N. (1992). Prognosis after hospital discharge of psychiatric hospital; clinical course during the acute stage and one-year
older medical patients with delirium. Journal of the American Geriatrics follow up. Acta Psychiatrica Scandinavica, 79, 579–85.
Society, 40, 601–6. Levkoff, S.E., et al. (1992). Delirium: the occurrence and persistence of
Francis, J., Martin, D., and Kapoor, W.N. (1990). A prospective study symptoms among the elderly hospitalised patients. Archives of Internal
of delirium in hospitalized elderly. Journal of the American Medical Medicine, 152, 334–40.
Association, 263, 1097–101. Levkoff, S.E., et al. (1996). Subsyndromal delirium. American Journal of
George, J., Bleasdale, S., and Singleton, S.J. (1997). Causes and prognosis of Geriatrics, 4, 320–9.
delirium in elderly patients admitted to district general hospital. Age and Lipowski, Z.J. (1990). Delirium: Acute confusional states. Oxford University
Ageing, 26, 423–427. Press, New York.
Golinger, R.C., Peet, T., and Tune, L.E. (1987). Association of elevated plasma Liptzin, B., et al. (2005). Donepezil in the prevention and treatment of
anticholinergic activity with delirium in surgical patients. American post-surgical delirium. American Journal of Geriatric Psychiatry, 13,
Journal of Psychiatry, 144, 1218–20. 1100–6.
Grover, S., Kumar, V., and Chakrabarti, S. (2011). Comparative efficacy Lofgren, R.P., et al. (1989). Mechanical restraints on the medical wards: are
study of haloperidol, olanzapine and risperidone in delirium. Journal of protective devices safe? American Journal of Public Health, 79, 735–8.
Psychosomatic Research, 71, 277–81. Lonergan, E., et al. (2007). Antipsychotics for delirium. Cochrane Database
Gustafson, Y., et al. (1988). Acute confusional states in elderly patients treated of Systematic Reviews, Issue 2, CD005594. doi: 10.1002/14651858.
for femoral neck fractures. Journal of the American Geriatrics Society, 36, CD005594.pub2.
525–30. Lundström, M., et al. (2005). A multifactorial intervention program reduces
Gustafson, Y., et al. (1991). A geriatric anesthesiologic program to reduce the duration of delirium. Journal of the American Geriatrics Society, 53,
acute confusional states in elderly patients treated for femoral neck 622–8.
fractures. Journal of the American Geriatrics Society, 39, 655–62. Mach, J.R., et al. (1995). Serum anticholinergic activity in hospitalized older
Han, L., et al. (2001). Use of medications with anticholinergic effect predicts persons with delirium: a preliminary study. Journal of the American
clinical severity of delirium symptoms in older medical in patients. Geriatrics Society, 43, 491–5.
Archives of Internal Medicine, 161, 1099–105. McCusker, J., et al. (2001). Delirium in older medical inpatients and
Hart, B. (1936). Delirious states. British Medical Journal, 2, 745–9. subsequent cognitive and functional status; a prospective study. Canadian
Inouye, S.K. (1997). Delirium and cognitive decline: does delirium lead to Medical Association Journal, 165, 575–83.
dementia? In: Cognitive decline: strategies for prevention: proceedings of a McIntosh, T.K., et al. (1985). Beta-endorphin, cortisol and postoperative
White House Conference on aging (eds H.M. Fillit and R.N. Butler), pp. delirium: A preliminary report. Psychoneuroendocrinology, 10, 303–13.
85–107. Greenwich Medical Media, London. Marcantonio, E.R., et al. (1994). The relationship of post operative delirium
Inouye, S.K., (1998a). Does delirium contribute to poor hospital outcomes? with psychoactive medications. Journal of the American Medical
A three site epidemiologic study. Journal of General Internal Medicine, Association, 272, 1518–22.
13, 234–42. Marcantonio, E.R., et al. (2000). Delirium is independently associated with
Inouye, S.K. (1998b). Delirium in hospitalized older patients. Clinical poor functional recovery after hip fractures. Journal of the American
Geriatric Medicine, 14, 745–64. Geriatrics Society, 48, 618–24.
Inouye, S.K. and Charpentier, P.A. (1996). Precipitating factors for delirium Marcantonio, E.R., et al. (2001). Reducing delirium after hip fracture: a
in hospitalized persons. Journal of the American Medical Association, randomised trial. Journal of the American Geriatrics Society, 49, 516–22.
275, 825–7. Marcantonio, E.R., et al. (2003). Delirium symptoms in post acute care:
Inouye, S.K., et al. (1990). Clarifying confusion: the confusion assessment prevalent, persistent and associated with poor functional recovery.
method. A new method for detection of delirium. Annals of Internal Journal of the American Geriatrics Society, 51, 4–9.
Medicine, 113, 941–8. Mattice, M. (1989). Intrahospital room transfers: a potential link to delirium
Inouye, S.K., et al. (1993). A predictive model for delirium in hospitalized in the elderly. Perspectives, 13, 10–12.
elderly medical patients bade on admission characteristics. Annals of Moss, J.H. (1991). Sedative and hypnotic withdrawal states in hospitalized
Internal Medicine, 119, 474–81. patients. Lancet, 338, 575.
Moss, J.H. and Lanctot, K.L. (1998). Iatrogenic benzodiazepine withdrawal Rosenberg, S., Lotze, M., and Yang, J. (1989). Experience with the use of
delirium in hospitalised older patients. Journal of the American Geriatrics high-dose interleukin-2 in the treatment of 652 cancer patients. Annals
Society, 43, 1020–2. of Surgery, 210, 474–84.
Murray, A.M., et al. (1993). Acute delirium and functional decline in the Schor, J.D., et al. (1992). Risk factors for delirium in hospitalized elderly.
hospitalized elderly patient. Journal of Gerontology, 48, M181–6. Journal of the American Medical Association, 267, 827–31.
Murray, J.A.H. (ed.) (1897). A New English dictionary on historical principles, Shigeta, H., et al. (2001). Postoperative delirium and melatonin levels in
Vol. 3, p. 165. Clarendon Press, Oxford. elderly patients. American Journal of Surgery, 182, 449–54.
NICE (2010). Delirium: diagnosis, prevention and management. Clinical Simpson, C.J. and Kellett, J.M. (1987). The relationship between preoperative
Guideline 103. National Institute for Health and Clinical Excellence. anxiety and post-operative delirium. Journal of Psychosomatic Research,
<http://www.nice.org.uk/nicemedia/live/13060/49909/49909.pdf> 31, 491–497.
(accessed 09.01.2012). Sullivan-Marx, E.M. (1994). Delirium and physical restraint in the
O’Keeffe, S.T. and Devlin, J.G. (1994). Delirium and the dexamethasone hospitalized elderly. Image, 26, 295–300.
suppression test in the elderly. Neuropsychobiology, 30, 153–6. Treloar, A.J. and Macdonald, A.J.D. (1995). Recognition of cognitive
O’Keeffe, S. and Lavan, J.N. (1997). The prognostic significance of delirium impairment by day and night nursing staff among acute geriatric patients.
in older hospital patients. Journal of the American Geriatrics Society, 45, Journal of the Royal Society of Medicine, 88, 196–8.
174–8. Treloar, A.J. and Macdonald, A.J.D. (1997a). Outcome of delirium: Part 1.
O’Keeffe, S.T., et al. (2005). Use of serial MMSE scores to diagnose and Outcome of delirium diagnosed by DSM-III-R, ICD-10 and CAMDEX
monitor delirium in elderly hospital patients. Journal of the American and derivation of the Reversible Cognitive Dysfunction Scale among
Geriatrics Society, 53, 867–70. acute geriatric in patients. International Journal of Geriatric Psychiatry,
Olsson, T. (1999). Activity in the hypothalamic-pituitary-adrenal axis and 12, 609–13.
delirium. Dementia and Geriatric Cognitive Disorders, 10, 345–9. Treloar, A.J. and Macdonald, A.J.D. (1997b). Outcome of delirium: Part 2.
Pompei, P., et al. (1994). Delirium in hospitalised older persons: outcomes Clinical features of reversible cognitive dysfunction—are they the same
and predictors. Journal of the American Geriatrics Society, 42, 809–15. as accepted definitions of delirium? International Journal of Geriatric
Potter, J. and George, J. (2006). The prevention, diagnosis and management Psychiatry, 12, 614–18.
of delirium in older people: concise guidelines. Clinical Medicine, 6, Trzepacz, P.T. (1994). The neuropathogenesis of delirium. Psychosomatics, 35,
303–8. 374–91.
Rabins, P.V. and Folstein, M.F. (1982). Delirium in dementia: diagnostic Trzepacz, P.T. (1996). Anticholinergic model for delirium. Seminars in
criteria and fatality rates. British Journal of Psychiatry, 140, 149–53. Clinical Neuropsychiatry, 1, 294–303.
Rahkonen, T., et al. (2001). Systematic intervention for supporting Trzepacz, P.T. (1999). Update on the neuropathogenesis of delirium. Dementia
community care of elderly people after a delirium episode. International and Geriatric Cognitive Disorders, 10, 330–4.
Psychogeriatrics, 13, 37–49. Trzepacz, P.T. and Francis, J. (1990). Low serum albumin and risk of delirium.
Roche, V. (2003). Etiology and management of delirium. American Journal of American Journal of Psychiatry, 147, 675.
Medical Science, 325, 20–30. Trzepacz, P. and van der Mast, R. (2002). The neuropathophysiology of
Robinson, S. and Vollmer, C. (2010). Undermedication for pain and delirium. In: Delirium in old age (eds J. Lindesay, K. Rockwood, and A.
precipitation of delirium. Medsurg Nursing, 19, 79–83. Macdonald), pp. 51–90. Oxford University Press, Oxford.
Rockwood, K. (1989). Acute confusion in elderly medical patients. Journal of Tune, L.E., et al. (1981). Association of postoperative delirium with elevated
the American Geriatrics Society, 37, 150–4. serum levels of anticholinergic drugs. Lancet, 1, 651.
Rockwood, K., et al. (1994). Increasing the recognition of delirium in elderly Wengel, S.P., Roccaforte, W.H., and Burke, W.J. (1998). Donepezil improves
patients. Journal of the American Geriatrics Society, 42, 252–6. symptoms of delirium in dementia: implications for future research.
Rockwood, K., et al. (1999). The risk of dementia and death after delirium. Journal of Geriatric Psychiatry and Neurology, 11, 159–61.
Age and Ageing, 28, 551–6. WHO (1993). International Classification of Diseases, 10th revision (ICD-10).
Rogers, M.P., et al. (1989). Delirium after elective orthopaedic surgery, World Health Organization, Geneva.
risk factors and natural history. International Journal of Psychiatry in Williams, M., et al. (1985). Reducing acute confusional states in elderly
Medicine, 19, 109–21. patients with hip fractures. Research in Nursing and Health, 8, 329–337.
Romano, J. and Engel, G.L. (1944). Delirium. 1: electroencephalographic Wong, C.L., et al. (2010). Does this patient have delirium? Journal of the
data. Archives of Neurology and Psychiatry, 51, 356–77. American Medical Association. 304, 779–86.
CHAPTER 41
The experience of depression
Judith Boast
I am a physically fit 71-year-old who is just emerging from the third My next bout of depression was after my mother’s death in 1990.
experience of depression in quite a long life. During its progress, I In many ways, that too was explicable as my mother, a widow who
was sensitively treated by a psychiatrist who asked me to write this lived half a mile from us, had been ill for some time and I had tried
brief account. to be a carer, full-time teacher, and mother to two occasionally
I experienced no depressive episodes until after the birth of my challenging teenage boys. My low mood started 3 months after my
first child in 1968. In fact, if anything, I had an upbeat and at times mother’s death and lasted about 2 years. It took me twice as long to
almost overconfident nature. I did well in school and went on to prepare the work as normal and I lacked my usual ability to relax
Oxford. During these years, I travelled extensively on my own and with students and to rely on a keen sense of humour. I felt I was
spent short vacation periods working with European families as putting on an act of being myself. During this time I did seek psy-
a carer/English language teacher. I taught English in secondary chiatric help and took a course of antidepressants. I never felt they
schools throughout my life, including when I was bringing up two did me a great deal of good since I felt permanently as if I were
sons, and retired as a head of English at the age of 61. I very much living under a thick grey blanket, but they may have kept me going
enjoyed my subject and my teaching and usually received an enthu- till the illness abated.
siastic response from my students. My latest bout of depression started in the summer of 2009 and
Given all this, the horrible feeling of misery and not coping that is more or less over now in October 2011. During it, I have had
enveloped me when my first son was 3 months old was a great the help of a very sympathetic GP, as it happens an ex-student of
shock. I thought I was going mad and had no future in front of me. mine, and psychiatrist. I found this bout the worst as I can give
Looking back 43 years, I can see that I had not prepared myself no reasonable explanation for its arrival. I was physically fit and
very well for childbirth. We were living as postdoctoral students enjoying retirement. I continued to teach part-time and also took
in the US. We had very little money and the baby was not planned. part in walking, charity work, book clubs, etc. I enjoyed all these
However, the American citizenship my son gained has been of great activities and I am lucky enough to have a good many close friends,
use to him over the years as he has experienced enjoyable periods a supportive family with two delightful grandchildren, and enough
of study and work in Princeton and New York. At his birth, I had money to travel, which, when I am well, I greatly enjoy. The depres-
only a few hours in a Los Angeles hospital with no support, since sion thus seemed to come out of the blue and to completely change
husbands were only there to pay the bills. I had no further medi- my life for the worse. One of the first symptoms was a break in my
cal help from then on and found living in a rented flat far from usual ability to sleep soundly. I took a long time getting to sleep
family and friends with a newborn baby a considerable challenge. and, when I woke during the night, it was with irrational feelings
This first bout of depression lasted just over a year. I have distant of dread. I tried to deal with them by making a cup of tea, reading,
memories of pushing my son round town in an old-fashioned pram or listening to the radio, but they were very distressing. The other
feeling tearful and cut off from everybody. Finally I was asked to go bugbear is a loss of my usual confidence. I have not stopped doing
back to teaching—it was unusual to be a working mother in those anything required of me during this last episode, but again I felt
days. I dreaded it and found the first 2 months very stressful, but as if was acting myself in a tense manner. I find it difficult to talk
on a particular day I was walking the short distance from my home about my depression except to two close fiends. I am aware that in
to the school when I felt the depression departing from me. After the past the mood has gone away and, certainly, when I was work-
that, I had some of the most productive working years I have expe- ing, I thought the less I talked of my feelings the better, as I did not
rienced. I felt my knowledge of depression made me particularly want to lose my much-valued job. My husband is supportive up to a
sympathetic to teenagers in my care and I was swiftly promoted. point. He has not suffered from this wretched condition himself so
We took great care in selecting the carers for our sons and they all his solution to my moods is to suggest I write a book or something
turned out to be excellent. My second son was born in my home of that kind, when I am feeling that putting one foot in front of the
town after a very short labour and I experienced no down-turn in other is quite enough. As a result, I have given up talking to him
mood after his birth. I thus convinced myself that depression was about the subject. I find a glass of wine or two during a social occa-
solely connected to childbirth and I was sterilized when my second sion I am dreading helps me, but I am very aware of the dangers of
son was 2 years old. this. Lately I have found an excellent therapist whom I see once a
week. I can talk easily to her—I have tried therapists in the past and from pancreatitis than from depression. What I dislike most about
not found them helpful, so this is a great relief. I am also helped by the condition is that it robs me of my get up and go. As I have said,
exercise; I walk with a group once a week and swim in a leisure cen- I have done everything asked of me—from extensive part-time
tre as many times as I can. I am also aware I am much better in the teaching to making a successful speech at my son’s wedding—but
summer, which is causing me some melancholic feelings just now, I rarely initiate anything at present. Previously, one of the ways I
but I have plans for some post-Christmas winter sun. managed a busy working life was to give myself small treats such
My attitude to antidepressants during this last episode has been as a cup of coffee and reading a newspaper on my way home from
complicated. I took mirtazapine for about 3 months. I feel I did get teaching. I also spent most half-terms walking abroad with the
some help from this antidepressant, which, above all, granted me Ramblers. I find that the pleasure I took in little things has largely
a full night’s sleep. However, I did feel sluggish and I developed a escaped me, though often, when I look back at an event that I found
minor tremor which displeased me. I am an avid reader of websites/ challenging at the time, such as a trip abroad, I realize I did gain a
articles on anything prescribed to me and I think I was probably lot from it. I cannot fault the help I have received. I am sleeping
over-ready to experience side effects. In the end, the tremor caused well once more and have long periods when I do not think about
me to come off antidepressants and since then I have managed on depression. However, at times I still feel like bursting into tears for
my own, with occasional recourse to sleeping pills. My psychia- no reason. In fact, I have not cried at all during this episode. I look
trist has been very encouraging throughout my progress and has forward to still further improvement. I also intend to volunteer to
allowed me to do what I considered best as far as taking medication help other sufferers from this disease. I still do not know how much
was concerned. I am really in control or how much the depression has gained the
I would much rather suffer physically than mentally. Fortunately upper hand, but I give thanks for the fact that, even at its worst,
I have been very healthy up to now, but I did have a week in hos- it has not prevented me from reading and escaping myself into a
pital 3 years ago and I found it much easier to deal with recovery nineteenth-century novel!
CHAPTER 42
Depression in older people
Alan Thomas
What Is Depression and How Common Is It? in its next (DSM-V) incarnation. ICD takes a slightly different
approach, with no category corresponding to MDD, but the great
The term depression is an elusive one, since it can refer to a symptom majority of research has been conducted using DSM criteria,
of low mood, the well-defined clinical syndrome of major depressive making it difficult to apply to ICD categories of depression. More
disorder, or to a range of less well-defined syndromes and constel- important for old age psychiatry is that both systems extrapo-
lations of symptoms. It is easy to forget this important reality and late criteria derived from younger adults to the older population.
proceed as if findings from research on major depressive disorder Such an approach may be criticized because it does not consider
can be applied to all other patterns of depression in the older popu- the age-related (gerontological) and disease-related (geriatric)
lation. DSM-defined (unipolar) major depressive disorder (MDD) changes that affect the presentation of depression in older peo-
is diagnosed when someone has one or more major depressive epi- ple. This leads to the next broad category of depression in older
sodes, which are episodes of depressed mood or loss of interest plus people, namely that of depression comorbid with other illnesses.
at least four other key symptoms from a list of nine (see Box 42.1). Depression is recognized to occur at high levels in people with
Importantly, this is not simply a checklist of symptoms, since the physical illnesses generally (Goodwin, 2006), but, as might be
DSM manual details these clinical features, which need to persist predicted, it is especially associated with brain diseases. It occurs
over at least 2 weeks, and this collection of symptoms needs to pro- in 20–40% of people with Parkinson’s disease (Lieberman, 2006),
duce clinically important impairment in everyday functioning. The in 25–50% of people with dementia (Ballard et al., 1996a), and
most recent revision of the NICE guidelines on the management about a third of people after stroke (Gaete and Bogousslavsky,
of depression emphasizes these key points, especially that duration 2008). Related to this is that rates of depression in older people
and disability due to depression should be carefully assessed when in residential care facilities are about three times higher than in
determining the appropriate management (which then follows a the community (Blazer, 2003).
rational stepped care approach) (NICE, 2009). The DSM manual It seems clear that, unlike dementia which shows a clear associa-
divides MDD into several subtypes, depending on severity (mild, tion with increasing age, MDD does not increase with age and, if
moderate, and severe); presence or absence of psychotic features; organic disorders such as dementia are excluded and disability is
and remission state. MDD is probably best regarded as the most controlled for, depression more broadly does not appear to change
severe depressive illness and its prevalence has been repeatedly in prevalence with age (Jenkins et al., 1997, Roberts et al., 1997;
reported at 1–3%, a rate similar to that in younger adults (Beekman Blazer, 1999). Women are still more likely to be affected than men,
et al., 1999). The other main category in DSM is dysthymic disor- though this sex difference may narrow with increasing age (Jorm,
der, which is diagnosed when depressive symptoms persist for over 1987; Bebbington et al., 1998). The lack of association with ageing
2 years without meeting criteria for MDD. contradicts widespread views that ageing itself is a risk factor for
In the older population there is a large group with less well-defined depression.
depressions, sometimes termed ‘subsyndromal depression’. This
term derives from the fact that individuals with such depression
fall short of the full criteria for MDD, but they do so for different Clinical Features of Late-Life Depression
reasons: some have partially responded to treatment; others have It has often been suggested there is a different symptom profile
partially spontaneously recovered; some have never attainted MDD characteristic of depression in older people. Evidence for this is lim-
status and wax and wane in symptom severity and frequency; some ited and, reviewing the literature, Baldwin (1994) concluded there
have chronic, low-grade symptoms. Studies of these ill-defined cat- was little evidence for phenomenological differences between older
egories, essentially assessing the presence of substantial depressive and younger depressed patients. However, certainty about such a
symptoms, yield prevalence rates about three times higher than for conclusion is hampered by much of the research being done using
MDD (Meeks et al., 2011). However, it should be emphasized that DSM criteria for MDD, which are tight criteria that arguably might
symptom duration and related disability are crucial for informing conceal differences in symptom patterns in younger and older peo-
management. ple; such findings of no differences may therefore be somewhat
The discussion thus far has been based around the DSM sys- self-fulfilling. Certain clinical features may be more characteristic
tem for depression, which will remain essentially unchanged of older people, but this is not necessarily due to depressive illness.
Box 42.1 DSM-V criteria for major depression
A. Five or more of the following symptoms have been present during the same 2-week period and represent change from previous
functioning; at least one of the symptoms is either (1) depressed mood or (2) loss of interest or pleasure.
Note: symptoms that are clearly due to a general medical condition are not included, nor are mood-incongruent delusions or
hallucinations.
(1) Depressed mood most of the day, nearly every day, as indicated by either subjective report (e.g., feels sad or empty) or observa-
tion made by others (e.g. appears tearful).
(2) Markedly diminished interest or pleasure in all, or almost all, activities most of the day, nearly every day (as indicated by either
subjective account or observation made by others).
(3) Significant weight loss when not dieting or weight gain (e.g. a change of more than 5% of body weight in a month), or decrease
or increase in appetite nearly every day.
(4) Insomnia or hypersomnia nearly every day.
(5) Psychomotor agitation or retardation nearly every day (observable by others, not merely subjective feelings of restlessness or
being slowed down).
(6) Fatigue or loss of energy nearly every day.
(7) Feelings of worthlessness or excessive or inappropriate guilt (which may be delusional) nearly every day (not merely self-reproach
or guilt about being sick).
(8) Diminished ability to think or concentrate, or indecisiveness, nearly every day (either by subjective account or as observed by
others).
(9) Recurrent thoughts of death (not just fear of dying), recurrent suicidal ideation without a specific plan, or a suicide attempt or
a specific plan for committing suicide.
B. The symptoms do not meet criteria for a mixed episode.

C. The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning.
D. The symptoms are not due to the direct physiological effects of a substance (e.g. a drug of abuse, a medication) or a general medical
condition (e.g. hypothyroidism).
For example, the concept of ‘masked depression’, in which older impairment is discussed in the section Cognitive Impairment,
people complain of somatic problems and not low mood, has not Depression, and Dementia.
generally been supported and the idea may have reflected a cohort
effect, with older people not wishing to bother their doctor and
only doing so for physical ailments (Baldwin, 1994). Some stud- Early-Onset and Late-Onset Depression
ies have directly compared symptoms in older and younger adults An alternative way to approach depression in older people is to dis-
with depression. One reported the older group to be more severely tinguish depression beginning earlier in life (early-onset depression,
depressed and to have significantly more delusions, agitation, and EOD) from depression presenting for the first time in older peo-
appetite loss (Brodaty et al., 1991), and the other also found older ple (late-onset depression, LOD). Different studies have employed
patients to be more severely depressed and to have more psycho- different age cut-offs, but LOD is most commonly, but arbitrarily,
motor disturbance, psychosis (both hallucinations and delusions), defined as a first episode after age 60. Although it is often thought
and melancholia, and also more appetite loss and weight loss that LOD has a different symptom profile, this does not appear to
(Brodaty et al., 1997). Other studies report an increase in somatic be the case, with three studies of inpatients reporting no differences
complaints, insomnia (Husain et al., 2005), and hypochondriasis in symptom profiles between people with EOD and LOD (Brodaty
(Gurland et al., 1976). Other symptoms such as irritability, hyper- et al., 1997, 2001; Alvarez et al., 2011).
somnia, and pessimism may become less frequent (Husain et al., In summary, the symptoms used for diagnosing depression
2005). Other studies have, however, noted few age-related differ- remain the same in older people and this makes detecting any
ences (Blazer, 2003). Study bias is likely to play a role in such stud- possible differences in symptoms difficult. It is not clear that any
ies as they emanate from specialist units and it is not clear how particular depressive symptom occurs more or less frequently in
well they generalize to other situations. The one type of clinical older people, although in selected hospitalized subjects there may
feature that is clearly more prominent in older people with depres- be some change in their balance, with psychotic and psychomotor
sion is cognitive impairment. Whilst cognitive symptoms are not symptoms being more common. The only symptoms that are more
themselves diagnostic symptoms of depression, they are a critical severe in late-life depression are cognitive impairments, but these
element to examine when assessing older people with depression, are not used for diagnosis and usually specialized tests are needed
and the complex relationship between depression and cognitive to identify them. It is also unclear whether people with an older age
CHAPTER 42 depression in older people 547
at onset of their depressive illness show a different balance of clini- evidence of cognitive decline is evident. A key behavioural change
cal features, but they may show more apathy and less anxiety. to consider is the emergence in later life of alcohol dependence,
which may be a manifestation of a maladaptive strategy for dealing
Assessment of Depression with a depression.
Assessment is discussed in detail in Chapter 9 and only key points
relevant to depression will be discussed here. A major difficulty in Cognitive Impairment, Depression,
assessing depression in older adults is comorbidity (i.e. depression and Dementia
coexisting with physical illness). Both DSM and ICD take an aetio-
logical approach, so that people whose symptoms can be attributed Perhaps the most important clinical difference in depression in
to physical illness are not diagnosed as having depression by these older people is that cognitive impairment is more prominent. The
criteria. Whilst this appears conceptually reasonable, it is often dif- complexity of the relationship of depression to cognitive impair-
ficult in practice for a clinician to determine whether a symptom, ment and dementia is well known and four main relationships
such as insomnia, is due to depression or the physical illness, such between them may be recognized (see Fig. 42.1) (Thomas and
as discomfort arising from shortness of breath. A careful history, O’Brien, 2008).
paying attention to the onset of depressive symptoms in relation
to the comorbid illness, is essential, but frequently it will not be Depression in Pre-Existing Dementia
possible to attribute symptoms to depression with much certainty.
People with dementia commonly have depression, with 24% hav-
This difficulty in deciding whether to attribute a symptom to the
ing significant symptoms in one major community-based study
physical illness or depression is compounded by the recognition
(Lyketsos et al., 2000). Depression is common in Alzheimer’s disease
that, frequently, older people with depression complain of somatic
and is further increased in vascular dementia (VaD) and dementia
symptoms rather than low mood. Pain complaints in all ages are
with Lewy bodies (DLB), with studies reporting major depression
a well-recognized presentation of depression (Katona et al., 2005).
in 10% of subjects with AD but 29% with VaD (Reichman and
These are typically nonspecific, including headache, abdominal
Coyne, 1995). Depressive symptoms were reported in 38% of DLB
pain, and back ache, and may be multiple. The high prevalence
subjects but only 16% of AD subjects (McKeith et al., 1992), and
of pathologies potentially causing such pain in older people, e.g.
major depression was found significantly more often in the DLB
lumbar osteoarthritis, makes it especially difficult to distinguish
subjects (33%) compared with 13% in AD (Ballard et al., 1999). The
physical from mental causation, and a mood-related exacerbation
variation in these figures arises for several reasons, such as the use
of such pathological pain is common. Although part of this somatic
of different samples and assessments, but a difficulty in assessing
phenomenon may be cohort related, i.e. earlier generations found it
major depression arises from the aetiological approach to diagno-
more comfortable to not speak of depressed mood, there are other
sis in DSM, since distinguishing symptoms due to depression from
possible reasons, such as an increased awareness of physiological
those due to dementia is inevitably inexact and subject to variation
changes associated with ageing and disease and a consequent focus
between raters. One consequence is that specific criteria have been
on these. In practice, complaints of pain and other somatic symp-
proposed for diagnosing depression in the presence of Alzheimer’s
toms in the absence of a clearly attributable physical cause should
disease (Olin et al., 2002) which include symptoms not emphasized
raise the strong suspicion of depression and lead to appropriate
in ICD-10 or DSM-IV, such as reduced positive affect, social isola-
mood-related questions. Similarly, the emergence (or significant
tion, and withdrawal.
worsening) of anxiety or obsessional symptoms in an older adult
should lead to questions about mood, as the onset of a pure anxiety
disorder is unusual in old age but it has long been recognized by Depression as a Prodrome of Dementia
clinicians that anxiety, agitation, and obsessional symptoms are fre- Major depression, especially when presenting for the first time
quent features of depressive illness later in life (Post, 1972). Late-life in later life (late-onset depression), can also turn out to be a
depression can present as an anxiety disorder and lead the unwary prodromal presentation of a dementia. In secondary care the
to not recognize depression and treat this appropriately. Deliberate
self-harm may be less frequent in older people, but it should always
lead to careful investigation for depression (or other mental ill-
Depression Dementia
ness) because such acts are more likely to be associated with men-
With Cognitive With Depressive
tal illness and to presage a more serious suicidal act (Manthorpe Impairment/ Symptoms/
and Iliffe, 2010). Older men (more than 75 years) are the group at Pseudodementia Depression
highest risk of suicide and most older people who commit suicide
have major depression (Manthorpe and Iliffe, 2010). As in younger
adults, it is precarious to try to gauge the seriousness of the act from Depression 1 or 2 Years Dementia
its objective level of risk, and close supervision, usually as an inpa-
Age, sex
tient, and treatment are required in such cases (this important topic
is considered in more detail in Chapter 43). Behavioural change Depression
Many Years
Dementia
in an older person may also be a consequence of depression. In Other risk
such cases it is well recognized that a careful cognitive assessment factors
is needed, since the behaviour may be an aspect of an early demen- Fig. 42.1 Four types of relationship of between depression and dementia
tia, but mood assessment is also important, especially where no (Thomas and O’Brien, 2008).
conversion rate was reported as four- to five-fold greater than in on a cognitive test but the pattern of the deficits. Neurocognitive
nondepressed older people (Alexopoulos et al., 1993a), whilst in a impairment in depression has a characteristic pattern of inatten-
primary care study it was found to be one- to two-fold increased tion and executive and amnesic deficits, but aphasia, agraphia,
(Lenoir et al., 2011). Thus those presenting for the first time with apraxia, and acalculia/alexia all suggest the presence of a demen-
depression in old age should be carefully assessed for cognitive tia (Butters et al., 2004; O’Brien et al., 2004). Consistent with the
and functional impairment, with a view to trying to determine view that neurocognitive impairment is a core feature of depres-
whether the extent of such impairments is consistent with the sion, it persists after remission of clinical symptoms (Bhalla et al.,
severity of mood symptoms. In some cases it can be recognized 2006) for up to at least 4 years (Kohler et al., 2010a). Although in
that this is not the case and more detailed neurocognitive and some cases this impairment may only be detectable by detailed
neuroimaging assessments may be indicated to identify a demen- neuropsychological examination, in other cases it is apparent at
tia. In other cases, whilst there may be a strong suspicion, regular interview, and fitting with this long-standing clinical observation
review will be needed to monitor the course of these impairments of cognitive impairments in older people with depression a large
and their relationship to mood symptoms. Although significant proportion (40% (O’Brien et al., 2004) and 54% (Lee et al., 2007))
improvement in mood with antidepressant treatment can be asso- of those with late-life depression were found to meet criteria for
ciated with improvements in cognition, this is not necessarily the mild cognitive impairment (MCI).
case, and when it does occur it may not rule out an underlying It has been proposed that much of this impairment results from
dementing illness. Even in those with unipolar disorder, cogni- a core deficit in information processing, i.e. the speed and accu-
tive impairments are now known to persist after depression itself racy with which information is handled (Butters et al., 2004),
remits (see Cognitive Impairment (and ‘Pseudodementia’) as a but executive dysfunction (Sheline et al., 2006) and vascular risk
Feature of Depression). factors (Sheline et al., 2006; Kohler et al., 2010a) have also been
reported as underlying explanatory variables for higher cortical
deficits. Community-based studies in primary care subjects have
Depression Is a Risk Factor for Dementia also reported similar deficits related to processing speed (Baune et
A thorough meta-analysis of depression studies showed that al., 2006), suggesting that such impairments do not only occur in
depression was associated with a doubling of the risk for subse- late-life major depression.
quent dementia, even when it occurred many years before the onset Whatever the factors underlying these cognitive deficits in
of the dementia; i.e. depression, even in middle age, is an independ- depression, this pattern (poorer information processing, amnesia,
ent risk factor for dementia (Ownby et al., 2006). Depression in this and executive dysfunction) is very similar to that of ageing (Austin
report was a more potent risk factor when it occurred earlier in life et al., 1999) and thus, as expected, age explains a larger portion of
rather than later. It is not clear, however, whether this increased these deficits (Sheline et al., 2006; Kohler et al., 2010a).This raises
risk applies equally to all types of dementia. For example, a 9-year the question as to whether the greater severity of cognitive impair-
follow-up study finding depression to be a risk factor for dementia ment characteristic of late-life depression compared with depres-
reported that depression increased the risk of VaD by three-fold sion in younger adults is simply due to ageing effects. This does
but did not increase the risk of Alzheimer’s disease (Kohler et al., not seem to be the case. Studies that have carefully excluded peo-
2011). ple with early dementia and compared older and younger adults
with depression matched for depression severity have found that,
even after controlling for the age-related increase in impairment,
Cognitive Impairment (and adults with late-life depression have more severe impairments in
‘Pseudodementia’) as a Feature memory and executive functions (Lockwood et al., 2002; Thomas
et al., 2009).
of Depression Thus people with late-life depression who are not developing
It is now well recognized that in adults of all ages without demen- dementia have more severe neurocognitive impairments due
tia who develop depression and who do not progress to dementia to factors associated with depression in older age, in addition
over the following few years, cognitive impairment is a com- to the well-recognized age-related increases. What might these
mon feature and arguably is part of the core of depressive illness depression-related factors be? Some studies have related cogni-
(Austin et al., 2001). Various terms have been applied to the clini- tive impairment to dysfunction in the hypothalamic pituitary
cal complex of cognitive impairment occurring in the context of adrenal (HPA) axis and associated hypercortisolaemia (O’Hara
depression, of which ‘pseudodementia’ is the most well known. et al., 2007), and possession of the short allele of the serotonin
Pseudodementia is poorly defined, which may be related to its transporter has also been linked to cognitive deficits (Hickie et
widespread use. It may mean simply someone with depression al., 2007; O’Hara et al., 2007). The most robust and consistent
who complains of ‘memory problems’ and is said to be forget- relationship is between white matter hyperintensities (WMH)
ful; or when someone with depression scores in the mid-20s on and these deficits (Thomas and O’Brien, 2008). Most studies
the MMSE; or when someone with depression has clinically sig- have been cross-sectional, but WMH at baseline also are the best
nificant impairment on testing with associated functional impair- predictor of persistence of these impairments over the following
ments (i.e. they appear to have a true dementia associated with years (Kohler et al., 2010b). Apart from WMH, detailed stud-
the depression); or it may be used when detailed neurocognitive ies examining correlates between regional and structural brain
testing reveals deficits that are greater than expected for age. The changes and cognition have not been reported, and neither have
term pseudodementia is therefore best avoided because of its correlations between functional brain changes in the frontal lobe
vagueness. Clinically it is important to examine not just a score and impaired cognition been shown.
Differential Diagnosis of Depression, Mild to be multifactorial, involving a complex and poorly understood
interaction between vulnerability factors that predispose people
Cognitive Impairment, and Dementia to developing depression and precipitating factors that initiate a
A clinical conundrum is differentiating from amongst patients with depressive episode; in many cases, these precipitating factors can
‘mixed depression and dementia’ those who have major depression act to perpetuate the episode too. Vulnerability factors are often
with associated cognitive impairment (and are not in prodromal permanent (e.g. sex and other genetic risk) or at least difficult to
dementia) from those who have an early dementia and prominent alter (e.g. social isolation, disability, and white matter ischaemia),
depressive symptoms (or a true double diagnosis of depression and becoming factors that persist beyond a specific episode.
dementia). The evidence in the following sections supporting different
Research studies involve a quality and depth of cognitive test- aetiological factors in late-life depression is pragmatically divided
ing beyond what can usually be carried out in clinical prac- into two categories. The first groups together social risk factors,
tice. As discussed (in the section Cognitive Impairment (and personality factors, and life events, because these factors are gen-
‘Pseudodementia’) as a Feature of Depression), such studies are erally recognized to be interrelated and because research in these
able to identify both prominent impairments, which might be areas has tended to investigate subjects with milder, more broadly
identified in the clinic, and more subtle abnormalities. In clini- defined levels of depression. The second group of biological factors
cal practice, impaired information processing may manifest itself also interact with each other and research into these risk factors has
as poor concentration, and memory and executive problems may generally focused on more severe major depression, usually drawn
be identified by routine tests in clinic, but these typically appear from secondary care samples. This division is not intended to sug-
much milder than is found in people with (even early) dementia. gest there are two different types of depression or that factors in one
Where such deficits are more marked, then a prodromal demen- category are not relevant to those in the other.
tia should be carefully considered. Special attention should in such
cases be paid to identifying the timing of the onset of mood symp- Social factors, personality, and life events
toms in relation to the cognitive impairments and to assessing the There has been little research into the relationship of personality
impact on everyday living, where more marked problems are typi- to the risk of developing depression in old age and evidence points
cal of those with early dementia. However, the pattern of cognitive to (different) personality traits as being both risk factors and pro-
deficits is often more helpful in distinguishing depression-related tective factors. Historically, experts have claimed that older peo-
impairments from those occurring in an early and prodromal ple with depression have had stronger personalities than younger
dementia. Problems with praxis and language (dyspraxias and dys- adults growing old with depression, indicating these were not risk
phasias) are characteristic of early dementia and rarely detectable factors in the development of their depression (Roth, 1955), or
in ‘pure’ depression, and thus the presence of either and especially that older people with less severe community-type depression had
both strongly suggests that an underlying dementia is present and more abnormal personality traits than older people with major
this should then be investigated accordingly with neuroimaging depression (Post, 1972). Such views are close to a more widespread
and more detailed neurocognitive assessment. opinion that more severe depression (major depression) is more
biological in its development, whilst less severe forms of depres-
Aetiology sion are more influenced by personality and psychosocial risk fac-
tors. A difficulty in interpreting the limited data in this field is that
The reason why some older people develop depression and others the direction of causality between these risk factors and depres-
do not is not understood. Most of the risk (and protective) factors sion is unclear, as most studies have been cross-sectional and thus
considered in the following sections affect far more people than unable to determine whether a depression preceded the event or
those who develop depression (see Boxes 42.2 and 42.3). For this followed it.
reason, the genesis of a depressive disorder is widely considered Life-long lack of capacity for intimacy has been reported as a
risk factor for late-life depression in both sexes (Murphy, 1982), in
men only (Emmerson et al., 1989), or women only (Pakhala, 1990).
Box 42.2 Risk factors for depression
Whilst ‘capacity for intimacy’ is not itself a personality variable,
it is related to personality but also to one’s social situation, which
Lack of intimacy Female
in turn is related to personality, making assessment of this diffi-
Poor social support Family history of depression cult. Such assessments are also confounded by current mood state.
Being a carer Physical illness Another study failed to identify any relationship with intimacy
but found depression associated with a lack of general social con-
Bereavement Brain disease
tact (Henderson et al., 1986), and the perception that depression
in older people is frequently related to loneliness and social isola-
tion is a common one (Blazer and Hybels, 2005). More positively,
Box 42.3 Protective factors for depression
self-efficacy and mastery over one’s environment are reported to
be protective factors for late-life depression (Blazer and Hybels,
Self-efficacy
2005). Along the same lines is evidence that psychological resil-
Mastery of environment ience (an ability to properly interpret and understand events and
Wisdom persevere through them) and adaptive coping styles reduce the
risk of depression after adverse life events (Arean and Reynolds,
Religious practice
2005). Wisdom in older people, which like capacity for intimacy
and psychological resilience is an ill-defined but generally under- to distinguish symptoms of normal grief from similar symptoms
stood concept, has also been reported to be associated with greater of depression. In the first month, most bereaved people experience
life satisfaction (Reichstadt et al., 2010) and to be protective against low mood, anorexia, insomnia, episodes of weeping, fatigue, loss
depression (Blazer and Hybels, 2005). Such personality factors not of interest, and guilt. Such symptoms tend not to be as persistent
only are difficult to define and influenced by one’s social setting, or severe as in depression and when guilt is present the focus is
upbringing, and experience, but also in terms of the development on perceived failures towards the deceased, and general pessimism
of depression interact with life events. That is, different people han- and suicidal thinking are rare. These symptoms gradually wane as
dle the same life events differently and thus not everyone who loses grief is dealt with and by about a year are largely gone. However, in
a loved one or suffers other adverse life events develops depression. older people who have lost a spouse after many years of marriage
Furthermore, although research generally examines negative life it is understandable that some grief symptoms continue for a long
events, people experience positive ones, e.g. marriages of children time episodically following the loss, e.g. occasional weeping when
or births of grandchildren, which can be mood elevating. reminded of the loved one. Such symptoms are likely to worsen at
The role of social isolation and lack of social support in influenc- key dates, e.g. the wedding anniversary. Importantly, normally the
ing the development of depression has been mentioned and this bereaved person is able to return to a prebereavement level of func-
is a risk factor in African (Gureje et al., 2011) and Japanese (Kaji tioning, and whilst such low-level grief symptoms may return from
et al., 2010) as well as western societies. Poverty and being a vic- time to time, they are not continuous and the persistence of symp-
tim of crime, which are interrelated, have also been reported to be toms, especially when they affect day-to-day functioning, suggests
risk factors (Arean and Reynolds, 2005). Being a carer of someone, a depression has developed. In addition to such a failure to resolve
an increasingly common experience for older people, is another grief symptoms, other markers of concern are: generalized guilt,
important risk factor for depression, with one study reporting that severe feelings of worthlessness, wishing one were dead, suicidal
a quarter of carers of people with dementia were depressed (Ballard thinking, psychomotor changes, and ‘mummification’ (where grief
et al., 1996b). Religious practice has been reported to protect against is perpetuated by a failure to adapt one’s lifestyle appropriately to
depression in older adults in both western (Braam et al., 2001) and the loss). In interpreting the impact of bereavement it is important
eastern (Bosworth et al., 2003) cultures. to remember that grief is a highly personal and individual experi-
Life events have been associated with the development of depres- ence. Thus loss of a spouse may be a major blow to one man who
sion in older people, although it is important to remember that had enjoyed a warm and supportive relationship with his wife but
the impact of life events on mood is mediated by cognitive coping a relief to another who had endured many years of tension, argu-
style (Meyer et al., 2010). One study reported that 48% of depressed ments, and violence. Moreover, different people have differing lev-
older people had had at least one severe life event in the previous els of social support to help buffer the impact of grief and different
year compared with 23% of their nondepressed peers (Murphy, personality traits, which may confer either resistance to developing
1982). These events included: death of a loved one, life-threatening depression or vulnerability. Thus a quarter of those experiencing
illness to somebody close, major financial problems, and having to a major adverse event did so without developing depression. It is
give up one’s home suddenly. Although broadly similar to findings clear that not all adverse events are followed by depression and not
in younger adults, an important difference was the role of physical all depressive episodes are preceded by adverse life events.
illness, either a chronic disabling one or a severe and recent illness.
The former relationship is supported by other findings of a relation- Medical and biological factors
ship between disability and handicap and depression. A study in Depression in all ages occurs more frequently in women (Cole and
London of 654 people over 65 found broadly defined depression Dendukuri, 2003) and a previous history of either major depression
associated with living alone, low income, and low social support, or dysthymia increases the risk of developing another depressive
but that the strongest association was with disablement, particu- episode in old age (Cole and Dendukuri, 2003). It is often said that
larly handicap (Prince et al., 1997a). Life events (most strongly genetic risk for depression declines with age, but it is important to
bereavement, personal illness, and theft), especially in the preced- clarify that if present this decreased risk applies to late-onset depres-
ing 6 months, also had associations with depression (Prince et al., sion. Studies comparing people with a first episode of depression in
1997b). Handicap may simply be a summary factor for all the social old age with those whose first episode was earlier in life have con-
disadvantage and chronic difficulties associated with depression, sistently reported that the former have fewer relatives with depres-
but a prospective analysis of this cohort found disablement, espe- sion (e.g. Hopkinson, 1964; Brodaty et al., 1991, 1997; Baldwin
cially handicap, to be strongly predictive of incident depression and Tomenson, 1995) and this has been confirmed by systematic
(Prince et al., 1997b). The problem is how to interpret this broadly review (Levinson, 2006). In contrast, a large Danish twin study
defined construct, since it appears to include contributions from all found no change in the heritability of depression with increasing
the physical illness and psychosocial factors that have been shown age (Johnson et al., 2002) and a Genome Wide Association Study
to be risk factors for depression. Life events in childhood, as well (GWAS) of over 3500 subjects also found no change in heritabil-
as recent life events, have been associated with an increased risk ity of depression with increasing age (Demirkan et al., 2011). As
of depression, but the former did not increase the risk of the latter has long been reported, this study confirmed that heritability in
producing depression;i.e. childhood events did not confer vulner- depression at all ages appears to be due to multiple genetic loci
ability for late-life depression (Comijs et al., 2007). of small effect rather than it being related to a few genes of large
Bereavement is reported as a frequent precipitant of depres- effect. Although candidate genes related to monoamine systems
sion in older people. However, the relationship of bereavement to have been linked to major depression in some studies and other
depression is complex and appears to be mediated by social context genetic loci implicated, e.g. polymorphism of the serotonin trans-
and personality factors (Pai and Carr, 2010). First there is the need porter promoter region (5-HTTLPR), there remains no convincing
evidence for a significant effect on depression from any single gene al., 2002). However, such data are very difficult to interpret because
(Levinson, 2006). people are taking medication for pre-existing diseases which are
Several major diseases have been associated with high rates of themselves related to depression, and detailed assessments trying to
depression. As already noted, neurological disorders, especially tease out the contribution due to drugs themselves have not clearly
those causing dementia, are strongly associated with high rates of supported a role for them in causing depression. For example, after
depression. But vascular diseases and risk factors have an especially reports that calcium channel blockers were associated with depres-
strong relationship with depression. Ischaemic heart disease (IHD) sion a review of these demonstrated this was not the case (Dunn et
is associated with a two- to three-fold increase in depression, and al., 1999), and secondary analyses of studies in hypertension have
stroke disease with a similar increased risk, and it has long been reported possible benefits of calcium channel blockers on mood
clear that the relationship between each of these and depression is (Ried et al., 2005). Whilst it is prudent to reflect on the possible role
bidirectional, with depression being an independent risk factor for of medication when depression appears to start after a treatment
IHD and stroke as well as these predicting increases in subsequent is commenced, it would be rash to consider switching medication
prevalence rates of depression. It is also now clear that the same is unless a convincing relationship is present, given the precarious
true of diabetes, with depression being a stronger risk factor for the nature of the current evidence base. Furthermore, since most older
development of diabetes than diabetes is for depression (Mezuk et people are on multiple medications, it is likely to be difficult to decide
al., 2008). The situation is less clear for blood pressure and hyper- which if any medication may be worth stopping or changing.
tension, with some studies finding that depression predicts the
development of hypertension (e.g.Davidson et al., 2000; Meyer et
al., 2004), but others finding no such relationship (e.g. Simonsick et
Neurobiology of Late-Life Depression
al., 1995; Jones-Webb et al., 1996), and others reporting a relation- The most widely known neurobiological abnormalities are those
ship of depression to low blood pressure (Lenoir et al., 2008) or that involve the monoaminergic neurotransmitters serotonin
orthostatic hypotension (Richardson et al., 2009). The cause of the and noradrenaline, because all conventional antidepressants tar-
relationship of these to depression varies and involves several com- get these two neurochemicals. However, research on monoam-
ponents. In some cases (especially the neurological and vascular ines is only a small part of the much wider evidence identifying
disorders) there is good evidence for a biological relationship and neurobiological abnormalities in older people with depression,
this is considered in detail in the section Neurobiology of Late-Life although most of the evidence comes from studies of major
Depression). In others, pain and discomfort related to physical ill depression.
health, including hip fracture and chronic painful conditions such
as rheumatoid arthritis, appear to mediate depression via psycho- Neuroanatomy
logical pathways which are thought to be bidirectional (Lin et al., Neuroimaging research in psychiatry has been central to the rec-
2003). Another aspect of the complex relationship between depres- ognition that several neurally linked key brain areas underpin the
sion and physical ill health is that involving the handicap induced symptomotology of depression and much of this research has focused
by the illness. The physical impairments of a significant illness cre- on late-life depression. Studies in primates initially identified five
ate disability that in turn places the sick person at a societal dis- frontal-subcortical circuits (FSC) and three of these, involving the
advantage, which constitutes their handicap (Prince et al., 1998). dorsolateral prefrontal cortex (DLFPC)-head of caudate nucleus
Thus an impairment induced by an illness will produce differing (DLPFC circuit), the anterior cingulate cortex (ACC)-nucleus
degrees of handicap in different people because of their differing accumbens (ACC circuit), and the orbitofrontal cortex (OFC)-head
social situations. For example, one person with chronic obstructive of caudate nucleus (OFC circuit), have been implicated in the regu-
pulmonary disease living alone may become highly socially isolated lation of mood (Lichter and Cummings, 2001). Evidence implicates
and depressed due to this loss of personal contact, whilst another dysfunction in these circuits in the onset and outcome of late-life
person with the same condition living with a spouse is able to get depression (Alexopoulos, 2005), with changes most pronounced in
out and about and enjoy company at home. Understanding the rela- late-onset cases (Lloyd et al., 2004).
tionship of physical illness to depression involves careful considera- Structural imaging studies have shown that frontal lobe volume is
tion for each individual of the interplay between these biological, reduced in depression (Kumar et al., 2000; Lai et al., 2000; Almeida
psychological, and social elements. et al., 2003; Chang et al., 2011). Structural studies have identified
The possible role of drugs, for older people mainly medica- specific areas of the prefrontal cortex as reduced in volume and
tion and alcohol, should also be considered. It is well recognized functional imaging has confirmed these areas have reduced activity
that alcohol lowers mood and that high alcohol intake may lead to and/or bloodflow in depression. Thus the ACC is smaller (Drevets et
depression or aggravate it. Moreover, often excess alcohol consump- al., 1997) and has reduced blood flow (Drevets et al., 1997; Mayberg
tion is a consequence of a depression, with it being consumed to et al., 1997, 1999), the DLPFC is smaller (Chang et al., 2011) and
try to minimize the painful depression-related feelings. Assessment has reduced blood flow (Baxter et al., 1989; Bench et al., 1992), and
of depression should always include a sensitive consideration of the the OFC is smaller (Bremner et al., 2002; Ballmaier et al., 2004).
role of alcohol. Some reports have claimed that different types of Subcortically, decreased grey matter volumes in limbic and striatal
prescribed medication are associated with depression. For example, structures have also been described in older depressed subjects in
an epidemiological study from the Netherlands estimated the popu- the amygdala, thalamus (Andreescu et al., 2008), and hippocampus
lation attributable risk percentage (PAR%, the proportion of depres- (Bell-McGinty et al., 2002; O’Brien et al., 2004; Andreescu et al.,
sion in the population attributed to a drug) for a range of common 2008; Steffens et al., 2011) as well as caudate (Krishnan et al., 1992;
treatments and found the PAR% for β-blockers was 2.5%, for cal- Parashos et al., 1998; Butters et al., 2009) and putamen (Andreescu
cium antagonists 5%, and for benzodiazepines 15.42% (Dhondt et et al., 2008).
The role of the FSC in the development of late-life depression et al., 1984) so that HPA axis dysfunction is more marked in late-life
is further supported by other areas of research. Thus ischaemic depression (O’Brien et al., 1993).
WMH (see section Neuropathology) are especially prominent Animal studies indicate that prolonged exposure to glucocorti-
in the frontal lobes and basal ganglia (Herrmann et al., 2008) coids may reduce dendritic density and increase neuronal death,
and diffusion tensor imaging studies have reported white matter and since the hippocampus has the highest concentration of glu-
abnormalities in these same frontal areas (Alexopoulos et al., 2002; cocorticoid and mineralocorticoid steroid receptors in the brain,
Taylor et al., 2004). Neuropathological abnormalities have also it may be especially vulnerable to such toxicity. High cortisol levels
been identified in these same brain areas (see Neuropathology) have also been reported to lower brain-derived neurotrophic fac-
(Khundakar and Thomas, 2009). Studies of neurocognitive tor (BDNF) in the hippocampus, with antidepressants reversing
impairment in late-life depression (see Cognitive Impairment this change (Nestler and Carlezon, 2006). Hippocampal atrophy
(and ‘Pseudodementia’) as a Feature of Depression) have also has been frequently reported in depression in older people (see
shown a pattern (impaired information processing and memory Neurochemistry) and amnesia is a prominent problem on detailed
and executive function) consistent with this neuroanatomy, and cognitive assessment. Thus HPA axis dysfunction may help explain
direct evidence of a relationship between the anatomical location the prominence of amnesic deficits in late-life depression. Although
of WMH and this neurocognitive impairment has been reported one study reported a relationship between potential exposure to
(Sheline et al., 2008). high levels of cortisol and hippocampal atrophy (Sheline et al.,
These neurobiological investigations in late-life depression impli- 1996), other studies have not replicated this (O’Brien et al., 2004),
cating the FSC have not focused on the monoaminergic nuclei in nor is it apparent that amnesic deficits are related to HPA axis dys-
the brainstem. The raphe nuclei (serotonin) and locus coeruleus function and so it is not clear how important this mechanism is in
(noradrenaline) in the brainstem are the other key brain areas late-life depression.
forming the neuroanatomical substrates for depression.
Immune system and inflammatory cytokines
Neurochemistry The suggestion of a relationship between depression and the
The neurotransmitters relevant to depression can be conceived of immune system dates back to at least 200 ad when Galen claimed
as having executive and modulatory components. The former are that melancholic women were more susceptible to breast cancer
the main neurotransmitters of the major projection neurons in the due to immune deficiency. Research during the last two decades
FSC, glutamate and gamma-amino butyric acid (GABA), and some has made it clear that the nervous, endocrine, and immune sys-
evidence implicates abnormalities in these in depression (Sanacora tems are so closely connected that it is suggested that they should
and Saricicek, 2007; Zarate et al., 2010). The modulatory neuro- be regarded as a single network, and their intimate relationship has
transmitters are the monoamines (serontonin, noradrenaline, and led to the development of the new discipline of psychoneuroim-
also dopamine) projecting from their brainstem nuclei to wide- munology. Initially it was thought that psychological stress and
spread parts of the cortex, having dense input to the FSC areas. depression compromised immune function (Irwin et al., 1990),
Abnormalities in serotonergic and noradrenergic neurotransmis- but studies failed to replicate initial findings and more recent stud-
sion are undoubtedly present in late-life depression, and this is ies indicate that depression is associated with immune activation,
most clearly supported by the proven efficacy of antidepressants including increased peripheral blood mononuclear cells (such as
which all target these transmitter systems (see Management of neutrophils, monocytes, and activated T-lymphocytes), increased
Depression). Dopaminergic changes also may be present (Nestler secretion of prostaglandin E2, and the presence of a moderate acute
and Carlezon, 2006) and may be especially relevant where there is phase reaction (with increased haptoglobin, C-reactive protein, and
apathy, reflecting loss of initiative and drive subsequent to loss of α1-antitrypsin) (Joyce et al., 1992).
dopaminergic function; this is often the case in ‘vascular depression’ Linked to this immune response is evidence of increases in
(see ‘Vascular Depression’). Although impairments in monoam- proinflammatory cytokines and reductions in anti-inflammatory
inergic neurotransmission are present in late-life depression, it cytokines in depression (Anisman et al., 1999). Cytokines are pro-
is not clear there are any age-associated changes in these neuro- teins (such as interleukins (IL), tumour necrosis factor (TNF), and
chemicals which might make older people more or less vulnerable interferons) secreted by a variety of cells, including glial cells in the
to depression (Veith and Raskind, 1988). Moreover, the failure of a CNS, which play a key role in mediating immune and inflammatory
significant proportion of older depressed people to respond to the responses. Brain cytokines have a broad range of immunological,
conventional monoamine-based antidepressants indicates that not neurochemical, neuroendocrine, and behavioural effects (Rothwell
only other transmitters may also need to be targeted, e.g. GABA and Luheshi, 2000) and their release during illness is associated with
and dopamine, but also other pathophysiological processes are so-called sickness behaviour, including symptoms such as fatigue,
likely to be important, including those affecting vascular, immune, anorexia, depressed mood, hopelessness, anhedonia, and poor con-
and neuroendocrine systems. centration. Most studies in depression demonstrating increases in
cytokines have been on younger adults, but in late-life depression,
Neuroendocrine system similar changes have been observed. IL-1β, IL-6, and TNF-α have
Dysfunction in the HPA axis is recognized to occur at all ages in all been reported to be raised in both depression and dysthymia in
depression and is associated with hypercortisolaemia, nonsup- both community- (Dentino et al., 1999; Penninx et al., 2003) and
pression on the dexamethasone suppression test, and a loss of the hospital-based studies (Thomas et al., 2005) and there appears to
characteristic circadian rhythm of the HPA axis. In addition, ageing be a dose-related effect (Dentino et al., 1999; Penninx et al., 2003;
itself is associated with a similar pattern of changes (Alexopoulos Thomas et al., 2005).
Neuropathology Other neuropathological investigations have directly examined

