Abstract
Background
Schizophrenia is a severe mental illness characterised by delusions and hallucinations. Antipsychotic drugs does reduce these symptoms, but at least half of people given these drugs do not comply with the treatment regimen prescribed.
Objectives
To assess the effects of compliance therapy on antipsychotic medication adherence for people with schizophrenia.
Search methods
Cochrane Schizophrenia Group Trials Register (June 2005).
Selection criteria
We included all randomised controlled trials of 'compliance therapy' for people with schizophrenia or related severe mental disorders.
Data collection and analysis
We independently extracted data and, for dichotomous data, calculated the relative risk (RR), its 95% confidence interval (CI) on an intention to treat basis. We present continuous data using the weighted mean difference statistic.
Main results
We included one trial with relevant and available data (n=56, duration 2 years) comparing compliance therapy with non‐specific counseling. The primary outcome 'non‐compliance with treatment' showed no significant difference between compliance therapy and non‐specific counseling (n=56, RR 1.23 CI 0.74 to 2.05). The compliance therapy did not substantially effect attitudes to treatment (n=50, WMD DAI score ‐2.10 CI ‐6.11 to 1.91). Very few people (˜10%) left the study by one year (n=56, RR 0.5 CI 0.1 to 2.51). Mental state seemed unaffected by the therapy (n=50, WMD PANSS score 6.1 CI ‐4.54 to 16.74) as was insight (n=50, WMD SAI ‐0.5 CI ‐2.43 to 1.43), global functioning (n=50, WMD GAF ‐4.20 CI ‐16.42 to 8.02) and quality of life (n=50, WMD QLS ‐3.40 CI ‐16.25 to 9.45). At both one and two years the average number of days in hospital was non‐significantly reduced for those allocated to the compliance therapy.
Authors' conclusions
There is no clear evidence to suggest that compliance therapy is beneficial for people with schizophrenia and related syndromes but more randomised studies are justified and needed in order for this intervention to be fully examined.
Plain language summary
Compliance therapy for schizophrenia
Relapse in people with schizophrenia is common and in many cases attributable to poor compliance with antipsychotic medication. Compliance therapy was developed to specifically address non‐compliance with antipsychotic medication. We only found one reasonably good but small trial. It did not show that compliance therapy really effected compliance with medication, psychotic symptoms, or quality of life but it was always too small really to show this for certain. The study did, however, suggest that the compliance therapy may help people spend shorter times in hospital across a two year period, when compared with standard care. There is a need for more studies and we have proposed a design that could be conducted within the confines of routine care for outcomes of interest to everyone involved.
Background
Antipsychotic medication has proven efficacy in the treatment of positive symptoms of schizophrenia and the prevention of relapse. In spite of this almost 90% of patients will relapse within the first five years of treatment following an acute episode (Robinson 1999) and in general the illness has a tendency to recur or become chronic (Mason 1996). Several factors have been shown to increase the chance of relapse, but probably the single most important determinant of relapse is the discontinuation of effective antipsychotic drug therapy (Green 1988).
The discontinuation of effective medication has been termed non‐compliance, non‐adherence or non‐concordance. The term non‐compliance emphasises the role of doctors as providing a solution and the patients health beliefs as barriers to be overcome. Other models have been proposed that emphasise the equality between patient and doctor in determining the best treatment option and convey the importance of the therapeutic alliance. The latter model usually describes the discontinuation of effective prescribed medication as non‐concordance and emphasises the role of the patient as the most important determinant (Marinker 1997).
Non‐concordance with drug therapy is common in schizophrenia; approximately 50% of patients are non‐concordant one year after being discharged from hospital (Bartko 1988). Effective interventions for improving antipsychotic drug adherence therefore have the potential to reduce the severity of symptoms of the disorder, prevent future relapse and may improve overall outcome.
Several effective psychosocial interventions are currently available for the treatment of schizophrenia; these include family therapy (Pharoah 2000) and psychoeducational approaches (Pekkala 2002). The mode of effectiveness is however uncertain and may simply result from an improvement in drug compliance. Compliance therapy is a therapy specifically designed to improve concordance with treatment for those with major mental illnesses, but its effectiveness in patients with schizophrenia or other related severe mental illness is uncertain.
Objectives
The primary objective of this review is to assess the effect of 'compliance therapy' on adherence with antipsychotic medication in people with schizophrenia or related psychoses compared to treatment as usual.
The secondary objective is to determine whether 'compliance therapy' is superior to any other intervention in promoting adherence with antipsychotic medication.
Methods
Criteria for considering studies for this review
Types of studies
We included all relevant randomised controlled trials (RCTs) in any language.
Types of participants
We included all people with schizophrenia and non‐affective psychotic mental illness irrespective of mode of diagnosis, age, sex or duration of illness.
Types of interventions
1. Compliance therapy An intervention based on motivational interviewing (Miller 1991) where the participant is invited to review their history of illness, symptoms and side‐effects and consider the benefits and drawbacks of drug treatment (Hayward 1995). The therapist's focus should be on the discrepancy between the participants beliefs and maladaptive behaviours, and should focus on adaptive behaviours. Other components may include a consideration of issues relating to the stigma of mental illness though this is not necessary or sufficient to fulfil the definition of 'compliance therapy'.
2. Standard care We defined standard care as the normal level of psychiatric care provided in the area where the trial was undertaken.
