Clinical Characteristics of

Late-Onset Schizophrenia
and Delusional Disorder

by Ramzy Yassa and

Barbara Suranyi-Cadotte

Abstract
We compared 20 patients with
late-onset schizophrenia, 7 with
delusional disorder with hallucinations (paraphrenia), and 13
with delusional disorder without
hallucinations (late-onset paranoia). We found that these three
categories could be distinguished
from each other on some clinical
parameters. Late-onset schizophrenia was characterized by bizarre delusions; auditory hallucinations; to a lesser degree,
first-rank and negative symptoms; and premorbid personality
of the paranoid or schizoid type.
Paraphrenia was associated with
predominantly nonbizarre delusions, auditory hallucinations,
earlier onset of symptoms, and
paranoid or schizoid personality.
Paranoia (late-onset) was characterized by late onset of symptoms, nonbizarre delusions,
relatively intact premorbid personality, and an underlying
physical stratum.

Kraepelin (1919/1971) applied the

diagnosis "paraphrenia" to schizophrenia patients who developed
psychosis in later life 0este et al.
1991). Paraphrenic patients were
characterized by delusions and hallucinations of a paranoid nature.
Kraepelin introduced the term
"paranoia" to describe a chronic
illness characterized by wellorganized delusions without hallucinations, formal thought disorder,
or personality deterioration (Flint
et al. 1991).
Since Kraepelin's time, these two
diagnostic categories have been
poorly studied and their conceptual boundaries blurred. ICD-9
(World Health Organization 1978)
follows the definitions set by

Kraepelin, while DSM-III-R

(American Psychiatric Association
1987) defines delusional disorder
as the exhibition of wellsystematized delusions with or
without hallucinations; at the same
time, DSM-1II-R recognizes that
schizophrenia can occur for the
first time after the age of 45.
Thus, there are now two sets of
criteria that differentiate between
late-onset schizophrenia (paraphrenia) and late-onset paranoia
(delusional disorder). The main
problem seems to be the presence
or absence of hallucinations.
Studies on late-onset schizophrenia continue to include patients with delusions but no hallucinations (Roth 1955; Kay and
Roth 1961; Post 1966; Naguib and
Levy 1987). Hint et al. (1991) used
Kraepelin's criteria and compared
21 patients with late paraphrenia
with 12 patients with late-onset
paranoia. They found that cerebral
organicity is an important risk
factor for late-onset paranoia
that may contribute to a poor
prognosis.
The present study focuses on a
cohort of patients admitted to our
psychogeriatric unit during a
7-year period. The aim of the
study is twofold:
1. To compare the general
characteristics of late-onset paranoia and late-onset schizophrenia,
using Kraepelin's and DSM-III-R
criteria (i.e., to compare late-onset
schizophrenia, delusional disorder
with hallucinations, and delusional
disorder without hallucinations).
2. To provide a long-term fol-

Reprint requests should be sent to

Dr. B. Suranyi-Cadotte, Douglas Hospital Centre, 6875 LaSalle Blvd., Verdun, Quebec, H4H 1R3, Canada.

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at King's College London - Journals Dept on November 25, 2014

VOL 19, NO. 4, 1993

SCHIZOPHRENIA BULLETIN

702

Materials and Methods

The patient population consisted of
all admissions (n = 486: 159 men,
327 women) to our acute psychogeriatric unit (for patients age 65
years and older) during the 7-year
period beginning September 1,
1984. Our department is the only
inpatient unit serving a catchment
area of approximately 300,000
persons.
We used two sets of criteria to
diagnose our patients:
1. Kraepelin's and ICD-9 criteria,
in which late-onset paranoia patients present with delusions but
not hallucinations. Paraphrenia patients have both symptoms.
2. DSM-III-R criteria of delusional disorder with prominent
nonbizarre delusions and "nonprominent hallucinations." Lateonset schizophrenia patients have
active symptoms develop after
age 45.
Diagnoses were made during routine clinical interviews. All of the
patients were interviewed by each
author separately; for a patient to
be included in the study, the two
authors had to agree on the diagnosis. The definitions of bizarre
and nonbizarre delusions followed
the DSM-III-R criteria; that is,
nonbizarre delusions involve situations that occur in real life, such
as being followed, poisoned, infected, loved at a distance, having
a disease, or being deceived by
one's spouse or lover, while bizarre delusions involve a phenomenon that one's culture would
regard as totally implausible.
Negative symptoms were also ascertained by clinical interview.
Seven detailed case reports of late-

