Attempting to Separate Placebo Effects from Exercise in Chronic Pain: A Systematic Review and Meta-analysis.

Exercise is widely recommended for chronic musculoskeletal pain, but it has been unclear how much of its benefit comes from the exercise itself versus placebo-like effects such as attention, expectation, or simply the passage of time. This systematic review and meta-analysis pooled 79 randomised trials comparing exercise training against placebo, true no-treatment controls, or usual care. When compared directly to placebo treatments, based on only 4 trials, there was no statistically significant difference in pain reduction. But exercise did show statistically significant benefits over doing nothing and over usual care.

The authors rated all of this evidence as very low quality, with very high variability between studies, meaning the true size and reliability of exercise's specific pain-relieving effect remains genuinely uncertain.

If you have ongoing musculoskeletal pain, you've probably been told exercise helps. This research looked at whether that benefit comes specifically from the exercise itself, or partly from other things like attention from a therapist, expectation of improvement, or pain naturally easing over time. When researchers compared exercise directly against placebo treatments, drawing on only 4 small trials, there was no clear difference detected between the two, so it's genuinely unclear whether exercise beats a placebo for pain.

However, exercise did show a statistically significant benefit over no treatment at all and over usual medical care, across a much larger set of studies. The researchers rated the overall evidence as very low quality, with results varying a lot from study to study. That means these findings should be read as uncertain rather than a settled answer about how exercise actually works for chronic pain.

This is a summary of research, not medical advice. Talk to your own healthcare provider about anything affecting your care.

This review addresses whether the pain-relieving effect of exercise training in chronic primary musculoskeletal pain reflects a genuine physiological effect or largely placebo and non-specific factors. Based on 79 randomised controlled trials, exercise showed a statistically significant effect versus true no-intervention controls (g=0.99, 95% CI 0.66-1.32, 42 studies) and versus usual care (g=0.64, 95% CI 0.44-0.83, 33 studies), but when compared directly against non-exercise placebo treatments (only 4 studies), no statistically significant difference was detected (g=0.94, 95% CI -0.17 to 2.06).

All estimates were graded very low quality with high heterogeneity (I2 up to 92%). This sharpens how exercise's clinical benefit over no-treatment or usual care should be framed with patients: part of that benefit may reflect attention, expectation, or natural history rather than exercise-specific mechanisms, since the placebo-controlled comparison rests on very few trials and showed no significant difference. The main limitation is the very small and heterogeneous evidence base specifically isolating exercise from placebo, which limits confidence in any claim about exercise's true specific effect.