Epilepsy

Psychological treatments for epilepsy (Review)
Ramaratnam S, Baker GA, Goldstein LH
This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library 2011, Issue 3 http://www.thecochranelibrary.com
Psychological treatments for epilepsy (Review) Copyright 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AUTHORS CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.1. Comparison 1 Relaxation therapy versus control, Outcome 1 50% or greater reduction in seizure frequency. Analysis 1.2. Comparison 1 Relaxation therapy versus control, Outcome 2 Seizure free. . . . . . . . . . . Analysis 1.3. Comparison 1 Relaxation therapy versus control, Outcome 3 Reduction in seizure frequency. . . . . Analysis 2.1. Comparison 2 Cognitive behavioural therapy versus control, Outcome 1 50% or greater reduction in seizure frequency (Group CBT). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 2.2. Comparison 2 Cognitive behavioural therapy versus control, Outcome 2 Seizure free (Group CBT). . Analysis 2.3. Comparison 2 Cognitive behavioural therapy versus control, Outcome 3 Effect of Individual/Group CBT on seizures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 3.1. Comparison 3 Galvanic skin response biofeedback, Outcome 1 50% or greater reduction in seizure frequency. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 4.1. Comparison 4 Cognitive behavioural therapy versus yoga, Outcome 1 Seizure Free. . . . . . . . Analysis 4.2. Comparison 4 Cognitive behavioural therapy versus yoga, Outcome 2 50% or greater reduction in seizure frequency. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 4.3. Comparison 4 Cognitive behavioural therapy versus yoga, Outcome 3 50% or greater reduction in seizure duration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FEEDBACK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WHATS NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 2 2 3 3 5 9 10 11 13 29 30 31 31 32 33 34 35 35 36 36 36 38 39 39 40 40 40
[Intervention Review]
Psychological treatments for epilepsy

Sridharan Ramaratnam1 , Gus A Baker2 , Laura H Goldstein3
1 Department
of Neurology, Apollo Hospitals, Chennai, India. 2 University Department of Neurological Science, Clinical Sciences Centre for Research & Education, Liverpool, UK. 3 Department of Psychology, Institute of Psychiatry, London, UK Contact address: Sridharan Ramaratnam, Department of Neurology, Apollo Hospitals, 21 Greams Lane, Off Greams Road, Chennai, Tamil Nadu, 600006, India. rsridharan52@gmail.com. rsridharan@vsnl.com. Editorial group: Cochrane Epilepsy Group. Publication status and date: New search for studies and content updated (no change to conclusions), published in Issue 3, 2011. Review content assessed as up-to-date: 14 February 2011. Citation: Ramaratnam S, Baker GA, Goldstein LH. Psychological treatments for epilepsy. Cochrane Database of Systematic Reviews 2008, Issue 3. Art. No.: CD002029. DOI: 10.1002/14651858.CD002029.pub3. Copyright 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
ABSTRACT Background Psychological interventions such as relaxation therapy, cognitive behaviour therapy, bio-feedback and educational interventions have been used alone or in combination in the treatment of epilepsy, to reduce the seizure frequency and improve the quality of life. Objectives To assess whether the treatment of epilepsy with psychological methods is effective in reducing seizure frequency or leads to a better quality of life (QOL), or both. Search methods We searched the Cochrane Epilepsy Group Specialized Register (11 January 2011), the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library Issue 4, 2010) and MEDLINE (1948 to December 2010). No language restrictions were imposed. We checked the reference lists of retrieved studies for additional reports of relevant studies. Selection criteria Randomized or quasi-randomized studies assessing one or more types of psychological or behaviour modication techniques for people with epilepsy. Data collection and analysis Two review authors independently assessed the trials for inclusion and extracted data. Primary analyses were by intention to treat. Outcomes included reduction in seizure frequency and quality of life. Main results We found three small trials (total 50 participants) of relaxation therapy. They were of poor methodological quality and a meta-analysis was therefore not undertaken. No study found a signicant effect of relaxation therapy on seizure frequency. Two trials found cognitive behavioural therapy (CBT) to be effective in reducing depression among people with epilepsy and a depressed affect, whilst another did not. Two trials of CBT found improvements in QOL scores. One trial of group cognitive therapy found no signicant effect on seizure frequency while another trial found a statistically signicant reduction in seizure frequency as well as seizure index (product of seizure frequency and seizure duration in seconds) among participants treated with CBT.
Psychological treatments for epilepsy (Review) Copyright 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. 1
One small trial compared CBT with yoga and found similar rates of seizure freedom and 50% responder rates at the end of one year. Compared to pretest scores, the CBT group showed an improvement in the World Health Organization quality of life instrument, short version (WHO QOL-BREF) (P 0.01) while the yoga group had an improvement in their QOL according to the satisfaction with life scale (SWLS) (P 0.05). Two trials of combined relaxation and behaviour therapy, one of electroencephalographic (EEG) bio-feedback and four of educational interventions did not provide sufcient information to assess their effects on seizure frequency. One small study of galvanic skin response biofeedback reported signicant reduction in seizure frequency. Combined use of relaxation and behaviour modication was found benecial for anxiety and adjustment in one study. In one study EEG bio-feedback was found to improve cognitive and motor functions in individuals with the greatest seizure reduction. Educational interventions were found to be benecial in improving knowledge and understanding of epilepsy, coping with epilepsy, compliance to medication and social competencies. Two trials investigated the effects of a self management program on QOL measures. One found no signicant differences in overall QOL except for an improvement in the QOL in persons with epilepsy (QOLIE-89) role limitations - emotional score. The other study found no signicant benets in any of the outcomes studied except for more positive attitudes about the outcomes associated with taking medications. Authors conclusions In view of methodological deciencies and the limited number of individuals studied, we have found no reliable evidence to support the use of psychological treatments and further trials are needed. Since the last version of this review, none of the new relevant studies have provided additional information to change the conclusions.
PLAIN LANGUAGE SUMMARY Psychological treatments for epilepsy No reliable evidence supports psychological treatments for people with epilepsy. Psychological interventions are used in attempts to reduce seizure frequency, improve quality of life and avoid adverse effects of drugs among people with epilepsy. Of the trials found, some were of poor methodological quality whilst others had contradictory results. We found no evidence that relaxation therapy, cognitive behaviour therapy, electroencephalographic (EEG) or galvanic skin response biofeedback, used alone or in combination, have an impact on seizures or quality of life. Educational interventions may reduce anxiety, improve medication compliance and social competency but further well designed trials are needed.
BACKGROUND
It has been postulated that there is a relationship between behavioural, physiological and psychological states and the probability of seizure occurrence (Fenwick 1992). Epilepsy is often associated with anxiety, depression, behavioural problems and cognitive dysfunction. These co-morbidities may reect a common single cause or be due to the stigma and social handicaps associated with epilepsy, or a combination of both. Psychological interventions such as psychotherapy; individual, group or family counselling; progressive relaxation therapy and cognitive behaviour therapy have been used to treat psychopathology associated with epilepsy
(Davis 1984; Miller 1994). According to anecdotal reports, such treatments not only alleviate anxiety, depression and behavioural problems but also reduce the frequency of seizures. Epileptic seizures have been known to be precipitated by psychological triggers (internal precipitants), such as stress, anxiety, anger and emotions, as well as by mental tasks and thoughts (Betts 1992; Fenwick 1994; Temkin 1984). In reex epilepsies such as musicogenic, photogenic, movement induced, eating or reading epilepsy, external factors may precipitate a seizure. Self induction of seizures by some people with epilepsy (waving a hand in front of the eyes or
2
blinking to induce photosensitive seizures) provides evidence that some people are aware of stimuli that precipitate their seizures. There are anecdotal reports of people with epilepsy using behavioural methods to avoid seizures, for example an individual may recognize precipitating factors or prodromal symptoms of a seizure and initiate countermeasures (Fenwick 1994; Pritchard 1985; Wolf 1997). Hence the question arises as to whether psychological and behavioural methods can be used to treat epilepsy. The broad aims of psychological approaches for treating epilepsy include (1) increasing the seizure threshold by modulating electroencephalographic (EEG) activity or altering the level of arousal; (2) modulating seizure precipitating factors; and (3) preventing the spread of epileptic activity and thus the generalization of the seizure. The psychological methods of treatment of epilepsy that have been experimentally studied include rstly, reward management consisting of overt reward, covert reward, denial of reward, punishment programmes and relief avoidance; secondly, self control strategies that allow the individual to gain control of seizure activity by using cognitive processes (Cabral 1976; Dahl 1988) comprise of (a) self (patient) identied strategies to inhibit seizure activity, (b) relaxation (progressive muscle relaxation), (c) desensitization to the occurrence of a seizure by exposure to seizureprovoking situations, (d) avoidance of seizure precipitating stimuli, (e) psychotherapy - individual, group or family, and (f ) autohypnosis. Thirdly, psycho-physiological approaches have been attempted and include (a) classical conditioning, habituation and extinction of seizure precipitating factors (Forster 1968; Forster 1969), (b) EEG bio-feedback training to modulate 12 to 16 Hz sensorimotor rhythm, or alpha activity or slow cortical potentials (Kuhlman 1978; Lubar 1981; Quy 1979; Sterman 1980a), and (c) diaphragmatic breathing with end-tidal per cent carbon dioxide (CO2 ) bio-feedback (Fried 1990). Fourthly, are miscellaneous approaches such as stress-management programs, physical training (Nakken 1990), neck or body massage, assertive therapy, rational emotive therapy, psychotherapy, and counselling pertaining to the vocational, educational, genetic and marital problems of the individual. A fth approach is a combination of the above measures individualized to a particular individual (Andrews 1992; Reiter 1987). Previously published critical reviews on the use of psychological treatments for epilepsy have highlighted the methodological deciencies in many of the studies (Goldstein 1990; Goldstein 1997; Krafft 1982; Mostofsky 1993). These deciencies include recruiting too small a number of participants; poor operational denitions; a lack of controls, randomization or blinding; highly selected patient groups; lack of documentation of inter-observer agreements; and poor descriptions of the interventions. This is the rst systematic review to examine the evidence from randomized controlled trials investigating psychological treatments.
Glossary of some terms used in this review Washington PsychoSocial Inventory (WPSI): this is a standardized battery of tests to assess the adjustment in various spheres (measure of psychosocial difculties) in persons with epilepsy. Minnesota Multiphasic Personality Inventory (MMPI): this consists of a battery of standardized tests to assess the personality (psychopathology). Becks Depression Inventory (BDI): standardized scale to assess depression. Depression Adjective Check List (DACL): scale to assess depression. Relaxation therapy: a technique that trains people to control muscle tension. Cognitive behavioural therapy (CBT) (also known as stress management training): includes those interventions designed to teach skills for identifying and controlling stress and minimizing the effects of stress. This is a broad category which includes various self control strategies wherein the individual is trained to use their intellectual experience to correct and retrain faulty behaviour. Biofeedback: physiological responses such as heart rate, skin conductivity (GSR) or brain wave pattern (EEG) are monitored and visual or auditory on-line feedback is provided to the individual to help him or her actively control the physiological response. Galvanic skin response (GSR): an accessible and sensitive index of peripheral sympathetic nervous activity, reecting peripheral autonomic change and alterations in skin conductance. EEG bio-feedback: a technique wherein the individual is taught to regulate their EEG activity and to produce certain EEG waves which are believed to increase the threshold for seizures. WHO QOL-BREF: World Health Organization quality of life instrument, short version SWS: Satisfaction with life scale QOLIE 89: a validated disease specic instrument to assess the quality of life in persons with epilepsy. QOLIE-89 contains 17 multi-item measures of overall quality of life; emotional well-being; role limitations due to emotional problems; social support; social isolation; energy or fatigue; worry about seizure; medication effects; health discouragement; work, driving and social function; attention and concentration; language; memory; physical function; pain; role limitations due to physical problems; and health perceptions.
OBJECTIVES
To examine the effects of psychological treatments for people with epilepsy.
METHODS
3
Criteria for considering studies for this review
more detailed description of this activity is given in the Specialized register section of the Cochrane Epilepsy Group module. In addition, we carried out searching as follows. (1) Electronic databases We searched the following databases. There were no language restrictions. (a) The Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2010, Issue 4) using the search strategy outlined in Appendix 1. (b) MEDLINE (Ovid) (1948 to December week 4, 2010) using the search strategy outlined in Appendix 2. (2) References from published studies
Types of studies Randomized controlled trials (RCTs) or quasi-randomized studies (for example where the randomization is according to the day of the week or date of birth). The studies may have been single or double blind, or unblinded. Types of participants Men, women and children of any age with any type of epilepsy, with or without learning disabilities, who were either taking or not taking antiepileptic drugs (AEDs). Types of interventions Psychological or behavioural modication therapies such as the use of relaxation therapy, cognitive behavioural therapy, bio-feedback, counselling, suggestion, hypnotherapy, conditioning, systematic desensitization, behavioural countermeasures at seizure onset applied by the patient or another person, physical therapies, massage, aromatherapy, music or dance therapy. These interventions may have been given singly or in combination, either alone or as an add on to AEDs. Types of outcome measures
We reviewed the reference lists of retrieved studies to search for additional reports of relevant studies. (3) Other sources We contacted colleagues to ask if they were aware of any studies that we had missed. The identied abstracts were independently checked by two review authors (Sridharan Ramaratnam and Gus Baker) to assess their relevance for inclusion in the review.
Data collection and analysis

