Inclusion criteria for the review were clearly defined and several relevant databases were searched. There was potential for language bias, as only articles in English were included. Publication bias was detected but the meaningfulness of analysis with less than ten studies was limited. Attempts were made to reduce reviewer error and bias during study selection and quality assessment, but the same methods were not reported for data extraction.
Quality assessment indicated that the quality of the evidence base was variable. There were differences across the trials in terms of intervention, participants and control. The authors noted that most trials had a short duration. Where possible, data were combined using meta-analysis and statistical heterogeneity was reported; analyses showed high level of statistical heterogeneity, which the authors acknowledged.
Overall, the evidence base had several limitations, which reduces the reliability of the pooled results, thus the authors’ conservative conclusions seem appropriate.