HealthProtocols
← All sources

Exercise for depression

Cochrane review (2008) of 28 trials: for exercise versus no treatment or control interventions, pooled SMD across 23 trials (907 participants) was −0.82 (95% CI −1.12 to −0.51); restricting to three trials with adequate allocation concealment, ITT, and blinded outcomes yielded pooled SMD −0.42 (95% CI −0.88 to 0.03), not statistically significant.

Design

  • Cochrane systematic review (Oct 2008 update)
  • Included: 28 RCTs comparing exercise vs standard care, no treatment, or placebo in adults with depression (per trial definitions)

Pooled depression outcomes (abstract)

  • Broad pool (23 trials, n = 907): SMD −0.82 (95% CI −1.12 to −0.51) vs no treatment/control
  • Low-bias subset (3 trials): SMD −0.42 (95% CI −0.88 to 0.03) — not statistically significant
  • Comparators: effect not significantly different from cognitive therapy in available comparisons (limited data)

Evidence hygiene

  • Older Cochrane snapshot—pair with Krogh 2017 and Recchia 2022 network meta-analyses already indexed on related protocols; use this row for historical context on risk-of-bias sensitivity.

Publication

Mead GE, Morley W, Campbell P, et al. Cochrane Database Syst Rev. 2008 Oct 8;(4):CD004366. PMID 18843656.

Outcomes

  • Effect Size (Cohen's d / SMD)
    -0.82
    d (Cohen's d)
  • Effect Size (Cohen's d / SMD)
    -0.42
    d (Cohen's d)
View original paper →