Trials were independently assessed for inclusion by two review authors (Sridharan Ramaratnam and Gus Baker) with disagreements resolved by mutual discussion. The same two review authors independently extracted the following data, and again any differences of opinion were resolved by mutual discussion. (1) Study methods: (a) design (e.g. parallel or crossover design); (b) randomization method (allocation concealment and list generation); (c) blinding method. (2) Participants: (a) number (total/per group); (b) age and sex distribution; (c) seizure type and epilepsy syndrome; (d) duration of epilepsy; (e) aetiology of epilepsy; (f ) presence or absence of learning disability. (3) Type of intervention and control. (4) Duration of follow up. (5) Outcome data, as described earlier. We reported the results for each intervention separately. In view of the poor quality of included studies we made no attempt to undertake a meta-analysis but results for individual studies are given.
4
Primary outcome measures
Seizure reduction (1) A 50% or greater reduction in seizure frequency (2) Seizure free (3) Percentage change in seizure frequency Secondary outcome measures (1) Quality of life (QOL) (2) Seizure severity, provided a standardized and validated scale was used
Search methods for identication of studies

We searched the Epilepsy Group Specialized Register (11 January 2011). This register contains reports of trials identied from regular searches of the Cochrane Central Register of Controlled Trials (CENTRAL) and MEDLINE. Relevant reports are also identied by handsearching selected journals and conference proceedings. A
Where the interventions were the same or similar enough, we had planned to synthesize results for seizure control in a meta-analysis provided that seemed appropriate and there was no important clinical or statistical heterogeneity. We had planned to assess clinical heterogeneity by examining the distribution of important prognostic variables between studies (see Methods of review: section 2(a) to (f )). We planned to check statistical heterogeneity using a Chi2 test (P < 0.1 indicating heterogeneity). For dichotomous data we quoted Peto odds ratio with 95% condence intervals (CIs), and used weighted mean difference (WMD) for continuous data. In view of the difculty in combining neuropsychological data from various studies, we have tabulated results rather than attempt to combine results in a meta-analysis. In view of the heterogeneous approach to these outcomes in the studies included, we have not explored the possibility and utility of undertaking a meta-analysis.
Lundgren 2008 compared acceptance concomitant therapy with yoga among 18 adults (12 men and 6 women) with refractory epilepsy.
Bio-feedback and epilepsy Lantz 1988 studied the effect of EEG bio-feedback in a trial on 24 adults randomized to three groups: contingent EEG bio-feedback, non-contingent feedback, and no intervention controls. Nagai 2004 investigated the effects of galvanic skin response (GSR) biofeedback training or sham feedback on seizure frequency in 18 patients with treatment resistant epilepsy.
Educational interventions and epilepsy Four studies (Helgeson 1990; Lewis 1990; May 2002; Olley 2001) assessed the outcome of educational interventions on psychosocial functioning. Helgeson 1990 tested the effect of a two-day psycho-educational program (Sepulveda Epilepsy Education) among 100 adults with epilepsy. Lewis 1990 investigated the effect of a child centred, family focussed, educational program on 252 children with epilepsy, aged between 7 and 14 years. Self competence, changes in the childrens knowledge about seizures, changes in the childrens and their parents behaviour were studied. Olley 2001 evaluated the efcacy of a two-day modular didactic psycho-educational program on adjustment to epilepsy, stigma, psycho-neurotic traits, depression, and knowledge of epilepsy among adult Nigerian patients. May 2002 studied the effect of a modular didactic educational program (MOSES: Modular Service Package Epilepsy) on health related quality of life, self esteem, depression, restriction in daily life, epilepsy related fear, stigma, mobility and leisure, knowledge of epilepsy, coping with epilepsy, adaptation to epilepsy, and contentedness with drug therapy among adult German speaking patients from 22 epilepsy centres in Germany, Austria and Switzerland. Apart from one study (May 2002) which assessed the seizure frequency categorized in a 0 to 5 scale (0 = no seizures, 5 = 1 or more seizure/day), the other studies did not investigate the outcome of their interventions on seizures.
RESULTS
Description of studies
See: Characteristics of included studies; Characteristics of excluded studies; Characteristics of studies awaiting classication. Relaxation therapy and epilepsy Three studies investigated the effect of relaxation therapy on seizure control (Dahl 1987; Puskarich 1992; Rousseau 1985). In total these studies included 50 adults with uncontrolled epilepsy (23 in the treatment group and 27 controls; 18 men and 32 women). One study enrolling 32 individuals (age and gender composition not stated) evaluated the effect of relaxation therapy on psychosocial functioning (Snyder 1983). Cognitive behavioural therapy and epilepsy We found ve studies evaluating the effect of cognitive behavioural therapy (CBT) on seizure control or psychological functioning, or both (Davis 1984; Lundgren 2006; Lundgren 2008; Martinovic 2006; Tan 1986). Davis 1984 studied the effect of CBT among 15 adults with epilepsy who were also depressed. Tan 1986 investigated the efcacy of group CBT for the alleviation of psychosocial problems and reduction of seizures among 30 adults with epilepsy. Lundgren 2006 investigated the effects of acceptance concomitant therapy for drug refractory epilepsy among 27 South African adults who were institutionalised or day workers in a centre for epilepsy. Martinovic 2006 studied 30 adolescents with newly diagnosed epilepsy and subthreshold depression who were given either a cognitive behavioural intervention or treatment with counselling as usual; they assessed the effects on depression and quality of life.
Relaxation and behavioural therapy and epilepsy The combined use of relaxation and behavioural therapy was evaluated in two studies. Dahl 1985 investigated the effect of a broad spectrum behaviour modication therapy on seizures among 18 children with uncontrolled epilepsy; the children were randomized to three groups: behaviour modication, attention control, and a control group. Sultana 1987 studied 150 adults with uncontrolled epilepsy who were randomized into a treatment group which received Jacobsons muscle relaxation and behavioural therapy and a control group in a 2:1 ratio. The outcomes studied included seizure frequency as well as psychological measures. Self management programs and epilepsy
5
Two studies (Dilorio 2009; Pramuka 2007) investigated the effects of self management programs for people with epilepsy. Dilorio 2009 studied the utility of a telephone based self management program among 15 men and 7 women with epilepsy. Pramuka 2007 investigated the effects of self management program among 55 adults with epilepsy. These studies focussed on quality of life; seizure outcomes were not assessed. The details of the methods, participants, interventions and outcomes assessed in these studies are given in the table Characteristics of included studies.
behavioural therapy in people with epilepsy and a depressed affect (Davis 1984) the concomitant antidepressant therapy and the intervention given to controls were not described. Lundgren 2008 randomized their participants using a computerized randomization table but did not use blinding. The baseline seizure frequency in the acceptance concomitant therapy (ACT) group was higher compared to the yoga group, probably due to small sample size.
Bio-feedback and epilepsy The published report of the study by Lantz 1988 did not state the randomization method. The study on GSR biofeedback (Nagai 2004) was a single blind (patients were blinded) randomized controlled study where the allocation was by random number tables, determined at trial onset.
Other therapies and epilepsy We did not nd any randomized studies that investigated the effects of counselling, suggestion, hypnotherapy, conditioning, systematic desensitization, behavioural countermeasures, physical therapies, massage, aromatherapy, music or dance therapy.
Educational interventions and epilepsy
Risk of bias in included studies
Relaxation therapy and epilepsy
(1) Relaxation therapy and seizure control Two studies used a quasi-randomisation method, one using sequential alternation (Rousseau 1985) and the other alternating blocks of ve (Puskarich 1992). In the third study (Dahl 1987) the randomization method was not mentioned in the published report. There was considerable variation in the seizure frequencies at baseline among the randomized groups in the three studies, which may be a reection of the small sample size and the randomization method.
In the study by Helgeson 1990 100 individuals with epilepsy were randomized, 50 to each group, but the method of randomization was not described. However, only 23 in the treatment group and 20 in the control group actually participated and 20 in the treatment group and 18 in the control group completed the study. In the study by Lewis 1990 randomization was by random number assignment. The randomization in the third study (Olley 2001) was by alternate allocation matched according to seizure type and frequency of seizures. The randomization method was not described in the study by May 2002, which randomized 383 individuals among whom only 250 completed the study and the results of 242 participants were analysed (excluding 8 protocol violators).
Relaxation and behavioural therapy and epilepsy The method of randomization was not described in the published report of the study by Dahl 1985. The randomization was by sequential allocation of participants to the treatment and control groups in a 2:1 ratio in the study by Sultana 1987. Self management programs and epilepsy Pramuka 2007 used a random number table to generate randomization sequence and sealed, consecutively numbered envelopes for concealment of randomization. However, the results of only 38 completers out of 55 randomized participants were analyzed. The method of list generation and concealment of randomization was not clear in the other study (Dilorio 2009). Overall, the majority of the included studies were of poor methodological quality and therefore prone to considerable bias. Further details regarding the individual studies are available in the table Characteristics of included studies.
(2) Relaxation therapy and psychosocial functioning The randomization was by alternate allocation in the study by Snyder 1983 and only 50% of enrolled participants completed the study.
Cognitive behavioural therapy and epilepsy The randomization method was not mentioned in three studies (Davis 1984; Martinovic 2006; Tan 1986). Randomization was done using a computerized randomisation table in the study by Lundgren 2006. In the study by Tan 1986, two individuals in each group had probable to denite co-existing nonepileptic seizures and the mean baseline weekly seizure frequency in the three groups differed considerably. In a study assessing the efcacy of cognitive
Effects of interventions
6
(1) Relaxation therapy
(b) CBT and psychosocial outcomes Tan 1986 found no signicant differences between the treatment group and the two control groups for the WPSI, MMPI and BDI. Only the therapists global ratings of psychological adjustment improved for both the CBT and control groups. In a study of the use of CBT in people with epilepsy and a depressed affect, Davis 1984 found signicantly greater reduction in dysphoria and depression in the treatment group compared to controls. There was also a signicant decrease in self reported anxiety, stress and anger. There was increased involvement in social activities for individuals in the treatment group as measured by the Community Adjustment Questionnaire. The study did not report the effect of treatment on seizures. Lundgren 2006 found signicant improvements in the satisfaction with life scale (SWLS) at six months (P < 0.05) and 12 months (P < 0.001) after intervention and the WHO QOLBREF at one year follow up in the intervention group (P < 0.05). Martinovic 2006 found that subthreshold depressive disorder signicantly improved at follow up (at six and nine months) in the CBT group compared to the control group (P < 0.05). The QOL scores were also better in the treatment group compared to controls at six and nine months follow up (P < 0.05). Participants who were seizure free had better QOL compared to those with uncontrolled seizures (P < 0.05). (2B) Cognitive behaviour therapy (CBT) compared to yoga
(a) Relaxation therapy and seizures The publications by Dahl 1987, Rousseau 1985 and Puskarich 1992 provided sufcient seizure frequency data to calculate 50% responder rates. Results for two studies suggested a nonsignicant advantage for relaxation therapy, with Peto odds ratios (OR) with 95% condence intervals (CIs) of 2.56 (95% CI 0.45 to 14.44) (Puskarich 1992) and 2.54 (95% CI 0.17 to 37.01) (Rousseau 1985). One study (Dahl 1987) found a signicant advantage for relaxation therapy with a Peto OR of 15.64 (95% CI 1.57 to 155.75). However, the condence intervals were wide and given that this study was of poor methodological quality this result is unreliable. In view of the methodological deciencies already discussed, we decided against a formal meta-analysis. In total, only one participant in the relaxation therapy group (in the study by Rousseau 1985) and none among the controls were seizure free. For the percentage reduction in seizure frequency outcome, compared to the attention control group, there was a nonsignicant advantage for treatment in all three studies, with weighted mean differences (WMDs) of 358.96 (95% CI -49.33 to 767.25), 27.08 (95% CI -10.49 to 64.65) and 32.68 (95% CI -5.56 to 70.92) respectively. We did not undertake a formal meta-analysis. (b) Relaxation therapy and psychosocial functioning Snyder 1983 found no signicant difference in WPSI scores between the treatment and control groups after relaxation therapy; no data regarding the seizure frequency were available.
