Assessment of study quality
RCTs were assessed and scored using the PEDro rating scale, which covers the following: randomisation; allocation concealment; baseline similarity of the treatment groups; blinding of the patient, therapist and outcome assessors; outcomes measured for more than 85% of those randomised; data analysed on an intention-to-treat basis; statistical comparison between treatments; point measures and variability of the outcome measures reported; and side-effects reported. The maximum possible PEDro score was 10 points. Two reviewers scored the RCTs, and any disagreements were resolved with the aid of a third reviewer where required. The validity of cohort and case-control studies was assessed using the Newcastle-Ottawa Scale.