The neuroimaging evidence identifying structural and functional WMH to assess their pathology. In a study comparing 20 people
changes in FSC in depression has led to post-mortem neuropatho- with late-life depression and 20 age-matched controls, DWMH were
logical investigations seeking to identify the cellular correlates of identified on post-mortem imaging and then examined microscopi-
these abnormalities. These studies have been able to take advan- cally. Whilst some lesions in controls were due to myelin pallor that
tage of the development of sophisticated computer-based image was not ischaemic, all WMH in depression were ischaemic and the
analysis methods, especially unbiased stereological studies, to esti- difference from controls was most marked in the DLPFC (Thomas
mate changes in the size of neurons and density of neurons and et al., 2002b). This indicates that not only are DWMH increased
glia (Khundakar and Thomas, 2009). In younger adults, such stud- in late-life depression but also such DWMH are due to cerebrov-
ies have consistently reported a reduction in glial density in major ascular disease and that such disease is more frequent in late-life
depression in the ACC (Ongur et al., 1998; Cotter et al., 2001), depression. In contrast to these findings in DWMH, examination
DLPFC (Rajkowska et al., 1999), and OFC (Rajkowska et al., 1999), of periventricular WMH found that most, whether in control or
but no differences in neuronal density or number. In contrast, stud- depressed subjects, were not ischaemic but due usually to disrup-
ies of late-life MDD have reported no evidence of reduced glial tion of the ependymal lining of the ventricles (Thomas et al., 2003).
density in cortical areas (Khundakar and Thomas, 2009). A study These findings support earlier work on the relationship of vascular
examining glial and neuronal density in 15 older subjects and 11 risk factors (VRFs) to WMH, which indicates that whilst DWMH
age-matched controls in the six laminae from the rostral region of are strongly related to VRFs and are due to brain ischaemia, PVH
the OFC (BA 47) reported a significant (30%) reduction in pyrami- are not. Other studies identified an increase in cell adhesion mol-
dal neuron density in MDD subjects in the OFC overall, largely due ecules, markers of an ischaemia-induced inflammatory reaction, in
to significant reductions in layers 3 and 5 of the OFC (Rajkowska the DLPFC in late-life depression in both the grey and white matter,
et al., 2005). The study also showed a negative correlation between suggesting that ischaemic brain disease in late-life major depres-
age and overall neuronal density in both MDD and control groups. sion extends beyond WMH into other white matter and grey mat-
However, a second study examining the OFC in late-life depres- ter (Thomas et al., 2000, 2002a).
sion found no evidence of any differences in neuronal density or Post-mortem studies have also identified evidence of an increase
volume (Khundakar et al., 2011b). In the DLPFC, evidence of in atheromatous disease in depression in older people (Thomas et
pyramidal cell pathology was found through a reduction in volume al., 2001) which was present even though subjects were matched
across all layers, and analysis of individual cortical layers revealed for vascular risk factors, and this was supported by a clinical study
lamina-specific reductions in pyramidal neuronal volume in layer 3 also identifying an increase in atheroma in late-life depression
and, more markedly, layer 5 of the DLPFC (Khundakar et al., 2009). (Tiemeier et al., 2004). Although such studies provide evidence
But again, another study of this region did not replicate these find- of an increase in vascular and cerebrovascular disease, there is no
ings, reporting no evidence of glial or neuronal abnormalities com- evidence of an increase in neurodegenerative disease in late-life
pared with age-matched controls (Van Otterloo et al., 2009). An depression in subjects who were cognitively intact at death in either
examination of the ACC in late-life depression also did not find the cortex (O’Brien et al., 2001; Tsopelas et al., 2011) or brainstem
evidence of either neuronal or glial cellular changes (Khundakar (Hendricksen et al., 2004; Syed et al., 2005).
et al., 2011c).
White matter hyperintensities
The amygdala and the entorhinal cortex have been examined in
tissue blocks acquired from either left or right hemispheres of older Cerebral white matter lesions (usually called white matter hyper-
MDD (seven in the amygdala, six in the entorhinal cortex) and intensities (WMH)) identified on magnetic resonance imaging
control (ten) cases (Bowley et al., 2002). Glial density was signifi- (MRI) are frequently reported in older people (Jeerakathil et al.,
cantly reduced in the amygdala in MDD cases, mainly as a result of 2004), but many studies have demonstrated they are increased in
reductions in the sections obtained from the left hemisphere. Glial depression in older people (Herrmann et al., 2008). Box 42.4 sum-
density was also reduced in the entorhinal cortex, though not sig- marizes their relationship to depression in older people. Although
nificantly. No changes were found in neuronal density in either the most of the research has been on major depression from hospital-
amygdala or the entorhinal cortex and neuronal size was not ana- derived samples, several community-based studies of depression
lyzed. Further histopathological examination appeared to suggest
that two of the subjects (one MDD and one control) had indicators
Box 42.4 MRI white matter hyperintensities and depression
of preclinical Alzheimer’s disease; both of the subjects had higher
glial densities on average in both diagnostic groups. The caudate Increase in volume
nucleus was examined in 13 late-life MDD and nine age-matched
Increase in volume over time
controls and a reduction in neuronal density, but not in neuronal
volume, was reported but again no change in glial cells (Khundakar More prevalent in frontal-subcortical areas
et al., 2011a). Associated with vascular risk factors
Overall, the findings are suggestive, but not conclusive, that sub-
Due to ischaemic brain disease
tle neuronal abnormalities may be present in key areas of the FSC
in late-life depression. The prefrontal pyramidal neurons are mainly Predict future depressive symptoms and episodes
glutamatergic and, as well as projecting to other cortical areas, also Associated with poorer remission rates
project to the striatum, and it may be that ischaemic WMH affect
the axons of these neurons, inducing these subtle changes. Predict chronic cognitive impairment
in older people have also reported increases in WMH (Steffens that cerebral ischaemic damage to frontal-subcortical circuits pre-
et al., 1999; de Groot et al., 2000). WMH are more common in disposes to and/or perpetuates depression in older people. This
LOD than EOD (Teodorczuk et al., 2007; Herrmann et al., 2008). concept has received criticism because delineation of a recogniza-
These lesions have frequently been associated with hypertension ble subgroup of ‘vascular depression’ has proven difficult (Baldwin,
(Dufouil et al., 2001; Jeerakathil et al., 2004) and other vascular 2005). More importantly, several community studies of late-life
risk factors, such as carotid atherosclerosis (Pico et al., 2002) depression have not identified an increase in clinically determined
and smoking (Jeerakathil et al., 2004). WMH have been associ- VRFs (Lyness et al., 1998; Lyness et al., 1999; Kim et al., 2004;
ated with VRFs in depression specifically (O’Brien et al., 1996; Naarding et al., 2007), casting further doubt on the importance of
Smith et al., 2010), and additionally in late-life depression WMH specific VRFs in the genesis of late-life depression. It is important
have been associated with orthostatic hypotension and drops in to note, however, that the same cohorts followed prospectively have
systolic blood pressure (Richardson et al., 2009; Vasudev et al., reported an association of depression with some VRFs (Lyness et
2011). They have therefore been widely regarded as due to cer- al., 2000; Kim et al., 2006) and that the increases in WMH occur
ebrovascular disease and this has been demonstrated to usually independently of the association with VRFs (Herrmann et al.,
but not always be the case in post-mortem studies (Thomas et al., 2008). This has been directly demonstrated in a large study match-
2002b, 2003; Fernando et al., 2006), and in late-life major depres- ing depressed and control groups for VRFs (Sheline et al., 2008).
sion WMH are more frequently due to cerebral ischaemia than in The explanation for these discrepant findings probably reflects the
age-matched controls (Thomas et al., 2002b). WMH are especially insensitivity of (peripherally measured) conventional VRFs for
increased in the frontal lobes and basal ganglia in older people actual ischaemic disease in the brain and the differences in sam-
with depression (Herrmann et al., 2008) and have been related ples, with the MRI and pathology studies assessing major depres-
to the neurocognitive impairments characteristic of depression sion in secondary care patients, whilst the community studies
(Sheline et al., 2008). sample milder cases of depression and ones that may have a differ-
Prospective studies have reported WMH to increase in volume ent aetiology. The ‘vascular depression’ model postulates that cer-
over time in older people with depression (Taylor et al., 2003) and ebrovascular disease is an important aetiological factor in late-life
that these lesions have a bidirectional relationship with late-life depression and likely to be an important one in a proportion of
depression. Baseline severity of WMH predicts the development such people, but does not propose a unifying mechanism involving
(Godin et al., 2008) and worsening (Teodorczuk et al., 2007, 2010) vascular disease for all late-life depression. Estimates of the propor-
of depression, but depression also predicts the worsening of WMH tion of people with late-life major depression where vascular dis-
(Godin et al., 2008). Although the impact of antidepressants on the ease is important indicate this is likely to be the case in about 50%
development and progression of WMH has not been examined, it of cases (Sneed et al., 2008a), and this delineation was based on uti-
has been reported that antihypertensive treatment reduces the pro- lizing evidence of significant WMH on MRI rather than VRFs. Two
gression of WMH (Dufouil et al., 2005; Firbank et al., 2007). Since intervention studies using the calcium channel blocker nimodipine
it is clear that WMH have an adverse impact on treatment response have also provided encouraging results (Taragano et al., 2001,
and longer-term outcomes (see Management of Depression), the 2005), suggesting that the ‘vascular depression’ model may provide
amelioration of these lesions by medication raises the hope that a way of developing new treatment approaches in late-life depres-
their likely role in the development and maintenance of depression sion, and such approaches are likely to target people with WMH
could be attenuated. who are those who generally have poorer responses and outcomes
with current treatments.
‘Vascular Depression’
Evidence clearly demonstrates that depression has a bidirectional Management of Depression
relationship with vascular disease (Thomas et al., 2004). Prospective The clinical management of depression requires a thorough assess-
studies of vascular diseases have reported that previous major ment and diagnosis, since diagnosis in older people is frequently
depression or depressive symptoms predicts a two- to three-fold difficult because of confounding cognitive impairments and physi-
increase in coronary heart disease (e.g. Pratt et al., 1996; Ford et al., cal illnesses, and because one neglected reason for nonresponse to
1998) and stroke disease (e.g. Jonas and Mussolino, 2000; Salaycik treatment and treatment resistance is misdiagnosis. Assessment
et al., 2007). The increase in WMH, which are ischaemic lesions, should aim to generate an understanding of the individual impact of
in late-life depression provides direct evidence for a role of cer- the current depressive episode on patients and their everyday lives,
ebral ischaemia, and this is further strengthened by reports from and not merely be a process of identifying and counting depres-
diffusion tensor imaging studies of increased diffusivity in late-life sive symptoms, since the number of symptoms is not a good guide
depression (Taylor et al., 2004), which is a further indicator of the to the severity of depression. Assessing the degree of functional
presence of white matter tract disruption due to cerebral ischaemia. impairment or disability related to the depression is key to develop-
This increased diffusivity occurs in areas without WMH, suggesting a good management strategy. This allows the potential benefits
ing a more widespread impact of cerebrovascular disease than that and risks of treatment to be carefully weighed. This approach was
identified by MRI WMH, and this is further supported by neuropa- proposed in the British Association of Psychopharmacology (BAP)
thology reports of increased cell adhesion molecule expression in guidelines (Anderson et al., 2008) and has been adopted by NICE
the DLPFC, showing vascular changes in frontal grey as well as in its most recent Guidelines for the Management of Depression
white matter (Thomas et al., 2002a). (NICE, 2009). These both emphasize that identifying duration
This evidence has led to and provided support for the ‘vascular of symptoms and the disability associated with them is crucial
depression’ hypothesis (Alexopoulos et al., 1997), which postulates to informing rational management and leads to a stepped care
approach based on the combination of burden of symptoms, dura- of relapse. During the acute phase of treatment, management of
tion of illness, and depression-related functional impairment. suicide risk and possibly self-neglect are important, and treat-
Assessment has been discussed in detail in Chapter 9 and ear- ment aims at more than symptom relief by trying to achieve a
lier in this chapter and here a few brief points relevant to depres- restoration of previous levels of everyday functioning. Once this
sion management will be emphasized. A careful risk assessment, is achieved, the management focus moves to reducing the like-
including sensitive questioning about suicidal thinking and previ- lihood of future depressive episodes through continuation and
ous acts, is an integral part of the assessment of depression and, maintenance treatment. Continuation treatment usually means
where depression is severe, suicidal thinking should be suspected continuing with whatever treatment got someone well for at least
and careful monitoring instituted, even where such thinking is not 6 months after remission. Whilst in younger adults with a first
volunteered. In such cases, it is also important to probe for symp- episode, treatment may then be withdrawn, it is argued in the
toms of psychosis, especially mood congruent delusions. A cogni- next section that for major depression in older adults continua-
tive assessment should be carried out whenever there is suspicion tion should always be followed by maintenance; all late-life major
of cognitive impairment; however, this need not always be the case depression (but not necessarily milder forms of depression) needs
and in those presenting with a ‘pure’ depression this may be unhelp- lifelong treatment.
ful, suggesting a dementia is suspected, which might damage the
therapeutic relationship. In those with a previous history of depres-
sion it is important to obtain details of treatments for these earlier Prescribing in Late-Life Depression
episodes, whether they were successful or unsuccessful, as this can There are several general topics that need to be considered when
be very helpful in guiding treatment of the current episode, since prescribing antidepressant medication in older adults. With age-
patients tend to respond and not respond to the same treatments. ing, there are pharmacokinetic changes that alter the bioavailability
of drugs. These are considered in detail in Chapter 13 and so only
General principles of management key aspects will be summarized here. There is reduction in hepatic
As when treating younger adults, it is helpful to consider treatment metabolism which reduces biotransformation and the impact of
as having three stages: an acute phase, a continuation phase, and a first-pass metabolism, allowing more active drug to enter the sys-
maintenance phase (Frank et al., 1991). The stepped care approach temic circulation; volume of distribution is typically increased
recommended by NICE and BAP applies to the acute phase, and with ageing due to the relative increase in adipose tissue; and,
ties in with an important finding in clinical trials of antidepres- most importantly, there is an age-related reduction in glomerular
sants: the more severe the depression, the stronger the evidence filtration rate. Together these changes lead to higher plasma and
for efficacy, and the greater the benefit from using this medication brain levels of active drug per unit dose compared with younger
(Khan et al., 2002; Kirsch et al., 2008). This increased benefit comes adults and consequently the need for lower doses. However, there
both from antidepressants bringing about a larger reduction in is much variability in these changes and clinicians are used to
depression when depressive illness is more severe and from pla- observing such variability. Some 75-year-olds are much ‘younger’
cebo having a smaller impact on depressive symptoms in the more physically than others and such ‘younger’ older people are likely
severely ill (Khan et al., 2002). The conclusion from this is that a cli- to need and to tolerate higher doses; it is prudent to remember
nician needs to weigh up the severity of the depressive illness when this if someone is not responding to a typical dose for an older
considering whether to use antidepressants and how vigorously to adult. A related factor that affects dosing at all ages is genetic vari-
treat. Thus, following the stepped care approach, those with few ability. Polymorphisms in key enzymes of the cytochrome P450
symptoms (‘subthreshold’) and little impact on everyday function- system which metabolize psychotropic drugs can lead to a wide
ing should not be offered antidepressants but instead be given sup- range of plasma levels for the same drug dose. Of course, in regu-
portive and low-intensity psychosocial interventions. In those with lar practice the clinician does not know which polymorphisms a
mild major depression (or persistent ‘subthreshold’ symptoms) a patient possesses, but it is helpful to be aware they exist because,
more intensive psychological treatment is indicated, but antide- as with variability in age-related pharmacokinetic changes, inad-
pressants are unlikely to be of benefit; these should be reserved equate response to a standard or even ‘high’ dose of antidepres-
for when such interventions fail and disability is significant. Those sant may be because a ‘high metabolizer’ is being undertreated.
with moderate or severe major depression are those most likely to Pharmacodynamic changes also occur with ageing and are prob-
benefit from antidepressants, and medication is then indicated as ably due to age-related changes in dendritic and synaptic density,
first-line treatment, and combining medication with intensive psy- and may explain why some older people are more prone to adverse
chological treatments is often appropriate. This leads back to the effects, e.g. antipsychotic-induced tardive dyskinesia (Jeste, 2000),
BAP recommendations of its ‘dimensional’ approach to prescrib- even at low drug levels. Furthermore, pharmacodynamic changes
ing antidepressants (Anderson et al., 2008). Those with MDD and may explain the propensity of older people to develop disabling
significant severity (taking into account both symptoms and the tremor even with low serum levels of lithium.
impact on everyday living) should be treated robustly. Similarly, In a different category from these biological alterations is poly-
when someone presents with a longer illness, say over 3 months, pharmacy. This continues to be a major issue in older adults and,
even though the severity is milder, antidepressants are indicated given the large and increasing numbers of effective available treat-
because spontaneous recovery or improvement is less likely. ments for illnesses strongly associated with ageing (statins, angi-
There are several goals in the management of depression, but otension receptor blockers, proton pump inhibitors, etc.), it seems
aiming to achieve clinical remission, rather than simply symp- impossible to avoid. Awareness that the likelihood of adverse
tom reduction, is vital because persistence of depressive symp- drug–drug interactions and related problems is proportional to the
toms (‘residual depression’) is strongly associated with high rates number of prescribed drugs should encourage regular medication
review, especially where new symptoms develop when a new drug Meta-analytic systematic reviews have reported clear benefits
is added (Chrischilles et al., 2009). The number of psychotropic from the use of antidepressants in late-life major depression (Nelson
drugs being prescribed has increased significantly in recent years, et al., 2008; Wilson et al., 2001). However, the reported trials have
making this an increasing problem (Mojtabai and Olfson, 2010). been far from unanimous in this verdict and it is important to con-
For example, the prescribing of antidepressants in nursing homes sider reasons why findings are heterogeneous. Some of the earlier
in the US more than doubled from 21.9% to 47.5% since the turn trials were less well designed, especially those of tricyclic antide-
of the century; interestingly, increased involvement by nursing pressants, and in particular were often underpowered (Wilson et al.,
staff was associated with an increase in antidepressant prescrib- 2001). However, these older antidepressants are used infrequently
ing, whereas more visits by physicians was associated with lower nowadays and so here there will be closer examination of ‘modern’
prescribing, perhaps due to a better understanding of depression compounds that are widely prescribed for older people. SSRIs have
and the benefits from antidepressants in the latter group (Hanlon the largest evidence base and overall are clearly efficacious in MDD,
et al., 2010). Levels of polypharmacy remain very high. In India, although some trials have been negative (Nelson et al., 2008). An
a study of two teaching hospitals reported that over 90% of those important reason for this in one trial in the ‘old-old’ (all subjects
over 60 were prescribed five or more medications and 45% were were over 75) was very large variation in outcome between different
receiving ten or more (Harugeri et al., 2010). A survey of over trial sites, which created so much noise in the outcome data that the
13,000 residents of nursing homes in the US found that over 40% signal of sertraline efficacy could not be detected. This problem is
were being prescribed nine or more drugs (Dwyer et al., 2010). probably widespread in clinical trials and likely had an impact on
Polypharmacy also raises the likelihood of clinically important another negative trial of fluoxetine and escitalopram (Kasper et al.,
additive side effects, whereby common side effects may be toler- 2005), which utilized 73 sites. Another important factor precipi-
ated from one medication but not when several are used together tating negative outcomes is the presence of cognitive impairment.
(Carnahan et al., 2006). For example, when prescribing a specific In a systematic review of randomized controlled trials (RCTs) of
serotonin reuptake inhibitor (SSRI) it is important to check for nontricyclic antidepressants in late-life depression (Nelson et al.,
other medication that can also cause gastric erosions and bleeding, 2008), it is noteworthy that those studies reporting no favourable
especially because treatments such as aspirin, donepezil, NSAIDs, outcome of medication over placebo (including the agents escita-
and steroids with such properties are commonly prescribed in lopram, citalopram, fluoxetine, and venlafaxine) included subjects
older people. If after review it is concluded that all such treatments with a significant amount of cognitive impairment. This is likely to
are needed, then it will be necessary to also prescribe something have contributed to the negative outcomes because (see Depression
such as a proton pump inhibitor for gastroprotection. Finally, con- in the Context of Other Illnesses) neurocognitive impairment is
sideration should be given to frailty and reduction in homeostatic associated with poorer outcomes generally. Indeed, a reanalysis of
reserve. These are two related but poorly defined and yet widely the data from one of these trials identified executive dysfunction
used terms which point to the disproportionate impact side effects as a predictor of poorer response to citalopram, with those with-
can have on older people. Frailty means older people are less out such impairments responding better than placebo (Sneed et al.,
robust and so, for example, anticholinergic effects from paroxetine 2010). It is worth adding the wider context including trials from
(especially if combined with other medication with anticholin- younger adults. A network meta-analysis of 117 RCTs (Cipriani et
ergic effects, such as valproate or carbamazepine or olanzapine), al., 2009) including 25,928 subjects reported sertraline, venlafaxine,
which may pass unnoticed in a healthy younger adult, may induce and escitalopram (along with mirtazapine) to have the most robust
delirium in an older adult with depression. evidence of efficacy of all antidepressants, supporting the view that
the negative trials of these agents in older people are probably due
Acute Phase Drug Treatment of Late-Life to other effects such as the presence of cognitive impairment. For
newer non-SSRIs the evidence base is much smaller in late-life
Major Depression MDD, but it seems reasonable to extrapolate this evidence from
The great majority of pharmacological studies of antidepressants younger adults since the evidence for SSRIs and TCAs has been
have been conducted in people diagnosed as having major depres- shown to reliably apply to older adults too.
sive disorder and so this will be the initial focus here. It is important Which antidepressant should one choose? There are now a large
to remember this because it is not clear if the abundant evidence number of antidepressants to choose from, emanating from sev-
for efficacy of antidepressants in this group can be extrapolated to eral distinct drug classes. In general terms, in making a choice one
the other types of depression occurring in older people. This is, of should consider tolerability, safety, side effects, drug interactions,
course, not an issue unique to late-life depression. In its 2008 guide- contraindications, and, where relevant, previous history of anti-
lines for the drug treatment of depression, the BAP recommended depressant use. Achieving adherence to antidepressants is clearly
a ‘dimensional’ approach to prescribing antidepressants for depres- important, since merely prescribing an antidepressant will achieve
sion (Anderson et al., 2008), and essentially the same approach nothing unless the patient is convinced of the need for taking it
has now been adopted by NICE in their Clinical Guidelines on and does so. Older people frequently do not adhere to medication
Depression Management (NICE, 2009). This means combining evi- in general and this is likely to be especially the case for antidepres-
dence of depression severity with the duration of the illness, rather sants. Hence engagement of the patient during assessment, sensi-
than simply making a diagnosis of depression and prescribing if tive explanation of the diagnosis, and education about the benefits
someone meets this diagnosis. Such a nuanced approach is perhaps and side effects of antidepressants are crucial. Antidepressants, con-
especially important in older patients, where the risk:benefit ratio is fused with benzodiazepines, are often thought to be ‘addictive’ and
less clear due to the increased propensity of older people to develop reassurance that this is not the case may be needed. Dose-related
adverse drug effects. and common side effects should be spelled out and encouragement
given that persistence with the treatment often is associated with findings are consistent with evidence indicating that antidepres-
these waning. In the author’s experience, patients often read and are sants are ineffective in depression in dementia (see Depression in
upset by the much larger number of potential adverse effects listed the Context of Other Illnesses) and with those studies in which
on the insert in the box, and it is helpful to explain that these con- those enrolling subjects with (mild) cognitive impairments showed
sist largely of common everyday symptoms that are not necessarily equivocal benefits for antidepressants. Whilst the evidence may not
related to the drug at all. Education about the speed of response to be robust enough to indicate that antidepressants should not be
antidepressant should also be given, since people may expect the used in such patients, it does suggest that more careful considera-
same kind of rapid response they have experienced with antibiot- tion is needed and that, if used, higher doses, augmentation, and
ics and so become discouraged when this does not happen. At this combination strategies may be needed.
stage, a medication review is worthwhile because simplifying the
medication regime not only is helpful to adherence to antidepres- Acute Phase Psychological Treatments
sants but also can improve adherence more generally and reduce the
number of drugs being taken. Often treatments have been added in of Late-Life Depression
from time to time over the years and, in liaison with the GP, some The quality of the evidence for using psychotherapies in late-life
can often be stopped and a once-daily or twice-daily treatment depression remains poor and the number of well-designed studies
regime can helpfully replace more complex ones. in this field continues to be disappointingly low. There are several
Regarding the specific antidepressant to use, depression guide- particular problems in evaluating the evidence. Much of the evi-
lines indicate the first antidepressant should be an SSRI (Anderson dence that is cited in support of psychotherapy for late-life depres-
et al., 2008; NICE, 2009), unless the patient has a clear history of sion is from nonrandomized studies, and it has been shown that
response to a drug from another class and/or failure to respond to the design of such studies inflates the outcomes in favour of the
or poor tolerability of an SSRI; this recommendation is supported intervention (Sneed et al., 2008b). Thus, for example, in one review
by these trials evidence in older adults and SSRIs are generally bet- (which did not include several important recent antidepressant stud-
ter tolerated than earlier antidepressants, especially tricyclic agents. ies included in the above meta-analysis (Nelson et al., 2008)) it was
SSRIs are generally free from significant interactions with other reported that the effect size in favour of psychotherapy in broadly
drugs, an important consideration in older adults who are usu- defined late-life depression was 1.09 compared with only 0.69 for
ally on other medication, and are particularly appropriate where antidepressants (Pinquart et al., 2006). However, this was really an
tiredness and excessive sleep is a problem because of their alerting apples versus oranges comparison, since randomized blinded anti-
properties. Mirtazapine, reported to be one of the four most effi- depressant trials were compared with unblinded, nonrandomized
cacious antidepressants in depression (Cipriani et al., 2009), has a psychotherapy studies. Its findings also contrast with those of the
side-effect profile that is particularly suited for use in older people. Cochrane review (Wilson et al., 2009) of psychotherapy in late-life
Its main adverse effects are weight gain and sedation, but in older depression which identified only 12 studies of sufficient quality to
people in whom sleep disturbance and weight loss are frequently include, and even then had to include reports comparing psycho-
problematic these can be helpful rather than problematic and the therapeutic interventions with waiting list controls, and most of the
clinical experience of mirtazapine is that it is very well tolerated. studies included people with depression whose baseline depres-
It has the added advantage that its pharmacological profile com- sion rating (using the Hamilton depression rating scale, HAM-D
plements that of SSRIs (and selective serotonin and noradrenaline score) only had to exceed 9 points (a low score, since less than 10 is
reuptake inhibitors (SNRIs)) as it has antagonism of 5-HT3 and often used as a definition of remission in drug studies). The largest
5-HT2 receptors which counteract the potential adverse effects of and best-designed study (Williams et al., 2000) found no benefit
sexual dysfunction and nausea. For these reasons mirtazapine has of problem-solving therapy over placebo (though paroxetine did
become a very widely prescribed antidepressant in late-life depres- have significant benefits) and overall the review concluded that the
sion and is frequently used second-line or in combination with an evidence for using psychotherapy in late-life depression is weak. It
SSRI. SSNRIs (duloxetine and venlafaxine) have similar side-effect was also observed that the patients included in these studies did
profiles to SSRIs and may be especially helpful in those resistant to not appear to be representative of those usually attending health
SSRIs where their dual action can prove helpful, and like SSRIs they services, since most had been recruited through adverts, which
can be combined well with mirtazapine. appears consistent with the low score on the HAM-D for inclusion.
An important caveat to the extrapolation of results from clinical Currently, there is clearly a need for larger, better designed studies
trials in younger adults is that the data may only be valid in older that enroll depressed subjects more typical of those seen in primary
adults who are healthy and cognitively intact, i.e. biologically like and psychiatric services. It can also be argued that in the meantime
younger adults. Physical illness morbidity has been shown to pre- the findings from studies in younger adults with depression can be
dict poor response to antidepressants (Reynolds et al., 2006), and extrapolated to older adults as with drug studies, though the same
cerebral ischaemic white matter disease (demonstrated by the pres- caveats apply about this probably only being valid for healthy, cog-
ence of WHM on MRI) is associated with a poorer response to anti- nitively intact older adults.
depressants (Alexopoulos et al., 2002, 2008), poorer longer-term When psychotherapy is needed there are added difficulties with
outcome, increased relapse rates (Hickie et al., 1996; O’Brien et al., availability and therapist experience. Availability of specific psy-
1998), and persistent cognitive impairment up to 4 years (Kohler chological interventions is still limited, especially in older adults,
et al., 2010b). Since WMH are associated with executive dysfunc- and identifying someone expert in the required therapy, e.g. cogni-
tion, it is not surprising that such impairment is itself associated tive behavioural therapy (CBT), can be especially difficult. When
with poorer response to antidepressants and increased relapse rates prescribing medication, the treatment is the same regardless of
(Kalayam and Alexopoulos, 1999; Alexopoulos et al., 2000). Such the prescriber, but in psychotherapy there is large variation in the
experience and expertise of therapists and this has been reported ECT and Other Interventions
to directly affect the efficacy of the intervention (Williams et al.,
2000). Thus even when psychotherapy is indicated, it can be dif- The other major treatment regularly used in clinical practice for
ficult to ensure that a quality of therapy is provided comparable to late-life depression is electroconvulsive therapy (ECT) (see Chapter
that in the published literature. 14). Its use is reserved for those who remain ill and disabled after
An important issue not covered in the literature is anxiety man- other interventions have been tried and those with life-threatening
agement in the context of depression. So-called comorbid anxi- illness (including high suicide risk). Concerns about using ECT in
ety is common in late-life depression and frequently persists once older people are often expressed, mainly about precipitating confu-
(other) depression symptoms have resolved. Anxiety manage- sion and the alleged increased risks of other physical adverse effects,
ment, involving psychoeducation on anxiety symptoms and tech- but the evidence strongly supports its use. Well-conducted studies
niques such as controlled relaxation, can in such circumstances be (comparing real ECT with sham ECT in which electrodes are placed
helpful and delivered by nurses or occupational therapists from on the head but no electricity is applied) have demonstrated the
local services. efficacy of ECT in depression, with improvements on the HAM-D
of over 5 points greater than antidepressants (UK ECT Review
Group, 2003). The quality and strength of this evidence has made
Multifaceted Interventions the conduct of such studies in older adults with depression unethi-
In contrast to the disappointing quality and quantity of the evi- cal. However, two large studies have directly compared ECT by age.
dence for using specific psychotherapies in late-life depression, The first, examining 268 adults with MDD who all had a HAM-D
there are an increasing number of studies examining a combination of greater than 20, found ECT to be of greater benefit in adults aged
of psychotherapeutic and other interventions to provide structured between 60 and 74 and in those 75 and over compared with younger
packages of care. Various ‘packages’ have been investigated and adults (Tew et al., 1999). The second of 253 adults, again with MDD
almost all of these studies have provided very encouraging findings. and a HAM-D score of over 20, again reported ECT to be nonsig-
One study of home-based treatment for older people with depres- nificantly more beneficial in adults aged 47–64 and in those 65 and
sion, for example, compared PEARLS (combined problem-solving over compared with younger adults, and reported a positive cor-
therapy (PST), increased physical activity, and increased socializa- relation between symptom improvement and age (O’Connor et al.,
tion) with usual care and found that a 50% depression symptom 2001). Importantly, both studies found no age-associated increase
reduction was achieved by 43% of the PEARLS group compared in adverse effects. These findings are very encouraging and support
with 15% with usual care, and remission rates were 36% and 12%, a confident use of ECT in late-life depression, though the patients in
respectively (Ciechanowski et al., 2004). The largest study (involv- such trials were by definition those who were physically fit enough
ing 1802 patients) of such an approach is the IMPACT study, in the first place. A clinical dilemma with ECT is what to do once
which compared usual care with a care management package, with someone is well again. Relapse rates after ECT are worryingly high,
patients being able to select between antidepressants and PST. At 12 with one study reporting almost all patients relapsing unless given
months, 45% of intervention patients had achieved a 50% reduction maintenance drug treatment, and even then relapse rates are high
in symptoms, compared to 19% in the usual care group (Unutzer et (Sackeim et al., 2001a). A large study of over 200 patients investi-
al., 2002). The PROSPECT study, using a collaborative interven- gating maintenance ECT found a reduction in relapse rates com-
tion based in primary care, has reported improved remission rates pared with placebo but not compared with combined maintenance
in late-life major depression at 6, 12, and 24 months and reduced pharmacotherapy (both lithium and nortriptyline) (Kellner et al.,
suicidal ideation (Alexopoulos et al., 2009). A similar but flexible 2006); over 50% of subjects relapsed or withdrew from the study. At
approach was used in a study in Austria comparing home-based the current time, maintenance pharmacotherapy is recommended
treatment by a multidisciplinary team with usual care, and report- following ECT, but the choice of medication is difficult. It seems
ing significant improvements in depressive symptoms, quality of prudent to avoid drugs to which the patient has failed to respond
life, and overall function, and a reduction in admissions to hospi- before ECT, and combination treatment offers the best outcome
tal and nursing homes over 1 year (Klug et al., 2010). These mul- (Sackeim et al., 2001a). The role of maintenance ECT remains con-
tifaceted interventions often include ‘exercise therapy’ of different troversial, but many clinicians, including the author, have known
kinds, and the evidence for benefit from exercise in depression is patients who only remain well with maintenance ECT and in such
similar to that for specific psychotherapies in that most of the stud- highly selected patients its use is justified pragmatically.
ies in the literature are of poor quality; a systematic review found Other stimulation treatments are available for depression, but
only 15 randomized studies, most of which were not blinded and there is little good quality evidence to justify their use and none
had inadequate follow-up (Sjosten and Kivela, 2006). The review specifically in late-life depression. Transcranial magnetic stimula-
concluded that the evidence justifying exercise therapy is modest tion (TMS) has received most attention and has good evidence for
and its role is probably best within such a multifaceted approach. efficacy in depression (Slotema et al., 2010). However, its efficacy
Overall, these multifaceted studies have applied their interventions is significantly less than that of ECT (Slotema et al., 2010) and it
in outpatients (Sirey et al., 2005), community patients (Banerjee et is therefore unclear which patients should receive it. In late-life
al., 1996; Unutzer et al., 2002; Ciechanowski et al., 2004), and even treatment-resistant depression, one study reported benefit from
in residential care (Llewellyn-Jones et al., 1999) and have consist- TMS (Fabre et al., 2004) but another did not (Mosimann et al.,
ently reported benefits. These findings provide robust evidence for 2002). TMS also has limited availability, and whilst it is unclear
enhancing the quality of community-based care by applying such which patients might be selected for its use it is difficult to justify
structured interventions in regular clinical practice, though unfor- providing it in clinical services. Vagal nerve stimulation has some
tunately this is often difficult to achieve. evidence for its benefit in selected patients with treatment resistance
to antidepressants (Sackeim et al., 2001b) but is expensive and very the same conclusion (Whyte et al., 2004). Predictors of a slower
limited in its availability. Similarly, deep brain stimulation (DBS) response are greater severity of depression, comorbid anxiety, and
is only available at a few selected centres. DBS involves surgically late-age at onset of the illness (Alexopoulos et al., 1996; Whyte et
placing an electrode in a selected brain area (such as the nucleus al., 2004). However, these encouraging findings need to be quali-
accumbens or subcallosal cingulate gyrus). The electrode is con- fied by acknowledging that the underlying research was conducted
nected by a wire to a neurostimulator placed subcutaneously in in physically healthy and cognitively intact older adults who were
the chest wall, which is then operated by an external device to set relatively young (almost all under 80). Other data indicate that
the frequency of stimulation. It can be effective in highly selected neurocognitive impairment and physical illness are also associated
treatment-resistant patients (Blomstedt et al., 2011), but there are with poorer response and remission rates (see Prognostic Factors).
currently very few studies and there are ethical concerns about its Thus in the ‘younger-old’ the same approach can be taken as with
use in depression (Synofzik and Schlaepfer, 2011). A similar group younger adults, but for people with these comorbidities a slower
of patients might be considered for psychosurgery, which also has response and lower remission rates are expected. A related issue is
some evidence of benefit, is only available in specialist centres, of possible differences in EOD versus LOD. Here the evidence is
and attracts similar ethical concerns; DBS appears to be replacing mixed, with studies reporting no differences (Reynolds et al., 1998),
psychosurgery in such treatment refractory patients (Sachdev and LOD to have a worse prognosis (Alexopoulos et al., 1993b), and
Chen, 2009). EOD to have poorer prognosis (Brodaty et al., 1993). Whilst dif-
ferences in the groups studied are likely to explain such discrepant
findings, it makes it difficult to draw firm conclusions. The wider
Achieving Remission in Late-Life Depression evidence for increases in WMH on MRI in LOD (Herrmann et al.,
The clinician managing someone with depression should be aiming 2008) and the robust evidence associating such lesions with poorer
to achieve remission, rather than simply symptomatic improvement, outcomes (O’Brien et al., 1998; Alexopoulos, 2002; Alexopoulos et
and so should persist and utilize a range of strategies to achieve al., 2008; Kohler et al., 2010b) leads the author to conclude that
this and restore people to normal functioning and a symptom-free LOD depression responds more slowly, and this is corroborated by
state. Persistence in applying a logical and structured approach clinical experience.
results in over 80% of patients achieving remission (Rush, 2007). How long should you wait before adjusting antidepressant treat-
As discussed, the management strategy used should be tailored ment? In ‘younger-old’ people with late-life depression, if there is
to the severity level of depression and stepped up if remission is no apparent change at all by 2 weeks then, assuming no adverse
not achieved at the starting level. But once several treatments have effects, the dose should be increased to the maximum tolerated
been utilized at different levels in the stepped care process, then a dose. In those with LOD and/or with cognitive impairment and/
structured and persistent approach is needed to combat the dan- or physical morbidity, it is prudent to wait until perhaps 4 weeks.
ger of fatalism and despair developing, and such a strategy largely If there is symptom improvement, then treatment should be con-
involves pharmacotherapy. tinued and the dose increased if further improvement ceases. An
First and most obviously, the antidepressant can be increased important factor is duration of treatment, since it has been shown
in dose or prolonged in its use. When dealing with older patients, that the longer a trial lasts, the greater the difference between drug
there remains a tendency to use lower doses and increase them cau- and placebo; that is, when a trial is conducted over a longer period,
tiously (applying the old age psychiatrist’s adage of ‘start low, go a larger proportion of people achieve response and, more impor-
slow’). Whilst appropriate with the frail older person, this is prob- tantly, remission on drugs, creating a larger difference from pla-
ably too cautious an approach for adults who are chronologically cebo and clearer evidence of benefit. Thus in a systematic analysis
‘old’ but biologically ‘young’. For example, in the open-label acute of this, it was reported that in clinical trials of 6–8 weeks’ dura-
treatment phase of a maintenance study in late-life depression using tion the odds ratio in favour of antidepressants was 1.2, whereas in
citalopram, it was reported that 50% of subjects tolerated 40 mg those studies with outcomes at 10–12 weeks after baseline the odds
and many were given 60 mg (Klysner et al., 2002). An additional ratio was substantially increased to 1.73 (Nelson et al., 2008). Such
caveat to this approach is to consider a patient’s previous history evidence supports persistence with a treatment rather than quickly
and especially the risk of relapse, so that earlier and more robust switching to an alternative because improvement seems to be too
treatment should be used in those with depressive episodes earlier slow. In practice, a treating doctor can come under both internal (‘I
in life whose symptoms are now returning. need to do something’) and external pressure to change treatment
How quickly should an older person with depression respond to and is then in danger of losing gains by switching to an alternative
treatment? Whilst it is often said that it takes 4–6 weeks for antide- or unnecessarily adding in a new treatment through impatience,
pressants to work, it is important to be clear what this means and thereby exposing the patient to a higher risk of adverse effects.
to distinguish between response and remission. Thus, in general, Patient explanation and support, and utilization of other aspects of
it is expected that some improvement in symptoms (i.e. there will treatment, e.g. group activity or PST, are perhaps better responses
be evidence of response to treatment) will occur within a couple of in such circumstances. Switching to an alternative antidepressant
weeks and such early response is predictive of stable later response should be carried out as soon as it is clear no benefit at all is being
(van Calker et al., 2009); the disappearance of all or nearly all symp- achieved (if some benefit is apparent then augmentation or combi-
toms (remission) will take some weeks longer. But are these time- nation approaches are indicated) or more commonly when adverse
lines also applicable to late-life depression? Two studies directly effects preclude its continuation.
comparing response and remission in older and middle-aged If the strategies of increasing the dose and prolonging the dura-
depressed adults reported no differences (Alexopoulos et al., tion of treatment when progress is being made are not successful
1996; Reynolds et al., 1996) and a review of the literature reached and remission is not achieved on the maximum tolerated dose,
combination and augmentation strategies should be considered. employed earlier due to severity and persistence of depression.
Unfortunately, as with switching and increasing the dose, the evi- After soliciting a second opinion, the more exotic stimulation and
dence for such strategies is very limited, with only lithium augmen- surgical treatment options might be investigated.
tation having replicated evidence of benefit in late-life depression
(Cooper et al., 2011). A combination approach (adding a second
antidepressant) is very popular but has little empirical evidence to Continuation and Maintenance Treatment
support it. Clinical experience supports it and a popular combi- Once remission is achieved, the clinical issue becomes keeping the
nation involves adding mirtazapine to an SSRI, since the latter is patient well. It is clear that in a first episode of depression, the treat-
in most cases the initial antidepressant. For the pharmacological ment that got someone well should be continued (Geddes et al.,
reasons discussed earlier, this is a rational strategy and experience 2003). Since the risk of relapse is highest in the first 6 months, this
suggests it is frequently helpful. Again, the mirtazapine, or other continuation treatment should be for at least 6–9 months for a first
combination agent, should be increased to the maximum tolerated uncomplicated episode from when the person achieves remission,
dose and other treatment aspects (group CBT, anxiety manage- and for at least a year in other situations (Anderson et al., 2008). In
ment, etc.) should be explored during this treatment phase. The younger adults, maintenance treatment (i.e. continuation of anti-
less-popular alternative is to augment the first-line antidepressant depressant treatment beyond this continuation phase) is not rec-
with a nonantidepressant. Historically, lithium has been the most ommended after a single episode of depression, but the situation is
frequently used augmentation agent and there is some evidence different for older adults. Full-dose maintenance treatment is effec-
to support this, though mainly from smaller, lower quality studies tive, with evidence supporting its use for up to 3 years. Several of
(Crossley et al., 2007). However, lithium has well-recognized adverse the key studies have specifically been conducted in late-life depres-
effects and it is not clear what serum level of lithium is appropri- sion and a meta-analysis of eight double-blind trials reported the
ate for augmentation. Generally, in older adults one should aim for number needed to treat to prevent one additional relapse or recur-
lower lithium doses and lower serum levels of around 0.4 mmol/L, rence to be only three (Kok et al., 2011). These maintenance stud-
but this is mainly driven by a desire to avoid or minimize adverse ies highlighted an important issue familiar to those working with
effects rather than by empirical investigation. Older patients appear older people with depression, which is that even in those with a first
especially susceptible to lithium side effects, with one study report- episode of depression in late life (late-onset depression), the risk of
ing polydipsia (50–74%), polyuria (25–58%), tremor (33–58%), relapse after treatment is stopped is very high, and when one cohort
and dry mouth (53%) as the most frequent adverse effects, but a of such LOD patients was followed-up when treatment was stopped
worryingly large percentage (11–23%) developed lithium toxic- after 2 years, over 60% had a further episode in the following 2 years
ity (Foster, 1992). In the author’s experience, tremor can be par- (Flint and Rifat, 1999). When considering maintenance treatment,
ticularly problematic and disabling, and although dosage reduction the impact of another depressive episode should be weighed and
often helps, in some cases it prohibits continued use of lithium. in older people with relatively few remaining years the impact is
Lithium augmentation has become less popular, probably because disproportionately high. The consideration in younger adults about
of the extra effort involved in monitoring and these concerns about exposing someone for many years to unnecessary antidepressant
toxicity, and because of the development of additional classes of treatment does not apply. Thus, in general, even when patients
antidepressants encouraging combination strategies ahead of aug- have a first episode of depression in late life (i.e. they have LOD)
mentation. One well-designed study has examined ‘augmentation’ they should be put on maintenance treatment, and by extension so
in late-life depression, though since the ‘augmentation’ agents used should all people with late-life depression. The important caveat to
included nortriptyline and bupropion (as well as lithium) it really this is that this evidence again comes from studies of major depres-
assessed both combination and augmentation. These second-line sive disorder, and for those with less severe depression who have
treatments were added to paroxetine (combined with interpersonal received antidepressants, such long-term use might not be war-
therapy (IPT)) and encouragingly an extra 20% of patients (over ranted. When planning continuation and especially maintenance
half those receiving ‘augmentation’) then achieved recovery (Dew treatment, education of the patient and significant others about
et al., 2007). The cholinesterase inhibitor donepezil has also been the patient’s illness and the risk of future relapse is crucial. Many
studied as an augmentation strategy in late-life depression, though older patients are reluctant to take medication, especially antide-
only as an add-on in the maintenance phase. It was found that whilst pressants, and to keep taking them when they feel well is especially
donepezil produced some modest and transient improvements in difficult. Thus, discussing the impact of the illness and the benefits
cognition, its use was associated with a significantly higher relapse of antidepressants, and explaining the high risk of relapse and how
rate (Reynolds et al., 2011). this risk can be substantially reduced by taking antidepressants
Beyond augmentation and combination with single agents there long-term, are absolutely vital to engage patients in managing their
are options of adding in other augmentation and antidepressant illness and so maximize their likelihood of long-term adherence to
agents. Evidence from the STAR-D study suggests, as might be their treatment.
expected, that a law of diminishing returns applies as more treat- Most of the literature has examined maintenance pharmaco-
ments are added, but that remission can be achieved in about 80% therapy, but two major studies have examined IPT for maintenance
of cases (Menza, 2006). However, this does leave 20% of patients, in late-life depression, alone and in combination with maintenance
a considerable proportion, who remain unwell, and in such drug treatment. Monthly IPT was reported in the first study to
treatment-resistant subjects it is prudent to revisit the diagnosis, reduce relapse rates compared with routine follow–up, and com-
considering, for example, the extent to which persistent symptoms bined IPT and nortriptyline was even more effective (Reynolds et
may be related to a physical illness or a personality disorder. At this al., 1999). However, the second study, involving the same design
point, ECT should also be considered, if it has not already been but substituting paroxetine for nortriptyline, found no benefit of
IPT alone or in combination in the maintenance phase (Reynolds of psychotherapy in poststroke depression found no evidence of
et al., 2006). It is worth noting, however, that this second study was benefit on any outcomes (Hackett et al., 2008). SSRIs are the most
on patients who on average were about 10 years older and included appropriate antidepressant to use in poststroke depression, not only
a significant proportion with late-onset of depression, which may because the best evidence from trials is for this drug class but also
have reduced the potential benefit from IPT. because they have been demonstrated to be safe to use in people
with coronary heart disease and by extension vascular disease more
broadly (Glassman et al., 2002). As with the management of major
Depression in the Context of Other Illnesses depressive disorder, the key issue is when to initiate such treatment,
Depression is highly prevalent in association with physical illness, given the need to balance exposure to predictable adverse drug
in general, and with specific (brain) disorders, e.g. Parkinson’s dis- effects with potential benefits on mood. Again, when depressive
ease, in particular. The difficulty of diagnosing depression as a sepa- symptoms are not severe or persistent it seems prudent to with-
rate entity when other illnesses are present was discussed earlier. hold medication, as the risk:benefit ratio is likely to be unfavour-
Antidepressants are frequently prescribed in such circumstances able and because spontaneous remission is common in this group.
in spite of the diagnostic difficulties, but evidence supporting their Conversely, if a poststroke depression is severe or persistent, then
use in such situations is limited. An analysis of one large mainte- SSRIs should be initiated, closely monitored, and titrated to achieve
nance trial in outpatients reported that a high burden of physical maximum effect. TMS has also been reported in one study to be
illness predicted a higher depression recurrence rate (Reynolds et significantly better than sham TMS in poststroke depression (Jorge
al., 2006) and this also applied to augmentation (Dew et al., 2007). et al., 2004), but although side effects are generally mild there is an
A systematic review found the prognosis for depressed inpatients increased risk of seizures and it has not been compared with anti-
to be very poor, with only about a fifth alive and well a year after depressants. If available locally, TMS may therefore be an option
the episode (Cole and Bellavance, 1997a). These data are difficult to consider in patients unable to tolerate or unwilling to take anti-
to interpret, since half the subjects had died, presumably mainly depressants for poststroke depression. Poststroke emotionalism is
from the illness for which they were hospitalized, but it does sug- often confused with depression. Here people with stroke disease
gest caution before prescribing antidepressants in such patients and have outbursts of tearfulness, or more rarely euphoria and laughter,
that a thorough psychiatric assessment to distinguish physical from which are triggered by insignificant emotional stimuli or no appar-
mental symptoms is appropriate to focus prescribing on those most ent stimuli. Unlike depression, the mood changes are rapid and
likely to respond. typically not associated with the appropriate subjective emotional
Depression in dementia is frequent and well recognized as a experience, e.g. they do not feel sad whilst crying. Although not
major issue and different classes of antidepressants have been tried well characterized or recognized in psychiatric classifications, it is a
in several clinical trials. However, until recently, these were often of frequently encountered clinical problem and antidepressants have
poor quality and were conducted in small numbers of subjects. Thus been studied in small trials. Overall, all classes of antidepressants
whilst the Cochrane review by Bains et al. (2002) found no evidence seem to produce significant benefits, which typically occur more
of benefit from using antidepressants, this conclusion was recog- rapidly than with response in depressive disorders (Hackett et al.,
nized to be subject to a type 2 error. More recently, two large studies 2010).
have been reported. DIADS-2, a multicentre US randomized trial There have also been few studies investigating the benefits of
of sertraline 100 mg versus placebo in depression in Alzheimer’s treatment of depression occurring in Parkinson’s disease. As with
disease, found no evidence of improvement in symptoms, response early studies of depression in dementia, the few studies have been
rates, or remission rates on sertraline at either 12 weeks (Rosenberg small and produced equivocal results. The Cochrane review by
et al., 2010) or 24 weeks (Weintraub et al., 2010). Moreover, those Ghazi-Noori et al. (2009) reported no evidence of benefit from CBT
on sertraline suffered a higher frequency of adverse effects. The UK or ECT and equivocal evidence for antidepressants. However, this
Study of Antidepressant Drugs and Depressive Symptoms (SADD) was based on three small antidepressant trials and, since this, two
was a 9-month randomized study at nine centres across England larger studies have been reported. One study in 52 subjects found
which compared sertraline, mirtazapine, and placebo for depres- that nortriptyline produced larger improvements in depression
sion in dementia. It also reported no evidence of benefit from symptoms than either paroxetine or placebo over 8 weeks (Menza
either antidepressant but an increase in adverse effects on medica- et al., 2009). The other study in 287 people found that pramipex-
tion (Banerjee et al., 2011). These findings strongly caution against ole produced a statistically greater, but clinically unimportant (less
the widespread use of antidepressants for depression occurring than 2-point difference), improvement on the Beck Depression
in people with dementia. However, these studies necessarily did Inventory at 12 weeks than placebo. But there were no differences
not include people with more severe depression, including those on the Geriatric Depression Scale (GDS), and of note is that depres-
at significant suicide risk, for whom antidepressants might still be sion was defined as having a low GDS score of only greater than 5
appropriate; this is especially the case if there is a previous history (Bxarone et al., 2010). Currently, it is not clear whether antidepres-
of major depression. sants or anti-Parkinson’s treatments produce significant clinical
Studies of antidepressants in stroke disease have reported evi- benefits for people with depression in Parkinson’s disease.
dence of efficacy, but this is offset by an increase in adverse effects
in those on antidepressants (Hackett et al., 2008). Thus the clini-
cian needs to carefully weigh up the potential benefits, determined Prevention of Late-Life Depression
by the severity of the current depressive episode and the history Since depression is so highly prevalent, there has been increasing
of previous episodes of depression and their response to treat- interest in preventing the development of depression, including
ment, with these risks of adverse events. Three randomized trials late-life depression, in the first place (Cuijpers et al., 2008; Beekman
et al., 2010). Unsurprisingly, strategies aimed at the general popula- depression (Cole and Bellavance, 1997a). At 3 months only 18%
tion (universal prevention) do not appear effective (Beekman et al., were rated as well, 43% were still depressed, and 22% were dead;
2010), but those targeting older people with subthreshold depres- by 12 months only 19% were well, 29% still depressed, and 53%
sive symptoms do reduce the development of depressive and anxi- had died. However, there are clearly difficulties in interpreting such
ety episodes at both 1 year (van’t Veer-Tazelaar et al., 2009) and 2 data and distinguishing physical illness causes from any additional
years (van’t Veer-Tazelaar et al., 2011). Targeting high-risk individ- depression effect. A detailed analysis controlling for physical illness,
uals (those living alone, having two or more chronic illnesses, sig- comorbidity, and previous history of depression found no increase
nificant anxiety symptoms, lack of mastery over their environment) in mortality for those diagnosed as depressed during their hospi-
also appears to reduce the development of depressive syndromes tal admission; this study also found that after taking these factors
(Smits et al., 2008). With both ‘high-risk’ and ‘subthreshold’ groups, into account, pre-existing depression was associated with a reduced
depression was reduced in primary care settings by ‘watchful wait- mortality (McCusker et al., 2006). For psychiatrists there is some
ing’, followed by bibliotherapy, PST, and if necessary medication. encouragement from a third analysis by these authors reporting
better outcomes in people with late-life depression who were seen
Prognosis in Late-Life Depression by hospital psychiatric services (Cole and Bellavance, 1997b). This
analysis of 1487 patients in 16 studies followed for at least a year
Outcome findings from research are based on aggregated data from found that 60% were either well or had relapses with good recovery,
follow-up studies on cohorts of patients, whereas the clinician is 14–22% were continuously ill, and the remainder again had poor
usually asked to predict the prognosis for a specific patient with outcomes such as death (ranging from 3–19% dead by 12 months)
his or her unique combination of individual characteristics. Such or developing dementia. Again, although many of the studies had
a transformation requires combining knowledge of the relevant weaknesses, the overall findings are similar to those in the larger
research findings with familiarity with the specific patient’s features, and better executed studies in the analysis (Baldwin and Jolley, 1986;
enhanced by wisdom acquired from clinical experience; it is an area Burvill et al., 1991; Baldwin et al., 1993). Whilst the mortality in
where such medicine-based evidence is especially required to mod- medical inpatients is unsurprising (and is not clearly directly related
ify the research-driven evidence. In theory, outcome data would to depression in any case), the poor outcome in community psychi-
come from both naturalistic studies of treatment-free patients and atric patients compared with psychiatric secondary care patients is
from treatment studies prospectively examining outcomes. In prac- counterintuitive, as the former are expected to have milder forms of
tice, there are very little purely naturalistic data available, and what depression and thus better outcomes than patients referred to psy-
was reported (e.g. Post, 1972) emanated from specialist inpatient chiatric services. This may reflect underdetection and undertreat-
units, making it difficult to compare with later data derived from ment in the community, with reports indicating only 20% (Cole et
very different clinical settings. al., 1999) and 10% (Sharma et al., 1998) of depressed older people in
the community receiving adequate treatment. Subsyndromal forms
Outcome of Depression of depression have a high conversion rate to major depression in
Reports from studies examining outcome comment on the high older people, about 8% per year, and whilst having better overall
mortality in patients with late-life depression and the generally poor outcomes than major depression such conditions still remit by 1
prognosis. For example, a meta-analysis of community depression year in only about 27% of cases (Meeks et al., 2011).
in older people (mean age over 60) involving a total of 1268 patients That depression in older people is associated with an increased
found that at 2 years after the index episode of depression, only 33% mortality has been long recognized, e.g. a three-fold increase
were well, 33% were still depressed, and 21% were dead. The rest (Murphy et al., 1988), but is depression itself the cause of this?
were classified as ‘other’ and appeared to represent a poor outcome Several studies have examined this question by following depressed
group, including people who developed dementia and had only older patients over several years and controlling for other risk fac-
partial remissions (Cole et al., 1999). Although most of the studies tors. A Finnish study examined 813 subjects over 5 years and found
had weaknesses, the two largest and best executed studies reported that 48% of those depressed at the beginning and end of the study
similar findings (Kennedy et al., 1991; Copeland et al., 1992). A had died compared with only 26% who were never depressed, but
more detailed prospective analysis of community depression over 6 those who were only depressed at baseline did not differ signifi-
years and including 14 assessments had more encouraging, though cantly (31% mortality) (Pulska et al., 1999). A 4-year study from
still poor, findings (Beekman et al., 2002), with 23% of patients the Netherlands found that those with major depression had a
having persistent remission, 12% remission with recurrence, 33% significant 1.83-fold increase in mortality, whilst in those with
a chronic-intermittent course, and 33% chronic depression. They minor depression only men had a significant increase (1.80-fold)
also reported that outcome was related to baseline diagnosis, with (Penninx et al., 1999). The same group reported similar findings
subthreshold depression having the best outcome, major depres- in another cohort of 4051 community-dwelling older people who
sion and dysthymia intermediate outcomes, and ‘double depres- were followed over 6 years (Schoevers et al., 2000). Survival analysis
sion’ (major depression on top of dysthymia) the poorest outcome. showed that only men with ‘neurotic’ depression had a significant
Older primary care patients with depression were found to have increase in mortality of 1.67-fold, whilst both men and women
episodes lasting a median of 18 months, and after 1 year of illness with ‘psychotic’ depression (approximately major depression) had
only 35% were well, by 2 years 60%, and by 3 years 68%. Physical ill significant increases in mortality of about two-fold. A US study
health and depression severity at baseline were predictors of poor following older people with broadly defined depression (20% had
outcome (Licht-Strunk et al., 2009). Bleak conclusions were also depression) over 6 years found a 24% increase in all-cause mortal-
reached in another meta-analysis of older medical inpatients with ity in this depressed group, after controlling extensively for other
risk factors for mortality (Schulz et al., 2000). A 10-year prospective
study in the Netherlands of 3746 community-dwelling subjects reported in one study to predict poorer outcome (Murphy, 1983),
with depression confirmed the increased mortality and reported a but other studies have not replicated this (e.g. Baldwin and Jolley,
dose-response relationship whereby more severe and more chronic 1986; Burvill et al., 1991). Adverse life events after the onset of
depression was positively correlated with this increased risk depression were reported in one study (Murphy, 1983) as predict-
(Schoevers et al., 2009). ing poor outcome, but not in another study assessing this (Burvill
Such studies uniformly confirm the high mortality associated et al., 1991). Again, Murphy (1983) found that the occurrence of
with late-life depression, whether major or minor depression, sug- adverse life events following the onset of depression had a particu-
gest men may have a higher risk, and this increase seems to be due larly adverse effect on outcome, but Burvill et al. (1991) found no
to depression itself rather than confounding factors associated with such relationship.
depression. The strong relationship of depression to cardiovascular
and cerebrovascular disease is likely to be an important factor in Conclusion
this relationship.
Depression in late life remains common and has a similar symptom
These findings would suggest that depression in older people has
profile to that in younger adults, but older people tend to have more
a poorer response to treatment than depression in younger adults
psychosis and cognitive impairments. The relative contributions of
and makes the findings from a review of this issue somewhat sur-
the proposed aetiologies need to be more precisely defined. Although
prising. Reviewing 24 publications between 1996 and 2004, it was
factors such as adverse life events and social isolation remain
reported that depression remitted as frequently in later life as in
important, physical illness, vascular brain disease, and other sub-
earlier life but was associated with a higher relapse rate (Mitchell
tle organic factors become important too, especially for those with
and Subramaniam, 2005). This increased risk of relapse was linked
LOD. Antidepressants are as effective in the ‘younger-old’ (those
to age at onset of depression, with EOD having a poorer outcome
without clinically significant cognitive impairments or physical ill-
because of multiple episodes in earlier life, and medical comorbid-
ness) as in younger adults. but probably less so for those afflicted by
ity, which worsened prognosis and was linked to a later onset.
such comorbidities. ECT remains an effective and safe treatment in
older people. The evidence supporting benefits from specific psy-
Prognostic Factors chotherapies is limited, though for the ‘younger-old’ they probably
remain as useful as in younger adults, and multifaceted interven-
There are several poor prognostic indicators recognized in late-life
tions involving care management applications of several interven-
depression, which are summarized in Box 42.5. Unsurprisingly,
tions appear highly effective. Continuation and maintenance use of
physical illness burden (Murphy, 1983; Baldwin and Jolley, 1986;
antidepressants is effective in reducing relapse and should probably
Burvill et al., 1991; Reynolds et al., 2006; Licht-Strunk et al., 2009;
be used in all late-life major depression. Longer-term outcome is
Azar et al., 2011), depression severity (, Reynolds et al., 2006; Licht-
poor in naturalistic studies but much improved in treated patients.
Strunk et al., 2009; Azar et al., 2011), comorbid anxiety (Reynolds
A high (two- to three-fold increased) mortality is evident and
et al., 2006; Azar et al., 2011), and cognitive impairment (Cole and
depression is an important risk factor for subsequent dementia. All
Bellavance, 1997b; Nelson et al., 2008) have been associated with a
this highlights the need for careful detection and thorough assess-
worse outcome in late-life depression, and clinically significant cog-
ment of depression in later life, combined with vigorous treatment
nitive impairment is strongly associated with an increased risk over
and regular follow-up to monitor maintenance treatment.
the following few years of dementia (Alexopoulos et al., 1993a). The
only other consistent predictors of outcome have been discussed
earlier, namely ischaemic WMH and executive dysfunction. WMH References
and related white matter changes leading to increased diffusion are Alexopoulos, G.S. (2002). Frontostriatal and limbic dysfunction in late-life
associated with a poorer response to antidepressants (Alexopoulos depression. American Journal of Geriatric Psychiatry, 10, 687–95.
et al., 2002, 2008), poorer longer-term outcome, increased relapse Alexopoulos, G.S. (2005). Depression in the elderly. Lancet, 365, 1961–70.
rates (Hickie et al., 1996; O’Brien et al., 1998), and persistent cogni- Alexopoulos, G.S., et al. (1984). Dexamethasone suppression test in geriatric
tive impairment up to 4 years (Kohler et al., 2010b). Executive dys- depression. Biological Psychiatry, 19, 1567–71.
Alexopoulos, G.S., et al. (1993a). The course of geriatric depression with
function predicts poorer response to antidepressants and increased
‘reversible dementia’: a controlled study. American Journal of Psychiatry,
relapse rates (Kalayam and Alexopoulos, 1999; Alexopoulos et al., 150, 1693–9.
2000; Sneed et al., 2010). Alexopoulos, G.S., Young, R.C., and Meyers, B.S. (1993b). Geriatric
Other factors examined do not seem to predict outcome. Once depression: age of onset and dementia. Biological Psychiatry, 34,
confounding factors have been accounted for, there appear to be no 141–5.
consistent differences due to sex or age. Depression duration was Alexopoulos, G.S., et al. (1996). Recovery in geriatric depression. Archives of
General Psychiatry, 53, 305–12.
Alexopoulos, G.S., et al. (1997). ‘Vascular depression’ hypothesis. Archives of
Box 42.5 Poor outcome predictors General Psychiatry, 54, 915–22.
Alexopoulos, G.S., et al. (2000). Executive dysfunction and long-term outcomes
Physical ill health of geriatric depression. Archives of General Psychiatry, 57, 285–90.
Depression severity Alexopoulos, G.S., et al. (2002). Frontal white matter microstructure and
Chronicity of depression treatment response of late-life depression: a preliminary study. American
Presence of white matter hyperintensities Journal of Psychiatry, 159, 1929–32.
Alexopoulos, G.S., et al. (2008). Microstructural white matter abnormalities
Cognitive impairment
and remission of geriatric depression. American Journal of Psychiatry,
(especially executive dysfunction) 165, 238–44.
Alexopoulos, G.S., et al. (2009). Reducing suicidal ideation and depression Baxter, L.R., Jr, et al. (1989). Reduction of prefrontal cortex glucose
in older primary care patients: 24-month outcomes of the PROSPECT metabolism common to three types of depression. Archives of General
study. American Journal of Psychiatry, 166, 882–90. Psychiatry, 46, 243–50.
Almeida, O.P., et al. (2003). Depression with late onset is associated with Bebbington, P.E., et al. (1998). The influence of age and sex on the prevalence
right frontal lobe atrophy. Psychological Medicine, 33, 675–81. of depressive conditions: report from the National Survey of Psychiatric
Alvarez, P., et al. (2011). Early- and late-onset depression in the older: no Morbidity. Psychological Medicine, 28, 9–19.
differences found within the melancholic subtype. International Journal Beekman, A.T., Copeland, J.R., and Prince, M.J. (1999). Review of community
of Geriatric Psychiatry, 26, 615–21. prevalence of depression in later life. British Journal of Psychiatry, 174,
Anderson, I.M., et al. (2008). Evidence-based guidelines for treating depressive 307–11.
disorders with antidepressants: a revision of the 2000 British Association Beekman, A.T., et al. (2002). The natural history of late-life depression:
for Psychopharmacology guidelines. Journal of Psychopharmacology, 22, a 6-year prospective study in the community. Archives of General
343–96. Psychiatry, 59, 605–11.
Andreescu, C., et al. (2008). Gray matter changes in late life depression—a Beekman, A.T., et al. (2010). Preventing depression in high-risk groups.
structural MRI analysis. Neuropsychopharmacology, 33, 2566–72. Current Opinion in Psychiatry, 23, 8–11.
Anisman, H., et al. (1999). Endocrine and cytokine correlates of major Bell-McGinty, S., et al. (2002). Brain morphometric abnormalities in geriatric
depression and dysthymia with typical or atypical features. Molecular depression: long-term neurobiological effects of illness duration.
Psychiatry, 4, 182–8. American Journal of Psychiatry, 159, 1424–7.
Arean, P.A. and Reynolds, C.F., 3rd (2005). The impact of psychosocial factors Bench, C.J., et al. (1992). The anatomy of melancholia-focal abnormalities
on late-life depression. Biological Psychiatry, 58, 277–82. of cerebral blood flow in major depression. Psychological Medicine, 22,
Austin, M.P., et al. (1999). Cognitive function in depression: a distinct 607–15.
pattern of frontal impairment in melancholia? Psychological Medicine, Bhalla, R.K., et al. (2006). Persistence of neuropsychologic deficits in the
29, 73–85. remitted state of late-life depression. American Journal of Geriatric
Austin, M.P., Mitchell, P., and Goodwin, G.M. (2001). Cognitive deficits in Psychiatry, 14, 419–27.
depression: possible implications for functional neuropathology. British Blazer, D. (1999). EURODEP Consortium and late-life depression. British
Journal of Psychiatry, 178, 200–6. Journal of Psychiatry, 174, 284–5.
Azar, A.R., et al. (2011). Remission in major depression: results from a Blazer, D.G. (2003). Depression in late life: review and commentary. Journals
geriatric primary care population. International Journal of Geriatric of Gerontology Series A-Biological Sciences and Medical Sciences, 58,
Psychiatry, 26, 48–55. 249–65.
Bains, J., Birks, J.S., and Dening, T.D. (2002). Antidepressants for treating Blazer, D.G. and Hybels, C.F. (2005). Origins of depression in later life.
depression in dementia. Cochrane Database of Systematic Reviews, 4, Psychological Medicine, 35, 1241–52.
CD003944. Blomstedt, P., et al. (2011). Deep brain stimulation in the treatment of
Baldwin, R.C. (1994). Is there a distinct type of major depression in the depression. Acta Psychiatrica Scandinavica, 123, 4–11.
elderly? Journal of Psychopharmacology, 8, 177–184. Bosworth, H.B., et al. (2003). The impact of religious practice and religious
Baldwin, R.C. (2005). Is vascular depression a distinct sub-type of depressive coping on geriatric depression. International Journal of Geriatric
disorder? A review of causal evidence. International Journal of Geriatric Psychiatry, 18, 905–14.
Psychiatry, 20, 1–11. Bowley, M.P., et al. (2002). Low glial numbers in the amygdala in major
Baldwin, R.C. and Jolley, D.J. (1986). The prognosis of depression in old age. depressive disorder. Biological Psychiatry, 52, 404–412.
British Journal of Psychiatry, 149, 574–83. Braam, A.W., et al. (2001). Religion as a cross-cultural determinant
Baldwin, R.C. and Tomenson, B. (1995). Depression in later life. A comparison of depression in elderly Europeans: results from the EURODEP
of symptoms and risk factors in early and late onset cases. British Journal collaboration. Psychological Medicine, 31, 803–14.
of Psychiatry, 167, 649–52. Bremner, J.D., et al. (2002). Reduced volume of orbitofrontal cortex in major
Baldwin, R.C., et al. (1993). Depression in old age. A reconsideration of depression. Biological Psychiatry, 51, 273–9.
cerebral disease in relation to outcome. British Journal of Psychiatry, 163, Brodaty, H., et al. (1991). Age and depression. Journal of Affective Disorders,
82–90. 23, 137–49.
Ballard, C., et al. (1996a). The prevalence, associations and symptoms of Brodaty, H., et al. (1993). Prognosis of depression in the elderly. A comparison
depression amongst dementia sufferers. Journal of Affective Disorders, with younger patients. British Journal of Psychiatry, 163, 589–96.
36, 135–44. Brodaty, H., et al. (1997). Increased rate of psychosis and psychomotor change
Ballard, C.G., et al. (1996b). A follow up study of depression in the carers of in depression with age. Psychological Medicine, 27, 1205–13.
dementia sufferers. British Medical Journal, 312, 947. Brodaty, H., et al. (2001). Early and late onset depression in old age: different
Ballard, C., et al. (1999). Psychiatric morbidity in dementia with Lewy bodies: aetiologies, same phenomenology. Journal of Affective Disorders, 66,
a prospective clinical and neuropathological comparative study with 225–36.
Alzheimer’s disease. American Journal of Psychiatry, 156, 1039–45. Burvill, P.W., et al. (1991). The prognosis of depression in old age. British
Ballmaier, M., et al. (2004). Anterior cingulate, gyrus rectus, and Journal of Psychiatry, 158, 64–71.
orbitofrontal abnormalities in elderly depressed patients: an Butters, M.A., et al. (2004). The nature and determinants of neuropsychological
MRI-based parcellation of the prefrontal cortex. American Journal of functioning in late-life depression. Archives of General Psychiatry, 61,
Psychiatry, 161, 99–108. 587–95.
Banerjee, S., et al. (1996). Randomised controlled trial of effect of intervention Butters, M.A., et al. (2009). Three-dimensional surface mapping of the
by psychogeriatric team on depression in frail elderly people at home. caudate nucleus in late-life depression. American Journal of Geriatric
British Medical Journal., 313, 1058–61. Psychiatry, 17, 4–12.
Banerjee, S., et al. (2011). Sertraline or mirtazapine for depression in Bxarone, P., et al. (2010). Pramipexole for the treatment of depressive
dementia (HTA-SADD): a randomised, multicentre, double-blind, symptoms in patients with Parkinson’s disease: a randomised,
placebo-controlled trial. Lancet, 378, 403–11. double-blind, placebo-controlled trial. Lancet Neurology, 9, 573–80.
Baune, B.T., et al. (2006). The association between depressive mood and Carnahan, R.M., et al. (2006). The Anticholinergic Drug Scale as a measure
cognitive performance in an elderly general population—the MEMO of drug-related anticholinergic burden: associations with serum
Study. Dementia and Geriatric Cognitive Disorders, 22, 142–9. anticholinergic activity. Journal of Clinical Pharmacology, 46, 1481–6.
Chang, C.-C., et al. (2011). Reduction of dorsolateral prefrontal cortex gray Dunn, N.R., Freemantle, S.N., and Mann, R.D. (1999). Cohort study
matter in late-life depression. Psychiatry Research, 193, 1–6. on calcium channel blockers, other cardiovascular agents, and the
Chrischilles, E.A., et al. (2009). Inappropriate medication use as a risk factor prevalence of depression. British Journal of Clinical Pharmacology, 48,
for self-reported adverse drug effects in older adults. Journal of the 230–3.
American Geriatrics Society, 57, 1000–6. Dwyer, L.L., et al. (2010). Polypharmacy in nursing home residents in
Ciechanowski, P., et al. (2004). Community-integrated home-based the United States: results of the 2004 National Nursing Home Survey.
depression treatment in older adults: a randomized controlled trial. American Journal of Geriatric Pharmacotherapy, 8, 63–72.
Journal of the American Medical Association, 291, 1569–77. Emmerson, J.P., et al. (1989). Life events, life difficulties and confiding
Cipriani, A., et al. (2009). Comparative efficacy and acceptability of 12 relationships in the depressed elderly. British Journal of Psychiatry, 155,
new-generation antidepressants: a multiple-treatments meta-analysis. 787–92.
Lancet, 373, 746–58. Fabre, I., et al. (2004). Antidepressant efficacy and cognitive effects of
Cole, M.G. and Bellavance, F. (1997a). Depression in elderly medical repetitive transcranial magnetic stimulation in vascular depression: an
inpatients: a meta-analysis of outcomes. Canadian Medical Association open trial. International Journal of Geriatric Psychiatry, 19, 833–42.
Journal, 157, 1055–60. Fernando, M.S., et al. (2006). White matter lesions in an unselected cohort
Cole, M.G. and Bellavance, F. (1997b). The prognosis of depression in old of the elderly: molecular pathology suggests origin from chronic
age. American Journal of Geriatric Psychiatry, 5, 4–14. hypoperfusion injury. Stroke, 37, 1391–8.
Cole, M.G. and Dendukuri, N. (2003). Risk factors for depression among Firbank, M.J., et al. (2007). Brain atrophy and white matter hyperintensity
elderly community subjects: a systematic review and meta-analysis. change in older adults and relationship to blood pressure. Brain
American Journal of Psychiatry, 160, 1147–56. atrophy, WMH change and blood pressure. Journal of Neurology, 254,
Cole, M.G., Bellavance, F., and Mansour, A. (1999). Prognosis of depression 713–21.
in elderly community and primary care populations: a systematic review Flint, A.J. and Rifat, S.L. (1999). Recurrence of first-episode geriatric
and meta-analysis. American Journal of Psychiatry, 156, 1182–9. depression after discontinuation of maintenance antidepressants.
Comijs, H.C., et al. (2007). Childhood adversity, recent life events and American Journal of Psychiatry, 156, 943–5.
depression in late life. Journal of Affective Disorders, 103, 243–6. Ford, D.E., et al. (1998). Depression is a risk factor for coronary artery
Cooper, C., et al. (2011). A systematic review of treatments for refractory disease in men: the precursors study. Archives of Internal Medicine, 158,
depression in older people. American Journal of Psychiatry, 168, 1422–6.
681–8. Foster, J.R. (1992). Use of lithium in elderly psychiatric patients: a review of
Copeland, J.R., et al. (1992). Alzheimer’s disease, other dementias, depression the literature Lithium, 3, 77–93.
and pseudodementia: prevalence, incidence and three-year outcome in Frank, E., et al. (1991). Conceptualization and rationale for consensus
Liverpool. British Journal of Psychiatry, 161, 230–9. definitions of terms in major depressive disorder. Remission, recovery,
Cotter, D., et al. (2001). Reduced glial cell density and neuronal size in relapse, and recurrence. Archives of General Psychiatry, 48, 851–5.
the anterior cingulate cortex in major depressive disorder. Archives of Gaete, J.M. and Bogousslavsky, J. (2008). Post-stroke depression. Expert
General Psychiatry, 58, 545–53. Review of Neurotherapeutics, 8, 75–92.
Crossley, N.A., et al. (2007). Acceleration and augmentation of antidepressants Geddes, J.R., et al. (2003). Relapse prevention with antidepressant drug treatment
with lithium for depressive disorders: two meta-analyses of randomized, in depressive disorders: a systematic review. Lancet, 361, 653–61.
placebo-controlled trials. Journal of Clinical Psychiatry, 68, 935–40. Ghazi-Noori, S., et al. (2009). Therapies for depression in Parkinson’s
Cuijpers, P., et al. (2008). Preventing the onset of depressive disorders: a disease. Cochrane Database of Systematic Reviews, 2, CD003465. doi:
meta-analytic review of psychological interventions. American Journal 10.1002/14651858.CD003465.
of Psychiatry, 165, 1272–80. Glassman, A.H., et al. (2002). Sertraline treatment of major depression
Davidson, K., et al. (2000). Do depression symptoms predict early in patients with acute MI or unstable angina. Journal of the American
hypertension incidence in young adults in the CARDIA study? Coronary Medical Association, 288, 701–9.
Artery Risk Development in Young Adults. Archives of Internal Medicine, Godin, O., et al. (2008). White matter lesions as a predictor of depression in
160, 1495–500. the elderly: the 3C-Dijon study. Biological Psychiatry, 63, 663–9.
de Groot, J.C., et al. (2000). Cerebral white matter lesions and depressive Goodwin, G.M. (2006). Depression and associated physical diseases and
symptoms in elderly adults . Archives of General Psychiatry, 57, symptoms. Dialogues in Clinical Neuroscience, 8, 259–65.
1071–6. Gureje, O., Oladeji, B., and Abiona, T. (2011). Incidence and risk factors for
Demirkan, A., et al. (2011). Genetic risk profiles for depression and anxiety in late-life depression in the Ibadan Study of Ageing. Psychological Medicine,
adult and elderly cohorts. Molecular Psychiatry, 16, 773–83. 41(9), 1897–906.
Dentino, A.N., et al. (1999). Association of interleukin-6 and other biologic Gurland, B., et al. (1976). The geriatric mental status interview (GMS).
variables with depression in older people living in the community. International Journal of Aging and Human Development, 7, 303–11.
Journal of the American Geriatrics Society, 47, 6–11. Hackett, M.L., et al. (2008). Interventions for treating depression after stroke.
Dew, M.A., et al. (2007). Recovery from major depression in older adults Cochrane Database of Systematic Reviews, 4, CD003437.
receiving augmentation of antidepressant pharmacotherapy. American Hackett, M.L., et al. (2010). Pharmaceutical interventions for emotionalism
Journal of Psychiatry, 164, 892–9. after stroke. Cochrane Database of Systematic Reviews, 2, CD003690.
Dhondt, T.D.F., et al. (2002). Iatrogenic depression in the elderly. Results Hanlon, J.T., Handler, S.M., and Castle, N.G. (2010). Antidepressant prescribing
from a community-based study in the Netherlands. Social Psychiatry and in US nursing homes between 1996 and 2006 and its relationship to
Psychiatric Epidemiology, 37, 393–8. staffing patterns and use of other psychotropic medications. Journal of
Drevets, W.C., et al. (1997). Subgenual prefrontal cortex abnormalities in the American Medical Directors Association, 11, 320–4.
mood disorders. Nature, 386, 824–7. Harugeri, A., et al. (2010). Prescribing patterns and predictors of high-level
Dufouil, C., et al. (2001). Longitudinal study of blood pressure and white polypharmacy in the elderly population: A prospective surveillance
matter hyperintensities: the EVA MRI cohort. Neurology, 56, 921–6. study from two teaching hospitals in India. American Journal of Geriatric
Dufouil, C., et al. (2005). Effects of blood pressure lowering on cerebral Pharmacotherapy, 8, 271–80.
white matter hyperintensities in patients with stroke: the PROGRESS Henderson, A.S., et al. (1986). Social support, dementia and depression
(Perindopril Protection Against Recurrent Stroke Study) magnetic among the elderly living in the Hobart community. Psychological
resonance imaging substudy. Circulation, 112, 1644–50. Medicine, 16, 379–90.
Hendricksen, M., et al. (2004). Neuropathological study of the dorsal raphe Khundakar, A. and Thomas, A.J. (2009). Morphometric changes in early and
nuclei in late-life depression and Alzheimer’s disease with and without late-life major depressive disorder: evidence from postmortem studies.
depression. American Journal of Psychiatry, 161, 1096–102. International Psychogeriatrics, 21, 844–54.
Herrmann, L.L., Le Masurier, M., and Ebmeier, K.P. (2008). White matter Khundakar, A., et al. (2009). Morphometric analysis of neuronal and glial
hyperintensities in late life depression: a systematic review. Journal of cell pathology in the dorsolateral prefrontal cortex in late-life depression.
Neurology, Neurosurgery and Psychiatry, 79, 619–24. British Journal of Psychiatry, 195, 163–169.
Hickie, I., et al. (1996). Subcortical hyperintensities on magnetic resonance Khundakar, A., et al. (2011a). Morphometric analysis of neuronal and glial
imaging in patients with severe depression-a longitudinal evaluation. cell pathology in the caudate nucleus in late-life depression. American
Biological Psychiatry, 42, 367–74. Journal of Geriatric Psychiatry, 19, 132–41.
Hickie, I.B., et al. (2007). Serotonin transporter gene status predicts caudate Khundakar, A., et al. (2011b). A morphometric examination of neuronal
nucleus but not amygdala or hippocampal volumes in older persons and glial cell pathology in the orbitofrontal cortex in late-life depression.
with major depression. Journal of Affective Disorders, 98, 137–42. International Psychogeriatrics, 23, 132–40.
Hopkinson, G. (1964). A genetic study of affective illness in patients over 50. Khundakar, A., et al. (2011c). Cellular pathology within the anterior cingulate
British Journal of Psychiatry, 110, 244–54. cortex of patients with late-life depression: a morphometric study.
Husain, M.M., et al. (2005). Age-related characteristics of depression: a Psychiatry Research, 194, 184–9.
preliminary STAR*D report. American Journal of Geriatric Psychiatry, Kim, J.M., et al. (2004). Vascular disease/risk and late-life depression in a
13, 852–60. Korean community population. British Journal of Psychiatry, 185,
Irwin, M., et al. (1990). Reduction of immune function in life stress and 102–7.
depression. Biological Psychiatry, 27, 22–30. Kim, J.M., et al. (2006). Vascular risk factors and incident late-life depression
Jeerakathil, T., et al. (2004). Stroke risk profile predicts white matter in a Korean population. British Journal of Psychiatry, 189, 26–30.
hyperintensity volume: the Framingham Study. Stroke, 35, 1857–61. Kirsch, I., et al. (2008). Initial severity and antidepressant benefits: a
Jenkins, R., et al. (1997). The National Psychiatric Morbidity surveys of meta-analysis of data submitted to the Food and Drug Administration.
Great Britain—initial findings from the household survey. Psychological PLoS Medicine/Public Library of Science, 5, e45.
Medicine, 27, 775–89. Klug, G., et al. (2010). Effectiveness of home treatment for elderly people with
Jeste, D.V. (2000). Tardive dyskinesia in older patients. Journal of Clinical depression: randomised controlled trial. British Journal of Psychiatry,
Psychiatry, 61, 27–32. 197, 463–7.
Johnson, W., et al. (2002). Frequency and heritability of depression Klysner, R., et al. (2002). Efficacy of citalopram in the prevention of recurrent
symptomatology in the second half of life: evidence from Danish twins depression in elderly patients: placebo controlled study of maintenance
over 45. Psychological Medicine, 32, 1175–85. therapy. British Journal of Psychiatry, 181, 29–35.
Jonas, B.S. and Mussolino, M.E. (2000). Symptoms of depression as a Kohler, S., et al. (2010a). The pattern and course of cognitive impairment in
prospective risk factor for stroke. Psychosomatic Medicine, 62, 463–71. late-life depression. Psychological Medicine, 40, 591–602.
Jones-Webb, R., et al. (1996). Relationships between depressive symptoms, Kohler, S., et al. (2010b). White matter hyperintensities, cortisol levels, brain
anxiety, alcohol consumption, and blood pressure: results from the atrophy and continuing cognitive deficits in late-life depression. British
CARDIA Study. Coronary Artery Risk Development in Young Adults Journal of Psychiatry, 196, 143–9.
Study. Alcoholism: Clinical and Experimental Research, 20, 420–7. Kohler, S., et al. (2011). Depressive symptoms and risk for dementia: a 9-year
Jorge, R.E., et al. (2004). Repetitive transcranial magnetic stimulation as follow-up of the maastricht aging study. American Journal of Geriatric
treatment of poststroke depression: a preliminary study. Biological Psychiatry, 19, 902–5.
Psychiatry, 55, 398–405. Kok, R.M., Heeren, T.J., and Nolen, W.A. (2011). Continuing treatment
Jorm, A.F. (1987). Sex and age differences in depression: a quantitative of depression in the elderly: a systematic review and meta-analysis of
synthesis of published research. Australian and New Zealand Journal of double-blinded randomized controlled trials with antidepressants.
Psychiatry, 21, 46–53. American Journal of Geriatric Psychiatry, 19, 249–55.
Joyce, P.R., et al. (1992). Elevated levels of acute phase plasma proteins in Krishnan, K.R., et al. (1992). Magnetic resonance imaging of the caudate
major depression. Biological Psychiatry, 32, 1035–41. nuclei in depression. Preliminary observations. Archives of General
Kaji, T., et al. (2010). Relationship between late-life depression and life Psychiatry, 49, 553–7.
stressors: large-scale cross-sectional study of a representative sample of Kumar, A., et al. (2000). Atrophy and high intensity lesions: complementary
the Japanese general population. Psychiatry and Clinical Neuroscience, neurobiological mechanisms in late-life major depression .
64, 426–34. Neuropsychopharmacology, 22, 264–74.
Kalayam, B. and Alexopoulos, G.S. (1999). Prefrontal dysfunction and Lai, T., et al. (2000). Reduction of orbital frontal cortex volume in geriatric
treatment response in geriatric depression. Archives of General Psychiatry, depression. Biological Psychiatry, 48, 971–5.
56, 713–18. Lee, J.S., et al. (2007). Persistent mild cognitive impairment in geriatric
Kasper, S., de Swart, H., and Friis Andersen, H. (2005). Escitalopram in the depression. International Psychogeriatrics, 19, 125–35.
treatment of depressed elderly patients. American Journal of Geriatric Lenoir, H., et al. (2008). Relationship between blood pressure and depression
Psychiatry, 13, 884–91. in the elderly. The Three-City Study. Journal of Hypertension, 26,
Katona, C., et al. (2005). Pain symptoms in depression: definition and clinical 1765–72.
significance. Clinical Medicine, 5, 390–5. Lenoir, H., et al. (2011). Depression history, depressive symptoms, and
Kellner, C.H., et al. (2006). Continuation electroconvulsive therapy vs incident dementia: the 3C Study. Journal of Alzheimer’s Disease, 26,
pharmacotherapy for relapse prevention in major depression: a multisite 27–38.
study from the Consortium for Research in Electroconvulsive Therapy Levinson, D.F. (2006). The genetics of depression: a review. Biological
(CORE). Archives of General Psychiatry, 63, 1337–44. Psychiatry, 60, 84–92.
Kennedy, G.J., Kelman, H.R., and Thomas, C. (1991). Persistence and Lichter, D.G. and Cummings, J.L. (2001). Introduction and overview. In:
remission of depressive symptoms in late life. American Journal of Lichter, D.G. and Cummings, J.L. (eds) Frontal subcortical circuits in
Psychiatry, 148, 174–8. psychiatric and neurological disorders. Guilford Press, London.
Khan, A., et al. (2002). Severity of depression and response to antidepressants Licht-Strunk, E., et al. (2009). Outcome of depression in later life in primary
and placebo: an analysis of the Food and Drug Administration database. care: longitudinal cohort study with three years’ follow-up. British
Journal of Clinical Psychopharmacology, 22, 40–5. Medical Journal, 338, a3079.
Lieberman, A. (2006). Depression in Parkinson’s disease—a review. Acta Murphy, E. (1982). Social origins of depression in old age. British Journal of
Neurologica Scandinavica, 113, 1–8. Psychiatry, 141, 135–42.
Lin, E.H., et al. (2003). Effect of improving depression care on pain and Murphy, E. (1983). The prognosis of depression in old age. British Journal of
functional outcomes among older adults with arthritis: a randomized Psychiatry, 142, 111–19.
controlled trial. (See Comment.) Journal of the American Medical Murphy, E., et al. (1988). Increased mortality rates in late-life depression.
Association, 290, 2428–9. British Journal of Psychiatry, 152, 347–53.
Llewellyn-Jones, R.H., et al. (1999). Multifaceted shared care intervention Naarding, P., et al. (2007). Clinically defined vascular depression in the
for late life depression in residential care: randomised controlled trial. general population. Psychological Medicine, 37, 383–92.
British Medical Journal, 319, 676–82. Nelson, J.C., Delucchi, K., and Schneider, L.S. (2008). Efficacy of second
Lloyd, A.J., et al. (2004). Hippocampal volume change in depression: late- generation antidepressants in late-life depression: a meta-analysis of the
and early-onset illness compared. British Journal of Psychiatry, 184, evidence. American Journal of Geriatric Psychiatry, 16, 558–67.
488–95. Nestler, E.J. and Carlezon, W.A., Jr (2006). The mesolimbic dopamine reward
Lockwood, K.A., Alexopoulos, G.S., and van Gorp, W.G. (2002). Executive circuit in depression. Biological Psychiatry, 59, 1151–9.
dysfunction in geriatric depression. American Journal of Psychiatry, 159, NICE (2009). Depression in adults: full guidance. Clinical Guidance 90. NICE,
1119–26. London.
Lyketsos, C.G., et al. (2000). Mental and behavioral disturbances in dementia: O’Brien, J., Ames, D., and Schweitzer, I. (1993). HPA axis function in
findings from the Cache County Study on Memory in Aging. American depression and dementia: a review. International Journal of Geriatric
Journal of Psychiatry, 157, 708–14. Psychiatry, 8, 887–98.
Lyness, J.M., et al. (1998). Cerebrovascular risk factors and later-life major O’Brien, J., et al. (1996). A magnetic resonance imaging study of white
depression. Testing a small-vessel brain disease model. American Journal matter lesions in depression and Alzheimer’s disease. British Journal of
of Geriatric Psychiatry, 6, 5–13. Psychiatry, 168, 477–85.
Lyness, J.M., et al. (1999). Cerebrovascular risk factors and depression in O’Brien, J., et al. (1998). Severe deep white matter lesions and outcome in
older primary care patients: testing a vascular brain disease model of elderly patients with major depressive disorder: follow up study. British
depression. American Journal of Geriatric Psychiatry, 7, 252–8. Medical Journal, 317, 982–4.
Lyness, J.M., et al. (2000). Cerebrovascular risk factors and 1-year depression O’Brien, J., et al. (2001). Cognitive impairment in depression is not associated
outcome in older primary care patients. American Journal of Psychiatry, with neuropathologic evidence of increased vascular or Alzheimer-type
157, 1499–501. pathology. Biological Psychiatry, 49, 130–6.
Manthorpe, J. and Iliffe, S. (2010). Suicide in later life: public health and O’Brien, J.T., et al. (2004). A longitudinal study of hippocampal volume,
practitioner perspectives. International Journal of Geriatric Psychiatry, cortisol levels, and cognition in older depressed subjects. American
25, 1230–8. Journal of Psychiatry, 161, 2081–90.
Mayberg, H.S., et al. (1997). Cingulate function in depression: a potential O’Connor, M.K., et al. (2001). The influence of age on the response of major
predictor of treatment response. Neuroreport, 8, 1057–61. depression to electroconvulsive therapy: a C.O.R.E. Report. American
Mayberg, H.S., et al. (1999). Early and late fluoxitene effects on regional Journal of Geriatric Psychiatry, 9, 382–90.
glucose metabolism in depression. Biological Psychiatry, 45, 1115. O’Hara, R., et al. (2007). Serotonin transporter polymorphism, memory and
McCusker, J., et al. (2006). Does depression in older medical inpatients hippocampal volume in the elderly: association and interaction with
predict mortality? Journals of Gerontology Series A-Biological Sciences cortisol. Molecular Psychiatry, 12, 544–55.
and Medical Sciences, 61, 975–81. Olin, J.T., et al. (2002). Provisional diagnostic criteria for depression of
McKeith, I.G., et al. (1992). Operational criteria for senile dementia of Lewy Alzheimer disease. American Journal of Geriatric Psychiatry, 10, 125–8.
body type (SDLT). Psychological Medicine, 22, 911–22. Ongur, D., Drevets, W.C., and Price, J.L. (1998). Glial reduction in the
Meeks, T.W., et al. (2011). A tune in ‘a minor’ can ‘b major’: a review subgenual prefrontal cortex in mood disorders. Proceedings of the
of epidemiology, illness course, and public health implications of National Academy of Sciences of the USA, 95, 13290–5.
subthreshold depression in older adults. Journal of Affective Disorders, Ownby, R.L., et al. (2006). Depression and risk for Alzheimer disease:
129, 126–42. systematic review, meta-analysis, and metaregression analysis. Archives
Menza, M. (2006). STAR*D: the results begin to roll in. American Journal of of General Psychiatry, 63, 530–8.
Psychiatry, 163, 1123. Pai, M. and Carr, D. (2010). Do personality traits moderate the effect of
Menza, M., et al. (2009). A controlled trial of antidepressants in patients with late-life spousal loss on psychological distress? Journal of Health and
Parkinson disease and depression. Neurology, 72, 886–92. Social Behaviour, 51, 183–99.
Meyer, C.M., et al. (2004). Incident hypertension associated with depression Pakhala, K. (1990). Social and environmental factors and depression in old
in the Baltimore epidemiologic catchment area follow-up study. Journal age International Journal of Geriatric Psychiatry, 5, 99–113.
of Affective Disorders, 83, 127–33. Parashos, I.A., et al. 1998. Magnetic-resonance morphometry in patients
Meyer, T.D., et al. (2010). Cognitive style and depressive symptoms in with major depression. Psychiatry Research, 84, 7–15.
elderly people—extending the empirical evidence for the cognitive Penninx, B.W., et al. (1999). Minor and major depression and the risk of
vulnerability-stress hypothesis. Behavioural Research Therapy, 48, death in older persons. Archives of General Psychiatry, 56, 889–95.
1053–7. Penninx, B.W., et al. (2003). Inflammatory markers and depressed mood in
Mezuk, B., et al. (2008). Depression and type 2 diabetes over the lifespan: a older persons: results from the Health, Aging and Body Composition
meta-analysis. Diabetes Care, 31, 2383–90. study. Biological Psychiatry, 54, 566–72.
Mitchell, A.J. and Subramaniam, H. (2005). Prognosis of depression in old Pico, F., et al. (2002). Longitudinal study of carotid atherosclerosis and white
age compared to middle age: a systematic review of comparative studies. matter hyperintensities: the EVA-MRI cohort. Cerebrovascular Diseases,
American Journal of Psychiatry, 162, 1588–601. 14, 109–15.
Mojtabai, R. and Olfson, M. (2010). National trends in psychotropic Pinquart, M., Duberstein, P.R., and Lyness, J.M. (2006). Treatments for later-life
medication polypharmacy in office-based psychiatry. Archives of General depressive conditions: a meta-analytic comparison of pharmacotherapy
Psychiatry, 67, 26–36. and psychotherapy. American Journal of Psychiatry, 163, 1493–501.
Mosimann, U.P., et al. (2002). Antidepressant effects of repetitive transcranial Post, F. (1972). The management and nature of depressive illnesses in
magnetic stimulation in the elderly: correlation between effect size and late life: a follow-through study. British Journal of Psychiatry, 121,
coil-cortex distance. Archives of General Psychiatry, 59, 560–1. 393–404.
Pratt, L.A., et al. (1996). Depression, psychotropic medication, and risk Sackeim, H.A., et al. (2001b). Vagus nerve stimulation (VNS) for
of myocardial infarction. Prospective data from the Baltimore ECA treatment-resistant depression: efficacy, side effects, and predictors of
follow-up. Circulation, 94, 3123–9. outcome. Neuropsychopharmacology, 25, 713–28.
Prince, M.J., et al. (1997a). Impairment, disability and handicap as risk Salaycik, K.J., et al. (2007). Depressive symptoms and risk of stroke: the
factors for depression in old age. The Gospel Oak Project V. Psychological Framingham Study. Stroke, 38, 16–21.
Medicine, 27, 311–21. Sanacora, G. and Saricicek, A. (2007). GABAergic contributions to the
Prince, M.J., et al. (1997b). Social support deficits, loneliness and life events pathophysiology of depression and the mechanism of antidepressant
as risk factors for depression in old age. The Gospel Oak Project VI. action. CNS and Neurological Disorders Drug Targets, 6, 127–40.
Psychological Medicine, 27, 323–32. Schoevers, R.A., et al. (2000). Association of depression and gender with
Prince, M.J., et al. (1998). A prospective population-based cohort study of the mortality in old age. Results from the Amsterdam Study of the Elderly
effects of disablement and social milieu on the onset and maintenance of (AMSTEL). British Journal of Psychiatry, 177, 336–42.
late-life depression. The Gospel Oak Project VII. Psychological Medicine, Schoevers, R.A., et al. (2009). Depression and excess mortality: evidence for a
28, 337–50. dose response relation in community living elderly. International Journal
Pulska, T., et al. (1999). Follow up study of longstanding depression as of Geriatric Psychiatry, 24, 169–76.
predictor of mortality in elderly people living in the community. British Schulz, R., et al. (2000). Association between depression and mortality in
Medical Journal, 318, 432–3. older adults: the Cardiovascular Health Study. Archives of Internal
Rajkowska, G., et al. (1999). Morphometric evidence for neuronal and glial Medicine, 160, 17618.
prefrontal cell pathology in major depression. Biological Psychiatry, 45, Sharma, V.K., et al. (1998). Outcome of the depressed elderly living in the
1085–98. community in Liverpool: a 5-year follow-up. Psychological Medicine, 28,
Rajkowska, G., et al. (2005). Prominent reduction in pyramidal neurons 1329–37.
density in the orbitofrontal cortex of elderly depressed patients. Biological Sheline, Y.I., et al. (1996). Hippocampal atrophy in recurrent major depression.
Psychiatry, 58, 297–306. Proceedings of the National Academy of Sciences USA, 93, 3908–13.
Reichman, W.E. and Coyne, A.C. (1995). Depressive symptoms in Alzheimer’s Sheline, Y.I., et al. (2006). Cognitive function in late life depression:
disease and multi-infarct dementia. Journal of Geriatric Psychiatry and relationships to depression severity, cerebrovascular risk factors and
Neurology, 8, 96–9. processing speed. Biological Psychiatry, 60, 58–65.
Reichstadt, J., et al. (2010). Older adults’ perspectives on successful aging: Sheline, Y.I., et al. (2008). Regional white matter hyperintensity burden in
qualitative interviews. American Journal of Geriatric Psychiatry, 18, 567–75. automated segmentation distinguishes late-life depressed subjects from
Reynolds, C.F., 3rd, et al. (1996). Treatment outcome in recurrent major comparison subjects matched for vascular risk factors. American Journal
depression: a post hoc comparison of elderly (‘young old’) and midlife of Psychiatry, 165, 524–32.
patients. American Journal of Psychiatry, 153, 1288–92. Simonsick, E.M., et al. (1995). Depressive symptomatology and
Reynolds, C.F., 3rd, et al. (1998). Effects of age at onset of first lifetime episode hypertension-associated morbidity and mortality in older adults.
of recurrent major depression on treatment response and illness course Psychosomatic Medicine, 57, 427–35.
in elderly patients. American Journal of Psychiatry, 155, 795–9. Sirey, J.A., Bruce, M.L. and Alexopoulos, G.S. (2005). The Treatment
Reynolds, C.F., 3rd, et al. (1999). Nortriptyline and interpersonal Initiation Program: an intervention to improve depression outcomes in
psychotherapy as maintenance therapies for recurrent major depression: older adults. American Journal of Psychiatry, 162, 184–6.
a randomized controlled trial in patients older than 59 years. Journal of Sjosten, N. and Kivela, S.-L. (2006). The effects of physical exercise on
the American Medical Association, 281, 39–45. depressive symptoms among the aged: a systematic review. International
Reynolds, C.F., 3rd, et al. (2006). Maintenance treatment of major depression Journal of Geriatric Psychiatry, 21, 410–18.
in old age. New England Journal of Medicine, 354, 1130–8. Slotema, C.W., et al. (2010). Should we expand the toolbox of psychiatric
Reynolds, C.F., 3rd, et al. (2011). Maintenance treatment of depression in treatment methods to include repetitive transcranial magnetic stimulation
old age: a randomized, double-blind, placebo-controlled evaluation (rTMS)? A meta-analysis of the efficacy of rTMS in psychiatric disorders.
of the efficacy and safety of donepezil combined with antidepressant Journal of Clinical Psychiatry, 71, 873–84.
pharmacotherapy. Archives of General Psychiatry, 68, 51–60. Smith, P.J., et al. (2010). Cerebrovascular risk factors and cerebral
Richardson, J., et al. (2009). A study of orthostatic hypotension in late-life hyperintensities among middle-aged and older adults with major
depression. American Journal of Geriatric Psychiatry, 17, 996–9. depression. American Journal of Geriatric Psychiatry, 18, 848–52.
Ried, L.D., et al. (2005). A Study of Antihypertensive Drugs and Depressive Smits, F., et al. (2008). An epidemiological approach to depression prevention
Symptoms (SADD-Sx) in patients treated with a calcium antagonist versus in old age. American Journal of Geriatric Psychiatry, 16, 444–53.
an atenolol hypertension Treatment Strategy in the International Verapamil Sneed, J.R., et al. (2008a). The vascular depression subtype: evidence of
SR-Trandolapril Study (INVEST). Psychosomatic Medicine, 67, 398–406. internal validity. Biological Psychiatry, 64, 491–7.
Roberts, R.E., et al. (1997). Does growing old increase the risk for depression? Sneed, J.R., et al. (2008b). Design makes a difference: a meta-analysis of
American Journal of Psychiatry, 154, 1384–90. antidepressant response rates in placebo-controlled versus comparator
Rosenberg, P.B., et al. 2010. Sertraline for the treatment of depression in trials in late-life depression. American Journal of Geriatric Psychiatry, 16,
Alzheimer disease. American Journal of Geriatric Psychiatry, 18, 136–45. 65–73.
Roth, M. (1955). The natural history of mental disorder in old age. Journal of Sneed, J.R., et al. (2010). Antidepressant medication and executive
Mental Science, 101, 281–301. dysfunction: a deleterious interaction in late-life depression. American
Rothwell, N.J. and Luheshi, G.N. (2000). Interleukin 1 in the brain: biology, Journal of Geriatric Psychiatry, 18, 128–35.
pathology and therapeutic target. Trends in Neurosciences, 23, 618–25. Steffens, D.C., et al. (1999). Cerebrovascular disease and depression symptoms
Rush, A.J. (2007). STAR*D: what have we learned? American Journal of in the cardiovascular health study. Stroke, 30, 2159–66.
Psychiatry, 164, 201–4. Steffens, D.C., et al. (2011). Change in hippocampal volume on magnetic
Sachdev, P.S. and Chen, X. (2009). Neurosurgical treatment of mood resonance imaging and cognitive decline among older depressed and
disorders: traditional psychosurgery and the advent of deep brain nondepressed subjects in the neurocognitive outcomes of depression
stimulation. Current Opinion in Psychiatry, 22, 25–31. in the elderly study. American Journal of Geriatric Psychiatry, 19,
Sackeim, H.A., et al. (2001a). Continuation pharmacotherapy in the prevention 4–12.
of relapse following electroconvulsive therapy: a randomized controlled Syed, A., et al. (2005). Depression in the elderly: pathological study of raphe and
trial. Journal of the American Medical Association, 285, 1299–307. locus ceruleus. Neuropathology and Applied Neurobiology, 31, 405–13.
Synofzik, M. and Schlaepfer, T.E. (2011). Electrodes in the brain-Ethical Thomas, A.J., et al. (2009). A comparison of neurocognitive impairment in
criteria for research and treatment with deep brain stimulation for younger and older adults with major depression. Psychological Medicine,
neuropsychiatric disorders. Brain Stimulation, 4, 7–16. 39, 725–33.
Taragano, F. E., et al. (2001). A double blind, randomized clinical trial Tiemeier, H., et al. (2004). Relationship between atherosclerosis and late-life
assessing the efficacy and safety of augmenting standard antidepressant depression: the Rotterdam Study. Archives of General Psychiatry, 61,
therapy with nimodipine in the treatment of ‘vascular depression’. 369–76.
International Journal of Geriatric Psychiatry, 16, 254–60. Tsopelas, C., et al. (2011). Neuropathological correlates of late-life depression
Taragano, F.E., Bagnatti, P., and Allegri, R.F. (2005). A double-blind, in older people. British Journal of Psychiatry, 198, 109–14.
randomized clinical trial to assess the augmentation with nimodipine UK ECT Review Group (2003). Efficacy and safety of electroconvulsive
of antidepressant therapy in the treatment of ‘vascular depression’. therapy in depressive disorders: a systematic review and meta-analysis.
International Psychogeriatrics, 17, 487–98. Lancet, 361, 799–808.
Taylor, W.D., et al. (2003). White matter hyperintensity progression and Unutzer, J., et al. (2002). Collaborative care management of late-life depression
late-life depression outcomes. Archives of General Psychiatry, 60, in the primary care setting: a randomized controlled trial. Journal of the
1090–6. American Medical Association, 288, 2836–45.
Taylor, W.D., et al. (2004). Late-life depression and microstructural van Calker, D., et al. (2009). Time course of response to antidepressants:
abnormalities in dorsolateral prefrontal cortex white matter. American predictive value of early improvement and effect of additional
Journal of Psychiatry, 161, 1293–6. psychotherapy. Journal of Affective Disorders, 114, 243–53.
Teodorczuk, A., et al. (2007). White matter changes and late-life depressive van Otterloo, E., et al. (2009). Reductions in neuronal density in elderly
symptoms: longitudinal study. British Journal of Psychiatry, 191, 212–17. depressed are region specific. International Journal of Geriatric Psychiatry,
Teodorczuk, A., et al. (2010). Relationship between baseline white-matter 24, 856–64.
changes and development of late-life depressive symptoms: 3-year results van’t Veer-Tazelaar, P.J., et al. (2009). Stepped-care prevention of anxiety and
from the LADIS study. Psychological Medicine, 40, 603–10. depression in late life: a randomized controlled trial. Archives of General
Tew, J.D., Jr., et al. (1999). Acute efficacy of ECT in the treatment of major Psychiatry, 66, 297–304.
depression in the old-old. American Journal of Psychiatry, 156, 1865–70. van’t Veer-Tazelaar, P.J., et al. (2011). Prevention of late-life anxiety and
Thomas, A.J. and O’Brien, J.T. (2008). Depression and cognition in older depression has sustained effects over 24 months: a pragmatic randomized
adults. Current Opinion in Psychiatry, 21, 8–13. trial. American Journal of Geriatric Psychiatry, 19, 230–9.
Thomas, A.J., et al. (2000). Elevation in late-life depression of intercellular Vasudev, A., et al. (2011). A study of orthostatic hypotension, heart rate
adhesion molecule-1 expression in the dorsolateral prefrontal cortex. variability and baroreflex sensitivity in late-life depression. Journal of
American Journal of Psychiatry, 157, 1682–4. Affective Disorders, 131, 374–8.
Thomas, A.J., et al. (2001). A neuropathological study of vascular factors in Veith, R.C. and Raskind, M.A. (1988). The neurobiology of aging: does it
late-life depression. Journal of Neurology, Neurosurgery and Psychiatry, predispose to depression? Neurobiology of Aging, 9, 101–17.
70, 83–7. Weintraub, D., et al. (2010). Sertraline for the treatment of depression in
Thomas, A.J., et al. (2002a). Cell adhesion molecule expression in the Alzheimer disease: week-24 outcomes. American Journal of Geriatric
dorsolateral prefrontal cortex and anterior cingulate cortex in major Psychiatry, 18, 332–40.
depression in the elderly. British Journal of Psychiatry, 181, 129–34. Whyte, E.M., et al. (2004). Time course of response to antidepressants in late-life
Thomas, A.J., et al. (2002b). Ischemic basis for deep white matter major depression: therapeutic implications. Drugs and Aging, 21, 531–54.
hyperintensities in major depression: a neuropathological study. Archives Williams, J.W., Jr, et al. (2000). Treatment of dysthymia and minor depression
of General Psychiatry, 59, 785–92. in primary care: a randomized controlled trial in older adults. Journal of
Thomas, A.J., et al. (2003). A neuropathological study of periventricular the American Medical Association, 284, 1519–26.
white matter hyperintensities in major depression. Journal of Affective Wilson, K., et al. (2001). Antidepressants versus placebo for the depressed
Disorders, 76, 49–54. elderly. Cochrane Database of Systematic Reviews, 1, CD000561.
Thomas, A.J., Kalaria, R.N., and O’Brien, J.T. (2004). Depression and vascular Wilson, K.C.M., Mottram, P.G., and Vassilas, C.A. (2009). Psychotherapeutic
disease: what is the relationship? Journal of Affective Disorders, 79, treatments for older depressed people. Cochrane Database of Systematic
81–95. Reviews, 1, CD004853. doi: 10.1002/14651858.CD004853.pub2.
Thomas, A.J., et al. (2005). Increase in interleukin-1 (beta) in late-life Zarate, C., et al. (2010). Glutamatergic modulators: the future of treating
depression. American Journal of Psychiatry, 162, 175–7. mood disorders? Harvard Review of Psychiatry, 18, 293–303.
CHAPTER 43
Suicide and attempted
suicide in older people
Helen Chiu and Joshua Tsoh
In most nations of the world, the risk of suicide increases with recorded as suicide. For instance, a narrative verdict may describe
age, particularly for older men (WHO, 2011). The magnitude of the circumstances of the death briefly without attributing the death
this global public health challenge will escalate rapidly with the to a category such as suicide (Carroll et al., 2012).
anticipated worldwide burgeoning of the older population. By Suicide rates are affected by age, period, and cohort factors
2020, the population aged over 60 is estimated to expand to over (Wasserman, 1989; Lindesay, 1991). In older people, suicide rates
1 billion, 70% of whom dwell in developing countries (United are higher in the age-band 75+ years compared to the age-band
Nations, 2009). Another factor highlighted by some researchers 65–74 years for men and women (Shah, 2007; Shah et al., 2007).
is the cohort effect from baby-boomers, who may have a higher Suicide rates are usually higher in men compared to women for
rate of suicide than their predecessors entering old age after the most nations (Dombrovski et al., 2008a). The ratio of men to
first decade of the millennium (Conwell et al., 2002a; Phillips et women suicide rates is around 3:1 in many western countries, but
al., 2010). In this book chapter, we first examine the scope of the this ratio is lower in many Asian countries (Chiu et al., 2001). In
problem and, from a public health perspective, address the risk and China, the gender ratio is notably close to 1:1 (WHO, 2011), and it
protective factors related to suicidal behaviours in late life, upon is noteworthy that rural suicide rates among older people are three
which evidence-based prevention and intervention (‘postvention’) to five times higher than the urban rates (Li et al., 2009).
measures might be mounted at accessible points along the suicide Some authors have speculated that cohort effects (the ensemble
trajectory (Conwell et al., 2011). We review the available studies of environmental factors that connote a certain generation and of
that have looked at possibilities for prevention of suicide in older its relative size (De Leo, 2002)) may further accentuate the older
people around the world. people suicide rate in the foreseeable future (Conwell et al., 2002a).
Higher suicide rates have been identified in the generations born in
Epidemiology and Trends the fertile post-war decades in the US (Conwell et al., 2010) and in
18 European nations (Pampel, 1996), especially among men. Suicide
Completed suicide rates tend to be higher in those age groups that comprise larger
Of the 1 million completed suicides that occur in the world annu- proportions of the total population, perhaps because of increased
ally, up to 60% take place in Asia, especially in China, India, and competition for the scarce societal resources available (Lindesay,
Japan (Beautrais, 2006). In most nations, rates of suicide in older 1991). However, because of different cultural influences, this cohort
people are higher than rates in younger age groups (WHO, 2011). effect is unlikely to be universally applicable across countries (De
Older people suicide rates are generally lowest in the Caribbean, Leo, 2002). Further studies are needed to examine the impact of
central American, and Arabic countries, and the highest in cen- this factor on the trends of older people suicide rates.
tral and eastern European, some Oriental, as well as some west- Period effects (e.g. from wars or major catastrophes experienced
ern European countries (Shah, 2007). However, caution is by the whole population at the same time) on suicide rates among
necessary when comparing suicide statistics across countries. First, older people are even less well understood. In Hong Kong, during the
there is a lack of reliable surveillance systems in some low- and community outbreak of severe acute respiratory syndrome (SARS)
middle-income countries. Even for the WHO databank on mortal- in 2003, the older people suicide rate in Hong Kong showed a sharp
ity, there is scarce information on suicide data from some regions of upturn from a previous downward trend, particularly in older women.
the world. Second, for some countries like China, suicide statistics It was postulated that factors such as the breakdown of social networks
are collected from a small percentage of the population, which may and limited access to healthcare might account for the findings: older
influence the accuracy of the data. Third, there are variations in the women, because of their previous readiness to utilize social and health
practice of ascertaining a death as suicide in different countries. In services instituted in the past decade, might have been more suscepti-
countries with a Coroner system, the extent of the use of ‘open ver- ble to the effects of temporary suspension of these socially supportive
dicts’ and ‘narrative verdicts’ might influence the number of deaths services during the epidemic (Chan et al., 2006).
Attempted suicide lethal means, neurobiological factors) (Conwell and Thompson,