Effect on seizures Lundgren 2008 reported that, at the end of one year, ve of the 10 participants in the acceptance concomitant therapy (ACT) group and four of the eight in the yoga group were seizure free at the end of trial (OR 1.0; 95% CI 0.16 to 6.42). A 50% or greater reduction in seizure frequency was found in nine of 10 participants in the ACT group and seven of eight participants in the yoga group (OR 1.29; 95% CI 0.07 to 24.38). Six of 10 participants in the ACT group and four of eight in the yoga group had a 50% or greater reduction in seizure duration (OR 1.5; 95% CI 0.23 to 9.8). The authors reported that ACT reduced seizure index (product of seizure frequency and duration) signicantly more than with yoga. There were no statistically signicant differences for the rates of seizure freedom, 50% or greater reduction in seizure frequency, or seizure duration between the two groups at the end of one year.
(2A) Cognitive behavioural therapy (CBT) compared to no active treatment
(a) CBT and seizures Tan 1986 found a greater than 67% reduction in seizure frequency in one of the 10 participants in the treatment group, two of the 10 in the attention control group, and one of the 10 controls. The Peto OR for CBT versus attention control was 0.47 (95% CI 0.04 to 5.19) and for CBT versus control the Peto OR was 1.0 (95% CI 0.06 to 17.25). One of the 10 participants in the treatment group and one in each of the two control groups were seizure free, giving a Peto OR of 1.0 (95% CI 0.06 to 17.25). Lundgren 2006 found a statistically signicant reduction in seizure frequency as well as in seizure index (product of seizure frequency and seizure duration in seconds) among participants treated with CBT compared to controls (WMD at 1 year -5.16; 95% CI -7.18 to -3.14). However, the results are difcult to interpret since the data were skewed. The other two studies (Davis 1984; Martinovic 2006) did not investigate the effect of CBT on seizures.
Quality of life (QOL) Pretest to post-test, the WHO QOL-BREF revealed signicant improvement among the participants in the ACT group (P 0.01) but the changes on SWLS over time were not signicant. The participants in the yoga group had signicant improvement in their quality of life according to the SWLS (P 0.05), but
7
did not show any signicant changes in the WHO QOL-BREF (Lundgren 2008).
(3) Bio-feedback
(a) EEG bio-feedback and seizures Lantz 1988 reported a statistically signicant reduction in seizure frequency following contingent training (P < 0.005). The median seizure reduction in the treatment group was 61%. However the data for the control group were not available and hence we are unable make a comparison between treatment groups.
2001 found signicant improvement in depression, neurotic disorders and knowledge of epilepsy in the treatment group compared to the control group after a two day psycho-educational program. May 2002 found that the MOSES educational program resulted in a signicant improvement in knowledge and coping with epilepsy, with the treatment group feeling more satised with therapy, and better tolerability and fewer side effects of AEDs. However there was no signicant improvement in the SF-36 questionnaire, depression, or other epilepsy specic measures such as self esteem, fear, stigma, mobility and leisure.
(5) Combined use of relaxation and behaviour therapy
(b) EEG bio-feedback and neuropsychological and quality of life measures These outcomes were reported by Lantz 1988, who found no signicant difference between treatment and control groups for the MMPI and the WPSI.
(a) Relaxation and behaviour therapy and seizures Dahl 1985 reported median seizure frequencies with ranges and a seizure index calculated by multiplying seizure frequency by the seizure duration (seconds). The authors found a sustained reduction in the median seizure index for the behaviour modication group at the end of one year and at eight years follow up, while in the two control groups there was an increase in seizure index at the end of one year. The published report did not provide the actual values for these observations. The study by Sultana 1987 reported mean seizure frequencies for each seizure type (number in each category not stated) and reported separately the mean seizure frequencies for those individuals who had less than 20 seizures per month at baseline and those who had greater than 20 per month. These data were however skewed and we were therefore unable to undertake analyses of effect size using the reported data.
(c) Galvanic skin response (GSR) bio-feedback and seizures Nagai 2004 found that six of the 10 patients in the treatment group had 50% or greater reduction in seizure frequency compared to none among the eight in the sham feedback group (Petos OR 12.81; 95% CI 1.88 to 87.06; P = 0.009).
(4) Educational interventions Helgeson 1990 found signicant differences between the treatment and control groups for the three major subscales of the 50item Sepulveda Epilepsy Education (SEE) true-false list. The treatment group demonstrated a signicant increase in overall understanding of epilepsy, a signicant decrease in fear of seizures, and signicant decreases in hazardous medical self management practices. Serum antiepileptic drug levels indicated better drug compliance in the treatment group. There was no statistically signicant difference between treatment and control groups for the WPSI, except that the control groups level of adjustment appeared to decrease over the course of the trial. Lewis 1990 found statistically signicant improvements in the treatment group for participants knowledge about epilepsy, specically regarding what not to do during a seizure (no objects in the mouth, do not restrain, emergency room visit not required); the purpose of the EEG; and the minimal restriction of activities required. Children in the treatment group were more likely to participate in normal activities compared to controls. Childrens perceived competencies in scholastic achievement, social skills and behaviour also improved. The anxiety of these childrens parents decreased. Olley
(b) Relaxation and behaviour therapy and psychological outcome Sultana 1987 found signicant improvements in state anxiety, trait anxiety and adjustment (home, health, social and emotional). There was also subjective improvement in inferiority feelings and fear of attack.
(6) Self management Pramuka 2007 found no signicant differences in overall quality of life except for an improvement in the QOLIE-89 role limitations - emotional score in the treatment group at follow up. Dilorio 2009 attempted to test the feasibility of a telephone-based self management program for adults with epilepsy. The participants in the treatment group reported signicantly higher levels of outcome expectancies for medications (those in the treatment group expressed more positive attitudes about the outcomes associated with taking medications). The treatment group also had
8
nonsignicantly higher levels of self efcacy, seizure management, and social aspects of epilepsy knowledge.
(7) Effects of psychological interventions on specic psychological measures
therapy. Helgeson 1990 found no signicant difference between the treatment and control groups for adjustment to seizures (WPSI) or for emotional, vocational and interpersonal adjustment. Sultana 1987 observed signicant improvements for the treatment group in home, health, social and emotional adjustment, whilst controls worsened in home adjustment.
(a) Effect of anxiety Sultana 1987 found a signicant reduction in state and trait anxiety scores in the treatment group with combined relaxation and behaviour therapy, while Helgeson 1990 who employed a psychoeducational approach found no signicant change in state-trait anxiety. Olley 2001 found that a two day educational program showed a signicant decrease in overall neurotic disorders (assessed by the Crown-Crisp experiential index - formerly known as the Middlesex Hospital Questionnaire) in the treated group compared to the control group.
(d) Effect on the Washington Psychosocial inventory (WPSI) Helgeson 1990 used a psycho-educational program and found no signicant change in WPSI scores. Tan 1986 did not nd any change in WPSI after CBT. Lantz 1988 found no change in WPSI following either the control period or contingent EEG biofeedback training. Snyder 1983 found no signicant changes in the overall mean WPSI scores after relaxation therapy.
(e) Effect on the Minnesota Multiphasic Personality Inventory (MMPI) (b) Effect on depression Davis 1984 found a signicant reduction in depression among people with epilepsy and depressed affect who were treated with CBT, as assessed by the BDI, Generalized Contentment Scale (GCS) and Lubins Depression Adjective Check List (DACL). Tan 1986 found no signicant improvement in the BDI or MMPI-D scales for individuals receiving CBT. Martinovic 2006 found that subthreshold depressive disorder signicantly improved at follow up (at six and nine months) in the CBT group compared to the control group (P < 0.05). The QOL scores were also better in the treatment group compared to controls at six and nine months follow up (P < 0.05). Participants who were seizure free had better QOL compared to those with uncontrolled seizures (P < 0.05). Sultana 1987 found a marginal but nonsignicant reduction in depression at one year for a combination of relaxation and behavioural therapy. Helgeson 1990 found no signicant reduction in depression between the group receiving the psycho-educational program and the control group for the BDI or DACL. Olley 2001 found a signicant reduction in depression with a two day psychoeducational program (both for inter-group and within-group pre and post-intervention analysis) as assessed by the BDI. May 2002 found no signicant benet on depression (assessed by a depressive mood scale) with the MOSES educational program. Lantz 1988 found no signicant differences in the MMPI between EEG bio-feedback and control groups. Tan 1986 found no significant change in MMPI after CBT. The psychological outcomes with the different interventions and the effect of the different interventions on psychological parameters are given in Additional tables (Table 1 to Table 7).
DISCUSSION
The methodological deciencies of these studies limit the validity of the reported results. The sample size was small in most studies and the inclusion and exclusion criteria were not uniform. There was a paucity of information regarding the seizure frequency at entry, whether individuals with co-existing nonepileptic seizures had been excluded and whether individuals with learning difculties or behavioural problems were excluded. The seizure types among the enrolled individuals, the aetiology of epilepsy, who determined the seizure type (neurologist, epileptologist or psychiatrist) and what investigations were done prior to enrolment in the study were not available in any report. A detailed description of the concomitant AED usage was not provided in most studies. In some studies (Sultana 1987; Tan 1986) the AEDs were adjusted by the treating neurologist, details of which were not known to the authors of the study. In another study (Puskarich 1992), two participants in the treatment group had a new AED added and one discontinued an AED during the study period; one of the participants who had a new AED added had a greater than 50% reduction in seizure frequency. Serum levels of AEDs were estimated in only two studies (Dahl 1987; Puskarich
9
(c) Adjustment to epilepsy Tan 1986 found that only the therapists global ratings of psychological adjustment improved with CBT and supportive counselling. Adjustment to seizures, emotional adjustment, vocational adjustment, and interpersonal adjustment did not change with
1992). It is therefore possible that seizure frequency might have been inuenced by changes in the medical management rather than by the psychological intervention. In one study (Lewis 1990) the medical records of only 50% to 60% of participants were available to the author of the study. In the absence of medical records, it is difcult to ascertain whether the prognostic variables have been equally distributed among the randomized groups. A study assessing the outcome of behavioural therapy in improving depression (Davis 1984) did not describe the concomitant antidepressant therapy. Seizure outcomes data were available only for a small number of participants. In many studies, there was a signicant drop out of randomized participants and the reported results pertain only to those completing the study, leading to biased results (Helgeson 1990; May 2002; Pramuka 2007). The results of these studies have not been reported in a uniform manner both regarding seizure outcomes and quality of life or psychological outcomes. The methodological deciencies of these studies pertaining to randomization have been discussed under the section on methodological quality of included studies. Most of the trials compared behavioural treatments with attention control or control (wait-list) conditions. Double blinding is impossible for most behavioural interventions, and effective single blinding is also difcult to achieve in most cases. Using no treatment controls (rather than having credible placebo groups) and no blinding makes behavioural trials more prone to bias than the traditionally designed drug trials, and one might expect to nd a spurious statistically signicant result (Type I error). Additionally, studies with small sample sizes (as used in most trials in this area) may lack sufcient statistical power to detect small but clinically important effects (Type II error) and may limit generalizing the ndings. The results of three studies employing relaxation therapy as an intervention revealed a benecial trend. These studies were not combined in a formal meta-analysis to avoid a Type I error. In view of methodological deciencies in these studies and small numbers of individuals studied, it would be prudent to await further studies to assess the efcacy of relaxation therapy for seizure control. The effects of psychological interventions on psychological and quality of life outcomes have not been studied in a uniform manner, making it difcult to combine the results from various studies. Four studies indicate possible benets from educational interventions in improving the understanding of epilepsy, adjustment to epilepsy and compliance to medication. The results of the various interventions on parameters such as anxiety, depression and adjustment to epilepsy appear to be contradictory. It is possible that the results of these psychological interventions may depend on the
baseline psychosocial functioning of the individuals included in the studies, seizure frequency, their education and level of intellectual functioning.
AUTHORS CONCLUSIONS Implications for practice

