Scolaris Content Display Scolaris Content Display

Cochrane Database of Systematic Reviews Protocol - Intervention

Exercise therapy for schizophrenia

This is not the most recent version

Collapse all Expand all

Abstract

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:

To determine the effects of exercise/physical activity programmes for people with schizophrenia or schizophrenia‐like illnesses.

Background

The health benefits of physical activity and exercise are well documented (Bouchard 1994). There is also an ever‐increasing body of evidence suggesting psychological benefits of such activity (DiLorenzo 1999). Exercise has been linked to improvements in anger and cynical distrust (Hassmen 2000), self‐esteem and physical self‐worth (McAuley 1997) as well as more general psychological conditions/states (Sorensen 1999) in non‐clinical samples. Research has also been carried out into the psychological effects of physical activity and exercise in people with mental illness.

Much of this research has been with people suffering from anxiety and/or depression. A recent systematic review of the effect of exercise interventions for depression identified 72 potentially relevant studies (Lawlor 2001). This review found the effect of exercise on depression to be similar to that of cognitive therapy. It was concluded, however, that the research to date was of poor quality and that it was not possible to determine from the available evidence the full effects of exercise in the management of depression.

With regard to the effects of exercise on schizophrenia, there is a much smaller body of research. Studies tend to consist mostly of case reports and anecdotes but some group studies do exist (Faulkner 1999). Adults with schizophrenia have been found to live quite unhealthy lifestyles even compared with people suffering from other mental illnesses (Brown 1999). For several measures related to physical activity and physical fitness, people with schizophrenia scored well below those free of mental illness and other psychotic patients (Chamove 1986). There have been studies reporting positive effects of exercise on various symptoms of schizophrenia including auditory hallucinations and researchers have suggested that exercise may be an effective adjunctive treatment for the illness (Faulkner 1999).

The active ingredient of exercise therapy for schizophrenia could be the social interaction involved in many forms of exercise. Another proposed mechanism for an effect is that the enhanced self‐esteem experienced by regular exercisers may be the trigger for more positive emotional experiences for people with schizophrenia. Exercise could also be a distraction from auditory hallucinations or emotional conflict for people with schizophrenia and this distraction could allow other positive effects to take root (Faulkner 1999).

Objectives

To determine the effects of exercise/physical activity programmes for people with schizophrenia or schizophrenia‐like illnesses.

Methods

Criteria for considering studies for this review

Types of studies

All relevant, randomised controlled trials. Where a trial was described as 'double‐blind' but it was implied that the study was randomised these trials were included in a sensitivity analysis. If there was no substantive difference within primary outcomes (see types of outcome measures) when these 'implied randomisation' studies were added, then they were included in the final analysis. If there was a substantive difference, only clearly randomised trials were utilised and the results of the sensitivity analysis described in the text. Quasi‐randomised studies, such as those allocating by using alternate days of the week, were excluded.

Types of participants

People diagnosed with schizophrenia or schizophrenia‐like illnesses using any criteria. Trials were included where it was implied that the majority of the participants had a severe mental illness which was likely to be schizophrenia. Trials were not excluded due to age, nationality or gender of participants. Trials were included with participants with any length of illness who were being treated in any treatment setting.

Types of interventions

1. Any intervention, used alone or in conjunction with others, where physical activity or exercise was considered to be the main or active element

Given that there was some evidence of an effect of acute exercise on mood, no minimum length of treatment was applied and interventions lasting for one session or more were included.

2. Standard care
We defined standard care as care that a person would normally receive had they not been included in the research trial. This would include interventions such as medication, hospitalisation, community psychiatric nursing input and day hospital.

3. Other treatments
This would include any other treatment (biological, psychological or social) such as medication, problem solving therapy, psycho‐education, social skills training, cognitive‐behavioural therapy, family therapy or psychodynamic psychotherapy.

For a study to be included, the experimental and comparison interventions had to have had a similar duration.

Types of outcome measures

Outcomes were divided into short term (less than three months), medium term (3‐12 months) and long term (longer than one year).

Primary outcomes

1. Global state
1.1 Relapse

2. Mental state (with particular reference to the positive and negative symptoms of schizophrenia)
2.1 No clinically important change in general mental state

Secondary outcomes

1. Global state
1.1 No clinically important change in global state (as defined by individual studies)
1.2 Average endpoint global state score
1.3 Average change in global state score

2. Mental state (with particular reference to the positive and negative symptoms of schizophrenia)
2.1 Average endpoint general mental state score
2.2 Average change in general mental state scores
2.3 No clinically important change in specific symptoms (positive symptom of schizophrenia negative symptoms of schizophrenia depression mania)
2.4 Average endpoint specific symptom score
2.5 Average change in specific symptom scores

3. General functioning
3.1 No clinically important change in general functioning
3.2 Average endpoint general functioning score
3.4 Average change in general functioning scores
3.5 No clinically important change in specific aspects of functioning, such as social or life skills
3.6 Average endpoint specific aspects of functioning, such as social or life skills
3.7 Average change in specific aspects of functioning, such as social or life skills

4. Behaviour
4.1 No clinically important change in general behaviour
4.2 Average endpoint general behaviour score
4.3 Average change in general behaviour scores
4.4 No clinically important change in specific aspects of behaviour
4.5 Average endpoint specific aspects of behaviour
4.6 Average change in specific aspects of behaviour

5. Adverse effects
5.1 Suicide and all causes of mortality
5.2 Clinically important general adverse effects
5.3 Average endpoint general adverse effect score
5.4 Average change in general adverse effect scores
5.5 Clinically important specific adverse effects
5.6 Average endpoint specific adverse effects
5.7 Average change in specific adverse effects