3. Other psychosocial interventions Additional psychological and/or social interventions, such as non‐specific counseling and supportive therapy and other 'talking therapies'.
Types of outcome measures
We, a priori, defined the time periods in which we would report outcomes (short term ‐ up to 12 weeks, medium term ‐ 13‐26 weeks and long term ‐ more than 26 weeks).
Primary outcomes
1. Patient adherence 1.1 Concordance with antipsychotic medication prescription 1.2 Concordance with follow‐up arrangements
Secondary outcomes
1. Death, suicide or natural causes.
2. Leaving the study early.
3. Clinical response 3.1 No clinically significant response in global state ‐ as defined by each of the studies 3.2 Average score/change in global state 3.3 No clinically significant response on psychotic symptoms ‐ as defined by each of the studies 3.4 Average score/change on psychotic symptoms 3.5 No clinically significant response on positive symptoms ‐ as defined by each of the studies 3.6 Average score/change in positive symptoms 3.7 No clinically significant response on negative symptoms ‐ as defined by each of the studies 3.8 Average score/change in negative symptoms. 3.9 No clinically significant response on insight ‐ as defined by each of the studies 3.10 Average score/change in insight
4. Extrapyramidal adverse effects 4.1 Incidence of use of antiparkinson drugs 4.2 No clinically significant extrapyramidal adverse effects ‐ as defined by each of the studies 4.3 Average score/change in extrapyramidal adverse effects
5. Quality of life/satisfaction with care for either recipients of care or carers 5.1 No significant change in quality of life/satisfaction ‐ as defined by each of the studies 5.2 Average score/change in quality of life/satisfaction
6. Costs 6.1 Direct and indirect
Search methods for identification of studies
Electronic searches
We searched the Cochrane Schizophrenia Group's Trials Register (latest update June 2005) using the phrase:
[((*complian* or *educat* or *concord* or *adhere* or *psychoed* or *non‐com* or *refus*) in REFERENCE) and ((*complian* or *educat* or *concord* or *adhere* or *psychoed* or *non‐com* or *refus*) in STUDY)]
This register is compiled by systematic searches of major databases, hand searches and conference proceedings (see Group Module).
Searching other resources
1. Reference searching We inspected the references of all identified studies for more trials.
Data collection and analysis
1. Selection of trials Two reviewers (AM,SL) independently searched for trials, identified potentially relevant abstracts and assessed full papers for inclusion and methodological quality. We resolved any disagreement by discussion.
2. Quality assessment The search for trials was performed independently by two reviewers, as described in the Cochrane Collaboration Reviewers' Handbook (Higgins 2005). When disputes arose as to which category a trial was allocated, resolution was attempted by discussion. When this was not possible and further information was necessary to clarify into which category to allocate the trial, data was not entered and the trial was allocated to the list of those awaiting assessment. We only included trials in category A or B in this review.
3. Data management 3.1 Data extraction This was performed independently by at least two reviewers and the authors of trials were contacted to provide missing data where possible.
3.2 Intention‐to‐treat analysis We excluded data from studies where more than 50% of participants in any group were lost to follow‐up. We were to have performed a sensitivity analysis to assess the impact of this decision. In studies with less than 50% dropout rate, we considered those who left early as having a negative outcome.
4. Data analysis 4.1 Binary data For binary outcomes we calculated an estimation of the relative risk (RR) and its 95% confidence interval (CI) and the number needed to treat statistic (NNT).
4.2 Continuous data 4.2.1 Skewed data: continuous data on clinical and social outcomes are often not normally distributed. To avoid the pitfall of applying parametric tests to non‐parametric data we applied the following standards to all data before inclusion: i. standard deviations and means were reported in the paper or were obtainable from the authors; ii. when a scale starts from a finite number (such as 0), the standard deviation, when multiplied by 2, was less than the mean (as otherwise the mean was unlikely to be an appropriate measure of the centre of the distribution (Altman 1996). Endpoint scores on scales often have a finite start and end point and this rule can be applied to them.
4.2.2 Summary statistic: for continuous outcomes we estimated a weighted mean difference (WMD) between groups.
4.2.3 Valid scales: we only included continuous data from rating scales if the measuring instrument had been described in a peer‐reviewed journal and the instrument was either a self report or completed by an independent rater or relative (not the therapist) (Marshall 2000).
4.2.4 Endpoint versus change data: where possible, we presented endpoint data and if both endpoint and change data were available for the same outcomes then we only reported the former in this review.
4.2.5 Cluster trials: Studies increasingly employ 'cluster randomisation' (such as randomisation by clinician or practice) but analysis and pooling of clustered data poses problems: Firstly, authors often fail to account for intra class correlation in clustered studies, leading to a 'unit of analysis' error (Divine 1992) ‐ whereby p values are spuriously low, confidence intervals unduly narrow and statistical significance overestimated ‐ causing type I errors (Bland 1997, Gulliford 1999). Secondly, RevMan does not currently support meta‐analytic pooling of clustered dichotomous data, even when these are correctly analysed by the authors of primary studies, since the 'design effect' (a statistical correction for clustering) cannot be incorporated.