(1) Psychological treatments and seizure control
(a) Relaxation therapy and seizure control The studies presently available indicate a possible benecial effect on seizure frequency. However, in view of methodological deciencies and the small number of individuals studied, denite conclusions cannot be made and further studies are needed.
(b) Cognitive behaviour therapy, EEG or GSR bio-feedback, educational interventions, combined use of relaxation and behaviour therapy, and self management programs No reliable conclusions can be drawn regarding their efcacy in controlling seizures. Though one study of GSR biofeedback reported benecial effect on seizure frequency, no denite conclusions can be drawn due to the small sample size and wide condence intervals. (2) Psychological treatments and psychological outcomes At present no reliable conclusions can be drawn regarding the effects of psychological interventions upon the quality of life of people with epilepsy. However, it appears that educational interventions may be of value in improving the persons knowledge and understanding of epilepsy, adjustment to seizures and medication compliance. Since the last version of this review, none of the new relevant studies have provided additional information to change the conclusions
Implications for research

Better designed and larger studies are imperative if researchers are to address whether psychological therapies can positively inuence seizure frequency and psychological functioning. Specically, the effect of relaxation therapy on seizures and the effect of cognitive behaviour therapy in improving depression among persons with epilepsy and comorbid depression need to be studied.
10
REFERENCES
References to studies included in this review

Dahl 1985 {published data only} Dahl J, Brorson LO, Melin L. Effects of a broad-spectrum behavioral medicine treatment program on children with refractory epileptic seizures: an 8-year follow-up. Epilepsia 1992 JanFeb;33(1):98102. [: ISSN 00139580] Dahl J, Melin L, Brorson LO, Schollin J. Effects of a broad-spectrum behavior modication treatment program on children with refractory epileptic seizures. Epilepsia 1985 JulAug;26(4):3039. [: ISSN 00139580] Dahl 1987 {published data only} Dahl J, Melin L, Lund L. Effects of a contingent relaxation treatment program on adults with refractory epileptic seizures. Epilepsia 1987 MarApr;28(2):12532. [: ISSN 00139580] Davis 1984 {published data only} Davis GR, Armstrong HE Jr, Donovan DM, Temkin NR. Cognitive-behavioral treatment of depressed affect among epileptics: preliminary ndings. Journal of Clinical Psychology 1984 Jul;40(4):9305. Dilorio 2009 {published data only} DiIorio C, Reisinger EL, Yeager KA, McCarty F. A telephone-based self-management program for people with epilepsy. Epilepsy & Behavior 2009;14(1):2326. Helgeson 1990 {published data only} Helgeson DC, Mittan R, Tan SY, Chayasirisobhon S. Sepulveda Epilepsy Education: the efcacy of a psychoeducational treatment program in treating medical and psychosocial aspects of epilepsy. Epilepsia 1990 JanFeb;31(1):7582. [: ISSN 00139580] Lantz 1988 {published data only} Lantz DL, Sterman MB. Neuropsychological assessment of subjects with uncontrolled epilepsy: effects of EEG feedback training. Epilepsia 1988 MarApr;29(2):16371. [: ISSN 00139580] Lewis 1990 {published data only} Lewis MA, Hatton CL, Salas I, Leake B, Chiofalo N. Impact of the childrens epilepsy program on parents. Epilepsia 1991 MayJun;32(3):36574. [: ISSN: 00139580] Lewis MA, Salas I, De La Sota A, Chiofalo N, Leake B. Randomized trial of a program to enhance the competencies of children with epilepsy. Epilepsia 1990 JanFeb;31(1): 1019. [: ISSN: 00139580] Lundgren 2006 {published data only} Lundgren T, Dahl J, Melin L, Kies B. Evaluation of acceptance and commitment therapy for drug refractory epilepsy: a randomized controlled trial in South Africa--a pilot study. Epilepsia 2006;47:21739. Lundgren 2008 {published data only} Lundgren T, Dahl J, Yardi N, Melin L. Acceptance and Commitment Therapy and yoga for drug-refractory epilepsy: a randomized controlled trial. Epilepsy & Behavior 2008;13(1):1028.
Martinovic 2006 {published data only} Martinovic Z, Simonovi P, Djoki R. Preventing depression in adolescents with epilepsy. Epilepsy & Behavior 2006;9: 61924. May 2002 {published data only} May TW, Pfafin M. The efcacy of an educational treatment program for patients with epilepsy (MOSES): Results of a controlled randomized study. Epilepsia 2002;43 (5):53949. Nagai 2004 {published data only} Nagai Y, Goldstein LH, Fenwick PBC, Trimble MR. Clinical efcacy of galvanic skin response biofeedback training in reducing seizures in adult epilepsy: a preliminary randomized controlled study. Epilepsy & Behavior 2004;3: 21623. Olley 2001 {published data only} Olley BO, Osinowo HO, Brieger WR. Psychoeducational therapy among Nigerian adult patients with epilepsy: a controlled outcome study. Patient Education and Counseling 2001;42:2533. Pramuka 2007 {published data only} Pramuka M, Hendrickson R, Zinski A, Van Cott AC. A psychosocial self-management program for epilepsy: a randomized pilot study in adults. Epilepsy & Behavior 2007; 11(4):53345. Puskarich 1992 {published data only} Puskarich CA, Whitman S, Dell J, Hughes JR, Rosen AJ Hermann BP. Controlled examination of effects of progressive relaxation training on seizure reduction. Epilepsia 1992 JulAug;33(4):67580. [: ISSN 00139580] Rousseau 1985 {published data only} Rousseau A, Hermann B, Whitman S. Effects of progressive relaxation on epilepsy: analysis of a series of cases. Psychological Reports 1985;57(3II):120312. Snyder 1983 {published data only} Snyder M. Effect of relaxation on psychosocial functioning in persons with epilepsy. Journal of Neuroscience Nursing 1983;15(4):2504. Sultana 1987 {unpublished data only} Sultana SM. A study on the psychological factors and the effect of psychological treatment in intractable epilepsy. PhD Thesis, University of Madras, India 1987. Tan 1986 {published data only} Tan SY, Bruni J. Cognitive behavior therapy with adult patients with epilepsy: a controlled outcome study. Epilepsia 1986 MayJun;27(3):22533. [: ISSN 00139580]
References to studies excluded from this review

Berger NM 2001 {published data only} Berger NM. The effect of a stress management training program on epilepsy. Epilepsia 2001;42 Suppl 7:245.
11
Earl 1986 {published data only} Earl WL. Job stability and family counseling. Epilepsia 1986;27(3):2159. Gillham 1990 {published data only} Gillham RA. Refractory epilepsy: an evaluation of psychological methods in outpatient management. Epilepsia 1990 JulAug;31(4):42732. [: ISSN: 00139580] Sterman 1980 {published data only} Sterman MB, Shouse MN. Quantitative analysis of training, sleep EEG and clinical response to EEG operant conditioning in epileptics. Electroencephalography and Clinical Neurophysiology 1980 Sep;49(5-6):55876. [: ISSN: 00134694] Tieffenberg 2000 {published data only} Tieffenberg JA, Wood EI, Alonso A, Tossutti MS, Vicente MF. A randomized eld trial of ACINDES: a child-centered training model for children with chronic illnesses (asthma and epilepsy). Journal of Urban Health: Bulletin of the New York Academy of Medicine 2000;77(2):28097.
Fenwick 1994 Fenwick P. Behavioral treatment of epilepsy generation and inhibition of seizures. Neurologic Clinics 1994 Feb;12(1): 175202. Forster 1968 Forster FM, Hansotia P, Cleeland CS, Ludwig A. A case of voice-induced epilepsy treated by conditioning. Neurology 1968;18(3):290. Forster 1969 Forster FM, Paulsen WA, Baughman FA Jr. Clinical therapeutic conditioning in reading epilepsy. Neurology 1969 Aug;19(8):71723. [: ISSN 00283878] Fried 1990 Fried R, Fox MC, Carlton RM. Effect of diaphragmatic respiration with end-tidal CO2 biofeedback on respiration, EEG, and seizure frequency in idiopathic epilepsy. Annals of New York Academy of Sciences 1990;602:6796. Goldstein 1990 Goldstein LH. Behavioural and cognitive-behavioural treatments for epilepsy: a progress review. British Journal of Clinical Psychology 1990 Sep;29(3):25769. [: ISSN: 01446657] Goldstein 1997 Goldstein LH. Effectiveness of psychological interventions for people with poorly controlled epilepsy. Journal of Neurology, Neurosurgery, and Psychiatry 1997 Aug;63(2): 13742. Higgins 2005 Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions 4.2.4 [updated March 2005]. The Cochrane Collaboration. Krafft 1982 Krafft KM, Poling AD. Behavioral treatments of epilepsy: methodological characteristics and problems of published studies. Applied Research in Mental Retardation 1982;3(2): 15162. [: ISSN: 02703092] Kuhlman 1978 Kuhlman WN. EEG feedback training of epileptic patients: clinical and electroencephalographic analysis. Electroencephalography and Clinical Neurophysiology 1978; 45:699710. Lubar 1981 Lubar JF, Shabsin HS, Natelson SE, Holder GS, Whitsett SF, Pamplin WE, et al.EEG operant conditioning in intractable epileptics. Archives of Neurology 1981 Nov;38 (11):7004. [: ISSN 00039942] Miller 1994 Miller L. Psychotherapy of epilepsy: seizure control and psychosocial adjustment. Journal of Cognitive Rehabilitation 1994;12:1430. Mostofsky 1993 Mostofsky DI. Behaviour modication and therapy in the management of epileptic disorders. In: Mostofsky DI, Loyning Y editor(s). The neurobehavioral treatment of epilepsy. New Jersey: Lawrence Earlbaum, 1993.
12
References to studies awaiting assessment

Rau 2006 {published data only} Rau J, May TW, Pffin M, Heubrock D, Petermann F. Education of children with epilepsy and their parents by the modular education program epilepsy for families (FAMOSES)--results of an evaluation study [Schulung von Kindern mit Epilepsie und deren Eltern mit dem Modularen Schulungsprogramm Epilepsie fr Familien (FAMOSES) Ergebnisse einer Evaluationsstudie]. Die Rehabilitation 2006;45(1):2739.
Additional references
Andrews 1992 Andrews DJ, Schonfeld WH. Predictive factors for controlling seizures using a behavioural approach. Seizure 1992;1(2):1116. Betts 1992 Betts T. Epilepsy and stress- Editorial. BMJ 1992;305 (6850):3789. Cabral 1976 Cabral RJ, Scott DF. Effects of two desensitization techniques, biofeedback and relaxation, on intractable epilepsy: follow-up study. Journal of Neurology, Neurosurgery, and Psychiatry 1976;39:5047. Dahl 1988 Dahl J, Melin L, Leissner P. Effects of a behavioural intervention on epileptic seizure behaviour and paroxysmal activity: a systematic replication of three cases of children with intractable epilepsy. Epilepsia 1988;29(2):17283. Fenwick 1992 Fenwick PBC. The relationship between mind, brain and seizures. Epilepsia 1992;33 Suppl 6:16.
Nakken 1990 Nakken KO, Bjorholt PG, Johannessen SI, Loyning T, Lind E. Effect of physical training on aerobic capacity, seizure occurrence, and serum level of antiepileptic drugs in adults with epilepsy. Epilepsia 1990 JanFeb;31(1):8894. [: ISSN: 00139580] Pritchard 1985 Pritchard PB III, Holstrom VL, Giacinto J. Self-abatement of complex partial seizures. Annals of Neurology 1985;18: 2657. Quy 1979 Quy RJ, Hutt SJ, Forrest S. Sensorimotor rhythm feedback training and epilepsy: some methodological and conceptual issues. Biological Psychology 1979;9(2):12949. Reiter 1987 Reiter J, Andrews DJ, Janis C. Taking control of your epilepsy.
Santa Rosa, California: The Basics Publishing Company, 1987. Sterman 1980a Sterman MB, Shouse MN. Quantitative analysis of training, sleep EEG and clinical response to EEG operant conditioning in epileptics. Electroencephalography and Clinical Neurophysiology 1980 Sep;49(5-6):55876. [: ISSN: 00134694] Temkin 1984 Temkin NR, Davis GR. Stress as a risk factor for seizures among adults with epilepsy. Epilepsia 1984;25(4):4506. Wolf 1997 Wolf P. Behavioral therapy. In: Engel J Jr, Pedley TA editor(s). Epilepsy: a comprehensive textbook. Philadelphia: Lippincott Raven, 1997:135964. Indicates the major publication for the study
13
CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]

