Exercise Can Improve Pain Save

Know-it-now

• Exercise yields a consistent, moderate short-term analgesic effect, comparable to many analgesic meds, but without their toxicity (SMD −0.59, or about 1.1 points on a 10-point VAS).
• Modality matters less than adherence. Tai chi, Pilates and yoga ranked highest, hence patient preference is important.
• Exercise effects in RA (−0.30) and hip OA (−0.30) are less than in fibromyalgia (−0.70) and low back pain (−0.92).
• Start "low and short" reflects adherence and confounding, not a true dose-response.
• This is a meta-meta-analysis of 157 systematic reviews covering 2,736 RCTs and 221,279 adults.
• The evidence is weaker than the headline. 

This pooled meta-analyses (n=197,392) shows that exercise reduced pain versus control immediately after the intervention (SMD −0.59; 95% CI −0.65 to −0.53; P<0.001). The benefit held across musculoskeletal, inflammatory, neurologic and cancer pain, and for every modality tested:

• Pilates −0.90
• Tai chi −0.85
• Yoga −0.74
• Aerobic −0.61
• Resistance −0.37

Exercise performance in Rheumatology disorders

The inflammatory arthritides sit at the low end of the range. On raw scales, the pooled VAS reduction was −1.10 on a 0–10 scale and −10.89 on a 0–100 scale. Neither WOMAC (−6.70; CI −17.96 to 4.56) nor SF-36 pain reached significance.

Counterintuitive Dose Findings

• Programs shorter than 12 weeks beat longer ones (SMD −0.59 vs −0.37; P=0.03).
• Low intensity beat moderate-to-vigorous intensity (−0.77 vs −0.47; P<0.01).
• Less than 120 min/week outperformed 120 min/week or more (P=0.02).
• Sessions per week (1–2 vs 3 or more) made no difference.

Limitations of this analysis

• Review quality: 147 of 157 reviews were rated low or critically low on AMSTAR-2.
• Publication bias: The funnel plot was asymmetric, and the bias-corrected estimate was −0.43.
• Heterogeneity: I² was 88%, yet GRADE was not downgraded for inconsistency.
• Comparators: Most were waitlist, usual care or no intervention. Blinding was impossible, so expectation and attention effects are built into the estimates.
• Dose subgroups: These are between-review comparisons and are confounded. "Low intensity" is largely tai chi and yoga. The estimate for 120 min/week or more (SMD +0.58; CI −0.27 to 1.42) is implausible.
• Labeling: The authors call −0.59 "large," but by Cohen's conventions it is moderate.
• Durability: There are no data beyond the end of the intervention.

The recommendation is to prescribe exercise to nearly every pain patient, and make the prescription specific and achievable. This study suggests moderate, short-term benefit, with an overall certainty of evidence as moderate, while also acknowledging limitations related to review quality and potential publication bias.. Expect smaller gains in RA and hip OA than in fibromyalgia.