Where clustering was not accounted for in primary studies, we would have presented these data in a table, with an (*) symbol ‐ to indicate the presence of a probable unit of analysis error. Subsequent versions of this review will seek to contact first authors of studies to seek intra‐class correlation co‐efficients of their clustered data and to adjust for these using accepted methods (Gulliford 1999). Where clustering has been incorporated into the analysis of primary studies, then we would have also presented these data in a table. No further secondary analysis (including meta‐analytic pooling) will be attempted until there is consensus on the best methods of doing so, and until RevMan, or any other software, allows this. A Cochrane Statistical Methods Workgroup is currently addressing this issue. In the interim, individual studies will be very crudely classified as positive or negative, according to whether a statistically significant result (p<0.05) was obtained for the outcome in question, using an analytic method that allowed for clustering.
5. Test for heterogeneity Firstly, we were to consider all the included studies within any comparison to judge clinical heterogeneity. Then we used visual inspection of graphs to investigate the possibility of statistical heterogeneity. This was supplemented using, primarily, the I‐squared statistic. This provides an estimate of the percentage of variability due to heterogeneity rather than chance alone. Where the I‐squared estimate was greater than or equal to 75%, we interpreted it as indicating the presence of high levels of heterogeneity (Higgins 2005). If inconsistency was high, we would not summate the data, but the data were to be presented separately and we would have investigated the reasons for heterogeneity. The studies responsible for heterogeneity were not to be added to the main body of homogeneous trials by us, but summated and presented separately and reasons for heterogeneity investigated.
6. Addressing publication bias We entered data from all identified and selected trials into a funnel graph (trial effect against trial size) in an attempt to investigate the likelihood of overt publication bias (Egger 1997).
7. Sensitivity analyses 7.1 We would have performed a sensitivity analysis to assess the impact of our decision to exclude trials with more than 50% loss of participants.
7.2 We would have performed a further sensitivity analysis to assess the impact of the inclusion of studies where the adherence of participants was not assessed by a 'highly objective method' such as pill counts or negligible blood levels.
Results
Description of studies
1. Excluded Several studies evaluated interventions based on cognitive behaviour therapy or motivational interviewing which had relevance to our review. Typically, however, the interventions under evaluation within studies involved more than simple compliance therapy. Therefore if family group therapy was also part of the treatment intervention we had to exclude these studies. We identified many such studies which we could have listed under excluded studies but we decided to describe only trials for which the focus was medication compliance and appeared to be based upon the principals of motivational interviewing.
We had to exclude Kemp 1996 because the participants also included some people with 'primary affective disorder'. We contacted the study authors but they were unable to provide data excluding ineligible patients. We made a post‐hoc decision on the minimum proportion of people with schizophrenia that should be in an included study. Following consultation with colleagues, we decided to widen our inclusion criteria to include studies where >80% of randomised people were diagnosed with schizophrenia or related psychotic illness. Kemp 1996 did not meet these revised inclusion criterion. Spooren 1998 also included people with mood and substance misuse disorders. In this case we did not contact the authors because the intervention also did not meet our inclusion criteria as it involved aspects of family therapy and psychoeducation. We felt this to be broader than our definition of compliance therapy. Finally we also excluded Barrowclough 2001 because it provided a family or caregiver as part of the experimental intervention.
2. Ongoing studies We did not identify any ongoing studies.
3. Awaiting assessment We identified one potentially relevant study published in Chinese (Xu 1999) which is currently awaiting assessment.
4. Included We only identified one small (n=56) study for inclusion in this review (O'Donnell 2003).
4.1 Length of trials The single included trial was conducted over a period of two years, although all data except 'inpatient days' were available only at one‐year follow up.
4.2 Participants in O'Donnell 2003 were consenting patients from 94 consecutive admissions to inpatient care. These people were aged 18‐65 with an an IQ of more than 80. All were fluent English speakers and had no evidence of organic disease. An operational (DSM III‐R) diagnosis of schizophrenia was confirmed using a structured clinical interview (SCID).
4.3 Setting O'Donnell 2003 was conduced in an urban catchment area of Dublin, Ireland.
4.4 Study size O'Donnell 2003 randomised 56 people.
4.5 Interventions O'Donnell 2003 defined compliance therapy as a cognitive behavioural intervention with techniques adopted from motivational interviewing, cognitive therapy and psychoeducation. Compliance therapy was conducted according to a manual (Kemp 1996) and delivered over five sessions, each lasting 30‐60 minutes. The patient's illness history, understanding of illness, ambivalence to treatment, maintenance treatment and stigma regarding mental illness were reviewed during the therapy sessions. The control therapy 'non‐specific counseling' also consisted of five sessions of 30‐60 minutes each. However, when patients raised matters relating to medication, they were advised to discuss these matters with their treating teams.
4.6 Outcomes and outcome scales/interviews
4.6.1 Patient adherence O'Donnell 2003 measured compliance with medication using a structured clinical interview (Adams 1993). This uses a four point scale: 1 for 0‐24% compliance (non‐compliant or consistently irregular), 2 for 25‐49% compliance (frequently irregular), 3 for 50‐74% compliance (irregular) and 4 for 75‐100% compliance (regular). The trialists sought further information from family members and health professionals to adjust compliance ratings. O'Donnell 2003 rated people who scored less than or equal to three on this scale as having sub‐optimal compliance.
4.6.2 Attitudes to medication Attitudes to medication were rated using the Drug Attitude Inventory (DAI, Hogan 1983). This is a self report questionnaire containing 30 (long form) or 10 (short form) items relating to the need and likely effects of medication. Each item is scored TRUE/FALSE with higher scores describing more positive attitudes to medication.