onset schizophrenia in this patient

group are included in a previous
4-year prevalence study (Yassa et
al., in press).
Exclusion criteria included evidence of organicity (Mini-Mental
State Exam [MMSE; Folstein et al.
1975] score of less than 25 at index admission) and conditions that
may mimic schizophrenia, such as
drug abuse, metabolic conditions,
or neurosyphilis.
After case identification, each
patient's family was interviewed to
determine the actual age at onset
of symptoms and to gain insight
into the premorbid personality of
each patient. In cases where there
was doubt, family members were
interviewed individually to assess
the accuracy of the information
provided.
In addition, the following tests
were conducted on each patient:
simultaneous multianalyzer chemistry including blood glucose, urea,
creatinine, calcium, phosphate, total
protein, albumin, cholesterol, total
bilirubin, uric acid, amylase, serum
glutamic peruvic transaminase,
serum glutamic oxalacetic transaminase, creatine kinase, alkaline
phosphatase, and lactate dehydrogenase; vitamin B12 and folic acid
levels; thyroid function tests;
electroencephalogram (EEG); and
serologic tests for syphilis. An
MMSE was carried out to detect
organicity.

Results
Of the 486 patients admitted to
the unit during the 7-year observation period, 40 were diagnosed as
having either late-onset schizophrenia or delusional disorder.
These patients were divided into
three diagnostic categories:
1. Group A consisted of late-

onset schizophrenia patients (n =

20) who met DSM-III-R criteria
for schizophrenia with an onset of
symptoms after age 45.
2. Group B consisted of patients
with delusional disorder with hallucinations (a DSM-III-R classification; n = 7). This diagnosis corresponds to paraphrenia as described
by Kraepelin and ICD-9. Patients
in this category exhibited nonbizarre, well-systematized delusions
accompanied by hallucinations.
3. Group C consisted of patients
with delusional disorder without
hallucinations (a DSM-III-R classification; n = 13). This diagnosis
corresponds to the category labeled
"late-onset paranoia" in ICD-9.
The nonbizarre, well-systematized
delusions of these patients were
not accompanied by hallucinations.
As seen in table 1, the
male:female ratio was almost identical in groups A and B. Women
were overrepresented in group C
in comparison with the other two
groups. Group C patients were
significantly older when first admitted than were patients in the
other two groups, although the
mean present age was similar in
the three groups.
Patients in group C were significantly more well-adjusted patients,
as reported by their families, than
were patients in group B (x2 =
4.9, df = 1, p < 0.05, with Yates'
correction); the difference between
patients in group A and B did not
reach significance (x2 = 1.5, rf/ = 1,
with Yates' correction). Fifty percent of the group A patients developed symptoms before age 60,
compared with 71 percent of
group B patients and 8 percent of
group C patients. The difference
between group A and C was statistically significant (x2 = 4.6, df =
1, p < 0.05, with Yates' correc-

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at King's College London - Journals Dept on November 25, 2014

lowup of these patients, using the

same diagnostic criteria.

VOL. 19, NO. 4, 1993

703

Group B
(n = 7)

Group C
(n = 13)

2:5

74.1 3.8
(70 - 80)
58.9 9.3
(47 - 73)2

1:12
77.3 7.2
(66 - 89)
71.3 9.0
(57-85) 1 2

16 13.1
(2 - 32)

15.3 8.2
(7 - 30)

5.8 4.2
(2-17)

