The effect of exercise on clinical pain: a systematic umbrella review and meta-meta-analysis

Exercise is recommended for managing pain, yet the consistency, magnitude, and certainty of effects across different pain conditions and exercises remain unclear.

This umbrella review aimed to synthesize the best available evidence on the analgesic effects of exercise by examining systematic reviews and meta-analyses of randomized controlled trials (RCTs).

Eleven databases were systematically searched from inception to August 2024. Eligible studies included systematic reviews with meta-analyses of RCTs comparing exercise to control conditions, with pain as a primary or secondary outcome. Reviews without meta-analyses, those not involving RCTs, or those primarily focused on experimentally induced or laboratory pain were excluded. Two reviewers independently extracted data and assessed methodological quality using AMSTAR-2. Standardized mean differences in pain were synthesized using random-effects meta-meta-analysis. Certainty of evidence was evaluated using GRADE, with subgroup and sensitivity analyses.

A total of 157 systematic reviews comprising 2,736 RCTs and 221,279 participants were included. Exercise significantly reduced pain compared with controls (pooled standardized mean differences = -0.59; 95% CI, -0.65 to -0.53; P < 0.001). Effects were observed across both chronic and acute pain conditions, encompassing musculoskeletal, neurological, inflammatory, and cancer populations. Aerobic, resistance, yoga, Pilates, and tai chi were effective. Greater effects were observed in lower-intensity, shorter-duration (<12 weeks) programs. Sensitivity analyses supported the robustness of findings, and the overall GRADE certainty was moderate.

This umbrella review provides compelling evidence that exercise offers consistent and clinically meaningful pain relief across diverse populations and conditions. Findings support the integration of even low-intensity, short-term exercise into routine pain management strategies.