4.6.3 Mental state Symptoms were measured using the positive and negative syndrome scale (PANSS, Kay 1987), a 30 item observer‐rated scale with each item rated from one (absent) to seven (extreme). The rater determines the severity of the symptom by reference to particular criteria. The scale is divided into three sub scales: negative symptoms (seven items), positive symptoms (seven items) and general psychopathology (16 items).
4.6.4 Insight This was measured using the schedule for assessment of insight (SAI, David 1990), a scale made up of three distinct components: (a) adherence to treatment, (b) recognition of having a mental illness and (c) ability to recognise psychotic phenomena as abnormal.
4.6.5 Global functioning Global functioning was assessed using the 90‐point global assessment of functioning scale (GAF, Endicott 1976). The GAF scale is rated with respect to psychological and occupational functioning.
4.6.6 Quality of life O'Donnell 2003 rates quality of life using the Heinrichs Quality of Life Scale (QLS, Heinrichs 1984), a 21 item scale with each item rated 0‐6. The scale is rated by semi‐structured interview providing information on symptoms and functioning in the previous four weeks. The scale provides scores along the following dimensions (interpersonal relations and social network, instrumental role functioning, intrapsychic foundations and common objects and activities).
Risk of bias in included studies
1. Randomisation O'Donnell 2003 stated that the study was randomised using "odd and even digits from a standard table of random numbers". The method of randomisation was poorly described and the method of allocation concealment was not stated.
2. Blinding of outcome assessment In O'Donnell 2003 outcomes were rated by a researcher who was said to be blind to the intervention delivered. No details are given regarding how this blinding was maintained or of testing of the blinding.
3. Non‐entry and treatment dropout O'Donnell 2003 accounted for all participants at each stage of the trial. However, reasons for dropout during and following period were not given, except in respect of one person who died before they could provide outcome data.
4. Outcome reporting Dichotomous outcomes were reported for all study people, whether they completed therapy or not. Continuous outcomes were reported at one year follow up (24 randomised to non‐specific counseling and 26 randomised to compliance therapy) although it is unclear whether all people who completed follow up provided data for every outcome.
5. Overall quality O'Donnell 2003 falls within category B of The Cochrane Collaboration Reviewers' Handbook (Higgins 2005) and is therefore at moderate risk of biased results.
Effects of interventions
1. The search The original search in March 2002 yielded 482 references from which we sought 102 papers in full text. We contacted the authors of Kemp 1996 for further data, but ultimately the study did not fulfil our inclusion criteria. The original search was updated in 2004, yielding 179 additional articles, from which we sought 20 articles in full text. One study met our inclusion criteria (O'Donnell 2003). The original search was updated again in June 2005 and a further 50 articles identified, none of which met our inclusion criteria.
2. COMPARISON 1. COMPLIANCE THERAPY versus NON‐SPECIFIC COUNSELING
2.1 Attitudes to treatment 2.1.1 Non‐compliance with medication Dichotomous data on medication compliance was given in O'Donnell 2003 which compared compliance therapy with non‐specific counseling, given as a supplement to usual care. The results at one year tended to favour the control group (n=56, RR 1.23 CI 0.74 to 2.05) but showed no statistically significant difference.
2.1.2 Attitude scores Continuous data on attitudes to treatment in one study also showed no significant difference between compliance therapy and non‐specific counseling (1 study, n=50, WMD ‐2.10 CI ‐6.11 to 1.91).
2.1.3 Compliance with follow‐up arrangements No useable data were available for this outcome.
2.2 Death, suicide or natural causes In O'Donnell 2003 one death is reported in the compliance therapy arm of the trial. No deaths were reported in patients randomised to the non‐specific counseling.
2.3 Leaving the study early O'Donnell 2003 reports how four people left the non‐specific counseling arm of the trial (two did not complete therapy and two refused follow‐up) and two left the compliance therapy arm of the trial (one did not complete therapy, one death occurred) (n=56, RR 0.5 CI 0.1 to 2.51).
2.4 Mental state We found no significant difference in terms of average PANSS scores (n=50, WMD 6.1 CI ‐4.54 to 16.74) although the results tended to non‐significantly favour non‐specific counseling over compliance therapy.
2.5 Insight No significant difference was found in terms of the average score on the scale used (SAI, n=50, WMD ‐0.5 CI ‐2.43 to 1.43).
2.6 Extrapyramidal adverse effects No study reported this outcome.
2.7 Global functioning O'Donnell 2003 found no significant difference in terms of the average scores on the GAF measure (n=50, WMD ‐4.20 CI ‐16.42 to 8.02).
2.8 Quality of life O'Donnell 2003 reports data but found no significant difference was found between compliance therapy and non‐specific counseling (n=50, WMD QLS ‐3.40 CI ‐16.25 to 9.45).
2.9 Service use The average number of days in hospital was reduced for those allocated to the compliance therapy but data are skewed and we are unclear if these differences are statistically significant.
2.10 Costs We found no useable data for this outcome.
3. Sensitivity analyses Sensitivity analyses were not possible given the paucity of data.
Discussion
1. General issues 1.1 Paucity of data The greatest problem we found in assessing the efficacy of compliance therapy for people with schizophrenia was the lack of suitably conducted trials. In Kemp 1996, where a trial assessing compliance therapy had indeed been conducted, unfortunately the sample included people with affective disorders. We were unable to obtain any new data from this study for only participants relevant to this review.