6. Service outcomes
6.1 Hospitalisation
6.2 Time to hospitalisation

7. Engagement with services

8. Satisfaction with treatment
8.1 Leaving the studies early
8.2 Recipient of care not satisfied with treatment
8.3 Recipient of care average satisfaction score
8.4 Recipient of care average change in satisfaction scores
8.5 Carer not satisfied with treatment
8.6 Carer average satisfaction score
8.7 Carer average change in satisfaction scores

9. Quality of life
9.1 No clinically important change in quality of life
9.2 Average endpoint quality of life score
9.3 Average change in quality of life scores
9.4 No clinically important change in specific aspects of quality of life
9.5 Average endpoint specific aspects of quality of life
9.6 Average change in specific aspects of quality of life

10. Economic outcomes
10.1 Direct costs
10.2 Indirect costs

Search methods for identification of studies

See: Collaborative Review Group search strategy.

Electronic searches

Cochrane Schizophrenia Group's trials register (May 2003) was searched using the phrase

[(*physical* and *fitness*) or (*exercise*) in title or (*physical* and *fitness*) or (*exercise*) in abstract, index or title terms of REFERENCE] or [Exercise Therapy* in interventions of STUDY]

The Schizophrenia Group's trials register is based on regular searches of BIOSIS Inside, CENTRAL, CINAHL, EMBASE, MEDLINE and PsycINFO; the hand searching of relevant journals and conference proceedings, and searches of several key grey literature sources. A full description is given in the Group's module.

Searching other resources

1. Reference searching
The reference list of all identified studies were inspected for more relevant citations.

2. Personal contact
The first author of each relevant study was contacted for information on unpublished trials.

Data collection and analysis

1. Selection of trials
Two reviewers independently assessed the abstracts of the studies returned by the searches for relevance. When the two reviewers disagreed or the abstract was unclear, the full report was obtained and the assessment process repeated. If a dispute still existed this was resolved by discussion and if there was still insufficient information to assess relevance the first author of the study was contacted.

2. Assessment of methodological quality
Two reviewers independently assessed the methodological quality of all relevant studies. Only the studies classified as A and B according to the criteria set out in the Cochrane Handbook (Clarke 2002) were included.

3. Data collection
Two reviewers extracted data from each study independently with disagreements resolved by documented discussion. Data did not need to be published to be included in this review.

4. Analysis
Data were entered into RevMan independently by each of the reviewers.

4.1. Binary Data
For dichotomous outcomes the fixed effects relative risk (RR) and its 95% confidence interval (CI) were calculated. If heterogeneity was found (see below) a random effects model was used.

4.2. Continuous data
In the case of continuous outcome measurement a weighted mean difference (WMD) between groups was estimated. Again if evidence of heterogeneity was found a random effects model was used.

Continuous data assessing mental health outcome is often not normally distributed and so to avoid the pitfall of applying parametric tests to non‐parametric data the following standards were applied to all data before inclusion: (a) standard deviations/errors and means were reported in the paper or obtainable from the author; (b) when a scale started from a finite number (such as zero), the standard deviation, when multiplied by two, was less than the mean (as otherwise the mean is unlikely to be an appropriate measure of the centre of the distribution ‐ Altman 1996). Endpoint scores on scales often have a finite starting point and top score; therefore this rule can be applied to them.

4.3. Scale derived data
Data obtained using a scale were only included if its psychometric properties (i.e. concerning reliability and validity) were reported in a peer reviewed journal (Marshall 2000). Furthermore these data were only included if the scale was either self‐report or completed by an independent rater or a relative and not the therapist or provider of the intervention.

4.4. Intention to treat analysis
For dichotomous data a 'once‐randomised‐always‐analyse' approach was taken and it was assumed that all those lost to follow‐up had relapsed. For continuous data a 'last‐observation‐carried‐forward' (LOCF) approach was acceptable although far from ideal. Outcome data were not used, however, when more than 50% of data were lost because of attrition.

4.5 Test for heterogeneity
A Chi‐square test was used, as well as visual inspection of graphs, to investigate the possibility of heterogeneity. A significance level less than 0.10 was interpreted as evidence of heterogeneity. If heterogeneity was found, the data were re‐analysed using a random effects model to see if this made a substantial difference. If it did, the studies responsible for heterogeneity were not added to the main body of homogeneous trials, but summated and presented separately and reasons for heterogeneity investigated.

4.6 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 presented the data in a table, with a (*) symbol ‐ to indicate the presence of a probable unit of analysis error. In subsequent versions of this review we will attempt to contact first authors of studies to seek intra‐class correlation co‐efficients of their clustered data and to adjust for this using accepted methods (Gulliford 1999). Where clustering has been incorporated into the analysis of primary studies, then we will also present these data as if from a non‐cluster randomised study, but adjusted for the clustering effect. We have sought statistical advice from the MRC Biostatistics Unit, Cambridge, UK. Dr Julian Higgins advised that the binary data as presented in a report should be divided by a 'design effect'. This is calculated using the mean number of participants per cluster (m) and the intraclass correlation co‐efficient (ICC) [Design effect = 1+(m‐1)*ICC]. Should the ICC not be reported it was assumed to be 0.1 (Ukoumunne 1999).

5. Publication bias
Data from all selected trials were entered into a funnel graph (trial effect against trial size) in an attempt to investigate the likelihood of overt publication or small study bias.