There is very limited epidemiologic data on attempted suicide in 2008; Conwell, et al., 2011; Van Orden and Conwell, 2011).
old age, as, unlike completed suicide, there is usually no systematic Mental health
registry for such events globally, even in the most developed coun-
tries. Attempted suicide is far less frequent in later life than among In a systematic review of global data from controlled PA studies
younger age groups (Moscicki, 1997). In adolescents, the ratio of on suicide in older people, Conwell et al. (2011) reported that of
attempted to completed suicide has been estimated to be 200:1. In all factors examined, psychiatric illness was consistently found to
contrast, there are approximately four attempts for each completed be most prominent: it is present in 71–97% of suicides, with affec-
suicide in later life (Conwell, et al., 2002a). Contributory factors tive disorder being the most common (with an odds ratio (OR) of
for the increased lethality of self-harm behaviours in older peo- 47.7–184.6 in studies with community controls) and, in particular,
ple include their greater planning and resolve and the violence of major depression. Psychotic disorders, including schizophrenia,
methods often used in their suicide acts, their diminished physical schizoaffective disorder, and delusional disorder, as well as anxi-
resilience, and greater isolation (thus less chance of timely availabil- ety disorder, tend to be present in lower proportions, unlike sui-
ity of medical interventions), as well as their reluctance to express cide among younger populations. Furthermore, the prevalence
or endorse suicide ideations (Conwell et al., 2002a). In a control- of comorbid substance use disorder was highly variable in these
led study on older suicide completers, attempters, and community studies, and this might be related to the different rates of problem
controls, it was further shown that the older suicide completers and drinking in Oriental and western populations (Chiu, et al., 2004;
hospital-treated attempters were largely overlapping populations Waern et al., 2002a; Conwell, et al., 2010, 2011).
and their acts were driven by very similar spectra of risk factors. Despite the prevalence of dementia in old age, it is infrequently
This is different from the situation in their younger counterparts, diagnosed in completed suicide using the PA method, which is
where major disparities were found in sociodemographic factors likely related to its intrinsic methodological limitations, as the
and diagnoses (Tsoh et al., 2005). Furthermore, it added to the evi- clinical picture of dementia, particularly in the early course, is less
dence that in late life, preceding suicide attempts were among the likely to be recognized by family members and other informants
strongest predictor for subsequent suicide (Beautrais, 2002; Waern (Conwell et al., 2011). There are biological (Rubio et al., 2001) and
et al., 2002a; Chiu et al., 2004; Conwell et al., 2010). Both findings epidemiological (Erlangsen et al., 2008) data that support the notion
show the need for high vigilance in the management of all cases of that dementia is associated with an amplified risk of suicide in late
deliberate self-harm in older people. life. Furthermore, more recently, a nationwide longitudinal study in
Denmark of persons aged 50 and over found that in both genders,
the diagnosis of dementia upon hospitalization was associated with
Risk and protective factors for suicide a significantly elevated risk of completed suicide (especially in the
first few years after the diagnosis was made) in the 10-year study
in older adults period (Erlangsen et al., 2008).
To understand and prevent suicide in older people, it is important to A prior history of attempted suicide is a significant risk factor
explore the related risk and protective factors from a public health for suicide (Beautrais,, 2002; Hawton and Harriss, 2006). Chiu
perspective. Such factors are often identified through retrospective et al. (2004) found that almost one-third of older people in a
case-control psychological autopsy (PA) studies. In PA research, Chinese community who committed suicide had a history of a
information about suicide decedents is gathered retrospectively suicide attempt, and that over 30% of these attempts had occurred
from relatives and other informants familiar with the decedents, as within 1 week prior to the completed suicide. On the other hand,
well as perusal of the medical records (Hawton et al., 1998). Despite retrospective PA studies indicate that nearly 75% of suicides died
its limitations, such as its retrospective nature and the potential for on their first attempt (Chiu et al., 2004; Conwell et al., 2010), so
reporting bias, case-controlled PA studies around the globe in the effective public health strategies towards suicide prevention pro-
past decade were striking in their close resemblance of the array grammes must include both high-risk (e.g. targeting the suicide
of factors associated with suicide in older people (Harwood et attempters) and population measures, in accordance with Rose’s
al., 2001; Beautrais, 2002; Waern, et al., 2002a; Chiu et al., 2004; theorem (Rose, 1985).
Conwell et al., 2010) (see Table 43.1 for details). Both studies in the West (as summarized by Conwell et al., 2008)
Other methods for ascertaining risk and protective factors of and the Orient (Chiu et al., 2004) indicate that although the rate
suicide include prospective follow-up studies of high-risk subjects of psychiatric diagnosis is high, only a minority of the decedents
(e.g. depressed individuals and suicide attempters) and case-control had consulted specialist psychiatrists before their fatal act. On the
studies of older suicide attempters (as their profiles resemble the other hand, up to 80% of them had consulted a primary care doc-
completers). The former is often limited by a low base rate of sui- tor in the preceding month. In old age, depressive symptoms are
cide and lack of statistical power to demonstrate the effect size of often attenuated or expressed as physical symptoms and thus eas-
the putative risk factors. The latter is limited by the lack of system- ily overlooked (Fountoulakis et al., 2003; Conwell and Thompson,
atic registry for suicide attempters. Nonetheless, analysis of the 2008). These findings underscore the importance of screening for
data from the aforementioned studies has consistently pointed to suicidal risk and depression, at the primary care level. A number
the finding that suicidal behaviours in seniors are multidetermined of collaborative stepped-care management programmes target-
tragic outcomes. The following risk and protective factors are con- ing older primary care clients with chronic physical illness using
sistently identified across several broad domains: mental health, algorithm-driven treatment of depression have shown that sui-
personality, physical diseases, social factors (e.g. network and sup- cide ideation is sustainably reduced among the participants after
port, life events), functional levels, and others (including access to completion of the programme, in addition to effective treatment
CHAPTER 43 suicide and attempted suicide in older people 573
Table 43.1 Odds ratios (OR) of suicide for risk and protective factors in different domains that are statistically significant in case-control
psychological autopsy studies of older adults
Harwood et al. (2001) Beautrais (2002) Waern et al. Chiu et al. (2004); Conwell et al. (2010)
(2002a, 2002b) Tsoh et al. (2005)
Location Central England New Zealand Goteborg, Sweden Hong Kong, China Monroe and Onondaga
Counties, New York, US
Age ≥ 60 ≥ 55 ≥ 65 ≥ 60 ≥ 50
Sample size 100 31 85 70 86
OR of risk/protective factors
Mental health
Any mood disorder 4.0 184.6 63.1 59.2 47.7
Major depressive – – 28.6 36.3 12.2
episode
Substance abuse ns 4.4 43.1 ns ns
disorder
Dementia 0.2 ns ns ns ns
Past suicide attempts – 41.9 – 18.9 13.6
Personality
NEO Personality – – – Neuroticism: 1.2 –
Inventory domains Extraversion: 0.8
OTE: 0.8
Conscientiousness: 0.8
Any personality disorder 4.0 – – – –
Physical health
Any serious medical – 19.4 Men: 4.2 – –
condition Women: 2.1
Overall: 3.0
Perceived poor health – 1.6 – – –
Number of physical – – – 1.9 –
conditions
Recent hospitalization – 25.8 (hospitalization in – 6.8 (hospitalization in the –
the past year) past 3 months)
IADL – – – 0.2 –
Social factors
Number of recent (past – – – 4.0 –
6 months) major life
events
Family discord – 17.2 – 41.0 –
Serious financial – 10.0 – ns –
problems
Low social interaction – 25.8 – – –
Poor social network – – – 1.1 –
(LSNS)
Other factors
Living with children – – – 0.5 –
Religion considered – – – 0.2 –
salient
IADL, instrumental activities of daily living (in the adapted version in Hong Kong, a lower score indicates worse functional level); LSNS, Lubben Social Network Scale; ; ns, not statistically
significant; OTE, openness to experience.
of depression, for instance, the Prevention of Suicide in Primary suicide, and that stroke, bone fracture, arthritis, and malignancy
Care Elderly: Collaborative Trial (PROSPECT) (Alexopoulos et al., were related to attempted suicide. Furthermore, a 10-year prospec-
2005, 2009). tive multisite study conducted in the US found that bone fractures
Several ecological studies have found an association between were associated with suicide in older people (Turvey et al., 2002).
increased antidepressant prescription and fewer completed sui- However, as the base rates of physical illness are very high in the
cides based on national registry and population statistics, which older population, their usefulness in identifying individual older
points to the possibility that antidepressant use may be associated people who need intervention is relatively weak (Conwell et al.,
with a reduced risk for completed suicide at a population level 2011). Conwell et al. also suggested that beyond individual physical
(Henriksson and Isacsson, 2006; Bramness et al.,2007). Such an illness, the perceived meaning of the illness burden and its impact
association was also found in senior populations in Australia (Hall on function, and limitations on autonomy and personal integrity
et al., 2003) and the UK (Shah and Lodhi, 2005). should not be overlooked. This is supported by case-controlled PA
studies that showed that the presence of any impairment in instru-
Personality mental activities of daily living (IADLs) was significantly associated
Personality is a relatively underresearched domain in suicide stud- with suicide case status (Conwell et al., 2010) and attempted suicide
ies among older people. Among the controlled studies, Harwood status (Tsoh et al., 2005) independent of the effects of physical and
et al. (2001) reported that levels of anankastic (obsessional) and mental health disorders.
anxious traits significantly distinguished suicide from natural
deaths, though personality disorder per se did not. Utilizing the Social factors
NEO Personality Inventory (Costa and McCrae, 1992), Duberstein Social factors related to suicide in older people can be broadly clas-
et al. (1994) found that of the five domains of personality measured sified into those related to stressful life events, erosion of social sup-
(neuroticism, extraversion, openness to experience (OTE), agreea- port and connectedness (Conwell et al., 2011), perceived financial
bleness, and conscientiousness), high neuroticism and low OTE difficulties, and religious/spiritual factors.
distinguished a group of individuals older than 50 years who com- Life events are often implicated as antecedents to suicide attempts
mitted suicide and an age- and gender-matched, living comparison and completed suicide in older people. Events related to physical
group. Low OTE was associated with reduced affective and hedonic illness, hospitalization, familial discord, interpersonal problems,
responses, a reduced range of interests, and a strong preference for retirement, bereavement, and social isolation are more typically
the familiar (Duberstein et al., 1994). associated with suicidal behaviour in later life (Heikkinen et al.,
Both low OTE and the similar attribute of anankastic personality 1995; Harwood et al., 2006). The association between stressful life
traits (as described by Harwood et al. (2001)) might render older events and completed suicide might be mediated through other
individuals more vulnerable to a negative impact from the chal- risk factors for suicide, in particular depression. A case-control PA
lenges of ageing and thus an escalated risk for suicide when stres- study in New Zealand (Beautrais, 2002) found that serious rela-
sors arise. More recently, the findings on OTE and neuroticism were tionship and financial problems distinguished older suicide victims
replicated in a study on Chinese older suicide attempters (Tsoh et and near-fatal suicide attempters from controls. The finding of the
al., 2005). In the same study, it was also found that as compared particular potency of familial discord has been replicated by other
to community comparison subjects, suicide decedents were less case-controlled PA studies in Sweden (Rubenowitz et al., 2001), the
extraverted, agreeable, and conscientious than their community US (Conwell et al., 2010), and Hong Kong (Chiu et al., 2004; Tsoh
counterparts. Furthermore, when suicide completers and attempt- et al., 2005).
ers were compared, it was found that the former group was distin- Bereavement from the death of a spouse is a relatively common
guished by their higher conscientiousness, agreeableness, and lower stressor in late life and is associated with elevated suicide risks after
neuroticism. These attributes might contribute to a higher chance the acute event (Harwood et al., 2006; Ajdacic-Gross et al., 2008).
of suicide attempters succeeding in their self-destructive act, and at On the other hand, widowhood per se is not uniformly identified to
the same time render their emotional distress less easily noticed by be a risk factor. A number of studies in Europe and Australia have
their relatives (Tsoh et al., 2005). Psychometric screening for early found an increased suicide risk in widowers (Yip, 1998; De Leo et
detection of impulsive behaviour and suicide proneness in suscep- al., 2001); however, the reverse is true for older Chinese women
tible groups of older people has been suggested to hold promise for in Hong Kong and Taiwan (Yip, 1998; Yip et al., 1998; Yeh et al.,
shaping future prevention strategies (De Leo, 2002). 2008). Furthermore, a study on all older suicides between 1987
and 2005 in Switzerland has shown that, for both genders, the risk
Physical health and functioning of suicide is elevated in the first week after bereavement. The risk
Several case-control PA studies have examined the independent remains elevated for men but virtually vanished after the first week
relation between common physical illnesses and suicide in older for women (Ajdacic-Gross et al., 2008). These intriguing findings
people. A study conducted in Sweden (Waern et al., 2002b) revealed deserve further study.
that visual impairment, neurological disorders, and malignant dis- Stress from caregiving to a spouse with chronic illness has been
ease were associated with elevated risk for suicide. A Canadian consistently identified in studies from the US (Cohen et al., 1998;
study also reported that chronic obstructive pulmonary disease Karch and Nunn, 2010) and Australia (Milroy et al., 1997) on
(COPD), congestive heart failure, and seizure disorder were signifi- homicide-suicide cases. This is alarming, given the rapid rise of
cantly related to suicide in older people, and that their effect was dementia prevalence with the ageing population.
cumulative (Juurlink et al., 2004). Studies conducted in Hong Kong Concerning social support, a case-controlled study conducted
(Chiu et al., 2004; Tsoh et al., 2005 demonstrated that COPD, arthri- in New Zealand demonstrated that a low level of social interac-
tis, bone fracture, and malignancy were associated with completed tion was found to be an independent risk factor for suicide, after
adjusting for physical and mental health variables (Beautrais, 2002). protective effects beyond general social connectedness (Rasic et al.,
Furthermore, Turvey et al. (2002) found that an increase in the 2009).
number of confidants was associated with a significantly reduced Moral and religious objections to suicide are reported to be
suicide risk in older adults. associated with less suicidal behaviour in depressed patients, and
The role of financial difficulties in older people’s suicide is relatively are proposed to act as a protective factor against suicidal behav-
underresearched. In psychological autopsies that have explored iour (Lizardi et al., 2008). This is also true for populations with
this factor, financial problems were present in around 13–19% of predominantly Islamic faith (e.g. Kok, 1988). So it is a surprising
the suicide decedents (Heikkinen et al., 1994; Rubenowitz et al., finding that, cross-nationally, there are significantly higher sui-
2001; Beautrais, 2002; Chiu et al., 2004; Tsoh et al., 2005). The cide rates in older men in Catholic countries compared to rates in
results in controlled studies have been divergent. In the studies by other countries, with a trend for similar findings among women,
Rubenowitz et al. (2001) and Beautrais (2002), monetary problems while Catholicism also condemns suicidal acts (Pritchard and
distinguished suicide completers from community controls, and Baldwin, 2000). Studies on negative religious coping or ‘religious
Conwell et al. (2010) also found lower annual income (less than US struggles’ in older persons (e.g. those with thoughts of illnesses
$35,000) distinguished the groups. However, no similar findings being signs of abandonment or punishment from God) might
were noted in the controlled study in Hong Kong (Chiu et al., 2004; potentially illuminate this apparently paradoxical finding (e.g.
Tsoh et al., 2005). Furthermore, in multivariate analyses, financial Pargament et al., 2001).
factors exerted no independent effect in the aforementioned stud- On the other hand, in Oriental belief systems, delineation of
ies by Rubenowitz et al. (2001), Beautrais (2002), and Conwell et religious denominations is much less clear-cut than the West. For
al. (2010). It is therefore likely that the influence of this stressor is instance, traditional Chinese religious beliefs have blended in herit-
mediated through independent factors such as depression. However, ages from Buddhism, Taoism, doctrines of Confucianism, regional
generalization from these studies is limited by the particular socio- folk religions, and ancestor worshipping (Cohen, 1992; Goossaert,
economic contexts in which these researches were conducted (e.g. 2005; Chamberlain, 2010). Moreover, collectively in traditional
the variable extent of welfare provisions for retired individuals). Chinese beliefs, ancestral spirits are commonly thought to dwell in
Financial factors may also act more at the societal than individual a different world and might impart positive influence on the filial
level; for instance, in a study by Shah et al. (2008), it was found descendents. Thus, during a perceived crisis, death may merely
that among the 87 global nations studied, elderly dependency ratios mean a different form of existence, or even an avenue to solve the
(EDR) were positively correlated with older suicide rates. This is of many impasses in this life through reincarnation (Goodrich, 1991;
concern, given the fast pace of worldwide population ageing, espe- Phillips et al., 1999).
cially in developing countries where income protection for senior Further studies in this area are obviously important, and hopefully
citizens is more likely to be limited. methodological advances will help to disentangle these complex
Research studies examining the possible influence of religious factors, leading to better interpretation of the data and application
and spiritual factors on suicide in older people are sparse, despite of the findings in evidence-based prevention programmes.
their potentially profound roles in the maintenance of emotional
health and coping with stress. In this domain, we should differenti- Other possible factors
ate between studies that examine the salience of religion towards Access to lethal means
an individual (or ‘religiousness’) and those that have looked at the Older adults are generally more ready to use more violent and
effect of affiliation to a specific religious denomination or particular potentially lethal methods for suicide attempts than younger age
set of spiritual beliefs. groups (Conwell et al., 2011). The prevalence of different methods
For the former, in a controlled study on persons aged 50 or over of suicide varies with the availability of lethal means across differ-
in the US, Nisbet and colleagues (2000) found that among 584 sui- ent sociocultural contexts. For instance, the use of a firearm is the
cides, participation in religious activities was probably associated commonest method of suicide by older men in the US (Conwell
with a lower risk of completed suicide. The odds for having never and Thompson, 2008). In the UK, suicidal older men mostly die
participated in religious activities were greater among suicide vic- by hanging and women by drug poisoning (Cattell, 2000). In Hong
tims compared with natural deaths, even after adjustment for sex, Kong, the most common methods of suicide in older people are
race, marital status, age, and frequency of social contacts. In a con- jumping from a height or hanging, while less violent methods (e.g.
trolled PA study and a controlled study on older suicide decedents asphyxia from charcoal burning, an increasingly common method
and attempters in Hong Kong (Chiu et al., 2004; Tsoh et al., 2005), in Asian Pacific cities) are infrequent among older people (Chiu et
it was found in both groups that those who considered religion as al., 2004; Chan et al., 2009). In rural China, the commonest method
salient in their life had a five-fold reduction in the odds for suicidal in older people is ingestion of organophosphate pesticides or rat
behaviours. Pargament et al. (2001), Van Ness and Larson (2002), poisons, followed by hanging (Li et al., 2009).
and June et al. (2009) have reported a generally protective effect Conwell et al. (2002b) found in a controlled PA study in the US
of religiousness on mental health in older Americans, possibly by that the presence of a firearm in the home was associated with a
way of conferring stronger reasons for living, which generally help two-fold risk for suicide, even after controlling for psychiatric ill-
to counter suicidal behaviours. Conwell et al. (2011) have postu- ness. Measures to restrict access to lethal means have been dem-
lated in a review article that religious and spiritual factors may be onstrated to have potential to lower overall suicide rates, especially
an important intrinsic constituent to the construct of social con- where the method in question is readily available, with examples
nectedness, and persons with low social connectedness are exposed of success reported following coal gas detoxification in Switzerland
to higher risk of self-destructive behaviours. However, there is and the UK, firearm restriction in Canada and Washington, DC, in
also new evidence that religiousness might impart independent the US (Mann et al., 2005), and legislations that limited the pack
size of paracetamol and salicylates sold over the counter in the UK Prevention and Intervention (‘Postvention’)
(Hawton et al., 2001).
In rural China, given the high rate of self-poisoning because of Suicide prevention interventions are usually classified as: univer-
the easy availability of these substances (Li et al., 2009), WHO has sal (targeting the entire population), selective (targeting sympto-
worked in recent years with Chinese governmental agencies to pro- matic or presymptomatic individuals or subgroups who have distal
mote the use of safer products, as well as safer methods of storage risk factors for suicide, or who have a higher-than-average risk for
of toxic substances (Chiu et al., 2012). There are now some pilot developing mental disorders), or indicated (targeting individuals
data showing that the use of ‘boxes with a lock’ for safe storage of who have detectable symptoms or other proximal risk factors for
pesticides in rural households is acceptable to the community in suicide) (Conwell and Thompson, 2008). Examples of universal
low-income countries (Konradsen et al., 2007). Further studies are measures include restriction to the means for self-harm, and pro-
necessary to examine the effectiveness of safe storage of pesticides motion of healthy ageing and destigmatization of mental illness.
in preventing suicide in rural communities. Improved access to rehabilitative care to promote functional inde-
pendence in individuals with chronic medical conditions could be
Neurobiological theories a selective measure for suicide prevention. Enhanced screening
Several studies have reported that depressed older people who for depression, availability of effective treatment of this condition,
attempt suicide could be distinguished from their nondepressed and adequate follow-up of suicide attempters (‘postventions’) are
counterparts by their poorer executive functioning on neuropsy- examples of indicated measures (Conwell and Thompson, 2008).
chological profiling (King et al., 2000; Keilp et al., 2001; Dombrovski A systematic review of suicide prevention strategies has found that
et al., 2008a, 2008b). This could possibly translate into poorer the most promising interventions are physician education, means
problem-solving capabilities or a higher propensity for impulsive restriction, and gatekeeper education, while other methods includ-
reactions in times of stress. In a community case-control study of ing public education, screening programmes, and media education
older adults aged 50 and over, clinical cerebrovascular risk factors need more research (Mann et al., 2005).
were significantly higher among suicide decedents than among To be effective, suicide prevention programmes must take into
community-dwelling age- and gender-matched comparison sub- account the known barriers to suicide prevention, which include
jects, after adjusting for age, sex, depression diagnosis, and func- the following:
tional status (Chan et al., 2007). Reviews of the neurotransmitter
◆ Older suicide victims are more determined in their acts, plan
disturbances mostly point to the possible age-related role of the ser-
their attempt with greater resolve, and display fewer external
otonergic, noradrenergic, and dopaminergic systems. For instance,
warnings of their intent compared to other age groups (Conwell
lower CSF levels of 5-HIAA were found in older depressed patients
and Thompson, 2008).
who attempted suicide compared with older depressed controls
(Conwell and Thompson, 2008). ◆ Older adults are less willing to communicate their suicidal intent
To sum up, there are preliminary findings that late-life suicide compared with younger people (Scocco and De Leo, 2002).
is associated with disrupted neural substrates that might play cru- ◆ The difficulty of timely identification of affective disturbances by
cial roles in impulse control, mood regulation, and cognition, and nonpsychiatric physicians when older people present with multi-
might represent exciting avenues for further understanding and ple somatic complaints (e.g. Heikkinen and Lonnqvist, 1995).
prevention of self-harm behaviours in older people.
◆ The unwillingness of older people to seek specialist psychiat-
ric care (and limited accessibility to such care in some nations)
Protective factors despite the high rate of psychiatric morbidities in people who kill
In contrast to the numerous studies on risk factors of suicide in themselves. However, most of them have seen a general practi-
older people, there is a relative paucity of studies on protective fac- tioner in the month before their suicide act (e.g. Chiu et al., 2004;
tors. The role of religious factors on suicide in older people has been Conwell et al., 2011).
examined in the section Social factors. Improving the social sup-
port to older people may be protective, as shown in a study that
◆ The therapeutic nihilism of some doctors towards older patients
an increase in the number of confidants was associated with lower with depression (e.g. Uncapher and Arean, 2000).
suicide risk in older people (Turvey et al., 2002). Two studies have Therefore it has been recommended that effective and compre-
attempted to improve resilience to suicidal behaviour. Lapierre et hensive prevention programmes should adopt a multicomponent
al. (2007) designed a programme for retired seniors based on a approach, with a combination of universal, selective, and indicated
cognitive-behavioural approach. It consisted of an 11-week work- interventions (Mann et al., 2005; Erlangsen et al., 2011).
shop and aimed to enhance meaning in the life of participants. Recent large-scale suicide prevention strategies in older people
Levels of depression, as well as suicidal ideas, were less in the inter- have adopted multifaceted or multilayered approaches, with par-
vention group compared with the control group. The other study ticular attention to mental health. The Telehelp/Telecheck service
by Heisel et al. (2009) consisted of a 16-week interpersonal psy- operated in Padua, Italy, and provided telephone-based outreach,
chotherapy progamme for older adults at increased risk of suicide, evaluation, and support services to older people who were iso-
to improve their social functioning so as to increase social support lated and functionally incapacitated. When assessed by De Leo et
and satisfaction. There was a reduction of depressive symptoms and al. (2002) over a decade later, a reduction of suicide rates among
lower scores on the Geriatric Suicide Ideation Scale (Heisel and women was identified. Oyama et al. reported that 5–10 years after
Flett, 2006) in the intervention group compared with the control the launching of a programme in rural Japan that combined uni-
group. Both of these studies are preliminary studies, but the results versal, selective, and indicated preventive interventions, the rela-
are encouraging. tive risk for suicide in older adults had dropped, with especially
a significant reduction (64–76%) in the risk for suicide in older Carroll, R., et al. (2012). Impact of the growing use of narrative verdicts
women (Oyama et al., 2005, 2006). Furthermore, in Hong Kong, by coroners on geographic variations in suicide: analysis of coroners’
the Elderly Suicide Prevention Programme (ESPP) was launched inquest data. Journal of Public Health, 34, 447–53.
Cattell, H. (2000). Suicide in the elderly. Advances in Psychiatric Treatment,
territory-wide in 2002. The initiative adopted a two-tiered, mul-
6(2), 102–8.
tifaceted, case-management model that offered prioritized and
Chamberlain, J. (2010). Chinese Gods: an introduction to Chinese folk religion,
intensive treatment of depressed or suicidal older people referred 3rd edition. Blacksmith Books, Hong Kong.
from primary care providers or other medical disciplines over a Chan, S.M., et al. (2006). Elderly suicide and the 2003 SARS epidemic
period of at least 6 months. An evaluation study suggested that the in Hong Kong. International Journal of Geriatric Psychiatry, 21(2),
programme was associated with a reduced suicide rate in women 113–18.
aged 85 and over, relative to the preintervention period (Chan et Chan, S.S., et al. (2009). What does psychological autopsy study tell us
al., 2011). about charcoal burning suicide—a new and contagious method in Asia?
A recent systematic review of 19 studies on suicide prevention Suicide, Life and Threatening Behaviour, 39(6), 633–8.
in older adults found that most of the interventions in these stud- Chan, S.S., et al. (2011). Outcomes of a two-tiered multifaceted elderly suicide
prevention program in a Hong Kong Chinese community. American
ies aimed to reduce suicide risk factors, e.g. depression screening
Journal of Geriatric Psychiatry, 19(2), 185–96.
and treatment (Lapierre et al., 2011). Most studies found an effect Chan, S.S., Lyness, J.M., and Conwell, Y. (2007). Do cerebrovascular risk
on reducing suicidal thinking and the suicide rate in older adults, factors confer risk for suicide in later life? A case-control study. American
but the programmes showed most benefits in women. The authors Journal of Geriatric Psychiatry, 15(6), 541–4.
suggested that strategies should aim to improve resilience, involve Chiu, H.F., Chan, S.M., and Lam, L.C. (2001). Suicide in the elderly. Current
community gatekeepers, use telecommunications to contact vul- Opinion in Psychiatry, 14, 395–9.
nerable older adults, and evaluate the effects of means restriction Chiu, H.F., et al. (2004). Elderly suicide in Hong Kong—a case-controlled
and physician education on suicide in older people (Lapierre et al., psychological autopsy study. Acta Psychiatrica Scandinavica, 109(4),
2011). 299–305.
Chiu, H.F., Chan, S.S., and Caine, E. (2012) Suicide prevention in the
Asia-Pacific region. Asia Pacific Psychiatry, 4 (1), 3–4 .
Conclusion Cohen, D., Llorente, M., and Eisdorfer, C. (1998). Homicide-suicide in older
persons. American Journal of Psychiatry, 155(3), 390–6.
Suicide prevention initiatives in older people have taken a great
Cohen, M.L. (1992). ‘Religion in a state society: China. In: Asia, case studies
stride forward in the past decade based on better understanding of in the social sciences : a guide for teaching (ed. M.L. Cohen), pp. 3–16.
the risk and protective factors pertaining to this great public health M.E. Sharpe, New York.
challenge. For further improvements to take place, sustained and Conwell, Y. and Thompson, C. (2008). Suicidal behavior in elders. Psychiatric
determined collaborations across clinicians, researchers, health Clinics of North America, 31(2), 333–56.
administrators, and policymakers are needed, given the complex, Conwell, Y., Duberstein, P.R. and Caine, E.D. (2002a). Risk factors for suicide
multidetermined nature of suicidal behaviours, a daunting task in in later life. Biological Psychiatry, 52(3), 193–204.
many nations for various reasons (De Leo, 2002). Conwell, Y., et al. (2002b). Access to firearms and risk for suicide in
Moreover, it is important to note that older men are generally middle-aged and older adults. American Journal of Geriatric Psychiatry,
10(4), 407–16.
at higher risk of suicide than women: in many countries, they use
Conwell, Y., et al. (2010). Health status and suicide in the second half of life.
more lethal means in their suicide acts, they are more suscepti- International Journal of Geriatric Psychiatry, 25(4), 371–9.
ble to the effects of bereavement and widowhood, and they have Conwell, Y., Van Orden, K., and Caine, E.D. (2011). Suicide in older adults.
responded less favourably to the most recent comprehensive sui- Psychiatric Clinics of North America, 34(2), 451–68, ix.
cide prevention programmes. Further research on the gender dif- Corcoran, P. (2009). The impact of widowhood on Irish mortality due
ferences of suicidal behaviours is urgently needed, to understand to suicide and accidents. European Journal of Public Health, 19(6),
the different underlying psychopathological mechanisms, and to 583–5.
adequately address the healthcare needs of older men, the largest Costa, P.T. and McCrae, R.R. (1992). Revised NEO Personality Inventory
group of completed suicides across the world. (NEO-PI-R) and NEO Five Factor Inventory (NEO-FFI) Professional
Manual. Psychological Assessment Resources, Odessa, Florida.
De Leo, D. (2002). Why are we not getting any closer to preventing suicide?
References British Journal of Psychiatry, 181, 372–4.
Ajdacic-Gross, V., et al. (2008). Suicide after bereavement: an overlooked De Leo, D., et al. (2001). Attempted and completed suicide in older subjects:
problem. Psychological Medicine, 38(5), 673–6. results from the WHO/EURO Multicentre Study of Suicidal Behaviour.
Alexopoulos, G.S., et al. (2005). Remission in depressed geriatric primary International Journal of Geriatric Psychiatry, 16(3), 300–10.
care patients: a report from the PROSPECT study. American Journal of De Leo, D., Dello, B.M. and Dwyer, J. (2002). Suicide among the elderly: the
Psychiatry, 162(4), 718–24. long-term impact of a telephone support and assessment intervention in
Alexopoulos, G.S., et al. (2009). Reducing suicidal ideation and depression northern Italy. British Journal of Psychiatry, 181, 226–9.
in older primary care patients: 24-month outcomes of the PROSPECT Dombrovski, A.Y., et al. (2008a). Sex differences in correlates of suicide
study. American Journal of Psychiatry, 166(8), 882–90. attempt lethality in late life. American Journal of Geriatric Psychiatry,
Beautrais, A.L. (2002). A case control study of suicide and attempted suicide 16(11), 905–13.
in older adults. Suicide, Life and Threatening Behaviour, 32(1), 1–9. Dombrovski, A.Y., et al. (2008b). Cognitive performance in suicidal depressed
Beautrais, A.L. (2006). Suicide in Asia. Crisis: Journal of Crisis Intervention elderly: preliminary report. American Journal of Geriatric Psychiatry,
and Suicide, 27(2), 55–7. 16(2), 109–15.
Bramness, J.G., Walby, F.A. and Tverdal, A. (2007). The sales of antidepressants Duberstein, P.R., Conwell, Y. and Caine, E.D. (1994). Age differences in the
and suicide rates in Norway and its counties 1980–2004. Journal of personality characteristics of suicide completers: preliminary findings
Affective Disorders, 102(1–3), 1–9. from a psychological autopsy study. Psychiatry, 57(3), 213–24.
Erlangsen, A., Zarit, S.H. and Conwell, Y. (2008). Hospital-diagnosed Konradsen, F., et al. (2007). Community uptake of safe storage boxes to
dementia and suicide: a longitudinal study using prospective, reduce suicide-poisoning from pesticides in rural Sri Lanka. BMC Public
nationwide register data. American Journal of Geriatric Psychiatry, Health, 7, 13.
16(3), 220–8. Lapierre, S., et al. (2007). Addressing suicidal ideations with the realization of
Erlangsen, A., et al. (2011). Key considerations for preventing suicide in older meaningful personal goals. Crisis, 28, 16–25.
adults. Crisis, 32,(2), 106–9. Lapierre, S. et al. (2011). A systematic review of elderly suicide prevention
Fountoulakis, K., et al. (2003). Unipolar late-onset depression: A programs. Crisis, 32, 88–98.
comprehensive review. Annals of General Hospital Psychiatry, 2(1), 11. Li, X., Xiao, Z., and Xiao, S. (2009). Suicide among the elderly in mainland
Goodrich, A.S. (1991). After death. In: Peking paper gods: a look at home China. Psychogeriatrics, 9(2), 62–6.
worship (ed. R. Malek), pp. 353–64. Steyler, Nettetal. Lindesay, J. (1991). Suicide in the elderly. International Journal of Geriatric
Goossaert, V. (2005). The concept of religion in China and the West. Diogenes, Psychiatry, 6(6), 355–61.
52(1), 13–20. Lizardi, D., et al. (2008). The role of moral objections to suicide in the
Hall, W.D., et al. (2003). Association between antidepressant prescribing and assessment of suicidal patients. Journal of Psychiatric Research, 42(10),
suicide in Australia, 1991–2000: trend analysis. British Medical Journal, 815–21.
326(7397), 1008. Mann, J.J., et al. (2005). Suicide prevention strategies. Journal of the American
Harwood, D., et al. (2001). Psychiatric disorder and personality factors Medical Association, 294(16), 2064–74.
associated with suicide in older people: a descriptive and case-control Milroy, C.M., Dratsas, M., and Ranson, D.L. (1997). Homicide-suicide in
study. International Journal of Geriatric Psychiatry, 16(2), 155–65. Victoria, Australia. American Journal of Forensic Medicine and Pathology,
Harwood, D.M., et al. (2006). Life problems and physical illness as risk 18(4), 369–73.
factors for suicide in older people: a descriptive and case-control study. Moscicki, E.K. (1997). Identification of suicide risk factors using epidemiologic
Psychological Medicine, 36, 1265–74. studies. Psychiatric Clinics of North America, 20(3), 499–517.
Hawton, K. and Harriss, L. (2006). Deliberate self-harm in people aged Nisbet, P.A., et al. (2000). The effect of participation in religious activities on
60 years and over: characteristics and outcome of a 20-year cohort. suicide versus natural death in adults 50 and older. Journal of Nervous
International Journal of Geriatric Psychiatry, 21, 572–81. and Mental Disease, 188(8), 543–6.
Hawton, K., et al. (1998). The psychological autopsy approach to studying Oyama, H., Watanabe, N., and Ono, Y. (2005). Community-based suicide
suicide: a review of methodological issues. Journal of Affective Disorders, prevention through group activity for the elderly successfully reduced
50(2–3), 269–76. the high suicide rate for females. Psychiatry and Clinical Neuroscience,
Hawton, K., et al. (2001). Effects of legislation restricting pack sizes of 59(3), 337–44.
paracetamol and salicylate on self poisoning in the United Kingdom: Oyama, H., et al. (2006). Outcomes of community-based screening for
before and after study. British Medical Journal, 322(7296), 1203–7. depression and suicide prevention among Japanese elders. Gerontologist,
Heikkinen, M., Aro, H., and Lonnqvist, J. (1994). Recent life events, social 46(6), 821–6.
support and suicide. Acta Psychiatrica Scandinavica Supplement, 377, Pampel, F.C. (1996). Cohort size and age-specific suicide rates: a contingent
65–72. relationship. Demography, 33(3), 341–55.
Heikkinen, M.E. and Lonnqvist, J.K. (1995). Recent life events in elderly Pargament, K.I., et al. (2001). Religious struggle as a predictor of mortality
suicide: a nationwide study in Finland. International Psychogeriatrics, 7, among medically ill elderly patients: a 2-year longitudinal study. Archives
287–300. of Internal Medicine., 161(15), 1881–5.
Heikkinen, M.E., et al. (1995). Age-related variation in recent life events Phillips, J.A., et al. (2010). Understanding recent changes in suicide rates
preceding suicide. Journal of Nervous and Mental Disease, 183(5), among the middle-aged: period or cohort effects? Public Health Reports,
325–31. 125(5), 680–8.
Heisel, M.J. and Flett, G.L. (2006). The development and initial validation Phillips, M.R., Liu, H., and Zhang, Y. (1999). Suicide and social change in
of the geriatric suicide ideation scale. American Journal of Geriatric China. Culture, Medicine and Psychiatry, 23(1), 25–50.
Psychiatry, 14(9), 742–51. Pritchard, C. and Baldwin, D. (2000). Effects of age and gender on elderly
Heisel, M., et al. (2009). Adapting interpersonal psychotherapy for older suicide rates in Catholic and Orthodox countries: an inadvertent neglect?
adults at risk for suicide: Preliminary findings. Professional Psychology: International Journal of Geriatric Psychiatry, 15(10), 904–10.
Research and Practice, 40,156–64. Rasic, D.T., et al. (2009). Spirituality, religion and suicidal behavior in a
Henriksson, S. and Isacsson, G. (2006). Increased antidepressant use nationally representative sample. Journal of Affective Discord, 114(1–3),
and fewer suicides in Jamtland county, Sweden, after a primary care 32–40.
educational programme on the treatment of depression. Acta Psychiatrica Rose, G. (1985). Sick individuals and sick populations. International Journal
Scandinavica, 114(3), 159–67. of Epidemiology, 14(1), 32–8.
June, A., et al. (2009). Religiousness, social support and reasons for living in Rubenowitz, E., et al. (2001). Life events and psychosocial factors in elderly
African American and European American older adults: an exploratory suicides—a case-control study. Psychological Medicine, 31(7), 1193–202.
study. Aging and Mental Health, 13(5), 753–60. Rubio, A., et al. (2001). Suicide and Alzheimer’s pathology in the elderly: a
Juurlink, D.N., et al. (2004). Medical illness and the risk of suicide in the case-control study. Biological Psychiatry, 49(2), 137–45.
elderly. Archives of Internal Medicine, 164(11), 1179–84. Scocco, P. and De Leo, D. (2002). One-year prevalence of death thoughts,
Karch, D.L. and Cole-Nunn, K. (2010). Characteristics of elderly and other suicide ideation and behaviours in an elderly population. International
vulnerable adult victims of homicide by a care giver: National Violent Journal of Geriatric Psychiatry, 17, 842–6.
Death Reporting System—17 US States, 2003–2007. Injury Prevention, Shah, A. (2007). The relationship between suicide rates and age: an analysis
16(Suppl 1), A9. of multinational data from the World Health Organization. International
Keilp, J.G., et al. (2001). Neuropsychological dysfunction in depressed suicide Psychogeriatrics, 19(6), 1141–52.
attempters. American Journal of Psychiatry, 158(5), 735–41. Shah, A. and Lodhi, L. (2005). The impact of trends in psychotropic
King, D.A., et al. (2000). A neuropsychological comparison of depressed prescribing on the method of suicide in the elderly. Medicine, Science
suicide attempters and nonattempters. Journal of Neuropsychiatry and and the Law, 45(2), 115–20.
Clinical Neurosciences, 12(1), 64–70. Shah, A., et al. (2007). Elderly suicide rates: cross-national comparisons and
Kok, L.P. (1988). Race, religion and female suicide attempters in Singapore. association with sex and elderly age-bands. Medicine, Science and the
Society of Psychiatry and Psychiatric Epidemiology, 23(4), 236–9. Law, 47(3), 244–52.
Shah, A., Padayatchi, M., and Das, K. (2008). The relationship between elderly Van Orden, K. and Conwell, Y. (2011). Suicides in late life. Current Psychiatry
suicide rates and elderly dependency ratios: a cross-national study using Reports, 13(3), 234–41.
data from the WHO data bank. International Psychogeriatrics, 20(3), Waern, M., et al. (2002a). Mental disorder in elderly suicides: a case-control
596–604. study. American Journal of Psychiatry, 159(3), 450–5.
Tsoh, J., et al. (2005). Attempted suicide in elderly Chinese persons: a Waern, M., et al. (2002b). Burden of illness and suicide in elderly people:
multi-group, controlled study. American Journal of Geriatric Psychiatry, case-control study. British Medical Journal, 324(7350), 1355.
13(7), 562–71. Wasserman, I.M. (1989). Age, period and cohort effects in suicide behavior
Turvey, C.L., et al. (2002). Risk factors for late-life suicide: a prospective, in the United States and Canada in the 20th century. Journal of Aging
community-based study. American Journal of Geriatric Psychiatry, 10(4), Studies, 3(4), 295–311.
398–406. World Health Organization (2011). Country reports and charts. <http://www.
Uncapher, H. and Arean, P.A. (2000). Physicians are less wiling to treat who.int/mental_health/prevention/suicide/country_reports/en/index.
suicidal ideation in older patients. Journal of the American Geriatrics html>, (accessed 30.12.2011).
Society, 48, 188–92. Yeh, J.Y., et al. (2008). Does marital status predict the odds of suicidal death
United Nations (2009). World population to exceed 9 billion by 2050: developing in taiwan? A seven-year population-based study. Suicide, Life and
countries to add 2.3 billion inhabitants with 1.1 billion aged over 60 and Threatening Behaviour, 38(3), 302–10.
1.2 billion of working age. Press release March 2011. Population Division/ Yip, P.S. (1998). Age, sex, marital status and suicide: an empirical study of
DESA, UN, New York. East and West. Psychological Reports, 82(1), 311–22.
Van Ness, P.H. and Larson, D.B. (2002). Religion, senescence, and mental Yip, P.S., Chi, I., and Yu, K.K. (1998). An epidemiological profile of elderly
health: the end of life is not the end of hope. American Journal of Geriatric suicides in Hong Kong. International Journal of Geriatric Psychiatry,
Psychiatry, 10(4), 386–397. 13(9), 631–7.
CHAPTER 44
Manic syndromes in old age
Akshya Vasudev
Classification Issues to be considered for a better understanding of the epidemiology of

the illness.
Bipolar disorder is a chronic disease of abnormal mood and is char-
acterized by episodes of either elevated mood or depression, or,
Age
much less frequently a mixed affective presentation of both elevated
mood and depression. To satisfy a clinical diagnosis of bipolar dis- Data from epidemiological studies suggest that bipolar disorder in
order there should be at least one hypomanic or manic episode, the general adult population has a prevalence rate of around 1% in
and the abnormal mood episodes (hypomanic/manic or depressed) the community (Regier et al., 1993). This may be an underestimate.
should have a detrimental effect on the social and occupational Evidence is accumulating that bipolar disorder and schizophrenia
functioning of the individual. Though the diagnostic criteria of this illnesses are two ends of a spectrum and often there is an overlap of
disorder are well demarcated in the younger adult population in symptoms and progression (Laursen et al., 2009). Indeed, if screen-
both the ICD-10 and DSM-IV (WHO, 1992; American Psychiatric ing tools like the Mood Disorder Questionnaire (MDQ) are used
Association, 1994), the same cannot be said for bipolar disorder in for estimation of prevalence, 4% of the population might be consid-
older people. The DSM-5, expected to be published in May 2013, is ered to have had a lifetime episode of hypomanic or manic symp-
also not likely to address age-related differences in presentation of toms (Hirschfeld et al., 2003).
bipolar disorder. This may be because bipolar disorder in older peo- In contrast, the prevalence and incidence rate of mania in old
ple is probably a heterogeneous illness, hence attempts to classify it age in the community is not very well established. This is because
remain controversial. In deference to this controversy, we choose methodological factors confound the data. Some of the concerns are
the broader term of ‘manic syndromes’ in old age as the title of our inherent in identifying individuals with mania in the community as
chapter. We recognize that recently there are reasons to believe that they are not readily amenable to interview or cooperation with sur-
it can be broadly divided into two main groups, the late-onset bipo- veys. Bipolar illness usually affects people by the age of 30 and it has
lar (LOB) patients and the early-onset bipolar (EOB), and we shall been estimated that 90% of cases are aged less than 50 when they
describe these in more detail later. Perhaps the most important have their first episode (Hirschfeld et al., 2003). This implies that
evidence concerning differences between younger adult and older 10% of bipolar patients will develop their illness after the age of 50
bipolar patients comes from genetic studies. Although bipolarity and one study did find that around 10% of a population sample had
is strongly influenced by genetic factors in the adult population a first incidence of mania after the age of 60 (Kennedy et al., 2005).
(Goodwin, 1990), there is less evidence of such a relationship in the Epidemiological studies conducted in the older population in the
geriatric population. In this chapter, we shall be elaborating differ- community using strict diagnostic criteria have found prevalence
ences between mania in younger adults and older people, between rates varying from 0.08% to 0.25%. Other prevalence studies also
LOB and EOB, discuss aetiopathological findings, and offer sugges- suggest that the figure for lifetime prevalence of bipolar disorder
tions for management. in older people in the community is around 0.1% (Weissman et al.,
Shakespeare (through Juliet) asks ‘What’s in a name? That which 1988; Unutzer et al., 1998; Hirschfeld et al., 2003). The reasons for
we call a rose by any other name would smell as sweet’. Whilst it is differences in prevalence rate between the general adult population
true that reality matters, not the name used to denote it, the oppo- (around 1%) and that in older people (around 0.1%) remains a mat-
site is not the case and using the same name for different things ter of controversy and no epidemiological factors are obvious for the
does matter (Vasudev and Thomas, 2010). Since older people discrepancy. Some suggest the influence of a relatively high mortal-
with LOB disorder seem to have a different symptom profile with ity rate from natural causes and from suicide, as well as evidence
a different epidemiology and different aetiologies compared with for ‘burn out’ of the disorder over a long-term course (Winokur,
younger people, it appears ‘bipolar disorder’ may not be an appro- 1975; Snowdon, 1991). These studies were, though retrospective in
priate name for this symptom constellation in older people. nature, reflecting the difficulty in following up patients with bipolar
disorder over their lifespan.
However, in selected populations such as nursing homes (9.7%;
Epidemiology and Clinical Presentation Koenig and Blazer, 1992) and inpatients (varying from 8 to 10%;
There is evidence suggesting that the age of the patient as well as Depp and Jeste, 2004), there are significantly higher preva-
the age of onset of the disorder are two different variables that need lence rates. The higher prevalence in such populations perhaps
is a reflection of the course of the illness; when older individu- that LOB seem to have higher medical and neurological comor-
als suffering from bipolar disorder do have an affective relapse bidity (Depp and Jeste, 2004). Some of these include dementias
they need higher levels of care and cannot be managed in the (Himmelhoch et al., 1980), neurological conditions (Shulman
community. et al., 1992), and cerebrovascular conditions (Subramaniam et
Also, until recently, it was believed the bipolar disorder had a al., 2007). LOB patients have also been shown to have fewer and
bimodal frequency distribution with a peak of episodes in early milder manic symptoms compared to the EOB or the young manic
adulthood and another peak in late life (Cassidy and Carroll, patient and a tendency to have irritable mood rather than elated
2002). However, this view has been challenged as data also suggest mood (Sajatovic, 2002). Kessing et al. (2006) used Denmark’s
a unimodal frequency distribution with no such peak later in life nationwide registry to describe a cohort of inpatients with LOB
(Almeida and Fenner, 2002; Kennedy et al., 2005). (onset over 50 years of age). Compared with a cohort of EOB
patients, the LOB had a lower prevalence of psychotic symptoms
Age at onset associated with mania, but they presented with more psychosis
Epidemiological data are better defined in late-life bipolar disor- associated with their depression. In a recent longitudinal study
der with regards to the age of onset of the illness. It is thought that comparing the outcome of LOB and EOB patients followed up
this variable helps distinguish the subtypes of mania and hence after a manic or a mixed affective relapse, it was reported that
lead to an improved understanding of pathogenesis and aetiology LOB patients had a quicker remission of their illness and a greater
(Young and Klerman, 1992; Leboyer et al., 2005). The older person proportion were discharged from an inpatient facility (Oostervink
with bipolar disorder may be considered to be of two types (see et al., 2009).
Fig. 44.1). The first group may be referred to as LOB, who have the In terms of disease characteristics, older bipolar patients (EOB
first episode of the affective disorder for the first time late in their and LOB considered together) seem to have a larger time gap
life; most studies consider this to be greater than 50 years of age. between a depressive episode and a manic one. In a prospective
The other is the EOB, who when older are ‘graduates’ of the disor- study (Jeremy and Robin, 1990), the mean time between the first
der, having had the onset of the illness at a younger age (less than episode of depression and the onset of mania was 17 years in the
50 years of age). The demarcation is useful as clinically these two older group versus 3 years in the younger group. This study also
groups are different. found that more older than younger manic patients had suffered
It has been argued that EOBs have a high familial rate, while three or more depressive episodes before their first manic episode.
LOBs do not show this pattern consistently (Depp and Jeste, 2004). An interesting finding was that the older manic patient was more
The estimated rates of familial penetrance in the older group per se likely to relapse into depression after mania.
vary between 24% and 88% (Glasser and Rabins, 1984; Shulman et These findings indicate that older people with bipolar disorder
al., 1992; Hays et al., 1998). The reason for this variance could be have a different clinical pattern, involving more depression and pos-
related to methodological issues such as the rigor of investigation sibly milder manic symptomatology, especially in the LOB group.
and the criteria used for a positive family history of bipolar disor- This could be a reflection of differences in comorbidity as discussed
der (e.g. whether only first-degree relatives are considered). Within in the next section Comorbidities.
the older group, it has been estimated that there is twice the famil-
ial penetrance rate in the EOB group compared to the LOB group
(Stone, 1989; Hays et al., 1998). Comorbidities
Using a cut-off point of 49 years, one study found the LOB Comorbidities may include psychiatric and physical disorders.
group of older patients to have more psychotic features and more Bipolar disorder in adult outpatients is often comorbid, with vari-
cerebrovascular risk factors (Wylie et al., 1999). Using a cut-off ous axis I conditions, including lifetime and current substance use,
of 50 years for age at onset, another study (Hays et al., 1998) also anxiety, and eating disorders (McElroy et al., 2001). Epidemiological
found more vascular comorbidity in older bipolar patients with studies such as this, however, have tended to include younger adults
late onset (mean age 74 years). Previous reviews have also shown as well as older people in the same group; in only some studies has
Manic syndromes in old age
Late onset bipolar Early onset bipolar
• Weaker family history • Strong family history

• High medical and neurological co-morbidity • High medical and neurological co-morbidity
• Fewer manic symptoms with earlier resolution • Manic and mixed symptoms predominate
• More psychotic features in context of depression • Longer inpatient stay
• Shorter inpatient stay
Fig. 44.1 Conceptual framework for differentiating manic syndromes in old age based on age of onset of bipolar disorder
CHAPTER 44 manic syndromes in old age 583
subgroup analysis been attempted. Cassidy et al. (2001) found a life- has not been well investigated. In fact, it might be that cognitive
time substance abuse rate of around 29% in a subgroup of older bipo- deficits in patients with bipolar disorder might be different from
lar patients. This contrasted with a substance abuse rate of around those who have MCI. One recent study compared cognitive deficits
48% in the younger adult bipolar population. Interestingly, the older in older adults with bipolar disorder (mean age 63 years) and those
bipolar patient tended to have more hospital admissions compared with MCI. They found that patients with MCI were more impaired
to the younger bipolar. In a more recent study (Goldstein et al., in verbal memory, whereas BD patients showed more deficits in
2006), the 1-year prevalence rate in late-life bipolar was found to be attention, motor initiative, calculation, and verbal abstraction
highest for alcohol abuse (38.1%), followed by panic attack (11.9%), (Silva et al., 2009).
generalized anxiety disorder (9.5%), and dysthymia (7.1%). A study by Gildengers et al. (2008a) has attempted to estimate
There are fewer data on the prevalence of axis II conditions other medical illness burden in patients with late-life bipolar dis-
in late-life bipolar. In a study on outpatients as well as inpatient order. They found that patients greater than 60 years of age with
populations of late-life affective disorders, 63% had a personality late-life bipolar had similar total medical morbidity, as measured
disorder based on their responses to a structured interview for per- by the Cumulative Illness Rating Scale-Geriatrics (CIRS-G), to
sonality. There was no difference in the prevalence rate between the age-matched depressed patients. Those with bipolar disorder did
depressed and the bipolar group (Molinari and Marmion, 1995). have a higher BMI and increased burden of endocrine/metabolic
This is similar to a lifetime odds ratio of having an axis II diagnosis and respiratory disease as measured by the CIRS-G subscores.
of 10.6 (95% CI 9.2–12.3) (equivalent to a percentage prevalence of Because CIRS-G scores in late-life depression are increased com-
53% by imputation assuming 5% prevalence in the adult popula- pared with nonpsychiatric controls, this study suggests the same is
tion) in the general adult bipolar population in a large community true for late-life bipolar disorder, although direct comparisons are
survey (Grant et al., 2005). needed (Baldwin, 2008).
As regards physical illness, a number of studies have shown an Another interesting finding is the link of affective disorders and
increased prevalence of neurological illnesses, as defined variously, dementia. Though the risk of developing dementia, particularly
but especially including dementia and cerebrovascular disease in Alzheimer’s disease, increases with age, this risk appears to increase
patients with LOB (Depp and Jeste, 2004). A large longitudinal even more in patients with affective disorders. The link between
study confirmed that the elderly bipolar have double the odds of unipolar depression and dementia is now well established, with
developing stroke compared to age matched controls over a period conservative estimates of an increased odds ratio of two times of
of 6 years (Lin et al., 2007). These findings appear similar to those developing a dementia syndrome in patients with late-life depres-
in unipolar disorder (Thomas et al., 2004). sion (Ownby et al., 2006; Gualtieri and Johnson, 2008; Brommelhoff
In a similar vein, the context of poststroke mania, or ‘secondary et al., 2009). Though there are fewer studies in patients with late-life
mania’ as described in psychiatric literature, tends to be similar to bipolar disorder, there is emerging evidence that patients with EOB
the concept of ‘disinhibition syndrome’ as preferred by neurolo- do present with cognitive impairment later in life (Tsai et al., 2007).
gists. Perhaps this is a reflection of the same pathology but labelled In this study, a multiple regression model was used for prediction
differently by different group of clinicians. It is now relatively well of cognitive impairment: years of education and the age at the last
established that normal mood is dependent on the integrity of the manic/hypomanic were the most important predictor variables;
frontal, limbic, and basal ganglia circuitry. To understand the pathol- other important predictors were age at the first depressive epi-
ogy observed in secondary mania, the relevant neuroanatomical sode and the presence of a first manic episode before the age of
domains are: frontal lobes modulate motivational and psychomotor 40 years. In another study of 33 late-life bipolar patients who were
behaviour; limbic connections modulate emotions; while the bio- followed-up longitudinally over a period of 3 years (Gildengers et
genic amine nuclei in the hypothalamus, amygdala, and brainstem al., 2009), there was evidence of more cognitive dysfunction and
modulate instinctive behaviours. It is therefore understandable more rapid cognitive decline than expected for their age and edu-
that any loss in these neural circuits can lead to mania (Starkstein cation. Consequently, it has been hypothesized that better control
and Robinson, 1997). In addition, it has been seen in a number of of bipolar disorder might reduce or delay dementia. One study
case reports that the right part of the cerebral cortex seems to be examining the use of mood stabilizers in preserving cognitive func-
more affected in patients with secondary mania (Braun et al., 1999). tion (Rybakowski et al. 2009) found a beneficial effect of lithium
Other authors also consistently argue that right-sided cerebral prophylaxis on executive cognitive function. There have also been
lesions subsequent to head injuries or other causes of organic brain interesting results from a large 10-year follow-up study (1995–
damage lead to ‘disinhibition syndrome’, similar to what is found in 2005) conducted in Denmark (Kessing et al., 2008). Two groups
‘ secondary mania’ (Starkstein et al., 1990; Strakowski et al., 1994; were compared: 16,238 persons who had purchased lithium at least
Verdoux and Bourgeois, 1995; Steffens and Krishnan, 1998). once in the community (implying a possible diagnosis of bipolar
Cognitive impairment is often found in patients with bipolar dis- disorder) and 1,487,177 persons from the general population who
order even in the euthymic stage in both the younger adult popula- had not purchased lithium. Patients who purchased lithium at least
tion (Martinez-Aran et al., 2004) and the older age group (Dhingra once had an increased rate of dementia compared with persons not
and Rabins, 1991). For instance, in a study of older bipolar patients exposed to lithium (relative risk, 1.47; 95% CI, 1.22–1.76), providing
who were euthymic, more than half scored one or more standard further evidence that bipolar disorder increases the risk of demen-
deviations below the mean of age- and education-matched com- tia. For persons who continued to take lithium, the rate of dementia
parison subjects on cognitive screening instruments, the MMSE, decreased to the same level as the rate for the general population,
and the Mattis Dementia Rating Scale (Gildengers et al., 2004). This suggesting a protective effect of lithium. More disappointing are
might suggest that a number of older bipolar patients have a comor- findings from another small study (Gildengers et al., 2008b). Twelve
bid diagnosis of mild cognitive impairment (MCI). This hypothesis older adults with bipolar disorder were prescribed donepezil for 12
weeks. There was no improvement in either ADL score or cognitive areas controlling emotions and mood, as well as its role in regu-
test scores at the time of completion of the study. lating memory. Surprisingly, increased temporal lobe volume has
These findings suggest that late-life bipolar disorder patients are been reported in some structural MRI studies in patients with adult
at increased risk of stroke, cognitive impairment, and dementia. bipolar disorder (Jones et al., 2009). However, there are contradic-
There are some data on a possible protective effect of lithium for tory reports of the extent of brain volume changes in late-life bipolar
preventing dementia, but good quality intervention studies are disorder. In a study (Sarnicola et al., 2009) comparing grey matter
needed to investigate this and to determine if better control of vas- volume, white matter volume, and total brain volume in 71 older
cular risk factors improves outcome. bipolar patients compared with 82 age-matched controls, there was
no evidence of greater volume changes in the bipolar group.
Prognosis Brain connectivity
Suicide has been consistently shown to be one of the important Diffusion tension imaging (DTI) is a relatively new neuroimaging
causes of mortality in patients with affective disorders. In a large technique that allows demarcation of microstructural modifica-
longitudinal study of patients suffering from affective disorder tions in white matter tracts. The nascent nature of this technique
admitted to an inpatient unit who were followed up for more than is reflected in the paucity of published literature using it in bipo-
34 years, it was observed that those patients who were on psycho- lar disorder. A recent review of DTI in bipolar disorder in adult
tropics (including antidepressants, neuroleptics, and lithium) had populations cited 10 published studies in the adult population
significantly reduced completed suicide rates compared to those (Vederine et al., 2011). The analysis suggests two significant clusters
not treated (Angst et al., 2002). Indeed, in a retrospective analysis of decreased connectivity on the right side of the brain. The first
of cases of older bipolar patients who had a suicide attempt, usage was located in the right white matter, close to the parahippocampal
of antidepressants and mood stabilizers was noted to have reduced gyrus. Four of the ten studies included contributed to this cluster.
suicide attempt rates compared to age- and sex-matched controls The second cluster was located close to the right anterior and sub-
(Aizenberg et al., 2006). genual cingulate cortex. In the only published study (Haller et al.,
2011) in older bipolar (LOB and EOB together) patients conducted
Aetiopathology on only 19 patients, there was significantly decreased connectivity
in the ventral part of the corpus callosum.
Analogous to the vascular depression hypothesis in late life, the
These findings show that our knowledge of the aetiopathology of
hypothesis that mania in late life may have a vascular cause has
late-life bipolar disorder is limited, but the best evidence supports a
generated much interest (Wijeratne and Malhi, 2007). In a study of
role of cerebrovascular disease.
patients with late-life bipolar disorder, higher Framingham Stroke
Risk Score was found in LOB versus EOB (Subramaniam et al., 2007).
Also, in a sample of older bipolar patients (n = 119), more vascu- Management
lar risk factors predicted poorer cognitive performance (Schouws This section will focus on those features of the management of
et al., 2010). Cerebrovascular disease is also reflected in the extent bipolar disorder that are distinct for older people. Most fundamen-
of brain white matter hyperintensities (WMH) as revealed by neu- tal, given the high prevalence of neurological morbidity in late-life
roimaging. There are consistent findings of an increase in WMH in bipolar disorder, is the need to search assiduously for localizing
patients with bipolar disorder (Lloyd et al., 2009) and even more neurological signs and symptoms. A careful history, being alert to
so in LOB (Tamashiro et al., 2008). Tamashiro et al. (2008) found evidence of head injury, cerebrovascular disease, and related risk
that there were more WMH in deep frontal, parietal, and putamen factors are essential in light of our current understanding of the
in LOB patients compared to age-matched older people with EOB. illness. Indeed, neuroimaging may be considered as an important
There is evidence that in unipolar disorder in older people, WMH component of the investigation of the older manic patient, espe-
are due to hypoxia-ischaemia (Thomas et al., 2002), but there have cially LOB (Van Gerpen, et al., 1999).
been no studies in bipolar disorder. The aetiology of WMH may be The evidence base for psychotherapeutic treatments in the
a result of deficits in vascular perfusion (Thomas et al., 2002). younger bipolar patient group is relatively strong, with different
Though it is relatively well established that traumatic brain injury approaches being effective in different phases of the illness. Family
in the younger adult population can cause secondary mania as a therapy, interpersonal therapy, and systematic care appear to be
neuropsychiatric sequela, the evidence for this association in older effective in preventing recurrences when initiated after an acute
people is much weaker. There are only a few case reports of second- episode, whereas cognitive-behavioural therapy and group psy-
ary mania in this population as a result of either anoxic encepha- choeducation appear to be most effective when initiated during
lopathy (Ku et al., 2006) or thalamic damage (Lopez et al., 2009). a period of recovery (Miklowitz, 2008). However, there has been
Although it seems reasonable that brain injury would have similar less investigation of the effectiveness of nonpharmacological treat-
or worse effects in older people, good studies are needed to dem- ment of late-life bipolar disorder. While there are some encour-
onstrate this. aging findings from psychosocial interventions trials (Fagiolini
et al., 2009), no randomized controlled trials (RCTs) have been
Brain volume changes reported.
Structural magnetic resonance imaging (MRI) can be used to Pharmacological treatment of any disorder in older patients is
estimate brain volume. The temporal lobe is one of the important challenging because of pharmacokinetic and pharmacodynamic
brain areas involved in the aetiopathogenesis of affective disorder changes associated with ageing, as well as an increased risk of
because of its cortical and subcortical connections with other brain drug interactions (see Chapter 13). In view of these challenges, it
is noteworthy that a ward-based study of older psychiatry patients replacement or else had elevated levels of thyroid stimulating hor-
found that 96% of prescriptions had a potential for drug–drug mone (Head and Dening, 1998), and another study of 1705 new
interactions, with an average of eight drugs prescribed for each lithium users over 65 years of age found treated hypothyroidism
patient (Vasudev and Harrison, 2008). in almost 6% of the sample, twice the prevalence expected among
a mixed-age population (Shulman et al., 2005b). Lithium can also
Mood Stabilizers cause frank encephalopathy when taken as an overdose or when
combined with various neuroleptics. Cautious slow withdrawal is
Lithium usually found to alleviate the symptoms (Swartz and Dolinar, 1995;
Lithium continues to be a commonly used mood stabilizer in old Boora et al., 2008).
age, although some clinicians and patients are reluctant to use it Drug interactions continue to be an ongoing concern with lith-
as first-line because of concerns regarding frequent blood level ium, particularly when thiazide diuretics are prescribed as these
monitoring, nephrotoxicity, and the narrow therapeutic window. can significantly lower lithium clearance and thereby put the
Evidence suggests it may help in reducing progression to demen- patient at risk of toxicity by increasing the serum lithium levels.
tia in late-life bipolar disorder (Nunes et al., 2007; Kessing et al., Other medications of concern include the angiotensin-converting
2008). These findings suggest that clinicians may become more enzyme (ACE) inhibitors and nonsteroidal anti-inflammatory
favourable towards the use of lithium in older people. In addition, drugs (NSAIDs) such as indomethacin. In an observational study
the BALANCE trial, a pragmatic RCT in working-age adults, has of 10,615 older lithium users, 3.9% had been admitted to hospital
revisited lithium again and concluded that for adults with bipolar I for lithium toxicity. In this study, initiation of a loop diuretic like
disorder, for whom long-term therapy is clinically indicated, both furosemide or an ACE inhibitor significantly increased the risk of
combination therapy with lithium plus valproate and lithium mon- lithium toxicity, while neither thiazide diuretics nor NSAIDs were
otherapy are more likely to prevent relapse than is valproate mono- independent risk factors (Juurlink et al., 2004).
therapy (Geddes et al., 2010). Unfortunately though, there are still
no RCTs of this agent in the pharmacological treatment of older Anticonvulsants
bipolar patients. There are, however, several open-label trials and Valproic acid or its congener, divalproex, are now frequently used in
retrospective reports that suggest that even older, frail patients can the treatment of bipolar disorder in the younger adult population.
be safely and effectively treated (Himmelhoch et al., 1980; Shulman Some evidence suggests that valproate therapy is now used more
and Post, 1980; Chen et al., 1999; Fahy and Lawlor, 2001; Gildengers frequently for older bipolar patients than even lithium (Shulman
et al., 2005). Lithium must be used with caution in the geriatric et al., 2003; Sajatovic et al., 2004). There are no RCT data available,
population as it is eliminated exclusively by the kidneys (Hardy et but there are a number of published case reports and case series
al., 1997). It has been found that lithium is excreted in older people (Risinger et al., 1994; Gildengers et al., 2005). These studies indi-
at a rate approximately half that of younger patients(Hardy et al., cate that valproate is an effective and well-tolerated mood stabilizer
1997). It is therefore recommended that in old age, only half the in the older population. Though there are recommendations that
adult dose should be prescribed. valproate should be uptitrated quickly for acute management of
Another factor to consider is the increase in sensitivity to lithium mania in the adult population, there are no such recommendations
by age. Adverse reactions and toxicity levels have been reported in in the older population. This is possibly because of the higher inci-
older people even with usual adult serum levels (Roose et al., 1979; dence of adverse effects even at therapeutic dosages. Principal side
Murray et al., 1983). However, under well-controlled conditions in effects include sedation and gastrointestinal disturbance, which
a specialized geriatric clinic, lithium is reported to be safe and well can be modulated by dosage reduction (Shulman and Herrmann,
tolerated (Parker, 1994). 1999). Indeed, valproate did not fare better than lithium in a rela-
One clinical question that old age psychiatrists sometimes face is tively large study comparing differences in hospitalization rate for
the long-term effect of lithium on the kidneys. Lithium is known to delirium in 2422 lithium users compared to 2918 valproate users
be nephrotoxic, especially if it is given over a long period of time. It (Shulman et al., 2005a). Use of valproate is associated with a rela-
has been known to cause nephrogenic diabetes insipidus through tively high incidence of thrombocytopenia, an often underrecog-
various mechanisms, including alterations in purinergic signalling nized side effect of this medication, which may occur frequently
(Zhang et al., 2009), disturbance in epithelial sodium channel func- in older people (Trannel et al., 2001). There may be merit in using
tioning in the collecting ducts (Kortenoeven et al., 2009), and dys- valproate in the treatment of rapid cycling illness in old age (Gnam
regulation of renal aquaporins and acid-base transporters(Nielsen and Flint, 1993), though as there are no RCTs, one must be cautious
et al., 2008). Perhaps in the future, newer treatments will be avail- in extrapolating results to the clinic.
able to counteract the effects of lithium on the kidneys, but at the Valproate is highly protein bound. There is a risk of higher
moment no specific recommendations are available for the man- adverse events when consumed at higher dosages, which may be
agement of lithium-induced nephrotoxicity, although amiloride has related to higher total as well as unbound drug levels in the blood.
been found to be beneficial to some extent (Bedford et al., 2008). This may result from complex protein binding. It has been shown
Prudence would suggest that the clinical management of such a that the protein binding of valproate decreases as the serum con-
patient should be done collaboratively with a nephrologist. centration increases (Felix et al., 2003). The drug is also an inhibi-
Additionally, lithium can cause a wide range of systemic adverse tor of CYP2D6 in the liver and hence has the potential for drug
reactions, including those on the central nervous system. Lithium interactions. It has been found to inhibit the metabolism of tri-
induces tremors, aggravates parkinsonian tremor, and causes spon- cyclic antidepressants and displace diazepam from protein bind-
taneous extrapyramidal symptoms. One community study found ing sites, thus increasing their plasma concentrations (Janicak,
that almost a third of older patients on lithium were on thyroid 1993).
Carbamazepine, a mood stabilizer preferred in the past, is In the European multicentre naturalistic follow-up study of 2761
still recommended in various guidelines for the management of patients, EMBLEM (Oostervink et al., 2009), it was found that LOB
bipolar disorder in younger adults. The British Association for patients tended to be maintained on typical antipsychotics, lithium,
Psychopharmacology recommends that it could be used for acute and anticholinergics, much more than EOB patients. However, after
mania that is mild in severity and for prevention of relapse in bipo- an episode of manic or mixed affective relapse there was an increase
lar I patients (Goodwin, 2009). Its clinical usage in the geriatric in prescription of atypical antipsychotics and a consequent decrease
population is low, perhaps because of its risk of drug interactions in typical antipsychotic use. There was a high use of antidepres-
and high neurotoxicity. Therapeutic target serum levels might be sants in the older group who were rapid cycling (40%) compared
a bit lower for the geriatric population and some authors suggest to non-rapid cycling older patients. There was a steady increase in
keeping serum levels below 9 μg/ml (Young, 1996). the prescription of antidepressants across all groups during active
Amongst other mood stabilizers, gabapentin has been found to treatment.
be beneficial in some case reports. Two case series used gabapen-
tin in a subsample of older patients (Ghaemi et al., 1998; Cabras Conclusion
et al., 1999). There is some evidence of lamotrigine’s efficacy in
bipolar disorder in those greater than 55 years of age; (Marcotte, Late-life bipolar disorder, especially LOB disorder, appears to be a
2004). In an open-label trial of treatment of refractory bipolar dis- different entity from bipolar disorder in the younger population.
order, a number of older patients appeared to tolerate lamotrigine Late-life bipolar disorder is frequently associated with vascular
(Calabrese et al., 1999). In a subanalysis of older subjects enrolled changes in the brain and these may be of aetiological significance.
in a placebo-controlled trial of maintenance therapy for bipolar I There is also cognitive impairment associated with late-life bipo-
disorder, 33 patients were treated with lamotrigine (Sajatovic et al., lar disorder and there is some evidence that lithium might prevent
2005), and it was found that lamotrigine was significantly better further deterioration in cognition. Good quality randomized con-
than placebo at delaying time to intervention for any mood disor- trolled trials of psychotropics in both acute and maintenance stages
der. The average age of these older subjects was, however, only 62. A are needed, as currently practice is based largely on extrapolation
recent multisite, open-label, prospective trial of lamotrigine for ger- of clinical evidence from younger bipolar patients, who may have a
iatric bipolar depression found nearly 60% remission and response different kind of illness.
rate in the study population of type I and II bipolar depression (n References
= 57, mean age 66 years) (Sajatovic et al., 2011). Dropout rates due
Aizenberg, D., et al. (2006). Suicide attempts amongst elderly bipolar patients.
to adverse events were modest (10.5%), and included rash (four
Journal of Affective Disorders, 91(1), 91–4.
patients), manic switch (one patient), and hyponatraemia (one Almeida, O.P. and Fenner, S. (2002). Bipolar disorder: similarities and
patient). A number of side effects were noted including reduced differences between patients with illness onset before and after 65 years
sleep, weight loss, increased dreams, polyuria, weight gain, dimin- of age. International Psychogeriatrics, 14(3), 311–22.
ished sexual desire, increased sleep, fatigue, and unsteady gait. American Psychiatric Association (1994). Diagnostic and statistical manual
Amongst other mood stabilizers there is also some evidence for of mental disorders: DSM-IV. American Psychiatric Association,
levetiracetam (Kyomen, 2006). Washington, DC.
Angst, F., et al. (2002). Mortality of patients with mood disorders: follow-up
Antidepressants over 34–38 years. Journal of Affective Disorders, 68(2–3), 167–81.
Baldwin, R. (2008). Mood disorders: depressive disorders. Oxford University
Antidepressants have a role in the acute treatment of a depressive
Press, Oxford.
episode in unipolar depressive disorder in older people as well as Bedford, J.J., et al. (2008). Lithium-induced nephrogenic diabetes insipidus:
prophylaxis in preventing relapse. For the treatment of bipolar dis- renal effects of amiloride. Clinical Journal of American Society of
order, monotherapy with antidepressants risks provoking a manic Nephrology, 3(5), 1324–31.
episode (Goodwin, 2009). Beyer, J.L., et al. (2008). Patterns of pharmacotherapy and treatment response
in elderly adults with bipolar disorder. Psychopharmacology Bulletin,
Antipsychotics 41(1), 102–14.
Some antipsychotics have displayed efficacy in late-life bipolar Boora, K., et al. (2008). Encephalopathy with combined lithium-risperidone
administration. Acta Psychiatrica Scandinavica, 117(5), 394–5; discussion
disorder, especially as an antimanic agent. In a post hoc analysis
396.
of noncontrolled data, quetiapine monotherapy was found to be Braun, C.M., et al. (1999). Mania, pseudomania, depression, and
effective for the acute treatment of bipolar mania in adults greater pseudodepression resulting from focal unilateral cortical lesions.
than 55 years of age (Sajatovic et al., 2008a). In an open-label trial, Neuropsychiatry, Neuropsychology and Behavioural Neurology, 12(1),
the same authors found evidence for efficacy of aripiprazole in a 35–51.
small (n = 20) study of older adults in acute mania (Sajatovic et al., Brommelhoff, J.A., et al. (2009). Depression as a risk factor or prodromal
2008b). feature for dementia? Findings in a population-based sample of Swedish
In a retrospective study on patterns of pharmacotherapy and twins. Psychological Aging, 24(2), 373–84.
treatment response following up 138 acutely unwell patients in a Cabras, P.L., et al. (1999). Clinical experience with gabapentin in patients
with bipolar or schizoaffective disorder: results of an open-label study.
single centre, it was observed that mood stabilizers were the most Journal of Clinical Psychiatry, 60(4), 245–8.
prescribed (68%), followed by antipsychotics (54%) and antide- Calabrese, J.R., et al. (1999). Spectrum of activity of lamotrigine in
pressants (34%). Combination therapy with these medications was treatment-refractory bipolar disorder. American Journal of Psychiatry,
more common than monotherapy (57% vs 38%). Remission was 156(7), 1019–23.
achieved in only 35% of subjects, while 32% showed no significant Cassidy, F. and Carroll, B.J. (2002). Vascular risk factors in late onset mania.
improvement (Beyer et al., 2008). Psychological Medicine, 32(2), 359–62.
Cassidy, F., et al. (2001). Substance abuse in bipolar disorder. Bipolar Head, L. and Dening, T. (1998). Lithium in the over-65s: who is taking it
Disorders, 3(4), 181–8. and who is monitoring it? A survey of older adults on lithium in the
Chen, S.T., et al. (1999). Efficacy of lithium vs. valproate in the treatment of Cambridge Mental Health Services catchment area. International Journal
mania in the elderly: a retrospective study. Journal of Clinical Psychiatry, of Geriatric Psychiatry, 13(3), 164–71.
60(3), 181–6. Himmelhoch, J.M., et al. (1980). Age, dementia, dyskinesias, and lithium
Depp, C.A. and Jeste, D.V. (2004). Bipolar disorder in older adults: a critical response. American Journal of Psychiatry, 137(8), 941–5.
review. Bipolar Disorders, 6(5), 343–67. Hirschfeld, R.M., et al. (2003). Screening for bipolar disorder in the
Dhingra, U. and Rabins, P.V. (1991). Mania in the elderly: a 5–7 year follow-up. community. Journal of Clinical Psychiatry, 64(1), 53–9.
Journal of the American Geriatrics Society, 39(6), 581–3. Janicak, P.G. (1993). The relevance of clinical pharmacokinetics and therapeutic
Fagiolini, A., et al. (2009). Enhancing outcomes in patients with bipolar drug monitoring: anticonvulsant mood stabilizers and antipsychotics.
disorder: results from the Bipolar Disorder Center for Pennsylvanians Journal of Clinical Psychiatry, 54 Suppl, 35–41; discussion 55–6.
Study. Bipolar Disorders, 11(4), 382–90. Jeremy, B. and Robin, J. (1990). Mania in old age: a first prospective study.
Fahy, S. and Lawlor, B.A. (2001). Discontinuation of lithium augmentation in an International Journal of Geriatric Psychiatry, 5(4), 215–22.
elderly cohort. International Journal of Geriatric Psychiatry, 16(10), 1004–9. Jones, L.D., et al. (2009). Temporal lobe volume in bipolar disorder:
Felix, S., et al. (2003). Dose-related pharmacokinetics and pharmacodynamics relationship with diagnosis and antipsychotic medication use. Journal of
of valproate in the elderly. Journal of Clinical Psychopharmacology, 23(5), Affective Disorders, 114(1–3), 50–7.
471–8. Juurlink, D.N., et al. (2004). Drug-induced lithium toxicity in the elderly: a
Geddes, J.R., et al. (2010). Lithium plus valproate combination therapy versus population-based study. Journal of the American Geriatric Society, 52(5),
monotherapy for relapse prevention in bipolar I disorder (BALANCE): a 794–8.
randomised open-label trial. Lancet, 375(9712), 385–95. Kennedy, N., et al. (2005). Incidence and distribution of first-episode mania
Ghaemi, S.N., et al. (1998). Gabapentin treatment of mood disorders: a by age: results from a 35-year study. Psychological Medicine, 35(6),
preliminary study. Journal of Clinical Psychiatry, 59(8), 426–9. 855–63.
Gildengers, A.G., et al. (2004). Cognitive functioning in late-life bipolar Kessing, L.V., et al. (2008). Lithium treatment and risk of dementia. Archives
disorder. American Journal of Psychiatry, 161(4), 736–8. of General Psychiatry, 65(11), 1331–5.
Gildengers, A.G., et al. (2005). A pilot study of standardized treatment in Koenig, H.G. and Blazer, D.G. (1992). Epidemiology of geriatric affective
geriatric bipolar disorder. American Journal of Geriatric Psychiatry, disorders. Clinical Geriatric Medicine, 8(2), 235–51.
13(4), 319–23. Kortenoeven, M.L., et al. (2009). Amiloride blocks lithium entry through the
Gildengers, A.G., et al. (2008a). Medical burden in late-life bipolar and major sodium channel thereby attenuating the resultant nephrogenic diabetes
depressive disorders. American Journal of Geriatric Psychiatry, 16(3), insipidus. Kidney International, 76(1), 44–53.
194–200. Ku, B.D., et al. (2006). Secondary mania in a patient with delayed anoxic
Gildengers, A.G., et al. (2008b). A 12-week open-label pilot study of donepezil encephalopathy after carbon monoxide intoxication. Journal of Clinical
for cognitive functioning and instrumental activities of daily living in Neuroscience, 13(8), 860–2.
late-life bipolar disorder. International Journal of Geriatric Psychiatry Kyomen, H.H. (2006). The use of levetiracetam to decrease mania in elderly
23(7), 693–8. bipolar patients. American Journal of Geriatric Psychiatry, 14(11), 985.
Gildengers, A.G., et al. (2009). The longitudinal course of cognition in older Laursen, T.M., et al. (2009). Bipolar disorder, schizoaffective disorder, and
adults with bipolar disorder. Bipolar Disorders, 11(7), 744–52. schizophrenia overlap: a new comorbidity index. Journal of Clinical
Glasser, M. and Rabins, P. (1984). Mania in the elderly. Age and Ageing, 13(4), Psychiatry, 70(10), 1432–8.
210–13. Leboyer, M., et al. (2005). Age at onset in bipolar affective disorders: a review.
Gnam, W. and Flint, A.J. (1993). New onset rapid cycling bipolar disorder in Bipolar Disorders, 7(2), 111–18.
an 87 year old woman. Canadian Journal of Psychiatry, 38(5), 324–6. Lin, H.C., et al. (2007). Increased risk of developing stroke among patients
Goldstein, B.I., et al. (2006). Comorbidity in bipolar disorder among the with bipolar disorder after an acute mood episode: a six-year follow-up
elderly: results from an epidemiological community sample. American study. Journal of Affective Disorders, 100(1–3), 49–54.
Journal of Psychiatry, 163(2), 319–21. Lloyd, A.J., et al. (2009). White matter lesions in euthymic patients with
Goodwin, F.K. and Jamison, K.R. (1990). Manic-depressive illness. Oxford bipolar disorder. Acta Psychiatrica Scandinavica, 120(6), 481–91.
University Press, New York. Lopez, J., et al. (2009). Late-onset bipolar disorder following right thalamic
Goodwin, G.M. (2009). Evidence-based guidelines for treating bipolar injury. Actas Españolas de Psiquiatría, 37(4), 233–5.
disorder: revised second edition—recommendations from the British Marcotte, D.B. (2004). Long-term use of lamotrigine for bipolar disorder in
Association for Psychopharmacology. Journal of Psychopharmacology, patients over 55 years of age. Proceedings of 157th Annual Meeting of the
23(4), 346–88. American Psychiatric Association, New York.
Grant, B.F., et al. (2005). Prevalence, correlates, and comorbidity of bipolar Martinez-Aran, A., et al. (2004). Cognitive function across manic or
I disorder and axis I and II disorders: results from the National hypomanic, depressed, and euthymic states in bipolar disorder. American
Epidemiologic Survey on Alcohol and Related Conditions. Journal of Journal of Psychiatry, 161(2), 262–70.
Clinical Psychiatry, 66(10), 1205–15. McElroy, S.L., et al. (2001). Axis I psychiatric comorbidity and its relationship
Gualtieri, C.T. and Johnson, L.G. (2008). Age-related cognitive decline in to historical illness variables in 288 patients with bipolar disorder.
patients with mood disorders. Progress in Neuropsychopharmacology, American Journal of Psychiatry, 158(3), 420–6.
Biology and Psychiatry, 32(4), 962–7. Miklowitz, D.J. (2008). Adjunctive psychotherapy for bipolar disorder: state
Haller, S., et al. (2011). Combined analysis of grey matter voxel-based of the evidence. American Journal of Psychiatry, 165(11), 1408–19.
morphometry and white matter tract-based spatial statistics in late-life Molinari, V. and Marmion, J. (1995). Relationship between affective disorders
bipolar disorder. Journal of Psychiatry and Neuroscience, 36(6), 391–401. and axis II diagnoses in geropsychiatric patients. Journal of Geriatric
Hardy, B.G., et al. (1997). Gradual discontinuation of lithium augmentation Psychiatry and Neurology, 8(1), 61–4.
in elderly patients with unipolar depression. Journal of Clinical Murray, N., et al. (1983). The influence of age on lithium efficacy and
Psychopharmacology, 17(1), 22–6. side-effects in out-patients. Psychological Medicine, 13(1), 53–60.
Hays, J.C., et al. (1998). Age of first onset of bipolar disorder: demographic, Nielsen, J., et al. (2008). Dysregulation of renal aquaporins and epithelial
family history, and psychosocial correlates. Depression and Anxiety, 7(2), sodium channel in lithium-induced nephrogenic diabetes insipidus.
76–82. Seminars in Nephrology, 28(3), 227–44.
Nunes, P.V., et al. (2007). Lithium and risk for Alzheimer’s disease in elderly Starkstein, S.E., et al. (1990). Mania after brain injury: neuroradiological and
patients with bipolar disorder. British Journal of Psychiatry, 190, 359–60. metabolic findings. Annals of Neurology, 27(6), 652–9.
Oostervink, F., et al. (2009). Bipolar disorder in the elderly; different effects of Steffens, D.C. and Krishnan, K.R. (1998). Structural neuroimaging and mood
age and of age of onset. Journal of Affective Disorders, 116(3), 176–83. disorders: recent findings, implications for classification, and future
Ownby, R.L., et al. (2006). Depression and risk for Alzheimer disease: directions. Biological Psychiatry, 43(10), 705–12.
systematic review, meta-analysis, and metaregression analysis. Archives Stone, K. (1989). Mania in the elderly. British Journal of Psychiatry, 155,
of General Psychiatry, 63(5), 530–8. 220–4.
Parker, K.L., Narayan, M., and Nelson, J.C. (1994). Lithium augmentation in Strakowski, S.M., et al. (1994). The co-occurrence of mania with medical
geriatric depressed outpatients: a clinical report. International Journal of and other psychiatric disorders. International Journal of Psychiatry and
Geriatric Psychiatry, 9, 995–1002. Medicine, 24(4), 305–28.
Regier, D.A., et al. (1993). One-month prevalence of mental disorders in the Subramaniam, H., et al. (2007). The role of vascular risk factors in late onset
United States and sociodemographic characteristics: the Epidemiologic bipolar disorder. International Journal of Geriatric Psychiatry, 22(8),
Catchment Area study. Acta Psychiatrica Scandinavica, 88(1), 35–47. 733–7.
Risinger, R.C., et al. (1994). Safety and efficacy of divalproex sodium in Swartz, C.M. and Dolinar, L.J. (1995). Encephalopathy associated with rapid
elderly bipolar patients. Journal of Clinical Psychiatry, 55(5), 215. decrease of high levels of lithium. Annals of Clinical Psychiatry, 7(4),
Roose, S.P., et al. (1979). Lithium treatment in older patients. American 207–9.
Journal of Psychiatry, 136(6), 843–4. Tamashiro, J.H., et al. (2008). Increased rates of white matter hyperintensities
Rybakowski, J.K., et al. (2009). Response to prophylactic lithium in bipolar in late-onset bipolar disorder. Bipolar Disorders, 10(7), 765–75.
disorder may be associated with a preservation of executive cognitive Thomas, A.J., et al. (2002). Ischemic basis for deep white matter
functions. European Neuropsychopharmacology, 19(11), 791–5. hyperintensities in major depression: a neuropathological study. Archives
Sajatovic, M. (2002). Aging-related issues in bipolar disorder: a health of General Psychiatry, 59(9), 785–92.
services perspective. Journal of Geriatric Psychiatry and Neurology, 15(3), Thomas, A.J., et al. (2004). Depression and vascular disease: what is the
128–33. relationship? Journal of Affective Disorders, 79(1–3), 81–95.
Sajatovic, M., et al. (2004). Age-related modifiers of clinical presentation Trannel, T.J., et al. (2001). Occurrence of thrombocytopenia in psychiatric
and health service use among veterans with bipolar disorder. Psychiatric patients taking valproate. American Journal of Psychiatry, 158(1),
Services, 55(9), 1014–21. 128–30.
Sajatovic, M., et al. (2005). Maintenance treatment outcomes in older patients Tsai, S.Y., et al. (2007). Cognitive impairment in later life in patients with
with bipolar I disorder. American Journal of Geriatric Psychiatry, 13(4), early-onset bipolar disorder. Bipolar Disorders, 9(8), 868–75.
305–11. Unutzer, J., et al. (1998). The treated prevalence of bipolar disorder in a large
Sajatovic, M., et al. (2008a). Quetiapine for the treatment of bipolar mania in staff-model HMO. Psychiatric Services, 49(8), 1072–8.
older adults. Bipolar Disorders, 10(6), 662–71. Van Gerpen, M.W., et al. (1999). Mania in the geriatric patient population:
Sajatovic, M., et al. (2008b). Aripiprazole therapy in 20 older adults with a review of the literature. American Journal of Geriatric Psychiatry, 7(3),
bipolar disorder: a 12-week, open-label trial. Journal of Clinical Psychiatry, 188–202.
69(1), 41–6. Vasudev, A. and Harrison, R. (2008). Prescribing safely in elderly psychiatric
Sajatovic, M., et al. (2011). Multisite, open-label, prospective trial of wards: Survey of possible drug interactions. Psychiatric Bulletin, 32(11),
lamotrigine for geriatric bipolar depression: a preliminary report. Bipolar 417–18.
Disorders, 13(3), 294–302. Vasudev, A. and Thomas, A. (2010). Bipolar disorder’ in the elderly: what’s in
Sarnicola, A., et al. (2009). No differential effect of age on brain matter a name? Maturitas, 66(3), 231–5.
volume and cognition in bipolar patients and healthy individuals. Bipolar Vederine, F.E., et al. (2011). A meta-analysis of whole-brain diffusion tensor
Disorders, 11(3), 316–22. imaging studies in bipolar disorder. Progress in Neuropsychopharmacology,
Schouws, S.N., et al. (2010). Risk factors for cognitive impairment in elderly Biology and Psychiatry, 35(8), 1820–6.
bipolar patients. Journal of Affective Disorders, 125(1–3), 330–5. Verdoux, H. and Bourgeois, M. (1995). Secondary mania caused by cerebral
Shulman, K.I. and Herrmann, N. (1999). The nature and management of organic pathology. Annals of Medical Psychology (Paris), 153(3),
mania in old age. Psychiatric Clinics of North America, 22(3), 649–65, ix. 161–8.
Shulman, K. and Post, P. (1980). Bipolar affective disorder in old age. British Weissman, M.M., et al. (1988). Affective disorders in five United States
Journal of Psychiatry, 136, 26–32. communities. Psychological Medicine, 18(1), 141–53.
Shulman, K.I., et al. (1992). Mania compared with unipolar depression in old WHO (1992). International Classification of Diseases-10. World Health
age. American Journal of Psychiatry, 149(3), 341–5. Organization, Geneva.
Shulman, K.I., et al. (2003). Changing prescription patterns for lithium Wijeratne, C. and Malhi, G.S. (2007). Vascular mania: an old concept in
and valproic acid in old age: shifting practice without evidence. British danger of sclerosing? A clinical overview. Acta Psychiatrica Scandinavica
Medical Journal, 326(7396), 960–1. Supplement, 434, 35–40.
Shulman, K.I., et al. (2005a). Incidence of delirium in older adults newly Winokur, G. (1975). The Iowa 500: heterogeneity and course in
prescribed lithium or valproate: a population-based cohort study. Journal manic-depressive illness (bipolar). Comprehensive Psychiatry, 16(2),
of Clinical Psychiatry, 66(4), 424–7. 125–31.
Shulman, K.I., et al. (2005b). New thyroxine treatment in older adults Wylie, M.E., et al. (1999). Age at onset in geriatric bipolar disorder. Effects
beginning lithium therapy: implications for clinical practice. American on clinical presentation and treatment outcomes in an inpatient sample.
Journal of Geriatric Psychiatry, 13(4), 299–304. American Journal of Geriatric Psychiatry, 7(1), 77–83.
Silva, D., et al. (2009). Cognitive deficits in middle-aged and older adults Young, R.C. (1996). Treatment of geriatric mania. In: Mood disorders across
with bipolar disorder and cognitive complaints: comparison with mild the life span (eds K.I. Shulman, M. Tohen, and S.P. Kutcher), pp. 411–25.
cognitive impairment. International Journal of Geriatric Psychiatry, Wiley-Liss, New York.
24(6), 624–31. Young, R.C. and Klerman, G.L. (1992). Mania in late life: focus on age at
Snowdon, J. (1991). A retrospective case-note study of bipolar disorder in old onset. American Journal of Psychiatry, 149(7), 867–76.
age. British Journal of Psychiatry, 158, 485–90. Zhang, Y., et al. (2009). Potential role of purinergic signaling in lithium-
Starkstein, S.E. and Robinson, R.G. (1997). Mechanism of disinhibition after induced nephrogenic diabetes insipidus. American Journal of Physiology
brain lesions. Journal of Nervous Mental Diseases, 185(2), 108–14. and Renal Physiology, 296(5), F1194–201.
CHAPTER 45
Anxiety disorders
in older people
Gerard Byrne
Fear and anxiety are phylogenetically ancient emotions that confer most GAD has its onset earlier in life. Another significant change
survival advantage across species (Darwin, 1872). They facilitate is mooted for criterion A of PTSD. It has been proposed to tighten
escape from present danger and prepare the individual to deal rap- up PTSD criterion A1, which describes the nature of the traumatic
idly with future threats. In contemporary life, a moderate increase exposure, and to delete criterion A2, which requires the person to
in anxiety is commonly associated with increased performance have experienced ‘intense fear, helplessness or horror’ at the time of
(Yerkes and Dodson, 1908). However, anxiety that is excessive or the traumatic exposure (Friedman et al., 2011).
prolonged is maladaptive and may represent a mental disorder.
Individual differences and contextual factors influence the final Epidemiology
form that an anxiety disorder takes. Like many complex behav-
iours, anxiety disorders result from interactions between genetic There is now a substantial literature on the prevalence of anxiety
and environmental factors. In older people, anxiety often compli- disorders in older people, although a firm consensus is yet to emerge
cates physical frailty and cognitive decline. From a nosological per- due to unresolved substantive and methodological issues. Bryant et
spective, anxiety can be conceptualized as both dimensional and al. (2008) have reviewed anxiety disorder prevalence estimates and
categorical. Both psychological and pharmacological treatments associated methodological issues. Most population-based surveys
are commonly applied to anxiety disorders in older people, with find that the prevalence rates for anxiety disorders fall significantly
moderate efficacy. This chapter deals with classification, epidemiol- after the age of about 50 years. The reasons for this oft-repeated
ogy, scientific underpinnings, phenomenology, and modern treat- observation are unclear. The observation that this effect begins
ment approaches to anxiety disorders in later life. in middle age makes it particularly challenging to explain using
conventional age-related arguments. It has been suggested that
the diagnostic interviews used to detect anxiety disorders are not
Classification appropriate for use in older people (O’Connor, 2006; O’Connor
The tenth edition of the International Classification of Diseases and Parslow, 2010), or that the diagnostic criteria themselves have
(ICD-10) includes anxiety disorders within a broader category of intrinsic age biases. Anxiety disorder prevalence rates are higher
neurotic, stress-related, and somatoform disorders (World Health in women than men at all ages, including later life. This is reflected
Organization, 2010). In contrast, DSM-IV has a separate section in the age by sex distribution of anxiety disorder prevalence in
for anxiety disorders (American Psychiatric Association, 2000). national population-based surveys (see Fig. 45.1). However, there
Both systems include obsessive-compulsive disorder (OCD) and is some evidence that the sex difference in anxiety prevalence might
post-traumatic stress disorder (PTSD) under these rubrics. Table decline in advanced old age (Pachana et al., 2012).
45.1 shows the classification of anxiety disorders under ICD-10 Table 45.2 provides a summary of nonhierarchical 12-month
and DSM-IV. There is substantial commonality between the lists population prevalence estimates for anxiety disorders according to
of anxiety disorders under the two dominant nosologies, reflecting DSM-IV and ICD-10 criteria in persons aged 65–85 years and in
convergence over time. Both of these classification systems are cur- persons aged 16–64 years, based on the 2007 Australian National
rently under revision, with ICD-11 due in 2015 and DSM-5 due in Survey of Mental Health and Wellbeing (NSMHWB) (Australian
April 2013. Several changes have been proposed to the diagnostic Bureau of Statistics, 2007). This population survey employed trained
criteria for anxiety disorders in DSM-5. Most significantly, it has lay interviewers, who used the Composite International Diagnostic
been proposed to reduce the minimum duration of symptoms in Interview (CIDI). The survey included 1905 community-residing
generalized anxiety disorder (GAD) from 6 months to 3 months persons aged 65 years and over, but it did not sample persons resid-
(Andrews et al., 2010). This is likely to have the effect of increasing ing in hospitals, nursing homes, or caravan parks. The two diag-
the overall prevalence of GAD in adults but not altering its sever- nostic systems generated somewhat different prevalence estimates
ity (Andrews and Hobbs, 2010). However, it is likely to have less for panic disorder, agoraphobia, and PTSD, with ICD-10 provid-
impact on the measured prevalence of GAD in older people, as ing higher estimates in each case. The survey did not measure the
Table 45.1 Anxiety disorder classification in ICD-10 and DSM-IV