Given that the only eligible study we found involved only 56 people, whatever the findings of this study, compliance therapy must therefore still be considered an experimental intervention.
1.2 Reporting of data O'Donnell 2003 is quite clearly reported. If better concealment of allocation had taken place the results may have been less prone to bias.
2. COMPARISON 1. COMPLIANCE THERAPY versus NON‐SPECIFIC COUNSELING
2.1 Attitudes to treatment By one year there is no convincing evidence that compliance therapy improves adherence with prescribed medication compared to non‐specific counseling, when given in addition to routine care. Of course ODonnell 2003 is small, and may have failed to detect a significant advantage of either compliance therapy or non‐specific counseling due to low statistical power. When the Drug Attitude Inventory is used to score attitude to medication, however, it too does not really suggest any effect of the compliance therapy.
2.2 Death There are far too little data for us to be able to draw any conclusion regarding this outcome.
2.3 Leaving the study early What is notable here is that few left this study by one year (˜10%) in stark contrast to many trials undertaken in order to evaluate drug treatments. There is no suggestion that compliance therapy is off putting but there are too few data for us to be certain.
2.4 Mental state, insight, adverse effects and global functioning There is no evidence from the O'Donnell 2003 study that compliance therapy is associated with a better clinical response than non‐specific counseling on the series of scales used to measure these parameters.
2.5 Quality of life O'Donnell 2003 used a quality of life measure but, again, found no evidence of a real change caused by the experimental intervention. Again, this small study may have failed to detect a significant advantage of either compliance therapy or non‐specific counseling on quality of life due to low statistical power.
2.6 Service use: Inpatient bed occupancy There is a real difficulty interpreting these continuous data that are so skewed. It is intriguing that both one and two year follow up data were suggesting that the compliance therapy group were consistently less using hospital than those allocated to the standard care. For this result, more than any other, this trial should be repeated and more data produced. O'Donnell 2003 has generated an important hypothesis that compliance therapy may not generate much change in standard measures of mental state, quality of life and global functioning, but it may reduce time spent in hospital care.
Authors' conclusions
Implications for practice.
1. For people with schizophrenia There is no evidence to suggest that compliance therapy is helpful in terms of adherence, psychotic symptoms or quality of life. The compliance therapy approach may, however, reduce the time spent in hospital, although this is not known for certain. People with schizophrenia should be reassured that this is an experimental approach that has not been shown to cause harm and may really help reduce hospitalization. It would seem that this approach is likely to be safe to be the focus of a randomised trial as there is no evidence it causes harm and some that it may do some tangible good.
2. For clinicians There is little or no evidence to support or refute the use of compliance therapy in usual clinical care. More trials are needed to clarify whether it has any beneficial effect for people with schizophrenia and it is here that clinicians could help create good evidence. If compliance therapy really could reduce admission times across 24 months it could be have a major part to play in the routine care of people with schizophrenia.
3. For managers and policy makers If the reduction of time spent in hospital could be replicated, compliance therapy could significantly contribute to improved care and cost savings. There is not enough evidence upon which to base policy but there should be.
Implications for research.
1. General O'Donnell 2003 was well reported but more information on allocation concealment would have been useful. Full compliance with CONSORT (Moher 2001) would have been helpful. Other data from two excluded studies would have been most helpful as, in reality, at least as many people with schizophrenia again as were in O'Donnell 2003 have been randomised to compliance therapy in trials. However, relevant data from Kemp 1996 and Spooren 1998 were not possible to extricate from information on the effects of the therapy on a series of other mental health conditions. A central repository of data from past trials would have been useful but is likely to be many years off.
2. More well designed, conducted and reported randomised trials Future trials, such as the one we have outlined in Table 2, should ensure that a clear description of the interventions is given and that any effects of compliance therapy are not confounded by other potentially active interventions. We do feel that such a study is justified from the results of O'Donnell 2003 as this small trial did suggest a real saving of days in hospital care. Such a study would only be meaningful, however, if undertaken within usual resources available to routine care and measure outcomes of relevance to clinicians and recipients of care as well as researchers. Study samples should include people with schizophrenia and closely related disorders, or at least allow data on this group of people to be extracted from the paper.
1. PICO table.
| Methods | Participants | Interventions | Outcomes | Notes |
| Allocation: randomised, concealment clear. Blindness: patients and therapists not blind, assessors blind. Duration: 2 years. Setting: in hospital at start of study, community follow up. | Diagnosis: people with schizophrenia or related disorders. N=300.* Age: working age adults. Sex: men & women. History: people in their first episode reported separately. | 1. Compliance therapy: administered according to manual of Kemp and David, given over 5 sessions of 30‐60 minutes. N=150. 2. Non‐specific' counseling: delivered over 5 sessions of 30‐60 minutes duration. N=150. | Service use: bed occupancy (primary outcome). Compliance: clinical interview. Other routinely recorded measures of mental state, quality of life, general functioning, adverse effects and service use. | * Powered to be reasonably confident of finding a 10% difference between groups for the primary outcome. |
What's new
| Date | Event | Description |
|---|---|---|
| 22 October 2008 | Amended | Converted to new review format. |
Acknowledgements
Simon Gilbody and Stefan Leucht provided helpful comments on several aspects of this review for which we are very grateful. Letters and telephone calls were made to Professor Anthony David and Dr Rosin Kemp and we thank them for responding to our correspondence.
