The review addressed a clear question, supported by appropriate eligibility criteria. Three database were searched for relevant studies, but the restriction to published English articles meant that some relevant trials may have been missed. No details were provided on whether any methods (e.g. independent, duplicate procedures) were used to reduce the risk of reviewer error and bias throughout the review process.
Trial quality was adequately assessed and was used in interpreting the results of the review. A narrative synthesis of standardised mean differences was presented, without confidence intervals, which made assessment of the reliability of individual results difficult; the lack of pooling, with appropriate weightings, also made interpretation of the review results problematic.
In light of this, the restricted search, the limited quality of the included trials, and the possible reviewer error and bias during the review process, the authors' conclusions should be interpreted with caution.