ICD-10 DSM-IV
F40.0 Agoraphobia, unspecified 300.22 Agoraphobia without history of panic disorder
F40.02 Agoraphobia without panic disorder
F40.1 Social phobia 300.23 Social phobia
F40.2 Specific phobia 300.29 Specific phobia
F41.0 Panic disorder 300.01 Panic disorder without agoraphobia
F40.01 Agoraphobia with panic disorder 300.21 Panic disorder with agoraphobia
F41.1 Generalized anxiety disorder 300.02 Generalized anxiety disorder
F41.2 Mixed anxiety and depressive disorder
F41.3 Other mixed anxiety disorders Proposed for DSM-5
F42 Obsessive compulsive disorder 300.3 Obsessive-compulsive disorder
F43.0 Acute stress disorder 308.3 Acute stress disorder
F43.1 Post-traumatic stress disorder 309.81 Post-traumatic stress disorder
F43.22 Adjustment disorder with anxiety 309.24 Adjustment disorder with anxiety
F43.23 Adjustment disorder with mixed anxiety and depressed mood 309.28 Adjustment disorder with mixed anxiety and depressed mood
F06.4 Anxiety disorder due to a known physiological condition 293.84 Anxiety disorder due to a general medical condition
Various disorders listed under F10–F19 292.89 Substance-induced anxiety disorder
F41.0 Anxiety disorder, unspecified 300.00 Anxiety disorder not otherwise specified
25 person simply to experience recurrent panic attacks (World Health

Organization, 2010), whereas DSM-IV requires the person to expe-
20 rience recurrent panic attacks and one or more sequelae, includ-
ing a persistent concern about having another attack, worry about
the implications or consequences of the attack, or a significant
15
%
change in behaviour following the attack (American Psychiatric
Association, 2000). Thus, in NSMHWB, similar proportions of
10 older people reported having experienced panic attacks in the past
12 months under the DSM-IV and ICD-10 criteria, but the absence
5 in ICD-10 of the requirement for one or more panic attack sequelae
meant that the prevalence of panic disorder was much higher under
0 this system than when the DSM-IV criteria were applied (see Table
45.2). So these different prevalence estimates do not reflect minor
0
5
–2
–2
–3
–3
–4
–4
–5
–5
–6
–6
–7
–7
–8
–8
16
21
26
31
36
41
46
51
56
61
66
71
76
81
methodological issues, but major differences in the way disorders

Age group (years)
are conceptualized or formalized. These differences might need to
Females Males be taken into consideration when basing service development or
Fig. 45.1 Twelve-month prevalence of anxiety disorders by age and sex. funding decisions on epidemiological data.
(Australian Bureau of Statistics (2007). National Survey of Mental Health and Wellbeing; ABS One method of attempting to sort out the validity of competing
Cat. No. 4326.0.) diagnostic approaches in a clinically meaningful way is to measure the
degree of distress and disability associated with a diagnosis. Using the
12-item Short Form Health Survey (SF-12) to gauge disability, Slade
prevalence of simple phobia. The population prevalence estimates and Andrews (2001) have shown in adults 18 years and over that
in Table 45.2 are associated with relatively wide confidence inter- DSM-IV GAD is associated with greater disability than ICD-10 GAD.
vals (CI), indicating the level of imprecision with which the esti- The Longitudinal Ageing Study Amsterdam (LASA) used a
mates were made. The variations in prevalence estimates between two-stage design to investigate the consequences of anxiety disor-
the DSM-IV and ICD-10 nosological systems can be explained by der and anxiety symptoms in people aged 55–85 years (de Beurs et
differences in their diagnostic criteria. To illustrate this, let us take al., 1999). The anxiety disorders studied were panic disorder, GAD,
panic disorder as an example. Both DSM-IV and ICD-10 require OCD, and phobic disorders. Older people with an anxiety disorder,
a person with panic disorder to experience multiple, unexpected or with anxiety symptoms, reported worse perceived health, greater
panic attacks, and the definitions of a panic attack are similar loneliness, worse life satisfaction, and greater health services utili-
under each system. The main difference is that ICD-10 requires the zation than age-matched controls (de Beurs et al., 1999). In a US
CHAPTER 45 anxiety disorders in older people 591
Table 45.2 Twelve-month prevalence rates (%) for DSM-IV and ICD-10 anxiety disorders
DSM-IV 12-month prevalence (95% CI) ICD-10 12-month prevalence (95% CI)
65–85 years 16–64 years 65–85 years 16–64 years

Generalized anxiety disorder 1.4 (0.8–1.9) 4.3 (3.8–4.7) 1.5 (1.0–2.1) 3.7 (3.3–4.1)
Panic disorder 0.5 (0.2–0.9) 2.2 (1.9–2.6) 0.9 (0.5–1.4) 3.1 (2.7–3.5)
Agoraphobia (+/− panic disorder) 0.3 (0.1–0.6) 1.6 (1.3–1.9) 1.0 (0.6–1.5) 3.4 (3.0–3.9)
Social phobia 1.2 (0.7–1.7) 5.4 (4.8–5.9) 1.3 (0.8–1.8) 5.8 (5.3–6.4)
Post-traumatic stress disorder 1.6 (1.1–2.2) 5.3 (4.8–5.8) 2.5 (1.8–3.2) 7.9 (7.3–8.6)
Obsessive-compulsive disorder 0.7 (0.4–1.1) 3.1 (2.7–3.5) 0.7 (0.4–1.1) 2.4 (2.0–2.7)
Any anxiety disorder 4.3 (3.3–5.2) 14.2 (13.4–15.0) 6.0 (4.9–7.1) 17.1 (16.1–17.9)
These population-weighted estimates were calculated from the confidentialized unit record file of the National Survey of Mental Health and Wellbeing, Australian Bureau of Statistics
(2007). All diagnoses are nonhierarchical.
study of people aged 60–80 years who were seeking treatment for disorder (Australian Bureau of Statistics,, 2007). In contrast, of
GAD and a matched control group, Wetherell et al. (2004) analysed older NSMHWB participants with a 12-month history of affective
quality of life using the 36-item Short Form Health Survey (SF-36). disorder, 45.5% also had a 12-month history of anxiety disorder.
In this study, GAD was associated with worse health-related quality These observations are important because there is a long-standing
of life across almost all SF-36 domains and these associations were myth that anxiety in pure culture is uncommon in older people.
present regardless of whether GAD was complicated by comorbid Bereavement reactions are also commonly complicated by anxiety
psychiatric disorder. In addition, Porensky et al. (2009) have dem- disorders, particularly GAD and PTSD (Zisook et al., 1990).
onstrated that older people with DSM-IV GAD have higher dis- Although the prevalence of anxiety disorders in
ability, worse health-related quality of life, and greater healthcare community-residing older people is lower than in young or
utilization than matched comparison participants without GAD. middle-aged persons, older people with other health problems have
The incidence of anxiety disorders in older people has been been shown to experience a high prevalence of anxiety disorder.
less well studied. However, using data from Wave 2 of the US Smalbrugge et al. (2005) found that 29.7% of 333 Dutch nursing
National Epidemiologic Survey on Alcohol and Related Conditions home residents had anxiety symptoms, 4.2% had subthreshold
(NESARC) based on the Alcohol Use Disorder and Associated anxiety disorders, and 5.7% had an anxiety disorder. Residents with
Disabilities Interview Schedule-IV (AUDADIS-IV), Chou et al. stroke or depression were at greater risk of an anxiety disorder.
(2011) estimated the weighted 3-year incidence of nonhierarchi- Chemerinski et al. (1998) assessed 398 patients with Alzheimer’s
cal DSM-IV anxiety disorders in 8012 persons aged 60 years and disease and found that 5% met diagnostic criteria for GAD over the
over as follows: GAD 1.63%; specific phobia 1.35%; panic disor- previous 4 weeks. In the Cache County study, Lyketsos et al. (2001)
der 0.76%; and social phobia 0.58%. Incidence estimates for other found that anxiety symptoms were present in 10–12% of patients
anxiety disorders were not reported. Participants aged 80 years and with Alzheimer’s disease of varying severities. Anxiety also occurs
over were at greater risk of incident panic disorder than partici- commonly in the context of mild cognitive impairment (MCI)
pants aged 60–79 years. PTSD at baseline predicted incident GAD, (Monastero et al. 2009) and its presence in MCI appears to be part
panic disorder, and specific phobia, whereas panic disorder at of the prodrome of Alzheimer’s disease (Gallagher et al., 2011).
baseline predicted incident social phobia. In addition, personality Moreover, it seems that individuals with normal cognition, but
disorder at baseline predicted incident GAD, panic disorder, and who are at increased risk of developing the clinical manifestations
social phobia, but not specific phobia. Interestingly, after adjust- of Alzheimer’s disease in the future, also have increased anxiety.
ment for other psychiatric disorders and sociodemographic fac- Lavretsky et al. (2009) found that anxiety symptoms in middle-aged
tors, neither adverse life events nor measures of general health were and older adults without dementia were associated with cerebral
related to incident anxiety disorder in this particular cohort (Chou amyloid load, as demonstrated by FDDNP-PET (positron emission
et al., 2011; Mackenzie et al., 2011). It is not appropriate to compare tomography) binding.
the incidence rates from NESARC with the prevalence rates from
NSMHWB as the age ranges, cohort characteristics, and diagnostic
instruments all differed. Biological Factors
A significant minority of older people with an anxiety disorder Although the prefrontal cortex is involved in social cognition, it
also meets diagnostic criteria for a comorbid psychiatric disorder. is the more phylogenetically ancient limbic system that is most
For example, of older people with a DSM-IV 12-month anxiety involved in fear and anxiety responses. The limbic system includes
disorder in the NSMHWB study, 60% met diagnostic criteria for a the insular cortex, the cingulate gyrus, the hippocampus, and the
single anxiety disorder, 14% met criteria for at least one additional amygdala. The amygdala is particularly important for the genera-
anxiety disorder, and 26% met criteria for at least one additional tion of panic (Davis, 1992) and there is evidence that fear responses
mood or substance use disorder. Similarly, of older people with a in the amygdala are ‘hard-wired’ and difficult to extinguish (Sah and
DSM-IV 12-month anxiety disorder, 78% had a comorbid physical Westbrook, 2008). The presence of fear-related circuits involving
the amygdala, insula, and anterior cingulate has been demonstrated Levels of trait anxiety are relatively stable over the lifespan, indi-
in humans on neuroimaging studies with both functional magnetic cating the influence of genetic and early environmental influences.
resonance imaging (fMRI) and PET (Sehlmeyer et al., 2009). In contrast, levels of state anxiety fluctuate over time, reflecting
The hippocampus modulates the activity of the hypothalamic– responses to aversive stimuli throughout life. However, state anxiety
pituitary–adrenal axis, dampening down stress responses (Davis, is also influenced by genetic factors. In the dimensional ‘five-factor’
1992). There is evidence that both hippocampal volume and neu- model of normal personality, high neuroticism is associated with
rogenesis are implicated in the response to stress. Although more anxiety symptoms and all anxiety disorders. In addition, high
research in humans is needed, individuals with greater hippocam- conscientiousness is associated with GAD and OCD, whereas low
pal volume and intact hippocampal neurogenesis appear to have extraversion is associated with social phobia (Rosellini and Brown,
greater resilience (Martin et al., 2009). In research on twins dis- 2011). In older adults, neuroticism and GAD are both moderately
cordant for combat exposure in Vietnam, those with smaller hip- heritable (0.47 and 0.27, respectively), and approximately one-third
pocampi were vulnerable to PTSD following combat exposure, of the genetic influence on GAD is shared with trait neuroticism
whereas twins with normal hippocampi or without combat expo- (Mackintosh et al., 2006). Even in people with Alzheimer’s dis-
sure did not have an elevated rate of PTSD (Gilbertson et al., 2002). ease, informant-rated premorbid neuroticism has been found to be
Thus, both the biological vulnerability demonstrated by small hip- associated with current anxiety on the Neuropsychiatric Inventory
pocampi and the environmental stimulus of combat exposure were (Archer et al., 2007).
necessary before PTSD developed. In both laboratory animals and humans, a new fear response
Studies in monozygotic and dizygotic twins indicate that about develops when a harmless context or cue (conditioned stimulus
30–40% of the variation in the risk of anxiety disorder is of genetic ori- (CS)) is paired with an aversive event (unconditioned stimulus
gin (Norrholm and Ressler, 2009). No genes of large effect have yet been (US)). Although the fear response to the CS can be extinguished
identified (Schumacher et al., 2011), so polygenic influences are likely. by habituation—repeated exposure to the CS in the absence of the
There is some evidence that genetic factors might play a greater part US—it is rapidly reinstated by either simultaneous re-exposure
in the aetiology of panic disorder than of GAD. Polymorphisms in the to the CS and US or sequential re-exposure to the US followed
promoter region of the 5-HT transporter gene have been linked to risk shortly by the CS, indicating that learnt fear is strongly retained
of anxiety disorder (Xie et al., 2009). Individuals who are homozygous (Sah and Westbrook, 2008). This observation is likely to at least
for the s allele of this promoter exhibit greater fMRI responses in partially explain the chronicity of many anxiety disorders and why
the amygdala to fearful stimuli and report higher levels of neuroti- the response to short-term treatment is often poorly sustained.
cism (Davis, 1992). In a study of survivors of the Rwandan genocide, Avoidance behaviour leads to a temporary reduction in anxi-
Kolassa et al. (2010) found that the risk of PTSD depended upon both ety, which negatively reinforces (increases) avoidance behaviour.
the traumatic load and the catechol-o-methyltransferase Val(158)Met However, avoidance is not a successful long-term strategy as it
polymorphism. Individuals with Val polymorphisms developed PTSD makes anxiety worse by preventing habituation to the feared stimu-
following trauma in a dose-response manner, whereas trauma-exposed lus. Avoidance is thus key to maintenance of the fear response, and
individuals homozygous for the Met/Met polymorphism were at high dealing with avoidance is critical to successful treatment of most
risk of PTSD regardless of the level of their traumatic exposure. This anxiety disorders. Exposure and behavioural activation paradigms
same genetic polymorphism might also predict response to treatment that reduce avoidance behaviours have been used as the basis for
(Lonsdorf et al., 2010), although more work is needed before these interventions for anxiety disorders (Cherbuin et al., 2008).
findings are translated into clinical practice. Hyperventilation is a common symptom of panic attacks, and
Experiments in laboratory animals demonstrate the importance for many years it was also considered to be a cause of panic (Griez
of the early environment in the development of anxiety. Rats sepa- and Schruers, 1998). However, inhaled carbon dioxide precipitates
rated from their mothers for several hours a day in the early postna- panic so it appears that it is hypercapnia not hypocapnia that leads
tal period demonstrate increased hormonal reactivity to stress and to panic. This makes physiological sense, and has been referred to
increased anxiety-related behaviours (Kalinichev et al., 2002). In as the suffocation false alarm theory (Klein, 1993; Maddock, 2001).
addition, rats raised by mothers with impaired licking and groom- The relationships between panic, acid-base metabolism, and hyper-
ing behaviour are at increased risk of developing anxiety, with ventilation are complex and poorly understood, although it has
cross-fostering experiments indicating that this effect is mainly been argued that chronic hyperventilation is a response to meta-
of environmental origin (Liu et al., 2000). Although it is clear that bolic acidosis (Sikter et al., 2007).
anxiety symptoms and anxiety disorders are due to a combina- There is evidence that anxiety disorders in later life are associ-
tion of genetic and environmental factors, a question that arises is ated with childhood abuse or neglect. Analysis of data from the
whether genetic influences are less important in later life. Gillespie US National Comorbidity Study Replication (NCS-R) showed that
et al. (2004) found that this was not the case for late-life anxiety and childhood sexual abuse was associated with social phobia, panic
depression, although there were differences between genetic influ- disorder, GAD, and PTSD, whereas childhood physical abuse was
ences in men and women. More specifically, there were additional associated with specific phobia and PTSD (Cougle et al., 2010).
midlife and late-life genetic influences in women. Thus, it seems There is evidence also that childhood physical abuse is associated
that genetic effects are still operating strongly in later life. with an earlier age of onset of GAD (Gonçalves and Byrne, 2012a).
From a psychodynamic perspective, anxiety can be understood
as arising out of uncontrolled fear and guilt that has its origins in
Psychological Factors earlier developmental periods. Emotions related to aggression and
Normal anxiety is conventionally divided into trait anxiety and state sexuality can be experienced as dangerous and be repressed, result-
anxiety. Both are dimensional rather than categorical constructs. ing in anxiety symptoms.
Anxiety Symptoms paradox that older people are the least likely group to be either vic-
tims or perpetrators of crime. Of course, older people have lived
Anxiety symptoms are diverse but are conventionally clustered longer than younger people and have thus had the opportunity to
under the three headings of physical, mental, and emotional. observe both more actual crime and more reports of crime. In a
Physical symptoms are multifarious and can be experienced in any small US study, predictors of fear of crime among older people were
bodily system. They include palpitations, dyspnoea, chest tightness, found to be female gender, non-Caucasian ethnicity, depressive
dizziness, sweating, itch, blushing, nausea, teeth grinding, choking, symptoms, and social isolation (Acierno et al., 2004). Housebound
butterflies in the stomach, diarrhoea, frequent urination, fatigue, older women might experience greater levels of fear of violence
weakness, muscle tension, tremor, and insomnia. Mental symp- than other groups (Barnett et al., 2007). In a UK study based on the
toms include worry, rumination, depersonalization, derealization, Whitehall II cohort (middle-aged adults), fear of crime was associ-
poor concentration, and memory impairment. Worry can be about ated with poorer physical and mental health and lower quality of
many different topics. Emotional symptoms include fear, dread, life (Stafford et al., 2007). The specific association of fear of crime
anger, depression, emotional numbness, irritability, and tearfulness. with anxiety disorders in later life needs further study.
Fear can be directed towards many objects or situations. Anxiety is
also associated with a variety of behaviours, including avoidance of Avoidance
feared objects or situations, and repetitive checking or washing. Avoidance behaviour in older people can take many different
forms and clinicians sometimes underestimate its severity and
Panic extent. Older people might avoid participating in everyday social
Panic attacks are conventionally defined as rapidly escalating anxi- activities by referring to commonplace physical complaints. They
ety episodes in which fear and apprehension rise to a crescendo might give up driving a motor vehicle prematurely or stop doing
within 10 min and then subside over several hours. Panic attacks the shopping. The risk for clinicians is that they might misinterpret
are associated with a wide variety of physical symptoms, including this social withdrawal as the inevitable consequence of retirement
shortness of breath, tachycardia, palpitations, a choking sensation, or ageing, when it actually reflects avoidance behaviour associated
chest tightness, sweating, tremor, paraesthesiae, nausea, and hot or with pathological anxiety. Identification of avoidance behaviour is
cold flushes. People experiencing a panic attack commonly cata- important because avoidance makes anxiety worse, leading to its
strophize about their symptoms and fear collapse, stroke, or death. exacerbation and persistence. The persistence of avoidance behav-
Sometimes they think they will ‘go crazy’ or embarrass themselves iour in older people is often supported by well-meaning family
in front of others. People having a panic attack often make fran- members or carers, and sometimes by domiciliary services. Severe
tic attempts to escape from the situation in which the panic attack avoidance behaviour of the sort seen in agoraphobia almost always
occurred. They often go on to develop anticipatory anxiety about means that the patients have one or more supporters who help them
having further panic attacks. maintain their behaviour. Even severe avoidance behaviour can be
concealed for many years before coming to clinical attention, often
Worry precipitated by a change in care arrangements. Sometimes the term
Worry is a core manifestation of generalized anxiety and can be ‘dependency’ is used to describe the situation in which an older
conceptualized as an attempt to solve feared events with uncertain person is overly reliant upon others, although in many instances
outcomes (Borkovec et al., 1983). Overall, older people report fewer this should really be reframed as avoidance behaviour related to
worries than young or middle-aged people (Lindesay et al., 2006; anxiety disorder.
Coelho et al., 2010). Worry in later life is often directed towards
different content areas than in younger adults. While older peo- Somatic symptoms
ple are more likely than younger adults to worry about the health Older people have many more somatic symptoms than younger peo-
and welfare of loved ones, they are less likely to worry about work, ple due to the exponential rise in the prevalence of general medical
health, and interpersonal relationships (Gonçalves and Byrne, problems with normal ageing. Older people are also thought to be
2013). Thus, there appear to be developmental aspects to worry more likely than younger people to minimize psychological symp-
content. At first glance, it might seem counterintuitive that older toms and to convert psychological distress into physical symptoms,
people should report fewer worries than younger people, given although it is unclear the extent to which this is a cohort effect or a
the common challenges of ageing and likely reduced opportunities direct effect of psychophysiological changes associated with ageing.
for older people to habituate through exposure. The consistently It is often difficult for the clinician to disentangle the symptoms
verified reduction in worry frequency with age has been attributed of general medical conditions from somatic symptoms of anxiety.
to several complementary factors, including increased emotional Indeed, both types of somatic symptoms are often present simulta-
regulation, changes in daily demands, the influence of cognitive neously in the same person.
habituation mechanisms, and cohort effects (Gonçalves and Byrne,
2013). Other factors, including data censoring due to premature Obsessions and compulsions
death or institutionalization of worriers, and the use of assessment Obsessional ruminations and compulsive rituals are core features of
measures that are inappropriate for older people, might also explain OCD but are also found in other mental disorders, especially major
some of the measured differences in worry prevalence (Gonçalves depressive disorder. Obsessional ruminations commonly involve
and Byrne, 2013). aggressive, sexual, repugnant, immoral, or religious themes, on the
Fear of crime among older people, especially older women, is a one hand; or doubt, contamination, checking, superstition, sym-
particular category of worry that is commonly promoted by mass metry, or order, on the other (Garcia-Soriano et al., 2011). They
media (Reiner, 2007). In most countries, it is associated with the seem to exist on a continuum with intrusive thoughts experienced
by individuals without OCD. The level of insight that accompanies have less trouble entering feared situations when accompanied by
obsessional phenomena varies, with some people with OCD believ- a trusted family member or friend. As time goes by, people with
ing that their intrusive thoughts will come true. In some cases of agoraphobia often gradually increase the range of situations they
severe OCD, obsessional thoughts may be difficult to distinguish avoid, until they are quite limited in the situations they feel they can
from delusions. Compulsive behaviours also appear to exist on a enter safely. People with agoraphobia tend to present for clinical
continuum with normal rituals and they can sometimes be difficult care many years after the first onset of their panic attacks or avoid-
to distinguish from one another (Muris et al., 1997). Rituals gener- ance behaviour. Agoraphobia is often a hidden problem in the sense
ally involve similar themes to obsessional ruminations, and often that both patients and their family supporters tend to conceal it due
involve washing and checking behaviours. True compulsions are to embarrassment.
generally more frequent, more intense, and associated with more
affect than normal rituals. Starcevic et al. (2011) investigated the Social phobia
functions of compulsions in adults with OCD and, not unexpect- People with social phobia experience fear, embarrassment, or
edly, found that compulsions are usually performed automatically humiliation in social situations, and experience anticipatory anxi-
to decrease stress or anxiety. ety about forthcoming events. They avoid situations in which they
feel other people may judge them. When avoidance is not feasible,
Flashbacks they feel anxious and self-conscious around other people. Social
During a flashback the older person re-experiences a traumatic phobia can be restricted to a certain social situation, such as pub-
event during the waking state, in contrast to dreams or nightmares lic speaking, or it can be generalized. Social phobia can be pre-
in which this occurs whilst asleep. Flashbacks are a core component cipitated by another psychiatric disorder, such as major depressive
of acute stress disorder and PTSD. disorder or panic disorder. It is also commonly associated with sub-
stance use disorders, particularly alcohol abuse and dependence.
Depersonalization and derealization Although social phobia is not an uncommon diagnosis in epidemi-
In depersonalization, patients feel as though they are watching ological surveys, older people with this condition rarely present for
themselves as if in a dreamlike state. There is a sense of unreality treatment. This might be because older people are not particularly
about the self. In contrast, in derealization, patients feel a sense of disabled by social phobia due to reduced social and occupational
unreality about the outside world. Depersonalization and dereali- demands in later life. There has been some controversy about the
zation are common phenomena in people with anxiety disorders, nosological status of social phobia, including concerns about medi-
and are part of the reason that panic attacks can be associated with calization of normal shyness.
a sense of impending doom or the thought that one is losing one’s
mind. Generalized anxiety disorder
As conceptualized in DSM-IV, the diagnosis of GAD requires the
Anxiety Disorders patient to have multiple worries that are experienced as excessive
and that have persisted for 6 months or more. Older people with
Panic disorder GAD report similar worry content to older worriers without GAD,
To meet DSM-IV diagnostic criteria for panic disorder, panic but greater worry frequency (Diefenbach et al., 2001). This obser-
attacks must be accompanied by at least four symptoms from a vation highlights the issue of whether GAD should be conceptual-
list, must peak within 10 min, and must be uncued. The practi- ized as a dimensional disorder rather than a categorical disorder. In
cal problems with this definition are that panicky feelings are comparison with nonpathological worriers, those with GAD differ
not always accompanied by the requisite four symptoms, do not also in the extent to which they can control their anxiety and in the
always rise to a crescendo within 10 min, and are often cued by level of distress associated with their anxiety (Ruscio, 2002). Worry
contextual factors. As a consequence, panicky episodes are more is the predominant symptom in GAD, to the extent that Andrews
prevalent than narrowly defined panic attacks, and subthreshold et al. (2010) have argued that the disorder should be renamed gen-
panic disorder is more prevalent than panic disorder. So panic in eralized worry disorder, major worry disorder, or pathological worry
older people is a larger clinical problem than the epidemiological disorder, in DSM-5. GAD appears to precede major depressive dis-
data would suggest. Whilst panic attacks are the defining feature of order (MDD) more often than MDD precedes GAD. Gonçalves
panic disorder, they occur in many other mental disorders, includ- et al. (2011) found that GAD in older people was associated with
ing major depressive disorder, GAD, alcohol withdrawal, and functional limitations, psychiatric comorbidity, and increased
delirium. Older people with panic disorder are seen in the emer- medication intake.
gency department, where panic attacks mimic a variety of medi-
cal emergencies, including asthma and acute coronary syndrome. Specific phobia
Panic attacks and panic disorder commonly precede the develop- Specific phobia involves avoidance of feared objects (e.g. spiders,
ment of agoraphobia. moths) or situations (e.g. flying, lifts). Although specific phobia is
highly prevalent, it is not often the focus of therapeutic attention in
Agoraphobia old age psychiatry practice, mainly because it is generally associ-
Agoraphobia involves fear and avoidance of situations from which ated with little distress or disability, apart from when the patient is
escape might be difficult or embarrassing, or in which help might actually confronted by the phobic stimulus. Older people are often
not be available if a panic attack should occur (American Psychiatric able to avoid their phobic stimuli. Some specific phobias resolve
Association, 2000). People with agoraphobia often avoid entering without formal treatment, but some cause significant distress and
supermarkets, restaurants, or crowded shopping malls. They often disability.
Fear of falling is a relatively distinct psychosyndrome in later life Mixed anxiety and depression
with features of both specific phobia and agoraphobia (Coelho et al., Some older people experience a mixture of anxiety and depressive,
2010). It is associated with considerable disability for many older although the prevalence of these mixed states is in dispute. As pre-
people. At least 30% of persons aged 65 years and over fall each viously noted, there is some comorbidity between GAD and major
year and one-fifth of these falls require medical attention (Stel et al., depressive disorder. In addition, there is a group of older people
2004). Falls are even more prevalent in older people with dementia who experience symptoms of both anxiety and depression with-
(Allan et al., 2009) and in those taking psychotropic medication out meeting diagnostic criteria for a mood or anxiety disorder. The
(Hill and Wee, 2012). Approximately one-quarter of those who fall diagnostic entity of mixed anxiety and depressive disorder already
develop fear of falling (McClure et al., 2005). exists in ICD-10 and has been proposed for DSM-5. Because of
Post-traumatic stress disorder the dimensionality of both anxiety and depressive symptoms, and
because mixed subthreshold states appear to be common among
PTSD develops in about 15–20% of older people exposed to older people, this is a useful diagnostic category.
life-threatening trauma. Classically, PTSD follows combat exposure,
armed holdup, rape, and violent assault. It is also seen in emergency Anxiety and agitation in dementia
workers required to deal with multiple trauma cases, and follow-
One question that arises is can anxiety be distinguished from the
ing disasters. In older people, PTSD has often developed earlier in
agitation associated with dementia? It is difficult to give a straight-
life during military service, but can also develop de novo following
forward answer to this question because both anxiety and agitation
adverse life events in later life. Sometimes PTSD does not develop
are words that describe complex arrays of human behaviours. There
immediately following exposure to the traumatic event, but follow-
is clearly some overlap between the constructs (Twelftree and Oazi,
ing an extended delay. Characteristic symptoms of PTSD include
2006), but also many differences between the two. The term agita-
intrusive re-experiencing of the traumatic event through flashbacks
tion is commonly used to describe a broad range of behaviours in
and nightmares. These intrusive phenomena are usually accompa-
people with dementia. Agitation is also used to describe a category
nied by hyperarousal, emotional numbness, and avoidance behav-
of behaviour in people with delirium and a different category of
iour. Symptoms of hyperarousal include impaired concentration,
behaviour in people with depression. So the legitimate question
lowered startle threshold, hypervigilance, and insomnia. Emotional
arises, can anxiety be reliably distinguished from agitation? In a
numbness leads individuals to feel that they can no longer experi-
large factor analytic study of untreated patients with dementia due
ence emotions normally and that their interpersonal relationships
to Alzheimer’s disease, Spalletta et al. (2010) found that anxiety and
are impaired. Avoidance behaviour commonly involves avoiding
agitation loaded on separate factors. Anxiety loaded on a factor
the scene of the traumatic exposure or avoiding reminders of the
with depression, whereas agitation loaded on a factor with irritabil-
trauma. Many other symptoms occur in PTSD, including disso-
ity and aberrant motor behaviour. Although this finding alone is
ciation and substance abuse. In older people, PTSD has often been
unlikely to settle the matter, it does suggest that these terms should
present for many years following combat exposure or wartime rape
not be conflated.
and other atrocities. Older refugees are particularly at risk of PTSD.
Older victims of natural disasters such as floods, fires, and earth- Anxiety and suicide
quakes are also at risk of PTSD.
There is robust evidence linking suicide attempts with anxiety
Obsessive-compulsive disorder disorders. Using data from the US NESARC study, Nepon et al.
(2010) found that among adults reporting a lifetime history of sui-
OCD is characterized by obsessional ruminations and compulsive cide attempts, 70% had an anxiety disorder. In multivariate mod-
rituals. Obsessions are thoughts, images, and impulses that lead to els, PTSD and panic disorder were independently associated with
psychological distress. Patient experience obsessions as being the suicide attempts. Individuals with PTSD or panic disorder who
product of their own mind and generally have insight into them. also had a personality disorder had an even greater risk of suicide
Compulsions can be observable behaviours (rituals) or unobserv- attempts (Nepon et al., 2010).
able behaviours (mental compulsions). Compulsions are consid-
ered to represent an attempt by the patient to undo obsessional
thoughts, prevent harm, or reduce anxiety. Although obsessions Assessment
and compulsions are the defining characteristics of OCD, they do Every psychiatric assessment interview should include screening
occur in other mental and neurological disorders. questions for anxiety symptoms, given the high prevalence of anxi-
Of particular relevance to OCD in older people, hoarding is often ety symptoms and anxiety disorders. At a minimum, worry, panic,
performed because of a perceived need for collected objects rather obsessions, and compulsions should all be covered. It is important
than for reasons common to other compulsions. Hoarding behav- to be aware of limited-symptom panic attacks. These subthreshold
iours appear to occur commonly in the general population (Timpano panic attacks are characterized by fewer than the four panic symp-
et al., 2011) and are linked to compulsive buying behaviours (Mueller toms required by DSM-IV definition of a panic attack. The clinician
et al., 2009). There is also evidence that hoarding is associated with should also make specific inquiry about the nature and extent of
multiple neurological and psychiatric disorders, including depression avoidance behaviour in older people presenting with anxiety symp-
and anxiety (Mataix-Cols et al., 2011; Reid et al., 2011). Hoarding toms. Avoidance behaviour can manifest as ‘taking to the bed’ with
also overlaps with severe domestic squalor (Snowdon and Halliday, nonspecific somatic symptoms, or be more subtle or camouflaged
2011). Moral or religious scrupulosity is often viewed as a symptom in some way, perhaps with the older person declining routine social
of OCD. Although more commonly encountered in older members invitations, only leaving the home in the company of others, not
of the clergy, it can be seen among laity as well. watching the television news, or avoiding talking about or even
thinking about certain topics. As anxiety increases the risk of sui- person’s likely alcohol intake. Clinicians should maintain a high
cide attempts, the clinician should ensure that suicidality is cov- index of suspicion for substance abuse and dependence in older
ered during the assessment of older people with anxiety. Anxiety people with anxiety disorders.
often flares up in response to adversity, so assessment of exposure
to adverse life events is essential. A life-history approach is often Medications associated with anxiety
useful to chart the correspondence between life events and anxiety Many medications used to treat mental disorders and general med-
symptoms. If a behavioural intervention is intended, then a behav- ical conditions are associated with anxiety symptoms, and some-
ioural analysis is a necessary preparatory step. times the most effective intervention is the cessation of unnecessary
medication. Sympathomimetics, corticosteroids, oestrogen, anti-
Informant interviews histamines, interferon, TNF-alpha, and thyroid hormones have all
Informant interviews are an essential component of the assessment been associated with anxiety symptoms. In addition, psychotropic
of older people with suspected cognitive impairment or dementia. drugs, including anxiolytics, antidepressants, mood stabilizers, and
They are also useful in older people with anxiety disorders. Some antipsychotics can cause anxiety symptoms.
older people minimize the extent of their avoidance behaviour or
the relationship of this to their anxiety symptoms. An appropriate Physical work-up
informant, usually a close family member, can often shed useful light Older people who present with anxiety symptoms for the first time
on this avoidance behaviour. An informant can also often provide in later life should be investigated for underlying general medical
collateral information about the use of substances, including alco- problems. Even those older people who have a distant past his-
hol and benzodiazepines. In most countries, older people are free to tory of anxiety symptoms but who have been symptom-free for
consult as many different doctors as they wish. In so doing, they are many years should undergo screening tests for potential underly-
at liberty to obtain multiple prescriptions for sedative and hypnotic ing general medical problems. At a minimum, a systems review
drugs. When older people with anxiety disorders are seen in special- should be undertaken and vital signs should be recorded. In
ist settings, they have generally been referred by only one of the doc- first-onset cases of anxiety disorder in later life, more detailed
tors with whom the patient has had contact. In such circumstances, neurological and cardiovascular examination is prudent. Selected
an informant can provide an important insight into the drug-seeking laboratory tests, including full blood examination, serum elec-
or drug-using behaviour of the patient. Of course, consent must be trolytes, serum glucose, thyroid stimulating hormone (TSH), and
obtained from the patient before consulting an informant. urinalysis should be ordered. If psychotropic medication is to be
prescribed, then pretreatment liver function tests and an elec-
Rating scales trocardiogram (ECG) should be added to the list. If there is any
It is often useful to measure the severity of anxiety during assess- suggestion of substance abuse then a urine drug screen should be
ment or treatment. Many scales are available to measure anxiety ordered. Electroencephalography (EEG) can help exclude epilepsy
in adults, but very few have been specifically designed for use in and delirium, if these are suspected clinically. Neuroimaging has a
older people. Self-report scales for use by older people, particu- low yield in late-life anxiety disorders, but if underlying cerebrov-
larly the frail older person, require certain modifications. Ideally, ascular disease is suspected then a structural brain scan (MRI or
such measures should avoid multiple somatic items, reversed CT) might be warranted to assess the extent of this. Serological tests
items, and complex response scales (Dennis et al., 2007). The Short for syphilis, hepatitis, HIV, and other infections should be reserved
Anxiety Screening Test (SAST) is a 10-item self-report measure for particular cases in which the history suggests greater than nor-
with a four-item response scale (Sinoff et al., 1999). The SAST was mal risk. Older patients with unusual presenting symptoms should
designed specifically for use in older people, although it does con- undergo more detailed physical investigation.
tain multiple somatic items. The Geriatric Anxiety Inventory (GAI)
is a 20-item self-rating scale designed to assess generalized anxiety Management
in older people (Pachana et al., 2007b). It employs straightforward
language and dichotomous (agree/disagree) ratings, and minimizes There is actually quite a lot for the clinician to do when managing
the use of somatic items. There is also a five-item version for screen- anxiety in older people. Paying close attention to some straightfor-
ing (Byrne and Pachana, 2011). The GAI is now available in multi- ward general principles is likely to greatly assist the patient, so these
ple languages. will be described here first. Anxiety of moderate severity or greater
is likely to require formal treatment with psychological treatment,
Substance abuse pharmacological treatment, or both. Several classes of drugs have
Older people with clinically significant anxiety often abuse alcohol been used in the treatment of anxiety symptoms and anxiety dis-
or benzodiazepines. They frequently minimize their intake of these orders in older people and several types of behavioural and psy-
substances or minimize the likely link between their intake and chological treatments have been used. A stepped-care approach
their anxiety symptoms. Substance abuse and dependence can be may be effective in reducing or even preventing symptoms (van’t
the direct cause of anxiety symptoms or can be the consequence of Veer-Tazelaar et al., 2009).
self-medication of a pre-existing anxiety disorder. In older people
with mild cognitive impairment or dementia, particularly those liv-
General measures
ing alone, alcohol abuse can be unsuspected, even by members of Substance use
their immediate family. In office practice, it helps to have a knowl- Both excessive consumption of nicotine, caffeine, and alcohol
edgeable informant who can provide collateral information about and withdrawal from these substances are commonly associ-
substance use. A domiciliary visit can often help clarify an older ated with anxiety symptoms. People with anxiety disorders also
self-medicate with these substances. Some older people require temperature, and ensuring the sleeping environment is both quiet
specific advice about minimizing their exposure to these legal sub- and dark. It is also important to reserve the bedroom for sleep and
stances. Medications used to treat general medical conditions may sexual activity, and not remain in bed if unable to sleep. In this way,
also be associated with anxiety and insomnia. These include sym- the bedroom is associated with sleep rather than wakefulness. It is
pathomimetics, corticosteroids, thyroid hormones, antimicrobials, also recommended that people with a sleep initiation problem do
oestrogen, and antihistamines. Diuretics commonly disrupt sleep not undertake vigorous physical exercise in the hours immediately
and are generally best taken earlier in the day. Stimulants includ- before bedtime, although physical exercise during the day generally
ing methylphenidate and dexamphetamine are used occasionally assists initiation and maintenance of sleep (Loprinzi and Cardinal,
to treat adult attention deficit hyperactivity disorder (ADHD) and 2011). Because of its potential to stimulate some people with
apathy in older people. These medications are commonly associ- insomnia, it is recommended that television should not be watched
ated with weight loss, insomnia, and anxiety. Some antidepressant in the bedroom. From a behavioural perspective, it is useful to have
medications have stimulant effects in vulnerable older people, and a bedtime routine that conditions one to expect sleep.
selective serotonin reuptake inhibitors (SSRIs) commonly cause an
initial increase in anxiety and insomnia. Psychological interventions
Although older people are less likely than young or middle-aged For pragmatic reasons, psychotropic medications are commonly
people to use illicit substances, there are cohort effects at work that used as first-line treatments for anxiety disorders in older adults,
might lead to increased rates of substance abuse in older people in particularly in primary care settings. However, expert opinion gen-
the future (Wu and Blazer, 2011). Cannabis, amphetamine, cocaine, erally advises the use of a psychosocial intervention for anxiety
and opiates can all cause anxiety symptoms. Thus, a careful review disorders of mild to moderate severity and combination treatment
of current substance use, both licit and illicit, is recommended. with a psychosocial intervention and a psychotropic medication for
General medical conditions more severe anxiety disorders. In old age psychiatry services, most
anxiety disorders will be at least moderate in severity and will often
Many general medical conditions are associated with anxiety, either
occur in comorbid relation to other disorders, including mood
directly or indirectly. Although a direct mechanism has not been
disorders, psychotic disorders, and substance use disorders. Thus,
fully elucidated, it is likely to involve cytokine production at sites
combination treatment with a psychosocial and psychopharmaco-
of inflammation. An indirect mechanism involves a psychologi-
logical intervention will generally be indicated in these specialist
cal reaction to pain, discomfort, and disability, as well as increased
mental health care settings.
dependency upon others and a sense of an altered future, associ-
ated with many diseases. Hyperthyroidism, paroxysmal supraven- Therapeutic alliance
tricular tachycardia, atrial fibrillation, asthma, chronic obstructive A supportive relationship with a mental health clinician is likely
lung disease, epilepsy, and systemic lupus erythematosis have all to be of considerable value to the anxious older person, and time
been associated with anxiety disorder. Thus, optimal management spent developing a durable therapeutic alliance is well worth the
of patients’ general medical conditions is an integral component of investment. Psychological and behavioural interventions that pro-
management of their anxiety disorder. vide older people with an improved set of coping skills are likely to
Physical activity be of enduring benefit to them. Older people with anxiety disorders
are often helped by a longer-term relationship with a mental health
In older people, physical activity has general benefits for health,
worker. Patients who are allocated to medical personnel who move
particularly in maintaining or improving bone density, muscle
rapidly through different treatment teams as part of their training
strength, and cardiovascular fitness. In addition, physical activi-
should also be allocated a case manager or staff psychologist who
ties such as walking, cycling, and swimming have anxiolytic effects.
has a continuing appointment. In this way, continuity of care can
Exposure to interoceptive stimuli, such as increased rate and depth
be provided. An effective therapeutic alliance is likely to be particu-
of breathing, increased heart rate, and perspiration, acts as a form
larly important if the clinician will be recommending to the patient
of nonspecific exposure in people with anxiety disorders. This
either formal exposure therapy (e.g. systematic desensitization) or
exposure to interoceptive stimuli is associated with reduced anxi-
a reduction in avoidance behaviour. The perceived counterintuitive
ety sensitivity (Barbour et al., 2007).If the physical activity is under-
nature of these interventions requires patients to trust their clini-
taken outdoors, or in a public setting such as a gym, it also acts as
cians to a greater extent than usual.
exposure to exteroceptive stimuli.
Psychoeducation
Sleep hygiene
Many older people with anxiety symptoms and anxiety disorders
Sleep architecture changes gradually with increasing age and sleep
suspect that they have serious physical problems. As a conse-
continuity declines in many older people (Harbison, 2002). Some
quence, psychoeducation is an essential component of any anxiety
people interpret these normal changes as indicative of a sleep dis-
management plan. In some cases, psychoeducation alone leads to a
order, when they are not. Nevertheless, insomnia does reduce the
clinically significant improvement in anxiety symptoms as the older
quality of life of many older people, particularly those who suppress
person reframes their problem. In people with cognitive impair-
their REM sleep with hypnotic medication, or those who worry
ment or intellectual disability it is essential that psychoeducation is
excessively about their sleep continuity or duration. Improvements
provided to their family and carers.
in sleep duration and quality can lead to reduced anxiety. Routine
sleep hygiene recommendations include ensuring that the sleeping Relaxation training
environment is satisfactory. In practice, this means having a com- Relaxation training is an essential component of the treatment of
fortable bed and pillow, maintaining the bedroom at an appropriate anxiety symptoms and anxiety disorders in older people. Many
different types of relaxation training have been described, although In OCD, compulsive rituals are generally treated with exposure and
three techniques are in common use: progressive muscular relaxa- response prevention. This is an effective treatment when rigorously
tion, reciprocal inhibition via visual imagery, and controlled breath- applied. Obsessional ruminations can be treated with either anti-
ing. Progressive muscular relaxation is most suitable for those older depressant medication or CBT, or both. Because exposure therapy
people without significant sarcopenia or joint deformity, whereas does lead to a temporary increase in psychological distress, it must
the other two techniques can be used by most older people without be undertaken with the informed consent of the patient. Deception
cognitive impairment. is not part of exposure therapy for anxiety disorders.
Cognitive behaviour therapy Interpersonal psychotherapy
There is a growing body of evidence in support of the use of cog- Interpersonal psychotherapy (IPT) uses four principal models to
nitive behaviour therapy (CBT) to treat GAD in older people. A engage the patient. These models address (1) issues that arise in
meta-analysis of CBT versus control conditions for GAD in older relation to grief following loss of a loved one, (2) conflict in rela-
people identified 11 small clinical trials with usable data (Gonçalves tionships, (3) adapting to change in life circumstances, and (4)
and Byrne, 2012b). CBT was superior to control conditions with social isolation. It is not difficult to see that generalized anxiety
a pooled odds ratio of 0.33 (95% CI: 0.17–0.66). The superiority symptoms and phobic avoidance behaviour could easily arise in
of CBT was demonstrated in those studies that used usual care or relation to each of these challenges that occur commonly in later
a waiting list as the comparison condition, but not in those that life. Despite this, there are limited data in support of the use of IPT
compared CBT to an active comparator. This observation raises in the management of anxiety disorders in older adults, although
questions about the specificity of CBT and suggests that other it is used for the management of depression in late life (Reynolds
interventions could have similar efficacy. Laidlaw et al. (2003) and et al., 1999).
Pachana et al. (2007a) have detailed suitable modifications for the
use of CBT or its components in older adults. These include an Problem-solving therapy
increased number of sessions to allow more time for participants to A reduced capacity to solve everyday problems has been linked,
develop an understanding of the therapy and its associated home- directly and indirectly, to anxiety. The indirect link is with impaired
work tasks, explicit learning aids, and manuals with larger print problem-solving and worry, which in turn lead to anxiety (D’Zurilla
size. CBT techniques are also used in the management of several and Nezu, 2007: 80–81). The clinical approach to problem-solving
other anxiety disorders in older people, although without the same therapy involves iterative cycles of problem definition and formu-
level of empirical evidence. lation, generation of alternative solutions through brainstorming
and other techniques, and identification of the most effective solu-
Behaviour therapy tion (D’Zurilla and Nezu, 2007: 95–148). Although much of the
In some older people with anxiety disorders, it is not practicable to research on problem-solving therapy has been conducted in young
implement the cognitive components of CBT. This can occur when and middle-aged persons, or in older people with depression rather
the older person has significant cognitive impairment or intellec- than anxiety, there has been some recent work in older adults with
tual disability. It can also occur when older people do not view their anxiety. In a pragmatic randomized controlled trial conducted in the
problem in psychological terms and remain impervious to psych- Netherlands, 6 weeks of bibliotherapy involving problem-solving
oeducation. In such situations it makes sense to apply behavioural techniques and systematic desensitization was superior to treatment
interventions alone. The principles of behavioural activation can be as usual in older primary care patients with an anxiety disorder but
employed (Ekers et al., 2011; Snarski et al., 2011). These emphasize no comorbid depression (Seekles et al., 2010). In contrast, a Hong
graded activity scheduling and a reduction in avoidance behaviours. Kong clinical trial that compared brief problem-solving therapy
In one study involving adults with major depression, behavioural with group viewing of health videos in primary care patients with
activation outperformed CBT and was comparable to antidepres- elevated Hospital Anxiety and Depression Scale scores did not find
sant medication (Dimidjian et al., 2006). There is preliminary evi- a significant difference between treatment arms (Lam et al., 2010).
dence that such an approach might be applicable to older adults There is evidence of the superiority of problem-solving therapy to
with anxiety disorders as well (Pachana et al., 2007a). supportive therapy in older people with depression (Alexopoulos
Exposure therapy is the treatment of choice for agoraphobia, et al., 2011), but further research is needed in older people with
social phobia, and specific phobia. As flooding is rarely appropri- anxiety disorders in order to establish the place of PST in late-life
ate or practicable, systematic desensitization is the usual exposure anxiety disorders.
method employed. This involves the preparation of a hierarchical
list of phobic cues in close consultation with the patient. The pho- Other types of therapy
bic cues are listed from least aversive to most aversive. Exposure If anxiety symptoms arise in the context of interpersonal conflict,
tasks are then designed to enable the patient to systematically work within a marital relationship or its equivalent, or within a family,
through the list from least aversive to most aversive. An individual- then consideration should be given to couples therapy or family
ized approach is necessary, as the time needed to master each step therapy. Psychoanalytic psychotherapy has been used historically
will vary from patient to patient. Many patients find themselves to treat chronic anxiety symptoms, but not commonly in older peo-
unable to engage in exposure therapy without the assistance of a ple. There is little conventional evidence for this approach in this
confederate. This should usually be a clinician rather than a family age group.
member, as family members often overtly or covertly sabotage expo-
sure activities. However, the ultimate aim of systematic desensitiza- Medication
tion is for patients to be able to repeatedly expose themselves to the Drug therapy is generally unhelpful in the management of specific
phobic object or situation without the assistance of another person. phobias, which usually respond to behaviour therapy, particularly
exposure-based treatments employing systematic desensitization. challenge for prescribers is that patients who have had experience
Drug therapy does have a role in the management of the other with the rapid onset of action of the benzodiazepines are often
anxiety disorders, particularly in combination with some type of dissatisfied with the relatively slow onset of action of antidepres-
psychotherapeutic intervention. A meta-analysis of clinical trials of sants. There appears to be little to choose between the available
psychotropic medication for GAD in older people identified nine SSRI antidepressants in terms of their anxiolytic effects. However,
brief clinical trials with usable data, including eight trials with a paroxetine has more anticholinergic effects than the other drugs
placebo comparator (Gonçalves and Byrne, 2012b). Medication in this class and for this reason is generally less favoured in older
was superior to the control condition with a pooled odds ratio of people. The long half-life of fluoxetine is associated with both risks
0.32 (95% CI: 0.18–0.54). Benzodiazepines, antidepressants, and and benefits, but fluoxetine is generally not the first choice in older
quetiapine all demonstrated short-term efficacy. people. Citalopram and escitalopram are associated with prolon-
Benzodiazepines gation of the electrocardiographic QTc interval and it has been
recommended that they be restricted to 20 mg and 10 mg daily,
Benzodiazepines operate at the GABAA receptors of the brain’s
respectively, in older people (Medicines and Healthcare Products
main inhibitory neurotransmitter GABA (γ-aminobutyric acid).
Regulatory Agency, 2011). Fluvoxamine is a sedating SSRI, which
There is crossreactivity with alcohol at this receptor complex.
is often well tolerated in anxious individuals, although it does have
Benzodiazepines remain in widespread use in older people with
a greater potential for drug–drug interactions than the other drugs
anxiety disorders, although the evidence in support of their use
in its class due to effects on the hepatic enzyme system.
is rather limited. They appear to have short-term efficacy in the
The SSRIs have significant adverse effects in older people. These
management of panic attacks, generalized anxiety, and insomnia.
include initial worsening of anxiety and insomnia, hyponatraemia,
However, their long-term effectiveness is uncertain and their use
sexual dysfunction, and falls. They are also associated with increased
is associated with a range of adverse effects in older people, includ-
bruising and bleeding in patients on anticoagulants and antiplatelet
ing anterograde amnesia, confusion, and falls. Tolerance develops
drugs. Hyponatraemia occurs most commonly in women on thi-
rapidly and abrupt cessation of benzodiazepines is often associated
azide diuretics, and often within 6 weeks of initiating antidepres-
with withdrawal phenomena. Both patients and clinicians find it
sant treatment. The symptoms of hyponatraemia can mimic those
difficult to distinguish benzodiazepine withdrawal phenomena
of worsening anxiety or depression, and can also include fatigue
from the pre-existing anxiety, thus perpetuating the use of these
and confusion. The true prevalence of hyponatraemia is unknown
agents. Benzodiazepines suppress REM sleep (Hemmeter et al.,
but it is a frequent occurrence on inpatient wards. The clinician
2000) and rebound effects can be encountered following their
is advised to check the serum sodium in older patients on antide-
withdrawal. Withdrawal phenomena appear most severe following
pressant medication whose clinical condition is deteriorating. All
long-term use of high doses of short-acting agents. A small propor-
antidepressants can precipitate mania and hypomania in vulnerable
tion of older people prescribed benzodiazepines develop paradoxi-
individuals, but this is relatively uncommon and is not a reason to
cal effects such as excitement or agitation. When benzodiazepines
avoid these drugs in older people with clinically significant anxiety.
are used in the management of anxiety symptoms, including
The reported increased risk of suicide in patients commenced on
insomnia, they should ideally be used for short periods (2–4 weeks)
SSRI antidepressants does not seem to be relevant to older people,
whilst other treatments are introduced. Drugs such as oxazepam
in whom treatment with SSRIs is associated with a reduced suicide
and temazepam, which do not undergo hepatic oxidation, are gen-
risk (Barbui et al., 2009).
erally preferred in older people. Because of its slow rate of absorp-
Other classes of antidepressants, including the serotonin and
tion, oxazepam is useful for the treatment of middle insomnia,
noradrenaline receptor inhibitors (e.g. venlafaxine, desvenlafaxine,
whereas temazepam is more useful for initial insomnia. Modern
duloxetine), drugs with mixed receptor activity (e.g. mirtazapine),
benzodiazepine replacements such as zopiclone and zolpidem have
and the older tricyclic drugs are also used in the pharmacological
similar pharmacological properties to the benzodiazepines and
management of anxiety in older people. Among the tricyclic anti-
seem to confer few advantages. Although the benzodiazepines have
depressants, nortriptyline appears to have the best safety profile in
short-term efficacy in many of the anxiety disorders, they are con-
older patients. Historically, clomipramine has had a special place
sidered ineffective in OCD and of limited value in PTSD.
in the treatment of severe OCD, although its adverse effect profile
Antidepressants makes it challenging to use in many older people. However, it is not
Antidepressants are preferred for the pharmacological treatment of entirely clear whether clomipramine is superior to intensive expo-
the anxiety disorders, although their use in this context does have sure and response prevention (Foa et al., 2005). The noradrenaline
some challenges. The SSRI antidepressants are the usual first-line reuptake inhibitor reboxetine and the older, irreversible, monoam-
medications in older people with panic, social phobia, GAD, ine oxidase inhibitor drugs are not usually used in the management
OCD, and PTSD. They appear to be of no value in specific phobia. of anxiety disorders in older people. The serotonin antagonist and
SSRIs take effect much more slowly than benzodiazepines, with reuptake inhibitor (SARI) trazodone does have anxiolytic effects,
longer-term effects on the sensitivity of autoreceptors leading to but it is not well studied in older people with anxiety disorders.
increased availability of synaptic 5-HT. The SSRIs are commonly The 5HT1A receptor partial agonist buspirone has been used in
associated with an initial increase in tremor, insomnia, and anxiety, treatment of GAD in older people for many years. It has a slow
and anxious older patients should be warned about this, lest they onset of action and seems to be most useful in benzodiazepine
become concerned that their underlying disorder is deteriorating. naïve patients. It has a number of drug–drug interactions and
It is sometimes necessary to use a low-dose benzodiazepine for a should not be used by people who are currently taking or who have
week or two to get the patient through the initial adverse effects recently been taking monoamine oxidase inhibitors. Agomelatine is
of an SSRI, particularly in somatically focused patients. A greater an antidepressant with melatonergic (MT1, MT2) receptor agonist
and 5-HT2C receptor antagonist properties that appears to reduce References