3.9 3.3
(1 - 3 )
26 49.8
(1 - 207)3

3.1 2.0
(1-7)
20.3 33.6
(2 - 95)4

1.7 0.9
(1-4)
8.4 i 13.5
(1 - 48)3'4

Group A
(n = 20)
Male:female ratio
Age at interview, yr, mean SD
(range)
Age at first admission, yr, mean SD
(range)
Duration of illness from first hospitalization,
mo, mean SD
(range)
Hospitalizations
Total number
(range)
Duration, mo
(range)
Marital status, n
Single
Married
Divorced
Widowed
Educational status, n
None
Primary
Secondary
University
Premorbid personality, n (%)
Well-adjusted
Paranoid
Schizoid
Obsessive
'F =
*F =
3F=
F=

6.6,
8.4,
1.5,
1.3,

df
df
df
df

=
=
=
=

1,31,
1,18,
1,31,
1,18,

p
p
p
p

<
<
<
<

3:17
78.7 8.0
(70-102)
62.1 10.71

2
6
6
6

2
2
2
1

2
6
1
4

0
9
9
2

0
3
4
0

1
9
3
0

10
6
3
1

(50)
(30)
(15)
(5)

1
4
2
0

(14)
(57)
(29)

10
2

(77)
(15)

(8)

0.02.
0.009.
0.2, NS.
0 3 , NS.

tion), as was the difference between groups B and C (x2 = 6.0,

df = 1, p < 0.025). None of the
patients in groups A and C had a
positive family history of psychiatric disorder. On the other hand,
three patients in group B had a
positive family history (a mother,
two sisters) of paranoid disorder.
The Clinical Picture. The distribution of clinical manifestations

is shown in table 2. Apart from

the delusions and hallucinations
that formed the basis for our definitions of the different groups,
first-rank symptoms, loose associations, and negative symptoms were
described exclusively in group A
patients. Delusional disorder patients were not differentiated from
each other clinically. Depressive
symptoms were present in all
three groups of patients but they
never met DSM-III-R criteria.

These symptoms were mainly categorized as sad affect, lack of appetite, and tearfulness during the
interview.
Concomitant Physical Disorders. More patients in group C
had concomitant physical disorders
than in the other two groups
(table 3). Hypertension was significantly more likely to be present in
group C patients than in group A
patients (x2 = 4.2, df = 1, p <

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at King's College London - Journals Dept on November 25, 2014

Table 1. Demographic characteristics of the patient population

SCHIZOPHRENIA BULLETIN

704

Manifestation
Hallucinations
Auditory
Visual
Tactile
Delusions
Bizarre
Persecutory
Ideas of reference
Religious
First-rank symptoms
Loose associations
Negative symptoms
Homicidal tendencies
Suicidal tendencies
Depressive symptoms

Group A
(n = 20)

Group B
(n = 7)

20(100)
2(10)
2(10)

7(100)
0
0

20(100)
17(85)
17(85)
2(10)
10(50)
4(20)
3(15)
3(15)
2(10)
3(15)

7(100)
4(57)
0
0
0
0

Group C

0
0
0
0

13(100)
8(62)
1 (8)
0
0
0

1 (14)

1 (8)

2 (29)

3(23)

Note.Values are n (%).

Table 3. Concomitant physical disorders

Disorder
None
Cardiovascular
Hypertension
Arteriosclerotic
heart diseases
Abdominal aneurysm
Diabetes mellitus
Chronic obstructive
lung disease
Cancer
Alcohol abuse
Sensory deprivation
Cataract
Deafness
Electroencephalogram
Normal
Abnormal

Group A
n = 20)

Group B
(" = 7)

Group C
(n = 13)

7(35)

2(29)

2(15)

4(20)

2(29)

8(61)

5(25)
1 (5)
2(10)

0
0

3(23)

1 (29)

2(15)

2(10)
2(10)
1 (5)

2(29)

1 (8)
1 (8)
1 (8)

1 (5)
1 (5)

0
1

17
3

0
0

5
2

2(15)
0
10
3

Note.Values are n (%).

