Why is exercise effective in reducing pain in people with osteoarthritis?

Key points

1. 01Exercise effect sizes for knee OA pain (ES 0.49) and function (ES 0.52) exceed those of NSAIDs (ES 0.29 for pain, 0.35 for function) based on meta-analyses of RCTs
2. 02At least 12 supervised exercise sessions appear superior to fewer sessions for pain relief in knee OA
3. 03Aerobic, strengthening, and neuromuscular exercise types are all effective; neuromuscular training may be superior to quadriceps strengthening alone in non-obese people with knee OA and varus thrust
4. 04Proposed mechanisms include endogenous analgesia (opioids, endocannabinoids, serotonin), improved muscle strength and neuromuscular control, reduced local joint inflammation, and effects on central pain processing
5. 05Evidence-based programs (GLA:D, BOA, ESCAPE Pain) implemented in multiple countries show real-world pain reductions of 33% on average with improved function and cost-effectiveness

How it was conducted

Design

Narrative review of translational and clinical literature, from mechanistic studies to real-world implementation registries

Focus

Hip and knee osteoarthritis; generalized OA and other individual joints are not covered due to limited data

Evidence pipeline

Translational studies on pain neuroscience, muscle, cartilage, and inflammation; systematic reviews and meta-analyses of RCTs; and population-level registry data

Comparators

Exercise types (aerobic, strengthening, neuromuscular) compared with each other and with pharmacological treatments (NSAIDs, paracetamol)

Implementation programs reviewed

GLA:D (Denmark, Canada, Australia, China, Switzerland, New Zealand), BOA (Sweden), ESCAPE Pain (UK)

What they found

• Exercise effect size for pain in symptomatic knee OA: ES 0.49; for function: ES 0.52 (Cochrane review)
• Exercise effect size for both pain and function in symptomatic hip OA: ES 0.46 (Cochrane review)
• NSAIDs effect size for knee OA: ES 0.29 for pain and 0.35 for function; hip OA: ES 0.26 for pain and 0.17 for function (meta-analysis of RCTs)
• Exercise resulted in mean absolute pain improvement of 14 points (95% CI 11-17) and function improvement of 11 points (95% CI 8-15) in knee OA (validated 0-100 scale)
• Exercise reduced pain by an average of 12 points (95% CI 10-14) and function by 10 points (95% CI 8-11) in hip OA
• Muscle strengthening exercises increased muscle mass (SMD 1.4; 95% CI 0.9-1.7) and cross-sectional area (SMD 1.3; 95% CI 0.8-1.3) in older adults with knee OA
• Resistance training increased quadriceps strength by 7.9% (ES 1.3)
• GLA:D registry data: within-group pain intensity reduced 33% on average with 13% increase in walking speed and 21% increase in number of chair stands
• More than 100 randomized trials show positive effects of exercise on pain and function in hip or knee OA
• Serum COMP returned to normal within 30 minutes of a 30-minute exercise session and remained at normal levels 6 weeks later
• IL-6 decreased modestly (-1.05 pg/mL) in knee OA patients performing aerobic exercise versus non-exercise control, but the effect was not statistically significant

Limitations

• Narrative rather than systematic review design; no formal quality assessment or meta-regression of primary studies
• Most mechanistic evidence is from knee OA; findings may not generalize to hip OA or other joints
• Studies of exercise effects on cartilage biomarkers and systemic inflammation are underpowered and heterogeneous, limiting causal conclusions
• The optimal exercise dose and type for specific OA subgroups remains unclear due to inconsistent outcomes across trials

Why it matters

For patients

Exercise, including group neuromuscular programs, is safe and reduces pain at least as much as common pain medications, and improvements can last a year or more if home exercise is continued.

For clinicians

Prescribing at least 12 supervised exercise sessions combined with patient education is supported by strong evidence and should be first-line treatment before or alongside pharmacotherapy for hip and knee OA.

For readers

This review synthesizes the biological reasons exercise works (neural, muscular, cartilage, and anti-inflammatory pathways) and shows that structured community programs in multiple countries achieve the same benefits seen in clinical trials at acceptable cost.