Data and analyses
Comparison 1. COMPLIANCE THERAPY versus NON‐SPECIFIC COUNSELLING.
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
|---|---|---|---|---|
| 1 Attitudes to treatment: 1. Non‐compliance with medication ‐ by 1 year | 1 | 56 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.23 [0.74, 2.05] |
| 2 Attitudes to treatment: 2. Average endpoint score ‐ by 1 year (DAI, high = poor) | 1 | 50 | Mean Difference (IV, Fixed, 95% CI) | ‐2.10 [‐6.11, 1.91] |
| 3 Death ‐ by 1 year | 1 | 56 | Risk Ratio (M‐H, Fixed, 95% CI) | 3.0 [0.13, 70.64] |
| 4 Leaving the study early for any reason ‐ by 1 year | 1 | 56 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.5 [0.10, 2.51] |
| 5 Mental state: Average endpoint score ‐ by 1 year (PANSS, high = poor) | 1 | 50 | Mean Difference (IV, Fixed, 95% CI) | 6.10 [‐4.54, 16.74] |
| 6 Insight: Average endpoint score ‐ by 1 year (Schedule for the Assessment of Insight, high = poor) | 1 | 50 | Mean Difference (IV, Fixed, 95% CI) | ‐0.5 [‐2.43, 1.43] |
| 7 Global functioning: Average endpoint score ‐ by 1 year (GAF, high = poor) | 1 | 50 | Mean Difference (IV, Fixed, 95% CI) | ‐4.20 [‐16.42, 8.02] |
| 8 Quality of life: Average endpoint score ‐ by 1 year (QLS, high = poor) | 1 | 50 | Mean Difference (IV, Fixed, 95% CI) | ‐3.40 [‐16.25, 9.45] |
| 8.1 Quality of life scale | 1 | 50 | Mean Difference (IV, Fixed, 95% CI) | ‐3.40 [‐16.25, 9.45] |
| 9 Service use: Inpatient bed occupancy (high = poor, skewed data) | Other data | No numeric data | ||
| 9.1 one year follow up | Other data | No numeric data | ||
| 9.2 two year follow up | Other data | No numeric data |
1.1. Analysis.
Comparison 1 COMPLIANCE THERAPY versus NON‐SPECIFIC COUNSELLING, Outcome 1 Attitudes to treatment: 1. Non‐compliance with medication ‐ by 1 year.
1.2. Analysis.
Comparison 1 COMPLIANCE THERAPY versus NON‐SPECIFIC COUNSELLING, Outcome 2 Attitudes to treatment: 2. Average endpoint score ‐ by 1 year (DAI, high = poor).
1.3. Analysis.
Comparison 1 COMPLIANCE THERAPY versus NON‐SPECIFIC COUNSELLING, Outcome 3 Death ‐ by 1 year.
1.4. Analysis.
Comparison 1 COMPLIANCE THERAPY versus NON‐SPECIFIC COUNSELLING, Outcome 4 Leaving the study early for any reason ‐ by 1 year.
1.5. Analysis.
Comparison 1 COMPLIANCE THERAPY versus NON‐SPECIFIC COUNSELLING, Outcome 5 Mental state: Average endpoint score ‐ by 1 year (PANSS, high = poor).
1.6. Analysis.
Comparison 1 COMPLIANCE THERAPY versus NON‐SPECIFIC COUNSELLING, Outcome 6 Insight: Average endpoint score ‐ by 1 year (Schedule for the Assessment of Insight, high = poor).
1.7. Analysis.
Comparison 1 COMPLIANCE THERAPY versus NON‐SPECIFIC COUNSELLING, Outcome 7 Global functioning: Average endpoint score ‐ by 1 year (GAF, high = poor).
1.8. Analysis.
Comparison 1 COMPLIANCE THERAPY versus NON‐SPECIFIC COUNSELLING, Outcome 8 Quality of life: Average endpoint score ‐ by 1 year (QLS, high = poor).
1.9. Analysis.
Comparison 1 COMPLIANCE THERAPY versus NON‐SPECIFIC COUNSELLING, Outcome 9 Service use: Inpatient bed occupancy (high = poor, skewed data).
| Service use: Inpatient bed occupancy (high = poor, skewed data) | ||||
|---|---|---|---|---|
| Study | Intervention | N | Mean (days) | SD |
| one year follow up | ||||
| O'Donnell 2003 | Compliance therapy | 26 | 26 | 45 |
| O'Donnell 2003 | Non‐specific counselling | 24 | 33 | 57 |
| two year follow up | ||||
| O'Donnell 2003 | Compliance therapy | 26 | 43 | 60 |
| O'Donnell 2003 | Non‐specific counselling | 24 | 50 | 70 |
Characteristics of studies
Characteristics of included studies [ordered by study ID]
O'Donnell 2003.
| Methods | Allocation: 'table of random numbers', concealment: unclear. Blindness: patients and therapists not blind, assessors blind, although no details are given. Duration: 2 years. Setting: in hospital at start of study. | |
| Participants | Diagnosis: schizophrenia (DSM‐III‐R) and IQ>80. N=56. Age: 18‐65, mean 32 years. Sex: 41 men, 15 women. History: 7 first episode schizophrenia. | |
| Interventions | 1. Compliance therapy: administered according to manual of Kemp and David, given over 5 sessions of 30‐60 minutes. N=28. 2. Non‐specific' counseling: delivered over 5 sessions of 30‐60 minutes duration. N=28. | |
| Outcomes | Compliance: clinical interview. Attitudes to treatment: DAI. Insight: SAI. Mental state: PANSS. Level of functioning: GAF. Quality of life: QLS. Service use: bed occupancy. | |
| Notes | ||
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Allocation concealment? | Unclear risk | B ‐ Unclear |
General abbreviations: IQ ‐ Intelligence Quotient N ‐ Number of people in the study
Diagnostic tools: DSM‐III‐R ‐ Diagnostic and Statistical Manual of Diseases, third revision, revised.
