Clinical heterogeneity was a common problem in Cochrane reviews of physiotherapy and occupational therapy

BACKGROUND AND OBJECTIVE: To identify the strategies used to deal with the clinical heterogeneity of interventions and multiple outcome measures used in Cochrane reviews on physiotherapy and occupational therapy. METHODS: A search for systematic reviews on physiotherapy and occupational therapy in the Cochrane Library was performed. Data on the method of categorization of interventions, on measures, and on the method of data synthesis were systematically extracted. RESULTS: 52 reviews were identified. In 22 (42%) reviews only one index intervention was evaluated, in the other 30 reviews index interventions were categorized. A large diversity in the number and type of outcome measures was found (median 6.5, range 1-23). In 48% of the reviews one or more primary outcome measures were defined. In 52% of the reviews no quantitative data synthesis was performed, whereas five different methods for qualitative data synthesis were applied in 11 reviews. CONCLUSIONS: Limitation to a few outcome measures and explicit procedures for the categorization of interventions might increase the transparency and reproducibility of systematic reviews on physiotherapy and occupational therapy. Qualitative data synthesis is not often applied, although it is a useful tool to summarize results if a quantitative synthesis is not appropriate. International consensus on a method for qualitative synthesis is clearly needed.