sleep latency without causing daytime drowsiness. There is some Acierno, R., et al. (2004). Predictors of fear of crime in older adults. Journal of
limited evidence for its efficacy in GAD in adults (Stein et al., 2008), Anxiety Disorders, 18, 385–96.
but there have been no clinical trials for this indication in older Ahearn, E.P., et al. (2011). A review of atypical antipsychotic medications
people. for posttraumatic stress disorder. International Journal of Clinical
In older people, antidepressants are generally commenced at Psychopharmacology, 26, 193–200.
half the usual starting dose used in younger adults and the dose Alexopoulos, G.S., et al. (2011). Problem-solving therapy and supportive
is titrated upwards more slowly. Sometimes the final antidepres- therapy in older adults with major depression and executive dysfunction:
sant dose is lower than that used in younger people, although many effect on disability. Archives of General Psychiatry, 68, 33–41.
Allan, L.M., et al. (2009). Incidence and prediction of falls in dementia: a
older people require the same final doses as younger people to
prospective study in older people. PLoS One, 4, e5521.
achieve efficacy. As a broad generalization, the antidepressant dose
required to treat anxiety disorders is often the same or greater than of mental disorder, 4th Edition, Text Revision. American Psychiatric
that required to treat depression. Publishing, Arlington, VA.
Antipsychotics Andrews, G. and Hobbs, M.J. (2010). The effect of the draft DSM-5 criteria
for GAD on prevalence and severity. Australian and New Zealand Journal
Atypical antipsychotic drugs, particularly quetiapine and risperi- of Psychiatry, 44, 784–90.
done, are sometimes used in the management of the symptoms Andrews, G., et al. (2002). The treatment of anxiety disorders: clinician guides
of both GAD and PTSD in adults. However, evidence for the use and patient manuals. Cambridge University Press, Cambridge.
of antipsychotics for these indications in older adults is meagre Andrews, G., et al. (2010). Generalized worry disorder: a review of DSM-IV
(Ahearn et al., 2011; Gonçalves and Byrne, 2012b). Cerebrovascular generalized anxiety disorder and options for DSM-V. Depression and
adverse events (strokes and transient ischaemic attacks) and excess Anxiety, 27, 134–47.
mortality have been described in association with the use of antip- Archer, N., et al. (2007). Premorbid personality and behavioural and
sychotic medication in older nursing home residents with dementia psychological symptoms in probable Alzheimer disease. American
Journal of Geriatric Psychiatry, 15, 202–13.
complicated by behavioural and psychological symptoms (Sacchetti
Australian Bureau of Statistics (2007). National survey of mental health and
et al., 2010), although a case-control study conducted in patients wellbeing. ABS Cat. No. 4326.0. <http://www.abs.gov.au>.
with dementia in the primary care setting did not identify similar Barbour, K.A., Edenfield, T.M., and Blumenthal, J.A. (2007). Exercise as a
problems (Laredo et al., 2011). Nevertheless, it would seem prudent treatment for depression and other psychiatric disorders. Journal of
to minimize the use of antipsychotic medication in older people Cardiopulmonary Rehabilitation and Prevention, 27, 359–67.
with nonpsychotic disorders in the absence of long-term efficacy Barbui, C., Esposito, E., and Cipriani, A. (2009). Selective serotonin reuptake
and safety data. inhibitors and risk of suicide: a systematic review of observational
studies. Canadian Medical Association Journal, 180, 291–7.
Anticonvulsants Barnett, K., et al. (2007). Older women’s fear of violence: the need for
There is some evidence for the use of anticonvulsants in adults with interventions that enable active ageing. Journal of Women and Aging, 19,
anxiety disorders, including the use of gabapentin in social phobia, 179–93.
lamotrigine in PTSD, and pregabalin in social phobia and GAD Borkovec, T.D., et al. (1983). Preliminary exploration of worry: some
(Mula et al., 2007). However, there is a scarcity of evidence relat- characteristics and processes. Behaviour Research and Therapy, 21,
9–16.
ing specifically to the use of anticonvulsants in older people with
Bryant, C., Jackson, H., and Ames, D. (2008). The prevalence of anxiety in
anxiety disorders.
older adults: Methodological issues and a review of the literature. Journal
Other drugs of Affective Disorders, 109, 233–50.
Lithium carbonate is rarely used for the treatment of anxiety Byrne, G.J. and Pachana, N.A. (2011). Development and validation of a Short
Form of the Geriatric Anxiety Inventory—the GAI-SF. International
disorders, apart from when it is used to augment other agents
Psychogeriatrics, 23(1), 125–31.
that are being used to treat comorbid disorders. Lithium is not Chemerinski, E., et al. (1998). Prevalence and correlates of anxiety in
recommended for the treatment of older people with anxiety dis- Alzheimer’s disease. Depression and Anxiety, 7, 166–70.
orders. Beta-adrenergic blockers, principally propranolol, have Cherbuin, N., et al. (2008). Hippocampal volume is positively associated
been used to reduce performance-related tremor in musicians. with behavioural inhibition (BIS) in a large community-based sample
Clonidine has been used to assist in the management of preop- of mid-life adults: the PATH through life study. Social Cognitive and
erative anxiety. Affective Neuroscience, 3, 262–9.
Chou, K.L., et al. (2011). Three-year incidence and predictors of first-onset
of DSM-IV mood, anxiety, and substance use disorders in older adults:
Conclusion results from Wave 2 of the National Epidemiologic Survey on Alcohol
Anxiety disorders occur commonly in older people and are and Related Conditions. Journal of Clinical Psychiatry, 72, 144–55.
associated with considerable emotional distress and disability. Coelho, C.M., et al. (2010). Specific phobias in older adults: characteristics
and differential diagnosis. International Psychogeriatrics, 22, 702–11.
Anxiety disorders in later life tend to be both underdiagnosed and
Cougle, J.R., et al. (2009). Examining the unique relationships between
undertreated. Anxiety disorders are often seen in comorbid rela- anxiety disorders and childhood physical and sexual abuse in the National
tion to other mental disorders and to general medical conditions. Comorbidity Survey-Replication. Psychiatry Research, 177, 150–5.
Effective treatments exist for anxiety disorders, although greater Darwin, C. (1872). The expression of the emotions in man and animals. John
effort is needed to make these treatments routinely accessible to Murray, London.
older people, including those with physical frailty or cognitive Davis, M. (1992). The role of the amygdala in fear and anxiety. Annual Review
impairment. of Neuroscience, 15, 356–75.
De Beurs, E., et al. (1999). Consequences of anxiety in older persons: its Kolassa, I.T., et al. (2010). The risk of posttraumatic stress disorder after
effect on disability, well-being and use of health services. Psychological trauma depends on traumatic load and the catechol-o-methyltransferase
Medicine, 29, 583–93. Val(158)Met polymorphism. Biology and Psychiatry, 67, 304–8.
Dennis, R.E., Boddington, S.J.A., and Funnell, N.J. (2007). Self-report Laidlaw, K., et al. (2003). Cognitive behaviour therapy with older people. Wiley,
measures of anxiety: are they suitable for older adults? Aging and Mental Chichester.
Health, 11, 668–77. Lam, C.L., et al. (2010) Brief problem-solving treatment in primary care
Diefenbach, G.J., et al. (2001). Anxiety, depression, and the content of worries. (PST-PC) was not more effective than placebo for elderly patients
Depression and Anxiety, 14, 247–50. screened positive of psychological problems. International Journal of
Dimidjian, S., et al. (2006). Randomized trial of behavioral activation, Geriatric Psychiatry, 25, 968–80.
cognitive therapy, and antidepressant medication in the acute treatment Laredo, L., et al. (2011). Risk of cerebrovascular accident associated with use
of adults with major depression. Journal of Consultant Clinical Psychology, of antipsychotics: population-based case-control study. Journal of the
74, 658–70. American Geriatrics Society, 59, 1182–7.
D’Zurilla, T.J. and Nezu, A.M. (2007). Problem-solving therapy. A positive Lavretsky, H., et al. (2009). Depression and anxiety symptoms are associated
approach to clinical intervention, 3rd edition. Springer, New York. with cerebral FDDNP-PET binding in middle-aged and older nondemented
Ekers, D., et al. (2011). Behavioural activation delivered by the non-specialist: adults. American Journal of Geriatric Psychiatry, 17, 493–502.
phase II randomised controlled trial. British Journal of Psychiatry, 198, Lindesay, J., et al. (2006). Worry content across the lifespan: an analysis of 16-
66–72. to 74-year-old participants in the British National Survey of Psychiatric
Foa, E.G., et al. (2005). Randomized, placebo-controlled trial of exposure and Morbidity 2000. Psychological Medicine, 36, 1625–33.
ritual prevention, clomipramine, and their combination in the treatment Liu, D., et al. (2000). Maternal care, hippocampal synaptogenesis and
of obsessive-compulsive disorder. American Journal of Psychiatry, 162, cognitive development in rats. Nature Neuroscience, 3, 799–806.
151–61. Lonsdorf, T.B., et al. (2010). The COMTval158met polymorphism is associated
Friedman, M.J., et al. (2011). Considering PTSD for DSM-5. Depression and with symptom relief during exposure-based cognitive-behavioral
Anxiety, 28, 750–69. treatment in panic disorder. BMC Psychiatry, 10, 99.
Gallagher, D., et al. (2011). Anxiety and behavioural disturbance as markers Loprinzi, P.D. and Cardinal, B.J. (2011). Association between
of prodromal Alzheimer’s disease in patients with mild cognitive objectively-measured physical activity and sleep, NHANES 2005–2006.
impairment. International Journal of Geriatric Psychiatry, 26, 166–172. Mental Health and Physical Activity, 4, 65–9.
Garcia-Soriano, G., et al. (2011). Symptom dimensions in obsessive-compulsive Lyketsos, C.G., et al. (2001). Neuropsychiatric disturbance in Alzheimer’s
disorder: from normal cognitive intrusions to clinical obsessions. Journal disease clusters into three groups: the Cache County study. International
of Anxiety Disorders, 25, 474–82. Journal of Geriatric Psychiatry, 16, 1043–53.
Gilbertson, M.W., et al. (2002). Smaller hippocampal volume predicts Mackenzie, C.S., et al. (2011). Prevalence and correlates of generalized
pathologic vulnerability to psychological trauma. Nature Neuroscience anxiety disorder in a national sample of older adults. American Journal
5, 1242–7. of Geriatric Psychiatry, 19, 305–15.
Gillespie, N.A., et al. (2004). Do the genetic or environmental determinants Mackintosh, M.A., et al. (2006). A twin study of lifetime generalized anxiety
of anxiety and depression change with age? A longitudinal study of disorder (GAD) in older adults: genetic and environmental influences
Australian twins. Twin Research, 7, 39–53. shared by neuroticism and GAD. Twin Research in Human Genetics, 9,
Gonçalves, D.C. and Byrne, G.J. (2012a). Sooner or later: age at onset of 30–7.
generalized anxiety disorder in older adults. Depression and Anxiety. 29, Maddock, R.J. (2001). The lactic acid response to alkalosis in panic
39–46. disorder: an integrative review. Journal of Neuropsychiatry and Clinical
Gonçalves, D.C. and Byrne, G.J. (2012b). Interventions for generalized Neuroscience, 13, 22–34.
anxiety disorder in older adults: systematic review and meta-analysis. Martin, E.I., et al. (2009). The neurobiology of anxiety disorders: brain
Journal of Anxiety Disorders, 26, 1–11. imaging, genetics, and psychoneuroendocrinology. Psychiatric Clinics of
Gonçalves, D.C. and Byrne, G.J. (2013). Who worries most? Worry prevalence North America, 32, 549–75.
and patterns across the lifespan. International Journal of Geriatric Mataix-Cols, D., Pertusa, A., and Snowdon, J. (2011). Neuropsychological
Psychiatry, 28(1), 41–9. and neural correlates of hoarding: a practice-friendly review. Journal of
Gonçalves, D.C., Pachana, N.A., and Byrne, G.J. (2011). Prevalence and Clinical Psychology, 67, 467–76.
correlates of generalized anxiety disorder among older adults in the McClure, R.J., et al. (2005). Population-based interventions for the prevention
Australian National Survey of Mental Health and Well-Being. Journal of of fall-related injuries in older people. Cochrane Database of Systematic
Affective Disorders, 132, 223–30. Reviews, 1, CD004441. doi: 10.1002/14651858.CD004441.pub2.
Griez, E. and Schruers, K. (1998). Experimental pathophysiology of panic. Medicines and Healthcare Products Regulatory Agency (2011). Citalopram
Journal of Psychosomatic Research, 45, 493–503. and escitalopram: QT interval prolongation—new maximum daily
Harbison, J. (2002). Sleep disorders in older people. Age and Ageing, 31-S2, dose restrictions (including in elderly patients), contraindications,
6–9. and warnings. Drug Safety Update 5(5). <http://www.mhra.gov.
Hemmeter, U., et al. (2000). Effect of zopiclone and temazepam on sleep uk/Safetyinformation/DrugSafetyUpdate/CON137769> (accessed
EEG parameters, psychomotor and memory functions in health elderly 30/12/2011).
volunteers. Psychopharmacology, 147, 384–96. Monastero, R., et al. (2009). A systematic review of neuropsychiatric
Hill, K.D. and Wee, R. (2012). Psychotropic drug-induced falls in older symptoms in mild cognitive impairment. Journal of Alzheimer’s Disease,
people: a review of interventions aimed at reducing the problem. Drugs 18, 11–30.
and Aging, 29, 15–30. Mueller, A., et al. (2009). The prevalence of compulsive hoarding and its
Kalinichev, M., et al. (2002). Long-lasting changes in stress-induced association with compulsive buying in a German population-based
corticosterone response and anxiety-like behaviors as a consequence sample. Behaviour Research and Therapy, 47, 705–9.
of neonatal maternal separation in Long- Evans rats. Pharmacology Mula, M., Pini, S., and Cassano, G.B. (2007). The role of anticonvulsant drugs
Biochemistry and Behavior, 73, 131–40. in anxiety disorders: a critical review of the evidence. Journal of Clinical
Klein, D.F. (1993). False suffocation alarms, spontaneous panics, and related Psychopharmacology, 27, 263–72.
conditions. An integrative hypothesis. Archives of General Psychiatry, 50, Muris, P., Merckelbach, H., and Clavan, M. (1997). Abnormal and normal
306–17. compulsions. Behaviour Research and Therapy, 35, 249–52.
Nepon, J., et al. (2010). The relationship between anxiety disorders and Sinoff, G., et al. (1999). Short anxiety screening test: a brief instrument
suicide attempts: findings from the National Epidemiologic Survey on for detecting anxiety in the elderly. International Journal of Geriatric
Alcohol and Related Conditions. Depression and Anxiety, 27, 791–8. Psychiatry, 14, 1062–71.
Norrholm, S.D. and Ressler, K.J. (2009). Genetics of anxiety and trauma-related Slade, T. and Andrews, G. (2001). DSM-IV and ICD-10 generalized anxiety
disorders. Neuroscience, 164, 272–87. disorder: discrepant diagnoses and associated disability. Social Psychiatry
O’Connor, D.W. (2006). Do older Australians truly have low rates of anxiety and Psychiatric Epidemiology, 36, 45–51.
and depression? A critique of the 1997 National Survey of Mental Health Smalbrugge, M., et al. (2005). Prevalence and correlates of anxiety among
and Wellbeing. Australian and New Zealand Journal of Psychiatry, 40, nursing home patients. Journal of Affective Disorders, 88, 145–53.
623–31. Snarski, M., et al. (2011). The effects of behavioral activation therapy
O’Connor, D.W. and Parslow, R.A. (2010). Differences in older people’s with inpatient geriatric psychiatry patients. Behavioural Therapy, 42,
responses to CIDI’s depression screening and diagnostic questions may 100–8.
point to age-related bias. Journal of Affective Disorders, 125, 361–4. Snowdon, J. and Halliday, G. (2011). A study of severe domestic squalor:
Pachana, N.A., Woodward, R.M., and Byrne, G.J. (2007a). Treatment of 173 cases referred to an old age psychiatry service. International
specific phobia in older adults. Journal of Clinical Interventions in Aging, Psychogeriatrics, 23, 308–14.
2, 469–76. Spalletta, G., et al., (2010) Neuropsychiatric symptoms and syndromes in
Pachana, N.A., et al. (2007b). Development and validation of the Geriatric a large cohort of newly diagnosed, untreated patients with Alzheimer
Anxiety Inventory. International Psychogeriatrics, 19(1), 103–14. disease. American Journal of Geriatric Psychiatry, 18, 1026–35.
Pachana, N.A., et al. (2012). Anxiety and depression in adults in their Spielberger, C.D. (1983). Manual for the State-Trait Anxiety Inventory (STAI).
eighties: do gender differences remain? International Psychogeriatrics, Consulting Psychologists Press, Palo Alto, California.
24, 145–50. Stafford, M., Chandola, T., and Marmot, M. (2007). Association between fear
Porensky, E.K., et al. (2009). The burden of late-life generalized anxiety of crime and mental health and physical functioning. American Journal
disorder: effects on disability, health-related quality of life, and healthcare of Public Health, 97, 2076–81.
utilization. American Journal of Geriatric Psychiatry, 17, 473–82. Starcevic, V., et al. (2011). Functions of compulsions in obsessive-compulsive
Reid, J.M., et al. (2011). Hoarding behaviors among nonclinical elderly adults: disorder. Australian and New Zealand Journal of Psychiatry, 45,
correlations with hoarding cognitions, obsessive-compulsive symptoms, 449–57.
and measures of general psychopathology. Journal of Anxiety Disorders, Stein, D.J., Ahokas, A.A., and de Bodinat, C. (2008). Efficacy of agomelatine
25, 1116–22. in generalized anxiety disorder: a randomized, double-blind,
Reiner, R. (2007). Media-made criminality: the representation of crime placebo-controlled study. Journal of Clinical Psychopharmacology, 28,
in the mass media. In: Maguire, M., Morgan, R., and Reiner, R. (eds) 561–6.
The Oxford handbook of criminology, 4th edition, Chapter 11. Oxford Stel, V.S., et al. (2004). Consequences of falling in older men and women and
University Press, New York. risk factors for health service use functional decline. Age and Ageing, 33,
Reynolds, C.F., et al. (1999). Nortriptyline and interpersonal psychotherapy 58–65.
as maintenance therapies for recurrent major depression: a randomized Timpano, K.R., et al. (2011). The epidemiology of the proposed DSM-5
controlled trial in patients older than 59 years. Journal of the American hoarding disorder: exploration of the acquisition specifier, associated
Medical Association, 281, 39–45. features, and distress. Journal of Clinical Psychiatry, 72, 780–7.
Rosellini, A.J. and Brown, T.A. (2011). The NEO Five-Factor Inventory: latent Twelftree, H. and Oazi, A. (2006). Relationship between anxiety and agitation
structure and relationships with dimensions of anxiety and depressive in dementia. Aging and Mental Health, 10, 362–7.
disorders in a large clinical sample. Assessment, 18, 27–38. van’t Veer-Tazelaar, P.J., et al. (2009). Stepped-care prevention of anxiety and
Ruscio, A.M. (2002). Delimiting the boundaries of generalized anxiety depression in late life: a randomized controlled trial. Archives of General
disorder: differentiating high worriers with and without GAD. Journal of Psychiatry, 66, 297–304.
Anxiety Disorders, 16, 377–400. Wetherell, J.L., et al. (2004). Quality of life in geriatric generalized anxiety
Sacchetti, E., Turrina, C., and Valsecchi, P. (2010). Cerebrovascular accidents disorder: a preliminary investigation. Journal of Psychiatric Research, 38,
in elderly people treated with antipsychotic drugs: a systematic review. 305–12.
Drug Safety, 33, 273–88. World Health Organization (2010). International statistical classification of
Sah, P. and Westbrook, R.F. (2008). Behavioural neuroscience: the circuit of diseases and related health problems, 10th Revision. WHO, Geneva.
fear. Nature, 454, 589–90. Wu, L.T. and Blazer, D.G. (2011). Illicit and nonmedical drug use among
Schumacher, J., et al. (2011). The genetics of panic disorder. Journal of older adults: a review. Journal of Aging and Health, 23, 481–504.
Medicine and Genetics, 48, 361–8. Xie, P., et al. (2009) Interactive effect of stressful life events and the serotonin
Seekles, W., et al. (2010). Effectiveness of guided self-help for depression and transporter 5-HTTLPR genotype on posttraumatic stress disorders
anxiety disorders in primary care: a pragmatic randomized controlled diagnosis in 2 independent populations. Archives of General Psychiatry,
trial. Psychiatry Research, 187, 113–20. 66, 1201–9.
Sehlmeyer, C., et al. (2009). Human fear conditioning and extinction in Yerkes, R.M. and Dodson, J.D. (1908). The relation of strength of stimulus
neuroimaging: a systematic review. PLoS One, 4, e5865. to rapidity of habit-formation. Journal of Comparative Neurology and
Sikter, A., et al. (2007). The role of hyperventilation: hypocapnia in the Psychology, 18, 459–82.
pathomechanism of panic disorder. Revista Brasileira de Psiquiatria, 29, Zisook, S., Schneider, D., and Shuchter, S.R. (1990). Anxiety and bereavement.
375–9. Psychiatric Medicine, 8, 83–96.
CHAPTER 46
Late-onset schizophrenia
Sarah Brunelle, Ipsit V. Vahia, and Dilip V. Jeste
While there is a growing recognition of the impact of late-onset Bleuler, a decade later, de-emphasized age of onset as a defining
schizophrenia (LOS), there remains a dearth of data on this topic feature (Bleuler, 1950). Kraepelin later recognized that sometimes
and it remains poorly understood (Jeste and Nasrallah, 2003). As of the illness presented later in life and he described cases with onset
2000, it was estimated that only 1% of the literature on schizophrenia as late as the seventh decade, although these accounted for only
focused on the older population, with an even smaller proportion 0.2% of the 1054 patients he studied.
specific to the late-onset subgroup (Cohen et al., 2000). Therefore, Manfred Bleuler, the son of Eugen, may be considered the first to
it is perhaps not surprising that schizophrenia with onset in late explicitly introduce the concept of LOS, in 1943. Setting the ground
life still remains a challenge to both clinicians and researchers, with criteria still applied today, he stated that these cases should
despite the 1-year prevalence rates of 0.6% and 0.2% among indi- begin after age 40, with symptoms not fundamentally different to
viduals aged 45–64 years and 65 years and over, respectively, and those seen in schizophrenia with an earlier onset, and that there
lifetime prevalence of 1% and 0.3% reported by the Epidemiologic should be no amnestic syndrome or physical sign that indicated
Catchment Area (ECA) study (Robins and Regier, 1991). a degenerative brain disease as the most likely aetiology (Bleuler,
Research on this topic has been hampered by lack of consen- 1943). Using this definition, he found that 15% of schizophrenic
sus regarding nomenclature and diagnostic criteria, and hetero- disorders began between 40 and 60 years of age, with only a small
geneity in clinical samples, often resulting in a variety of clinical number of cases presenting later. Prior to Manfred Bleuler’s work,
syndromes included under the same broad diagnostic category others in Germany had attempted to distinguish cases of later
(Castle and Morgan, 2008). For example, ‘late-life schizophrenia’ onset, including Gaupp in 1905, followed by Stransky (‘dementia
is an umbrella term that can be applied to two different groups of tardiva’), Berger (‘paranoia chronica’), Kleist (‘involutional para-
individuals: those diagnosed with schizophrenia in early adulthood noia), Albrecht (‘presenile paraphrenia’), Serko (‘involutional para-
and those who developed the illness in middle or late life (Palmer phrenia’), and Medow (‘stiffening involutional psychosis’). British
et al., 1999). psychiatrists employed the term ‘late paraphrenia’ to describe
The first U.S. baby-boomer turned 65 in 2011. A significant ‘a well-organized system of paranoid delusions with onset after
increase in the older population is expected, resulting in a doubling age 45, with or without hallucinations existing in the setting of a
of the number of seniors with severe mental illnesses (Cohen et al., well-preserved personality and affective response’ (Roth, 1955).
2000). More frequent cases of LOS are to be expected, but it needs Post doubted the frequency of schizophrenia with onset in late life
to be emphasized that individuals with onset of symptoms prior and felt that a heterogeneous group of disorders were more likely,
to age 45 will still account for approximately 85% of patients with hence the expression ‘persistent persecutory states of late-life’ (Post,
schizophrenia encountered in clinical settings (Cohen et al., 2000). 1966).
In this chapter, we will focus primarily on schizophrenia first mani- The Diagnostic and Statistical Manual (DSM) has changed its
festing in late life, and subsequently on chronic schizophrenia in stance on distinguishing late-onset from earlier-onset schizophre-
older people and its distinction from LOS. nia over the past three editions. Published in 1980, the DSM-III had
eliminated the category ‘involutional paranoid states/paraphrenia’
Nomenclature and did not allow for the diagnosis of schizophrenia if symptoms
emerged after the age of 45 (American Psychiatric Association,
Historical background 1980). This criterion was modified in the revised edition with the
While the current terms late-onset schizophrenia (LOS) and introduction of the ‘late-onset’ specifier for onset after 44 years
very-late-onset schizophrenia-like psychosis (VLOSLP) have gained old (American Psychiatric Association, 1987). However, the term
acceptance in recent years, previous terms such as late paraphrenia was eliminated from the DSM-IV (APA, 1994). The World Health
and the French term psychose hallucinatoire chronique remain in use Organization included no qualifier for late-onset in its International
occasionally in the world literature and are commonly used by cli- Classification of Diseases, 8th edition (ICD-8). In the ICD-9, how-
nicians, especially outside the US (Harris and Jeste, 1988; Dubertret ever, the category of ‘paranoid schizophrenia’ included the sub-
et al., 2004; Howard et al., 1994; Riecher-Rossler et al., 1995). type of ‘paraphrenic schizophrenia’. In the ICD-10 the entity was
In his use of the term ‘dementia praecox’, Kraepelin (1899) included as ‘paraphrenia (late)’ in the delusional disorders group
implies an early onset and rapidly deteriorating course. Eugen (World Health Organization, 1992).
Current terminology: international consensus 1987; Holden, 1987; Copeland et al., 1998; ). New onset of psycho-
In 2000, the International Late-Onset Schizophrenia Group pro- sis in late-adulthood therefore frequently occurs outside the con-
posed the terms ‘late-onset schizophrenia (LOS)’ for cases with onset text of a primary psychotic disorder.
between 40 and 60 years and ‘very-late-onset schizophrenia-like There is evidence that incidence rates are higher for women, and
psychosis (VLOSLP)’ for those presenting first after age 60. The women:men ratios vary from 1.9:1 to as high as 45:2 (Bleuler, 1943;
panel concluded that evidence supported these age cut-offs and that Herbert and Jacobson, 1967; Howard et al., 2000). Whether gender
these diagnoses had face validity (Howard et al., 2000). The distinc- ratios represent real differences in incidence rates based on true
tion of the VLSOLP was supported by strong empirical evidence, risk factors or whether they are impacted by research bias or cohort
with the LOS age criteria of 40 years more arbitrary; however, the effects remains subject to some debate.
experts felt that both had clinical usefulness and were intended to In a 1997 study conducted in the UK, African-Caribbean people
promote research in the field (Howard et al., 2000). Despite ini- were underrepresented among late-onset compared to younger-on-
tial concerns, subsequent research has not shown a significant set cases (4% and 40%, respectively), but older adults from this eth-
likelihood of affective disorders misclassified as LOS or VLOSLP. nic background appear more likely to be diagnosed with late-onset
Moreover, early-onset schizophrenia (EOS) and LOS appear to be psychosis (Castle et al., 1997; ; Reeves et al., 2001, 2002, 2003; Mitter
very stable diagnoses; they remained unchanged in as many as 93% et al., 2004, 2005).
of cases in a recent follow-up and only rarely were they reclassi-
fied as mood disorders (Taylor, 2001; Vahia et al., 2010). Although Risk Factors and Correlates
stronger evidence supports the distinction of VLOSLP as opposed
As with epidemiology, the literature on LOS is limited by lack of
to LOS, few studies have focused on this diagnosis specifically.
standardization in outcomes criteria, populations, and evaluation
In this chapter, we use the term LOS to indicate both LOS and
tools, as underscored in a recent systematic review by Brunelle
VLOSLP, except where specified. We also omit use of the term ‘late
et al. (2011). Longitudinal studies assessing predictors of LOS are
paraphrenia’, unless specified.
few and current knowledge relies primarily on cross-sectional data
(Köhler et al., 2007).
Epidemiology
Poorly defined diagnostic criteria, diverse terminology, and hetero- Genetic risk
geneity of samples have limited epidemiological data on psychotic A significant family history of psychotic disorders has not been
symptoms in older people. reported with LOS, although this probably reflects a lower familial
The prevalence of paranoid ideation in individuals over age 65 burden in LOS compared to EOS rather than a complete lack of
has been reported to be as high as 6%, although symptoms arising association. A wide variation can be noted in the reported prev-
in the context of cognitive impairment account for most of these alence rates of schizophrenia in relatives of both EOS and LOS
cases. Forsell and Henderson(1998) found an overall prevalence of subjects, which can be attributed to the different age cutoffs and/
12.1% in those with cognitive dysfunction, compared to 2.6% in or diagnostic criteria used. To our knowledge, only Köhler et al.
those without (Christenson and Blazer, 1984). Older individuals (2007) have prospectively studied the risk imparted by a positive
also seem to be more susceptible to both visual and auditory hal- family psychiatric history. Neither depressive nor psychotic disor-
lucinations, although women appear to be the most at risk in later ders appeared to increase risk, consistent with cross-sectional evi-
age than men and they experience a later peak prevalence (Tien, dence. Jeste et al. (1995) found no difference in family prevalence of
1991). Rates of 1-year psychotic symptom prevalence in the oldest depression in persons with LOS or EOS, or unaffected controls.
old (over 95 years) have been estimated to be as high as 7.4%. The More recently, data from linkage-analyses seem to suggest that
risk of psychosis appears directly proportional to age in the geriat- the age of onset is, at least in part, dependent on genetic predis-
ric population. Van Os et al. (1995) suggested that incidence rates position, and that genes also have an effect on the phenotype
increase by 11% every 5 years after age 60. From 10 per 100,000 (Hamshere et al., 2011). This may partly explain differences in clini-
person-years between 60 and 65 years, the incidence rate reached cal presentation of LOS and EOS. Several genetic risk factors have
25 per 100,000 person-years in people 90 years and older. These been proposed for LOS. The DRD2 gene and specifically one of its
rates reflect incidence for all nonaffective, nonorganic psychoses, polymorphisms, rs2734839, was strongly associated with older age
including schizophrenia. of onset, although the number of subjects above age 40 was insuf-
The incidence rates of schizophrenia peak between ages 16 and ficient to draw any firm conclusions (Voisey et al., 2011). A link
25 years and then again in the 46–55 years age group, and there between the CCR5 32-bp deletion allele and LOS has also been
may in fact be a third peak in incidence rates in individuals aged suggested (Rasmussen et al., 2006). Dopa-decarboxylase seems to
65 and older (Castle and Murray, 1993). A considerable overlap affect the age at first presentation, with certain genotypes contain-
between late-onset psychosis and schizophrenia exists in epide- ing the 1-bp deletion potentially increasing the likelihood of LOS in
miologic studies and incidence rates have been found to be lower, men (Borglum et al., 2001). These findings point toward a genetic
with an outcome restricted to schizophrenia only. In a prospective susceptibility to psychosis even in late-onset cases, which could be
study assessing new cases of schizophrenia fulfilling DSM-IV crite- both specific to LOS and shared with EOS, though the genetic sus-
ria, Bogren et al. (2010a) found an incidence rate of 8 per 100,000 ceptibility seems lower for LOS than for EOS.
person-years in the age group of 65 + years (APA, 1994). Rates of
15.6 and 17 per 100,000 person-years have been suggested for delu- Gender
sional disorder (based on DSM-III-R) and late-paraphrenia (as per The higher incidence rate of LOS in women does not necessarily
Kay and Roth’s criteria), respectively (Kay and Roth, 1961; APA, imply causality. While it may indicate higher risk, other factors like
CHAPTER 46 late-onset schizophrenia 605
access to care, proportion of individuals at risk, or even diagnostic 1976; Prager and Jeste, 1993). The association between schizophre-
bias may underlie this higher rate. It has been hypothesized that dif- nia and hearing impairment appears stronger for VLOSLP than
ferences in life-expectancy between genders could also contribute for LOS, but these associations have not been confirmed in lon-
to the increased incidence found in older women compared to men gitudinal studies. Less convincing evidence exists for an associa-
(Howard et al., 1994). As the gap in longevity between men and tion with visual deficits (Howard et al., 2000), though prospective
women has narrowed between 1970 and the twenty-first century, studies have demonstrated that people with visual deficits are at
and might continue to do so over the next decades, it would be significantly increased risk of psychosis, although it has not been
interesting to see if this is accompanied by a corresponding trend demonstrated specifically for LOS (Blazer et al., 1996; Forsell, 2000;
of the incidence rates for LOS to even out between both genders Kohler et al., 2007).
(Pinkhasov et al., 2010). Surprisingly, no difference in incidence Although the existing evidence suggests a correlation, specific
rates of late-onset psychosis was found in a systematic review of aetiological pathways linking sensory deficit and psychosis have
longitudinal risk factors, possibly because higher incidence in been disputed. Proposed theories include sensory loss, reinforc-
women does not translate into a higher risk ratio in prospective ing premorbid tendencies toward isolation (Corbin and Eastwood,
studies (Brunelle et al., 2011). Huang and Zhang (2009) studied 1986), or that persons with sensory loss would simply be less likely
Taiwanese individuals older than age 60 admitted with a diagnosis to seek treatment. Comparing patients with early- vs very-late-on-
of schizophrenia and reported that the difference in gender propor- set, Rodriguez-Ferrera et al. (2004) obtained a higher than expected
tions was not significant between the early- and late-onset groups, prevalence of hearing loss in their very-late-onset group (54%),
women representing 44.8% and 34.7% of all cases, respectively. with only 15% in the younger onset group including cases of EOS
Interestingly, as the familial burden of schizophrenia increases, and LOS, with a mean age of onset of 36 years (SD = 11.3). Almeida
the effect of gender on age of onset appears to lessen. Several authors et al. (1992) reported very similar rates of 41% and 6.6% for LOS
have reported comparable men-to-women ratios in patients with and EOS, respectively. Reported prevalence rates of hearing defi-
a positive family history, even in late-onset cases (Leboyer et al., cits in the literature in the general older population range between
1992; Rasanen et al., 2000; Abel et al., 2010). 25% and 80% (Newman and Sandridge, 2004). A recent review sug-
The observed gender differences in prevalence rates for LOS have gested that hearing deficits severe enough to impair conversation
led to the so-called oestrogen hypothesis of schizophrenia—the affect approximately 10% of the general population and as many
onset of symptoms in older women due to the loss of the previ- as 40% of adults over age 65, with 80% of all cases of hearing loss
ous protection conferred by the oestrogenic influence (Seeman and occurring in older people (Huang and Tang, 2010). In light of these
Lang, 1990; Hafner et al., 1998). Oestrogen modulates several neu- numbers, the high frequency of hearing impairments in patients
rotransmitter systems, including the dopaminergic, serotonergic, with VLOSLP and to a lesser extent LOS is hardly surprising.
GABA-ergic, noradrenergic, and cholinergic pathways, all of which There is very limited literature on medical comorbidity in people
have been implicated in psychotic disorders (Behl et al., 1995; Lee with LOS, especially considering the significant burden of medical
and McEwen, 2001; Kolsch and Rao, 2002; Brann et al., 2007). issues in the older population in general and in the psychiatric pop-
However, attempts to identify specific polymorphisms of the oes- ulation in particular. Indeed, some ‘geriatric syndromes’ are specific
trogen receptors responsible for schizophrenia or LOS have been to this group, which is also at higher risk of accumulating concur-
mostly inconclusive, and there are currently no data supporting the rent medical illnesses, with resulting adverse effect on functional
hypothesis that a genetic variant would have a specific influence on and cognitive status. At least 20–30% of people over 65 years of age
age of onset other than through the general mechanisms (Ouyang are believed to suffer from chronic diseases (De Luca d’Alessandro
et al., 2001). Using first admissions data for VLOSLP, Reeves et al. et al., 2011). Moreover, medical issues are often overlooked in this
(2002) found that male patients were significantly more likely to be population. Data from a chart review of 79 consecutive geriatric
lost at follow-up, which points to differences in help-seeking behav- psychiatry hospital admissions (University of California, San Diego,
iours between genders. Senior Behavioral Health Unit) indicate that 34% of the patients
In summary, the question of whether female gender is itself a sig- had unrecognized medical conditions (Woo et al., 2003).
nificant risk factor for the development of LOS is not fully under- An association between occlusive carotid and vertebral arteries
stood yet, but women constitute a higher proportion of the clinical disease and late-onset of psychiatric disorders has been reported
population. However, gender does not seem to exert any direct and cardiovascular risk factors seem to be increased in subjects
effect outside its association with age at onset (Leboyer et al., 1992; with LOS, but also late-onset affective disorders (van der Heijden
Hafner et al., 1998; Häfner, 2003). et al., 2010). Barak et al. (2002) indicated that medical comor-
bidities were frequent in both groups of older people with either
Sensory loss/medical comorbidity VLOSLP or chronic schizophrenia (71.4% vs 57.1%, respectively,
Sensory loss has been consistently implicated in the develop- although it was not significant), with hypertension being the most
ment of late-life psychosis since an association with hearing loss frequent medical condition, followed by ischaemic heart disease
was reported for the first time by Kay and Roth (Roth, 1955; Kay and diabetes. The rates of medical comorbidities seem to be higher
and Roth, 1961). Symptoms of paranoia following experimentally in late-onset delusional disorder than schizophrenia, with also more
induced deafness have been described and the literature suggests neurological conditions in the first group (Anita Riecher-Rossler
that adjustments to hearing aids can lead to the resolution of symp- et al., 2003). Evidence for potential medical risk factors from lon-
toms (Eastwood et al., 1981; Zimbardo et al., 1981; Khan et al., gitudinal studies is limited. Henderson et al. (1998) found that
1988). Authors have reported that hearing loss in those cases is current physical symptoms increased the risk of delusions and hal-
more often conductive and usually severe, bilateral, and appearing lucinations, although causality was not established for past physical
early in life (Cooper et al., 1974; Cooper, 1976; Cooper and Curry, illnesses. There is a dearth of data about the topic of a history of
obstetric or perinatal complications in LOS patients. To our knowl- highest severity of current depressive symptoms, which could have
edge, only one longitudinal study has been published discussing the introduced a significant recall bias (Schmidt et al., 1995). Contrary
rates of perinatal complications in LOS, retrospectively looking at a to these findings, there is evidence pointing toward a lack of aetio-
small sample of 12 LOS patients and concluding against any predis- logical association between depressive schemas and LOS. Indeed,
posing effect (Castle et al., 1997; Reulbach et al., 2007). McCulloch et al. (2006) found that not only was there no indica-
tion of depressive cognition or self-conception in individuals with
Cognitive and psychiatric symptomatology late-onset psychosis, but also there was nothing to suggest changes
Cognitive patterns and psychological mechanisms have been dis- over time; hence the possible contribution of depression to the
cussed extensively in the general literature on LOS. ‘Theory of onset of symptoms was limited.
mind’, ‘mentalization’, and ‘social cognition’ are related concepts Unfortunately, to our knowledge only two groups have assessed
that describe the processes used by human beings to make sense of this topic longitudinally. Köhler et al. (2007) found that lifetime
their environment and interact with others. It has been proposed depressive symptoms at baseline were not associated with the onset
that decreased social exploration is an expected phenomenon with of psychosis (not restricted to schizophrenia) in individuals over
normal ageing and that its interaction with cognitive impairment age 50, despite a significant effect in younger cohorts. Interestingly,
could increase the susceptibility to psychotic symptoms in people although there was a trend toward lower influence with older
with otherwise healthy development and mentalization patterns. age of onset, ‘neuroticism’ retained its significance as a contribu-
Similar to what has been reported for EOS patients, those with LOS tive factor in late-onset psychosis. This is congruent with the
have been found to make significantly more mentalizing errors analysis from Bogren et al. (2010b) of premorbid behavioural and
on testing than older controls, but they did not demonstrate the personality-related signs and symptoms predictive of psychosis,
impaired performance in probabilistic reasoning and exaggeration with only the two clusters ‘nervous-tense’ and ‘abnormal-antisocial’
in self-attribution bias typical of the early-onset individuals (Moore reaching significance for schizophrenia. The description of the lat-
et al., 2006). The literature on social cognition has been burgeoning ter cluster indicates what would now be more suggestive of cluster
over the last years in the field of schizophrenia in general, and the A personality disorders.
extent of the deficits appears to mirror those in neurocognition; Abnormal premorbid personality traits have often been impli-
whether this will prove relevant in the LOS population remains to cated. Fuchs (1999a) found high rates of personality pathology in
be determined (Hofer et al., 2010). Fromholt et al. (1999) described a sample of 38 late paraphrenia patients and indicated that 39%
the psychological characteristics of 20 individuals with late para- had met the criteria for paranoid or schizoid personality disorders
phrenia and found that the subjects’ emotional reactions and cop- prior to the psychosis, and the frequency of characteristic traits
ing mechanisms appeared congruent to their subjective views of has been reported to be as high as 70% in older literature. Subjects
the problem. That is, their reactions would have been deemed affected by late-onset psychotic disorders have been described by
rational or logical if the patients’ perceptions corresponded to the early British authors as ‘suspicious’, ‘quarrelsome’, ‘hostile’, ‘sensi-
reality. This was replicated in a later study by Quin et al. (2009). tive’, ‘unsociable’, ‘reticent’, ‘odd’, ‘eccentric’, ‘histrionic’, ‘pretentious’,
Fromholt et al. (1999) also noted that in all but one patient the delu- and exhibiting a long-standing difficulty to establish or maintain
sional ideas were plausible, i.e. they ‘did not violate physical laws or intimate relationships (Kay and Roth, 1961; Post, 1966; Herbert
include references to any supernatural phenomena’. and Jacobson, 1967; Howard et al., 1994; Giblin et al., 2004). In
In the historical literature on social tendencies and LOS, there LOS specifically, Pearlson et al. (1989) noted patterns of having
are contradictions. While the late paraphrenia described by Roth been ‘reclusive and introverted’, with ‘few friends, poor interper-
has been thought to occur mostly in older women with a life- sonal relationships and peculiar religious beliefs’. Brodaty et al.
long tendency toward solitariness and suspiciousness, Janzarik’s (1999) described persons with LOS as ‘odd and eccentric’, ‘suspi-
‘Kontakmangelparanoid’ was described in women with a previ- cious and detached’ compared to normal controls, and similar to
ously high level of vitality and poor tolerance to solitude who previously described traits in late paraphrenia. Longitudinally, no
found themselves newly isolated by factors inherent to the ageing premorbid condition has been clearly associated with the onset
process (Roth, 1955; Janzarik, 1957, 1973; Kojo, 2010). In most of psychotic symptoms; anxious manifestations may potentially
cases, a progressive deterioration in social networks often prefig- be predisposing and a prior diagnosis of post-traumatic stress
ured the onset of psychosis and initiated a process of ‘life-review’ disorder had been given to several LOS patients with significant
in which subjects reminisced over hostile relationships with some childhood trauma (Tien and Eaton, 1992; Sachdev et al., 2000;
family members, expressed regrets for past actions, and described Reulbach et al., 2007). Although substance-related disorders are
well-established patterns of solitary coping styles that could be more common in older people with schizophrenia than without,
related to the perception of having always been different or ‘outsid- LOS specifically has not been well studied (Mulsant et al., 1993;
ers’ as described by most patients (Quin et al., 2009). Whether they Jeste et al., 1996a). Reviews of the longitudinal studies were incon-
might be at the origin or a consequence of LOS remains unclear. clusive (Tien and Eaton, 1992; Wiles et al., 2006; Köhler et al.,
Comparing older people with late-onset of depression or schizo- 2007; Brunelle et al., 2011).
phrenia and unaffected age-matched controls, Giblin et al. (2004)
demonstrated that LOS patients had significantly higher scores in Other sociodemographic factors
the following domains on the Schema Questionnaire: ‘rejection Some sociodemographic characteristics of the subjects affected by
and disconnection’, ‘impaired autonomy and performance’, ‘oth- LOS are likely not predisposing factors and reflect the direct con-
er-directedness’, and ‘over vigilance and inhibition’. They were also sequences of later age of onset. However, they are important to the
more likely to have a low morale about the ageing process, even understanding of the pathogenesis of schizophrenia in late life, and
compared to those with major depression. However, they had the are mentioned here.
Studies consistently show that subjects with LOS may have similar 36% of those the subjects reported a sense of loss. Similar associa-
marriage rates as normal controls (especially for VLOSLP) and are tions with acute stresses or losses have been reported in other stud-
much more likely to be, currently or formerly, married than their ies (Kay and Roth, 1961; Dewi Rees, 1971; Janzarik, 1973; Fromholt
earlier-onset counterparts (Barak et al., 2002; Hassett, 2002; Girard et al., 1999; Yasuda and Kato, 2009). This is compatible with the
and Simard, 2008). It has also been suggested that they might have view of LOS and EOS as unique single disorders, with LOS mani-
higher rates of divorce (or initiating the divorce) and widowhood festing later in more resilient individuals (Schmid et al., 2011).
(Kay and Roth, 1961; Herbert and Jacobson, 1967; Dewi Rees, 1971;
Almeida et al., 1992; Howard et al., 1994; Jeste et al., 1995; Fuchs,
1999a; Girard and Simard, 2008). One exception emerged from a Neuropathology
comparison of EOS and LOS in chronically hospitalized patients in Neuropathological mechanisms remain central to contemporary
Taiwan, with high rates of marriage in both groups, possibly cor- conceptualizations of schizophrenia, including: (1) dysfunction
responding to a cultural phenomenon (Huang and Zhang, 2009). of the neurotransmitter pathways in the central nervous sys-
Compared to those with early-onset, patients with LOS might have tem, primarily implicating dopamine, glutamate, and serotonin;
fewer friends and fewer children than normal controls (Gurian (2) neurodevelopmental alterations, suggested by regional brain
et al., 1992; Semple et al., 1997; Brodaty et al., 1999; Barak et al., abnormalities on both structural and functional imaging; and (3)
2002; Rodriguez-Ferrera et al., 2004; Romero-Rubiales et al., 2004; modifications at the cellular level, with a postulated reduction
McCulloch et al., 2006). in oligodendrocytes and/or aberrant neuronal cytoarchitecture
While an association with academic achievement remains unclear, (Pickard, 2011). However, the extent to which the above processes
better work performance and employment history reflected a much contribute specifically to illness manifestation in late life and the
better level of premorbid adjustment for schizophrenia patients role of alternative mechanisms is not well understood.
diagnosed later in life (Jeste et al., 1995; Castle et al., 1997; Fuchs, The first brain imaging study of the late-onset population was
1999a; Barak et al., 2002; Girard and Simard, 2008; Vahia et al., published by Haug in 1962, involving seven patients with a first epi-
2010; Girard et al., 2011). These findings, however, have not been sode of ‘psychosis’ after the age of 45 years. Using the technique of
replicated in longitudinal studies (Brunelle et al., 2011). Functional pneumoencephalography, he reported evidence of cerebral atrophy
limitations have not been studied in relation to schizophrenia and and dilated ventricles (Haug, 1962). It took almost a quarter of a
there is inconsistent evidence as to whether they increase the risk of century for the next study reporting on brain imaging in patients
psychosis in older people directly (Blazer et al., 1996; Forsell, 2000). with LOS. In 1986, Miller et al. (1986) reported computer tomog-
Although there are very limited data on this topic, it is possible that raphy (CT) results on a small sample of five women with late para-
an urban status at birth might predispose to LOS, as it has been sug- phrenia, and noted that three had cortical or subcortical changes
gested already for EOS (Marcelis et al., 1998). and another showed signs of normal pressure hydrocephalus.
Several subsequent studies all determined that ventricle-to-brain
Adverse life events ratios (VBRs) were larger in the patients than the normal con-
Negative experiences early in life can affect developmental tasks trols group (Naguib and Levy, 1987; Rabins et al., 1987; Pearlson
and thereby increase susceptibility to psychosis when they occur and Rabins, 1988). More recent investigations have mostly used
early in life, but psychosocial stressors may exert a larger impact magnetic resonance imaging (MRI) techniques, in an attempt to
on the psychotic episode’s development. Both have been described assess periventricular and deep white matter changes better. The
in relation to LOS, although the specificity and wide range of first two studies (Miller et al., 1989; Breitner et al., 1990) described
adverse events described in the literature likely impairs generaliza- the MRIs of subjects with a broader diagnosis of ‘late-life psychosis’
tion. Results from a systematic review of longitudinal studies on or ‘late-onset paranoid illness’. In the first study, 20% of them had
LOS indicated that both the number and intensity of negative life evidence of ‘silent vascular disease’ and tumours were discovered
events can contribute to the onset of psychotic symptoms later in in another 12%. The second group reported vascular lesions in the
life, and exposure to the Holocaust has been implicated for schizo- pons or medulla of seven out of eight subjects, but none in normal
phrenia specifically (Reulbach et al., 2007; Brunelle et al., 2011). controls. Miller et al. (1991) later wrote that white matter lesions
Childhood traumas and experiences perceived as discriminat- affected 42% of subjects, presenting with first onset of ‘psychosis’
ing, humiliating, or threatening have been frequently identified in after age 45, as opposed to only 8% of the age-matched healthy
subjects with LOS (Gurian et al., 1992; Fuchs, 1994, 1999b, 1999a; comparison subjects, predominantly in the temporal and frontal
Rockwell et al., 1994). For those with EOS, early life events have lobes. However, no definite pattern of white matter lesions has been
not been associated with the persistence of psychotic symptoms in established.
older age, but the accumulation of stressors throughout the lifetime A literature review published in 2010 suggested that extracranial
has been implicated (Cohen et al., 2011). Psychosocial adversity arterial pathology also contributed to the development of LOS. A
can play a role during adulthood as well, and certain ethnic groups significantly higher burden of premorbid cardiovascular risk fac-
such as African-Americans have been diagnosed more frequently. tors was described in those with late-onset psychiatric disorders,
While diagnostic bias, perceived discrimination, and stress related including schizophrenia (van der Heijden et al., 2010). It is possible
to the migration have been implicated for these associations, none that this finding reflects a distinction between LOS and VLOSLP.
has emerged as a significant predisposing factor in longitudinal Volumetric studies have failed to pinpoint any abnormality spe-
evidence (Blazer et al., 1996; Forsell and Henderson, 1998; Forsell, cific to late-onset patients and findings are mostly concordant to
2000; Reeves et al., 2001, 2003; Mitter et al., 2004, 2005; Brunelle those with EOS. The only notable difference was the larger thalamic
et al., 2011). In a recent article on LOS from Japan, psychosocial volumes found in patients with LOS compared to their early-onset
stresses were deemed causally related in 65.8% of the cases, and in counterparts (Corey-Bloom et al., 1995; Barta et al., 1997; Symonds
et al., 1997; Sachdev and Brodaty, 1999b, 1999a, 1999, 2000; Rabins LOS is more aetiologically similar to EOS, although this is not com-
et al., 2000). However, there have been no demonstrated alterations pletely clear.
in the size of the frontal lobes, hippocampus, parahippocampus,
thalamus ,or basal ganglia structures in patients with LOS com- Clinical Features
pared to older controls (Howard et al., 1995). Sachdev et al. (2000)
later confirmed that the hippocampus and amygdala volumes in Positive and negative symptoms
older people with LOS and EOS were comparable and were not sig- The conclusions of the International Consensus on Late-Onset
nificantly associated with their cognitive performances. Schizophrenia Group pointed toward greater similarities than dif-
Diffusion tensor imaging (DTI) results have confirmed the integ- ferences in the clinical presentation of schizophrenia arising in early
rity of frontal lobes and frontal cortical tracts, which had already and late life, especially regarding the positive symptomatology and
been suggested by volumetric studies (Jones et al., 2005). However, between ages 40 and 60 years. However, in clinical samples, cases
studies using single photon emission tomography (SPECT) have with very late onset were associated with a low prevalence of formal
suggested that up to 83% of LOS patients suffered from frontal or thought disorder and affective blunting and a higher prevalence of
temporal hypoperfusion. In one study, the regional cerebral blood visual hallucinations (Kay and Roth, 1961; Post, 1966; Grahame,
flow of patients with LOS showed a different pattern of alteration 1984; Pearlson et al., 1989; Howard et al., 1993, 2000; Jeste et al.,
than that of subjects with EOS. Reduced blood flow was observed 1995, 1997).
bilaterally in the postcentral gyrus for LOS patients, while the pre- Whether LOS patients experience more severe positive symp-
central and inferior frontal gyri were more affected in the EOS toms overall compared to those with early-onset is not fully clear.
group. It was postulated that these changes were more likely attrib- Yasuda and Kato (2009) found that the paranoid subtype applied
utable to the age of onset rather than the chronological age of the to over one-half of their LOS sample and was significantly higher
subjects (Wake et al., 2011). than in the older group with earlier onset. However, Vahia et al.
Lohr et al. (1997)reported a significant difference in the frequency (2010) suggested similar rates for the early- and late-onset groups,
of minor physical anomalies in those two sets of patients. While the although positive symptoms were less severe overall among those
subjects with Alzheimer’s dementia did not differ from normal con- with LOS. The absence of significant difference in positive symp-
trols, older people with LOS or EOS, as well as those with unipo- tomatology has been replicated in other studies (Pearlson et al.,
lar depression, all had more anomalies. Pathological examinations 1989; Jeste et al., 1995; Rodriguez-Ferrera et al., 2004; Huang and
have been rarely reported, but the absence of Alzheimer’s disease as Zhang, 2009). Some specific features do appear to be more com-
defined by extensive neurofibrillary tangles seems to be consistent mon in the LOS population, such as persecutory, elaborate, and
(Bozikas et al., 2002; Casanova et al., 2002). Although the propor- systematized delusions. Delusions and hallucinations may become
tion of tangles was not abnormal, other neuritic changes have been more prevalent with older age rather than with older age of onset,
observed. On the basis of their studies, Casanova and colleagues but delusions of partition or ‘the belief that people, objects or
have proposed the concept of a ‘restricted limbic tauopathy’ that radiation can pass through what would normally constitute a bar-
affects LOS and, to some extent, EOS patients (Casanova et al., rier to such passage’ are especially frequent in those with LOS and
2002; Casanova and Lindzen, 2003). In patients with schizophre- much rarer in older and young EOS patients (Howard et al., 1992;
nia, the familial loading for Alzheimer’s, vascular, or Lewy body Howard, 2006). Individuals with a later onset are also more sus-
dementias has not been found to be higher in those with a later ceptible to experience visual, olfactory, or tactile hallucinations
compared to an earlier onset of symptoms (Howard et al., 1997; and in several sensory modalities simultaneously (Pearlson et al.,
Brodaty et al., 1999). From a systematic review of the longitudi- 1989; Howard et al., 1993; Castle et al., 1997; Wynn Owen and
nal studies on the risk factors for late-onset psychosis, cognitive Castle, 1999; Alici-Evcimen et al., 2003; Sato et al., 2004; Girard
impairment not reaching the threshold for dementia emerged as a and Simard, 2008).
probable predictor for psychotic symptoms at follow-up, but not for It has been increasingly recognized that a later onset might not be
schizophrenia per se (Tien and Eaton, 1992; Henderson et al., 1998; as protective as originally assumed, with recent evidence suggesting
Forsell, 2000; Brunelle et al., 2011). The Weschler Adult Intelligence comparable severities of negative symptoms in EOS and LOS and
Scale-Revised (WAIS-R) similarities subtest and the California even in VLOSLP (Jeste et al., 1995; Girard and Simard, 2008; Vahia
Verbal Learning Test (short- and long-delay free recall) may be et al., 2010). Indeed, VLOSLP seems to be characterized by promi-
the two most sensitive neuropsychological measures discriminat- nent positive symptoms but not less negative symptomatology than
ing between LOS and Alzheimer’s disease in patients already diag- LOS, although it is rarely specifically assessed on its own. The level
nosed with LOS or VLOSLP (Zakzanis et al., 2003; Girard et al., of psychomotor activity appears to distinguish the two conditions,
2011). However, neuropsychological testing in individuals without with LOS patients presenting with more apathy and abnormal psy-
psychosis was not shown to have any predictive value regarding the chomotor activity (Barak et al., 2002; Girard and Simard, 2008).
likelihood of subsequently exhibiting schizophrenia manifestations
(Blazer et al., 1996; Henderson et al., 1998). Cognition
In summary, LOS does not appear to be related to Alzheimer’s Cognitive changes are among the cardinal features of schizophrenia,
disease or other dementia pathology. However, it is still unclear regardless of age of onset, and are under consideration for inclusion
whether a neurodegenerative process is at play in this condition, as a distinct dimension in the DSM-V section on schizophrenia
mainly because of the lack of distinction in the literature between (Keefe and Fenton, 2007; Bora et al., 2010; Laughren, 2011).
cases of LOS and VLOSLP and the pooling of various psychotic dis- Consistent with the overall literature on schizophrenia, late-on-
orders together in studies. Current evidence indicates that VLOSLP set patients were found to have lower cognitive performance than
might be more associated with a neurodegenerative disease, while normal controls. Ting et al. (2010) reported that LOS patients were
more impaired than age-matched controls on most cognitive tests, patients to depression patients and healthy controls; both LOS and
with the exception of those for the recall of newly learned verbal normal subjects exhibited a better self-esteem than the depressed
information (using the California Verbal Learning Test: Woods et individuals. When comparing with older EOS patients, Jeste et al.
al., 2006). However, those with LOS outperformed patients with (1995) found no significant between-group differences, and a study
Alzheimer’s disease on every measure, including retrieval, the only by Rodriguez-Ferrera et al. (2004) showed a trend toward higher
notable exception being a nonsignificant difference in learning new depression levels in EOS, though significance was not reported.
verbal material. These more recent findings are very similar to the
first neuropsychological descriptions of LOS patients, attributed to Differential Diagnoses of Psychotic
Hopkins and Roth in 1953. Comparing older subjects belonging to
several different diagnostic categories, they reported that those with
Symptoms in Older People
late paraphrenia had a much better performance on all tests than Several conditions can present with psychotic symptoms in older
those with dementia. The severity of impairments was similar in patients. A detailed discussion of all possible causes is beyond
late paraphrenia and affective patients, but the two groups showed the scope of this section, but we summarize the major differential
distinct patterns on neuropsychological assessment (Hopkins and diagnoses.
Roth, 1953).
Rajji et al. (2009) recently conducted a meta-analysis on cog- Early-onset schizophrenia
nitive deficits found in subjects with a first-episode in adult life, Table 46.1 provides a summary of the differences between EOS and
youth-onset, or LOS. From the nine studies selected, the authors LOS.
concluded that LOS patients have an overall better cognitive per-
formance than those with youth-onset or first-episode schizophre- Delusional disorder and other primary psychoses
nia, though the patterns of cognitive deficits differ in each group. Significant overlap exists between LOS, delusional disorder (DD),
The LOS patients had more impairment on measures of attention, and late paraphrenia, to the extent that authors have challenged the
fluency, global cognition, IQ, and visuospatial construction than the validity of any diagnostic division (Jorgensen and Munk-Jorgensen,
other two subgroups, while arithmetic, digit symbol coding, and 1985; Howard et al., 1994). Riecher-Rössler et al. (2003) proposed
vocabulary were mostly preserved. This challenges the assumption that ‘since clear and aetiologically meaningful differentiation
of minimal differences in the cognitive functioning of patients with between these diagnostic categories is not possible, they should not
EOS and LOS (Heaton et al., 1994; Jeste et al., 1995; Sachdev et al., be separated by artificial diagnostic criteria for research purposes’.
1999). It is possible that this design allowed for the detection of However, other authors have reported some evidence to support
results that did not reach significance in isolated studies because of the clinical distinction of DD from schizophrenia with onset after
small sample sizes; but it might also reflect the patient groups in the age 40 (Evans et al., 1996). A diagnosis of DD is more often seen in
analyzed studies. Unfortunately, in most studies, older adults with older patients and is not an uncommon cause of psychosis in this
LOS are compared to older adults with EOS. It has been suggested population. For example, Alici-Evcimen et al. (2003) described 27
that the younger the age of onset, the greater the severity of cogni- cases of late-onset psychosis in the 420 inpatients admitted to the
tive and functional impairments; however, while chronicity con- sole geriatric psychiatry unit of a Turkish hospital between 1993
tributes significantly to the cognition scores, it has not been shown and 2002. From those 27 patients, five had EOS. Eight were diag-
to alter the effect of age of onset on specific measures. LOS subjects nosed as having LOS, six with VLOS, and eight had DD. In a study
most consistently perform better on measures of abstraction/flex- of older people with schizophrenia, it was found that 27% of those
ibility, semantic/verbal memory, and learning. These findings have diagnosed after age 60 met the criteria for DD, as opposed to none
been replicated by other authors, although some also noted lesser in the early-onset group (Rodriguez-Ferrera et al., 2004). In fact,
impairment in processing speed for the late-onset group (Jeste et diagnostic classification seems to rely heavily on age. Results from
al., 1997; Tuulio-Henriksson et al., 2004; Vahia et al., 2010; Girard another group suggest that age accounted for 22% of the variation
et al., 2011). in the two diagnoses of schizophrenia and DD (Riecher-Rossler
Comparisons of executive functions in LOS and EOS are incon- et al., 2003).
sistent and the current consensus is that impairments are manifest Schizophreniform disorder resembles schizophrenia, but lasts
regardless of age of onset (Rajji and Mulsant, 2008). A positive cor- less than 6 months. Older patients with this disorder often demon-
relation exists between family history of schizophrenia, younger strated good premorbid adaptation, and the sudden onset of florid
age of onset, and poorer cognitive performance, although familial psychotic symptoms generally leaves them perplexed and confused.
loading might not have such a significant effect in late-onset cases. The episode might end as abruptly as it had begun, often with a
How this contributes to the neurocognitive features in LOS has, return to the premorbid level of functioning (Jørgensen et al.,
however, not been well studied (Tuulio-Henriksson et al., 2004; 1997).
Goldberg et al., 2011). To our knowledge, there has not been any Schizoaffective illness in older patients has not been well described,
study comparing the neuropsychological functioning of individu- probably because it is often included with cases of LOS. Among
als with VLOSLP and LOS. 27 very-late-onset psychosis cases described by Rodriguez-Ferrera
et al. (2004), two received a diagnosis of schizoaffective disorder.
Mood Holden (1987) did describe a schizoaffective subtype of late para-
Compared to normal controls, patients with LOS appear to endorse phrenia associated with a high prevalence of auditory hallucina-
a higher level of depressive symptoms (Jeste et al., 1995; Moore et tions but almost no visual manifestations. Affected patients had low
al., 2006). However, there was no evidence of depression in the rates of psychiatric antecedents and sensory deficits and were most
LOS group in a study by McCulloch et al. (2006) comparing LOS likely to be alive at 10 years compared to the other subgroups.
Table 46.1 Characteristics of various types of late-life psychosis