Rating scales: DAI ‐ Drug Attitude Inventory SAI ‐ Schedule for the Assessment of Insight PANSS ‐ Positive and Negative Syndrome Scale GAF ‐ Global Assessment of Functioning QLS ‐ Heinrich's Quality of Life Scale
Characteristics of excluded studies [ordered by study ID]
| Study | Reason for exclusion |
|---|---|
| Barrowclough 2001 | Allocation: randomised. Participants: people with schizophrenia. Interventions: routine care versus motivational nterviewing, CBT and family/caregiver intervention, not simply compliance therapy. |
| Kemp 1996 | Allocation: randomised. Participants: people with schizophrenia, related psychotic disorders and primary affective disorder, unclear how many suffered from schizophrenia. |
| Spooren 1998 | Allocation: randomised. Participants: people with schizophrenia, substance misuse disorders, mood disorder and adjustment disorder, unclear how many suffered from schizophrenia. |
Contributions of authors
Andrew McIntosh ‐ initiation of the review, protocol production, searching, data extraction, analysis, data interpretation and writing the final report.
Louise Conlon ‐ searching and writing the final report.
Stephen Lawrie ‐ protocol production, searching, data interpretation and writing the final report.
Andrew Stanfield ‐ searching, data extraction, data interpretation and writing the final report.
Sources of support
Internal sources
Edinburgh University, UK.
National University of Ireland, Ireland.
External sources
No sources of support supplied
Declarations of interest
SL has been paid for speaking about critical appraisal by employees of the manufacturers of olanzapine, quetiapine, risperidone, and ziprasidone, and has been paid to speak about the management of schizophrenia by employees of the manufacturers of amisulpiride, olanzapine, risperidone, and clozapine.
Edited (no change to conclusions)
References
References to studies included in this review
O'Donnell 2003 {published data only}
- O'Donnell C, Donohoe G, Sharkey L, Owens N, Migone M, Harries R, Kinsella A, Larkin C, O'Callagham E. Compliance therapy: a randomised controlled trial in schizophrenia. BMJ 2003;327:834‐7. [DOI] [PMC free article] [PubMed] [Google Scholar]
References to studies excluded from this review
Barrowclough 2001 {published data only}
- Barrowclough C, Haddock G, Tarrier N, Lewis SW, Moring J, O'Brien R, Schofield N, McGovern J. Randomized controlled trial of motivational interviewing, cognitive behaviour therapy, and family intervention for patients with comorbid schizophrenia and substance use disorders. American Journal of Psychiatry 2001;158:1706‐13. [DOI] [PubMed] [Google Scholar]
Kemp 1996 {published and unpublished data}
- Hayward P, Chan N, David A, Kemp R, Youle S, David A. Medication self‐management: a preliminary report on an intervention to improve medication compliance. Journal of Mental Health 1995;4:511‐7. [Google Scholar]
- Kemp R, David A, Hayward P. Compliance therapy: an intervention targeting insight and treatment adherence in psychotic patients. Behavioural and Cognitive Psychotherapy 1996;4:331‐50. [Google Scholar]
- Kemp R, Hayward P, Applewhaite G, Everitt B, David A. Compliance therapy in psychotic patients: randomised controlled trial. BMJ 1996;312:345‐9. [MEDLINE: ] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kemp R, Kirov G, Everitt B, Hayward P, David A. Randomised controlled trial of compliance therapy. 18‐month follow‐up. British Journal of Psychiatry 1998;172:413‐9. [DOI] [PubMed] [Google Scholar]
- Kemp R, Krov G, Hayward P, David A. A randomised controlled trial of compliance therapy in psychosis: follow‐up results and. Schizophrenia Research 1996;18:135. [Google Scholar]
Spooren 1998 {published data only}
- Spooren D, Heeringen C, Jannes C. Strategies to increase compliance with out‐patient aftercare among patients referred to a psychiatric emergency department: a multi‐centre controlled intervention study. Psychological Medicine 1998;28:949‐56. [DOI] [PubMed] [Google Scholar]
References to studies awaiting assessment
Xu 1999 {published data only}
- Xu Z, Weng Y, Hou Y. Efficacy and follow‐up resarch of medication management module training for schizophrenic patients. Chinese Journal of Psychiatry 1999;32(2):96‐9. [Google Scholar]
Additional references
Adams 1993
- Adams SG Jr, Howe JT. Predicting medication compliance in a psychotic population. Journal of Nervous and Mental Disease 1993;181:558‐60. [DOI] [PubMed] [Google Scholar]
Altman 1996