Dahl 1985 Methods Unblinded, controlled clinical trial. Three treatment arms: I - behaviour modication; II - attention control; III - control. The randomization procedure is not clear. Pre-randomization baseline period was 10 weeks, treatment phase 6 weeks, follow up after 10 weeks, 1 year, 8 years 18 children with more than 1 seizure per month were enrolled. 6 participants were allocated to each of the 3 groups. The age ranged between 7 to 17 years. There were 13 boys and 5 girls. The mean age SD in the 3 groups were 14.22.14; 14.84.1 and 11.32.6 years respectively. The duration of epilepsy in the 3 groups were 7.83.5; 10.53.4 and 7.23.4 years respectively. The median seizure frequency at baseline in the 3 groups were 24, 4 and 6.5 respectively. Individuals with severe mental retardation and progressive neurological disease were excluded. All except 1 child in the behaviour modication group were receiving AEDs, which were continued The treatment group received broad-spectrum behaviour modication therapy, attention control group received equal amount of psychologists attention without any specic program or advice. The control group received no additional intervention. The intervention consisted of 1 hour sessions every week for 6 weeks Seizure frequency Seizure duration Seizure index (seizure frequency x seizure duration in seconds) No participant excluded from analysis. There were no withdrawals from the study. The baseline median seizure frequencies were more in the intervention group compared to the other two groups
Participants
Interventions
Outcomes
Notes
Risk of bias Item Allocation concealment? Dahl 1987 Methods Unblinded controlled clinical trial. Three treatment arms: I - relaxation training; II - attention control; III - control. The randomization procedure is not clear. Pre-randomization baseline period was 10 weeks, treatment phase 6 weeks, follow up after 10 weeks Authors judgement Unclear Description B - Unclear
14
Dahl 1987
(Continued)
Participants
18 adults having at least 1 seizure per month in spite of AED treatment. 6 participants were allocated to each of the 3 groups. There were 11 women and 7 men, aged 19 to 58 years with symptomatic epilepsy. The baseline mean SD seizure frequency in the 3 groups were 9.049.95; 1.680.95; 2.792.39 respectively. The duration of epilepsy ranged from 2 to 44 years. Individuals with progressive neurological disease and people unable to understand and follow instructions were excluded Group I received contingent relaxation, group II received supportive therapy and served as an attentioncontrol group, group III received no treatment. AEDs were continued in all groups. The contingent relaxation training was given for 1 hour every week for 6 weeks Seizure frequency No participant excluded from analysis. There was 1 drop-out each, in the groups I and II The baseline weekly seizure frequency in the intervention group is much higher than the other two control groups
Interventions
Outcomes Notes
Risk of bias Item Allocation concealment? Davis 1984 Methods Unblinded controlled clinical study. 2 treatment arms: 1 cognitive behavioural therapy, the other control. The randomization procedure is not stated. 15 adults with epilepsy and depression were randomized 8 to treatment group and 7 to control group. The mean age was 33.110.8 years. The mean duration of epilepsy was 13.6911.1 years. Subjects with IQ less than 70 and behaviour problems were excluded. 10 women and 3 men completed the study Treatment group received 6 weekly 2 hour classes designed to increase the participants skills in employing specic cognitive behavioural techniques for the management of depression Depression Adjective Checklist, Form-E (DACL) Hudson Generalized Contentment scale (GCS) BDI Community Adjustment Questionnaire (CAQ) - subjective scale The testings were done before randomization, post-treatment and at 6 weeks after treatment 2 participants in the control group withdrew from the study. Authors judgement Unclear Description B - Unclear
Participants
Interventions
Outcomes
Notes Risk of bias Item
Authors judgement
Description
15
Davis 1984
(Continued)
Allocation concealment? Dilorio 2009 Methods Participants
Unclear
B - Unclear
Randomized controlled trial, single centre study from USA. Twenty-two adults (15 men and 7 women) with epilepsy recruited from hospital-based epilepsy clinics. The mean age was 43 years. Inclusion criteria: English speaking adults diagnosed to have epilepsy for at least 1 year and receiving standard treatment for epilepsy Those in the intervention group received a ve-session intervention with a nurse trained in Motivational Interviewing (MI) counselling. Following an in-person introductory session, the remaining four sessions were conducted by phone. 95% of the 55 planned MI sessions and the 44 planned courtesy calls for those in the control group were completed Participants completed the baseline and a 3 month follow-up assessment using a computer administered self-interview with audio (ACASI) Self-reported medication adherence (measured at baseline and follow up) using the Antiretroviral General Adherence Scale (AGAS), revised for use with antiepileptic drugs Adherence using an electronic monitoring device MEMSTM (Medication Event Monitoring System) cap for the 3 month study period Epilepsy self management (the Epilepsy Self-Management Scale in the follow-up assessment only) comprising medication, information, safety, seizure and lifestyle management Self efcacy measured with the Epilepsy Self-Efcacy Scale Outcome expectancy - judgment of the likely consequences of practicing self-management strategies measured by three scales assessing outcomes related to epilepsy treatment, having a seizure, and general seizure management Knowledge about epilepsy - medical and social aspects - assessed with the Epilepsy Knowledge Questionnaire (EKQ) Seizure outcomes were not assessed. One from each group withdrew from the study: one withdrew for family reasons, and the other was lost to follow up
Interventions
Outcomes
Notes
Risk of bias Item Blinding? All outcomes Authors judgement No Description
16
Helgeson 1990 Methods Unblinded controlled clinical trial. 2 arms: 1 treatment with a psycho-educational program and the other control. Randomization method not specied. Intervention period was 2 days; follow up after 4 months 100 individuals with epilepsy were randomized 50 to each group. Only 23 in the treatment group and 20 in the control group actually participated. The mean age in the control group was 38.5610.67 years and in the treatment group 36.1512.81 years. The age at onset of epilepsy was 23.3916.14 for controls and 18.813.22 for treatment group. The seizure frequency in controls was 2.054.73 per month for controls; 2.473.98 per month for treatment group. The duration of epilepsy in the control group was 15.4411.14 years and in the treatment group 17.410.78. 28 women and 10 men completed the study 2-day psycho-educational program to improve the understanding of epilepsy versus none WPSI BDI Lubins depression adjective checklist 50-item true-false list to evaluate Sepulveda Epilepsy Education (SEE) program State-trait anxiety inventory Acceptance of disability scale Sherers self-efcacy scale administered before intervention and at four months follow up Only 20 participants in the treatment group and 18 in control group, completed the study
Participants
Interventions Outcomes
Notes Risk of bias Item Allocation concealment? Lantz 1988 Methods
Authors judgement Unclear
Description B - Unclear
Unblinded controlled study 3 arms: I - contingent EEG biofeedback; II - noncontingent feedback and III - no intervention control. Pre-randomization baseline = 6 weeks, treatment = 4 weeks, withdrawal = 4 weeks, follow up = 6 weeks. Randomization method not stated. 24 adults with uncontrolled epilepsy, 8 participants in each group. 15 men and 9 women aged between 15 to 53 years with mean of 28.37. The age at onset ranged from 1/2 to 50 years (mean 12.57). The duration of epilepsy ranged from three to 38 years (mean 15.1) and the seizure frequency from 3 to 70 (mean 21.2) per month. Individuals with progressive neurological disease and signicant psychopathology were excluded Contingent EEG bio-feedback training, versus noncontingent training (attention control) versus no intervention control. The intervention was given for 30 minutes, 3 times a week for 6 weeks. The 2 control groups were subsequently given contingent training, and all 3 groups were followed up Psychological assessment was done at the end of the 6 week baseline period, after the end of 6 week treatment period and for the control groups again after they had completed the contingent training.
17
Participants
Interventions
Outcomes
Lantz 1988
(Continued)
The psychological assessment consisted of 4 scales of WPSI, 71 item Mini-Mult form of the MMPI, word list recall using the common nouns of the Buschke Memory test, wonderlic personnel test, Wechsler Memory scale, Logical memory and visual reproduction, seashore tonal memory test, nger tip number writing perception, name writing, and the rst half of the Stroop colour word test Notes None withdrew from the study. The authors have also tried to discriminate individuals who are more likely to respond to the EEG bio-feedback treatment, based on neuropsychological assessment
Risk of bias Item Allocation concealment? Lewis 1990 Methods Randomized controlled study. Randomized to a child centred, family focussed, educational group or a control group by random number assignment. Unblinded study. Treatment phase lasted 4 weeks; follow up after 5 months This study included 252 children with epilepsy aged between 7 and 14 years, (126 in each group). Medical records were available to the authors only for 65% of participants in the experimental group and 47% of controls. The mean age of control group was 9.92.1 years and for treatment group 10.12.2 years. 123 males and 113 females completed the study The treatment group was given child centred, family focussed educational program while the control group received the same information related to epilepsy through traditional lecture followed by question and answer sessions. The intervention was given for 1.5 hours every week for 4 weeks Knowledge about seizures. Harters self competency scale (1982), perceived changes in the childrens knowledge and behaviour and in their parents behaviour. The tests were administered before and 5 months after the intervention 10% of children were re-interviewed to check reliability of data which was 96%. There were three withdrawals in the experimental group and 13 in the control group Authors judgement Unclear Description B - Unclear
Participants
Interventions
Outcomes
Notes
Risk of bias Item Allocation concealment? Authors judgement Yes Description A - Adequate
18
Lundgren 2006 Methods Randomized controlled single centre study from South Africa. There were two arms: acceptance concomitant therapy (ACT) arm and supportive therapy (ST) arm. Randomization by computerized randomization table. The intervention was carried out over an initial 5 week period. Follow up lasted 12 months 27 adults aged between 21 and 55 years, institutionalized or day care workers in a centre for epilepsy. Inclusion criteria: diagnosed epilepsy with minimum seizure frequency of 4 seizures in past 3 months. Subjects with concomittant progressive illnesses were excluded ACT or ST were administered in 4 sessions in the following sequence: one individual session, two group sessions followed by another individual session. The individual sessions lasted 1.5 hours each and group sessions 3 hours each. All participants were subsequently given individual boosters and followed up for an additional session at 6 and 12 months. The total therapy time over 12 months was 11 hours. The ACT sessions included general principles of values, self as context, defusion, acceptance, contact with present moment, committed action and empowerment. Seizure control techniques in the form of countermeasures were instructed. Subjects given supportive therapy received an equal amount of professional attention; subjects were made to reect on their lives and their problems in a nonjudgemental empathic and accepting environment without any active advice from the therapists Seizure frequency, seizure index (product of seizure frequency and seizure duration), WHO Quality of Life (WHO QOL-BREF) and satisfaction with life scale (SWLS) One subject in control arm excluded in analysis of seizure frequency
Participants
Interventions
Outcomes
Notes Risk of bias Item Adequate sequence generation? Allocation concealment? Blinding? All outcomes Free of selective reporting?
Authors judgement Yes Yes No
Description
A - Adequate
Yes
Lundgren 2008 Methods Single centre, randomized controlled trial with repeated measures. Randomization using computerized randomization table, not blinded. All participants had drug refractory seizures veried by EEG. Baseline period = 3 months; Study duration = 5 weeks; Follow up = 12 months. Two arms Arm A= ACT; Arm B=yoga Eighteen adult outpatients (12 men and 6 women) from India, aged between 18 and 55 years and having a minimum of three seizures during the past 3 months and absence of an ongoing progressive illness
19
Participants
Lundgren 2008
(Continued)
Interventions
One group (N=10) received Acceptance Concomitant Therapy (ACT). The other group received yoga (N=8). The treatments were administered over a 5 week period Yoga included Pranayama (controlled deep breathing), Asanas (physical postures), and Dhyana (meditation) Yama (harmony with others) and Niyama (harmony with yourself ). The yoga teacher integrated the teaching of Pranayama, Asanas, Dhyana and the teaching of Yama and Niyama into a four-session protocol. A yoga teacher from the outpatient clinic was responsible for the yoga treatment. ACT was given by two clinical psychologists trained in ACT and behavioral treatment of epilepsy. ACT was given in four sessions (one individual session, two group sessions, and one individual session). Booster sessions occurred at 6 and 12 months.The individual sessions and booster sessions were each 1.5 hours long and the group sessions were 3 hours, for a total therapy time of 12 hours, for each participant. The ACT protocol was adjusted to the Indian context after discussions with the participating staff to avoid unnecessary misunderstandings caused by language problems and cultural differences Dosages of antiepileptic drugs remained constant throughout the study Seizure index (seizure frequency X seizure duration) was assessed at a 3 month baseline and during the 12 months of follow up using a seizure diary. Quality of life was assessed (Satisfaction with Life Scale (SWLS), and the World Health Organization Quality of Life instrument, short version (WHOQOL-BREF)) prior to the initiation of treatment, after treatment, and at the 6 and 12 month follow ups No withdrawals from study; Seizure frequency was more in the ACT group compared to yoga group
Outcomes
Notes Risk of bias Item Adequate sequence generation? Allocation concealment? Blinding? All outcomes Martinovic 2006 Methods
Authors judgement Yes Yes No
Description
Computerized randomization table
Randomized single centre study from Belgrade, Yugoslavia. Adolescents (age 13-19 years) referred from general practices in Belgrade with subthreshold depression. Subjects with epilepsy due to progressive cerebral lesion, co-morbid depression, psychosis, mental retardation were excluded. Treatment phase Follow-up phase There were 32 subjects (18 women and 12 men + 2 subjects who dropped out from the study). The mean age was 17.41.6y
Participants
20
Martinovic 2006
(Continued)
Interventions
One group received cognitive behavioural intervention (CBI). The other group treated with counselling as usual (TAU). AEDs were continued in all. Beck Depression Inventory (BDI), Center for Epidemiological Study on Depression (CES-D), Hamilton depression scale (HAMD) and Quality of Life in Epilepsy Inventory (QOLIE-31) The scales were administered at baseline and at 6 and 9 month follow up 1 subject in the treatment and control groups withdrew from the study
Outcomes
Notes Risk of bias Item Allocation concealment? May 2002 Methods
Authors judgement Unclear
Description B - Unclear
