Adaptive Interventions for Enhancing Participation Poststroke: A Systematic Review and Meta-Analysis.

Not every included trial could be pooled. Reporting varied enough that some studies had to be left out of the meta-analysis, and attempts to get missing data from the original authors did not always succeed. The main analysis missed statistical significance and its confidence interval was wide, running from a small negative effect to a large positive one, so the true effect on personal goals could be nothing or could be substantial. Only one learning-focused intervention (CO-OP) has been tested in an eligible trial, and only one trial of occupational performance coaching was pooled, so claims about either rest on very little trial evidence. Follow-up periods were too varied to combine, so the review cannot say whether any gains last. By the review's own quality ratings, only 2 of the 6 learning-focused trials and 6 of the 14 coping-focused trials met its higher-quality criteria, and the reviewers did not apply a formal certainty-of-evidence rating. Most trials were small; individual sample sizes ran from 8 to 400 people.

The authors state that they received no financial support for the research, authorship or publication of the article. No other funding source or competing interest is declared in the text.

Do not take this review as proof that CO-OP or occupational performance coaching improves the goals stroke survivors set. Its pooled result for those personal goals did not reach significance (p = .07), and the only statistically significant finding came from four small trials of a secondary outcome, everyday household activities. The authors' call to adopt these interventions goes further than their own numbers do.