EOS LOS VLOSLP PoD
Family history of schizophrenia + + – –
Female preponderance – + ++ –
Minor physical anomalies + + – –
Specific brain abnormalities (MRI) – – + +/–
Dementia-like cognitive decline – – + ++
Magnitude of cognitive impairment + + ++ +++
Paranoid subtype + ++ ++? N/A
Visual vs auditory hallucinations +/– + +? ++
Complex vs simple delusions ++ + +? +/–
Thought disorder +/++ + – –
Negative symptoms ++ + – –
Required neuroleptic dose ++ + + +/–
(Adapted with permission from Iglewicz, A., et al. (2011).)
EOS, early-onset schizophrenia; LOS, late-onset schizophrenia; MRI, magnetic resonance imaging; N/A, not applicable; PoD, psychosis of dementia; VLOSLP, very-late-onset schizophrenia-
like psychosis; +, moderately present; ++, strongly present; +++ very strongly present; –, not likely to be present; ?, only partially supported by the literature.
Isolated hallucinations can be found in visually impaired indi- Association, 2000; Khouzam et al., 2005; Blazer and Wu, 2011).
viduals (Schadlu et al., 2009). They have been described as a com- Delirium from any cause should also be suspected in first onset
monly occurring phenomenon in persons who have experienced of psychosis and behavioural disturbances in late life (Reeves and
widowhood, described in approximately 4% of widowers (Dewi Brister, 2008).
Rees, 1971; Khouzam et al., 2005).
Delusions of misidentification, e.g. the Capgras syndrome, might Dementia
occur in people with primary psychotic disorders but should herald Psychotic manifestations, especially visual hallucinations, can
the possibility of an underlying organic condition. They are rela- present in virtually every type of dementia and occur in up to 50%
tively common in Alzheimer`s dementia and other neurological of patients at some point over the course of the illness. They are
or medical conditions (Khouzam et al., 2005). Paranoid personal- a characteristic feature of Lewy body dementia. Also, people with
ity can increase the risk of late-life psychosis, but premorbid sus- dementia are especially vulnerable to anticholinergic side effects of
piciousness can also worsen with age (Manford and Andermann, medications and the resulting delirium. Psychosis can be a poor
1998; Paulsen et al., 2000). prognostic marker in this population, where antipsychotic treat-
ment is associated with an increased mortality rate (American
Affective disorders Psychiatric Association, 2000; Paulsen et al., 2000; Hardy, 2003;
Psychosis is an especially frequent accompanying feature to Kales et al., 2007).
depressive episodes in later life. Delusions or hallucinations are
mood-congruent in the majority of the cases and somatic mani- Management
festations are especially common (Jørgensen et al., 1997; Khouzam
et al., 2005). Clinical evaluation
Grandiose delusions commonly appear during manic episodes. The evaluation of older people presenting with schizophrenia man-
Although mania tends to emerge as an early feature in the course ifestations requires assessment similar to that for any psychiatric
of bipolar disorder, it can occur for the first time in late life or can illness in late life. Extra focus should be given to collateral infor-
reappear as a consequence of medication changes. Mania in older mation, owing to the high risk of guardedness and poor insight
people should be differentiated from frontotemporal dementia due to psychosis and cognitive impairment. It is also important to
(Almeida and Fenner, 2002; Sajatovic and Chen, 2011). assess the impact of the disease on the social network. Assessment
of safety includes ruling out the risk of violence toward others, even
Delirium and substance-induced psychotic disorders though this is reportedly more common in younger rather than
While substance use disorders are comparatively less common in older adults (Martinez-Martin et al., 2011).
the older population, they are still observed with some frequency. The pharmacological history should focus on potentially contrib-
Also, older adults are at risk for withdrawal due to sudden lack of uting medications, including over-the-counter, herbal remedies,
access because of financial, mobility-related, or medical condi- interactions, recent changes, and medication adherence (e.g. anti-
tions. Alcohol and benzodiazepines are associated with especially cholinergic medications are often implicated as a cause of psychotic
high risk of psychotic phenomena during both intoxication and symptoms in older adults). Sexual and substance use histories
withdrawal and their use is not uncommon (American Psychiatric should be gathered considering the risk of intoxication, withdrawal,
neoplastic lesions, and sexually transmitted diseases. A complete older individuals are more prone. Benefits have been observed in
review of personal and familial antecedents, both medical and psy- older patients initially treated with risperidone or olanzapine after
chiatric, with a specific focus on psychotic and neurodegenerative the switch from a typical antipsychotic (Jeste et al., 1999a, 1999b;
disorders, is mandated and should include a discussion of the pre- Ritchie et al., 2003, 2006). In older people, as many as 2–30% of
morbid level of functioning. Information should be obtained on the patients per year may develop TD with typical antipsychotics
current pattern of impairment, course of symptoms, and accom- and they are also less likely to experience remission compared to
panying physical or psychological manifestations. During physi- younger individuals. Newer agents have a safer side-effect pro-
cal examination, the clinician should look for evidence of neglect, file but are not risk-free; TD still emerges in approximately 5% of
abuse, decreased hygiene, incontinence, or cachexia, as well as for treated older patients each year, in keeping with the young to old
any sign of delirium or any indication of its aetiology. Neurological risk ratio of 1:5 observed with first-generation medications (Correll
assessment should rule out increased intracranial pressure, local- et al., 2004). Older people are also more sensitive to EPS, even with
izing lesions, or neurodegenerative conditions like Huntington’s second-generation antipsychotics like risperidone, and this risk
and Parkinson’s disease or Lewy body dementia, lesions of the basal may be higher for those with EOS (Jeste et al., 1995; Lemmens et
ganglia or cerebellum, or seizures. A basic cognitive evaluation is al., 1999). Other related side effects include orthostatic hypoten-
mandated in all patients and should be repeated over time to rule sion, cardiac conduction abnormalities, agranulocytosis, and neu-
out fluctuations in the context of delirium. roleptic malignant syndrome (NMS). Hyperprolactinaemia from
A basic work-up should consist of a complete blood count, blood dopamine blockade is associated with osteoporosis, increasing the
chemistry profile (glucose, electrolytes with calcium and magne- risk of hip fractures (Boyce and Walker, 2008; Reeves and Brister,
sium, blood urea nitrogen, creatinine and liver function tests), uri- 2008; Sachdev and Brodaty, 1999b). Both classes of antipsychotics
nalysis (and urine culture, if indicated), thyroid function studies, have been implicated in increasing cardiovascular and all-cause
toxicology screening, vitamin B12 and folate levels, and serologi- mortality (Schneider et al., 2005; Kales et al., 2007; Isaac and Koch,
cal tests for syphilis. Brain imaging is recommended, especially for 2010; Kelly et al., 2010). Age-related changes in liver and kidney
cases with acute deterioration, altered consciousness, or abnor- function, decreased lean body mass, and polypharmacy may addi-
malities on neurological examination. Some authors also present tionally impact the pharmacokinetics and pharmacodynamics of
evidence for electrocardiography and routine chest radiography antipsychotics (Sable and Jeste, 2002; Tsuboi et al., 2011).
(Howard et al., 2000; Boyce and Walker, 2008; Reeves and Brister, The International Consensus suggested that one-half to
2008). It may be useful to obtain a baseline lipid panel, weight, one-quarter the doses of that used in younger patients could be
height, and waist circumference before an antipsychotic trial, in used in LOS patients and as low as one-tenth in VLOSLP. The initi-
order to monitor metabolic side effects. ation of therapy should be made at a very low dosage, with cautious
Ancillary investigations should focus on ruling out other sus- increments up to the lowest effective dose (Howard et al., 2000).
pected aetiology based on the history and physical examination. For example, risperidone should be introduced at 0.25–0.50 mg/day
The decision to perform comprehensive neuropsychological evalu- and titrated by no more than 0.50 mg/day to a target range of 1.25–
ation should be guided by clinical judgement, especially in settings 3.50 mg/day. Starting and target dose-ranges of olanzapine should
with limited resources. be 1–2.5 mg/day titrated up to 5–15 mg/day. With quetiapine these
should be 12.5–25 mg/day titrated to 100–300 mg/day (Jeste et al.,
1996b; Alexopoulos et al., 2004; Reeves and Brister, 2008). Recent
Pharmacology data suggest the efficacy of aripiprazole with doses as high as 15–30
Antipsychotics mg/day, but some recommend the more modest target of 10–15 mg/
No pharmacologic treatment guidelines for LOS have been estab- day, introduced at 2 mg/day (Madhusoodanan et al., 2004; Coley
lished to date, mostly due to the limited evidence in this specific et al., 2009; Kohen et al., 2010; Rado and Janicak, 2010). Lower
population. The bulk of the literature in this regard pertains to dosing strategies have been suggested for late-life psychosis in the
older patients with chronic schizophrenia; moreover, studies on the context of delusional disorder or dementia (Tsuboi et al., 2011).
late-onset population usually failed to distinguish between LOS and Amisulpride has also been demonstrated to be comparable to risp-
VLOSLP. A Cochrane systematic review on antipsychotic drugs use eridone in older patients and doses of 200–400 mg/day have been
in older people with LOS by Arunpongpaisal et al. was published used, introduced at 50 or 100 mg/day (Psarros et al., 2009). Depot
in 2003 and revised in 2012. In both editions there was no rand- medications may have a more continuous delivery and are often
omized controlled trial meeting their inclusion criteria and upon used in nonadherent patients, minimizing the plasma level fluctua-
which to base any guidelines; therefore clinicians must for now use tions that can be especially problematic in older people (Lasser et al.,
‘clinical judgment and habit to guide prescribing’ (Arunpongpaisal 2004). Long-acting risperidone at doses starting at 25 mg every
et al., 2003; Essali and Ali, 2012). However, a more recently pub- 2 weeks was shown to be safe in older patients followed for 1 year.
lished study by Jin et al in 2013, which specifically studied older Some patients received and tolerated doses as high as 75 mg every
adults on antipsychotic medications, found that quetiapine may 2 weeks (Lasser et al., 2004; Kissling et al., 2007; Singh and O’Connor,
have the highest incidence of adverse events. The study also noted 2009). As in the younger population, depot formulations have not
that older adults on antipsychotics had high discontinuation rates, been found to be more effective than oral antipsychotics if adher-
limited improvement in psychotic symptoms and a 36% incidence ence is good (Reeves et al., 2002). Paliperidone extended-release
of metabolic syndrome (Jin et al., 2013). (ER) consists of the active metabolite 9-hydroxyrisperidone in a
Atypical antipsychotics are now favoured, mainly because of tablet using patented extended-release technology to provide con-
the propensity of first-generation neuroleptics to induce extrapy- tinual and consistent delivery over 24 h. It has theoretical advan-
ramidal symptoms (EPS) and tardive dyskinesia (TD), to which tages in older people, like a once-daily schedule, minimal hepatic
metabolism, potentially fewer side effects, and limited plasma example, a socially stimulating group intervention targeting lonely
fluctuations in cases of missed doses (Turkoz et al., 2011). In a older individuals with schizophrenia has been associated with bet-
30-week trial involving patients around age 70 with EOS and LOS, ter cognitive functioning, improvements in wellbeing, and lower
paliperidone ER was found to be safe and well tolerated, with an mortality (Routasalo et al., 2009; Pitkala et al., 2011). This is partic-
age-related increase in somnolence and tachycardia. Age of onset ularly relevant to the LOS population impacted by high prevalence
did not impact these findings. Paliperidone also seemed to be effec- of social isolation, which has also been found to increase the risk
tive at reducing symptoms, but the study was not powered to assess of dementia (Wilson et al., 2007). Three nonpharmacological treat-
this outcome. Doses ranged between 3 and 12 mg/day and started ments have been designed specifically for the older population with
at 6 mg/day (Tzimos et al., 2008). schizophrenia and they are all group based.
According to a survey of expert clinicians in the US, risperidone Functional Adaptation Skills Training (FAST) and its modi-
should be the drug of choice in late-life schizophrenia, and quetiap- fied version for older Latinos—Programa de Entrenamiento de
ine, olanzapine, and aripiprazole are good second-line options Aptitudes para Latinos (PEDAL)—are manualized therapies imple-
(Alexopoulos et al., 2004). There is more limited evidence regarding mented in the group format. Both have been found to improve
the use of ziprasidone and clozapine and none for LOS specifically. social and everyday functional skills and to decrease the short-term
Clozapine especially should be used with caution considering its use of emergency medical services (Patterson et al., 2003, 2006;
side-effect profile and important anticholinergic activity, and poly- Mausbach et al., 2008).
pharmacy using more than one antipsychotic should be avoided. Cognitive-behavioural social skills training (CBSST) is another
group intervention designed for older people with schizophrenia. It
Adjunctive medications combines two modalities with efficacy in younger patients. CBSST is
Although younger age is correlated with a more frequent use of based on the concept of challenging the common beliefs that inter-
adjunctive mood stabilizers for schizophrenia, recent data support fere with treatment in this population and by providing repetitive
their use in older people as well. Adjunct ER valproate (mean dose practice of behaviours to improve retention and skill development.
587.50 mg/day) was effective and well tolerated in an open-label It has been associated with learning of new coping skills and bet-
study on 20 older chronically ill patients, and was associated with ter social functioning after treatment and also at 1-year follow-up.
symptomatic improvement, better global function, and lower However, gains in cognitive insight following CBSST appear to not
depression scores (Sajatovic et al., 2008; Sim et al., 2011). The be maintained at follow-up. This suggests that improvements in
resolution of comorbid depressive symptoms with adjunctive phy- function are not necessarily an outcome of better insight (McQuaid
toestrogens in a woman with LOS was discussed in a case report et al., 2000; Granholm et al., 2005, 2007).
(Rakesh et al., 2011). Cholinergic neurotransmission has been Enhanced skills training (ST) and healthcare management (HM)
implicated in psychosis and visual hallucinations, but there is cur- combine skills training for daily living and medication manage-
rently no evidence regarding the use of cholinesterase inhibitors for ment plus preventive nursing visits to address medical comorbidi-
this purpose, although there seem to be a role for them in manage- ties in older adults with severe mental illnesses (SMI). ST + HM has
ment of associated cognitive deficits (Patel et al., 2010; Ribeiz et al., been associated with improved social functioning and independent
2010). There is also a theoretical rationale for the use of memantine living skills, whereas functioning remained constant or declined for
and reports of a positive impact in cases of catatonic schizophrenia the HM-only group. The two groups receiving HM demonstrated
(Zdanys and Tampi, 2008). To our knowledge, there are no data increased use of preventive health services and identification of
supporting the addition of these agents in older adults, particularly previously undetected medical disorders (Bartels et al., 2004). A
those with LOS. derivative, Helping Older People Experience Success (HOPES), is
Addition of SSRIs, specifically citalopram, has been shown to be an integrated model of psychosocial rehabilitation and healthcare
effective in treating subsyndromal depressive symptoms in older management and appears helpful in improving community living
persons with schizophrenia. More importantly, such treatment may skills (Pratt et al., 2008).
reduce risk of suicide. The potential benefits of employment are numerous and older
There is supporting evidence regarding electroconvulsive ther- adults with SMI often manifest a desire to work despite low rates
apy (ECT) and repetitive transcranial magnetic stimulation (rTMS) of paid jobs (Auslander and Jeste, 2002; Twamley et al., 2005).
for refractory symptoms in younger patients with schizophrenia Ageing with schizophrenia can be associated with challenges in
(Matheson et al., 2010). To date, there are no publications report- the workplace (Jeste et al., 2003; Kurtz, 2005). However, competi-
ing LOS treatment with ECT, but rTMS in LOS has been the focus tive employment and, specifically, supported employment (SE) has
of a case report, describing one individual with auditory hallucina- been associated with better outcomes compared to conventional
tions successfully treated with both acute and maintenance rTMS vocational rehabilitation (CVR). The goal of SE is rapid and indi-
(Poulet et al., 2008). vidualized placement in competitive work with on-site training if
required, following the ‘place-then-train’ philosophy. CVR reflects
Psychosocial treatments the ‘train-then-place’ approach with prevocational training and
Most trials of psychosocial interventions for older persons with volunteering and gradual contact with competitive work (Twamley
schizophrenia have targeted patients with EOS. Keeping in mind et al., 2005, 2008).
the significant differences in premorbid psychosocial functioning
between those with EOS and those with LOS, there is a need for Treatment adherence and service utilization
studies comparing results of psychosocial interventions in early- No study has assessed treatment adherence in patients with LOS
and late-onset patients. Psychosocial interventions can have sig- specifically, but both psychosis and ageing have been linked to
nificant benefits other than immediate symptomatic relief. For adherence problems. Partial adherence to treatment is observed in
at least 20–50% of patients in the general population, but these rates of 30 months. They compared this group to 21 older inpatients
approach 70–80% in persons with psychotic disorders. In older with onset of schizophrenia before age 40. Worsening of symp-
patients taking antipsychotics and medications for hypertension, toms was reported in only one VLOSLP subject. Most had a sin-
diabetes, or hyperlipidaemia, researchers have reported similar gle episode with full or partial remission, while almost one-half of
adherence for psychiatric and nonpsychiatric medications. These the EOS patients experienced a relapse. The choice of the control
rates ranged from 52–64%. Several factors related to old age may group might have biased the findings, as the literature suggests that
result in poor adherence, including sensory impairments, cognitive ageing is associated with an overall improvement in psychosocial
deficits, osteoarthritis, restricted mobility, lack of transportation, function and psychopathology in older adults with EOS, although
social isolation, financial insecurity, polypharmacy, and increased most remain impaired (Jeste et al., 2003; Auslander and Jeste, 2004;
sensitivity to side effects. Poor adherence is associated with recur- Jeste et al., 2011). The findings in LOS patients are in keeping with
rence of symptoms and readmission, and with significant personal results for the VLOSLP population reported by Reeves et al. (2002),
and societal costs (Masand and Gupta, 2003). with 52% having a single admission. Similarly, Brodaty et al. (2003)
Studies have shown that among the population with chronic psy- followed individuals with first onset of schizophrenia at 50 years or
chotic disorders, older age is associated with a lower use of all men- above, and observed a considerable decrease in symptomatology
tal health services, with the exception of case management. This is over time; 68.4% of them fulfilled DSM-IV criteria 1 year after the
in keeping with the observed decline in mental health expenditures diagnosis, and only 16.7% met criteria at 5-year follow-up.
in older people and the fact that the higher costs of case manage- However, the broader literature tends to suggest partial, rather
ment and inpatient/crisis residential services among older persons than complete, symptomatic improvement over time in LOS and
are attributed largely to the population with schizophrenia. A sig- VLOSLP, which mirrors the literature on late paraphrenia from the
nificant drop in the use of outpatient services has been noted with 1960s (Kay and Roth, 1961). Jeste et al. (1995) reported that most
age; this might explain the more frequent use of psychiatric emer- of the LOS patients in their sample had a chronic course with ill-
gency response teams and psychiatric emergency unit admissions ness duration longer than 2 years, with an average of 5.7 years SD ±
(Jin et al., 2003; Gilmer et al., 2006). 5.5. Copeland et al. (1998) described residual symptoms and no case
These findings are consistent with those of McNulty et al. (2003) with complete recovery in subjects who had developed schizophre-
describing the high level of unmet care needs in a community nia after age 65, although worsening was not seen either. In a recent
sample of older people with schizophrenia or related disorders in study by Meesters et al. (2011), no significant difference in rates of
Scotland, 59% of them with onset of symptoms after age 45. Reeves symptomatic remission was observed among those patients with
et al. (2002) also observed that only 59% of older people previously early, late, or very-late onset. Literature on psychotic disorders in late
diagnosed with VLOSLP were still in contact with psychiatric serv- life seems to indicate that female gender, good treatment adherence,
ices after 3 years. and schizoaffective illness are predictors of a favourable outcome.
It is also critical to assess capacity for healthcare and financial Findings are more inconsistent regarding the impact of cognition,
decisions. There is considerable heterogeneity in the level of deci- sensory deficits, and ethnicity on the course of illness and psychotic
sional capacity among patients with schizophrenia, and age itself symptoms specifically, although cognitive impairment has been asso-
has not been found to strongly predict the lack of capacity to con- ciated with worse functional status (Harvey, 2001; Palmer et al., 2002;
sent. Negative and cognitive symptoms have a more significant Auslander and Jeste, 2004; Granholm et al., 2008). People diagnosed
impact on the decision-making capacity than psychosis per se, and with LOS and delusional disorder do not seem to differ in terms of
capacity should be assessed when appropriate, by using instru- outcomes. A dose-related effect was shown regarding antipsychotic
ments such as the MacArthur Competence Assessment Tool for use. Contact with a community psychiatric nurse also appears to lead
Treatment (MacCAT-T) (Grisso and Appelbaum, 1997). Advanced to a more favourable course in LOS and late paraphrenia (Holden,
care directives should be reviewed in any patient, but even more so 1987; Howard and Levy, 1992; Hassett, 2002; Reeves et al., 2002;
in older adults, and should minimally include a discussion on living Riecher-Rossler et al., 2003; Meesters et al., 2011).
will and proxy decision-making. Partial improvement in symptomatology can be achieved in most
patients, though this does not necessarily translate into functional
Prognosis recovery (Patterson et al., 2003). Brodaty et al. (2003) demon-
strated that even though the majority of LOS patients did not meet
Course of symptoms and functioning DSM-IV criteria for schizophrenia at follow-up, they were still
Considerable reductions in schizophrenia symptoms have been functionally impaired, with GAF scores still below 50. Nineteen of
observed in persons with LOS treated with antipsychotics. Open the 21 patients included in a follow-up study by Mazeh et al. (2005)
studies of typical antipsychotics have reported full remission were able to stay at home for 30 months following discharge. Mild
in 48–61% of patients and higher proportions of at least partial difficulties in the activities of daily living (ADLs), still within the
response (Howard et al., 2000; Sable and Jeste, 2002). Better results normal range, characterized the 13 subjects with LOS described
have been noted with second-generation antipsychotics, even in the by Laks et al. (2006), with no significant deterioration in the func-
VLOSLP population, with substantial improvements noted in up to tional scores at the 1-year reassessment. These results underscore
77% of patients, with a more favourable response in those from out- the heterogeneity in clinical and functional outcomes in the LOS
patient vs inpatient settings. In older people treated with atypical and VLOSLP populations.
neuroleptics, a more positive outcome could be seen in those with
LOS as opposed to EOS (Barak et al., 2002; Scott et al., 2010). Course of cognition
Mazeh et al. (2005) followed 21 inpatients fulfilling criteria for Although there is no indication of an increased risk of Alzheimer’s
schizophrenia with first onset at age 70 or later for a mean duration disease in patients with LOS, some data support a general
neurodegenerative pathophysiology. In follow-up studies on LOS specifically higher risk is not sure. Most outcome studies report-
or psychosis, significant cognitive deficits have been described in ing deceased subjects did not include a control group. The onset of
up to 50% of patients, especially in very-late-onset cases (Brodaty psychotic symptoms in late life was associated with double the risk
et al., 2003; Korner et al., 2008, 2009a, 2009b). of mortality at follow-up by Henderson et al. (1997), while Östling
Yet, most of the evidence points towards relative stability in the et al. (2007) found no such association. Although the risk of sui-
cognitive abilities of patients with LOS. A study by Palmer et al. cide might be somewhat lower in older patients with schizophrenia
(2003) resulted in no evidence of cognitive decline for schizophrenia by approximately 5% compared to younger affected individuals,
patients in both late- and early-onset groups compared to normal they are still at risk of suicidal behaviours, and age of onset was not
controls, while two Alzheimer’s disease comparison groups (per- found to mitigate this risk one way or another (Barak et al., 2004).
sons with very mild deficits and persons with psychotic symptoms
at baseline) manifested progressive impairments. The absence of References
significant change in neuropsychological performance over time in Abel, K.M., Drake, R., and Goldstein, J.M. (2010). Sex differences in
individuals with LOS has been replicated. Although people with LOS schizophrenia. International Review of Psychiatry, 22 (5), 417–28.
do not experience a greater worsening of their cognition compared Alexopoulos, G.S., et al. (2004). ‘Using antipsychotic agents in older patients.
to age-matched controls, they do not show the same level of improve- Journal of Clinical Psychiatry, 2, 5–99.
ment following intervention (Granholm et al., 2010). This could have Alici-Evcimen, Y., et al. (2003). Case series with late-onset psychosis
treatment implications, as it suggests a ceiling effect in response to hospitalized in a geriatric psychiatry unit in Turkey: experience in 9
years. International Psychogeriatrics, 15 (1), 69–72.
interventions that have a cognitive remediation component.
Almeida, O.P. and Fenner, S. (2002). Bipolar disorder: similarities and
Quality of life differences between patients with illness onset before and after 65 years
of age. International Psychogeriatrics, 14 (3), 311–22.
In older patients with schizophrenia, symptomatic remission has Almeida, O.P., et al. (1992). Late paraphrenia: a review. International Journal
been associated with a higher reported quality of life (QoL), inter- of Geriatric Psychiatry, 7 (8), 543–8.
mediate between that of younger patients and healthy comparison American Psychiatric Association (1980). Diagnostic and statistical manual
subjects (Auslander and Jeste, 2004; Bankole et al., 2008). A study of mental disorders: DSM-III. American Psychiatric Association,
by Folsom and colleagues suggested that older age may, in fact, be Washington, DC.
associated with greater mental health-related QoL in this population American Psychiatric Association (1987). Diagnostic and statistical manual
(Folsom et al., 2009). On the other hand, Meesters et al. (2011) could of mental disorders: DSM-III-R. American Psychiatric Association,
Washington, DC.
not demonstrate any correlation between symptomatic remission
and QoL in older people with schizophrenia, although the remission of mental disorders: DSM-IV. American Psychiatric Association,
rates in their sample were lower than those reported from conven- Washington, DC.
ience samples. This is consistent with studies indicating significantly American Psychiatric Association (2000). Diagnostic and statistical manual
greater improvement in psychological and health-related QoL in of mental disorders: DSM-IV-TR. American Psychiatric Association,
patients treated with olanzapine rather than risperidone (Ritchie et Washington, DC.
al., 2003, 2006). In light of the significant interplay between cognitive Arunpongpaisal, S., et al., (2003). Antipsychotic drug treatment for elderly
status and social situation with regard to overall function and QoL, people with late-onset schizophrenia. Cochrane Database of Systematic
the failure to include measures of QoL in most outcome studies on Reviews, (2), CD004162.
Auslander, L.A. and Jeste, D.V. (2002). Perceptions of problems and needs for
LOS or VLOSLP as well as the exclusion of late-onset patients from
service among middle-aged and elderly outpatients with schizophrenia
some papers on QoL area is a drawback (Meesters et al., 2010). and related psychotic disorders. Community Mental Health Journal, 38
A similar effect to medications has not been as clearly estab- (5), 391–402.
lished with psychosocial interventions. No impact on QoL was Auslander, L.A. and Jeste, D.V. (2004). Sustained remission of schizophrenia
demonstrated following participation in FAST (Patterson et al., among community-dwelling older outpatients. American Journal of
2003, 2006). Improvement in QoL resulting from better function- Psychiatry, 161 (8), 1490–3.
ing following CBSST was postulated by the authors, but as QoL Bankole, A., et al. (2008). Symptomatic remission in a multiracial urban
was not part of the outcomes measured, this remains hypotheti- population of older adults with schizophrenia. American Journal of
cal (Emmerson et al., 2009). We did not find any mention of QoL Geriatric Psychiatry, 16 (12), 966–73.
Barak, Y., et al. (2002). Very late-onset schizophrenia-like psychosis:
in relation to ST + HM (Bartels et al., 2004). However, vocational
clinical and imaging characteristics in comparison with elderly patients
rehabilitation, particularly supported employment, does appear with schizophrenia. Journal of Nervous and Mental Disease, 190 (11),
to lead to better QoL (Twamley et al., 2005; Twamley et al., 2008; 733–6.
Bartels and Pratt, 2009). This could indicate that better functioning Barak, Y., Knobler, C.Y., and Aizenberg, D. (2004). Suicide attempts amongst
seems to be associated with greater wellbeing; other avenues also elderly schizophrenia patients: a 10-year case-control study. Schizophrenic
need to be considered. The importance of positive social support Research, 71 (1), 77–81.
and interactions cannot be minimized, as married middle-aged and Barta, P.E., et al. (1997). Quantitative MRI volume changes in late onset
older adults with schizophrenia reported better QoL than those schizophrenia and Alzheimer’s disease compared to normal controls.
who were not, despite the presence of depressive symptoms in both Psychiatry Research, 68 (2–3), 65–75.
Bartels, S.J. and Pratt, S.I. (2009). Psychosocial rehabilitation and quality of
groups (Nyer et al., 2010).
life for older adults with serious mental illness: recent findings and future
research directions. Current Opinions in Psychiatry, 22 (4), 381–5.
Mortality
Bartels, S.J., et al. (2004). Enhanced skills training and health care management
An increase in morbidity and mortality rates has been observed for older persons with severe mental illness. Community Mental Health
in schizophrenia in general, but whether late-onset cases herald a Journal, 40 (1), 75–90.
Behl, C., et al. (1995). 17-beta estradiol protects neurons from oxidative Cooper, A.F. (1976). Deafness and psychiatric illness. British Journal of
stress-induced cell death in vitro. Biochemical and Biophysical Research Psychiatry, 129, 216–26.
Communications, 216 (2), 473–82. Cooper, A.F. and Curry, A.R. (1976). The pathology of deafness in the
Blazer, D.G. and Wu, L.T. (2011). The epidemiology of alcohol use paranoid and affective psychoses of later life. Journal of Psychosomatic
disorders and subthreshold dependence in a middle-aged and elderly Research, 20 (2), 97–105.
community sample American Journal of Geriatric Psychiatry, 19 (8), Cooper, A.F., et al. (1974). Hearing loss in paranoid and affective psychoses
685–94. of the elderly. Lancet, 2 (7885), 851–4.
Blazer, D.G., Hays, J.C., and Salive, M.E. (1996). Factors associated with Copeland, J.R., et al. (1998). Schizophrenia and delusional disorder in older
paranoid symptoms in a community sample of older adults. The age: community prevalence, incidence, comorbidity, and outcome.
Gerontologist, 36 (1), 70–5. Schizophrenia Bulletin, 24 (1), 153–61.
Bleuler, E. (1950). Dementia praecox or the group of schizophrenias. Oxford Corbin, S.L. and Eastwood, M.R. (1986). Sensory deficits and mental
University Press, Oxford. disorders of old age: causal or coincidental associations? Psychological
Bleuler, M. (1943). ‘Clinical picture of the late schizophrenia. Fortschritte der Medicine, 16 (2), 251–6.
Neurologie und Psychiatrie und Ihrer Grenzgebiete, 15 (9), 259–90. Corey-Bloom, J., et al. (1995). Quantitative magnetic resonance imaging of
Bogren, M., et al. (2010a). Incidence of psychotic disorders in the 50 year the brain in late-life schizophrenia. American Journal of Psychiatry, 152
follow up of the Lundby population. Australian and New Zealand Journal (3), 447–9.
of Psychiatry, 44 (1), 31–9. Correll, C.U., Leucht, S., and Kane, J.M. (2004). Lower risk for tardive
Bogren, M., et al. (2010b). Predictors of psychosis: a 50-year follow-up of dyskinesia associated with second-generation antipsychotics: a
the Lundby population. European Archives of Psychiatry and Clinical systematic review of 1-year studies. American Journal of Psychiatry, 161
Neuroscience, 260 (2), 113–25. (3), 414–25.
Bora, E., Yucel, M., and Pantelis, C. (2010). Cognitive impairment in De Luca d’Alessandro, E., Bonacci, S., and Giraldi, G. (2011). Aging
schizophrenia and affective psychoses: implications for DSM-V criteria populations: the health and quality of life of the elderly. Clinica
and beyond. Schizophrenia Bulletin, 36 (1), 36–42. Terapeutica, 162 (1), e13–18.
Borglum, A.D., et al. (2001). Dopa decarboxylase genotypes may influence Dewi R. W. (1971). The hallucinations of widowhood. British Medical Journal,
age at onset of schizophrenia. Molecular Psychiatry, 6 (6), 712–17. 4 (5778), 37–41.
Boyce, N. and Walker, Z. (2008). Late-onset schizophrenia and very late-onset Dubertret, C., Ades, J., and Gorwood, P. (2004). Clinical and etiopathogenic
schizophrenia-like psychosis. Psychiatry, 7 (11), 463–6. specificities of the French concept of psychose hallucinatoire chronique
Bozikas, V.P., et al. (2002). Neurofibrillary tangles in elderly patients with late compared to schizophrenia. Schizophrenia Bulletin, 30 (1), 173–84.
onset schizophrenia. Neuroscience Letters, 324 (2), 109–12. Eastwood, R., Corbin, S., and Reed, M. (1981). Hearing impairment and
Brann, D.W., et al. (2007). Neurotrophic and neuroprotective actions of paraphrenia. Journal of Otolaryngology, 10 (4), 306–8.
estrogen: basic mechanisms and clinical implications. Steroids, 72 (5), Emmerson, L.C., et al. (2009). Insight and treatment outcome with
381–405. cognitive-behavioral social skills training for older people with
Breitner, J.C., et al. (1990). Cerebral white matter disease in late-onset schizophrenia. Journal of Rehabilitation Research and Development, 46
paranoid psychosis. Biology of Psychiatry, 28 (3), 266–74. (8), 1053–8.
Brodaty, H., et al. (1999). Schizophrenia with onset after age 50 years. I: Essali, A. and Ali, G. (2012). Antipsychotic drug treatment for elderly people
phenomenology and risk factors. British Journal of Psychiatry, 175, with late-onset schizophrenia. Cochrane Database of Systematic Reviews,
410–15. 12 Feb (2), Epub.
Brodaty, H., et al. (2003). Long-term outcome of late-onset schizophrenia: Evans, J.D., et al. (1996). A clinical and neuropsychological comparison of
5-year follow-up study. British Journal of Psychiatry, 183, 213–19. delusional disorder and schizophrenia. Journal of Neuropsychiatry and
Brunelle, S., Cole, M.G., and Elie, M. (2011). Risk factors for the late-onset Clinical Neuroscience, 8 (3), 281–6.
psychoses: a systematic review of cohort studies. International Journal of Folsom, D.P., et al. (2009). Physical and mental health-related quality of life
Geriatric Psychiatry, 6 (10), 240–52. among older people with schizophrenia. Schizophrenia Research, 108
Casanova, M.F. and Lindzen, E.C. (2003). Changes in gray-/white-matter (1–3), 207–13.
ratios in the parahippocampal gyri of late-onset schizophrenia patients. Forsell, Y. (2000). Predictors for depression, anxiety and psychotic symptoms
American Journal of Geriatric Psychiatry, 11 (6), 605–9. in a very elderly population: data from a 3-year follow-up study. Society
Casanova, M.F., et al. (2002). Disentangling the pathology of schizophrenia of Psychiatry and Psychiatric Epidemiology, 35 (6), 259–63.
and paraphrenia. Acta Neuropathologica, 103 (4), 313–20. Forsell, Y. and Henderson, A.S. (1998). Epidemiology of paranoid symptoms
Castle, D.J. and Morgan, V. (2008). Epidemiology. Clinical handbook of in an elderly population. British Journal of Psychiatry, 172, 429–32.
schizophrenia, pp. 14–24. Guilford Press, New York. Fromholt, P., Bender, L., and Elleberg, B. (1999). Coping with a paraphrenic
Castle, D.J. and Murray, R.M. (1993). The epidemiology of late-onset world in late life. Journal of Clinical Geropsychology, 5 (4), 301–8.
schizophrenia. Schizophrenia Bulletin, 19 (4), 691–700. Fuchs, T. (1994). Uprooting and late-life psychosis. European Archives of
Castle, D.J., et al. (1997). Schizophrenia with onset at the extremes of adult Psychiatry and Clinical Neuroscience, 244 (3), 126–30.
life. International Journal of Geriatric Psychiatry, 12 (7), 712–17. Fuchs, T. (1999a). Patterns of relation and premorbid personality in late
Christenson, R. and Blazer, D. (1984). Epidemiology of persecutory ideation paraphrenia and depression. Psychopathology, 32 (2), 70–80.
in an elderly population in the community. American Journal of Fuchs, T. (1999b). Life events in late paraphrenia and depression.
Psychiatry, 141 (9), 1088–91. Psychopathology, 32 (2), 60–9.
Cohen, C.I., et al. (2000). Schizophrenia and older adults: an overview: Giblin, S., et al. (2004). Psychosocial correlates of late-onset psychosis: life
directions for research and policy. American Journal of Geriatric experiences, cognitive schemas, and attitudes to ageing. International
Psychiatry, 8 (1), 19–28. Journal of Geriatric Psychiatry, 19 (7), 611–23.
Cohen, C.I., et al. (2011). The relationship between trauma and clinical Gilmer, T., et al. (2006). Costs of community-based public mental health
outcome variables among older adults with schizophrenia spectrum services for older adults: variations related to age and diagnosis.
disorders. American Journal of Geriatric Psychiatry, 21, 21. International Journal of Geriatric Psychiatry, 21 (12), 1121–6.
Coley, K.C., et al. (2009). Aripiprazole prescribing patterns and side effects Girard, C. and Simard, M. (2008). Clinical characterization of late- and very
in elderly psychiatric inpatients. Journal of Psychiatric Practice, 15 (2), late-onset first psychotic episode in psychiatric inpatients. American
150–3. Journal of Geriatric Psychiatry, 16 (6), 478–87.
Girard, C., et al. (2011). Late-onset-psychosis: cognition. International Howard, R. (2006). Commentary. International Journal of Geriatric Psychiatry,
Psychogeriatrics, 23 (8), 1301–16. 21 (1), 5.
Goldberg, X., et al. (2011). Neurodevelopmental liability to schizophrenia: Howard, R. and Levy, R. (1992). Which factors affect treatment response
the complex mediating role of age at onset and premorbid adjustment. in late paraphrenia? International Journal of Geriatric Psychiatry, 7 (9),
Schizophrenic Research, 11, 11. 667–72.
Grahame, P.S. (1984). Schizophrenia in old age (late paraphrenia). British Howard, R., et al. (1992). Permeable walls, floors, ceilings and doors.
Journal of Psychiatry, 145, 493–5. Partition delusions in late paraphrenia. International Journal of Geriatric
Granholm, E., et al. (2005). A randomized, controlled trial of cognitive Psychiatry, 7 (10), 719–24.
behavioral social skills training for middle-aged and older outpatients with Howard, R., et al. (1993). A comparative study of 470 cases of early-onset and
chronic schizophrenia. American Journal of Psychiatry, 162 (3), 520–9. late-onset schizophrenia. British Journal of Psychiatry, 163, 352–7.
Granholm, E., et al. (2007). Randomized controlled trial of cognitive Howard, R., Almeida, O., and Levy, R. (1994). Phenomenology, demography
behavioral social skills training for older people with schizophrenia: and diagnosis in late paraphrenia. Psychological Medicine, 24 (2),
12-month follow-up. Journal of Clinical Psychiatry, 68 (5), 730–7. 397–410.
Granholm, E., et al. (2008). Neuropsychological predictors of functional Howard, R., et al. (1995). Magnetic resonance imaging volumetric
outcome in Cognitive Behavioral Social Skills Training for older people measurements of the superior temporal gyrus, hippocampus,
with schizophrenia. Schizophrenic Research, 100 (1–3), 133–43. parahippocampal gyrus, frontal and temporal lobes in late paraphrenia.
Granholm, E., et al. (2010). Age-related practice effects across longitudinal Psychological Medicine, 25 (3), 495–503.
neuropsychological assessments in older people with schizophrenia. Howard, R.J., et al. (1997). A controlled family study of late-onset non-affective
Neuropsychology, 24 (5), 616–24. psychosis (late paraphrenia). British Journal of Psychiatry, 170, 511–14.
Grisso, T., and Appelbaum, P.S. (1997). The MacCAT-T: a clinical tool to Howard, R., et al. (2000). Late-onset schizophrenia and very-late-onset
assess patients’ capacities to make treatment decisions. Psychiatric schizophrenia-like psychosis: an international consensus. The
Services, 48, 1415–19. International Late-Onset Schizophrenia Group. American Journal of
Gurian, B.S., Wexler, D., and Baker, E.H. (1992). Late-life paranoia: possible Psychiatry, 157 (2), 172–8.
association with early trauma and infertility. International Journal of Huang, C. and Zhang, Y.-L. (2009). Clinical differences between late-onset
Geriatric Psychiatry, 7 (4), 277–84. and early-onset chronically hospitalized elderly schizophrenic patients
H ä fner, H. (2003). Gender differences in schizophrenia in Taiwan. International Journal of Geriatric Psychiatry, 24 (10),
Psychoneuroendocrinology, 28 Suppl 2, 17–54. 1166–72.
Hafner, H., et al. (1998). Causes and consequences of the gender difference in Huang, Q. and Tang, J. (2010). Age-related hearing loss or presbycusis.
age at onset of schizophrenia. Schizophrenia Bulletin, 24 (1), 99–113. European Archives of Otorhinolaryngology, 267 (8), 1179–91.
Hamshere, M.L., et al. (2011). Phenotype evaluation and genomewide linkage Iglewicz, A., Meeks, T.W., and Jeste, D.V. (2011). New wine in old bottle:
study of clinical variables in schizophrenia. American Journal of Medicine late-life psychosis. Psychiatric Clinics of North America, 34 (2), 295–318.
and Genetics Series B Neuropsychiatric Genetics, 29 (10), 31240. Isaac, M. and Koch, A. (2010). The risk of death among adult participants
Hardy, J. (2003). The relationship between Lewy body disease, Parkinson’s in trials of antipsychotic drugs in schizophrenia. European
disease, and Alzheimer’s disease. Annals of the New York Academy of Neuropsychopharmacology, 20 (3), 139–45.
Science, 991, 167–70. Janzarik, W. (1957). Zur Problematik schizophrener Psychosen im höheren
Harris, M.J. and Jeste, D.V. (1988). Late-onset schizophrenia: an overview. Lebensalter. [On the problems of schizophrenic psychoses in advanced
Schizophrenia Bulletin, 14 (1), 39–55. age.]. Der Nervenarzt, 28, 535–42.
Harvey, P.D. (2001). Cognitive and functional impairments in elderly patients Janzarik, W. (1973). Uber das Kontaktmangelparanoid des höheren Alters
with schizophrenia: a review of the recent literature. Harvard Review of und den Syndromcharakter schizophrenen Krankseins. Der Nervenarzt,
Psychiatry, 9 (2), 59–68. 44 (10), 515–26.
Hassett, A. (2002). Schizophrenia and delusional disorders with onset in later Jeste, D.V. and Nasrallah, H.A. (2003). Schizophrenia and aging no more
life. Revista Brasileira de Psiquiatria, 24, 81–6. dearth of data? American Journal of Geriatric Psychiatry, 11 (6), 584–7.
Haug, J.O. (1962). Pneumoencephalographic studies in mental disease. Acta Jeste, D.V., et al. (1995). Clinical and neuropsychological characteristics of
Psychiatrica Scandinavica Supplementum, 38 (165), 1–104. patients with late-onset schizophrenia. American Journal of Psychiatry,
Heaton, R., et al. (1994). Neuropsychological deficits in schizophrenics. 152 (5), 722–30.
Relationship to age, chronicity, and dementia. Archives of General Jeste, D.V., et al. (1996a). Medical comorbidity in schizophrenia. Schizophrenia
Psychiatry, 51 (6), 469–76. Bulletin, 22 (3), 413–30.
Henderson, A. and Kay, D. (1997). The epidemiology of functional psychoses Jeste, D.V., et al. (1996b). Management of late-life psychosis. Journal of
of late onset. European Archives of Psychiatry and Clinical Neuroscience, Clinical Psychiatry, 3, 39–45.
247 (4), 176–89. Jeste, D.V., et al. (1997). Nondementia nonpraecox dementia praecox?
Henderson, A.S., et al. (1998). Psychotic symptoms in the elderly: a Late-onset schizophrenia. American Journal of Geriatric Psychiatry, 5 (4),
prospective study in a population sample. International Journal of 302–17.
Geriatric Psychiatry, 13 (7), 484–92. Jeste, D.V., et al. (1999a). Conventional vs. newer antipsychotics in elderly
Herbert, M.E. and Jacobson, S. (1967). Late paraphrenia. British Journal of patients. American Journal of Geriatric Psychiatry, 7 (1), 70–6.
Psychiatry, 113 (498), 461–9. Jeste, D.V., et al. (1999b). Lower incidence of tardive dyskinesia with
Hobbs F.B. and Damon B.L. (eds) (1996). 65 + in the United States. Washington risperidone compared with haloperidol in older patients. Journal of the
Bureau of the Census, Washington, DC. American Geriatric Society, 47 (6), 716–19.
Hofer, A., et al. (2010). Neurocognition and social cognition in patients with Jeste, D.V., et al. (2003). Aging and outcome in schizophrenia. Acta
schizophrenia or mood disorders. Neuropsychiatry, 24 (3), 161–9. Psychiatrica Scandinavica, 107 (5), 336–43.
Holden, N.L. (1987). Late paraphrenia or the paraphrenias? A descriptive Jeste, D.V., Wolkowitz, O. M., and Palmer, B. W. (2011). Divergent trajectories
study with a 10-year follow-up. British Journal of Psychiatry, 150 (5), of physical, cognitive, and psychosocial aging in schizophrenia.
635–9. Schizophrenia Bulletin, 37 (3), 451–5.
Hopkins, B. and Roth, M. (1953). Psychological test performance in patients Jin, H., et al. (2003). Patterns of public mental health service use by age in
over sixty. II. Paraphrenia, arteriosclerotic psychosis and acute confusion. patients with schizophrenia. American Journal of Geriatric Psychiatry, 11
British Journal of Psychiatry, 99 (416), 451–63. (5), 525–33.
Jin, H., et al (2013). Comparison of longer-term safety and effectiveness Leboyer, M., et al. (1992). No gender effect on age at onset in familial
of 4 atypical antipsychotics in patients over age 40: a trial using schizophrenia? American Journal of Psychiatry, 149(10),1409.
equipoise-stratified randomization. Journal of Clinical Psychiatry, 74(1), Lee, S.J. and McEwen, B.S. (2001). Neurotrophic and neuroprotective actions
10–8 of estrogens and their therapeutic implications. Annual Review of
Jones, D.K., et al. (2005). A diffusion tensor magnetic resonance imaging Pharmacology and Toxicology, 41, 569–91.
study of frontal cortex connections in very-late-onset schizophrenia-like Lemmens, P., Brecher, M., and Van Baelen, B. (1999). A combined analysis
psychosis. American Journal of Geriatric Psychiatry, 13 (12), 1092–9. of double-blind studies with risperidone vs. placebo and other
Jørgensen, P., et al. (1997). Acute and transient psychotic disorder: a 1-year antipsychotic agents: factors associated with extrapyramidal symptoms.
follow-up study. Acta Psychiatrica Scandinavica, 96 (2), 150–4. Acta Psychiatrica Scandinavica, 99 (3), 160–70.
Jorgensen, P. and Munk-Jorgensen, P. (1985). Paranoid psychosis in the elderly. Lohr, J.B., et al. (1997). Minor physical anomalies in older patients with
A follow-up study. Acta Psychiatrica Scandinavica, 72 (4), 358–63. late-onset schizophrenia, early-onset schizophrenia, depression, and
Kales, H.C., et al. (2007). Mortality risk in patients with dementia treated Alzheimer’s disease. American Journal of Geriatric Psychiatry, 5 (4),
with antipsychotics versus other psychiatric medications. American 318–23.
Journal of Psychiatry, 164 (10), 1568–76. Madhusoodanan, S., et al. (2004). Clinical experience with aripiprazole
Kay, D.W. and Roth, M. (1961). Environmental and hereditary factors in treatment in ten elderly patients with schizophrenia or schizoaffective
the schizophrenias of age (‘late paraphrenia’) and their bearing on the disorder: retrospective case studies. CNS Spectrum, 9 (11), 862–7.
general problem of causation in schizophrenia. Journal of Mental Science, Manford, M. and Andermann, F. (1998). Complex visual hallucinations.
107, 649–86. Clinical and neurobiological insights. Brain, 121 (10), 1819–40.
Keefe, R.S. and Fenton, W.S. (2007). How should DSM-V criteria for Marcelis, M., et al. (1998). Urbanization and psychosis: a study of 1942–1978
schizophrenia include cognitive impairment? Schizophrenia Bulletin, 33 birth cohorts in the Netherlands. Psychological Medicine, 28 (4), 871–9.
(4), 912–20. Martinez-Martin, N., et al. (2011). Self-perceived needs are related to violent
Kelly, D.L., et al. (2010). Cardiovascular disease mortality in patients with behavior among schizophrenia outpatients. Journal of Nervous Mental
chronic schizophrenia treated with clozapine: a retrospective cohort Diseases, 199 (9), 666–71.
study. Journal of Clinical Psychiatry, 71 (3), 304–11. Masand, P.S. and Gupta, S. (2003). Long-acting injectable antipsychotics
Khan, A.M., Clark, T., and Oyebode, F. (1988). Unilateral auditory in the elderly: guidelines for effective use. Drugs and Aging, 20 (15),
hallucinations. British Journal of Psychiatry, 152, 297–8. 1099–110.
Khouzam, H.R., et al. (2005). Psychoses in late life: evaluation and management Matheson, S.L., et al. (2010). Quality assessment and comparison of evidence
of disorders seen in primary care. Geriatrics, 60 (3), 26–33. for electroconvulsive therapy and repetitive transcranial magnetic
Kissling, W., et al. (2007). Long-term safety and efficacy of long-acting stimulation for schizophrenia: a systematic meta-review. Schizophrenic
risperidone in elderly psychotic patients. Human Psychopharmacology, Research, 118 (1–3), 201–10.
22 (8), 505–13. Mausbach, B.T., et al. (2008). Reducing emergency medical service use
Kohen, I., Lester, P.E., and Lam, S. (2010). Antipsychotic treatments for the in patients with chronic psychotic disorders: results from the FAST
elderly: efficacy and safety of aripiprazole. Neuropsychiatric Disease intervention study. Behavioural Research Therapy, 46 (1), 145–53.
Treatment, 6, 47–58. Mazeh, D., et al. (2005). Patients with very-late-onset schizophrenia-like
Köhler, S., et al. (2007). Psychosis risk as a function of age at onset: a psychosis: a follow-up study. American Journal of Geriatric Psychiatry,
comparison between early- and late-onset psychosis in a general 13 (5), 417–19.
population sample. Society of Psychiatry and Psychiatric Epidemiology, 42 McCulloch, Y., et al. (2006). Psychological processes underlying delusional
(4), 288–94. thinking in late-onset psychosis: a preliminary investigation. International
Kojo, K. (2010). Late-onset schizophrenic syndromes in socially isolated Journal of Geriatric Psychiatry, 21 (8), 768–77.
situations: a comparison of Janzarik’s ‘Kontaktmangelparanoid’ and late McNulty, S.V., et al. (2003). Care needs of elderly people with schizophrenia.
paraphrenia. Psychogeriatrics, 10 (2), 83–9. Assessment of an epidemiologically defined cohort in Scotland. British
Kolsch, H. and Rao, M.L. (2002). Neuroprotective effects of estradiol-17beta: Journal of Psychiatry, 182, 241–7.
implications for psychiatric disorders. Archives of Women’s Mental McQuaid, J.R., et al. (2000). ‘Development of an integrated
Health, 5 (3), 105–10. cognitive-behavioral and social skills training intervention for older
Korner, A., et al. (2008). Delusional disorder in old age and the risk of patients with schizophrenia. Journal of Psychotherapy and Practice
developing dementia: a nationwide register-based study. Aging and Research, 9 (3), 149–56.
Mental Health, 12 (5), 625–9. Meesters, P.D., et al. (2010). Social functioning among older
Korner, A., et al. (2009a). Acute and transient psychosis in old age and the community-dwelling patients with schizophrenia: a review. American
subsequent risk of dementia: a nationwide register-based study. Geriatric Journal of Geriatric Psychiatry, 18 (10), 862–78.
Gerontology International, 9 (1), 62–8. Meesters, P.D., et al. (2011). Symptomatic remission and associated factors in
Korner, A., et al. (2009b). Late and very-late first-contact schizophrenia and a catchment area based population of older patients with schizophrenia.
the risk of dementia—a nationwide register based study. International Schizophrenic Research, 126 (1–3), 237–44.
Journal of Geriatric Psychiatry, 24 (1), 61–7. Miller, B.L., et al. (1986). Late-life paraphrenia: an organic delusional
Kraepelin, E. (1899). Allgemeine psychiatrie. Barth, Leipzig. syndrome. Journal of Clinical Psychiatry, 47 (4), 204–7.
Kurtz, M.M. (2005). Neurocognitive impairment across the lifespan in Miller, B.L., et al. (1989). Brain white-matter lesions and psychosis. British
schizophrenia: an update. Schizophrenic Research, 74 (1), 15–26. Journal of Psychiatry, 155, 73–8.
Laks, J., et al. (2006). Absence of dementia in late-onset schizophrenia: a one Miller, B.L., et al. (1991). ‘Brain lesions and cognitive function in late-life
year follow-up of a Brazilian case series. Arquivos de Neuro-Psiquiatria, psychosis. British Journal of Psychiatry, 158, 76–82.
64 (4), 946–9. Mitter, P.R., et al. (2004). The effect of ethnicity and gender on first-contact
Lasser, R.A., et al. (2004). Efficacy and safety of long-acting risperidone rates for schizophrenia-like psychosis in Bangladeshi, black and white
in elderly patients with schizophrenia and schizoaffective disorder. elders in Tower Hamlets, London. International Journal of Geriatric
International Journal of Geriatric Psychiatry, 19 (9), 898–905. Psychiatry, 19 (3), 286–90.
Laughren, T. (2011). FDA Perspective on the DSM-5 approach to Mitter, P., et al. (2005). Migrant status, age, gender and social isolation in very
classification of ‘cognitive’ disorders. Journal of Neuropsychiatry and late-onset schizophrenia-like psychosis. International Journal of Geriatric
Clinical Neuroscience, 23 (2), 126–31. Psychiatry, 20 (11), 1046–51.
Moore, R., et al. (2006). Misunderstanding the intentions of others: an Prager, S. and Jeste, D.V. (1993). Sensory impairment in late-life schizophrenia.
exploratory study of the cognitive etiology of persecutory delusions Schizophrenia Bulletin, 19 (4), 755–72.
in very late-onset schizophrenia-like psychosis. American Journal of Pratt, S.I., et al. (2008). Helping older people experience success: an integrated
Geriatric Psychiatry, 14 (5), 410–18. model of psychosocial rehabilitation and health care management for
Mulsant, B.H., et al. (1993). Schizophrenia in late life: elderly patients older adults with serious mental illness. American Journal of Psychiatric
admitted to an acute care psychiatric hospital. Schizophrenia Bulletin, 19 Rehabilitation, 11 (1), 41–60.
(4), 709–21. Psarros, C., et al. (2009). Amisulpride for the treatment of very-late-onset
Naguib, M. and Levy, R. (1987). Late paraphrenia: neuropsychological schizophrenia-like psychosis. International Journal of Geriatric Psychiatry,
impairment and structural brain abnormalities on computed tomography. 24 (5), 518–22.
International Journal of Geriatric Psychiatry, 2 (2), 83–90. Quin, R.C., et al. (2009). ‘Not of this world’: the subjective experience of
Newman, C.W. and Sandridge, S.A. (2004). Hearing loss is often late-onset psychosis. Aging and Mental Health, 13 (6), 779–87.
undiscovered, but screening is easy. Cleve Clinical Journal of Medicine, Rabins, P., et al. (1987). Increased ventricle-to-brain ratio in late-onset
71 (3), 225–32. schizophrenia. American Journal of Psychiatry, 144 (9), 1216–18.
Nyer, M., et al. (2010). The relationship of marital status and clinical Rabins, P.V., et al. (2000). MRI findings differentiate between late-onset
characteristics in middle-aged and older patients with schizophrenia and schizophrenia and late-life mood disorder. International Journal of
depressive symptoms. Annals of Clinical Psychiatry, 22 (3), 172–9. Geriatric Psychiatry, 15 (10), 954–60.
Ostling, S., Palsson, S.P., and Skoog, I. (2007). The incidence of first-onset Rado, J. and Janicak, P.G. (2010). Aripiprazole for late-life schizophrenia.
psychotic symptoms and paranoid ideation in a representative population Journal of Clinical Interventions in Aging, 5, 253–8.
sample followed from age 70–90 years. Relation to mortality and later Rajji, T.K. and Mulsant, B.H. (2008). Nature and course of cognitive function
development of dementia. International Journal of Geriatric Psychiatry, in late-life schizophrenia: a systematic review. Schizophrenia Research,
22 (6), 520–8. 102 (1–3), 122–40.
Ouyang, W.C., et al. (2001). Estrogen receptor alpha gene polymorphism in Rajji, T.K., Ismail, Z., and Mulsant, B.H. (2009). Age at onset and cognition
schizophrenia: frequency, age at onset, symptomatology and prognosis. in schizophrenia: meta-analysis. British Journal of Psychiatry, 195 (4),
Psychiatric Genetics, 11 (2), 95–8. 286–93.
Palmer, B.W., Heaton, S.C., and Jeste, D.V. (1999). Older patients with Rakesh, G., et al. (2011). Treatment with phytoestrogens for depressive
schizophrenia: challenges in the coming decades. Psychiatric Services, 50 symptoms in late-onset schizophrenia: a case report. Primary Care
(9), 1178–83. Companion to the Journal of Clinical Psychiatry, 13 (3), pii.
Palmer, B.W., et al. (2002). Heterogeneity in functional status among older Rasanen, S., et al. (2000). Sex differences in schizophrenia: a review. Nordic
outpatients with schizophrenia: employment history, living situation, Journal of Psychiatry, 54 (1), 37–45.
and driving. Schizophrenic Research, 55 (3), 205–15. Rasmussen, H.B., et al. (2006). Association between the CCR5 32-bp deletion
Palmer, B.W., et al. (2003). Are late-onset schizophrenia spectrum disorders allele and late onset of schizophrenia. American Journal of Psychiatry,
neurodegenerative conditions? Annual rates of change on two dementia 163 (3), 507–11.
measures. Journal of Neuropsychiatry and Clinical Neurosciences, 15 (1), Reeves, R.R. and Brister, J.C. (2008). Psychosis in late life: emerging issues.
45–52. Journal of Psychosocial Nursing and Mental Health Services, 46 (11),
Patel, S.S., et al. (2010). Acetylcholinesterase inhibitors (AChEIs) for the 45–52.
treatment of visual hallucinations in schizophrenia: a review of the Reeves, S.J., et al. (2001). Increased first-contact rates for very-late-onset
literature. BMC Psychiatry, 10, 69. schizophrenia-like psychosis in African- and Caribbean-born elders.
Patterson, T.L., et al. (2003). Functional adaptation skills training (FAST): British Journal of Psychiatry, 179, 172–4.
a pilot psychosocial intervention study in middle-aged and older Reeves, S., Stewart, R., and Howard, R. (2002). Service contact and
patients with chronic psychotic disorders. American Journal of Geriatric psychopathology in very-late-onset schizophrenia-like psychosis:
Psychiatry, 11 (1), 17–23. the effects of gender and ethnicity. International Journal of Geriatric
Patterson, T.L., et al. (2006). Functional adaptation skills training (FAST): Psychiatry, 17 (5), 473–9.
a randomized trial of a psychosocial intervention for middle-aged and Reeves, S., et al. (2003). Are Black Caribbean patients more likely to receive
older patients with chronic psychotic disorders. Schizophrenic Research, an incorrect diagnosis of very-late-onset schizophrenia-like psychosis
86 (1–3), 291–9. than their white British counterparts? American Journal of Geriatric
Paulsen, J.S., et al. (2000). Neurobehaviors and psychotic symptoms in Psychiatry, 11 (6), 674–7.
Alzheimer’s disease. Journal of the International Neuropsychological Reulbach, U., et al. (2007). Late-onset schizophrenia in child survivors of the
Society, 6 (7), 815–20. holocaust. Journal of Nervous and Mental Disease, 195 (4), 315–19.
Pearlson, G. and Rabins, P. (1988). The late-onset psychoses. Possible risk Ribeiz, S.R., et al. (2010). Cholinesterase inhibitors as adjunctive therapy in
factors. Psychiatric Clinics of North America, 11 (1), 15–32. patients with schizophrenia and schizoaffective disorder: a review and
Pearlson, G.D., et al. (1989). A chart review study of late-onset and early-onset meta-analysis of the literature. CNS Drugs, 24 (4), 303–17.
schizophrenia. American Journal of Psychiatry, 146 (12), 1568–74. Riecher-Rossler, A., et al. (1995). Late-onset schizophrenia and late
Pickard, B. (2011). Progress in defining the biological causes of schizophrenia. paraphrenia. Schizophrenia Bulletin, 21 (3), 345–54.
Expert Reviews in Molecular Medicine, 28 (13), e25. Riecher-Rossler, A., et al. (2003). Late-onset schizophrenia versus paranoid
Pinkhasov, R.M., et al. (2010). ‘Are men shortchanged on health? Perspective psychoses: a valid diagnostic distinction? American Journal of Geriatric
on life expectancy, morbidity, and mortality in men and women in the Psychiatry, 11 (6), 595–604.
United States. International Journal of Clinical Practice, 64 (4), 465–74. Ritchie, C.W., et al. (2003). The impact upon extra-pyramidal side effects,
Pitkala, K.H., et al. (2011). Effects of socially stimulating group intervention clinical symptoms and quality of life of a switch from conventional to
on lonely, older people’s cognition: a randomized, controlled trial atypical antipsychotics (risperidone or olanzapine) in elderly patients
American Journal of Geriatric Psychiatry, 19 (7), 654–63. with schizophrenia. International Journal of Geriatric Psychiatry, 18 (5),
Post, F. (1966). Persistent persecutory states of the elderly. Pergamon Press, 432–40.
Oxford, New York. Ritchie, C.W., et al. (2006). A comparison of the efficacy and safety of
Poulet, E., et al. (2008). Maintenance treatment with transcranial magnetic olanzapine and risperidone in the treatment of elderly patients with
stimulation in a patient with late-onset schizophrenia. The American schizophrenia: an open study of six months duration. International
Journal of Psychiatry, 165 (4), 537–8. Journal of Geriatric Psychiatry, 21 (2), 171–9.
Robins, L.N. and Regier, D.A. (1991). Psychiatric disorders in America: Symonds, L.L., et al. (1997). Lack of clinically significant gross structural
the epidemiologic catchment area study. Free Press, New York; Collier abnormalities in MRIs of older patients with schizophrenia and related
Macmillan Canada; Maxwell Macmillan International. psychoses. J Neuropsychiatry Clin Neurosci, 9 (2), 251–8.
Rockwell, E., et al. (1994). Late-onset psychosis with somatic delusions. Taylor, M.A. (2001). Late-onset schizophrenia-like psychosis. The American
Psychosomatics, 35 (1), 66–72. Journal of Psychiatry, 158 (8), 1334–5.
Rodriguez-Ferrera, S., Vassilas, C.A., and Haque, S. (2004). Older people Tien, A.Y. (1991). Distributions of hallucinations in the population. Soc
with schizophrenia: a community study in a rural catchment area. Psychiatry and Psychiatric Epidemiology, 26 (6), 287–92.
International Journal of Geriatric Psychiatry, 19 (12), 1181–7. Tien, A.Y. and Eaton, W.W. (1992). Psychopathologic precursors and
Romero-Rubiales, F., Reeves, S., and Howard, R. (2004). Have risk factors for sociodemographic risk factors for the schizophrenia syndrome. Archives
very late-onset schizophrenia-like psychosis changed in the last 40 years? of General Psychiatry, 49 (1), 37–46.
International Journal of Geriatric Psychiatry, 19 (8), 806–7. Ting, C., et al. (2010). Differentiating the cognitive profile of schizophrenia
Roth, M. (1955). The natural history of mental disorder in old age. Journal of from that of Alzheimer disease and depression in late life. PLoS One, 5
Mental Science, 101 (423), 281–301. (4), e10151.
Routasalo, P.E., et al. (2009). Effects of psychosocial group rehabilitation on Tsuboi, T., Suzuki, T., and Uchida, H. (2011). A tipping point in drug
social functioning, loneliness and well-being of lonely, older people: dosing in late-life schizophrenia. Current Psychiatry Report, 13 (3),
randomized controlled trial. Journal of Advances in Nursing, 65 (2), 225–33.
297–305. Turkoz, I., et al. (2011). Paliperidone ER and oral risperidone in patients
Sable, J.A. and Jeste, D.V. (2002). Antipsychotic treatment for late-life with schizophrenia: a comparative database analysis. BMC Psychiatry,
schizophrenia. Current Psychiatry Report, 4 (4), 299–306. 11 (1), 21.
Sachdev, P. and Brodaty, H. (1999a). Quantitative study of signal Tuulio-Henriksson, A., et al. (2004). Age at onset and cognitive functioning
hyperintensities on T2-weighted magnetic resonance imaging in in schizophrenia. British Journal of Psychiatry, 185, 215–19.
late-onset schizophrenia. American Journal of Psychiatry, 156 (12), Twamley, E.W., et al. (2005). Work rehabilitation for middle-aged and older
1958–67. people with schizophrenia: a comparison of three approaches. Journal of
Sachdev, P.S. and Brodaty, H. (1999b). Mid-sagittal anatomy in late-onset Nervous Mental Disease, 193 (9), 596–601.
schizophrenia. Psychological Medicine, 29 (4), 963–70.
Twamley, E.W., et al. (2008). Supported employment for middle-aged
Sachdev, P., et al. (1999). Schizophrenia with onset after age 50 years. 2: and older people with schizophrenia. American Journal of Psychiatric
neurological, neuropsychological and MRI investigation. British Journal Rehabilitation, 11 (1), 76–89.
of Psychiatry, 175, 416–21.
Tzimos, A., et al. (2008). Safety and tolerability of oral paliperidone
Sachdev, P., et al. (2000). Hippocampus and amygdala volumes in elderly extended-release tablets in elderly patients with schizophrenia: a
schizophrenic patients as assessed by magnetic resonance imaging. double-blind, placebo-controlled study with six-month open-label
Psychiatry and Clinical Neurosciences, 54 (1), 105–12. extension. American Journal of Geriatric Psychiatry, 16 (1), 31–43.
Sajatovic, M. and Chen, P. (2011). Geriatric bipolar disorder. Psychiatric
Vahia, I.V., et al. (2010). Is late-onset schizophrenia a subtype of schizophrenia?
Clinics of North America, 34 (2), 319–33.
Acta Psychiatrica Scandinavica, 122 (5), 414–26.
Sajatovic, M., et al. (2008). Adjunct extended-release valproate semisodium
van der Heijden, F.M., Zeebregts, C.J., and Reijnen, M.M. (2010). Does
in late life schizophrenia. International Journal of Geriatric Psychiatry,
extracranial arterial pathology play a role in late-onset psychiatric
23 (2), 142–7.
disorders? Cognitive Behaviour and Neurology, 23 (3), 147–51.
Sato, T., et al. (2004). Psychopathology of early-onset versus late-onset
van Os, J., et al. (1995). Increasing age is a risk factor for psychosis in the
schizophrenia revisited: an observation of 473 neuroleptic-naive patients
elderly. Society of Psychiatry and Psychiatric Epidemiology, 30 (4),
before and after first-admission treatments. Schizophrenia Research, 67
161–4.
(2–3), 175–83.
Voisey, J., et al. (2011). A novel DRD2 single-nucleotide polymorphism
Schadlu, A.P., Schadlu, R., and Shepherd, J.B., 3rd (2009). Charles Bonnet
associated with schizophrenia predicts age of onset: HapMap
syndrome: a review. Current Opinions in Ophthalmology, 20 (3), 219–22.
Tag-single-nucleotide polymorphism analysis. Genetic Testing and
Schmid, L.A., Lasser, M.M., and Schroder, J. (2011). Symptoms and cognition
Molecular Biomarkers, 23, 23.
in geriatric schizophrenia. Fortschritte der Neurologie-Psychiatrie, 79 (5),
267–76. Wake, R., et al. (2011). The finding of brain perfusion by 99mTc-ECD single
photon emission computed tomography (SPECT) in patients with the
Schmidt, N.B., et al. (1995). The Schema Questionnaire: investigation of
late-onset schizophrenia. Asian Journal of Psychiatry, 4, S76.
psychometric properties and the hierarchical structure of a measure of
maladaptive schemas. Cognitive Therapy and Research, 19 (3), 295. Wiles, N.J., et al. (2006). Self-reported psychotic symptoms in the general
population: results from the longitudinal study of the British National
Schneider, L.S., Dagerman, KK.S., and Insel, P. (2005). Risk of death with
Psychiatric Morbidity Survey. British Journal of Psychiatry, 188,
atypical antipsychotic drug treatment for dementia. Journal of the
519–26.
American Medical Association, 294 (15), 1934–43.
Scott, J., et al. (2010). Atypical (second generation) antipsychotic treatment Wilson, R.S., et al. (2007). Loneliness and risk of Alzheimer disease. Archives
response in very late-onset schizophrenia-like psychosis. International of General Psychiatry, 64 (2), 234–40.
Psychogeriatrics, 1, 1–7. Woo, B.K., et al. (2003). Unrecognized medical disorders in older psychiatric
Seeman, M.V. and Lang, M. (1990). The role of estrogens in schizophrenia inpatients in a senior behavioral health unit in a university hospital.
gender differences. Schizophrenia Bulletin, 16 (2), 185–94. Journal of Geriatric Psychiatry and Neurology, 16 (2), 121–5.
Semple, S.J., et al. (1997). The social networks of older schizophrenia patients. Woods, S.P., et al. (2006). The California Verbal Learning Test—second
Clinical Research Center on Late Life Psychosis Research Group. edition: test-retest reliability, practice effects, and reliable change indices
International Psychogeriatrics, 9 (1), 81–94. for the standard and alternate forms. Archives of Clinical Neuropsychology,
Sim, K., et al. (2011). Adjunctive mood stabilizer treatment for hospitalized 21 (5), 413–20.
schizophrenia patients: Asia psychotropic prescription study (2001– World Health Organization (1992). The ICD-10 classification of mental and
2008). International Journal of Neuropsychopharmacology, 14 (9), behavioural disorders: clinical descriptions and diagnostic guidelines.
1157–64. WHO, Geneva.
Singh, D. and O’Connor, D.W. (2009). Efficacy and safety of risperidone Wynn Owen, P.A. and Castle, D.J. (1999). Late-onset schizophrenia:
long-acting injection in elderly people with schizophrenia. Journal of epidemiology, diagnosis, management and outcomes. Drugs and Aging,
Clinical Interventions in Aging, 4, 351–5. 15 (2), 81–9.
Yasuda, M. and Kato, S. (2009). Clinical psychopathological research on Zdanys, K. and Tampi, R.R. (2008). A systematic review of off-label uses of
late-onset schizophrenia—mainly patients with schizophrenia from a memantine for psychiatric disorders. Progress in Neuropsychopharmacology
hospital psychiatric ward. Seishin Shinkeigaku Zasshi, 111 (3), 250–71. and Biological Psychiatry, 32 (6), 1362–74.
Zakzanis, K.K., et al. (2003). Neuropsychological differentiation of Zimbardo, P.G., Andersen, S.M., and Kabat, L.G. (1981). Induced
late-onset schizophrenia and dementia of the Alzheimer’s type. Applied hearing deficit generates experimental paranoia. Science, 212 (4502),
Neuropsychology, 10 (2), 105–14. 1529–31.
CHAPTER 47
Personal experience
of lifelong illness
Anonymous in collaboration with Sue Green
My problems started when I was about 13, more than 60 years ago. shower room. They had a spray from the ceiling and sprayed you
I was at the village school and very successful at sport, but even as down about four at a time. There was no privacy, none at all.
this success came my way, wicked and disturbed thoughts started After we had washed we were put in a short night-dress again,
to come into my head. If I heard about a crime on the news I would even during the day. Mostly we only came out of the cells to eat
tell my mother I thought I had done it; if anyone got an illness, I and wash. In the summer they used to take us for walks. There
believed I had it; if anyone committed a murder, I’d done it; if any- were more staff than there were patients, but we were still in the old
one had a road accident, I’d caused it. It happened every day. After gowns, tied up. We weren’t allowed proper clothes or underwear or
a bit it got to the stage where I used to go up the road to the police anything like that.
station and they informed my GP. After a little while I got used to it and it was not quite so bad, but
One day I went to see my doctor because I was worried I was I still did not know why it was happening. I still did not understand
losing my hair. He made me an appointment to go to a derma- why I was there. The nurses were in their 30s, perhaps with sons or
tologist. I remember the day so clearly. It was a Friday afternoon daughters of my age. They seemed to feel sorry for me and made
and I was cycling to the hospital. There were two young children me as comfortable as possible. They were kind and I was never
walking along the road and when I saw them I started to worry ill-treated but I didn’t know what was happening; I was lost. I can’t
that I had harmed them. I kept looking round and I kept pedal- explain how severe it was. As days went on, some of the nurses used
ling. I could not get to the hospital quickly enough. When I got to bring in books that their children had had. However, I was still
there I told the dermatologist about these worries, not about my spending most of the day on my own in the padded cell. I had had
hair. That had gone out of the window. Before I knew anything he no treatment and no one explained. I think at first they thought I
had security there. They put me into an ambulance, took my bike had really committed an offence, but once you were there you were
and everything, and I was whisked off to the psychiatric hospital. forgotten about.
No one explained or went into details. I would have been about 15 Things went on like this until a new doctor arrived and then
years old. things changed a lot. This doctor wanted to know about us. He went
I was taken up old concrete steps onto a wooden floor. There were from one end of the ward to the other until he came to me and
all these iron cages which I learned were padded cells. They were talked to me. The nurses said, ‘The new doctor will help you,’ and
lined with the heating in the roof. I had to take all my clothes off they lifted my spirits a bit with that. One day shortly afterwards,
and was stood there naked. I did not say anything. The nurses came two or three administrators came. They had got me a case and a
and helped me put on a big white sheet with sleeves cut in it. I sat in load of new clothes. I got dressed normally and moved to a new
a cell with just a chair and table. I did not want to tell my parents. ward, but when I got there I was put in pyjamas and into bed. I was
I thought I would soon come home but really I did not know what back to square one. I thought, ‘I’ve got to do something,’ so when
to think. my doctor came round with the nurses I jumped out of bed and
They got in touch with my parents and found out where I lived. asked him if he could help me. He started me on sodium Amytal,
One sister told me my bike was safe. I was pleased about that, but the first time I had had any treatment for my problems.
I asked ‘What am I doing here?’ She said, ‘Well, it is because of the After that I was let out onto the hospital grounds at weekends.
circumstances you came here. We have to sort that out before we do My mother could visit and she brought me cakes. My father never
anything.’ But it went on for months, for years. In the end I was in did come to see me.
the cell for about 18 months. I was stunned. I never had ECT but I had insulin treatment. My doctor went to
In the evenings they would come and push two cages together. Japan and America to learn this treatment, a wonderful treatment
The woman in the next cell used to scream her head off. She was a they said. They had eight men for the first session and said there
little older that me. It was the first time I’d seen a naked lady. We was a great improvement in every patient. One of the nurses, my
had these trolleys with red blankets on; they would line you up, mentor, wanted me to have it, but there was no room in that group
about 12 of you, and come round with a bottle. They would make for me. I went in the next group with the women.
you drink it. It was this thick black horrible stuff, that knocked you My doctor explained what was happening and what I would see.
right out. In the morning we went two by two and were put into a He told me not to be afraid. He introduced me to one of the sisters
and told me that if anything worried me I should see her. I went in I have had to deal with stigma through my life a lot. I have, in
the dormitory and they put me in a corner with screens around. my life, taken one overdose. It was because I had the flu. I couldn’t
The rest of the patients having treatment were women so they did get rid of it. I took four tablets and then another four. Then my
not allow any male staff on that ward. We were kept on our own wife came home and found me on the floor. The next thing I knew
as a group. I felt comfortable because I could talk to the sister. We I was in hospital. My wife followed the ambulance and when I
had to have 60 comas to complete the course. I was there 5 months got there I gradually felt ten times better because of all the tablets
or more. I had had my 60 comas so I thought I would come out I had taken. The doctor asked about the leucotomy. There was a
of treatment, but no, the course did not finish until the last one in nurse there who pointed to my wife and asked, ‘Are you related to
the group had 60 comas. I carried on and they reduced the dose of him?’ She said ‘Yes, he’s my husband’. The nurse said, ‘You mar-
insulin so I had shorter comas. I saw plenty of nudity and they saw ried a lunatic. I’d never do that,’ and walked out. My wife did not
me. I can’t believe it even to this day, although it did not hurt me. say anything but I felt sick as a pig. I told my doctors the first
When you read in the papers about mixed wards I think to myself, chance I got and they wanted my wife to complain, but she had
‘Well, little do they know what I’ve been through’. had enough, she didn’t want to. That was a hard one. I get upset
I also had a leucotomy. I was about 18. It was done before I went when I hear anything about lunatic asylums. I wish I didn’t know
on insulin. There were two other patients having the treatment. They anything about it.
shaved the top of your head and made a small cut. One of the other When I met my wife my mother told me I must tell her every-
two unfortunately died; the other later committed suicide. I asked thing. She knew I was a patient. My doctor saw me and my wife
why I was alright and was told they did not cut so deep with me. That together. They said it was better to try. If they had not done that I
was why I was alright, but it didn’t do me any good. I don’t think so. might never have got married or had children. I was a patient from
Then I gradually did get better. I formed friendships, made friends the time I was 15 to the time I was nearly 50. I worked at the hospi-
with one or two of the nurses. I started working in light industries. tal and cycled home at the weekend.
I had already had some work experience. I had left school at 14 so Now I have grandchildren. Sometimes I have cycled through the
I had already worked about a year and a half. They had a machine villages and been 20 miles away and started worrying I have left my
that was broken down and they could not afford to have it repaired. pills beside my bed where the children could get them. I’ve worried
I said, ‘Switch off the electricity and I will do it,’ and that way I got a they would come to my house and go upstairs. I have turned round
job. I got really into it. I was still a patient but I got 5 shillings a week. and cycled 20 miles back to check. I never go out much because I
Quite a lot of patients were doing jobs like that in the hospital. worry about something at home.
One day I was working in light industry and the bloke in charge When I left the hospital, the community there was breaking up
asked me to take a letter upstairs. ‘Just take it up,’ he said. If I had and there was no more work. Someone helped by finding other
known what it said I doubt I would have had the nerve to do it. work for me. I had already learnt to drive. I had stopped a van from
There were others there including the head engineer. They shook running away once when I was a patient. I shunted it to the porter’s
my hand and offered me a job on the staff. I was 21. I was intro- lodge and he said, ‘Thank you so much’. He could have got in trou-
duced to the other lads. I was still a patient living there all the time ble. He said, ‘I can get you to pass your test’. I went driving with him
but now working on the staff. I stayed there until I was 49 when this and I passed first time. That meant when I left hospital I worked
part of the hospital closed. on various other NHS properties. I also worked on people’s houses
As the weeks and months went by, the disturbed thoughts I was doing carpentry and decorating.
having got less but they were still there. I was doing some roofing How have I coped over the years with my illness? Sodium Amytal
and I was up on the roof and there were a couple of old tramps near deadens the thoughts for a period, and if I am active or distracted,
there. They worried me to death for some money and I gave them that period is longer. Work, TV, cricket, and cycling all helped to
some. Then I thought, ‘Did I do them any harm?’ I have had that deal with the thoughts. I was keen on sport and I loved cycling. I
worry ever since, for years. These thoughts were there all the time cycled home through the villages. That helped me physically to be a
but they got easier to cope with. very good cyclist. I have got medals and cups for cycling
My doctor pushed me to the limit and encouraged me to go out Once the building jobs dried up I was offered a job in an antique
with young ladies: ‘Enjoy life, you are only on this earth once.’ I shop just down the road. That was a lovely job, I enjoyed it. As soon
got friendly with some of the nurses. I was really looking forward as that finished, then my illness, my thoughts, came back. I haven’t
to the hospital ball with a nurse one time. We went and got me done any cycling in the last year because my knees hurt. I feel very
measured for a suit. She bought the tickets and everything. The ball dependent on sodium Amytal now.
was in the main hall at the hospital. The men on the door were my However, talking to the right person still helps. It is very impor-
mates. We played snooker and all that. They said, sorry, I could not tant to me that I talk to professional people and this is the best cure
go in because I was a patient. The nurse I was with went loopy. She I have found. I would never be able to see my doctors and nurses
said, ‘Come on,’ and we went somewhere else to a Greek restau- every few days, but I used to have a little book and would write
rant. I remember standing back looking up at the dark blue sky with down what they said to me, for example, ‘My GP is not going to
all the stars and I thought, ‘Yes, you are going to come across this strike me off her list’. If I had a worry or a fear I would look it up in
throughout your life’. I made the best of it. these books.
CHAPTER 48
Severe and enduring
mental illness
Catherine Hatfield and Tom Dening
This chapter addresses the needs of people growing old with et al., 1993; Trieman et al., 1999; Leff and Trieman, 2000). Even
long-term, severe mental illness. These comprise mainly patients among those initially considered unsuitable for community place-
with schizophrenia and other psychotic mental illness, but also a ments, nearly half were able to be placed within the next 5 years
smaller number with bipolar affective disorder and other diagnoses (Trieman and Leff, 2002).
including obsessive compulsive disorder, personality disorders, and
substance misuse. Much of the clinical research concerns patients
with schizophrenia, but the service considerations also apply to The Community Care Era
those with other diagnoses. In 1990, the NHS and Community Care Act made local authorities
responsible for providing appropriate community mental health
Historical Context services. This formalized the shift in attitudes over the preceding
three decades. It was envisaged that people with chronic mental
Graduates from institutional care illness would live independently with support from community
In the early part of the twentieth century, people with long-term mental health teams and social care services and access mainstream
functional mental illness in the UK were cared for in large county education, leisure, and employment facilities. Unfortunately, not
asylums. At the peak of provision in 1955 there were 152,000 inpa- all of the expected benefits of community care were realized. The
tient psychiatric beds. From the 1960s onwards this model of care promise of integration into the community was not fulfilled and
was questioned, as the disadvantages of long-term institutionaliza- many former patients were merely ‘transinstitutionalized’ rather
tion became clear and bed numbers were gradually reduced, with a than deinstitutionalized, as they were moved into nursing homes
shift of focus to care in the community. By 1986 the inpatient popu- that differed very little from the wards from which they had come.
lation had halved and the old asylums began to be closed down Services were criticized for being fragmented and failing to meet
(Jolley et al., 2004). need (Killaspy, 2006). Between 1954 and 1996, 110,000 psychiatric
Older people with chronic psychosis were among the last to ben- hospital beds were closed and only 13,000 community placements
efit from this new approach. In 1975 just over half of the 41,864 replaced them (Lelliott et al., 1996). Some people with mental illness
people resident in mental hospitals in England for more than 5 have undoubtedly been diverted to the prison or homeless popula-
years were aged 65 and over (Department of Health and Social tions (Priebe et al., 2005). Crane (1998) found that two-thirds of a
Security, 1978). In addition to the impairments associated with sample of 219 homeless people aged over 55 from four UK cities
their illness they were further handicapped by their long hospital suffered from a mental illness and that for at least 18% of them it
stay, which had left them with no independent living skills. Clifford was a factor in their becoming homeless.
et al. (1991), surveying the population of five hospitals destined for Since the demise of the large institutions it has become harder
closure, found that of those with functional diagnoses hospitalized to identify people with enduring mental illness, as they are scat-
for over a year, the mean duration of admission was 24.5 years and tered between different living situations and cared for by different
their mean age 64.5. services (if indeed they are receiving any service at all). This has led
Nonetheless, studies, in particular from the Team for the to them being all but invisible in terms of research and policy. The
Assessment of Psychiatric Services (TAPS) group studying a cohort National Service Framework (NSF) for Mental Health (Department
of people discharged from Friern Barnet and Claybury hospitals, of Health (DH), 1999) applied only to working age adults. The
have found some positive outcomes. Only 71 out of 130 were alive 3 separate NSF for Older People (DH, 2001) included mental health
years later (not significantly different from those remaining in hos- alongside physical health conditions, rather than as a main focus.
pital) and, of those, half had returned to hospital. However, those The NICE guidance on interventions and management of schizo-
remaining in the community had more social contacts and were phrenia (National Collaborating Centre for Mental Health, 2010)
more satisfied with their lives than those in hospital (Anderson refers only to working age adults.
Epidemiology Andreasen et al. (2005) proposed consensus criteria for defin-

ing remission of the absence or low intensity of eight core positive
The adult psychiatric morbidity surveys in England found 1-year and negative symptoms sustained for a period of at least 6 months.
prevalence rates for probable nonorganic psychosis in those aged They distinguished this from recovery which they conceptualized
65–74 of 0.4% in 2000 and 0.1% in 2007 (this compares to 0.5% as a broader concept encompassing cognitive, functional, and psy-
for those aged 16–74) (McManus et al., 2009). Similarly in the US, chosocial criteria. Recent studies employing these remission cri-
National Comorbidity Survey Replication also reported a lifetime teria in older patients have shown approximately 50% remission
prevalence rate of 0.3% which fell to 0.1% among people over the rates in two outpatient populations in the US (Bankole et al., 2008;
age of 60 (Kessler et al., 2005). These figures are likely to represent Leung et al., 2008) but only 29.4% in a catchment area sample in
an underestimate of the prevalence of older people with psychotic the Netherlands containing both community-dwelling and institu-
mental illness, both because of the exclusion of those living in insti- tionalized subjects (Meesters et al., 2011). Earlier studies employ-
tutional care and because even those living at home are unlikely to ing stricter remission criteria found even smaller rates of 7–8%
respond to surveys. (Marneros et al., 1992; Auslander and Jeste, 2004).
Smaller community surveys have given prevalence rates of A different approach to outcome is to ask people with schizo-
0.1–0.5%, including Copeland et al. (1998) in Liverpool, Castle phrenia about their subjective quality of life (QOL). Although there
and Murray (1993) in Camberwell, and McNulty et al. (2003) in is no universally accepted definition of QOL the WHO defines it as
Lanarkshire. Regional prevalence figures will be affected by the ‘an individual’s perception of their position in life in the context of
historical structure of services, in particular the presence of large the culture and value systems in which they live and in relation to
asylums in some areas. Rodriguez-Ferrera et al. (2004) found their goals, expectations and standards’ (Saxena and Orley, 1997).
approximately half the number of cases in rural Suffolk as McNulty Health-related quality of life (HR-QOL) refers to the impact of a
et al. in Lanarkshire and that a much higher proportion of subjects disease on a person’s wellbeing. Studies have found that older people
in their sample lived independently. They thought this was prob- with schizophrenia have lower HR-QOL than community controls
ably due to some patients being placed in the county asylum and (Patterson et al., 1996; Cohen et al., 2003). Depressive symptoms
not returning to their original area. and cognitive impairment have consistently been shown to predict
Prevalence rates in older people are much lower than the 1% life- lower QOL (Cohen et al., 2003; Mittal et al., 2006). However, recent
time prevalence usually quoted for schizophrenia overall. Some of studies have suggested that mental health-related QOL actually
the difference might be accounted for by incomplete ascertainment, improves with ageing despite reduced physical HR-QOL (Reine et
but excess mortality due to suicide and physical ill health earlier in al., 2005; Folsom et al., 2009). In a qualitative study, participants
life are also major determinants. described reduced impact of symptoms and better self-management
strategies with ageing (Shepherd et al., 2012).
Course and Outcome Overall, it appears that whilst complete and sustained sympto-
matic remission is uncommon, this does not preclude a positive
Schizophrenia functional outcome (Cohen et al., 2008; Jeste et al., 2011). Most
The outcome of schizophrenia in older age is a contentious issue. older people with schizophrenia remain symptomatic and impaired,
Of necessity, the long-term studies needed to address this question but the course of the disorder appears largely stable, with no real
encompass eras with differing diagnostic definitions, treatment evidence for either progressive decline or spectacular improvement
options, and social contexts, and this makes their findings difficult (Jeste et al., 2003a). There may be even greater cause for optimism
to interpret. In addition, patients included in long-term observa- in the future as the cohort of people now entering later life with
tional studies have tended to be those institutionalized patients schizophrenia did not have the benefit of early recognition and
with a poorer prognosis. effective treatment of their illness and were faced with even greater
Despite these limitations, long-term outcome studies of patients stigma and social exclusion than exists today. With reduced dura-
in the US and Europe followed up over 22–37 years have consistently tion of untreated psychosis (Perkins et al., 2005) and avoidance of
shown that approximately half to two-thirds of patients achieve sig- secondary disability, outcomes may improve still further for today’s
nificant improvement or recovery in the long term (Bleuler, 1972; cohorts.
Huber et al., 1975; Tsuang et al., 1979; Ciompi, 1980; Harding et al.,
1987; Marneros et al., 1992). The WHO International Study of Complications
Schizophrenia (ISS) provides particularly powerful evidence as it
included 1633 patients from 18 centres around the world followed Cognitive impairment
up over 15–25 years. Overall, 50% had a good outcome, increasing Cognitive impairment has long been recognized as a component
to 60% in some centres in the developing countries (Harrison et al., of the clinical syndrome of schizophrenia. Semantic memory and
2001). executive functioning seem to be the most prominently impaired
Rates of recovery are dependent to a very large degree on the (Heinrichs and Zakzanis, 1998).
criteria used, specifically whether these are entirely symptom based There is evidence that cognitive deficits are present at first epi-
or incorporate aspects of psychosocial functioning. The relation- sode and indeed may even predate the onset of illness (Jones et al.,
ship of symptoms to function is not simple, as patients may develop 1994), but whether these deficits remain static throughout life or
coping mechanisms that allow improved functioning despite ongo- continue to deteriorate over time remains controversial (Rund,
ing psychotic symptoms. There is good evidence to indicate that 1998; see Radhakrishnan et al. (2012) for a review). Cross-sectional
cognitive impairment is a better predictor of adaptive function than studies of mostly younger community-dwelling patients with schiz-
psychotic symptoms (Green, 1996; Velligan et al., 1997). ophrenia showed no deterioration in performance over time scales
CHAPTER 48 severe and enduring mental illness 625
of up to 10 years (Heaton et al., 1994; Eyler Zorilla et al., 2000), patients have been increasing over the past two decades, perhaps
although they were grossly impaired compared to normal subjects. as a result of deinstitutionalization (Westermeyer, 2006). Dual
In contrast, studies of older institutionalized patients have shown diagnosis is associated with worse outcomes in terms of mortal-
deterioration over time periods of as little as 2.5 years (Harvey et ity, frequent hospitalization, and criminal offending (Schmidt et al.,
al., 1999; Friedman et al., 2001). Studies of long-term institutional- 2011).
ized patients may be confounded by the effects of institutionaliza-
tion itself, use of long-term psychotropic medication, and selection Social exclusion
for those with a poorer prognosis. People with schizophrenia are less likely to marry and have chil-
Pharmacological strategies to ameliorate cognitive dysfunction dren than the general population. About two-fifths never marry
have not been successful to date. Antipsychotic medication has only and a further quarter are divorced. Two-fifths do not have children
a modest effect on cognitive impairment (Keefe et al., 2007), and (Cohen and Talavera, 2000). They are also less likely to have been
apparent benefits are likely to be due to improvement in psychosis in employment and their lives may be disrupted by substantial peri-
rather than in cognition per se. Indeed, antipsychotic drugs often ods of hospitalization. All of these factors lead to them having sub-
have sedative effects that would tend to impair rather than improve stantially reduced social networks. Of those contacts that they do
cognition. Furthermore, despite some early promise in open-label have, most are based on provision of sustenance or support (59%
trials, cholinesterase inhibitors have not been shown to be effective vs 15% for age-matched peers) and few are intimate, confiding rela-
in randomized placebo controlled trials (Akhondzadeh et al., 2008; tionships (44% vs 67%) (Cohen et al., 1996).
Lindenmayer and Khan, 2011). On measures of social skills they are significantly impaired in
accepting and initiating contact with others, group participation,
Psychiatric comorbidity and making friendships (Bartels, 1997a). Social withdrawal may be
Older people with functional mental illness are no more likely to a mechanism to cope with distressing psychotic symptoms, such
have the neuropathological brain abnormalities of Alzheimer’s disas persecutory delusional beliefs or thought interference, or may
ease than the general population (Baldessarini et al., 1997; Arnold result from reduced capacity to enjoy friendships due to negative
et al., 1998; Arnold, 2001). Even subjects with marked cognitive symptoms.
impairment do not show neuropathological abnormalities suf- Despite these deficits, older people with enduring mental illness
ficient for a diagnosis of Alzheimer’s disease (Rapp et al., 2010). do identify social needs as important. Auslander and Jeste (2002)
Given the high prevalence of dementia in older age groups, how- examined the self-reported needs of a sample of 72 people (41–80
ever, the conditions may co-occur. years). Around half of them lived independently and the remainder
It can be very difficult to distinguish the cognitive impairments (47%) lived in an assisted care facility. The study excluded people
associated with schizophrenia from a new onset of dementia. with dementia or with a comorbid alcohol or substance abuse prob-
Short-term memory loss, word-finding difficulties, and disorienta- lem. The main priorities they identified were around social rela-
tion are more associated with Alzheimer’s disease, whereas fron- tionships (developing and sustaining friendships), managing their
tal executive deficits are more associated with schizophrenia. The illness, and improvements in mood, physical health, and memory.
time-course of cognitive decline should also be helpful in making
the distinction. Physical health
Approximately 60% of people with schizophrenia suffer a major People with schizophrenia have a two- to three-fold increased mor-
depression during the course of their illness and around a quarter tality rate compared to the general population and on average their
experience depression following an acute schizophrenic episode life-expectancy is reduced by 10–15 years (Harris and Barraclough,
(Martin et al., 1985). Depression is associated with poorer outcomes, 1998; Laursen, 2011). Much of this excess mortality is accounted for
including poorer response to drug treatments, longer duration of by suicides in earlier life, but there is also an excess of deaths from
inpatient care, chronic course, and increased rates of relapse and natural causes (Brown et al., 2000). There is an increased risk of
suicide (Cohen et al., 1996; Siris, 2000; Jin et al., 2001). Diwan et death from cardiovascular and respiratory disease, infections, and
al. (2007) found that 32% of a community cohort of people over 55 type 2 diabetes (Jeste et al., 1996; Harris and Barraclough, 1998).
with schizophrenia were clinically depressed versus 11% of controls. This is likely to be multifactorial and related to lifestyle factors, side
Depression was significantly associated with physical illness, positive effects of medication, and reduced access to healthcare.
symptoms, reduced number of confidants, and a coping strategy of Patients with schizophrenia are more likely than the general pop-
medication use. Depression may be confused with negative symptoms ulation to have lifestyle risk factors for cardiovascular disease. They
but can be distinguished by the presence of low mood, early morning were found to be more likely to smoke, less likely to exercise, and
waking, and loss of appetite (Felmet et al., 2011). Although there is a more likely to have high fat, low fibre diets, even when the study
paucity of evidence, there is reason to believe that treatment is likely population was controlled for socioeconomic status (Brown et al.,
to be successful. Kasckow et al. (2001) demonstrated improvement in 1999; McCreadie et al., 2003; Osborn et al., 2008).
19 chronically hospitalized patients aged over 55 with schizophrenia Side effects of antipsychotic medication may also contribute to
and depressive symptoms treated using citalopram in an open-label increased morbidity and mortality. The cardiovascular side effects
trial. Zisook et al. (2009) also demonstrated improvements in mood, of antipsychotic medication include lengthening of the QT inter-
social functioning, and QOL in a double-blind randomized control- val which may contribute to arrhythmia and sudden cardiac death
led trial of citalopram in 198 patients with schizophrenia and subsyn- (Glassman and Bigger, 2001). Significant weight gain, dyslipidae-
dromal depression aged 40–75. mia, and new onset of type 2 diabetes are particularly associated
Substance misuse may become more of a problem as today’s with newer atypical antipsychotic medications. Conventional
cohorts age: rates of substance misuse amongst schizophrenic antipsychotic medications may significantly elevate prolactin levels,
which interferes with sexual functioning and reduces bone density, liver metabolism may also be reduced, leading to increased plasma
increasing the risk of osteoporosis. concentration of drugs such as benzodiazepines.
Patients with schizophrenia are less likely to receive treatment In clinical practice, one is often faced with the dilemma of weigh-
for physical health problems (Kilbourne et al., 2008; Vahia et al., ing up whether to reduce antipsychotic doses as patients become
2008). Data from the Patient Outcomes Research Team (PORT) older and frailer versus the risk of relapse of psychotic symptoms
study in the US (Dixon et al., 2000) showed that 30% of patients in someone who may have been stable for many years. There is lit-
reporting a physical health condition were not receiving treatment. tle research evidence to guide decision-making, but in one study,
Mitchell and Lord (2010) showed that following a myocardial inf- Harris et al. (1997) found that it was possible to reduce antipsy-
arction, patients with schizophrenia were less likely to be offered chotic medication doses by about 40% in a sample of patients aged
revascularization procedures and less likely to receive optimum over 45 without an increase in symptoms. It is often possible to
medication regimes. Patients may lack the motivation to attend achieve significant dose reductions over time. How rapidly this is
appointments, or cognitive deficits or delusional ideas may make done depends on various factors, such as the potential severity of
it hard for them to comply with treatment plans, but is also likely relapses, and also on the wishes of the patient, as some people will
that negative attitudes of care providers towards those with chronic be more cautious about going on to a lower dose than will oth-
mental illness play a role. Studies have shown that GPs were less ers. It is also often possible to replace depot injections with oral
likely to offer screening for cardiovascular risk factors to people medication, especially if the person with schizophrenia is starting
with schizophrenia than to those with asthma or another diagnosis to receive personal care. This obviously makes it easier to assess the
(Roberts et al., 2007). effects of altered dosages.
Older people are more susceptible to side effects from medication.
Management TD is a particular problem with typical antipsychotic medications
such as haloperidol and the risk is five to six times higher in older
Medication patients (Jeste, 2004). Studies of older hospitalized patients have
There is a very poor evidence base to guide pharmacological shown very high rates of TD of 60–90% (Quinn et al., 2001). TD
treatment in older people due to their exclusion from large-scale is often irreversible even after stopping medication, and although
randomized controlled trials (RCTs). Only two RCTs have been patients are usually not aware of the movements, they are obvious
conducted specifically in older patients. One trial comparing olan- to others and can be stigmatizing. In severe cases, TD can cause
zapine with haloperidol in 117 patients aged over 60 found that difficulties with eating and swallowing.
olanzapine was both more efficacious and better tolerated than Other common side effects are those related to anticholinergic
haloperidol (Kennedy et al., 2003). A further RCT comparing olan- actions, such as delirium, constipation, and urinary retention. All
zapine and risperidone in 175 older patients with schizophrenia of these can be more severe in older people due to the interaction
found no significant differences between the two treatments (Jeste with physiological ageing. For example, prostatic hypertrophy in
et al., 2003b). Older patients have been included in other trials, but older men puts them at higher risk of urinary retention. In some
the number of older subjects is either not specified or too small for cases, different side effects combine to increase the risk of an
meaningful subgroup comparisons. adverse outcome, e.g. antipsychotic medications cause both ortho-
Clozapine is the only antipsychotic medication shown to have static hypotension, which can lead to falls, and reduced bone den-
greater efficacy in treatment-resistant schizophrenia (Kane et al., sity, which increases the risk of fracture. Older people are also likely
1988) and has negligible risks of tardive dyskinesia (TD), but its use to be taking multiple physical health medications and this increases
is tightly controlled due to rare but serious adverse effects includ- the potential for drug interactions.
ing agranulocytosis, cardiomyopathy, and seizures. More common
adverse effects include sedation and postural hypotension. There Nonpharmacological treatments
is very little evidence to guide its use in older patients and this is There is now considerable evidence of benefit from cognitive
likely to become an increasingly important issue, as patients who behavioural therapy (CBT) in younger patients with schizophre-
were prescribed clozapine for treatment-resistant schizophrenia nia (Tai and Turkington, 2009). It is recommended by NICE for
as young adults age. One RCT comparing it with chlorpromazine improving persistent psychotic symptoms and increasing insight
in patients over 55 found no differences between the two drugs and treatment adherence and can be delivered in 1:1 sessions with
(Howanitz et al., 1999). Barak et al. (1999) reviewed the literature a psychologist or in a group. Other psychological therapies such
and identified 139 reports of patients aged over 65 treated with as social skills training and cognitive remediation have also been
clozapine for a psychiatric indication. The mean dose used was 135 shown to be valuable in younger adults. Skills training is usually
mg. Clozapine was discontinued in 25% due to side effects, non- done in a group and may include learning about verbal and nonver-
compliance, or inefficacy and the rate of leucopenia was 5%. In the bal communication and rules of social interaction using role play
absence of any evidence-based guidance, it seems reasonable to and exercises.
make attempts at gradual reduction of clozapine dosage in older Recent studies have evaluated the effectiveness of combined
patients with psychosis. interventions including elements of these therapies. Patterson et al.
Older patients would be expected to require lower doses of med- (2006) developed functional adaptive skills training (FAST) and
ication due to physiological changes associated with ageing. For evaluated it in an RCT in 240 patients aged over 40. They showed
example, there is an increase in the proportion of body fat with significant improvements in social skills and activities of daily living
ageing which increases the volume of distribution of lipid-soluble and reduced use of emergency services in the intervention group.
drugs and prolongs their action; glomerular filtration rate decreases, Granholm and colleagues (2005, 2007) also used a group cogni-
reducing clearance of renally excreted drugs such as lithium; and tive behavioural and social skills training intervention (CBSST) in
76 older outpatients with schizophrenia or schizoaffective disor- Psychiatric services