Multicentre randomized controlled study involving 22 epilepsy centers in Germany, Austria and Switzerland. Randomization method not known. Treatment group completed a questionnaire immediately before and 6 months after the completion of the course. Controls completed the questionnaire 6 months before their course and immediately before their course. Individuals who had mental retardation, acute psychiatric illness, nonepileptic seizures and people less than 16 years of age were excluded 113 individuals were in control group, 129 in treatment group. Their age ranged between 16 to 80 years The mean age of the control group was 38.413.5 years and that of the treatment group was 37.513.7 years. Male:female ratio in the control group was 43.4:56.6; while in the treatment group was 42.5:57.5. The median age at onset of epilepsy in the control group was 15 years and in the treatment group was 17 years. The median duration of epilepsy among controls was 18.2 years and in the treatment group was 13.5 years. 62.8% in each group had partial epilepsy. 15.5% of controls and 20.4% in treatment group had generalized epilepsy. 41.2% of controls and 43.3% of treatment group had >1 seizure/month 2 day patient education program, developed for German speaking countries - MOSES (Modular service package epilepsy). Has 9 units, covering living with epilepsy, epidemiology, basic knowledge, diagnostics, therapy, self control, prognosis, psychosocial aspects and network - 14 lessons 60 minutes each. The control group received no intervention but were given the course 6 months later, after the completion of the study 1. HRQOL by SF-36 questionnaire 2. Rosenberg self esteem scale 3. Depression scale D-S by von Zerssen 4. Epilepsy specic instruments to assess: restrictions in daily life; epilepsy related fear; stigma, mobility and leisure; epilepsy knowledge; coping with epilepsy and adaptation to epilepsy Seizure frequency graded scale of 6 categories 0 = no seizures, 5 = 1or more per day
21
Participants
Interventions
Outcomes
May 2002
(Continued)
Notes
383 adults were randomized; only 342 returned the rst questionnaire, 250 returned the second questionnaire as well but 8 violated inclusion criteria and hence only 242 were analysed
Risk of bias Item Allocation concealment? Nagai 2004 Methods Single centre, single blind, randomized controlled parallel group trial. Adult patients with epilepsy of more than 2 years and average seizure frequency of 2-3/month for at least 3 months and stable medication for more than 1 month prior to trial and who were able to keep seizure diary were included. Patients with psychiatric illness, severe head injury, drug abuse, non compliant patients and those with learning disabiliies were excluded. 2 arms viz: active (bio-feedback) and control (sham biofeedback). Patients were unaware of the allocated group. Randomization using random number tables determined at trial onset. The study comprised of 3 phases - baseline period of 3 months, treatment period of 1 month and 3 month follow-up period 18 adult patients were randomized: 10 (5 men and 5 women) to bio-feedback group and 8 (4 men and 4 women) to sham biofeedback. All had drug refractory epilepsy. In the bio-feedback group 7 had complex partial seizures, 2 had generalized tonic clonic seizures and one had generalized absence seizures while in the control group ve had complex partial seizures, 2 had generalized tonic clonic seizures and 1 had myoclonic seizures. The mean age was 42.56.8 years in the treatment group and 41.3813.28 in control group Active group received bio-feedback based on the patients skin resistance -galvanic skin response (GSR) . In the sham feedback, patients watched a video of the same computer generated animation but the animation changes were unrelated to GSR changes. There were 12 sessions - 3/week each lasting 40 minutes. Participants were asked to practise the skills at home without the bio-feedback machine. 2 patients were not receiving AEDs by choice, while others were on polytherapy Primary: seizure frequency Secondary: GSR change and correlation with seizure frequency 3 patients withdrew from the control group due to lack of improvement Authors judgement Unclear Description B - Unclear
Participants
Interventions
Outcomes
Notes Risk of bias Item Allocation concealment?
Authors judgement Yes
Description A - Adequate
22
Olley 2001 Methods Randomized controlled study. Randomization was by alternate allocation. Two treatment arms: one group received a 2 day psycho-educational program while the control group received no intervention. Pre-assessment testing of knowledge of epilepsy and psychopathology were done and follow-up assessment was done after 2 months. Inclusion criteria were: (a) diagnosis (clinico-EEG) of epilepsy by neurologist/neuropsychiatrist; (b) adults 21-65 yrs; (c) no mental subnormality or major neurological disorder; (d) on follow up for at least 3 months. 300 consecutive individuals with epilepsy on routine follow up, at Neuro-Psychiatry Hospital, Aro, Nigeria were assessed for interictal psychopathology. Among them 53 individuals living in the study area were considered for treatment intervention - of whom 30 participated. 15 were allocated to treatment group and 15 to waiting list control group 2 day modular, didactic, psycho-educational program: Day 1: preliminary assessment and establishment of rapport; Day 2: didactic modular educational session ( covering background information, diagnosis and management of epilepsy, day-to-day adjustments, seizure control and stigma) with time for group discussions and clarications. The control group received no intervention and were told that their session would be conducted at a later date (1) Knowledge about epilepsy (2) BDI (3) Crown Crisp Experiential index (Middlesex Hospital Questionnaire) to assess psychoneurotic traits
Participants
Interventions
Outcomes
Notes Risk of bias Item Allocation concealment? Authors judgement No Description C - Inadequate
Pramuka 2007 Methods A randomized, controlled trial. A random number table was used by the study coordinator to generate a randomization sequence, with letters describing the randomization arm sealed into consecutively numbered envelopes 61 adults with epilepsy recruited from two centres in USA; 55 adults were randomized The psycho-educational program was developed to be delivered in a group format for approximately 6-12 individuals, meeting for a 2 hour block of time once per week for 6 consecutive weeks. The curriculum included a mixture of psycho-education, medical information, and advocacy topics framed in the self management activities of self evaluation, self monitoring, stimulus control, and self rewards. Additional topics were oriented toward group process and peer
23
Participants Interventions
Pramuka 2007
(Continued)
support. Controls received treatment as usual (TAU) They completed the self-report psychosocial measures at entry into the study (baseline), 1 month later, and 6 months after baseline, but received no additional services Outcomes Quality of Life in Epilepsy-89 Inventory (QOLIE-89), the Washington Psychosocial Seizure Inventory (WPSI), a locus of control scale (LOC), and the Epilepsy Self-Efcacy Scale-2000 (ESES) , Millon Clinical Multiaxial Inventory-III (MCMI-III) Seizure outcomes were not assessed. 9 subjects randomized to intervention group did not participate; 7 subjects dropped out during the trial (two from intervention and 5 from TAU group). One subject in intervention group was excluded from analysis for data integrity issues. Only data of 38 subjects were included in nal analysis
Notes
Risk of bias Item Adequate sequence generation? Allocation concealment? Authors judgement Yes Yes A random number table used to generate randomization sequence. Sealed consecutively numbered envelopes were used for concealment Description
Blinding? All outcomes Puskarich 1992 Methods
No
Unblinded controlled trial. Two arms: I - progressive relaxation training and II - quietly sitting. The randomization was in alternating blocks of 5. Unblinded study. Pre-randomization baseline = 8 weeks, treatment phase = 8 weeks, follow up = 8 weeks 24 adults with uncontrolled epilepsy, having at least 6 seizures during a baseline period of 8 weeks. There were 13 individuals in the treatment group and 11 controls. The mean age in the treatment group (n= 13) was 39.1 years and 39.7 years in the control group (n=11). There were 8 males and 16 females. The mean age at onset in the treatment group was 17.7 years and in the control group 16.2 years. The mean duration of epilepsy in the treatment and control groups were 20.8 and 23.5 years respectively. Individuals with mental subnormality and psychiatric disorders were excluded Progressive relaxation training, compared to quietly sitting. 6 treatment sessions were given over 4 to 8 weeks Seizure frequency Participant expectations in both groups before the study did not differ. 2 participants in the treatment group had a new AED added and 1 discontinued an AED. 1 of the individuals who had a new AED added
24
Participants
Interventions
Outcomes Notes
Puskarich 1992
(Continued)
had greater than 50% reduction in seizure frequency. The baseline seizure frequency in the treatment group was 1712.22 while in the control group 10.333.38. Overall 53 individuals met the inclusion criteria for the study. 29 did not enter the study or could not complete the study Risk of bias Item Allocation concealment? Authors judgement No Description C - Inadequate
Rousseau 1985 Methods Unblinded controlled trial. Two arms: I - progressive relaxation therapy; II - sham treatment. Randomization was on a sequential alternating basis. Pre-randomization baseline phase 3 weeks; treatment phase 3 weeks 3 men and 5 women with uncontrolled epilepsy (6 or more seizures during 3 week baseline) and normal intelligence. Mean age was 23 years for the treatment group and 20.7 years for controls. Mean duration of epilepsy in the treatment group was 17.2 years and in controls 10 years. Mean baseline seizure frequency was 10.5 in the treatment group and 56 in control group. Mentally handicapped and persons who did not maintain reliable records were excluded Progressive relaxation training, compared to quiet sitting. Seizure frequency. None excluded from the study after randomization.
Participants
Interventions Outcomes Notes Risk of bias Item Allocation concealment?
Authors judgement No
Description C - Inadequate
Snyder 1983 Methods Unblinded, controlled study. Two arms: one treatment, other control. They were alternately allocated to a treatment and control group 32 people with epilepsy were included in the study among whom only 16 participants (11 in the experimental group and 5 in the control group) completed the 6 month study. Age and sex composition not mentioned. Only 3/16 had seizures for less than 5 years. AED therapy not mentioned Progressive relaxation training versus none. Progressive relaxation was taught in 6 sessions. The timing between sessions varied from once every 2 to 3 days for hospitalized individuals to weekly sessions for community participants
25
Participants
Interventions
Snyder 1983
(Continued)
Outcomes
The assessment was by mean scores on the 57 item WPSI. The assessment was done before and 6 months after the treatment Only 50% of participants enrolled completed the study.
Notes Risk of bias Item Allocation concealment?
Authors judgement No
Description C - Inadequate
Sultana 1987 Methods Unblinded, controlled clinical trial. Two arms: one treatment arm, comprising multiple behavioural interventions and the other control. Every third participant was allocated to control group. Duration of baseline, treatment and follow-up phases were not clear This study included 150 adults with uncontrolled epilepsy, as judged by treating neurologist, and were allocated to the treatment and control groups in 2:1 ratio.The mean age of the participants in the treatment group was 29.068.34 for males and 257.4 for females while in the control group it was 29.138.79 for men and 26.148.47 for women. There were 43 men and 42 women in the treatment group besides 15 others who dropped out in the early part of the study. In the control group there were 24 men and 26 women. The baseline monthly seizure frequency (meanSD) in the treatment group and among controls were 5.784.05 and 4.443.7 for simple partial seizures; 5.444.18 and 4.542.59 for complex partial seizures; 1.762.36 and 1.791.16 for GTCS. AED therapy not described The treatment group received Jacobsons muscle relaxation, behaviour therapy such as conditioning, desensitization, aversion, assertive training and supportive psychotherapy. The controls received no treatment. Seizure frequency The psychological test battery included the Bender Visual Motor Gestalt Test, Wechsler Memory Scale - Form 1; the Bhatias battery of performance test of intelligence; self analysis scale; Spielbergers anger expression scale; the personality trait inventory; the adjustment rating inventory (Bells 1967), behaviour rating scale and interpersonal attitude scale Seizure outcomes available for 72 participants in the treatment group and 36 in the control group. Psychological outcomes available for 59 in treatment group and 29 controls
Participants
Interventions
Outcomes
Notes
Risk of bias Item Allocation concealment? Authors judgement No Description C - Inadequate
26
Tan 1986 Methods Unblinded, controlled clinical trial. Three arms: one - group cognitive behaviour therapy; second - supportive counselling and third - none (waiting list controls). The randomization method is not clear. Baseline phase = 2 to 4 weeks; treatment phase = 8 weeks; follow up after 4 months 30 adult individuals who had signicant psychological problems and inadequate seizure control (as judged by the treating neurologist) were randomized 10 to each group. 10 men and 17 women completed the study. They had a mean age of 33.411.1 years. Two participants in each group had probable to denite co-existing hysterical seizures. The duration of epilepsy was 15.58.9 years. The mean baseline weekly seizure frequency in the 3 groups were 10.75, 1.53 and 4.64 respectively. Individuals with mental retardation and psychotic individuals were excluded. All were continued on AEDs Group I received group cognitive behaviour therapy, group II received supportive counselling and served (according to authors as attention-control group) while group III received no intervention. The treatment was given in 2 hour sessions once a week for 8 weeks Seizure frequency Minnesota Multiphasic Personality Inventory WPSI BDI The tests were administered before intervention and at 4 months follow up Two participants from the cognitive behaviour treatment group and one from the waiting list control group did not complete the study. There is considerable variation in the mean baseline weekly seizure frequencies in the three groups
Participants
Interventions
Outcomes
Notes
Risk of bias Item Allocation concealment? Authors judgement Unclear Description B - Unclear
AED: antiepileptic drug BDI: Beck Depression Inventory EEG: electroencephalogram WPSI: Washington Psychosocial Inventory
Characteristics of excluded studies [ordered by study ID]
Study Berger NM 2001
Reason for exclusion This study was published as a conference abstract. We were unable to get details of study and outcomes data despite several attempts to contact the author
27
(Continued)
Earl 1986
The study includes 27 individuals with epilepsy, who were chronic job losers and studies the effect of family counselling. The methods states that the clients were randomly allocated. However, group I (controls) had 11 participants, group II (single session with the family) 6 and group III (family therapy) 10 participants. It is not clear whether there were a large number of drop outs in group II. Moreover, the number of participants in group I is given as 11 in Table I and 10 in Table III. The number of participants in group III is given as 10 in Table I and 11 in Table III This is a controlled study comparing 2 groups of adults with epilepsy with signicant psychological disorder, treated with 2 psychological treatments (self control and alleviation of psychological disorder) in a balanced crossover design and a third group of adults with epilepsy without signicant psychological disturbance treated with self control measures. Since the objective of this review is not to compare the efcacy of one form of psychological intervention with the other, this study has not been included This is a study comparing the efcacy of EEG operant conditioning in 8 individuals with epilepsy, randomized to 2 different EEG feedback training groups. Since the objective of this review is not to compare the efcacy of one form of bio-feedback intervention with the other, this study has not been included This study is a child centred training model for self management of chronic illnesses such as epilepsy and asthma. It is not clear whether the randomization was done separately for each disease. Among people with epilepsy only 52.4% in the treatment group and 70.3% in the control group completed the study. Outcomes data in Table III is based on still fewer individuals
Gillham 1990
Sterman 1980
Tieffenberg 2000
EEG: electroencephalogram
Characteristics of studies awaiting assessment [ordered by study ID]