- Altman DG, Bland JM. Detecting skewness from summary information. BMJ 1996;313:1200. [DOI] [PMC free article] [PubMed] [Google Scholar]
Bartko 1988
- Bartko G, Herczeg I, Zador G. Clinical symptomatology and drug compliance in schizophrenic patients. Acta Psychiatrica Scandinavica 1988;77(1):74‐6. [DOI] [PubMed] [Google Scholar]
Bland 1997
- Bland JM, Kerry SM. Statistics notes. Trials randomised in clusters. BMJ 1997;315:600. [DOI] [PMC free article] [PubMed] [Google Scholar]
David 1990
- David AS. Insight and psychosis. British Journal of Psychiatry 1990;156:798‐808. [DOI] [PubMed] [Google Scholar]
Divine 1992
- Divine GW, Brown JT, Frazer LM. The unit of analysis error in studies about physicians' patient care behavior. Journal of General Internal Medicine 1992;7:623‐9. [DOI] [PubMed] [Google Scholar]
Egger 1997
- Egger M, Davey SG, Schneider M, Minder C. Bias in meta‐analysis detected by a simple graphical test. BMJ 1997;315:629‐34. [DOI] [PMC free article] [PubMed] [Google Scholar]
Endicott 1976
- Endicott J, Spitzer RL, Fleiss JL, Cohen J. The Global Assessment Scale. A procedure for measuring overall severity of psychiatric disturbance. Archives of General Psychiatry 1976;33:766‐71. [DOI] [PubMed] [Google Scholar]
Green 1988
- Green JH. Frequent rehospitalization and noncompliance with treatment. Hospital & Community Psychiatry 1988;39(9):963‐6. [DOI] [PubMed] [Google Scholar]
Gulliford 1999
- Gulliford MC, Ukoumunne OC, Chinn S. Components of variance and intraclass correlations for the design of community‐based surveys and intervention studies: data from the Health Survey for England 1994. American Journal of Epidemiology 1999;149:876‐83. [DOI] [PubMed] [Google Scholar]
Hayward 1995
- Hayward P, Chan N, David A, Kemp R, Youle S, David A. Medication self‐management: a preliminary report on anintervention to improve medication compliance. Journal of Mental Health 1995;4:511‐7. [Google Scholar]
Heinrichs 1984
- Heinrichs DW, Hanlon TE, Carpenter WT Jr. The quality of life scale: an instrument for rating the schizophrenic deficit syndrome. Schizophrenia Bulletin 1984;10:338‐98. [DOI] [PubMed] [Google Scholar]
Higgins 2005
- Higgins JPT, Green S, editors. Assessment of study quality. Cochrane Handbook for Systematic Reviews of Interventions 4.2.5 [updated May 2005];. Chichester: John Wiley & Sons, Ltd., 2005. [Google Scholar]
Hogan 1983
- Hogan TP, Awad AG, Eastwood R. A self‐report scale predictive of drug compliance in schizophrenics: reliability and discriminative validity. Psychological Medicine 1983;13:177‐83. [DOI] [PubMed] [Google Scholar]
Kay 1987
- Kay SR, Fiszbein A, Opler LA. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophrenia Bulletin 1987;13(2):261‐76. [DOI] [PubMed] [Google Scholar]
Marinker 1997
- Marinker M. Personal paper: writing prescriptions is easy. BMJ 1997;314:747‐8. [DOI] [PMC free article] [PubMed] [Google Scholar]
Marshall 2000
- Marshall M, Lockwood A, Adams C, Bradley C, Joy C, Fenton M. Unpublished rating scales ‐ a major source of bias in randomised controlled trials of treatments for schizophrenia?. British Journal of Psychiatry 2000;176:249‐52. [DOI] [PubMed] [Google Scholar]
Mason 1996
- Mason P, Harrison G, Glazebrook C, Medley I, Croudace T. The course of schizophrenia over 13 years. A report from the International Study on Schizophrenia (ISoS) coordinated by the World Health Organization. British Journal of Psychiatry 1996;169(5):580‐6. [97086857] [DOI] [PubMed] [Google Scholar]
Miller 1991
- Miller WR, Rollnick S. Motivational interviewing: preparing people to change. New York: Guilford Press, 1991. [Google Scholar]
Moher 2001
- Moher D, Schulz KF, Altman D. The CONSORT statement: revised recommendations for improving the quality of reports of parallel‐group randomized trials. JAMA 2001;285:1987‐91. [DOI] [PubMed] [Google Scholar]
Pekkala 2002
- Pekkala E, Merinder L. Psychoeducation for schizophrenia. Cochrane Database of Systematic Reviews 2002, Issue 2. [DOI: 10.1002/14651858.CD002831; CD002831] [DOI] [PubMed] [Google Scholar]
Pharoah 2000
- Pharoah FM, Mari JJ, Streiner D. Family intervention for schizophrenia. Cochrane Database of Systematic Reviews 2000, Issue 4. [DOI: 10.1002/14651858.CD000088.pub2] [DOI] [PubMed] [Google Scholar]
Robinson 1999
- Robinson D, Woerner MG, Alvir JM, Bilder R, Goldman R, Geisler S, Koreen A, Sheitman B, Chakos M, Mayerhoff D, Lieberman JA. Predictors of relapse following response from a first episode of schizophrenia or schizoaffective disorder. Archivs of General Psychiatry 1999;56(3):241‐7. [DOI] [PubMed] [Google Scholar]