0.05). On the other hand, the difference between the two delusional
disorder groups, B and C, was not

statistically significant (x2 = 0.9,

df = 1).
Two patients in group C had a

recent history of cerebrovascular

accidents. Sensory deprivation was
not differentially distributed among
the three groups.
Electroencephalogram disturbances were present equally often
in the three groups. Diffuse slowwave disturbances without localization were present in two group A
patients, one group B patient, and
two group C patients. Slow-wave
disturbances with localization were
present in one group A patient
(right temporal area), one group B
patient (left temporal area), and
one group C patient (anterior
area).
Followup. Only followup information for the index observation
period is presented, even if patients were admitted before the index period.
Our definitions of status at followup are as follows:
1. Stablethe patient showed no
evidence of recurrence of symptoms during the index observation
period following the index admission, with or without neuroleptic
treatment.
2. Intermittentthe patient's
original symptoms reappeared during the index observation period,
necessitating a change in neuroleptic dosage but not readmission.
3. Readmittedthe patient's condition necessitated readmission.
Of the 20 patients in group A, 8
(40%) were first admitted during
the index period. They remained
in the hospital during the index
admission for a mean standard
deviation of 6.1 8.1 months
(range = 2-26). Five of these patients (two died and one was lost
to followup) were followed for 3.8
2 . 1 years (range = 2-7) after the
index admission. Three were discharged to their homes and two to

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at King's College London - Journals Dept on November 25, 2014

Table 2. Clinical manifestations

705

VOL 19, NO. 4, 1993

were admitted during the index

period. The mean duration of the
index admission was 8.3 15.7
months (range = 1-48). One patient was lost to followup, one
died, and one remained in hospital
during the observation period. Of
the seven group C patients who
were followed for a mean of 4.6
1 . 9 years (range = 2-7) following index admission, one was discharged home, three to various
foster homes, and three to various
nursing homes. At followup, six
remained stable and one required
readmission. The mean neuroleptic
dosage was 114.3 91.1 mg
chlorpromazine equivalents per
day (range = 25-300).
Of the three group C patients
who were admitted before the index observation period, two stayed
for 1 month each and the third
for 2 months during the index admission period. One died, one was
lost to followup, and one was
readmitted during the index observation period.
In conclusion, after excluding
those who died and those who
were lost to followup, we found
that 4 of 15 (26%) patients in
group A, 2 of 5 (40%) patients in
group B, and 6 of 9 (66%) patients
in group C remained stable during
the index observation period. None
of these results is statistically significant, however.
On the other hand, significantly
more patients in group A than in
group C stayed in the hospital or
were readmitted during the observation period (10 of 15 vs. 3 of 9;
X2 = 4.9, df = 1, p < 0.05).
However, if we compare the status of all the.patients who were
admitted during the index observation period (n = 19) with the status of all those who were admitted before the observation period
( = 21), excluding patients who

died or were lost to followup, we

find that 10 of 14 (71%) in the
former group, compared with 2 of
15 (13%) in the latter group, remained stable during the index
observation period (x2 = 7.96, df =
1, p < 0.005, with Yates'
correction).

Discussion
This comparison of patients in
three diagnostic categorieslateonset schizophrenia (group A), delusional disorder with hallucinations (group B), and delusional
disorder without hallucinations
(group C)indicates that there are
characteristics that can differentiate
these patients. Patients in groups
A and B developed their symptoms significantly earlier than did
group C patients. In addition,
groups A and B contained more
paranoid and schizoid premorbid
personalities, whereas group C patients were more often described
as "well adjusted."
Apart from the differences in
delusions and hallucinations that
were used as a basis for diagnosing the three categories, only
group A patients had first-rank
and negative symptoms similar to
those seen in younger schizophrenia patients. However, there
was no difference in symptom distribution among patients with delusional disorder.
Howard et al. (1992) described
first-rank Schneiderian (Schneider
1959) symptoms in 39 percent of
41 patients diagnosed as suffering
from late-onset schizophrenia.
Pearlson et al. (1989) compared patients with early- and late-onset
schizophrenia and found no difference in the prevalence of firstrank symptoms, whereas negative
symptoms were less common in ,

