This review addressed a clear question. The inclusion criteria were generally clear but left some scope for subjective interpretation. The search was limited to one database and selected journals, was restricted to English language publications, and only a few keywords were used, so it is possible that some relevant studies could have been missed. The risk of publication bias was not assessed. The authors did not assess validity, which makes it difficult to comment on the reliability of the included studies and the synthesis derived from them. In additions, since review methods were not reported, the risk of bias and error during the review process is uncertain. Adequate details of the included studies were provided in the text and tables. The studies were combined by meta-analysis but, since the authors do not appear to have assessed heterogeneity, it is unclear whether it was appropriate to pool them. The authors' conclusions are in line with the evidence presented, but should be treated with caution in view of the methodological and reporting limitations of the review.