This review addressed a clear question and had inclusion criteria for participants, intervention/comparator and study design. Inclusion criteria for outcomes were broad. The authors searched a limited range of sources. Search terms and language restrictions were not reported, which made it difficult to assess the risk that relevant studies were missed. Some attempts were made to locate unpublished studies. Risk of publication bias was not assessed.
Relevant aspects of study validity were assessed and the results were used in the synthesis. Study selection, validity assessment and data extraction were conducted by two or more independent reviewers, reducing the risk of reviewer errors or bias. Limited details of included trials were provided. A narrative synthesis was appropriate in view of the clinical heterogeneity of the included trials.
The authors' conclusions reflect the limitations of the evidence and appear reliable.