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at King's College London - Journals Dept on November 25, 2014

a nursing home. Three were stable, one showed intermittent

symptoms, and one was readmitted. All patients continued to
receive neuroleptics, mainly
haloperidol. The mean dosage for
this population was 231.3 175.1
mg (range = 50-500) chlorpromazine equivalents (Davis 1976) per
day.
Twelve of the group A patients
were first admitted before the index observation period. The mean
duration of their index admission
was 13.1 20.6 months (range =
2-60). Two were lost to followup,
three remained in hospital during
most of the observation period,
and seven were followed for a
mean duration of 5.9 1 . 2 years
(range = 4-7) following the index
admission. Only one of the seven
remained stable; six were readmitted. The mean neuroleptic dosage
was 260 301.4 mg chlorpromazine equivalents per day (range =
25-750).
One of the seven patients in
group B was first admitted during
the index observation period. She
stayed for 2 months in the hospital and was discharged to a nursing home. She remained stable
during the 7-year observation
period. She received 0.5 mg
haloperidol per day.
Six patients from group B were
admitted before the index observation period. The mean duration of
the index admission was 11.5
23.8 months (range = 1-60). One
patient died, one was lost to followup, and one remained in the
hospital during the observation
period. Three were followed after
the index admission. One went
home and two to a nursing home.
One remained stable, one had intermittent symptoms, and the third
was readmitted to hospital.
Ten of the patients in group C

SCHIZOPHRENIA BULLETIN

706

1. Patients with late-onset schizophrenia had worse outcomes, in

terms of more admissions during
the index observation period, than
did patients with delusional disorder with or without hallucinations.
2. Patients with earlier onset of
psychotic symptoms seemed to
fare worse than those who developed psychotic symptoms later in
life (3 of 14 admitted vs. 13 of 15,
X2 = 10.0, df = 1, p < 0.005).
Thus, we find that these three
categories can be distinguished
from each other by some parameters. Late-onset schizophrenia is a
condition characterized by bizarre
delusions, auditory hallucinations,
and, to a lesser degree, first-rank
and negative symptoms. Premorbid
personality is commonly of the
paranoid or schizoid type. Delusional disorder with hallucinations

(paraphrenia) is characterized by
predominantly nonbizarre delusions
and auditory hallucinations. It
shares with late-onset schizophrenia an earlier onset of symptoms
and more paranoid or schizoid
personalities. Delusional disorder
without hallucinations (late-onset
paranoia) is characterized primarily
by late onset of symptoms, nonbizarre delusions, relatively intact
premorbid personality, and an underlying physical stratum. These
three conditions should be differentiated from each other in any
clinical or scientific study.

References
American Psychiatric Association.
DSM-III-R: Diagnostic and Statistical Manual of Mental Disorders. 3rd
ed., revised. Washington, DC: The
Association, 1987.
Davis, J.M. Dose equivalence of
antipsychotic drugs. In: Matthysse,
S., and Kety, S.S., eds. Catecholamines and Their Enzymes in the
Neuropathology of Schizophrenia.
Elmsford, NY: Pergamon Press,
1976. pp. 65-69.
Flint, A.J.; Rifat, S.L.; and Eastwood, R. Late-onset paranoia: Distinct from paraphrenia? International Journal of Geriatric
Psychiatry, 6:103-109, 1991.
Folstein, M.F.; Folstein, S.E.; and
McHugh, P.R. Mini-Mental State: A
practical method for grading the
cognitive state of patients for the
clinician. Journal of Psychiatric Research, 12:189-198, 1975.
Herbert, M.E., and Jacobson, S.
Late paraphrenia. British Journal of
Psychiatry, 113:461^69, 1967.
Howard, R.J.; Forstl, H.; Naguib,
M.; Burns, A.; and Levy, R. First
rank symptoms of Schneider in
late paraphrenia: Cortical structural

correlates. British Journal of Psychiatry, 160:108-109, 1992.