Rau 2006 Methods A prospective, controlled, multi-centre, pre-post study design was used to examine the efcacy of the program in the treatment group compared to the waiting group (control group) 55 parents of the treatment group and 48 parents of control group completed the questionnaires three months before the course and three months later. 31 children, who participated in the program, and 19 children in the control group completed the questionnaires immediately before the course and three months later Modular educational program for children with epilepsy and their parents (FAMOSES) compared to waiting list controls Knowledge, coping, treatment outcome, emotional and practical adaptation to epilepsy. Questionnaires included epilepsy specic scales regarding knowledge, attitudes, restrictions in daily living, epilepsy related fears, coping with the chronic disease and generic instruments (quality of life, KINDL) Publication is in German language. The author has been contacted to obtain the full publication
Participants
Interventions
Outcomes
Notes
28
DATA AND ANALYSES
Comparison 1. Relaxation therapy versus control
Outcome or subgroup title 1 50% or greater reduction in seizure frequency 1.1 Relaxation therapy versus attention control 1.2 Relaxation therapy versus no intervention 2 Seizure free 2.1 Relaxation therapy versus attention control 2.2 Relaxation therapy versus no intervention 3 Reduction in seizure frequency 3.1 Relaxation therapy versus no intervention 3.2 Relaxation therapy versus attention control
No. of studies 3 3 1 3 3 1 3 1 3
No. of participants
Statistical method Peto Odds Ratio (Peto, Fixed, 95% CI) Peto Odds Ratio (Peto, Fixed, 95% CI) Peto Odds Ratio (Peto, Fixed, 95% CI) Peto Odds Ratio (Peto, Fixed, 95% CI) Peto Odds Ratio (Peto, Fixed, 95% CI) Peto Odds Ratio (Peto, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI)
Effect size Totals not selected Not estimable Not estimable Totals not selected Not estimable Not estimable Totals not selected Not estimable Not estimable
Comparison 2. Cognitive behavioural therapy versus control
Outcome or subgroup title 1 50% or greater reduction in seizure frequency (Group CBT) 1.1 Group cognitive behaviour therapy versus no intervention 1.2 Group cognitive behaviour therapy versus attention control 2 Seizure free (Group CBT) 2.1 Group cognitive behaviour therapy versus attention control 2.2 Group cognitive behaviour therapy versus no intervention 3 Effect of Individual/Group CBT on seizures 3.1 6 months follow-up 3.2 1 year follow-up
No. of studies 1 1 1 1 1 1 1 1 1
No. of participants
Statistical method Peto Odds Ratio (Peto, Fixed, 95% CI) Peto Odds Ratio (Peto, Fixed, 95% CI) Peto Odds Ratio (Peto, Fixed, 95% CI) Peto Odds Ratio (Peto, Fixed, 95% CI) Peto Odds Ratio (Peto, Fixed, 95% CI) Peto Odds Ratio (Peto, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI)
Effect size Totals not selected Not estimable Not estimable Totals not selected Not estimable Not estimable Subtotals only -5.16 [-7.18, -3.14] -5.18 [-7.14, -3.22]
27 27
Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI)
29
Comparison 3. Galvanic skin response biofeedback
Outcome or subgroup title 1 50% or greater reduction in seizure frequency
No. of studies 1
No. of participants 18
Statistical method Peto Odds Ratio (Peto, Fixed, 95% CI)
Effect size 12.81 [1.88, 87.06]
Comparison 4. Cognitive behavioural therapy versus yoga
Outcome or subgroup title 1 Seizure Free 2 50% or greater reduction in seizure frequency 3 50% or greater reduction in seizure duration
No. of studies 1 1 1
No. of participants 18 18 18
Statistical method Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI)
Effect size 1.0 [0.16, 6.42] 1.29 [0.07, 24.38] 1.5 [0.23, 9.80]
Analysis 1.1. Comparison 1 Relaxation therapy versus control, Outcome 1 50% or greater reduction in seizure frequency.
Review: Psychological treatments for epilepsy
Comparison: 1 Relaxation therapy versus control Outcome: 1 50% or greater reduction in seizure frequency
Study or subgroup
Treatment n/N
Control n/N
Peto Odds Ratio Peto,Fixed,95% CI
1 Relaxation therapy versus attention control Dahl 1987 Puskarich 1992 Rousseau 1985 2 Relaxation therapy versus no intervention Dahl 1987 4/6 0/6 15.64 [ 1.57, 155.75 ] 4/6 5/13 2/4 0/6 2/11 1/4 15.64 [ 1.57, 155.75 ] 2.56 [ 0.45, 14.44 ] 2.54 [ 0.17, 37.01 ]
0.1 0.2
0.5
10
Favours control
Favours treatment
30
Analysis 1.2. Comparison 1 Relaxation therapy versus control, Outcome 2 Seizure free.
Comparison: 1 Relaxation therapy versus control Outcome: 2 Seizure free
Study or subgroup
Treatment n/N
Control n/N
1 Relaxation therapy versus attention control Dahl 1987 Puskarich 1992 Rousseau 1985 2 Relaxation therapy versus no intervention Dahl 1987 0/6 0/6 0.0 [ 0.0, 0.0 ] 0/6 0/13 1/4 0/6 0/11 0/4 0.0 [ 0.0, 0.0 ] 0.0 [ 0.0, 0.0 ] 7.39 [ 0.15, 372.38 ]
0.1 0.2
0.5
10
Favours control
Favours treatment
Analysis 1.3. Comparison 1 Relaxation therapy versus control, Outcome 3 Reduction in seizure frequency.
Comparison: 1 Relaxation therapy versus control Outcome: 3 Reduction in seizure frequency
Study or subgroup
Treatment N Mean(SD)
Control N Mean(SD)
Mean Difference IV,Fixed,95% CI
1 Relaxation therapy versus no intervention Dahl 1987 5 49.82 (28.18) 6 0.04 (18.99) 49.78 [ 20.78, 78.78 ]
2 Relaxation therapy versus attention control Dahl 1987 Puskarich 1992 Rousseau 1985 5 13 4 49.82 (28.18) 37.16 (33.8) 55.19 (28.73) 5 11 4 -309.14 (464.95) 10.08 (55.46) 22.51 (26.4) 358.96 [ -49.33, 767.25 ] 27.08 [ -10.49, 64.65 ] 32.68 [ -5.56, 70.92 ]
-10
-5
10
Favours control
Favours treatment
31
Analysis 2.1. Comparison 2 Cognitive behavioural therapy versus control, Outcome 1 50% or greater reduction in seizure frequency (Group CBT).
Comparison: 2 Cognitive behavioural therapy versus control Outcome: 1 50% or greater reduction in seizure frequency (Group CBT)
Study or subgroup
Treatment n/N
Control n/N
1 Group cognitive behaviour therapy versus no intervention Tan 1986 1/10 1/10 1.00 [ 0.06, 17.25 ]
2 Group cognitive behaviour therapy versus attention control Tan 1986 1/10 2/10 0.47 [ 0.04, 5.19 ]
0.1 0.2
0.5
10
Favours control
Favours treatment
32
Analysis 2.2. Comparison 2 Cognitive behavioural therapy versus control, Outcome 2 Seizure free (Group CBT).
Comparison: 2 Cognitive behavioural therapy versus control Outcome: 2 Seizure free (Group CBT)
Study or subgroup
Treatment n/N
Control n/N
1 Group cognitive behaviour therapy versus attention control Tan 1986 1/10 1/10 1.00 [ 0.06, 17.25 ]
2 Group cognitive behaviour therapy versus no intervention Tan 1986 1/10 1/10 1.00 [ 0.06, 17.25 ]
0.1 0.2
0.5
10
Favours control
Favours treatment
33
Analysis 2.3. Comparison 2 Cognitive behavioural therapy versus control, Outcome 3 Effect of Individual/Group CBT on seizures.
Comparison: 2 Cognitive behavioural therapy versus control Outcome: 3 Effect of Individual/Group CBT on seizures
Study or subgroup
Treatment N Mean(SD)
Control N Mean(SD)
Weight
1 6 months follow-up Lundgren 2006 14 0.7 (0.89) 13 5.86 (3.62) 100.0 % -5.16 [ -7.18, -3.14 ]
Subtotal (95% CI)