der, and showed significant improvements in social function and There has been much debate about who should provide services
insight which were sustained over a year despite no improvement to older people with functional mental illness, and practice var-
in symptoms. ies across the country. Patients may continue to be provided for
In summary, the evidence suggests that older people are able to by general adult or rehabilitation services indefinitely or they may
benefit from psychological therapies, but sadly in clinical practice automatically ‘graduate’ to the older people’s service on attaining
these interventions are not yet widely available to older people. the age of 65. In addition, some may have been discharged from or
Accommodation lost contact with secondary care and instead they receive their care
from their GP and social care services.
From the 1970s and 1980s onwards, with the decline of long-stay People growing older with functional mental illnesses may
wards and psychiatric hospitals, other types of provision have arisen perceive a great injustice in being discharged from services they
to fill the void. Hostels and care homes run by private sector and have previously been able to access due to an arbitrary age cut-off
voluntary organizations vary from those with 24-h staffing by qual- of 65 years. Older people’s mental health services are often more
ified staff to those with daytime cover only from a smaller number poorly resourced than rehabilitation services for younger adults
of staff without specialist qualifications. Group homes are unstaffed and focused around dementia rather than functional illness. Access
facilities where small groups of residents with chronic mental ill- to services such as crisis resolution and home treatment teams
ness live together supported by visits from staff (Macpherson et al., for acute relapses or assertive outreach teams may be restricted to
2004). A more recent model of care is where each individual has working age adults. On the other hand, older people’s services may
his or her own flat or bedsit but within a unit where there is a staff be better placed to meet the needs of those with physical frailty or
member present and often some communal facilities. There is evi- cognitive impairment, whatever their age.
dence that this form of support is preferred by patients (Tanzman, In the UK, the Royal College of Psychiatrists (RCPsych, 2002)
1993). published a joint report from the faculties of Old Age, General and
It should not be forgotten that many people with enduring men- Community and Rehabilitation Psychiatry which recommended
tal illness live with their families. Tsai et al. (2011), examining that an individual’s health and social care needs should be regularly
data from the large Clinical Antipsychotic Trials of Intervention reviewed, that transfer should be based on need rather than chron-
Effectiveness (CATIE) study of outpatients aged 18–65 with schizo- ological age, and that there should be agreed protocols in place.
phrenia in the US, found 46% living with family members (vs 18% When this report was updated in 2009 (RCPsych, 2009), two-thirds
living independently, 17% in an institution, and 14% with no stable of services that responded to a survey of policy implementation had
housing). There are no corresponding national data for older peo- transition protocols in place, but there was very little audit or data
ple. These percentages will vary depending on cultural expectations collection to demonstrate effectiveness (Bawn et al., 2007).
and on provision of alternative resources. For instance, 60–80% of Services for younger people with enduring mental illness are
patients from ethnic minorities live with their families in the US now heavily influenced by the ‘Recovery Model’. This originated as
(Lefley, 1987). a service user-led movement in the US in the 1990s but has now
As people with enduring mental illness age, physical disabili- become a fundamental influence on mainstream mental health
ties or loss of support networks may mean that their needs can policy on both sides of the Atlantic (Silverstein and Bellack, 2008).
no longer be met in their previous setting. Andrews et al. (2009) It has aimed to redefine recovery away from a purely medical focus
studied admissions to nursing homes over a 10-year period in on remission of symptoms and towards the attainment of personal
New Hampshire, US. They found a much younger age of entry to goals that provide hope, purpose, and meaning in life beyond men-
a nursing home amongst those with a diagnosis of schizophrenia tal illness (Anthony, 1993). Recovery-orientated services aim to
(65 vs 80 for those with no mental illness). The rates of admission promote shared decision-making, choice, and autonomy for serv-
started to diverge significantly between the ages of 40 and 65 where ice users. Social inclusion in mainstream employment and leisure
nursing home admission risk was 3.9 times greater than for peers opportunities rather than provision of segregated services is encour-
with no mental illness. The reasons for this are not known, but one aged and peer support from those with lived experience of mental
could speculate that loss or burnout of the primary carer may have health problems is facilitated. Whilst these concepts are starting to
contributed. Bartels et al. (1997b) found that cognition, function- be widely implemented in services for working age adults, there has
ing, and behaviour were important predictors of nursing home thus far been little uptake in older people’s services. Nonetheless,
as opposed to community residence, with cognitive impairment some of the enshrined ideas around acceptance and hope seem
especially important as it was also a predictor of a person’s level of applicable to the older age group too.
functioning. Auslander et al. (2001) also found that among older
patients with schizophrenia (age range 40–97), residence in assisted
living facilities was associated with never having been married, the Mental Capacity
presence of cognitive impairment, and poorer quality of wellbeing. Lack of insight is very common and can lead to difficult decisions
Because their numbers are relatively small, those with functional about whether a patient has capacity to refuse care or treatment
illness are often placed in settings designed mainly for people with and whether or when to intervene in a person’s best interests. The
severe dementia that may be poorly placed to meet their needs. Mental Capacity Act provides a clear framework for making deci-
Opportunities for social interaction and participation in particular sions, but in the case of older people with functional mental ill-
are likely to be severely limited and freedoms curtailed. These prob- ness, the decision often hinges on the complex issue of their ability
lems might be overcome by better staff education and individual to weigh evidence and make a decision rather than on simpler
care planning and risk assessment. issues of understanding or retention of information. Impaired
decision-making capacity is not by any means a universal finding. Arnold, S.E. (2001). Contributions of neuropathology to understanding
In one recent study, 69–89% of subjects with psychotic mental ill- schizophrenia in late life. Harvard Review of Psychiatry, 9, 69–76.
ness were able to score in the same range as control or medically Arnold, S.E., et al. (1998). Absence of neurodegeneration and neural injury
in the cerebral cortex in a sample of elderly patients with schizophrenia.
ill subjects on a test of decisional capacity (Candilis et al., 2008).
Archives of General Psychiatry, 55, 225–32.
Cognitive impairment is the best predictor of impaired capacity,
Auslander, L.A. and Jeste, D.V. (2002). Perceptions of problems and needs for
but understanding can be improved by repetition of information service among middle-aged and elderly outpatients with schizophrenia
(Moser et al., 2006; Palmer and Jeste, 2006). and related psychotic disorders. Community Mental Health Journal, 38,
391–401.
Future Challenges Auslander, L.A. and Jeste, D.V. (2004). Sustained remission of schizophrenia
among community-dwelling older outpatients. American Journal of
As the population ages, the numbers of people surviving into old Psychiatry, 161, 1490–3.
age with enduring mental illness will increase. However, the num- Auslander, L.A., et al. (2001). A comparison of community-dwelling
bers will always be relatively small in comparison to those with older schizophrenia patients by residential status. Acta Psychiatrica
dementia, who form the overwhelming majority of patients of old Scandinavica, 103, 380–6.
age psychiatry services. Baldessarini, R.J., et al. (1997). Meta-analysis of postmortem studies of
There are specific areas of expertise in working with older patients Alzheimer’s disease-like neuropathology in schizophrenia. American
Journal of Psychiatry, 154, 861–3.
with long-term mental illnesses, e.g. around prescribing, medical
Bankole, A., et al. (2008). Symptomatic remission in a multiracial urban
support, housing and accommodation, and productive activities. population of older adults with schizophrenia. American Journal of
Staff working within older people’s mental health services do need Geriatric Psychiatry, 16, 966–73.
to develop this kind of experience and expertise and to have access Barak, Y., et al. (1999). Clozapine in elderly psychiatric patients: tolerability,
to appropriate training and resources. This needs to be recognized safety and efficacy. Comprehensive Psychiatry, 40, 320–5.
in planning and providing services, with suitable arrangements for Bartels, S. J., Mueser, K. T., and Miles, K. M. (1997a). Functional impairments
staff support and supervision. in elderly patients with schizophrenia and major affective disorders living
The lack of research in this age group means that there is a dearth in the community: social skills, living skills and behaviour problems.
of good quality evidence to guide practice. More research in par- Behaviour Therapy, 28, 43–63.
Bartels, S. J., Mueser, K. T., and Miles, K. M. (1997b). A comparative study
ticular into medication, psychological therapies and services is
of elderly patients with schizophrenia and bipolar disorder in nursing
urgently needed. homes and the community. Schizophrenia Research, 27, 181–90.
Bawn, S., et al. (2007). Transitions: graduating between general and old age
Conclusion psychiatry services in England and Wales. Mental Health Review, 12, 21–6.
Bleuler M. (1972). Die schizophrenen Geistesstörungen im Lichte langjahriger
Older people with functional mental illness represent a vulnerable, Kranken- und Familiengeschichten. Georg Thieme, Stuttgart. [Translated
marginalized group in society. Over the years, they are likely to have by Clemens S.M. (1978) as The schizophrenic disorders: long term patient
suffered invasive treatments, a lack of autonomy, and discrimina- and family studies. Yale University Press, New Haven, Connecticut.
tory attitudes. This history can increase their mistrust of healthcare Brown, S., et al. (1999). The unhealthy lifestyle of people with schizophrenia.
professionals and they will often have no family members or friends Psychological Medicine, 29, 697–701.
to act as advocates or supporters. They may be the victims of a dou- Brown, S., Inskip, H., and Barraclough, B. (2000). Causes of the excess
ble discrimination on the grounds of age and mental illness and are mortality of schizophrenia. British Journal of Psychiatry, 177, 212–17.
likely to receive poor quality care despite having high needs. Candilis, P.J., et al. (2008). A direct comparison of research decision-making
capacity: schizophrenia/ schizoaffective, medically ill, and non-ill
Older people with enduring mental illness need access to good
subjects. Schizophrenia Research, 99, 350–8.
physical and mental healthcare, including screening and health
Castle, D.J. and Murray, R.M. (1993). The epidemiology of late-onset
promotion, given their high risk of physical ill health, psychiatric schizophrenia. Schizophrenia Bulletin, 19, 691–700.
comorbidity, and drug side effects. Despite these challenges, posi- Ciompi, L. (1980). Catamnestic long term study on the course of life and
tive outcomes are possible and older people should not be excluded aging of schizophrenics. Schizophrenia Bulletin, 6, 606–18.
from provision of innovative service models and psychosocial Clifford, P., et al. (1991). Planning for community care: long stay populations
interventions. of hospitals scheduled for rundown or closure. British Journal of
References Cohen C.I., et al. (2003). Assessing quality of life in older persons with
Akhondzadeh, S., et al. (2008). A 12 week double blind placebo controlled schizophrenia. American Journal of Geriatric Psychiatry, 11, 658–66.
trial of donepezil adjunctive treatment to risperidone in chronic Cohen, C.I. and Talavera, N. (2000). Functional impairment in older
and stable schizophrenia. Progress in Neuropsychopharmacology and schizophrenic persons: towards a conceptual model. American Journal of
Biological Psychiatry, 32, 1810–15. Geriatric Psychiatry, 8, 237–44.
Anderson, J., et al. (1993). The TAPS project. 13: clinical and social outcomes Cohen, C. I., Talavera, N., and Hartung, R. (1996). Depression among aging
of long-stay psychiatric patients after one year in the community. British persons with schizophrenia who live in the community. Psychiatric
Journal of Psychiatry, 162 (Suppl 19), 45–56. Services, 47, 601–7.
Andreasen, N.C., et al. (2005). Remission in schizophrenia: proposed criteria Cohen, C.I., et al. (2008). Schizophrenia in later life: clinical symptoms and
and rationale for consensus. American Journal of Psychiatry, 162, 441–9. social well-being. Psychiatric Services, 59, 232–4.
Andrews, A.O., et al. (2009). Increased risk of nursing home admission Copeland, J.R et al. (1998). Schizophrenia and delusional disorder in older
among middle aged and older adults with schizophrenia. American age: community prevalence, incidence, comorbidity, and outcome.
Journal of Geriatric Psychiatry, 17, 697–705. Schizophrenia Bulletin, 24, 153–61.
Anthony, W.A. (1993). Recovery from mental illness: the guiding vision of Crane, M. (1998). The associations between mental illness and homelessness
the mental health service system in the 1990s. Psychosocial Rehabilitation among older people: an exploratory study. Aging and Mental Health, 2,
Journal, 16, 11–23. 171–80.
Department of Health (1999). National Service Framework for Mental Health. Jeste, D.V., et al. (2003a). Aging and outcome in schizophrenia. Acta
Department of Health, London. Psychiatrica Scandinavica, 117, 336–43.
Department of Health (2001). National Service Framework for Older People. Jeste, D.V., et al. (2003b). International multisite double-blind trial of the
Department of Health, London. atypical antipsychotics risperidone and olanzapine in 175 elderly patients
Department of Health and Social Security (1978). In-patient statistics from with chronic schizophrenia. American Journal of Geriatric Psychiatry, 11,
the Mental Health Inquiry for England: 1975. HMSO, London. 638–47.
Diwan, S., et al. (2007). Depression in older adults with schizophrenia Jeste, D.V., Wolkowitz, O.M., and Palmer, B.W. (2011). Divergent trajectories
spectrum disorders: prevalence and associated factors. American Journal of physical, cognitive and psychosocial aging in schizophrenia.
of Geriatric Psychiatry, 15, 991–8. Schizophrenia Research, 37, 451–5.
Dixon, L., et al. (2000). Prevalence and correlates of diabetes in national Jin, H., et al. (2001). Association of depressive symptoms with worse
schizophrenia samples. Schizophrenia Bulletin, 26, 903–12. functioning in schizophrenia: a study in older outpatients. Journal of
Eyler Zorrilla, L.T., et al. (2000). Cross-sectional study of older outpatients Clinical Psychiatry, 62, 797–803.
with schizophrenia and healthy comparison subjects: no differences Jolley, D., Kosky, N., and Holloway, F. (2004). Older people with long-standing
in age-related cognitive decline. American Journal of Psychiatry, 157, mental illness: the graduates. Advances in Psychiatric Treatment, 10,
1324–6. 27–34.
Felmet, K., Zisook, S., and Kasckow, J.W. (2011). Elderly patients with Jones, P., et al. (1994). Child development risk factors for adult schizophrenia
schizophrenia and depression: diagnosis and treatment. Clinical in the British 1946 birth cohort. Lancet, 344, 1398–402.
Schizophrenia and Related Psychoses, 4, 239–50. Kane, J.M., et al. (1988). Clozapine in treatment-resistant schizophrenics.
Folsom, D.P., Depp, C., and Palmer, B.W. (2009). Physical and mental Psychopharmacology Bulletin, 24, 62–7.
health-related quality of life among older people with schizophrenia. Kasckow, J.W., et al. (2001). Citalopram augmentation of antipsychotic
Schizophrenia Research, 108, 207–13. treatment in older schizophrenia patients. International Journal of
Friedman, J.I., et al. (2001). Six-year follow-up study of cognitive and Geriatric Psychiatry, 16, 1163–7.
functional status across the lifespan in schizophrenia: a comparison with Keefe, R.S., et al. (2007). Neurocognitive effects of antipsychotic medications
Alzheimer’s disease and normal aging. American Journal of Psychiatry, in patients with chronic schizophrenia in the CATIE trial. Archives of
158, 1441–8. General Psychiatry, 64, 633–47.
Glassman, A.H. and Bigge, J.T. (2001). Antipsychotic drugs: prolonged QT Kennedy, J.S., et al. (2003). Olanzapine versus haloperidol in geriatric
interval, torsade de pointes and sudden death. American Journal of schizophrenia: analysis of data from a double-blind controlled trial.
Psychiatry, 158, 1774–82. International Journal of Geriatric Psychiatry, 18, 1013–20.
Granholm, E., et al. (2005). A randomized, controlled trial of cognitive Kessler, R.C., et al. (2005). The prevalence and correlates of nonaffective
behavioral social skills training for middle-aged and older outpatients psychosis in the National Comorbidity Survey Replication (NCS-R).
with chronic schizophrenia. American Journal of Psychiatry, 162, 520–9. Biological Psychiatry, 58, 668–76.
Granholm, E., et al. (2007). Randomized controlled trial of cognitive Kilbourne, A.M., et al. (2008). Quality of care for cardiovascular
behavioral social skills training for older people with schizophrenia: disease-related conditions in patients with and without mental disorders.
12-month follow-up. Journal of Clinical Psychiatry, 68, 730–7. Journal of General Internal Medicine, 23, 1628–33.
Green, M.F. (1996). What are the functional consequences of neurocognitive Killaspy, H. (2006). From the asylum to community care: learning from
deficits in schizophrenia? American Journal of Psychiatry, 153, 321–30. experience. British Medical Bulletin, 79–80, 245–58.
Harding, C. M., et al. (1987). The Vermont longitudinal study of persons Kurtz, M.M. (2005). Neurocognitive impairment across the lifespan in
with severe mental illness. II: long-term outcome of subjects who schizophrenia: an update. Schizophrenia Research, 74, 15–26.
retrospectively met DSM-III criteria for schizophrenia. American Journal Laursen, T.M. (2011). Life expectancy among persons with schizophrenia or
of Psychiatry, 144, 727–35. bipolar affective disorder. Schizophrenia Research, 131, 101–4.
Harris, E.C. and Barraclough, B. (1998). Excess mortality of mental disorder. Leff, J. and Trieman, N. (2000). Long-stay patients discharged from psychiatric
British Journal of Psychiatry, 173, 11–53. hospitals: social and clinical outcomes after five years in the community.
Harris, M.J., et al. (1997). Neuroleptic dose reduction in older psychotic The TAPS Project 46. British Journal of Psychiatry, 176, 217–23.
patients. Schizophrenia Research, 30, 241–8. Lefley, H.P. (1987). Aging parents as caregivers of mentally ill adult children:
Harrison, G., et al. (2001). Recovery from psychotic illness: a 15- and 25-year an emerging social problem. Hospital and Community Psychiatry, 38,
international follow up study. British Journal of Psychiatry, 178, 506–17. 1063–70.
Harvey, P.D., et al. (1999). Cognitive decline in late-life schizophrenia: a Lelliott, P., et al. (1996). The mental health residential care study: classification
longitudinal study of geriatric chronically hospitalized patients. Biological of facilities and description of residents. British Journal of Psychiatry,
Psychiatry, 45, 32–40. 169, 139–47.
Heaton, R., et al. (1994). Neuropsychological deficits in schizophrenics: Leung, W.W., Bowie, C.R., and Harvey, P.D. (2008). Functional implications
relationship to age, chronicity, and dementia. Archives of General of neuropsychological normality and symptom remission in older
Psychiatry, 51, 469–76. outpatients diagnosed with schizophrenia: a cross-sectional study.
Heinrichs, R.W. and Zakzanis, K.K. (1998). Neurocognitive deficit in Journal of the International Neuropsychological Society, 14, 479–88.
schizophrenia: a quantitative review of the evidence. Neuropsychology, Lindenmayer, J.-P. and Khan, A. (2011). Galantamine augmentation of
12, 426–45. long-acting injectable risperidone for cognitive impairments in chronic
Howanitz, E., et al. (1999). The efficacy and safety of clozapine versus schizophrenia. Schizophrenia Research, 125, 267–77.
chlorpromazine in geriatric schizophrenia. Journal of Clinical Psychiatry, Macpherson, R., Shepherd, G., and Edwards, T. (2004). Supported
60, 41–4. accommodation for people with severe mental illness: a review. Advances
Huber, G., Gross, G., and Schüttler, R. (1975). A long-term follow-up study in Psychiatric Treatment, 10, 180–8.
of schizophrenia: psychiatric course of illness and prognosis. Acta Martin, R.L., et al . (1985). Frequency and differential diagnosis of
Psychiatrica Scandinavica, 52, 49–57. depressive syndromes in schizophrenia. Journal of Clinical Psychiatry,
Jeste, D.V. (2004). Tardive dyskinesia rates with atypical antipsychotics in 46, 9–13.
older adults. Journal of Clinical Psychiatry, 65 (Suppl 9), 21–4. Marneros, A., Deister, A., and Rohde, A. (1992). Comparison of long term
Jeste, D.V., et al. (1996). Medical comorbidity in schizophrenia. Schizophrenia outcome of schizophrenic, affective and schizoaffective disorders. British
Bulletin, 22, 413–30. Journal of Psychiatry, 161 (Suppl 18), 44–51.
McCreadie, R.G. and Scottish Schizophrenia Lifestyle Group (2003). Rodriguez-Ferrera, S., Vassilas, C., and Haque, S. (2004). Older people
Diet, smoking and cardiovascular risk in people with schizophrenia: with schizophrenia: a community study in a rural catchment area.
descriptive study. British Journal of Psychiatry, 183, 534–9. International Journal of Geriatric Psychiatry, 19, 1181–7.
McManus, S., et al. (2009). Adult psychiatric morbidity in England, 2007: Royal College of Psychiatrists (2002). Caring for people who enter old age
results of a household survey. NHS Information Centre for Health and with enduring or relapsing mental illness (CR110). Royal College of
Social Care, London. Psychiatrists, London.
McNulty, S.V., et al. (2003). Care needs of elderly people with schizophrenia. Royal College of Psychiatrists (2009). Links not boundaries: service transitions
Assessment of an epidemiologically defined cohort in Scotland. British for people growing older with enduring or relapsing mental illness (CR153).
Journal of Psychiatry, 182, 241–7. Royal College of Psychiatrists, London.
Meesters, P.D., et al. (2011). Symptomatic remission and associated factors in Rund, B.R. (1998). A review of longitudinal studies of cognitive functions in
a catchment area based population of older patients with schizophrenia. schizophrenia patients. Schizophrenia Bulletin, 24, 425–35.
Schizophrenia Research, 126, 237–44. Saxena, S. and Orley, J. (1997). Quality of life assessment: the World
Mitchell, A.J. and Lord, O. (2010). Do deficits in cardiac care influence Health Organisation perspective. European Psychiatry, 12 (Suppl 3),
high mortality rates in schizophrenia? A systematic review and pooled 263–6.
analysis. Journal of Psychopharmacology, 24 (Suppl 4), 69–80. Schmidt, L.M., Hesse, M., and Lykke, J. (2011). The impact of substance
Mittal, D., et al. (2006). Correlates of health-related quality of well-being use disorders on the course of schizophrenia: a 15 year follow up study.
in older patients with schizophrenia. Journal of Nervous and Mental Schizophrenia Research, 130, 228–33.
Disease, 194, 355–60. Shepherd, S., et al. (2012). Perspectives on schizophrenia over the lifespan: a
Moser, D.J., et al. (2006). Using a brief intervention to improve decisional qualitative study. Schizophrenia Bulletin, 38, 295–303.
capacity in schizophrenia research. Schizophrenia Bulletin, 32, 116–20. Silverstein, S.M. and Bellack, A.S. (2008). A scientific agenda for the concept
National Collaborating Centre for Mental Health (2010). Schizophrenia: the of recovery as it applies to schizophrenia. Clinical Psychology Review, 28,
NICE guideline on core interventions in the treatment and management 1108–24.
of schizophrenia in primary and secondary care, updated edition. British Siris, S.G. (2000). Depression in schizophrenia: perspective in the era of
Psychological Society and Royal College of Psychiatrists, Leicester and ‘atypical’ antipsychotic agents. American Journal of Psychiatry, 157,
London. 1379–89.
Osborn, D.P., et al. (2008). Relative risk of diabetes, dyslipidaemia, Tai, S. and Turkington, D. (2009). The evolution of cognitive behavior
hypertension and the metabolic syndrome in people with severe mental therapy for schizophrenia: current practice and recent developments.
illnesses: systematic review and metaanalysis. BMC Psychiatry, 8, 84. Schizophrenia Bulletin, 35, 865–73.
Palmer, B.W. and Jeste, D.V. (2006). Relationship of individual cognitive Tanzman, B. (1993). An overview of surveys of mental health consumers’
abilities to specific components of decisional capacity among middle-aged preferences for housing and support services. Hospital and Community
and older patients with schizophrenia. Schizophrenia Bulletin, 32, Psychiatry, 44, 450–55.
98–106. Trieman, N. and Leff, J. (2002). Long term outcome of long-stay psychiatric
Patterson, T.L., et al. (1996). Quality of well-being in late life psychosis. in-patients considered unsuitable to live in the community: TAPS Project
Psychiatry Research, 63, 199–210. 44. British Journal of Psychiatry, 181, 428–32.
Patterson, T.L., et al. (2006). Functional adaptation skills training (FAST): Trieman, N., Leff, J., and Glover, G. (1999). Outcome of long stay psychiatric
a randomized trial of a psychosocial intervention for middle-aged and patients resettled in the community: prospective cohort study. British
older patients with chronic psychotic disorders. Schizophrenia Research, Medical Journal, 319, 13–16.
86, 291–9. Tsai, J., Stroup, T.S., and Rosenheck, R.A. (2011). Housing arrangements
Perkins, D.O., et al. (2005). Relationship between duration of untreated among a national sample of adults with chronic schizophrenia living in
psychosis and outcome in first-episode schizophrenia: a critical review the United States: a descriptive study. Journal of Community Psychology,
and meta-analysis. American Journal of Psychiatry, 162, 1785–804. 39, 76–88.
Priebe, S., et al. (2005). Reinstitutionalisation in mental health care: Tsuang, M., Woolson, R., and Fleming, J. (1979). Long term outcome of
comparison of data on service provision from six European countries. major psychoses: I. Schizophrenia and affective disorders compared with
British Medical Journal, 330, 123–6. psychiatrically symptom-free surgical conditions. Archives of General
Quinn, J, et al. (2001). Vulnerability to involuntary movements over a Psychiatry, 36, 1295–301.
lifetime trajectory of schizophrenia approaches 100%, in association Vahia, I.V., et al. (2008). Adequacy of medical treatment among older persons
with executive (frontal) dysfunction. Schizophrenia Research, 49, 79–87. with schizophrenia. Psychiatric Services, 59, 853–9.
Radhakrishnan, R., Butler, R., and Head, L. (2012). Dementia in schizophrenia. Velligan, D.I., et al. (1997). The functional significance of symptomatology
Advances in Psychiatric Treatment, 18, 144–53. and cognitive function in schizophrenia. Schizophrenia Research, 25,
Rapp, M.A., et al. (2010). Cortical neuritic plaques and hippocampal 21–31.
neurofibrillary tangles are related to dementia severity in elderly Westermeyer, J. (2006). Comorbid schizophrenia and substance abuse: a
schizophrenia patients. Schizophrenia Research, 116, 90–6. review of epidemiology and course. American Journal of Addictions, 15,
Reine, G., et al. (2005). Assessing health-related quality of life in patients 345–55.
suffering from schizophrenia: a comparison of instruments. European Zisook, S, et al. (2009). Citalopram augmentation for subsyndromal depressive
Psychiatry, 20, 510–19. symptoms in middle-aged and older outpatients with schizophrenia and
Roberts, L., et al. (2007). Physical health care of patients with schizophrenia schizoaffective disorder: a randomized controlled trial. Journal of Clinical
in primary care: a comparative study. Family Practice, 24, 34–40. Psychiatry, 70, 562–71.
CHAPTER 49
Alcohol and substance
abuse in older people
Henry O’Connell and Brian Lawlor
Alcohol use disorders (AUDs) are common in older people and are medications for older people, which may at times be inappropriate,
associated with considerable morbidity and mortality. However, along with variable compliance, altered pharmacokinetics, reduced
clinical practice, medical research, public health initiatives, and functional ability, and increased levels of physical, psychiatric, and
media attention focus primarily on how such problems affect cognitive morbidity. As with AUDs, clinical features of IMU may
younger people, partly because of the more clinically ‘silent’ nature be atypical and masked by other conditions and thus go undetected
of AUDs in older people. As a result, AUDs are less likely to be and untreated (Beers et al., 2000).
detected in older people, because of a general lack of awareness and Illicit drug use in older people is far less of a problem in com-
knowledge of older-specific aspects of clinical presentation and the parison to AUDs and IMU and so will receive less attention in this
use of inappropriate screening instruments and diagnostic criteria chapter. However, there is emerging evidence that, as with AUDs,
that are geared towards younger people. Furthermore, even when generations of people reaching old age in the coming decades may
such problems are detected in older people, they are less likely than carry with them higher levels of illicit drug use than current and
younger people to be treated adequately or referred on to specialist past generations of older people (Patterson and Jeste, 1999; Dowling
treatment facilities (Khan et al., 2002; O’Connell et al., 2003a; Dar, et al., 2008). Principles similar to those seen with AUDs and IMU
2006). apply, in that lower levels of drug intake are required to cause harm
The ageing of populations worldwide means that the absolute and presentation may be atypical and thus go undetected.
number of older people with AUDs is on the increase and, due to This chapter covers the many older-specific aspects of alco-
worldwide cultural changes in the past three to four decades, cohorts hol and substance use disorders and is divided in two ways.
of people reaching old age in the coming years, e.g. the so-called First, apart from the Conclusion, there are seven main sections:
baby boomers (Patterson and Jeste, 1999), are likely to have higher Definitions and Diagnosis; Epidemiology; Aetiology, Risk Factors,
prevalence rates of AUDs and other substance use disorders than and Associations; Clinical Features and Comorbidity; Clinical
current and past generations of older people. Morbidity associated Assessment, Investigations, and Screening; Management and
with AUDs in older people affects practically all aspects of the indi- Prevention, and finally Prognosis. Second, each of these seven
vidual’s physical, psychiatric, cognitive, and social health and well- sections is further divided into three subsections, relating to AUDs,
being, and makes a substantial contribution to premature mortality. IMU, and the abuse of illicit substances. We have also discussed
Such morbidity and mortality also has significant knock-on effects smoking in older people at the end of the chapter.
for the individual’s family and carers, for healthcare professionals
and service planners, and for society in general. Definitions and Diagnosis
Therefore, it is imperative that increased attention at all levels
be focused on this important area, with the aims of increasing AUDs in older people
detection rates and providing accessible, specialized and effective Alcohol use disorders is a general term used to include the wide
treatment services, thereby preventing the development of a ‘silent spectrum of problems associated with alcohol use, from excessive
epidemic’ of alcohol and substance use disorders in older people consumption of alcohol above recommended ‘safe’ or ‘healthy’ lev-
(O’Connell et al., 2003a; Dar, 2006; Reid and Anderson, 1997). els of intake to harmful use and alcohol dependence. AUDs and
Older people use more medications and are at a higher risk of problems associated with use of other substances forms section
inappropriate medication use (IMU), resulting in more adverse F1 of the ICD-10, ‘Mental and behavioural disorders due to psy-
events than any other age group (Chutka et al., 2004, 2005; Barry choactive substance use’ (see Table 49.1). The general criteria for
et al., 2006; Gallagher et al., 2008) and also with the potential for harmful use and dependence syndrome of all substances, includ-
development of abuse and dependence on psychoactive agents. ing alcohol, are given in Tables 49.2–49.6. The ICD-10 (World
Various factors combine to increase this risk, including high lev- Health Organization, 1992) and DSM-IV (American Psychiatric
els of prescribing of both psychotropic and nonpsychotropic Association, 1994) use largely similar diagnostic criteria (see, for
Table 49.1 ICD-10: Mental and behavioural disorders due to psychoactive substance use
F10. Mental and behavioural disorders due to use of alcohol
F11. Mental and behavioural disorders due to use of opioids
F12. Mental and behavioural disorders due to use of cannabinoids
F13. Mental and behavioural disorders due to use of sedatives or hypnotics
F14. Mental and behavioural disorders due to use of cocaine
F15. Mental and behavioural disorders due to use of other stimulants, including caffeine
F16. Mental and behavioural disorders due to use of hallucinogens
F17. Mental and behavioural disorders due to use of tobacco
F18. Mental and behavioural disorders due to use of volatile solvents
F19. Mental and behavioural disorders due to multiple drug use and use of other psychoactive substances
Table 49.2 ICD-10 criteria for acute intoxication

G1. There must be clear evidence of recent use of a psychoactive substance (or substances) at sufficiently high dose levels to be consistent with intoxication.
G2. There must be symptoms or signs of intoxication compatible with the known actions of the particular substance (or substances), as specified below, and of
sufficient severity to produce disturbances in the level of consciousness, cognition, perception, affect, or behaviour that are of clinical importance.
G3. Symptoms or signs present cannot be accounted for by a medical disorder unrelated to substance use and are not better accounted for by another mental or
behavioural disorder.
Acute intoxication frequently occurs in persons who have more persistent alcohol or drug-related problems in addition. Where there are such problems, e.g.
harmful use, dependence syndrome, or psychotic disorder, they should also be recorded.
Table 49.3 ICD-10 criteria for harmful use

A. There must be clear evidence that the substance use was responsible for (or substantially contributed to) physical or psychological harm, including impaired
judgement or dysfunctional behaviour.
B. The nature of the harm should be clearly identifiable (and specified).
C. The pattern of use has persisted for at least 1 month or has occurred repeatedly within a 12-month period.
D. The disorder does not meet the criteria for any other mental or behavioural disorder related to the same drug in the same time period (except for acute
intoxication).
example, Tables 49.4 and 49.5) and we will refer mainly to ICD-10 ICD-10 states that identification of the psychoactive substance
criteria in this chapter. (alcohol or otherwise) should be based on as many sources as possi-
Most research and clinical descriptions of alcohol status describe ble. These include self-report data, analysis of blood and other body
individuals as belonging to one of five main alcohol categories: fluids, characteristic physical and psychological symptoms, clinical
abstinent (for a specified period of time or for their entire life- signs and behaviour, and other evidence such as a drug being in the
time); moderate drinkers, i.e. individuals who drink within recom- patient’s possession or reports from informed third parties.
mended ‘safe’ or ‘healthy levels’ and who do not have criteria for An individual’s drinking status is not static and may vary
heavy use or other AUD; individuals with ‘heavy use’ of alcohol (i.e. throughout life, because of changes in life circumstances and health
those who drink above recommended levels but without obvious characteristics. For example, individuals with alcohol dependence
negative social, behavioural, or health consequences); individuals syndrome in their younger years may quit alcohol use completely
with alcohol dependence syndrome (see Tables 49.4 and 49.5); and and be described as alcohol abstainers in their later years, but con-
a fifth group whose alcohol use is problematic but milder in sever- tinue to have some of the adverse health characteristics of indi-
ity than dependence; these latter individuals may be described as viduals with active alcohol dependence syndrome, such as residual
being problem drinkers or having harmful use (ICD-10) or abuse physical or psychological health problems, social and occupational
(DSM-IV) of alcohol (see Table 49.3). ‘Binge drinkers’, i.e. individu- impairments acquired when their AUD was active, or inherent
als who do not fulfil criteria for alcohol dependence syndrome but personality traits such as impulsivity, hyperactivity, or antisocial
who have a pattern of regular heavy drinking sessions associated personality traits that put them at risk for other physical and men-
with adverse health consequences, may also be included in the lat- tal health problems. The presence of these so-called sick-quitters
ter group. among nondrinkers in observational studies may in part explain
CHAPTER 49 alcohol and substance abuse in older people 633
Table 49.4 ICD-10 criteria for dependence syndrome

Three or more of the following manifestations should have occurred together for at least 1 month or, if persisting for periods of less than 1 month, should have
occurred together repeatedly within a 12-month period:
(1) a strong desire or sense of compulsion to take the substance;
(2) impaired capacity to control substance taking behaviour in terms of its onset, termination, or levels of use, as evidenced by the substance being often taken in
larger amounts or over a longer period than intended, or by a persistent desire or unsuccessful efforts to reduce or control substance use;
(3) a physiological withdrawal state when substance use is reduced or ceased, as evidenced by the characteristic withdrawal syndrome for the substance, or by use
of the same (or closely related) substance with the intention of relieving or avoiding withdrawal symptoms;
(4) evidence of tolerance to the effects of the substance, such that there is a need for significantly increased amounts of the substance to achieve intoxication or
the desired effect, or a markedly diminished effect with continued use of the same amount of the substance;
(5) preoccupation with substance use, as manifested by important alternative pleasures or interests being given up or reduced because of substance use; or a great
deal of time being spent in activities necessary to obtain, take, or recover from the effects of the substance;
(6) persistent substance use despite clear evidence of harmful consequences, as evidenced by continued use when the individual is actually aware, or may be
expected to be aware, of the nature and extent of harm.
Table 49.5 DSM-IV criteria for dependence syndrome

As with ICD-10, three or more of the criteria listed below must have been met in the previous 12 months to constitute a diagnosis of substance dependence.
1. Tolerance, as defined by either of the following:
(a) need for markedly increased amounts of substance to achieve intoxication or desired effect; or
(b) markedly diminished effect with continued use of the same amount of the substance.
2. Withdrawal, as manifested by either of the following:
(a) characteristic withdrawal syndrome for the substance; or
(b) the same (or a closely related) substance is taken to relieve avoiding withdrawal symptoms.
3. The substance is often taken in larger amounts or over a longer period than was intended.
4. There is a persistent desire or unsuccessful efforts to cut down or control substance use.
5. A great deal of time is spent in activities necessary to obtain the substance (e.g. visiting multiple doctors or driving long distances), use the substance (e.g. chain
smoking), or recover from its effects.
6. Important social, occupational, or recreational activities are given up or reduced because of substance use.
7. Substance use is continued despite knowledge of having a persistent physical or psychological problem that is likely to have been caused or exacerbated by
the substance (e.g. current cocaine use despite recognition of cocaine-induced depression, or continued drinking despite a peptic ulcer made worse by alcohol
consumption).
Table 49.6 ICD-10 criteria for withdrawal state

G1. There must be clear evidence of recent cessation or reduction of substance use after repeated, and usually prolonged and/or high-dose, use of that substance.
G2. Symptoms and signs are compatible with the known feature of a withdrawal state for the particular substance or substances.
G3. Symptoms and signs are not accounted for by a medical disorder unrelated to substance use, and not better accounted for by another mental or behavioural
disorder.
Table 49.7 Definitions of a standard drink

A standard drink of beverage alcohol is equivalent to a 12-floz (US; 355 ml) domestic beer (alcohol content about 4%), a 5-fl oz (US; 148 ml) glass of table wine
(about 12% alcohol), or a mixed drink containing 1–1.5 fl oz (US; 29–40 ml) hard liquor (about 40% alcohol).
the less favourable health characteristics of alcohol abstainers in for women and 21 units per week for men, according to the Royal
comparison to moderate drinkers. Furthermore, such individuals Colleges Report (1995). These units should not all be consumed
may be at risk of developing a relapse of their AUD in the context of at the one sitting. However, these recommendations relate to the
health problems or changes in social circumstances in later life. general population of all ages, and ‘safe’ or ‘healthy’ levels of intake
Definitions of a unit of alcohol are given in Table 49.7. The rec- are likely to be lower for older people. Within the broadly defined
ommended upper limits of alcohol intake are 14 units per week group of older people, spanning as it does four decades in age,
even more detailed recommendations should apply. For exam- people on anticoagulant treatment may be unknowingly doing
ple, a 90-year-old woman who has multiple medical problems is themselves harm, even if their alcohol intake is moderate.
unlikely to tolerate the same amount of alcohol as a fit and healthy
65-year-old man. Abuse of illicit substances in older people
The US National Institute of Alcohol Abuse and Addiction ICD-10 defines specific diagnostic criteria for intoxication, harm-
(NIAAA) recommends no more than one drink per day for older ful use, dependence, and withdrawal state for a wide range of ille-
people (NIAAA, 1998), but there are no such older-specific criteria gal substances, including opioids (heroin), cannabinoids, cocaine,
in Europe or described by the ICD-10 or DSM-IV. Therefore, in other stimulants including caffeine, hallucinogens, tobacco, and
applying existing diagnostic criteria, AUDs in older people may be volatile solvents (see Table 49.1).As with AUDs, these diagnostic
missed, as older people may not display the classical symptoms or criteria may not be applicable for older people, as older people may
signs of AUDs seen in younger people, such as a craving or com- experience harm at lower levels of use, they may not display clas-
pulsion, and signs of intoxication, changes in tolerance, withdrawal sical features of craving or compulsion, and intoxication and with-
states, and other diagnostic criteria for AUDs may be atypical or drawal states may be masked by other medical or neuropsychiatric
masked by other health problems. Furthermore, older people may conditions. As with AUDs, low levels of clinical suspicion are likely
not experience the same degree of legal, family, and occupational to apply to older people with illicit drug use.
problems associated with AUDs as seen in working-age adults, and
biophysical screening measures for AUDs such as mean corpuscu- Epidemiology
lar volume (MCV) and abnormal liver function tests may not be
as sensitive in older people. Older people with covert AUDs may Epidemiology of AUDs in older people
present with a wide range of seemingly unrelated problems, such AUDs are common in older people and are associated with signifi-
as unexplained falls and fractures, confusion, treatment-resistant cant morbidity and mortality. The real prevalence of AUDs in older
depression, and adverse drug reactions, and the underlying AUD is people is often underestimated, however, for a variety of reasons, as
likely to go undetected unless a high degree of clinical suspicion is outlined in Table 49.8.
maintained (this is discussed in more detail in the section Clinical High proportions of community-dwelling older people have
Assessment, Investigations, and Screening; see also Table 49.10). been reported to drink alcohol. For example, in one study of
A stereotypical idea of the ‘down and out’ alcoholic with a high community-dwelling older people in the US, 62% were found to
consumption of alcohol is also likely to lead to missed diagnoses, drink alcohol (Mirand and Welte, 1996). However, levels of alcohol
as older people may experience significant health problems even at use vary widely between different populations of older people and
relatively low levels of intake, or experience new problems in later are likely to be influenced by numerous psychosocial and sociode-
life due to age-related reductions in tolerance, even while continu- mographic factors such as age, clinical characteristics, social class,
ing a lifelong pattern of moderate consumption. ethnicity, and religion.
There are few systematic studies that have examined the preva-
Inappropriate medication use (IMU) in older people lence of alcohol abuse/dependence in people over the age of 65.
Broadly speaking, medications may be divided into prescription-only A recent study (Blazer and Wu, 2011) examining the prevalence
(generally more potentially toxic) and ‘over the counter’ (OTC) of alcohol abuse, dependence, and subthreshold dependence
medications. However, such a distinction may be arbitrary, as peo- among middle-aged and older people in the US found that about
ple of all ages, including older people, may experience significant 6.7% (dependence 0.6%, abuse 0.9%, and subthreshold depend-
problems with abuse of common OTC medications, such as para- ence 5.2%) of those older than 65 reported alcohol abuse, depend-
cetamol and cough mixtures containing codeine. ence, or dependence symptoms. Among past-year alcohol users,
A more useful distinction may be to divide medications into psy- 15.4% (dependence 1.3%, abuse 2.1%, and subthreshold depend-
chotropic (abuse associated more with neuropsychiatric effects) ence 12.0%) of those older than 65 endorsed alcohol abuse or
and nonpsychotropic. The ICD-10 uses the same general princi-
ples of intoxication, harmful use, dependence, and withdrawal state
Table 49.8 Reasons for underdetection and misdiagnosis of AUDs in
that apply to alcohol for use of sedative and hypnotic medications
older people
(see Tables 49.2–49.6). As with AUDs, older-specific criteria are not
cited, but the same general principles apply: older people are likely ◆ Older people and their families and carers reluctant to disclose alcohol use
to experience harm at lower levels of use and clinical features guid- and AUDs
ing diagnosis are more likely to be atypical and masked by other ◆ Inaccurate recall of alcohol intake due to cognitive impairment
health problems.
◆ Lack of clinical suspicion on the part of healthcare workers
In addition to incorporating a framework of abuse and depend-
ence for psychoactive agents, the concept of IMU can be further ◆ Atypical and masked clinical presentation
extended to include overuse or potentially inappropriate use of ◆ Use of inappropriate screening and diagnostic instruments
medications (Beers et al., 2000). Overuse occurs when a drug is
◆ Recommended levels of alcohol intake inappropriately high for older people
used when no drug should be used at all. Drug misuse occurs when
the wrong drug is used or a drug is used at the wrong dose, at the ◆ Reduced likelihood of referral of older people for specialist AUD treatment
wrong schedule, or for the wrong duration. ◆ Therapeutic pessimism/nihilism
The concept of IMU may be further complicated due to interac-
◆ Alcohol use and AUDs in later life perceived as ‘a comfort’ and
tions with alcohol (even when taken only at moderate levels or in
‘understandable’
the context of an AUD) or with illicit drugs. For example, older
dependence symptoms. ‘Tolerance’ (48%) and ‘time spent using’ Apart from the timing of onset of AUD, other key differences, with
(37%) were the two symptoms most frequently endorsed by the implications for clinical outcome, have been noted to exist between
subthreshold group. these two groups: individuals with early-onset AUDs are more
Rates of recent alcohol use have been noted to decline after the likely to have a positive family history of AUDs, antisocial person-
age of 55, to 25% of people aged 85 and older (Ruchlin, 1997) and, ality traits, and more severe AUDs with greater associated damage
in general, levels of alcohol use and the prevalence of AUDs decline to their physical and psychological health and social networks. As
with age (Temple and Leino, 1989; Adams et al., 1990, 1995). a result, individuals with early-onset AUDs may be more difficult
However, there have also been reports of stable (Ekerdt et al., 1989) to engage in treatment (Liberto and Oslin, 1995). In contrast, those
or increased (Gordon and Kannel, 1983) levels of alcohol consump- with late-onset AUDs may have higher levels of income and educa-
tion in later life. tion and more circumscribed AUDs triggered by discrete stressful
The general decline in prevalence of AUDs with age may be life events such as bereavement or retirement (LaGreca et al., 1988).
due to a number of factors, including the premature deaths of There is evidence to suggest that those with late-onset AUDs have
those with early-onset AUDs. Furthermore, age-related changes better clinical outcomes (Schutte et al, 1994).
in pharmacokinetics leading to reduced physiological reserve,
along with an increased prevalence of medical conditions and dis- Epidemiology of IMU in older people
abilities, may lead to lower levels of alcohol intake in older people. Older people receive more prescriptions than any other age group,
Psychosocial factors such as diminished social networks, social and are also more likely to be dispensed multiple drug regimens
isolation, and financial problems may also lead to reduced alco- (Patterson and Jeste, 1999; McGrath et al., 2005). For example,
hol intake, although these factors may conversely lead to increased although older people comprise 13% of the US population, they
alcohol intake. An age cohort effect may exist, with the prevalence have been estimated to use more than 30% of prescription drugs
of AUDs differing between different generations, depending on the (Williams and Lowenthal, 1992; Avorn, 1995) and 35% of OTC
sociocultural norms of the time. In the US, for example, individuals drugs (Williams and Lowenthal, 1992). It has been estimated that
reared in the Prohibition era of the 1920s, when alcohol use was older people use prescription and OTC medications approximately
outlawed and stigmatized, are likely to drink less than the so-called three times as much as the general population and that the esti-
baby boomer generation who grew up in the more liberal climate of mated annual expenditure on prescription drugs by older people
the 1960s and who will start to reach the age of 65 in 2011. AUDs is four times that of younger people (Anderson et al., 1993; Jeste
at all levels of severity are far more common in men than women, and Palmer, 1998). While polypharmacy may be appropriate and
generally by a factor of four to six times. Other sociodemographic clinically indicated in older people, the risk of adverse drug reac-
factors may include socioeconomic group and geographical setting tions has been reported to double when utilization increases from
(i.e. urban or rural), although these relationships are not clear in one to four medications, and the risk increases 14-fold when seven
older people. drugs are used (Cadieux, 1989; Chrischilles et al., 1992). It has been
Culture and ethnicity are important associations of AUDs in gen- estimated that older people in the US use 5.8 prescription drugs
eral adult populations and these factors are also likely to be impor- concurrently along with 3.2 OTC drugs (Williams and Lowenthal,
tant in older people. Therefore levels of AUDs are likely to be lower 1992; Sintonen, 1994; Avorn, 1995).
in religious and ethnic groups where alcohol use is strictly sanc- In a large US retrospective cohort study involving 493,971 inpa-
tioned or not allowed, such as among Muslims and Orthodox Jews. tients (Rothberg et al., 2008), 49% received at least one potentially
In the UK, alcohol-related mortality may be marginally elevated in inappropriate medication and 6% received three or more. Patient,
people of Caribbean origin, and is substantially higher in people of physician, and hospital characteristics were all associated with
Irish and Indian origin (Harrison et al., 1997). potentially inappropriate medication use.
The prevalence of AUDs in older people also varies depending on These high levels of medication use, along with age-related
the restrictiveness of diagnostic criteria used and the clinical and impairments in pharmacokinetics, the high levels of polypharmacy
sociodemographic characteristics of the population being studied. (either appropriate or not), and the fact that older people are more
For example, the prevalence rate for ‘excessive alcohol consump- likely to be on long-term medications (e.g. antidiabetic or cardio-
tion’ in a particular population will be higher than that for alco- vascular medication) as opposed to short-term medications (e.g.
hol dependence syndrome. Community-based studies estimate the a single course of antibiotics) than is the case in younger people,
prevalence of alcohol misuse or dependence among older people mean that older people are far more likely to experience an adverse
as 2–4% (Adams and Cox, 1995), with much higher rates of 16% drug event than younger people, even when medications are taken
(men) and 2% (women) when looser criteria such as excessive alco- as directed. Added to this (as outlined in the section Epidemiology
hol consumption are used (UK Datasets, 1994; Greene et al., 2003). of AUDs in older people), medications may be overused, under-
Likewise, the prevalence of AUDs is higher in clinical populations used, used inappropriately, or used in combination with substances
such as medical and psychiatric inpatients than among primary such as alcohol or illegal drugs that may lead to harmful drug
care attendees and normal community-dwelling populations. For interactions.
example, the prevalence of AUDs is higher among older inpatients, Benzodiazepines are the most commonly prescribed psychotropic
with estimates of 14% for emergency department patients (Adams drugs in older people, with one study of community-dwelling older
et al., 1992), 18% for nursing home residents (Joseph et al., 1995), people in Ireland demonstrating that 17% of participants were pre-
and 23% for psychiatric inpatients (Speer and Bates, 1992). scribed benzodiazepines, with use in women being twice that in
It has been estimated that two-thirds of older people with AUDs men, and 18% of benzodiazepine users taking at least one other
started drinking at a younger age, with the remaining one-third psychotropic drug. Furthermore, 52% of benzodiazepine users were
having ‘late-onset’ AUDs (Council on Scientific Affairs, 1996). prescribed a long-acting benzodiazepine (Kirby et al., 1999a). It has
also been reported that depression in older community-dwelling attending for substance use treatment is increasing relative to
people is more likely to be detected if accompanied by anxiety younger adults, with an increasing illicit drug involvement (cocaine
symptoms, and such individuals are at risk of inappropriate treat- and heroin) in older adult admissions, are consistent with a cohort
ment with benzodiazepines (Kirby et al., 1999b). effect of ‘baby-boomers’.
In another study, the prevalence of psychotropic drug misuse
was reported to be four times greater in women than in men, and
other associations demonstrated included widowhood, lower edu-
Aetiology, Risk Factors, and Associations
cational level, lower income, poorer health, and reduced social sup- The aetiology, risk factors, and associations of AUDs, IMU, and ille-
port (King et al., 2003). gal substance use disorders all involve complex interactions among
Abuse of prescription drugs (most often sedative-hypnotic, several factors. These factors may be described as being distal (e.g.
anxiolytic, and analgesic) has been reported to account for 5% of positive family history of AUD) as opposed to proximal (e.g. recent
community-dwelling older mental health clinic attendees (Jinks bereavement leading to excessive drinking), they may be thought of
and Raschko, 1990), and 30% of residents of intermediate care as being fixed (e.g. female gender and risk of IMU) as opposed to
facilities were reported as being prescribed long-acting drugs not modifiable (e.g. rationalizing prescribed medication may reduce the
recommended for older people (Beers et al., 1988). risk of IMU), or they may be thought of as being biological/medical
Misuse of drugs in older people due to poor prescribing prac- (e.g. age-related pharmacokinetic changes leading to relatively higher
tices is also a significant public health problem, with 20% of alcohol toxicity and increased risk of AUD), social (e.g. unemployed
community-dwelling older people receiving a drug considered or divorced status and risk of AUDs), or psychological (e.g. anti-
inappropriate by experts and 40% of nursing home residents receiv- social personality traits or depression and risk of AUDs) in nature.
ing inappropriate medication (Avorn and Gurwitz, 1995). Finally, risk factors can also be described in another three broad
Use of opiate analgesia is common in older people and is liable groupings: predisposing factors; factors that may increase substance
to give rise to inappropriate use and problems with addiction. exposure and consumption level; and factors that may increase the
Therefore, use of these medications should be closely monitored, effects and abuse potential of substances (see Table 49.9).
with due consideration of dose, duration of treatment, and careful
tapering (Schneider, 2005). During the 8-year period of 1995–2002, Aetiology, risk factors, and associations of AUDs in
admission for substance use treatment increased 32% for people older people
over the age of 65 in the US. While the most common drug of abuse
Biological/medical factors
was alcohol, the number of older Americans reporting opiates as
their primary drug of abuse increased from 6.8% to 12% (Office of While the genetics of AUDs is complex, involving the interaction
Applied Studies, 2005). of several genes (Buckland, 2001), we know that positive family
Epidemiology of use of illicit drugs

Table 49.9 Risk factors for substance abuse in older people
Lifetime prevalence rates for illicit drug dependence have been esti-
mated as 17% for 18- to 29-year-olds, 4% for 30- to 59-year–olds, Predisposing factors
and less than 1% for those over the age of 60 (Hinkin et al., 2002).
Family history (alcohol)
Lifetime experience of cannabis in the 65- to 69-year-old age group
in the UK has been reported to be 7 per 1000 (Lawlor et al., 2003). Previous substance abuse
The Epidemiological Catchment Area (ECA) study found that less Previous pattern of substance consumption (individual and cohort effects)
than 0.1% of older people (over age 65) met DSM-III criteria for
Personality traits (sedative-hypnotics, anxiolytics)
illegal drug abuse or dependence in the previous month (Regier
et al., 1988), with corresponding rates for the same period of 3.5% Factors that may increase substance exposure and consumption level
for 18- to 24-year-olds. ECA data suggest a lifetime prevalence Gender (men—alcohol, illicit drugs; women—sedative-hypnotics, anxiolytics)
rate for illicit drug use of only 1.6% for older people (Anthony and
Chronic illness associated with pain (opioid analgesics), insomnia (hypnotic
Helzer, 1991). In the 2009 National Survey on Drug Use and Health drugs), and anxiety (anxiolytic)
(NSDUH) report from the US (Office of Applied Studies, 2009,
during the past year nonmedical use of prescription drugs was Long-term prescribing (sedative-hypnotics, anxiolytics)
more common than marijuana use among those over age 65 (0.8% Caregiver overuse of ‘as needed’ medication (institutionalized older people)
vs 0.4%); however, the rate of marijuana use for ‘baby-boomers’ Life stress, loss, social isolation
(those over age 50) was much higher at 4.2%, and we can therefore
expect a rise in the incidence and prevalence of illicit drug use in Negative affects (depression, grief, demoralization, anger) (alcohol)
older people by 2020 as this population ages. Family collusion and drinking partners (alcohol)
Several other sources of data suggest similarly low rates of Discretionary time, money (alcohol)
illegal drug use among older people. However, the ageing of the
Factors that may increase the effects and abuse potential of substances
‘baby-boomer’ generation is likely to result in a cohort of older
people who are healthier and have higher life-expectancies than Age-associated drug sensitivity (pharmacokinetic, pharmacodynamic factors)
previous generations of older people, but who also carry with them Chronic medical illnesses
higher rates of illegal drug use (Patterson and Jeste, 1999). The find-
Other medications (alcohol–drug, drug–drug interactions)
ings of Amdt and colleagues (2011) demonstrating that the pro-
portion of older adults (defined as 55 years or over in the study) (From Atkinson, R.M., Psychiatry in the Elderly, 2001.)
history is a risk factor for AUDs in the general population and precise mechanisms are complex and not fully clear. However, clin-
is also likely to be important in older people, in relation to both ical experience suggests that symptoms of AUDs may also be the
early-onset (Dahmen et al., 2005) and late-onset (Tiihonen et al., first presenting problems of cognitive impairment or dementia in
1999) AUDs. The genetic risk for AUDs may also overlap with risk older people.
for other mental disorders, such as antisocial personality disorder,
other drug use problems, anxiety disorders, and mood disorders Aetiology, risk factors, and associations of IMU in
(Nurnberger et al., 2004). older people
The higher prevalence of AUDs seen in clinical populations such
The general principles for aetiology, risk factors, and associations for
as medical and psychiatric inpatients (see Epidemiology) implies
substance misuse outlined in the previous section (see Table 49.9)
that AUDs may arise as a consequence of medical conditions or
apply to IMU. The much higher exposure of older people to prescribed
indeed may play a part in the aetiology and perpetuation of medi-
medications (see Epidemiology) along with age-related pharmacoki-
cal conditions.
netic changes and comorbid medical and neuropsychiatric conditions
Age-related changes in pharmacokinetics and reductions in
all interact to heighten the risk of IMU in older people.
physiological reserve also mean that older people are more likely
Biological and medical factors in the development of IMU may
to encounter problems associated with alcohol intake even when
include genetic predisposition (phenotypes associated with other
consumed at moderate levels, along with having an increased risk
factors such as alcohol dependence, illicit drug use, and antisocial
of developing alcohol–drug interactions (National Institute on
personality disorder (Nurnberger et al., 2004)), cognitive impair-
Alcohol Abuse and Alcoholism, 1998).
ment, chronic medical conditions requiring long-term prescribing
Social factors of medication such as pain (Helme and Katz, 1993), and age-related
Social factors are involved in the aetiology of AUDs in older people pharmacokinetic changes leading to an increased likelihood of
as in all age groups, and these factors may have a two-way relation- developing adverse reactions to medication use (Beers et al., 2000).
ship with the development of AUDs (i.e. play a causative role in the There may also be adverse drug interactions between prescribed
development of the AUD or arise as a consequence of the AUD). medications and alcohol.
Male gender is invariably found to be a risk factor for AUDs in The type of medication prescribed is also an important consid-
all age groups and cultures, and this may simply reflect prevail- eration. For example, benzodiazepines are among the medications
ing societal norms of heavier drinking among men that lead to an most commonly implicated in IMU in older people, and the risk
increased risk of developing AUDs. However, social characteristics of IMU with benzodiazepines increases with higher drug potency,
may also reflect underlying biological/medical and psychological shorter elimination half-life, and longer duration of prescription
risk factors and associations (e.g. increased risk of AUDs in men (Kirby et al., 1999). The high prevalence of pain (estimated at
may reflect their higher levels of antisocial personality disorder, a 20–50% in community-dwelling older people: Barkin et al., 2005)
risk factor for AUDs in itself). means that analgesics such as nonsteroidal agents and opiates are
An age cohort effect may also exist, with the prevalence of AUDs likely to be misused in older people and lead to the development
differing between generations, depending on the sociocultural norms of IMU.
of the time. Other important social factors include culture, ethnicity, Psychosocial factors identified to be associated with IMU include
social isolation, and marital status (sees Epidemiology, and Ekerdt older age, female gender, white race, lower educational level, and
et al., 1989; Bristow and Clare, 1992; Ganry et al., 2000). separated or divorced status (Swartz et al., 1991). Other mental dis-
orders associated with benzodiazepine dependence in the general
Psychological factors
population, and also likely to be relevant to older people, include
Personality types and traits associated with AUDs may differ
depression, panic disorder, AUDs and abuse of other substances,
between ‘early-onset’ and ‘late-onset’ types. The ‘early-onset’ type
generalized anxiety disorder, and personality disorders (Swartz et
may have a stronger association with antisocial personality traits,
al., 1991; Busto et al., 1996).
hyperactivity, and impulsivity. ‘Late-onset’ AUDs may have a
stronger relationship with ‘neuroticism’ and depression (Mulder,
2002). Furthermore, there is emerging evidence that the person- Aetiology, risk factors, and associations of illicit drug
alities of lifelong nondrinkers may differ from those of moderate use in older people
drinkers, being more introverted and more neurotic, thus explain- Predictably, older opioid maintenance patients have been reported
ing their poorer physical and psychological health characteristics as to have significantly more medical problems and worse general
they appear on the J-shaped curve (O’Connell et al., 2005). health, along with a later onset of use of illicit substances, than
AUDs in older people, as in all populations, may also have a younger opioid patients (Lofwall et al., 2005), and these findings
two-way relationship with psychiatric disorders such as depression are also likely to apply to older people who use other illegal drugs.
and anxiety disorders. For example, older people may begin drink- Certain illegal drugs may be associated with particular conditions,
ing in an effort to self-medicate depressive symptoms, or they may such as the association seen between pain and abuse of opioids
become depressed because of their drinking (Davidson and Ritson, (Trafton et al., 2004). As with AUDs and IMU, genetic factors may
1993). Previous history of alcohol use or AUD is also important, play a role in the aetiology of illegal drug use in the general popula-
as an individual with a resolved AUD from the past may relapse in tion and in older people (Kuhar et al., 2001).
later life in the context of changes in health characteristics or life Important social factors may include gender (male gender is
circumstances. associated with higher risk of illegal drug use), socioeconomic sta-
AUDs are involved in the aetiology of cognitive impairment and tus, and level of social supports and networks. Age cohort is also
dementia (see Clinical Features and Comorbidity), although the likely to be important, with ‘baby-boomers’ identified as carrying
with them into later life a higher prevalence of illegal drug use than Malignancies
previous generations of older people (Patterson and Jeste, 1999;
Amdt et al., 2011). Mouth, pharynx, larynx, oesophagus, hepatic, colorectal, pancreatic
Use of illegal drugs in the general population is associated with Cardiovascular
an increased prevalence of psychiatric illnesses and personality dis-
Ischaemic heart disease
orders, and this is also likely to be the case in older populations.
Hypertension
Clinical Features and Comorbidity Alcohol-induced arrhythmias
Clinical features and comorbidity in AUDs in older Congestive heart failure
people Alcoholic cardiomyopathy
AUDs in older people are associated with significant morbidity and
Haematological
mortality, affecting practically all aspects of physical, neuropsychi-
atric, and social health and wellbeing, as summarized in Table 49.10 Macrocytosis (acute effect of alcohol intake and due to vitamin B12 and folate
(Reid and Anderson, 1997; O’Connell et al., 2003a). deficiency in chronic AUD)
Physical aspects Anaemia (due to gastrointestinal problems)

Practically every organ system may be adversely affected by AUDs Musculoskeletal
in older people, as outlined in Table 49.10. Pharmacokinetic changes Falls and fractures
in older people, with reduced physiological reserve, reduced meta-
bolic efficiency, and an increased volume of distribution due to a Reduced bone density
higher fat to lean muscle ratio, mean that alcohol intake in older Myopathy
people leads to relatively higher blood alcohol concentrations
Metabolic
(Dufour and Fuller, 1995; Kalant, 1998). This increase in blood
alcohol concentration is associated with a higher risk of intoxica- Hypoglycaemia
tion and harmful effects. Furthermore, even when blood alcohol Hyperuricaemia
concentration is controlled for, alcohol-induced impairment in task
performance has been demonstrated to increase with advancing Elevated lipids
age (Vogel-Sprott and Barrett, 1984). Diabetes more difficult to control
These pharmacokinetic changes, along with the general effects
Neuropsychiatric
of physical and cognitive ageing, increasing frailty, reduced func-
tional ability, and higher levels of concomitant prescription drug Cognitive impairment and dementia
use, mean that alcohol is relatively more toxic to older people than Frontal lobe impairment
younger people. Furthermore, such toxic effects may be subtle and
may be missed or mistaken for other conditions. Wernicke–Korsakoff syndrome
Neuropsychiatric aspects Cerebellar cortical degeneration

AUDs in older people are associated with a wide range of mental Central pontine myelinosis
disorders, such as depression, psychosis, withdrawal syndromes, Marchiafava–Bignami disease
cognitive impairment, and dementia (see Table 49.10). The rela-
tionship between alcohol use and brain damage and dementia is Depression
complex (Nieman, 1998; Fig. 49.1): AUDs may increase the risk Psychosis
for many types of dementia (Letenneur, 2004) and there also exist
Intoxication
diagnostic entities known as ‘alcohol-induced persisting dementia’
(Table 49.11) and an amnesic syndrome associated with alcohol use Withdrawal syndrome (may be more difficult to treat in older people)
(Table 49.12), while light–moderate alcohol use may protect against Suicide
dementia (Ruitenberg et al., 2002). Sulcal widening and ventricular
Other
Alcohol–drug interactions
Table 49.10 Physical, neuropsychiatric, and sociodemographic aspects Aspiration pneumonia
of AUDs in older people
Road traffic and other accidents
Gastrointestinal Sociodemographic
Hepatic problems: elevated liver enzymes; fatty liver; alcoholic hepatitis;
Male gender
cirrhosis; malignancy
Divorced, widowed, and single status
Gastritis, peptic ulcer disease, and bleeding
Social isolation
Oesophageal varices
Upper and lower ends of socioeconomic spectrum
Acute and chronic pancreatitis
enlargement have been cited as the strongest neuropathological consequences of their drinking. In general terms, the nature and
findings in patients with ‘alcohol-induced dementia’, with addi- severity of psychiatric comorbidity is likely to differ between those
tional evidence for peripheral neuropathy, ataxia, sparing of lan- with early- and those with late-onset AUDs, with the early-onset
guage, and improved prognosis when compared to other types of group having higher levels of antisocial personality traits, a stronger
dementia (Smith and Atkinson, 1995). family history, and more severe and complicated AUDs.
There are several ways in which AUDs may cause cogni- The Liverpool longitudinal community study (Saunders et al.,
tive impairment and dementia (Fig. 49.1). AUDs may indirectly 1991) demonstrated that older men with a history of heavy drink-
increase the risk for vascular dementia and Alzheimer’s dementia ing for 5 years or more at some time in their lives had a greater than
by increasing vascular risk factors, and these conditions probably five-fold increased risk of suffering from a psychiatric disorder.
account for the majority of cognitive impairment and dementia Furthermore, it was concluded that those with a current psychi-
seen in older people with AUDs. However, there is also significant atric diagnosis had significantly higher alcohol consumption, and
epidemiological evidence demonstrating reduced risk of dementia the association between heavy alcohol consumption in earlier years
in light and moderate drinkers, although the underlying mecha- and psychiatric morbidity in later life was not simply explained by
nisms for this relationship are yet to be elucidated (Anstey et al., current drinking habits.
2009). AUDs are also associated with the Wernicke–Korsakoff syn-
drome, an amnestic syndrome arising from thiamine deficiency
due to AUDs or other causes. Table 49.11 DSM-IV diagnostic criteria for alcohol-induced persisting
AUDs in older people are frequently associated with other psy- dementia (APA, 1994)
chiatric disorders, most commonly depression. There may be a
A. The development of multiple deficits manifested by both:
two-way interaction between mental disorders and alcohol use. For
example, an individual with depression may begin to drink exces- 1. Memory impairment (impaired ability to learn new information or to
sively, or excessive drinking may lead to depression (Davidson and recall previously learned information); and
Ritson, 1993). Thus, comorbidity may arise as a result of cause or 2. One or more of the following cognitive disturbances:
effect, i.e. older individuals who develop psychological problems
(a) aphasia (language disturbance);
in later life may drink alcohol as a form of ‘self-medication’ (i.e.
‘late-onset’ AUD; see Epidemiology), and put themselves at risk (b) apraxia (impaired ability to carry out motor activities despite intact
of developing an AUD, or individuals with AUDs may develop motor function);
psychiatric disorders because of the physical, cognitive, or social (c) agnosia (failure to recognize or identify objects despite intact sensory
function); and/or
(d) disturbance in executive functioning (i.e. planning, organization,
Head injury
sequencing, abstracting).
B. The cognitive deficits in criteria A1 and A2 each cause significant
Alzheimer’s dis. impairment in social or occupational functioning and represent a significant
decline from a previous level of functioning.
Hypertension C. The deficits do not occur exclusively during the course of a delirium and
Diabetes persist beyond the usual duration of alcohol intoxication or withdrawal.
Ischaemic heart disease D. There is evidence from the history, physical examination, and laboratory
findings that the deficits are aetiologically related to the persisting effects of
Cerebrovascular dis. alcohol use.
Alcohol Use
Disorders Depression
Dementia Table 49.12 Amnesic syndrome (due to alcohol or other substances)

(WHO, 1992)
A. Memory impairment is manifest in both:
‘Alcoholic dementia’
(1) a defect of recent memory (impaired learning of new material) to a
degree sufficient to interfere with daily living; and
(2) a reduced ability to recall past experiences.
Direct toxic effects of alcohol B. All of the following are absent (or relatively absent):
Alcohol withdrawal (1) defect in immediate recall (as tested, e.g., by the digit span);
Electrolyte disturbances
(2) clouding of consciousness and disturbance of attention, as defined in
Liver disease F05, criterion A, of ICD-10;
(3) global intellectual decline (dementia).
C. There is no objective evidence from physical and neurological examination,
laboratory tests, and history of a disorder or disease of the brain (especially
involving bilaterally the diencephalic and medial temporal structures), other
Malnutrition Wernicke-Korsakoff
than that related to substance use, which can reasonably be presumed to
syndrome
be responsible for the clinical manifestations described under criterion A.
Fig. 49.1 Relationship between AUDs and dementia.
In the US, data from the National Longitudinal Alcohol Table 49.13 Clinical features and comorbidity associated with
Epidemiologic Survey demonstrated that major depression is three inappropriate medication use in older people
times more common in the over 65s who have an AUD compared
Neuropsychiatric (all psychotropic drugs; benzodiazepines particularly
to those who do not (Grant and Harford, 1995), and in another
prevalent)
survey, 30% of an older population with AUDs were found to have
concurrent psychiatric disorders (Moos et al., 1998). These high Delirium
levels of comorbidity are also reflected in the observation that up Day-time drowsiness
to 50% of older psychiatric inpatients have been noted to be heavy
Sleep disturbance
users of alcohol (Atkinson and Schuckit, 1983; McGrath et al.,
2005). Depression
Depression in older people with AUDs has a more complicated Anxiety
clinical course and has been demonstrated to be more severe and
Physical
chronic in nature than depression without AUDs (Cook et al.,
1991). Such individuals may be less likely to present to or engage Falls
with mental health services and may be less compliant with any Fractures
treatment strategies introduced. Drug–drug and drug–alcohol interactions
Other psychiatric disorders, such as anxiety disorders and psy-
chotic disorders, are also important but less common comorbidities Problems related to drug metabolism (e.g. renal and hepatic impairment)
of AUDs in older people.
In view of the high degree of psychiatric comorbidity associated
with AUDs in older people, it is not surprising that AUDs in older
marijuana is likely to have different clinical features and comorbidi-
people are also strongly associated with suicide, due to interaction
ties from an older intravenous heroin user. As with AUDs and IMU,
between the effects of drinking and other health factors such as
clinical features of illegal drug use in older people may be atypical
depressive symptoms, medical illness, negatively perceived health
and masked by other conditions, and problems are likely to arise at
status, and low social support (Blow et al., 2004). In a Swedish ret-
lower levels of drug use, due to reduced physiological reserve, phar-
rospective case-control study, alcohol dependence or misuse as
macokinetic changes, and comorbid physical and neuropsychiatric
defined by DSM-IV was present in 35% of male and 18% of female
conditions.
suicides, in comparison to only 2% of male and 1% of female con-
trols (Waern, 2003).
It must also be borne in mind that, even in the absence of a Clinical Assessment, Investigations,
clinically diagnosable AUD, the neuropsychiatric effects of alco- and Screening
hol intoxication (emotional changes, disinhibition, impulse
dyscontrol, impaired judgement, and increased propensity for
Clinical assessment, investigations and screening
violent self-harm) may increase suicide risk: alcohol intoxica- in AUDs
tion may be part of the individual’s suicide plan, or intoxication Clinical assessment
may lead on to suicidal behaviour, particularly if the individual The assessment of AUDs in older people should be based on
is currently experiencing psychological distress (O’Connell and the standard clinical interview, mental state examination, physi-
Lawlor, 2005a). cal examination, and collateral history, if available and with the
patient’s consent. The standard interview and examination should
Clinical features and comorbidity in IMU in older be supplemented by specific questions relating to alcohol use,
people keeping in mind the potentially subtle and atypical nature of
AUDs in older people. Questions should be framed in a sensitive
Clinical features and comorbidities associated with IMU in older
and nonjudgemental way, as patients may disengage and be lost to
people will vary widely depending on the drug being used and
treatment and follow-up if they feel threatened by the assessment
patient characteristics, such as age, gender, and presence of other
procedure.
physical and neuropsychiatric problems. An outline of clinical fea-
The history should involve questions about current and past levels
tures and comorbidities is given in Table 49.13. The most common
of alcohol use, frequency and quantity of intake, the types of alco-
reported adverse consequence of inappropriate medication use has
holic beverage consumed, and the context in which drinking takes
been increased risk of falls, usually due to long-acting benzodi-
place. Changes in levels of intake may be significant, and should be
azepines and medications with anticholinergic properties (Berdot
interpreted in the context of the overall history. For example, while
et al., 2009).
older individuals may have reduced their level of alcohol intake in
recent years, suggesting that they have no current AUD, this reduc-
Clinical features and comorbidity of illicit drug use in tion may have occurred due to decreased tolerance to the effects of
older people alcohol on a background history of a chronic AUD, and the person
As outlined in the section Epidemiology, illicit drug use is uncom- may continue to experience alcohol-related problems even at this
mon in current and past cohorts of older people and there is thus lower level of intake.
a dearth of information on clinical features and comorbidity. Such Specific questions relating to the features of alcohol intoxication
features will also vary widely depending on the drug in question (Table 49.2), harmful use (Table 49.3), alcohol dependence (Tables
and the mode of administration. For example, an older smoker of 49.4 and 49.5), and a previous history of alcohol withdrawals
(Table 49.6) should be asked if indicated by the initial history and The main methods of screening used in older people involve
examination, along with specific questions relating to AUDs in self-report screening measures, such as the CAGE (Ewing, 1984)
older people (see Table 49.10). Concomitant medication and illegal and the Alcohol Use Disorders Identification Test (AUDIT; Saunders
drug use should be recorded. A family history of psychiatric illness et al., 1993), and older-specific versions of self-report instruments,
should be elicited, focusing especially on AUDs and mood disor- such as the Michigan Alcohol Screening Test – Geriatric Version
ders. Relevant aspects of the personal and social history include (MAST-G; Blow, 1991), along with biophysical measures, such as
early life experiences and exposure to AUDs in others, age of onset blood tests checking mean corpuscular volume and liver function
of drinking, occupational record, relationship history, and legal or tests. In practice, different screening measures are often used in
forensic problems. the assessment of an individual. It is important to point out that
The mental state examination should take into account level screening instruments are designed to detect at-risk individuals,
of alertness and a preliminary cognitive assessment or cognitive but they are not diagnostic tools in themselves, their use may yield
screening measure (e.g. Mini-Mental State Examination: Folstein false-positives and false-negatives, and they should not serve as a
et al., 1975); current mood state; any evidence of psychosis; ideas substitute for a thorough clinical interview and examination.
of hopelessness, suicide, or deliberate self-harm; and psychiatric A systematic review of self-report alcohol screening instru-
manifestations of alcohol withdrawal (Table 49.6). ments in older people (O’Connell et al., 2004) found that the CAGE
The physical examination should include recording of vital signs (Ewing, 1984) was the most widely studied instrument, followed by
such as blood pressure, pulse, temperature, and respiratory rate. the MAST (Selzer, 1968) and variations of the MAST, the AUDIT
Objective evidence of acute alcohol withdrawals should be recorded. (Saunders et al., 1993) and variations of the AUDIT, and others.
A more detailed physical examination should be performed if indi- Sensitivity and specificity of self-report alcohol screening instru-
cated, focusing on the different organ systems involved in AUDs ments was found to vary widely, depending on the prevalence of
(see Table 49.10). AUDs in the population under study, the clinical characteristics
A collateral history is also very valuable for the overall assess- of the population, and the type of AUD being detected. The Cyr–
ment, as people may underestimate their level of alcohol intake and Wartman (Cyr and Wartman, 1988) and CAGE questionnaires are
associated AUD either deliberately, because of embarrassment and the briefest tests and can be administered in less than 30 s, making
reluctance to engage in treatment, or because of faulty recall related these screening instruments the easiest to use. Far more research
to cognitive impairment. has focused on the CAGE, and its sensitivity and specificity in older
people are superior to those of the Cyr–Wartman questionnaire.
Investigations
Despite the advantage of ease of use of the CAGE, deficiencies
Following a detailed history and examination, other investigations
were highlighted in a number of studies of older populations, and
may be indicated and should be directed by the patient’s clinical
the AUDIT-5 (Philpot et al., 2003), another brief test, may prove
status. For example, an individual admitted for treatment of alcohol
to be more useful in older people with psychiatric illness. The
withdrawals will require blood tests to check at least the following:
Alcohol-Related Problems Survey (ARPS) and its shortened ver-
urea and electrolytes; full blood count; liver function tests; vitamin
sion (shARPS) (Fink et al., 2002) may prove to be more useful in
B12 and folate levels. Further investigations may also be indicated,
older people with medical comorbidity, but their use is limited by
depending on the clinical findings and the organ system involved
the fact that they take, respectively, 10 and 9 min to apply.
(Table 49.10). These may include the following: neuroimaging (CT
A recent Finnish study (Aalto et al., 2011) examining the utility
or MRI brain); gastrointestinal investigations such as ultrasound,
of the AUDIT and its derivatives in screening for heavy drinking in
CT, or MRI examinations of the abdomen, upper gastrointestinal
older people found that the AUDIT and AUDIT-C are accurate if
endoscopy, and liver biopsy; basic cardiovascular investigations
the cut points are tailored to this age group (cut points of 4 or more
such as electrocardiogram and other more detailed investiga-
on AUDIT-C and 5 or more on AUDIT).
tions if indicated, e.g. echocardiogram and 24-h blood pressure
In order to use AUD screening in routine clinical practice,
monitoring.
the demographic characteristics of the target population are an
Further psychological investigations include detailed neuropsy-
important consideration. For example, routinely screening all
chological assessment of cognition if impairment is detected
older women attending an active retirement centre may yield few
in the initial assessment (as outlined in Clinical assessment).
cases of AUDs, but it may prove to be a costly and intrusive exer-
Psychological assessment may also be required to assess for patient
cise. However, screening of high-risk populations, such as older
suitability and level of motivation for addiction counselling or other
medical and psychiatric inpatients (see section Epidemiology) is
psychotherapeutic interventions.
advised, as this practice may yield high levels of AUDs that direct
Social investigations include more detailed collateral histories if
the development of appropriate and much-needed treatment serv-
available, along with an assessment of financial status, social sup-
ices. Furthermore, ease of use, patient acceptability, and sensitivity
ports, and accommodation arrangements.
and specificity of the screening instrument must all be taken into
Screening consideration.
As already alluded to throughout this chapter, screening for AUDs As with self-report screening instruments, the utility of biophysi-
in older people should focus on the broad spectrum of such prob- cal screening measures such as carbohydrate-deficient transferrin,
lems and not merely clear-cut cases such as the ‘down and out’ liver function tests, or the mean corpuscular volume may be less
alcoholic. Screening should also aim to detect the subtler but also reliable in older people (Luttrell et al., 1997), because of higher lev-
damaging effects of AUDs in older people, such as drinking moder- els of comorbid physical illnesses leading in themselves to abnor-
ately while taking medications that interact with alcohol. mal results.
Clinical assessment, investigations, and screening for except for those with particularly high risk, such as the homeless
IMU in older people and prison inmates, or people already known to have a history of
AUDs and IMU.
Clinical assessment
As with AUDs, a standard clinical assessment involving a history,
mental state and physical examination, and collateral history should Management and Prevention
be performed. Prescription drug abuse of benzodiazepines and opi- Management and prevention can be broadly divided into primary
ates can be difficult to detect in older people, and effects of abuse prevention, secondary prevention, and tertiary prevention or treat-
such as irritability, depression, drowsiness, and memory lapses can ment. Primary prevention refers to the prevention of problems aris-
be falsely attributed to normal ageing. A list of all prescribed and ing for the first time (e.g. late-onset AUDs); secondary prevention
OTC medications being used, along with their indications for use, aims to prevent the onset or worsening of problems in at-risk indi-
should be recorded. Ideally, patients should be asked to bring with viduals (e.g. individuals with ‘heavy use’ of alcohol); and tertiary
them all medications in their containers, as this will also give an indi- prevention refers to treatment for established problems (e.g. indi-
cation as to levels of adherence or compliance. Any reported adverse viduals with established alcohol dependence syndrome).
effects should be recorded, along with symptoms and signs indicat-
ing underuse, overuse, or intermittent use of medication. Clinical Moderate alcohol consumption and health
features will vary depending on the medications being used. In considering the prevention of problems of alcohol misuse, it
Investigations is important to be aware of the complex relationship between
moderate alcohol consumption and health. Up to a hundred epi-
Blood levels of some prescribed medications may be checked in
demiological studies conducted over the past three decades have
order to assess levels of compliance and to establish if the blood
demonstrated that, in relation to physical, social, psychiatric, and
level is within the therapeutic window for the drug in question (e.g.
cognitive health and wellbeing, light–moderate drinkers seem to
lithium, carbamazepine). Other biophysical measures may also be
be generally healthier than both nondrinkers and heavy drinkers or
indicated that provide proxy measures of medication compliance,
those with AUDs, a phenomenon that has been referred to as the J-
such as random or fasting glucose levels and levels of glycosylated
or U-shaped curve (Doll et al., 1994; O’Connell and Lawlor, 2005b).
haemoglobin, to assess for level of diabetes control and compliance
As a result, it has been argued that light–moderate consumption of
with hypoglycaemic agents or insulin.
alcohol, particularly red wine (the so-called French paradox), has
Screening beneficial effects on health. Furthermore, there is some empirical
There are no routinely used screening measures for IMU in older evidence linking light–moderate alcohol intake and an improved
people. Clinicians should have a high index of suspicion for pre- profile of cardiovascular biomarkers (Rimm et al., 1999).
scription abuse where an older person is on benzodiazepines or However, there are a number of potential problems with the
opiates and has ongoing symptoms of pain, depression, or anxiety, epidemiological evidence. Nondrinkers are a heterogenous group,
or is complaining of adverse side effects such as drowsiness, falls, comprised of both lifelong nondrinkers and individuals with a
memory loss, and confusion. The use of a measure such as Beers’ past history of AUD (so-called sick quitters), and the latter group
criteria (Beers, 1997) may be a useful addition to the overall assess- may be less healthy than moderate drinkers for reasons other than
ment of an older person if potentially inappropriate use of medica- their current drinking habits (Shaper, 1995). The ‘French paradox’,
tion is suspected. In a study updating Beers’ criteria (Fick et al., whereby moderate intake of red wine has been proposed as the fac-
2003), a modified Delphi method was used, which is a set of proce- tor responsible for the lower levels of cardiovascular disease in the
dures and methods for formulating a group judgement for a subject French compared to British and American populations, has been
matter in which precise information is lacking. Identified were 48 questioned (Law and Wald, 1999). Moderate drinkers may also
individual medications or classes of medications to avoid in older have other favourable physical (Shaper, 1995) and psychological
adults and 20 diseases/conditions and medications to be avoided health characteristics, such as differing personality type (O’Connell
in older adults with these conditions (Tables 49.14 and 49.15). et al., 2005), in comparison to nondrinkers.
Sixty-six of the inappropriate medications identified were consid- Furthermore, no controlled trials of light–moderate alcohol
ered by the panel to have adverse outcomes of high severity. The intake and effects on cardiovascular or other health characteristics
STOPP (Screening Tool of Older Persons’ Potentially Inappropriate have been performed and such trials are likely to encounter ethical
Prescriptions) has recently been demonstrated to be superior to barriers due to the neuropsychiatric effects of alcohol and the exist-
Beers’ criteria when applied to acutely ill older medical inpatients ence of proven treatments in the form of conventional medication.
(Gallagher and O’Mahony, 2008). Therefore, despite the epidemiological and biochemical evidence
for potential health benefits of light–moderate alcohol intake,
Clinical assessment, investigations, and screening for there is insufficient evidence at present to support advising mod-
illicit drug use in older people erate alcohol intake for the entire population, particularly when
As with AUDs and IMU, a thorough history of any current or past the many potential harmful effects of excessive alcohol intake are
use of illicit drugs should be recorded, along with mental state and considered.
physical examinations and collateral history, if available. Further
investigations will be directed by the type of drug or drugs used, the
Management and prevention of AUDs
route of administration, and the clinical findings. Primary prevention
The low levels of illicit drug use in current generations of older Primary prevention aims to prevent the development of de novo
people mean that screening is unwarranted in most populations AUDs in older people. Primary prevention of AUDs in older people
Table 49.14 Criteria for potentially inappropriate medication use in older adults: independent of diagnoses and conditions (Fick et al., 2003)
Drug Concern Severity rating
(high or low)
Propoxyphene (Darvon) and combination products (Darvon Offer few analgesic advantages over acetaminophen, yet have the adverse Low
with ASA, Darvon-N, Darvocet-N) effects of other narcotic drugs.
Indomethacin (Indocin, Indocin SR) Of all available NSAIDs, this drug produces the most CNS adverse effects. High
Pentazocine (Talwin) Narcotic analgesic that causes more CNS adverse effects, including confusion High
and hallucinations, more commonly than other narcotic drugs. Additionally, it is
a mixed agonist and antagonist.
Trimethobenzamide (Tigan) One of the least effective antiemetic drugs, yet it can cause extrapyramidal High
adverse effects.
Muscle relaxants and antispasmodics: methocarbonal Most muscle relaxants and antispasmodic drugs are poorly tolerated by High
(Robaxin), carisoprodal (Soma)chlorozoxazone (Paraflex), older patients, since these cause anticholinergic adverse effects, sedation, and
meaxalone (Skelaxin), cyclobenzaprine (Flexeril), oxybutynin weekness. Additionally, their effectiveness at doses tolerated by older patients is
(Ditropan). Do not consider the extended-release Ditropan XL. questionable.
Flurazepam (Dalmane) This benzodiazepine hypnotic has an extremely long half-life in older patients High
(often days), producing prolonged sedation and increasing the incidence of falls
and fracture. Medium- or short-acting benzodiazepines are preferable.
Amitriptyline (Elavil), chlorodiaazepaxide-amitriptyline Because of its strong anticholinergic and sedation properties, amitriptyline is High
(Limbitrol), perphenazine-amitriptyline (Triavil) rarely the antidepressant of choice for older patients.
Doxepin (Sinequan) Because of its strong anticholinergic and sedation properties, doxepin is rarely High
the antidepressant of choice for older patients.
Meprobamate (Miltown, Equanil) This is a highly addictive and sedating anxiolytic. Those using meprobamate High
for prolonged periods may become addicted and may need to be withdrawn
slowly.
Doses of short-acting benzodiazepines (doses greater than): Because of increased sensitivity to benzadiazepines in older patients, smaller High
larazepam (Ativan) 3 mg; oxazepam (Serax) 60 mg; alprozolam doses may be effective as well as safer. Total daily doses should rarely exceed the
(Xanax) 2 mg; tenazepam (Restoril) 15 mg; triazolam (Halcion) suggested maximums.
0–25 mg
Long-acting benzodiazepines: chlorodiazepoxide Have a long half-life in older patients (often several days), producing High
(Libriurn), chlordiazepoxide-amitriptyline (Limbitrol) prolonged sedation and increasing the risk of falls and fractures. Short- and
clidinium-chlorodiazepaxide (Librax), diazepam (Valium), intermediate-acting benzodiazepines are preferred if a benzodiazepine is
quazepam (Doral), halasepam (Paxipam), chlorazepate required.
(Tranxene)
Disopyraximide (Norpace, Norpace CR) Of all the antiarrhrythmic drugs, this is the most potent negative inotrope High
and therefore may induce heart failure in older patients. It is also strongly
anticholinergic. Other antiarrhythmic drugs should be used.
Digoxin (Lanoxin) (should not exceed > 0.125 mg/day except Decreased renal clearance may lead to increased risk of toxic effects. Low
when treating atrial arrhythmias)
Short-acting dipyridamole (Persantine). Do not consider the May cause orthostatic hypotension. Low
long-acting dipyridamole (which has better properties than the
short-acting in older adults) except with patients with artificial
heart valves
Methyldopa (Aldomet) and methyldopa-hydrochlonothiazide May cause bradycardia and exacerbate depression in older patients. High
(Aldoril)
Reserpine at doses > 0.25 mg May induce depression, impotence, sedation, and orthostatic hypotension. Low
Chlorpropamide (Diabinese) Has a prolonged half-life in older patients and could cause prolonged High
hypoglycaemia. Additionally, it is the only oral hypoglycaemic agent that causes
SIADH.
Gastrointestinal antispasmodic drugs: dicyclomine Highly anticholinergic and have uncertain effectiveness. These drugs should be High
(Bentyl), hyoscyamine (Levsin and Levsinex), propantheline avoided (especially for long-term use).
(Pro-banthine), belladonna alkaloids (Donnatal and others),
clidinium-chlordiazepoxide (Librax)
(Continued)

(high or low)
Anticholinergics and antihistamines: chloropheniramine All nonprescription and many prescription antihistamines may have potent High
(Chlor-Trimetan), diphenhydramine (Banadryl), hydroxyzine anticholinergic properties. Nonanticholinergic antihistamines are preferred in
(Vistaril and Atarax), cyproheptadine (Periactin), promethazine older patients when treating allergic reactions.
(Phenergan), tripellannamine, dexchlorpheniramine
(Polaramine)
Diphenhydramine (Benadryl) May cause confusion and sedation. Should not be used as a hypnotic, and when High
used to treat emergency allergic reactions, it should be used in the smallest
possible dose.
Ergot mesyloids (Hydergine) and cyclandelate (Cyclospasmol) Have not been shown to be effective in the doses studied. Low
Ferrous sulphate more than 325 mg/day Doses more than 325 mg/day do not dramatically increase the amount Low
absorbed but greatly increase the incidence of constipation.
All barbiturates (except phenobarbital) except when used to Highly addictive and cause more adverse effects than most sedative or hypnotic High
control seizures drugs in elderly patients.
Meperidine (Demerol) Not an effective oral analgesic in doses commonly used. May cause confusion High
and has many disadvantages compared with other narcotic drugs.
Ticlopidine (Ticlid) Has been shown to be no better than aspirin in preventing clotting and may be High
considerably more toxic. Safer, more effective alternatives exist.
Ketorolac (Toradal) Immediate and long-term use should be avoided in older persons, since a High
significant number have asymptomatic GI pathologic conditions.
Amphetamines and anorexic agents Have potential for causing dependence, hypertension, angina, and myocardial High
infarction.
Long-term use at full dosage, longer half-life, non-COX-selective Have the potential to produce GI bleeding, renal failure, high blood pressure, High
NSAIDs: naproxen (Naprosyn, Avaprox, Aleve), oxaprozin and heart failure.
(Daypro), piroxicam (Feldene)
Daily fluoxetine (Prozac) Long half-life of drug and risk of producing excessive CNS stimulation, sleep High
disturbances, and increasing agitation. Safer alternatives exist.
Long-term use of stimulant laxatives: bisacodyl (Dulcolax), May exacerbate bowel dysfunction. High
cascara sagrada, and Neoloid except in the presence of opiate
analgesic use
Amiadarone (Cordarone) Associated with QT interval problems and risk of provoking torsades de pointes. High
Lack of efficacy in older adults.
Orphenadrine (Norflex) Causes more sedation and anticholinergic adverse effects than safer alternatives. High
Guanethidine (Ismelin) May cause orthostatic hypotension. Safer alternatives exist. High
Guanadrel (Hyloral) May cause orthostatic hypotension. High
Cyclandelate (Cyclospasmol) Lack of efffcacy. Low
Isoxsurpine (Vasodilan) Lack of efffcacy. Low
Nitrofurantoin (Macrodantin) Potential for renal impairment. Safer alternatives are available. High
Doxazosin (Cardura) Potential for hypotension, dry mouth, and urinary problems. Low
Methyltestosterone (Android, Virilon, Testrad) Potential for prostatic hypertrophy and cardiac problems. High
Thioridazine (Mellaril) Greater potential for CNS and extrapyramidal adverse effects High
Mesoridazine (Serentile) CNS and extrapyramidal adverse effects. High
Short-acting nifedipine (Procardia, Adalat) Potential for hypotension and constipation. High
Clondinine (Catapres) Potential tor orthostatic hypotension and CNS adverse effects. Low
Mineral oil Potential for aspiration and adverse effects. Safer alternatives are available. High
Cimetidine (Tagamet) CNS adverse effects including confusion. Low
Ethacrynic acid (Edecrin) Potential for hypertension and fluid imbalances. Safer alternatives are available. Low
(Continued)

