Smith BE, et al. - A comparative appraisal was conducted of the effect of exercises where the pain was allowed/encouraged with non-painful exercises on pain, function or disability in patients with chronic musculoskeletal pain. The protocols using painful exercises yielded a minor yet prominent benefit over pain-free exercises in the short term, with a moderate quality of evidence. In the medium and long term, no clear superiority of one treatment over another was reported. Pain during therapeutic exercise for chronic musculoskeletal pain did not serve as a barrier to successful outcomes. The effectiveness of loading and resistance programmes into pain for chronic musculoskeletal disorders required additional analyses.

Methods

• The study selection and appraisal was done by 2 authors with regard to the risk of bias.
• Methodological quality was assessed via the Cochrane risk of bias tool, and the Grading of Recommendations Assessment system gauged the quality of evidence.

Results

• 9081 potentially eligible studies were selected.
• Nine papers (from seven trials) with 385 subjects met the inclusion criteria.
• A short- term prominent variation was reported in pain, with moderate quality evidence for a small effect size of -0.27 (-0.54 to -0.05) favouring painful exercises.
• No notable change was seen for pain in the medium and long term, and function and disability in the short, medium and long term.