Jeste, D.V.; Manley, M.; and
Harris, M.J. Psychoses. In:
Sadavoy, }.; Lazarus, L.W.; and
Jarvik, L.F., eds. Comprehensive Review of Geriatric Psychiatry. Washington, DC: American Psychiatric
Press, 1991. pp. 353-368.
Kay, D.W., and Roth, M. Environmental and hereditary factors in
the schizophrenias of old age and
their bearing on the general problem of causation in schizophrenia.
Journal of Mental Science, 107:649686, 1961.
Kraepelin, E. Dementia Praecox and
Paraphrenia. (1919) Translated by
R.M. Barclay. Huntington, NY:
Robert E. Kreiger Publishing Company, 1971.
Munro, A. Delusional disorders:
Etiological and taxonomic considerations: I. The possible significance of brain factors in the etiology of delusional disorders.
Canadian Journal of Psychiatry,
33:171-174, 1988.
Naguib, M., and Levy, R. Late
paraphrenia: Neuropsychological
impairment and structural brain
abnormalities on computed tomography. International Journal of
Geriatric Psychiatry, 2:83-90, 1987.
Pearlson, G.D.; Kreger, L.; Rabins,
P.V.; Chase, G.A.; Cohen, B.;
Wirth, J.B.; Schlaepfer, T.B.; and
Tune, L.E. A chart review study of
late-onset and early-onset schizophrenia. American Journal of Psychiatry, 146:1568-1574, 1989.
Post, F. Persistent Persecutory States
of the Elderly. Oxford, England:
Pergamon Press, 1966.
Roth, M. The natural history of
mental disorder in old age. Journal
of Mental Science, 101:281-301,
1955.

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at King's College London - Journals Dept on November 25, 2014

patients with late onset (7%) than

in those with early onset (22%).
These figures are similar to our
findings.
We were unable to confirm a
high percentage of sensory deprivation in our patients, as previously reported for a similar population (Herbert and Jacobson
1967), but we found that hypertension and a history of cerebrovascular accidents were more common
in patients in group C (delusional
disorder without hallucinations)
than in the other two groups. This
result confirms the findings of
some authors (Seidel 1977; Munro
1988; Flint et al. 1991) but not
others (Van Valkenburg and
Winokur 1984).
It is difficult to generalize about
course and prognosis in these disorders from our study because of
the relatively small number of subjects. However, we found two
trends in the outcomes of our
patients:

707

VOL 19, NO. 4, 1993

International Classification of Dis-

ogy. Translated by M.W. Hamilton.

New York, NY: Grune & Stratton,
1959.
Seidel, U.P. Hypertension and paranoid states. [Letter] Lancet, 1:906,
1977.
Van Valkenburg, C , and Winokur,
G. Hypertension and paranoia.

eases. 9th ed. Geneva, Switzerland:

The Organization, 1978.
Yassa, R.; Dastoor, D.; Nastase, C;
Camille, Y.; and Belzile, L. The
prevalence of late-onset schizophrenia in a psychogeriatric popu-

American Journal of Psychiatry,

141:999-1000, 1984.
World Health Organization. ICD-9:

Back Issues
Available

lation. Journal of Geriatric Psychiatry and Neurology, in press.

The Authors
Ramzy Yassa, M.D., F.R.C.P.(c)

Several back issues of the Schizophrenia Bulletin are still available to

requesters:
Schizophrenia Bulletin, Vol. 17, No. 1,
1991
(Issue theme: Schizophrenia and
Related Disorders in DSM-W)
Schizophrenia Bulletin, Vol. 17, No. 4,
1991
(Featured topics: Diagnosis and
Neuropathology)
Schizophrenia Bulletin, Vol. 18, No. 1,
1992
(Issue theme: Schizophrenia Research in Japan)

(deceased), was Professor of Psychiatry, McGill University,

Montreal, and Staff Psychiatrist,
Douglas Hospital Centre, Quebec,
PQ, Canada; and Barbara SuranyiCadotte, B.E., M.D., F.R.C.P.(c) is
Psychiatrist, Psychogeriatric Department of Douglas Hospital Centre,
Quebec, and Associate Professor
Department of Psychiatry, McGill
University, Montreal, Quebec,
Canada.

Schizophrenia Bulletin, Vol. 18, Nos.

2 and 3, 1992
(Issue theme: First-Episode
Psychosis)
If any or all of these issues are
missing from your collection, let
us know, and we will send you
a copy free of charge. Requests
should be sent to the following
address:
Research Projects and Publications Branch
National Institute of Mental
Health
5600 Fishers Lane, Rm. 18C-06
Rockville, MD 20857

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at King's College London - Journals Dept on November 25, 2014

Schneider, K. Clinical Psychopathol-