(high or low)
Desiccated thyroid Concerns about cardiac effects. Safer alternatives are available. High
Amphetamines (excluding methylphenedate hydrochloride CNS stimulant adverse effects. High
and anorexics)
Oestrogens only (oral) Evidence of the carcinogenic (breast and endometrial cancer) potential of these Low
agents and lack of cardioprotective effect in older women.
CNS, central nervous system; COX, cyclo-oxygenase; GI, gastrointestinal; NSAIDs, nonsteroidal anti-inflammatory drugs; SIADH, syndrome of inappropriate antidiuretic hormone secretion.
Table 49.15 Criteria for potentially inappropriate medication use in older adults: considering diagnoses or conditions (Fick et al., 2003)
Disease or condition Drug Concern Severity rating
(high or low)
Heart failure Disopyramide (Norpace), and high sodium content drugs Negative inotropic effect. Potential to promote High
(sodium and sodium salts (alginate bicarbonate, biphosphate, fluid retention and exacerbation of heart failure.
citrate, phosphate, salicylate, and sulphate))
Hypertension Phenylpropanolamine hydrochloride (removed from the May produce elevation of blood pressure High
market in 2001), pseudoephedrine; diet pills, and amphetamines secondary to sympathomimetic activity.
Gastric or duodenal ulcers NSAIDs and aspirin (>325 mg) (coxibs excluded) May exacerbate existing ulcers or produce new/ High
additional ulcers.
Seizures or epilepsy Clozapine (Clozaril), chlorpromazine (Thorazine), thioridarine May lower seizure thresholds. High
(Mellaril), thiothixene (Navane)
Blood clotting disorders Aspirin, NSAIDs, dipyridamole (Persantin), ticlopidine (Ticlid), May prolong clotting time and elevate INR High
or receiving anticoagulant clopidogrel (Plavix) values or inhibit platelet aggregation, resulting
therapy in an increased potential for bleeding.
Bladder outflow Anticholinergics and antihistamines, gastrointestinal May decrease urinary flow, leading to urinary High
obstruction antispasmodics, muscle relaxants, oxybutynin (Ditrapan), retention.
flavoxate (Urispas), anticholinergics, antidepressants,
decongestants, tolterodine (Detrol)
Stress incontinence α-Blockers (Doxazosin, Prazosin, Terazosin), anlicholinergics, May produce polyuria and worsening of High
tricylic antidepressants (imipramine hydrochloride, doxepin incontinence.
hydrochloride, amitriptyline hydrachloride), long-acting
benzodiazepines
Arrhythmias Tricyclic antidepressants (imipramine hydrochloride, doxepin Concern due to proarrhythmic effects and High
hydrochloride, amitriptyline hydrochloride) ability to produce QT interval changes.
Insomnia Decongestants, theophylline (Theodur), methylphenidate Concern due to CNS stimulant effects. High
(Ritalin), MAOIs, and amphetamines
Parkinson’s disease Metoclopramide (Reglan), conventional antipsychotics, tacrine Concern due to their antidopaminergic/ High
(Cognex) cholinergic effects.
Cognitive impairment Barbiturates, anticholinergics antispasmodics, muscle Concern due to CNS-altering effects. High
relaxants. CNS stimulants: dextroamphetamine (Adderall),
methylphenidate (Ritalin), methylphentamine (Desoxyn),
pemolin
Depression Long-term benzodiazepine use. Sympatholytic agents: May produce or exacerbate depression. High
methyldopa (Aldomet), reserpine, guanethidine (Ismelin)
Anorexia and malnutrition CNS stimulants: dextroamphetamine (Adderall), Concern due to appetite-suppressing effects. High
methylphenidate (Ritalin), methamphetamine (Desoxyn),
pemolin, fluoxetine (Prozac)
Syncope or falls Short- to intermediate-acting benzodiazepine and tricyclic May produce ataxia, impaired psychomotor High
antidepressants (imipramine hydrochloride, doxepin function, syncope, and additional falls.
hydrochloride, amitriptyline hydrochloride)
(Continued)

Disease or condition Drug Concern Severity rating
(high or low)
SIADH/hyponatraemia SSRIs: fluoxetine (Prozac), citalopram (Celexa), fluvoxamine May exacerbate or cause SIADH. Low
(Luvox), paroxetine (Paxil), sertraline (Zoloft)
Seizure disorder Bupropion (Wellbutrin) May lower seizure threshold. High
Obesity Olanzapine (Zyprexa) May stimulate appetite and increase weight Low
gain.
COPD Long-acting benzodiazepines: chlordiazepoxide CNS adverse effects. May induce respiratory High
(Librium), chlordiazepoxide-amitriptyline (Limbitrol), depression. May exacerbate or cause respiratory
clidinium-chlordiazepoxide (Librax), diazepam (Valium), depression.
quazepam (Doral), halazepam (Paxiparn), chlorozepate
(Tranxene). β-blockers: propranolol
Chronic constipation Calcium channel blockers, anticholinergics, tricyclic May exacerbate constipation. Low
antidepressant (imipramine hydrochloride, doxepin
hydrochloride, amitriptyline hydrochloride)
CNS, central nervous system; COPD, chronic obstructive pulmonary disease; INR, international normalized ratio; MAOIs, monoamine oxidase inhibitors; NSAIDs, nonsteroidal anti-
inflammatory drugs; SIADH, syndrome of inappropriate antidiuretic hormone secretion; SSRIs, selective serotonin reuptake inhibitors.
is important, considering that up to 1 in 3 older people with AUDs ward, particularly if there is a history of alcohol withdrawal sei-
may develop such problems for the first time in later life (Adams zures or delirium tremens. Fluid and electrolyte imbalances should
and Waskel, 1991). Clinicians should watch for the development of be corrected and cognitive state should be monitored regularly in
AUDs when an older person encounters stressful life circumstances view of the risk of developing delirium.
and major changes or losses, particularly if that individual has a Care should be taken with benzodiazepine-assisted with-
personal or family history of AUDs. drawal in older people, in view of the elevated risk of overseda-
Primary prevention can also be seen as a strategy directed at the tion, confusion, and falls. There are no older-specific guidelines on
entire population, targeting factors such as ease of access to alcohol, benzodiazepine-assisted alcohol withdrawal. However, it has been
alcohol pricing, restrictions on alcohol advertising, and education established that older people have more severe alcohol withdrawals
about the adverse effects of drinking. Such primary prevention and and receive higher doses of chlordiazepoxide as a result (Liskow
public health initiatives tend to be directed towards younger indi- et al., 1989). Lorazepam has been identified as the safest choice of
viduals, but they should also take into account the more clinically benzodiazepine for treatment of alcohol withdrawal in older peo-
‘silent’ AUDs that may develop in older people (O’Connell et al., ple, in view of the fact that advancing age and liver disease have lit-
2003b). tle impact on its metabolism, and absorption by the intramuscular
Secondary prevention route is predictable (Peppers, 1996). Use of an objective measure
of alcohol withdrawal, such as the Clinical Institute Withdrawal
Secondary prevention strategies should focus on older people who
Assessment for Alcohol-Revised Version (CIWA-Ar), is advisable
already have ‘at-risk’ drinking, either currently or in the past, and
in defining the severity of alcohol withdrawal and monitoring clini-
who are at risk of developing worsening problems in the context of
cal course, although this scale is not older-specific and areas such
diverse factors such as bereavement, social isolation, adjustment to
as ‘orientation and clouding of sensorium’ may be disproportion-
retirement, and physical or psychiatric health problems. As with
ately affected in older people, especially if cognitive impairment is
primary prevention, there should be a high index of clinical suspi-
present (Naranjo and Sellers, 1986) (Table 49.16).
cion when assessing such people. There is emerging evidence that
Parenteral or oral thiamine should be given to prevent develop-
formalized brief psychological interventions in primary care may
ment of the Wernicke–Korsakoff syndrome. A recent review has
have a positive impact on AUDs in older people (Fleming et al.,
concluded that, in the emergency department setting, oral thia-
1999; Blow and Barry, 2000). However, older people should also
mine administration is as effective as parenteral administration
be referred on to specialist mental health and addiction services if
(Jackson and Teece, 2004). However, there are no older-specific
problems persist.
guidelines, and individual patient characteristics must be taken
Tertiary prevention into account, such as general health, ability to take oral medication,
Tertiary prevention involves treatment of existing AUDs. Treatment and compliance.
modalities can be divided into biological/medical, social, and psy- There is limited evidence available on the use of abstinence
chological. Biological/medical treatments are most important in medications such as Disulfiram, Naltrexone, and Acamprosate in
the acute setting, where detoxification may be required. In view older people, and they are probably best avoided in view of the
of increased physical frailty and evidence for more severe alcohol elevated risk of adverse effects. However, there is some evidence
withdrawals in older people (Brower et al., 1994), it is advisable to that Naltrexone may be safe and effective in treating AUDs in older
admit older people requiring detoxification, preferably to a medical people (Oslin et al., 1997a, 1997b).
Table 49.16 The Clinical Institute Withdrawal Assessment for clinical assessment should be discouraged. Physicians should be
Alcohol-Revised Version (CIWA-Ar) aware of the potential for prescription drug abuse among older
patients, particularly for opiate abuse and dependence, which has
Items 1–9 are scored from 0–7 and item 10 from 0–4. Maximum possible
been steadily rising in the older age group, and should consider
score is 67.
these problems when evaluating an older person.
1. Nausea and vomiting Community pharmacists should have an active role in advis-
2. Tremor ing on the appropriate use of prescription and OTC medications,
3. Paroxysmal sweats in cautioning patients on inappropriate use and on interactions
4. Anxiety with alcohol and other medications, and in alerting the prescrib-
5. Agitation ing physician when abuse, overuse, underuse, or inappropriate use
6. Tactile disturbances of medication (either iatrogenically or due to patient behaviour)
7. Auditory disturbances
is suspected. Older patients should be advised at all times on the
use of the appropriate dose of medication and on the dangers of
8. Visual disturbances
escalating the dose without medical supervision, particularly ben-
9. Headaches and fullness in head
zodiazepines and opiates.
10. Orientation and clouding of sensorium Patients should be educated about all aspects of their medication,
Severity of alcohol withdrawal including physical descriptions of the medication, the clinical indi-
Mild: 10 cations, dose, and frequency and common side effects. If patients
are unable to manage administration of their own medication due
Moderate: 10–20 to cognitive or other impairments, provision should be made for a
Severe: 20+ family member or caregiver to arrange this.
(From Taylor et al., 2005.)
Several physical and cognitive impairments impact on the ability
of an older person to open medicine containers. One study of older
people aged 81 years and older, living in the community and in
institutions in Sweden (Beckman et al., 2005), found that 14% were
Social aspects of treatment include identifying and addressing unable to open a screw-cap bottle, 32% a bottle with a snap lid, and
problems in such diverse areas as personal finances, housing, employ- 10% a blister pack. Less than half of those who were unable to open
ment, and levels of social contacts, as continuing problems in these one or more of the containers received help with their medication,
areas may serve to perpetuate the AUD. Psychological treatments and only 27% of those living in their own homes received help.
include specific psychotherapies focusing on AUDs (e.g. motivational Dosette boxes may be useful, but it has been pointed out that
interviewing, addiction counselling) and psychotherapies aimed at those patients most in need of them are the least likely to be able to
comorbid mood or other psychiatric disorders. While there is lit- manage them, and problems may occur with both filling the devices
tle evidence on the effectiveness of psychotherapeutic approaches to and the taking of medications from them (Levings et al., 1999). One
addiction in older people, one review has concluded that only those study found that the use of a combination pack for medication used
studies involving behavioural and cognitive-behavioural interven- to treat osteoporosis was associated with improved understanding
tions have provided empirical support for treatment effectiveness of medication directions and improved patient satisfaction (Ringe
(Schonfeld and Dupree, 1995). There is also some evidence that et al., 2006). A more recent study (Zedler et al., 2011) suggests that
older people may respond better to psychotherapy in same-age set- calendar packaging of medication through use of calendar blister
tings, i.e. among other older people (Kofoed et al., 1987; Schonfeld packaging (CBP) and calendar pill organizers (CPO), especially in
and Dupree, 1995), presumably because of a shared experience of combination with education and reminder strategies, may improve
the older-specific aspects of AUDs. medication adherence.
The wide variety of older-specific aspects of AUDs, along with Active management of physical and psychiatric conditions also
a projected increase in the numbers of older people with AUDs helps in primary prevention of IMU. For example, adequate treat-
in future years, and evidence for improved treatment response in ment of depression and anxiety should lead to a reduced risk of
same-age settings mean that the development of screening pro- benzodiazepine overuse, and adequate management of pain should
grammes and treatment facilities geared towards older people, which lead to a reduced risk of overuse or abuse of opiates and other anal-
may be based on population-defined sectors, is now needed. Such gesics. Criteria such as those of Beers (Beers, 1997; Fick et al., 2003;
screening and treatment programmes should involve collaboration Tables 49.14 and 49.15) should be taken into account when consid-
between old age psychiatry and geriatric medicine and a multidisci- ering the most appropriate medication to prescribe, and the ones
plinary team should deliver treatment, targeting the many medical, best avoided, for a particular condition in an older person.
social, and psychological aspects of AUDs in older people.
Secondary prevention
Management and prevention of IMU Secondary prevention of IMU in older people should focus on
Primary prevention those with a past history of IMU, and the medical conditions and
medications listed among Beers’ criteria in Tables 49.14 and 49.15.
Along with patients themselves, healthcare workers, family mem-
bers, and carers all have important roles in the primary and sec- Tertiary prevention
ondary prevention of IMU in older people. Prescriptions should be Tertiary prevention of IMU in older people will depend on the
reviewed regularly with a view to simplification and rationalization medication in question and the clinical and sociodemographic pro-
if possible, and the practice of giving ‘repeat prescriptions’ without file of the patient. Admission to a medical or psychiatric ward may
be required to facilitate reduction or stopping of certain medica- Comorbid psychiatric illness, gender (with possibly poorer prog-
tions, e.g. benzodiazepine and opiate detoxification, as outpatient nosis in men), levels of social supports, and the availability of alter-
detoxification in older people may be hazardous. native and healthier social outlets (which may in turn be related to
factors such as culture and ethnicity) may influence motivation for
Management and prevention of illicit drug use individuals to change their drinking habits.
The relative rarity of illicit drug use in older people means that there In view of the wide range of older-specific aspects of AUDs, it
is a dearth of information and guidelines on primary and second- makes intuitive sense that older people would be best treated by
ary prevention. However, it is predicted that by 2020, the number specialists in same-age settings, among other older people who
of people over the age of 50 needing substance abuse treatment may share similar problems, and there is evidence to suggest that
will double (Han et al., 2009). The ageing ‘baby-boomer’ genera- such an approach is associated with a better outcome (Kofoed et al.,
tion in the US has been cited as a potential source of older illegal 1987). However, despite the mounting evidence for the extent of
drug users (Patterson and Jeste, 1999) and so clinical experience in AUDs in older people, older-specific treatment settings are few in
this area and the need for primary prevention strategies focused on number. Furthermore, engaging with such services may be depend-
both the individual and older people in general will need to expand ent on an intuitive and motivated physician with sufficient training
in the future. and expertise to identify the AUD and treat or refer appropriately.
Secondary prevention strategies are likely to focus on those with
a past history of illegal drug use, medical (e.g. pain) and psychiatric Prognosis in IMU
(e.g. AUDs) disorders that may increase the risk for illegal drug use, Similar prognostic indicators that apply to AUDs are likely to be
and social and environmental factors. relevant to IMU, and centre on the individual’s clinical and sociode-
Tertiary prevention or treatment will depend on the drug in mographic characteristics, levels of support, and available services.
question and the clinical and sociodemographic characteristics of The duration of inappropriate use and abuse and the medication or
the patient. medications in question are also of relevance. For example, an older
person who has been overusing benzodiazepines for decades is
Prognosis more likely to encounter adverse effects and difficulties with cessa-
tion than someone who has been overusing mild analgesics because
Prognosis in AUDs of a recent worsening of arthritic pain. As with AUDs, the relation-
The available literature on the topic suggests that older people are at ship with the individual’s physician, and the ability of the physician
least as likely, if not more likely, to benefit from treatment of AUDs to identify and treat or refer appropriately, is vital. Likewise, moti-
as younger people (Curtis et al., 1989; Oslin et al., 2002). However, vation to address the IMU may be related to the individual’s mental
prognosis in older people is likely to vary widely depending on a health, level of cognition, and social circumstances.
number of factors relating to individuals themselves and the nature
of their AUD, the presence of family and other support systems, and Prognosis in illicit substance use
the availability of treatment services, particularly services that are There is a dearth of literature in this area, but similar general prin-
tailored to older people. Individuals with late-onset AUDs are gen- ciples of good and poor prognostic indicators that relate to AUDs
erally felt to have a better prognosis than those with early-onset or and IMU are likely to apply to the use of illegal substances.
life-long AUDs (Babor et al., 1992). Better prognosis may be related
to the shorter history and milder severity of AUDs in this group,
more intact social supports, higher income levels, and the presence Smoking in Older People
of potentially modifiable precipitants such as depression, bereave- Although use of tobacco (primarily through cigarette smoking) may
ment reactions, and social isolation. In contrast, the older individual be classified as a mental and behavioural disorder due to psychoac-
with an early-onset or life-long AUD may have accumulated signifi- tive substance use and may fulfil criteria for harmful use, depend-
cant physical, psychiatric, cognitive, and social deficits that are more ence, and withdrawal states (Tables 49.2–49.6), and despite the fact
difficult to address. The higher prevalence of antisocial personality that nicotine use is arguably associated with more morbidity and
disorder and the higher risk of comorbid substance use in this group mortality in older people than alcohol and all other substance use
may also lead to increased difficulties in engagement with therapy. disorders (Atkinson, 2001), this particular problem receives rela-
Of particular relevance to the prognosis of AUDs in older peo- tively little attention in the psychiatric literature. This is perhaps
ple is the role of cognitive impairment, which is likely to act as a due to the fact that the neuropsychiatric effects of smoking are sub-
barrier to engaging with treatment. However, one study comparing tle and are not generally clinically significant, and there may be a
cognitively impaired and cognitively intact outpatients enrolled in perception that smoking palliates psychological distress. Indeed, a
an intensive treatment programme found no significant intergroup complex and circular relationship between depression, smoking,
differences in outcome; a wide range of treatment gains were seen and medical illness has been described (Wilhelm et al., 2004).
in both groups, albeit with a higher level of treatment drop-out in We know that significant proportions of older people (approxi-
the impaired group (Teichner et al., 2002). mately 10%) smoke (Bratzler et al., 2002), and this figure is likely
The level of cognitive impairment is likely to be higher in the to be higher again for older people with psychiatric disorders such
early-onset AUD group, but clinical experience suggests that there as depression (Covey et al., 1998). The health impact of smoking
may be a subgroup of individuals with late-onset AUDs whose is well documented elsewhere and includes malignancies (lung,
AUD arises as a result of cognitive impairment, with an increase in oesophageal, bladder, etc.), cardiovascular disease (ischaemic
alcohol intake a behavioural manifestation of an early dementing heart disease, cerebrovascular disease, peripheral vascular disease,
process. etc.), respiratory disease (chronic obstructive airway disease), and
countless other problems. Furthermore, the effects of smoking are References

cumulative and age-related and it has been estimated that 70% of Aalto, M., et al. (2011). The Alcohol Use Disorders Identification Test
the excess mortality attributed to smoking in the US occurs in those (AUDIT) and its derivatives in screening for heavy drinking among the
over the age of 60 (Burns, 2000). elderly. International Journal of Geriatric Psychiatry, 26(9), 881–5.
The key message from recent research on smoking in older people Adams, S.L. and Waskel, S.A. (1991). Late onset of alcoholism among older
is that smoking cessation is possible in this age group and is worthy Midwestern men in treatment. Psychology Report, 68, 432.
of active consideration in the individual clinical setting and in a wider Adams, W.L. and Cox, N.S. (1995). Epidemiology of problem drinking among
public health context, in view of the considerable health benefits that elderly people. International Journal of Addiction, 30, 1693–716.
are likely to accrue. LaCroix and Omenn (1992) found that older Adams, W.L., et al. (1990). Alcohol intake in the healthy elderly. Journal of the
smokers who quit have a reduced risk of death compared with cur- American Geriatrics Society, 38, 211–16.
rent smokers within 1–2 years of quitting, and overall risk of death Adams, W.L., et al. (1992). Alcohol abuse in elderly emergency department
patients. Journal of the American Geriatrics Society, 40, 1236–40.
approaches that of those who never smoked after 15–20 years of absti-
Alcoholism in the Elderly (1996). Council on Scientific Affairs, American
nence. Although the benefits of smoking cessation for longevity are Medical Association. Journal of the American Medical Association, 275,
most pronounced in younger people, a large study of smoking ces- 797–801.
sation has demonstrated that men over the age of 65 gained 1.4–2.0 Amdt, S., Clayton, R., and Schultz, S.K. (2011). Trends in substance abuse
years of life and women gained 2.7–3.7 years (Taylor et al., 2002). treatment 1998–2008: increasing older adult first-time admissions for
Systematic reviews on the Cochrane database have demonstrated illicit drugs. American Journal of Geriatric Psychiatry, 19(8), 704–11.
that nicotine replacement therapy through any mode (e.g. gum, spray, American Psychiatric Association (1994). Diagnostic and statistical manual of
or transdermal patch) increases quit rates by 1.5- to two-fold regard- mental disorders (4th edition). American Psychiatric Press, Washington,
less of setting (Silagy et al., 2002). Another such systematic review DC.
has demonstrated that the antidepressants bupropion and nortriptyl- Anderson, G.M., et al. (1993). Trends and determinants of prescription drug
expenditures in the elderly: data from the British Columbia Pharmacare
ine aid long-term smoking cessation, but the serotonin selective
program. Inquiry, 30, 199–207.
reuptake inhibitors (SSRIs) do not (Hughes et al., 2004). There is a
Anstey, K.J., Mack, H.A., and Cherbuin, N. (2009). Alcohol consumption
dearth of research on such pharmacological approaches to smoking as a risk factor for dementia and cognitive decline: meta-analysis of
cessation specifically in older populations, and it must be borne in prospective studies. American Journal of Geriatric Psychiatry, 17(7),
mind that any such therapies may be associated with higher levels 542–55.
of adverse effects in older people. In any comprehensive approach to Anthony, J.C. and Helzer, J.E. (1991). Syndromes of drug abuse and dependence.
improving the health of older people, however, consideration should In: Robins, L.N. and Regier, D.A. (eds) Psychiatric disorders in America:
be given to pharmacological therapies, along with advice and educa- the Epidemiologic Catchment Area study. Free Press, New York.
tion on the many health benefits of smoking cessation. Atkinson, J.H. and Schuckit, M.A. (1983). Geriatric alcohol and drug misuse
and abuse. Advances in Substance Abuse, 3, 195–237.
Atkinson, R.M. (2001). Alcohol and drug abuse in the elderly. In: Jacoby, R.
Conclusion and Oppenheimer, C. (eds) Psychiatry in the elderly. Oxford University
Press, Oxford.
This chapter has highlighted the importance of AUDs, IMU, and Avorn, J. (1995). Medication use and the elderly: current status and
use of illicit drugs in older people, in terms of their prevalence and opportunities. Health Affairs, 14(1), 276–86.
their important contribution to morbidity and mortality. AUDs Avorn, J. and Gurwitz, J.H. (1995). Drug use in the nursing home. Annals of
are underdetected, misdiagnosed, and often completely missed Internal Medicine, 123, 195–204.
in older populations. However, despite ageist and therapeutically Babor, T.F., et al. (1992). Types of alcoholics, 1. Evidence for an empirically
pessimistic assumptions, AUDs in older people are as amenable derived typology based on indicators of vulnerability and severity.
to treatment as in younger people, and treating an AUD in indi- Archives of General Psychiatry, 49(8), 599–608.
viduals of any age can lead to significant benefits in their quality Barkin, R.L., Barkin, S.J., and Barkin, D.S. (2005). Perception, assessment,
of life. Likewise, the wide variety of IMU in older people may be treatment and management of pain in the elderly. Clinical Geriatric
Medicine, 21(3), 465–90.
associated with addiction to medication and the undertreatment
Barry, P.J., et al. (2006). Inappropriate prescribing in the elderly: a comparison
and inappropriate treatment of medical and psychiatric conditions. of the Beers criteria and the improved prescribing in the elderly tool
Considering that older people are the highest consumers of pre- (IPET) in acutely ill elderly hospitalised patients. Journal of Clinical
scription medications, screening and treatment programmes for Pharmacology and Therapy, 31(6), 617–26.
IMU should also lead to considerable improvements in quality of Beckman, A., et al. (2005). The difficulty of opening medicine containers in
life, along with financial and other savings. Misuse of illicit drugs old age: a population-based study. Pharmacy World and Science, 27(5),
by older people is not generally a major problem at present, but it 393–8.
is virtually certain that consumption of illegal substances by people Beers, M.H. (1997). Explicit criteria for determining potentially inappropriate
over 65 will increase in the future. medication use by the elderly. An update. Archives of Internal Medicine,
157(14), 1531–6.
Greater awareness amongst physicians and other healthcare pro-
Beers, M., et al. (1988). Psychoactive medication use in intermediate-care
viders of the possibility of AUDs and IMU in their older patients
facility residents. Journal of the American Medical Association, 260,
should lead to the development of more comprehensive and 3016–20.
age-appropriate prevention and treatment strategies. At the lev- Beers, M.H., et al. (2000). Part 1: the problems facing managed care. American
els of everyday clinical practice and public health policy, greater Journal of Managed Care, 6, 1313–20.
emphasis should be placed on AUDs and IMU in older people, and Berdot, S., et al. (2009). Inappropriate medication use and risk of falls—a
further development and evaluation of dedicated ‘same-age’ treat- prospective study in a large community dwelling elderly cohort. BMC
ment services and settings should be performed. Geriatrics, 9, 30.
Blazer, D.G. and Wu, L.T. The epidemiology of alcohol use disorders and Ekerdt, D.J., et al. (1989). Change in drinking behaviours with retirement:
subthreshold dependence in a middle-aged and elderly community findings from the normative ageing study. Journal of Studies on Alcohol,
sample. American Journal of Geriatric Psychiatry, 19(8), 685–94. 50, 347–53.
Blow, F. (1991). Michigan Alcoholism Screening Test—Geriatric Version Ewing, J.A. (1984). Detecting alcoholism: the CAGE questionnaire. Journal of
(MAST-G). University of Michigan Alcohol Research Center, Ann the American Medical Association, 252, 1905–7.
Arbor. Fick, D.M., et al. (2003). Updating the Beers criteria for potentially
Blow, F.C. and Barry, K.L. (2000). Older patients with at-risk and problem inappropriate medication use in older adults: results of a US consensus
drinking patterns: new developments in brief interventions. Journal of panel of experts. Archives of Internal Medicine, 163(22), 2716–24.
Geriatric Psychiatry and Neurology, 13(3), 115–23. Fink, A., et al. (2002). The alcohol-related problems survey: identifying
Blow, F.C., Brockmann, L.M., and Barry, K.L. (2004). Role of alcohol in hazardous and harmful drinking in older primary care patients. Journal
late-life suicide. Alcoholism: Clinical and Experimental Research, 28(5 of the American Geriatrics Society, 50(10), 1717–22.
Suppl), 48S–56S. Fleming, M.F., et al. (1999). Brief physician advice for alcohol problems in
Bratzler, D.W., Oehlert, W.H., and Austelle, A. (2002). Smoking in the older adults: a randomized community-based trial. Journal of Family
elderly—it’s never too late to quit. Journal of the Oklahoma State Medical Practitioners, 48(5), 378–84.
Association, 95(3), 185–91. Folstein, M.F., Folstein, S.E., and McHugh, P.R. (1975). ‘Mini-mental state’.
Bristow, M.F. and Clare, A.W. (1992). Prevalence and characteristics of at A practical method for grading the cognitive state of patients for the
risk drinkers among elderly acute medical inpatients. British Journal of clinician. Journal of Psychiatric Research, 12(3), 189–98.
Addiction, 87, 291–4. Gallagher, P. and O’Mahony, D. (2008). STOPP (Screening Tool of Older
Brower, K.J., et al. (1994). Severity and treatment of alcohol withdrawal in Persons’ potentially inappropriate Prescriptions): application to acutely
elderly versus younger patients. Alcoholism: Clinical and Experimental ill elderly patients and comparison with Beers’ criteria. Age and Ageing,
Research, 18(1), 196–201. 37(6), 673–9.
Buckland, P.R. (2001). Genetic association studies of alcoholism—problems Gallagher, P.F., et al. (2008). Inappropriate prescribing in an acutely ill
with the candidate gene approach. Alcohol, 36(2), 99–103. population of elderly patients as determined by Beers’ criteria. Age and
Burns, D.M. (2000). Cigarette smoking among the elderly: disease Ageing, 37(1), 96–101.
consequences and the benefits of cessation. American Journal of Health Ganry, O., et al. (2000). Prevalence of alcohol problems among elderly
Promotion, 14(6), 357–61. patients in a university hospital. Addiction, 95, 107–13.
Busto, U.E., Romach, M.K., and Sellers, E.M. (1996). Multiple drug use and Gordon, T. and Kannel, W.B. (1983). Drinking and its relation to smoking,
psychiatric comorbidity in patients admitted to the hospital with severe BP, blood lipids and uric acid. Archives of Internal Medicine, 143,
benzodiazepine dependence. Journal of Clinical Psychopharmacology, 1366–74.
16(1), 51–7. Grant, B.F. and Harford, T.C. (1995). Comorbidity between DSM-IV alcohol
Cadieux, R.J. (1989). Drug interactions in the elderly: how multiple drug use use disorders and major depression: results of a national survey. Drug
increases risk exponentially. Postgraduate Medicine, 86, 179–86. and Alcohol Dependence, (39), 197–206.
Chrischilles, E.A., Segar, E.T., and Wallace, R.B. (1992). Self-reported adverse Greene, E., et al. (2003). Self-reported alcohol consumption in the Irish
drug reactions and related resource use. A study of community-dwelling community dwelling elderly. Irish Journal of Psychological Medicine,
persons 65 years of age and older. Annals of Internal Medicine, 117, 20(3), 77–9.
634–40. Han, B., et al. (2009). Substance use disorder among older adults in the
Chutka, D.S., Takahashi, P.Y., and Hoel, R.W. (2004). Inappropriate United States in 2020. Addiction, 104(1), 88–96.
medications for elderly patients. Mayo Clinic Proceedings, 79(1), Harrison, L., Sutton, M., and Gardiner, E. (1997). Ethnic differences in
122–39. substance use and alcohol use related mortality in first generation
Chutka, D.S., Takahashi, P.Y., and Hoel, R.W. (2005). Inappropriate medication migrants to England and Wales. Journal of Substance Use and Misuse,
use in the elderly. Essential Psychopharmacology, 6(6), 331–40. 32, 849–76.
Cook, B.L., et al. (1991). Depression and previous alcoholism in the elderly. Helme, R.D. and Katz, B. (1993). Management of chronic pain. Medical
British Journal of Psychiatry, 158, 72–5. Journal of Australia, 158(7), 478–81.
Covey, L.S., Glassman, A.H., and Stetner, F. (1998). Cigarette smoking and Hinkin, C., et al. (2002). Screening for drug and alcohol abuse among older
major depression. Journal of Addictive Diseases, 17(1), 35–46. adults using modified version of CAGE. American Journal of Addiction,
Curtis, J.R., et al. (1989). Characteristics, diagnosis and treatment of 10, 319–26.
alcoholism in elderly patients. Journal of the American Geriatric Society, Hughes, J., Stead, L., and Lancaster, T. (2004). Antidepressants for smoking
37(4), 310–16. cessation. Cochrane Database of Systematic Reviews, (4), CD000031.
Cyr, M.G. and Wartman, S.A. (1988). The effectiveness of routine screening Jackson, R. and Teece, S. (2004). Oral or intravenous thiamine in the
questions in the detection of alcoholism. Journal of the American Medical emergency department. Emergency Medical Journal, 21, 501–2.
Association, 259(1), 51–4. Jeste, D.V. and Palmer, B. (1998). Secondary psychoses: an overview. Seminars
Dahmen, N., et al. (2005). Tyrosine hydroxylase Val-81-Met polymorphism in Clinical Neuropsychiatry, 3, 2–3.
associated with early-onset alcoholism. Psychiatric Genetics, 15(1), Jinks, M.J. and Raschko, R.R. (1990). A profile of alcohol and prescription
13–16. drug abuse in a high risk community-based elderly population. Annals
Dar, K. (2006). Alcohol use disorders in older people: fact or fiction? Advances of Pharmacotherapy 24, 971–5.
in Psychiatric Treatment, 12, 173–81. Joseph, C.L., Ganzini, L., and Atkinson, R.M. (1995). Screening for alcohol
Davidson, K.M. and Ritson, E.B. (1993). The relationship between alcohol use disorders in the nursing home. Journal of the American Geriatrics
dependence and depression. Alcohol, 28, 147–55. Society, 43, 368–73.
Doll, R., et al. (1994). Mortality in relation to consumption of alcohol: 13 Kalant, H. (1998). Pharmacological interactions of aging and alcohol. In:
years’ observations on male British doctors. British Medical Journal, Gomberg, E.S.L., Hegedus, A.M., and Zucker, R.A. (eds) Alcohol problems
309(6959), 911–18. and aging. NIAAA Research Monograph No. 33. NIH Pub. No. 98–4163.
Dowling, G.J., Weiss, S.R., and Condon, T.P. (2008). Drugs of abuse and the NIAAA, Bethesda, Maryland.
ageing brain. Neuropsychopharmacology, 33(2), 209–18. Khan, N., et al. (2002). Drinking patterns among older people in the
Dufour, M. and Fuller, R.K. (1995). Alcohol in the elderly. Annual Review of community: hidden from medical attention? New Zealand Medical
Medicine, 46, 123–32. Journal, 115(1148), 72–5.
King, C., et al. (1994). Diagnosis and assessment of substance abuse in older O’Connell, H., et al. (2004). A systematic review of the utility of self-report
adults. Current strategies and issues. Addictive Behaviours, 19, 41–5. alcohol screening instruments in the elderly. International Journal of
Kirby, M., et al. (1999a). Benzodiazepine use among the elderly in the Geriatric Psychiatry, 19, 1074–86.
community. International Journal of Geriatric Psychiatry, 14(4), 280–4. O’Connell, H., et al. (2005). Neuropsychiatric aspects of alcohol use disorders
Kirby, M., et al. (1999b). Influence of anxiety on treatment of depression in in older people: findings from the Dublin Healthy Ageing Study. Paper
later life in primary care: questionnaire survey. British Medical Journal, presented at International Psychogeriatric Association, Stockholm.
318, 579–80. Office of Applied Studies. (2005). Older adults in substance abuse treatment:
Kofoed, L.L., et al. (1987). Treatment compliance of older alcoholics: an update. Substance Abuse and Mental Health Administration Services,
elder-specific approach is superior to ‘mainstreaming.’ Journal of Studies Rockville, MD.
on Alcohol, 48, 47. Office of Applied Studies. (2009). Results from the 2008 National Survey on
Kuhar, M.J., Joyce, A. and Dominguez, G. (2001). Genes in drug abuse. Drug Drug Use and Health. National Findings (DHSS Publication No. SMA
and Alcohol Dependencies, 62(3), 157–62. 09–4434, NSDUH Series H-36). Substance Abuse and Mental Health
LaCroix, A.Z. and Omenn, G.S. (1992). Older adults and smoking. Clinics in Services Administration, Rockville, MD.
Geriatric Medicine, 8, 69–87. Oslin, D., et al. (1997a). Tolerability of naltrexone in treating older,
LaGreca, A.J., Akers, R.L., and Dwyer, J.W. (1988). Life events and alcohol alcohol-dependent patients. American Journal of Addiction, 6(3), 226–70.
behaviors among older adults. The Gerontologist, 28, 552–8. Oslin, D., et al. (1997b). Naltrexone as an adjunctive treatment for older
Lawlor, D.A., Patel, R., and Ebrahim, S. (2003). Association between falls in patients with alcohol dependence. American Journal of Geriatric
elderly women and chronic disease and drug use. British Medical Journal, Psychiatry, 5(4), 324–32.
327, 712–17. Oslin, D.W., Pettinati, H., and Volpicelli, J.R. (2002). Alcoholism treatment
Letenneur, L. (2004). Risk of dementia and alcohol and wine consumption: a adherence: older age predicts better adherence and drinking outcomes.
review of recent results. Biological Research, 37, 189–93. American Journal of Geriatric Psychiatry, 10(6), 740–7.
Levings, B., Szep, S., and Helps, S.C. (1999). Towards the safer use of dosettes. Patterson, T.L. and Jeste, D.V. (1999). The potential impact of the baby-boom
Journal of Quality in Clinical Practice, 19, 69–72. generation on substance abuse among elderly persons. Psychiatric
Liberto, J.G. and Oslin, D.W. (1995). Early onset versus late onset alcoholism Services, 50(9), 1184–8.
in the elderly. International Journal of Addiction, 30, 1799–818. Peppers, M.P. (1996). Benzodiazepines for alcohol withdrawal in the elderly
Liskow, B.I., et al. (1989). Alcohol withdrawal in the elderly. Journal of Studies and in patients with liver disease. Pharmacotherapy, 16(1), 49–57.
on Alcohol, 50(5), 414–21. Philpot, M., et al. (2003). Screening for problem drinking in older people
Lofwall, M.R., et al. (2005). Characteristics of older opioid maintenance referred to a mental health service: a comparison of CAGE and AUDIT.
patients. Journal of Substance Abuse and Treatment, 28(3), 265–72. Aging and Mental Health, 7(3), 171–5.
Luttrell, S., et al. (1997). Screening for alcohol misuse in older people. Regier, D.A., et al. (1988). One-month prevalence of mental disorders in the
International Journal of Geriatric Psychiatry, 12, 1151–4. United States based on five Epidemiologic Catchment Area sites. Archives
McGrath, A., Crome, P., and Crome, I.B. (2005). Substance misuse in the of General Psychiatry, 45, 977–86.
older population. Postgraduate Medical Journal, 81, 228–31. Reid, M.C. and Anderson, P.A. (1997). Geriatric substance misuse disorders.
Mirand, A.L. and Welte, J.W. (1996). Alcohol consumption among the elderly Medical Clinics of North America, 81, 999–1016.
in a general population, Erie County, New York. American Journal of Rimm, E.B., et al. (1999). Moderate alcohol intake and lower risk of coronary
Public Health, 86, 978–84. heart disease: meta-analysis of effects on lipids and haemostatic factors.
Moos, R., Brennan, P., and Schutte, K. (1998). Life context factors, treatment, British Medical Journal, 319(7224), 1523–8.
and late-life drinking behaviour. In: Gomberg, E.S.L., Hegedus, A.M., Ringe, J.D., van der Geest, S.A., and Moller, G. (2006). Importance of calcium
and Zucker, R.A. (eds) Alcohol problems and aging. NIAAA Research co-medication in bisphosphonate therapy of osteoporosis: an approach
Monograph No. 33. NIH Pub. No. 98–4163. NIAAA, Bethesda, to improving correct intake and drug adherence. Drugs and Aging, 23(7),
Maryland. 569–78.
Mulder, R.T. (2002). Alcoholism and personality. Australian and New Zealand Rothberg, M.B., et al. (2008). Potentially inappropriate medication use in
Journal of Psychiatry, 36, 44–52. hospitalised elders. Journal of Hospital Medicine, 3(2), 91–102.
Naranjo, C.A. and Sellers, E.M. (1986). Clinical assessment and Royal Colleges of Physicians, Psychiatrists and General Practitioners (1995).
pharmacotherapy of the alcohol withdrawal syndrome. Recent Alcohol and the heart in perspective: sensible limits reaffirmed. Royal
Developments in Alcohol, 4, 265–81. College of Physicians, London.
National Institute on Alcohol Abuse and Alcoholism (1998). Alcohol alert Ruchlin, H.S. (1997). Prevalence and correlates of alcohol use among older
No. 40. NIAAA, Bethesda, MD. <www.niaaa.nih.gov/publications/aa40. adults. Preventive Medicine, 26, 651–7.
htm>. Ruitenberg, A., et al. (2002). Alcohol consumption and risk of dementia: the
Nieman, J. (1998). Alcoh as a risk factor for brain damage: neurologic aspects. Rotterdam study. Lancet, 359, 281–6.
Alcohol and Clinical Experimental Research, 22(7 Suppl), 346S–51S. Saunders, J.B. (1993). Development of the Alcohol Use Disorders Identification
Nurnberger, J.I., et al. (2004). A family study of alcohol dependence: Test (AUDIT). Addiction, 88, 791–804.
coaggregation of multiple disorders in relatives of alcohol-dependent Saunders, P.A., et al. (1991). Heavy drinking as a risk factor for depression
probands. Archives of General Psychiatry, 61(12), 1246–56. and dementia in elderly men. Findings from the Liverpool longitudinal
O’Connell, H. and Lawlor, B.A. (2005a). Recent alcohol intake and suicide: community study. British Journal of Psychiatry, 159, 213–16.
a neuropsychological perspective. Irish Journal of Medical Science, 174, Schneider, J.P. (2005). Chronic pain management in older adults: with coxibs
51–4. under fire, what now? Geriatrics, 60(5), 26–8, 30–1.
O’Connell, H. and Lawlor, B. (2005b). Alcohol and heart disease—what do Schonfeld, L. and Dupree, L.W. (1995). Treatment approaches for older problem
you prescribe? Irish Medical Journal, 98, 230–1. drinkers. International Journal of Addiction, 30(13–14), 1819–42.
O’Connell, H., et al. (2003a). Alcohol use disorders in elderly people— Schutte, K.K., Brennan, P.L., and Moos, R.H. (1994). Remission of late-life
redefining an age old problem in old age. British Medical Journal, 327, drinking problems: a 4-year follow-up. Alcohol and Clinical and
664–7. Experimental Research, 18, 835–44.
O’Connell, H., Chin, A.V., and Lawlor, B.A. (2003b). Alcohol use in Ireland— Selzer, M.L. (1968). Michigan Alcoholism Screening Test (MAST):
can we hold our drink? Irish Journal of Psychological Medicine, 20, preliminary report. University of Michigan Medical Center Journal, 34(3),
109–10. 143–5.
Shaper, A.G. (1995). Mortality and alcohol consumption. Non-drinkers Tiihonen, J., et al. (1999). Association between the functional variant of the
shouldn’t be used as a baseline. British Medical Journal, 310(6975), 325. catechol-O-methyltransferase (COMT) gene and type 1 alcoholism.
Silagy, C., et al. (2002). Nicotine replacement therapy for smoking cessation. Molecular Psychiatry,4(3), 286–9.
Cochrane Database of Systematic Reviews, (4), CD000146. Trafton, J.E., et al. (2004). Treatment needs associated with pain in substance
Sintonen, H. (1994). Outcomes measurement in acid-related diseases. use disorder patients; implications of concurrent treatment. Drug and
Pharmaco Economics, 5(Suppl 3), 17–26. Alcohol Dependency, 73, 23–31.
Smith, D.M. and Atkinson, R.M. (1995). Alcoholism and dementia. UK National Digital Archive of Datasets (1994). General household survey,
International Journal of Addiction, 30(13–14), 1843–69. UK. <http://ndad.ulcc.ac.uk/CRDA/28/DS/1/detail.html>.
Speer, D.C. and Bates, K. (1992). Comorbid mental and substance disorders Vogel-Sprott, M. and Barrett, P. (1984). Age, drinking habits and the effects of
among older psychiatric patients. Journal of the American Geriatrics alcohol. Journal of Studies on Alcohol, 45, 517–21.
Society, 40, 886–90. Waern, M. (2003). Alcohol dependence and misuse in elderly suicides.
Swartz, M., et al: (1991). Benzodiazepine anti-anxiety agents: prevalence and Alcohol, 38, 249–54.
correlates of use in a southern community. American Journal of Public Wilhelm, K., et al. (2004). Grey lungs and blue moods: smoking cessation in
Health, 81, 592–6. the context of lifetime depression history. Australian and New Zealand
Taylor, D.H., et al. (2002). Benefits of smoking cessation for longevity. Journal of Psychiatry, 38(11–12), 896–905.
American Journal of Public Health, 92, 990–6. Williams, L. and Lowenthal, D.T. (1992). Drug therapy in the elderly. Southern
Taylor, D., et al. (2005). The Maudsley 2005–2006 prescribing guidelines. Medical Journal, 85(2), 127–31.
Taylor and Francis, London. World Health Organization (1992). The ICD-10 classification of mental and
Teichner, G., et al. (2002). Substance abuse treatment outcomes for cognitively behavioural disorders. WHO, Geneva.
impaired and intact outpatients. Addictive Behaviors, 25, 751–63. Zedler, B.K., et al. (2011). Does packaging with a calendar feature improve
Temple, M.T. and Leino, E.V. (1989). Long-term outcomes of drinking: a adherence to self-administered medication for long-term use? A
twenty year longitudinal study of man. British Journal of Addiction, 84, systematic review. Clinical Therapy, 33(1), 62–73.
889–93.
CHAPTER 50
Older people with
learning disabilities
Maria Luisa Hanney
Older people with learning disabilities (LD) are an emerging het- Classification of Disease (ICD-10) (WHO, 1993) and the revised
erogeneous clinically complex population to whom little clinical or edition of the American Psychiatric Association’s Diagnostic and
research interest has been devoted. They pose unique medical and Statistical Manual of Mental Disorders (DSM-IV-TR (APA, 2000).
social challenges as they grow older and become frailer and infirm Both classification systems use similar diagnostic criteria:
(Kim et al., 2011), and they no longer fit within the group of their
A. Significantly subaverage intellectual functioning: an Intelligence
younger peers.
Quotient of approximately 70 or below on an individually
The cost of caring for people with LD makes up a large propor-
administered IQ test (for infants, a clinical judgement of signifi-
tion of healthcare spending in western Europe. For example, older
cantly subaverage intellectual functioning).
adults with LD comprise about 0.15–0.25% of the population of
England but consume up to 5% of the total personal care budget. It B. Concurrent deficits or impairments in present adaptive func-
seems very likely that general services for older people will need to tioning (i.e. the person’s effectiveness in meeting the standards
prepare themselves for the inclusion of this subgroup of the ageing expected for his or her age by his or her cultural group) in at
population. This will have huge implications for future planning of least two of the following areas: communication, self-care, home
services (Strydom et al., 2010). living, social/interpersonal skills, use of community resources,
This chapter will start by reviewing the concept and diagnostic self-direction, functional academic skills, work, leisure, health
criteria for LD and what is meant by ‘older’ in LD, followed by a and safety.
summary of the present knowledge about functional psychiatric C. The onset is before the age of 18 years.
illness and dementia in this group. A separate section concentrates
on issues related to Down syndrome (DS) and their increased risk
of dementia. This is followed by a section on assessment and man- What Is Meant by ‘Older’ in Learning
agement of dementia in LD in general, including pharmacological
treatment and end-of-life and best-interest decisions, with refer- Disabilities, and Statistics
ence to relevant best practice guidelines. There is no agreed definition of what constitutes ‘older’ in people
with LD. Various age thresholds for the ageing population with LD
Definition and Diagnostic Criteria for have been adopted in the literature which range from 40–65 years
of age (Jenkins et al., 1994).
Learning Disabilities Little is known about the prevalence of LD among the general
Mental retardation (MR), intellectual disabilities (ID), and learning older population. A study in New Zealand in 1996 by Hand and
disabilities (LD) are often used as interchangeable terms. The latter Reid estimated the age-specific national prevalence of people with
term is used in this chapter as defined by the UK Department of LD born before 1940 on a total population of 7,433,560 as 1.43 per
Health (DH, 2001: 14–15). 1000 for people with LD between 51 and 88 years of age (range
Learning disability includes the presence of a significantly 1.49–2.82 per 1000). A total of 1063 older people with LD were
reduced ability to understand new or complex information, to learn identified; 56% between 51 and 60 years of age; 32% between 61
new skills (impaired intelligence), with a reduced ability to cope and 70 years; and 12% between 71 and 88 years. Emerson and
independently (impaired social functioning) which started before Hatton (2004) estimated that 985,000 people in England have an
adulthood, with a lasting effect on development. ‘Learning disabil- LD (2% of the general population). This figure includes 828,000
ity’ does not include all those who have a ‘learning difficulty’, which adults (aged 18 and over). Of these adults, it was estimated that
is more broadly defined in education legislation. 177,000 were known users of LD services in England (equivalent
The term MR is used in the two main diagnostic systems for to 0.47% of this adult population) and 174,000 were aged 60 or
mental disorders, the World Health Organization’s International more (21.4%).
Functional Psychiatric Disorders in Older diagnosis of dementia in people with LD has been proposed, as
opposed to the DSM-IV (APA, 1994), because the ICD-10 criteria
People with Learning Disabilities place more emphasis on ‘noncognitive’ aspects of dementia, which
The psychiatric assessment of people with LD is complicated, have often been reported to be part of the first signs of dementia in
given the potential communication difficulties displayed by this people with LD (Aylward et al., 1997). In 2001, the Royal College
group (Sigelman et al., 1981). In the older person with LD, this is of Psychiatrists modified the ICD-10 diagnostic criteria for demen-
even more difficult due to added age-related sensory and cogni- tia for use with people with LD, publishing the DC-LD) (RCPsych,
tive difficulties and often a lack of collateral information from car- 2001). However, Strydom and colleagues (2007) have examined
ers who had known the person for long enough (Holland, 2000). the validity of the three dementia diagnostic criteria in an epide-
Furthermore, diagnostic criteria for mental health used in the gen- miological sample of older adults with LD not due to DS, showing
eral population (ICD-10 and DSM-IV-TR) tend to underestimate that DSM-IV dementia criteria were more inclusive than DC-LD
the prevalence of mental illness in this population, as compared or ICD-10, and that diagnosis using ICD-10 excluded people with
with the Royal College of Psychiatrists Diagnostic Criteria for even moderate dementia.
Psychiatric Disorders for use with adults with Learning Disabilities
(DC-LD) (RCPsych, 2001; Cooper et al., 2007). Difficulties in the diagnosis of dementia in learning
Cooper et al. (1997) compared the prevalence of psychiatric disabilities
disorder between a randomly selected younger group of people Due to the high rates of physical and mental ill health, and sen-
with LD and a group of identified older people with LD (over 65 sory and mobility impairment in this population, any cognitive and
years old) in a defined geographical area. In the older group, an functional decline can be very difficult to distinguish from early
additional psychiatric disorder was present in 68.7% of people as signs of dementia.
compared with 47.9 % in the younger group. The higher rates of Given the wide range of abilities in the LD population, a diagnosis
psychiatric diagnosis in the older group were due mainly to a high of dementia requires a change from a premorbid level of function-
rate of dementia of 21.6%, as compared with 2.7% in the younger ing specific to the individual in question. Longitudinal assessments
group. This high rate was not accounted for by the presence of DS, that document both baseline and present cognitive and behavioural
as the older group included only 3.7% of people with DS, compared functioning over a period of at least 6 months is necessary before
with 28.8% of the younger group. The older group also had higher sufficient information can be obtained to make a confident diagno-
rates of generalized anxiety disorder (9.0% compared with 5.5%) sis of dementia (Burt and Aylward, 1998).
and depression (6.0% vs 4.1%). The presence of behaviour disor- Decline in cognitive function and behaviour in individuals with
ders, pervasive developmental disorders, anxiety disorders, and mild LD who develop dementia can be very similar to that seen
psychotic disorders was equal in both groups. Schizophrenia/delu- in the general population with Alzheimer’s disease (AD). However,
sional disorders were present in 3% (vs 2.7%) and mania in 0.7% the extremely limited cognitive and communication abilities of
(vs 0.0%). There were no cases of obsessive-compulsive disorder or people with severe and profound LD is necessarily associated with
alcoholism. Behavioural disorder was present in 14.9% (vs 15.1%). diagnostic uncertainty (Strydom et al., 2009).
Mental ill health has been reported to be less prevalent in adults To be indicative of dementia, any changes over time must be
with DS than for other adults with LD (Cooper and van der Speck, greater than those related to normal ageing in adults with LD
2009). Mantry and colleagues (2008), in a longitudinal cohort study (Burt and Aylward, 1999). People with LD as a whole tend to age
of 168 people with DS (mean age = 41.1 years; range = 16–74 years; faster and to show signs of ‘ageing’ earlier than the general popula-
standard deviation = 11.8 years), reported a point prevalence of tion (Jenkins et al., 1994). Nevertheless, there is little information
mental ill health of any type, excluding specific phobias, of 23.7% regarding the impact of ageing in this population.
by clinical diagnosis, 19.9% by DC-LD (RCPsych, 2001), 11.3% Functional ability decreases with age in individuals with LD,
by ICD-10 criteria, and 10.8% by DSM-IV-TR criteria. Compared but, as a group, functional ability improves in later life because
with persons with LD not due to DS, the standardized rate for prev- of differential mortality rates, leading to a shorter life-expectancy
alence of mental illness in people with DS was 0.6 (0.4–0.8), or 0.4 for people with more severe learning disabilities and for people
(0.3–0.6) if organic disorders were excluded. with DS (Moss, 1991). The extents to which physical disabili-
Depression, however, has been reported to be the most common ties contribute to age-related morbidity in people with LD have
mental health problem in people with DS (Määtä et al., 2006) and been summarized by Day and Jancar (1994) and more recently by
is often associated with the presence of dementia in this popula- Haveman and colleagues (2011). These studies report an increas-
tion (Prasher and Filler, 1995). Detection of depression in DS is ing prevalence of musculoskeletal, cardiovascular, respiratory,
problematic as people with DS do not tend to verbalize symptoms and neoplastic illnesses with age. However, in the case of some
of depression and the diagnosis is often inferred by the presence of disorders (e.g. neoplasia), age-related increases in prevalence
prominent vegetative symptoms and, on occasions, prominent hal- are relatively small. Cataract, hearing disorder, diabetes, hyper-
lucinations (Myers and Pueschel, 1991). tension, osteoarthritis/arthrosis, and osteoporosis were more
strongly associated with advancing age; allergies and epilepsy
Dementia in Learning Disabilities were negatively associated.
Definition and diagnostic criteria for dementia Dementia in people with learning disabilities not due
in learning disabilities to Down syndrome
There is no definition of dementia that is specific for individu- It is well known that people with DS tend to develop dementia more
als with LD. The use of the ICD-10 (WHO, 1993) criteria for the often and at an earlier age than the general population. It is perhaps
CHAPTER 50 older people with learning disabilities 655
generally less well known that people with LD not due to DS also (NFT) and/or neuritic plaques (NP) was observed in 63.4% of all
suffer from dementia at higher rates and at an earlier age than the cases and varied with age. Recommended age-specific quantitative
general population (Thompson, 1951; Tredgold, 1952; Patel et al., criteria for the diagnosis of AD (Khachaturian, 1985) were met in
1993; Strydom et al., 2009). 9.5% of those cases below 50 years of age, 54.2% between 50 and 65,
70% between 66 and 75, and 87% of the cases greater than 75 years
Prevalence, types of dementia, and risk factors
of age. However, in most cases the NP did not have a neuritic com-
The study by Strydom and colleagues (2009) has confirmed that ponent containing paired helical filaments, and substantial num-
using the criteria defined, dementia is two to three times more bers of NFT were seen in frontal cortex, contrasting with results
common in the LD population not due to DS than in the general reported in the literature for the general population. The number
population. They assessed 281 adults with LD not due to DS aged of NP per square millimetre consistently increased with age for all
60 years or over using defined diagnostic criteria for ‘probable’ sub- areas examined, with the largest age-associated increases in den-
types of dementia. Overall, prevalence for criteria-defined demen- sity seen in frontal and temporal regions. In contrast, NFT density
tia was 13.1% in those aged 60 years or over and 18.3% in those increased with age only within hippocampus and parahippocampal
aged over 65 years. AD diagnosed using NINCDS–ADRDA criteria gyrus, but not neocortex.
(McKhann et al., 1984) was the most common type of dementia, In a similar study, Silverman and colleagues (1993) reported sta-
with a prevalence of 8.6% in those of 60 years and over and 12% in tistical independence of regional amyloid plaque and NFT densities
those over 65 years. Dementia with Lewy bodies, diagnosed using with NP lesions within neocortex being more diffusely distributed
the McKeith et al. (1996) criteria, was the second most common, across regions for older cases compared to younger cases, while
with a prevalence of 5.9% in those of 60 years and over and 7.7% in no similar age-associated change in the topography of NFT was
those over 65 years. Frontotemporal dementia using the McKhann observed.
et al. (2001) criteria was 3.2% in those of 60 years and over and
4.2% in those over 65 years and was more common than vascular
dementia diagnosed using the NINDS–AIREN criteria (Roman et Down syndrome
al., 1993), for which prevalence was 2.7% in those of 60 years and DS is the most commonly identified genetic form of LD and the
over and 3.5% in those over 65 years. Prevalence rates did not differ leading cause of specific birth defects and medical conditions
between mild, moderate, and severe LD groups. Age was a strong (Sherman et al., 2007). People with DS are a heterogeneous group
risk factor for dementia and was not influenced by gender or LD with marked individual differences in cognitive and functional
severity. ability.
Clinical presentation Epidemiology
Despite problems in diagnosing dementia in this group, there
The prevalence of DS for the 16 years and over population in the
are significant measurable differences in several neurocognitive
UK has been reported as 5.9 per 10,000 general population (Mantry
domains between people with LD with and without dementia. Test
et al., 2008). The prevalence of pregnancies affected by DS has
performance on measures of attention and executive functions,
increased in the last 20 years, but there has been little change in live
language, memory, and learning have been reported as significantly
birth prevalence. Increasing maternal age and improved survival of
lower for persons with LD and dementia compared with matched
children with DS have offset the effects of prenatal diagnosis, fol-
controls of people LD without dementia (Palmer, 2006).
lowed by the termination of pregnancy and declining general birth
It has been suggested that clinical presentation of dementia may
rate (Irving et al., 2008). Morris and Alberman (2009), in the UK,
differ between people with LD and DS and those with LD due to
have reported an increase in antenatal and postnatal diagnoses of
other causes. Cooper and Prasher (1998) conducted a small-scale
DS of 71% between 1989/90 and 2007/08 (from 1075 in 1989/90 to
study comparing a group of 19 people with DS and dementia,
1843 in 2007/08, despite the number of births being similar dur-
and a group of 26 people with LD of other causes and dementia.
ing that period of time), but the numbers of live births with DS
Maladaptive behaviours and psychiatric symptomatology were
decreased only by 1% (752 to 743; 1.10 to 1.08 per 1000 births) in
assessed in both groups. The group with DS had a higher preva-
the same period of time.
lence of low mood, restlessness and overactivity, disturbed sleep,
being excessively uncooperative, and auditory hallucinations.
Life-expectancy and mortality
Aggression occurred with greater frequency in those subjects with
LD not due to DS. Life-expectancy of individuals with DS has increased in the last
60 years from 12 years of age in 1949 to nearly 60 today (Penrose,
Neuropathology 1949; Bittles and Glasson, 2004). However, life-expectancy in peo-
AD neuropathology develops in many individuals with LD not due ple with DS is still lower than in the general population and in other
to DS. In a post-mortem series, Malamud (1972) reported no evi- forms of LD.
dence of AD pathology among adults in this population for people Mortality in people with DS increases faster after the age of 40
under the age of 40 and a prevalence of 14% in individuals over than in other forms of LD (Strauss and Eyman, 1996). Disorders
40 years old. Barcikoswska and colleagues (1989) reported a preva- in persons with DS that are related to mortality include dementia,
lence of neuropathology for AD of 31% in people with LD not due mobility restrictions, feeding and swallowing difficulties, aspiration
to DS who were over 65 years of age at the time of death. pneumonia (Coppus et al., 2008), visual impairment, and epilepsy,
Popovitch and colleagues (1990) examined the brains of 385 but not cardiovascular diseases (Strauss and Eyman, 1996). Levels
adults aged 23–90 years with LD without DS, metabolic disorder, or of intellectual disability and institutionalization are also associated
hydrocephalus. The presence of one or more neurofibrillary tangles with mortality (Coppus et al., 2008).
Age-related changes in physical health Esbensen and colleagues (2008) found evidence of age-related
Adults with DS experience premature age-related health problems change in a longitudinal study over 9 years in a group of 150 adults
(see Esbensen (2010) for a comprehensive review). Amongst these with DS and 240 adults with LD due to other causes. The majority
are increased risk for dementia, skin and hair changes (Madan of participants were younger than 40 years of age. They reported
et al., 2006), early-onset menopause (Seltzer et al., 2001), visual and a decline in both groups for functional abilities relating to per-
hearing impairments (McCarron et al., 2005), adult-onset seizure sonal care and mobility, age-related improvement in housekeep-
disorder (Menéndez, 2005), thyroid dysfunction (Coleman, 1994), ing skills, and stability over the 9-year period in abilities relating
obesity (Melville et al., 2005), sleep apnoea (Resta et al., 2003), tes- to meal-related activities. They also reported age-related improve-
ticular cancer (Hasle et al., 2000), musculoskeletal problems such ments in behaviour problems in both groups. In another prospec-
as osteoporosis, osteoarthritis, decreased muscle strength, impaired tive cross-sectional study (Dressler et al., 2010), 75 individuals with
balance, falls (Dacre and Huskisson, 1988; Center et al., 1998), and DS without dementia (age range 4–52 years) underwent detailed
mitral valve prolapse (Barnhart and Connolly, 2007). There are assessments. The authors reported that individuals with DS con-
conflicting reports of whether people with DS are at increased risk tinue to increase their competence in functional skills until the
of diabetes type 1 (Haveman et al., 1989; Yang et al., 2002; Hill et al., age of 30 years, even when cognitive abilities reach a plateau, and
2003) and decreased risk of diabetes type 2 (Silverman, 2010). detected no major decline in middle adulthood. In general, most
Adults with DS are, however, at a lower risk than the general pop- studies report no major functional decline before the age of 40
ulation for several medical conditions, including malignant solid years (Esbensen et al., 2008).
tumors (Satgé et al., 1998), cerebrovascular and noncongenital car-
diovascular disease (Marino and Pueschel, 1996), and digestive sys- Dementia in people with Down syndrome
tem disease (Glover and Ayub, 2010). They have lower resting heart The association between dementia and DS was first scientifically
rates and lower blood pressure than the general population (Prasher, reported by Fraser and Mitchell (1876), well before Alois Alzheimer
1994). Hypertension is not common in this population (Kerins et al., described the case of the first patient with the disease that later bore
2008). Compared with people with LD not due to DS, they have lower his name (Alzheimer, 1907). It was Jervis who in 1948 first clinically
rates of emphysema, fractures, and hypercholesterolaemia (Kerins et described the association between DS and AD (Jervis, 1948).
al., 2008). While significant respiratory problems are not common The reason why AD is more frequent in individuals with DS is still
(Minihan and Dean, 1990), as mobility declines with age, recurrent not known. It is believed that people with DS have a genetic predis-
pneumonia with incomplete recovery has been found to occur more position to develop this type of dementia due to the triplication of
often (Van Allen et al., 1999). This is significant as respiratory illness the APP gene on chromosome 21, leading to overproduction and
is a common cause of mortality in adults with DS (Esbensen, 2010). deposition of amyloid beta (Aβ).
Age-related cognitive changes Prevalence

The prevalence of dementia among adults with DS has been
Cognitive changes occur with age in DS and are more common
estimated to be approximately 20% after the age of 40 (Janicki
than dementia. Several studies have compared the cognitive profile
and Dalton, 2000) and 45% after the age of 55 (Silverman et al.,
of older people with DS without dementia to young DS and older
1998). After the age of 60, only just over half of individuals with
people with DS and dementia (Haxby, 1989; Schapiro et al., 1992).
DS (56%) have a diagnosis of dementia (Janicki and Dalton, 2000;
They have shown that older people with DS without dementia do
Margallo-Lana et al., 2007).
not differ from the younger DS groups on tests of overall abilities,
Even more optimistic outlook for this population has been
such as the Stanford–Binet Test (Terman and Merril, 1973), tests of
reported in the largest longitudinal study published yet by Coppus
attention or language (Schapiro et al., 1992), and immediate ver-
and colleagues (2006) in the Netherlands. Their study included 506
bal memory spans (digit and object pointing spans). Older people
people with DS over the age of 45 years who were followed-up for
with DS without dementia had lower visuospatial memory span
a mean period of 3.3 years. They reported an overall prevalence of
(block tapping) than younger DS subjects. This suggests that there
dementia of 16.8%. Up to the age of 60, the prevalence of dementia
is a greater loss of immediate visuospatial memory than immediate
doubled with each 5-year interval. From age 45–49, the prevalence
verbal memory with ageing in DS even in the absence of dementia.
was 8.9%, from 50–54, it was 17.7%, from 55–59, it was 32.1%,
Age-related changes in verbal communication in people with DS
and from 60 and above, it was 25.6%. They argued that the lack
include a decline in social discourse involving changes in conver-
of increase in prevalence after the age of 60 may be explained by
sational style, literal understanding, and verbal expression in social
the increased mortality among older DS patients in comparison
contexts (Nelson et al., 2001; Couzens et al., 2011).
with patients without dementia during their 3.3-year follow-up.
Other studies looking at measures of attention, executive func-
However, they found no decrease in incidence of dementia in the
tion, and memory in people with DS without dementia suggest that
age group of 60 and above. In a 15-year follow-up community study
people with DS over the age of 40 experience an average annual glo-
of 92 people with DS who had previously been institutionalized, the
bal cognitive decline of 11%, indicating that progressive cognitive
years at risk for dementia for those aged 45 or under was found to
decline occurs from midlife onwards in the absence of dementia
be zero, whereas it reached a maximum of 8.75/100 person years in
(Margallo-Lana et al., 2003).
those aged 60–64 years (Margallo-Lana et al., 2007).
Age-related changes in functional skills Clinical presentation
The literature about age-related changes in functional skills for Clinical presentation of dementia in people with DS will be influ-
people with DS without dementia is controversial. For example, enced by the severity of their LD. The milder the LD severity is, the
CHAPTER 50 older people with learning disabilities 657
closer the symptomatology will resemble that seen in the general epilepsy incidence, with peaks in childhood, early adulthood, and
population with AD. over 50 years of age (Pueschel et al., 1991). Individuals after the
Forgetfulness, impairment of recent memory with relatively age of 50 tend to present with late-onset myoclonic epilepsy in
intact long-term memory, and confusion are all common, and adults with DS (LOMEDS) (Möller et al., 2001), characterized by
present early in dementia among adults with DS. A general slow- a progressive deterioration of cognitive function a few years before
ness including slowness in activities and speech, other language the onset of the epilepsy, featuring myoclonic as well as general-
problems, loss of interest in activities, social withdrawal, balance ized tonic-clonic seizures (Li et al., 1995). EEG changes show gen-
problems, sleep problems, loss of pre-existing skills, along with the eralized fast spike-waves or polyspikes or polyspike-waves with or
emergence of emotional and behavioural problems are common without bilateral myoclonic jerks, especially at awakenings. Photo
presentations of dementia among adults with DS (Deb et al., 2007). paroxysmal response (11–21 Hz) with bilateral myoclonic jerks has
Recent studies have suggested that people with DS tend to have also been described as part of LOMEDS (Crespel et al., 2007).
early manifestation of frontal lobe symptomatology rather than
Dysphagia
memory changes (Ball et al., 2010). This is not surprising given
the brain developmental abnormalities affecting the frontal lobe There is no published research examining the effect of ageing or
present in people with DS at birth (Teipel and Hampel, 2006) and dementia on drinking, eating, and swallowing in adults with DS
the fact that the earliest deposition of Aβ in the brain of people with (see Lazenby (2008) for a literature review on the subject). As in any
DS occurs in the frontal lobe and entorhinal cortex (Azizeh et al., degenerative disease affecting the central nervous system, people
2000; Nelson et al., 2011). with DS and dementia develop feeding problems as their disease
Among adults with more severe and profound LD, the diagnosis progresses. People with DS have specific oropharyngeal problems
of dementia, in the absence of any measurable cognitive decline, (Hennequin et al., 1999), swallowing dysfunction (Frazier and
needs to be made on the basis of behavioural changes character- Friedman 1996), high prevalence of oesophageal motor disorders
istic of progressive neurodegenerative disease, such as decline in (Bianca et al., 2002), and high prevalence of reflux (Wallace, 2007),
everyday skills, loss of interest in surroundings, daytime sleepiness, which increase the risk of dysphagia and aspiration.
wandering and getting lost, decreasing mobility, onset of epilepsy, Clinicians working in the field report that progressive dysphagia
increasing incontinence (Burt et al., 1992; Margallo-Lana, 2007), and frequent choking may be observed in people with DS in the
abnormality of the posture or gait (Lai and Williams, 1989), and early stages of dementia but are more obvious in the middle stage
rigidity or myoclonic jerks (McVicker et al., 1994). These symp- (Alvarez, 2011). Sometimes carers find it difficult to identify dys-
toms are common in people with DS and dementia, and are used to phagia in the early stages, as people with DS and dementia may just
support the diagnosis of dementia when cognitive decline cannot become more selective with the type of food they eat or restrict flu-
be reliably measured. ids and solids altogether. Early-stage dysphagia might not be diag-
nosed until the patient presents with severe weight loss or recurrent
Cognitive changes associated with dementia in Down respiratory infection.
syndrome
Motor and gait disorders and falls
The studies mentioned in the section Age-related cognitive changes
When people with DS develop dementia, motor disorders could
comparing the cognitive profile of younger and older people with
manifest early in the disease and would depend on the premorbid
DS, with and without dementia, have shown that older people with
level of mobility. They usually become more obvious in the mid-
DS and dementia show a more global pattern of deficits when com-
dle and advanced stage of the disease. They include general motor
pared to older people with DS without dementia and young DS.
slowness, difficulty in performing complex motor tasks, such as
Older people with DS and dementia also showed deficits in lan-
getting in and out of a car, going up and down stairs, negotiating
guage and verbal immediate span tests, when compared to young
changes in levels such as in kerbs, progressive gait disorder, falls,
people with DS, and deficits in tests of new long-term memory and
and, in some patients, a parkinsonian syndrome (Alvarez, 2011).
visuospatial construction when compared to older people with DS
As in the general population with dementia, in people with DS and
who had no dementia.
dementia mobility progressively deteriorates, and in the advanced
Behavioural changes stages they become bedridden, with little voluntary movement.
Behavioural changes are common in people with DS in the early Although it has recently been reported that young people with
stages of dementia (Määtä et al., 2006). It has been reported that DS have a lower risk of falling compared with their peers with other
people with DS and dementia tend to display less aggressive behav- types of LD (Finlayson et al., 2010), others (Cox et al., 2010) have
iour but higher degree of low mood, restlessness, hyperactivity, dis- found that increased age on its own is a risk factor for falls among
turbed sleep, oppositional behaviour, and auditory hallucinations the LD population, including people with DS. Nevertheless, there is
compared to people with dementia and LD of other causes (Cooper no research published looking at motor disorders and falls in older
and Prasher, 1998). Temple and Konstantareas (2005) have also people with DS and dementia.
reported that people with DS and dementia have less problem
Natural history
behaviour and delusions than patients with AD.
There is little research devoted to the natural history of dementia in
Other symptoms often associated with dementia in people people with DS. In the 15-year follow-up study of 92 people with
with Down syndrome DS mentioned in the section Prevalence, types of dementia, and
Epilepsy risk factors (Margallo-Lana et al., 2007), the mean age of onset of
Up to 84% of people with DS and dementia will develop seizures dementia was 55.5 years (range 45–74) and the mean age of death
(Puri et al., 2001). People with DS have a triphasic distribution of with dementia was 59.1 ± 9.6 years, with a survival period from
onset of dementia until death of 3.5 ± 2.2, range 1–8 years. The samples of blood from healthy blood donors in the same geograph-
majority of deaths were due to bronchopneumonia. Respiratory ical area.
tract infection is a common reported cause of death in older people It has been reported that E4 allele frequency of older DS patients
in general, in the general population with AD (Keene et al., 2001), is about half that of younger ones, suggesting premature death of
and has previously been reported as a common cause of death of people with DS of those with this allele (Folin et al., 2003). ApoE2
people with DS and dementia (Evenhuis, 1990). has been associated with increased longevity and decreased fre-
quency of dementia (Royston et al., 1994). The majority of studies
Risk factors for Alzheimer’s disease in Down syndrome
have found that the apoE allele distribution in people with DS is
Age
similar to that found in the general population without AD (Hardy
As in the general population, age is the strongest risk factor influ-
et al., 1994; van-Gool et al., 1995; Avramopoulos et al., 1996).
encing the age of onset of dementia in people with DS (Bush and
Several authors (Deb et al., 2000; Coppus et al., 2010) have found
Beail, 2004; cited in Zigman and Lott, 2007). More than 50% of
that on meta-analysis, the frequency of apoE4 was higher in adults
people with DS aged over 50 years develop dementia.
with dementia compared with adults with DS without dementia, but
Gender there was no significant reduction in the frequency of apoE2. They
The existing literature regarding rates of dementia according to concluded that apoE4 acts as a risk factor for age specific manifesta-
gender shows conflicting results. For example, Schupf and col- tion of AD in people with DS. Some researchers have reported that
leagues (1998) reported that, compared with women, men with DS apoE4 might influence the age of onset of dementia and mortality
are three times as likely to develop AD. Lai and colleagues (1999) risk in DS, even in the absence of dementia (Zigman et al., 2005;
found that women were 1.77 times as likely to dement as men at any Coppus et al., 2008).
given point in time (P = 0.04).
Gene polymorphisms
However, there seems to be a significant relationship between
A gene polymorphism in the prion protein gene has been reported
age of onset of menopause and the onset of dementia (Schupf et
to be associated with earlier decline in intellectual ability in adults
al., 2006; Coppus et al., 2010) in women with DS. Coppus and col-
with DS (Del Bo et al., 2003). Margallo-Lana and colleagues (2004)
leagues (2010) recently reported that early age at menopause is
reported a 13-year difference in the age at onset of dementia in DS
associated with a 1.8-fold increased risk of dementia.
associated with the number of tetranucleotide repeats on intron 7
APP, BACE-1,and BACE-2 alleles in the APP gene. The results suggested that APP is an impor-
APP is expressed at levels that are four- to five-fold higher in DS tant locus predicting the age at onset of dementia in people with
than in the general population (Beyreuther et al., 1993). Aß pep- DS.
tide is generated from the APP by sequential action of ß secretase
amyloid precursor protein cleaving enzymes 1 and 2 (BACE1 and Atypical karyotypes
BACE2) and γ secretase. It has been reported that APP C99, the Translocations, partial trisomies, and varying degrees of mosaicism
major ß secretase product, and Aß are increased in DS (Busciglio et are reported to be associated with better survival and lower risk of
al., 2002; Sun et al., 2006). The gene for BACE2 is located on chro- AD than the full trisomy 21 (Schupf 2002). There have been cases
mosome 21, probably within the obligate region for DS (Korenberg reported in the literature of people with DS of advanced age who
et al., 1990). BACE2 protein levels have been found higher in fetal died showing no signs of cognitive decline; for example, a case of
tissue of individuals with DS (Barbiero et al., 2003) and in fibrob- a woman who at 78 years of age was diagnosed with partial tri-
lasts of adults with DS, than in controls from the general popula- somy excluding the APP region who died at the age of 83 showing
tion (Motonaga et al., 2002; Barbiero et al., 2003). However, the no clinical symptoms of dementia or neuropathological lesions of
roles of BACE1 and BACE2 in the neuropathogenesis of AD in DS AD (Prasher et al., 1998). Other reports mentioned in the litera-
have been questioned (Sun et al., 2006; Cheon et al., 2008). It has ture include the cases of two women with 25% and 86% disomy for
been reported that APP overexpression seems to be absent during chromosome 21 who at the ages of 83 and 74, respectively, showed
the development of the DS brain up to weeks 18–19 of gestational no signs of clinical dementia (Chicoine and McGuire, 1997; W.B.
age and that its overexpression in adult DS brain could lead to dis- Zigman, personal communication, 2000).
turbance of normal function of APP, contributing to neurodegen- Disease modifiers
eration. The comparable expression of BACE1 and BACE2 in brains
The risk of dementia in DS increases with age. The majority of peo-
of adults with DS and during early fetal brain development would
ple with DS will have AD-like changes in their brains by the age of
speak against the hypothesis that increased ß secretase results in (or
40. However, it seems that not every individual with DS will inevi-
even underlies) increased production of amyloidogenic Aß frag-
tably develop dementia and there is a wide range of age of onset.
ments in people with DS (Cheon et al., 2008).
Head and Lott (2004) suggest that there may be additional provoca-
Apolipoprotein E4 tive factors in the disorder (e.g. apoE genotype, gender, individual
The literature about the effect of apolipoprotein E (apoE) on AD differences in Aβ deposition) acting in parallel with compensatory
in DS is inconclusive (Coppus et al., 2008). Bálint and colleagues events that would maintain brain function despite the accumulation
(2000) studied a group of 56 ethnically homogeneous Hungarian of AD pathology. Compensatory events reported in people with DS
people, looking at the apoE distribution in amniotic fluid of 15-week include growth response in the hippocampus in people with DS at
conceptuses with trisomy 21 from a genetic screening programme, ages immediately before the development of full-blown AD pathol-
thus including those that might have not reached full-term gesta- ogy (Head et al., 2003), and hypermetabolism in the temporal cor-
tion. The authors found no significant difference in the distribution tex of adults with DS, shown in PET scans prior to the onset of
of apoE alleles in the group of trisomy 21 fetuses compared with dementia (Haier et al., 2003). Zigman and Lott (2007) suggest that

(Oxford Textbooks in Psychiatry) Tom Dening, Alan Thomas-Oxford Textbook of Old Age Psychiatry-Oxford University Press (2013) PDF

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

(Oxford Textbooks in Psychiatry) Tom Dening, Alan Thomas-Oxford Textbook of Old Age Psychiatry-Oxford University Press (2013) PDF

Hochgeladen von

Copyright:

Verfügbare Formate

Oxford Textbook of

Old Age Psychiatry

Oxford Textbook of Community Mental Health

Contributors xv 12 Neuroimaging 177

case studies from anthropology 39

24 Memory assessment services 319 40 Delirium 529

59 Elder abuse 779 63 The law relating to mental capacity

Editors Carol Brayne

Jane Fossey John Holmes

Catherine Hatfield Christopher Kipps

Verena Heise Alexander Kurz

Julie Livingston Carole Proctor

Kay Wheat Gill Windle

for the degradation of proteins, and without the requirement for

and post-translational errors (accounting for as much as a third of Accrual Removal

Genetics and epigenetics Cellular Mechanisms of Ageing

Table 2.1 (Continued)

Test name Author Country Function

transition’, the older population is growing proportionally faster

century and carried on for approximately 100 years. As fertility

to explore the respondents’ symptoms and behaviour at interview, Dementia Men

Table 6.1 Major protein aggregates in neurodegenerative diseases

Hippocampal Sclerosis protein, an inhibitor of phospholipase D2, and an active participant

0.75 0.75 PD no dementia

changes contributing to the movement disorder characteristics of Acetylcholine—presynaptic changes

Dementia is a syndrome that can result from a variety of diseases Neuropathology of AD

Table 8.1 Genome-wide association studies of Alzheimer’s disease

Assessment in Residential Homes The Psychiatric Assessment

response to antidementia treatment; patients ‘typically’ decline Physical examination

Table 10.1 Discriminating features in common neurodegenerative diseases

—progressive nonﬂuent aphasia Mispronunciation of words with insertion of repeated or incorrect

◆ forgetting recent personal and family events (appointments, Working memory

Addenbrooke’s cognitive examination—ACE-R

Executive and Frontal Lobe Function

Fig. 10.4 Examples of poor visuospatial performance when

Neglect cube, interlocking pentagons, or constructing a clock-face with

Formal Neuropsychological Assessment Conclusion

Table 10.2 Associated neurological features in dementia

Stalevo tablets with associated drooling, and this should prompt a

Being expert in the art of the ‘nontouch’ examination tech-

◆ There is excess wax in the ear—patient’s ears need to be checked.

patient’s first language, consider the possibilities of aphasia (usually

Skin Suspicious bruising includes: thumb marks (usually on limbs);

Investigations in an uncooperative patient Learning point

phenothiazines muscle weakness, involuntary movements)

◆ Tremor is common in older people and may be atypical and diﬃcult to

Adrenal disease Urinary tract infections

Table 11.7 Common family concerns and likely explanation

Fig. 12.1 Computed X-ray tomography (CT).

Fig. 12.2 Structural MRI. Examples of (a) T1-weighted

Blood ﬂow magnetic resonance imaging

major direction of diffusion, can automatically reconstruct entire

(a) Gamma Camera (b) PET Camera

Fig. 12.6 Methods of determining site of

abnormal amyloid scans. A potential shortcoming of the ligands

Spatial resolution [mm]

Table 12.2 MRI used in the diagnostic criteria of some dementias

Abnormal areas FDG

Clinical Geriatrics and Psychopharmacology Compliance

b Yale et al., 2005

Gerontological factors Gerontological factors

• Reduced hepatic extraction, • nil

• Increased volume of • Induced microsomal enzymes