Heterogeneity: not applicable
14
13
100.0 %
-5.16 [ -7.18, -3.14 ]
Test for overall effect: Z = 5.00 (P < 0.00001) 2 1 year follow-up Lundgren 2006 14 0.62 (0.86) 13 5.8 (3.51) 100.0 % -5.18 [ -7.14, -3.22 ]
Subtotal (95% CI)

14
13
100.0 %
-5.18 [ -7.14, -3.22 ]
Test for overall effect: Z = 5.18 (P < 0.00001) Test for subgroup differences: Chi2 = 0.00, df = 1 (P = 0.99), I2 =0.0%
-10
-5
10
Favours treatment
Favours control
34
Analysis 3.1. Comparison 3 Galvanic skin response biofeedback, Outcome 1 50% or greater reduction in seizure frequency.
Comparison: 3 Galvanic skin response biofeedback Outcome: 1 50% or greater reduction in seizure frequency
Study or subgroup
Treatment n/N
Control n/N 0/8
Weight
Nagai 2004
6/10
100.0 %
12.81 [ 1.88, 87.06 ]
Total (95% CI)

10
100.0 %
12.81 [ 1.88, 87.06 ]
Total events: 6 (Treatment), 0 (Control) Test for overall effect: Z = 2.61 (P = 0.0091) Test for subgroup differences: Not applicable
0.01
0.1
10
100
Favours Control
Favours Treatment
Analysis 4.1. Comparison 4 Cognitive behavioural therapy versus yoga, Outcome 1 Seizure Free.
Comparison: 4 Cognitive behavioural therapy versus yoga Outcome: 1 Seizure Free
Study or subgroup
CBT n/N
Yoga n/N 4/8
Odds Ratio M-H,Fixed,95% CI
Weight
Lundgren 2008
5/10
100.0 %
1.00 [ 0.16, 6.42 ]
Total (95% CI)

Total events: 5 (CBT), 4 (Yoga) Heterogeneity: not applicable
10
100.0 %
1.00 [ 0.16, 6.42 ]
Test for overall effect: Z = 0.0 (P = 1.0) Test for subgroup differences: Not applicable
0.01
0.1
10
100
Favours Yoga
Favours CBT
35
Analysis 4.2. Comparison 4 Cognitive behavioural therapy versus yoga, Outcome 2 50% or greater reduction in seizure frequency.
Comparison: 4 Cognitive behavioural therapy versus yoga Outcome: 2 50% or greater reduction in seizure frequency
Study or subgroup
CBT n/N
Yoga n/N 7/8
Weight
Lundgren 2008
9/10
100.0 %
1.29 [ 0.07, 24.38 ]
Total (95% CI)

10
100.0 %
1.29 [ 0.07, 24.38 ]
0.01
0.1
10
100
Favours Yoga
Favours CBT
Analysis 4.3. Comparison 4 Cognitive behavioural therapy versus yoga, Outcome 3 50% or greater reduction in seizure duration.
Comparison: 4 Cognitive behavioural therapy versus yoga Outcome: 3 50% or greater reduction in seizure duration
Study or subgroup
CBT n/N
Yoga n/N 4/8
Weight
Lundgren 2008
6/10
100.0 %
1.50 [ 0.23, 9.80 ]
Total (95% CI)

10
100.0 %
1.50 [ 0.23, 9.80 ]
0.01
0.1
10
100
Favours Yoga
Favours CBT
36
APPENDICES Appendix 1. CENTRAL search strategy

#1 epilep* #2 MeSH descriptor Epilepsy explode all trees #3 seizure* or convuls* #4 MeSH descriptor Seizures explode all trees #5 (#1 OR #2 OR #3 OR #4) #6 MeSH descriptor Psychotherapy explode all trees #7 biofeedback #8 hypnotherapy or hypnosis #9 relaxation therapy #10 behav* modication #11 behav* therapy #12 suggestion* #13 counsel*ing #14 cognitive behav* therapy #15 conditioning #16 relaxation training #17 desensitisation or densitization #18 physical training #19 aromatherapy #20 (massage or music or dance) therapy #21 behav* countermeasures #22 (#6 OR #7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15 OR #16 OR #17 OR #18 OR #19 OR #20 OR #21) #23 (#22 AND #5)
Appendix 2. MEDLINE search strategy

The following search is based on the Cochrane highly sensitive search strategy for identifying randomized trials in MEDLINE as set out in Appendix 5b of the Cochrane Handbook for Systematic Reviews of Interventions (version 4.2.4, updated March 2005) (Higgins 2005). 1. randomized controlled trial.pt. 2. controlled clinical trial.pt. 3. exp Randomized Controlled Trials/ 4. exp Random Allocation/ 5. exp Double-Blind Method/ 6. exp Single-Blind Method/ 7. clinical trial.pt. 8. Clinical Trial/ 9. (clin$ adj trial$).ab,ti. 10. ((singl$ or doubl$ or trebl$ or tripl$) adj (blind$ or mask$)).ab,ti. 11. exp PLACEBOS/ 12. placebo$.ab,ti. 13. random$.ab,ti. 14. exp Research Design/ 15. or/1-14 16. (animals not humans).sh. 17. 15 not 16 18. (epilep$ or seizure$ or convulsion$).tw. 19. exp epilepsy/
20. exp seizures/ 21. 18 or 19 or 20 22. exp Psychotherapy/ 23. biofeedback.tw. 24. (hypnotherapy or hypnosis).tw. 25. (relaxation adj therapy).tw. 26. (behav$ adj (modication or therapy)).tw. 27. suggestion.tw. 28. counsel$ing.tw. 29. (cognitive adj behav$ adj therapy).tw. 30. conditioning.tw. 31. (relaxation adj training).tw. 32. (desensitisation or desensitization).tw. 33. (physical adj training).tw. 34. aromatherapy.tw. 35. ((massage or music or dance) adj therapy).tw. 36. (behav$ adj countermeasures).tw. 37. or/22-36 38. 21 and 17 and 37
FEEDBACK
Query relating to EEG bio-feedback
Summary The review authors were asked if they were aware of any ongoing trials in EEG bio-feedback in epilepsy. Reply We only found one completed RCT on the use of EEG bio-feedback in epilepsy and this was included in our review. We are unaware of any ongoing RCTs on the use of EEG bio-feedback in epilepsy. Contributors Comment made by Harry McConnell (UK). Sridharan Ramaratnam replied to the comment on behalf of the authors of the review.
EEG biofeedback studies
Summary This review has neglected a very signicant body of literature on EEG biofeedback. I have supplied more than 3 dozen references to the author, Dr. Ramaratnam. The review only cites 1 controlled study of EEG biofeedback with uncontrolled epilepsy, but there are several. This literature was reviewed by: Sterman, M. B. (1990). . Basic concepts and clinical ndings in the treatment of seizure disorders with EEG operant conditioning. Clinical Electroencephalography, 32(1), 45-55. This review notes that in 82% of cases, there was a signicant reduction in seizure rate in patients with uncontrolled epilepsy across the studies reviewed. Below I will provide references to studies that were neglected by Stermans review, or which have occurred since his review. Studies that were not included in the Sterman review that show very signicant outcomes from EEG biofeedback include: Kotchoubey, B. et al. (1999). Changes in EEG power spectra during biofeedback of slow cortical potentials in epilepsy. Applied Psychophysiology
& Biofeedback, 24(4), 213-233; Kotchoubey, D., et al. (2001). Modication of slow cortical potentials in patients with refractory epilepsy: A controlled outcome study. Epilepsia, 42(3), 406-416. Birbaumer, N. et al. (1991). Clinical psychological treatment of epileptic seizures: a controlled study. Chapter in A. Ehlers et al. (Eds.), Perspectives and Promises in Clinical Psychology. New York: Plenum Press; Rockstroh, B. et al. (1993). Cortical self-regulation in patients with epilepsies. Epilepsy Research, 14, 63-72; Uhlmann, C., & Froscher, W. (2001). Biofeedback treatment in patients with refractory epilepsy: Changes in depression and control orientation. Seizure, 10(1), 34-38. It is my hope that the Ramaratnam, Baker and Goldstein review will be updated to include these very important studies which clearly demonstrate the effectiveness of EEG biofeedback in assisting with the treatment of uncontrolled epilepsy. Reply Thank you for your comments regarding our Cochrane review Psychological Treatments for Epilepsy. We, as part of the Cochrane collaboration, welcome criticisms and are always on the look out for studies, which may have been missed by our searches. The objective of the collaboration is to produce systematic reviews with the minimum of bias. Because randomized controlled trials (RCTs) are the least biased means of assessing the effects of medical interventions, our review included only RCTs. Unfortunately, most studies of EEG biofeedback (including the references you have provided) are not RCTs and hence are prone to considerable bias. The majority are either case reports or before and after comparisons or N1, N2, N3 comparisons. Many of them are unblinded, with a small sample size. Most studies do not report seizure outcomes in detail. Some studies have investigated cognitive outcomes, depression or EEG only. These methodological deciencies have been highlighted in the text of the review. I once again thank you for your interest in the review. Contributors Comment made by Professor D. Corydon Hammond (USA). Sridharan Ramaratnam replied to the comment on behalf of the authors of the review.
WHATS NEW
Last assessed as up-to-date: 14 February 2011.
Date 21 January 2010
Event New search has been performed
Description Searches updated 11 January 2011; three new studies have been added (Dilorio 2009; Lundgren 2008; Pramuka 2007). Conclusions remain unchanged.
HISTORY
Protocol rst published: Issue 2, 2000 Review rst published: Issue 4, 2001
39
Date 24 October 2008
Event Amended
Description Search strategy amended to comply with RevMan 5 format. Substantive update.
1 May 2008 1 May 2008
New search has been performed
New citation required but conclusions have not We re-ran our searches on 1 July 2007. We have changed added two new studies of cognitive behaviour therapy (Lundgren 2006; Martinovic 2006). Changes have been made in all sections (description of studies, methodological quality, results and discussion) dealing with cognitive behaviour therapy. Data tables have been modied. However, there is no change in the conclusions Amended Converted to new review format.
30 April 2008 4 July 2005
New citation required but conclusions have not Substantive amendment. changed
CONTRIBUTIONS OF AUTHORS
Sridharan Ramaratnam and Gus Baker identied the studies, independently extracted data, and co-wrote the review. Laura Goldstein commented on the draft review.
DECLARATIONS OF INTEREST
None known
INDEX TERMS Medical Subject Headings (MeSH)

Biofeedback, Psychology; Cognitive Therapy; Epilepsy [ therapy]; Patient Education as Topic; Psychotherapy [ methods]; Relaxation Therapy
MeSH check words

Humans
40

Epilepsy

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

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

Epilepsy

Hochgeladen von

Copyright:

Verfügbare Formate

Psychological treatments for epilepsy (Review)

Ramaratnam S, Baker GA, Goldstein LH

Psychological treatments for epilepsy

Criteria for considering studies for this review

Data collection and analysis

Primary outcome measures

Search methods for identication of studies

Educational interventions and epilepsy

Risk of bias in included studies

Relaxation therapy and epilepsy

(1) Relaxation therapy

(2A) Cognitive behavioural therapy (CBT) compared to no active treatment

(5) Combined use of relaxation and behaviour therapy

(7) Effects of psychological interventions on specic psychological measures

AUTHORS CONCLUSIONS Implications for practice

Implications for research

References to studies included in this review

References to studies excluded from this review

References to studies awaiting assessment

Characteristics of included studies [ordered by study ID]

Notes Risk of bias Item

Allocation concealment? Dilorio 2009 Methods Participants

Risk of bias Item Blinding? All outcomes Authors judgement No Description

Notes Risk of bias Item Allocation concealment? Lantz 1988 Methods

Authors judgement Unclear

Authors judgement Yes Yes No

Authors judgement Yes Yes No

Computerized randomization table

Notes Risk of bias Item Allocation concealment? May 2002 Methods

Authors judgement Unclear

Notes Risk of bias Item Allocation concealment?

Authors judgement Yes

Blinding? All outcomes Puskarich 1992 Methods

Interventions Outcomes Notes Risk of bias Item Allocation concealment?

Notes Risk of bias Item Allocation concealment?

Risk of bias Item Allocation concealment? Authors judgement No Description C - Inadequate

Characteristics of excluded studies [ordered by study ID]

Study Berger NM 2001

Characteristics of studies awaiting assessment [ordered by study ID]

DATA AND ANALYSES

Comparison 1. Relaxation therapy versus control

Comparison 2. Cognitive behavioural therapy versus control

Comparison 3. Galvanic skin response biofeedback

Outcome or subgroup title 1 50% or greater reduction in seizure frequency

Statistical method Peto Odds Ratio (Peto, Fixed, 95% CI)

Effect size 12.81 [1.88, 87.06]

Comparison 4. Cognitive behavioural therapy versus yoga

Peto Odds Ratio Peto,Fixed,95% CI

Peto Odds Ratio Peto,Fixed,95% CI

Comparison: 1 Relaxation therapy versus control Outcome: 2 Seizure free

Peto Odds Ratio Peto,Fixed,95% CI

Peto Odds Ratio Peto,Fixed,95% CI

Comparison: 1 Relaxation therapy versus control Outcome: 3 Reduction in seizure frequency

Mean Difference IV,Fixed,95% CI

Mean Difference IV,Fixed,95% CI

Peto Odds Ratio Peto,Fixed,95% CI

Peto Odds Ratio Peto,Fixed,95% CI

Peto Odds Ratio Peto,Fixed,95% CI

Peto Odds Ratio Peto,Fixed,95% CI

Mean Difference IV,Fixed,95% CI

Mean Difference IV,Fixed,95% CI

Subtotal (95% CI)

-5.16 [ -7.18, -3.14 ]