To synthesise the research regarding the effectiveness of occupational therapy (OT) in the restoration of valued roles, tasks, and activities of persons who have had a stroke.

Dissertation Abstracts (from 1980 to 2000), MEDLINE (from 1984 to 2000), Index Medicus (from 1980 to 1983), PsycLIT (from 1980 to 2000), CINAHL (from 1981 to 2000), the Social Sciences Citation Index (from 1988 to 2000), and Excerpta Medica Volume 19: Rehabilitation Medicine (from 1980 to 2000) were searched. The keywords used were 'occupational therapy' AND 'stroke' OR 'cerebrovascular accident (CVA)', cerebrovascular disorders', hemiplegia', hemiparesis' or 'hemiplegics'. In addition, eight OT journals were handsearched. The reference sections of relevant articles were examined for additional studies until no new studies were found. Only English language publications, and studies published in a peer-reviewed journal or as a thesis or dissertation during 1980 to 2000, were included.

Study designs of evaluations included in the review

The inclusion criteria for study design were not specified. The included studies were randomised controlled trials (RCTs), non-randomised trials and uncontrolled studies.

Specific interventions included in the review

Studies that measured the effect of OT treatment were included. Only studies where the treatment was designated as OT treatment, had been administered by an occupational therapist, or where the article was written by an occupational therapist, were included. The following types of studies were excluded: treatments usually regarded as OT (e.g. cognitive-perceptual training), but not regarded as such by the reviewers; treatments administered by professionals other than occupational therapists; treatments administered by occupational therapists that are not usually regarded as OT with effects not reported in functional terms (e.g. fabrication and application of ankle-foot orthoses); OT delivered in combination with other rehabilitation treatments; and interventions delivered to participants post-stroke, who had other concomitant conditions.

Participants included in the review

Participants receiving OT who had had a stroke were included. Of the 895 participants who completed the trials, 411 had a left cerebrovascular accident and 425 had a right cerebrovascular accident. The mean age ranged from 63.1 to 76.0 years (overall mean 70.3 years). In 4 studies, the participants were 12 months or greater post-stroke, while in 10 studies they were 6 months or less post-stroke. In one study, the participants were of mixed chronicity.

Outcomes assessed in the review

The inclusion criteria were not specified for outcomes. The outcome measures in the included studies were classified into the International Classification of Functioning, Disability and Health (ICIDH-2) categories of participation (role), activity (instrumental activities of daily living, and basic activities of daily living), and structural and functional integrity (mental and physical abilities and capacities).

How were decisions on the relevance of primary studies made?

Two reviewers independently selected the papers and any disagreements were resolved through discussion.

The studies were evaluated for level of evidence using a system developed for the American Occupational Therapy Association's Evidence-based Literature Review Project. The system was based on study design, sample size, internal validity and external validity. Two reviewers independently assessed validity and any disagreements were resolved through discussion.

Two reviewers independently verified the statistics to be noted. The authors do not state how many of the reviewers performed the data extraction. The extracted data included: citation details, participant details, type of treatment, length of treatment, outcome and p-values, effect size, and comments on threats to internal validity. The product-moment correlation was used as the estimate of effect size. If significance levels or effect sizes were not reported, they were calculated provided sufficient information was available (method given in review).

How were the studies combined?

A narrative synthesis was undertaken. Where appropriate, the mean effect sizes of statistically homogeneous studies of similar treatments were combined in a meta-analysis. The effect sizes were weighted by sample size.

How were differences between studies investigated?

The studies were grouped in the narrative synthesis according to the outcome: participation in life roles, instrumental or extended activities of daily living, and basic activities of daily living. The authors reported a statistical test for heterogeneity for a subgroup of studies only.

Fifteen studies (n=984) were included; 10 RCTs, 2 non-randomised controlled trials and 3 uncontrolled studies.

Participation in life roles (3 studies).

The effect of OT appeared to be dependent on the length of treatment and goal specificity. Two studies where the intervention was infrequent and less structured found no improvement in role performance or satisfaction in the treatment group, compared with the control group. The remaining study found that 9 hours of treatment focusing on the practice of skills required for preferred leisure activities was related to a successful outcome.

Instrumental or extended activities of daily living (8 studies).

From the better quality studies it appeared that task-specific training in the home results in improved outcomes for instrumental activities of daily living. Two studies of poorer quality found improved performance on specific instrumental activities of daily living tasks after treatment to improve the abilities that were thought to underlie functional performance.

Basic activities of daily living (9 studies).

Five studies reported an improvement in basic activities of daily living between the intervention and control groups, conditions or times. An overall effect size for the 5 studies could not be calculated due to statistical heterogeneity (chi-squared 22.906, p<0.001). After the elimination of a study with a much larger effect size than the other studies, the 4 remaining studies were considered suitable for pooling (chi-squared 5.805, p>0.01); this resulted in a medium effect size (r=0.30).

OT effectively improves participation and activity after stroke.

This review addressed a very broad review question using adequately specified inclusion and exclusion criteria. The literature search was reasonably comprehensive, although the exclusion of non-English language articles may have resulted in studies being missed. No attempt was made to identify unpublished research, and the restriction to articles from peer-reviewed journals and theses or dissertations may have resulted in the exclusion of relevant studies. A validity assessment was carried out and the results of the studies were discussed in the context of their quality. A heterogeneity analysis was reported for one subgroup of studies only; it would have been helpful to have had a more complete report of the heterogeneity analysis. The authors' conclusions follow broadly from the results, and they appropriately point out that it is unclear from the review which elements of the intervention lead to improved performance.

Practice: The authors state that opportunities should be provided for the practice of client-chosen activities, preferably in a familiar context, using structured instruction. There should also be provision of necessary adaptations in addition to training in their use.

Research: The authors state that research is required to establish which elements of OT interventions are the key to improved outcomes. They also state that further research is required on strategy training, environment-focused therapy, and habit formation or dissolution.

Supported in part by the American Occupational Therapy Association.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.