Resistance training for depression: a systematic review and meta-analysis of randomized controlled trials.

Introduction: Depression is a prevalent and disabling mental disorder worldwide. Resistance training (RT) has emerged as a promising adjunct intervention, but comprehensive quantitative synthesis on its efficacy and optimal exercise prescription remains limited. Objective: To evaluate the effects of RT on depressive symptoms in adults with a clinically diagnosed depressive disorder, and to explore-exploratorily-whether participant characteristics and prescription components modify outcomes. Methods: We searched PubMed, Embase, Web of Science, Cochrane Central, and CNKI from inception through August 2024 for randomized controlled trials (RCTs) comparing RT to a non-exercise control in adults with depression (PROSPERO CRD42024583413). Two reviewers independently screened studies and extracted data in accordance with PRISMA 2020. Depression outcomes were pooled as standardized mean differences (SMDs) with 95% confidence intervals (CIs) using a random-effects model, with unit-of-analysis safeguards for multi-arm trials. Pre-specified exploratory analyses evaluated potential effect modifiers (e.g., clinical phenotype [primary vs. comorbid], training frequency, age, baseline severity, duration, intensity, weekly volume). Risk of bias was assessed by two independent reviewers using the Cochrane tool; publication bias was evaluated by funnel plot and Begg's test, noting limited power with this study count. Sensitivity analyses included exclusion of high-risk studies and leave-one-out influence checks to test robustness. Results: Twenty-nine RCTs (N = 2,036) met inclusion criteria. RT significantly reduced depressive symptoms compared to controls (pooled SMD = -0.94, 95% CI: -1.16 to -0.72, p < 0.001), though heterogeneity was high (I 2 ≈ 80%). Benefits were observed in both primary depressive disorder (SMD - 1.12, 95% CI - 1.43 to -0.81) and comorbid depression (SMD - 0.66, -0.96 to -0.36), with a modest between-subgroup contrast (Q_between = 4.41, p = 0.036). Effects were directionally consistent across self-report and observer-rated measures and across frequency strata (<3 vs. ≥ 3 sessions/week), with no compelling between-subgroup differences; beyond these key strata, exploratory subgroup analyses across age, baseline severity, duration, intensity, and weekly volume likewise did not reveal consistent between-group differences, and estimates were imprecise in small strata. Sensitivity analyses-excluding high-risk studies and via leave-one-out influence checks-yielded estimates of similar magnitude. The funnel plot appeared broadly symmetric and Begg's test was non-significant, while acknowledging limited power with this study count. Conclusion: RT meaningfully reduces depressive symptoms in adults with clinically diagnosed depression. Given substantial heterogeneity and measurement (self-report vs. observer-rated) and clinical (primary vs. comorbid) variability, any apparent effect modifiers are interpreted cautiously and considered exploratory/hypothesis-generating. To improve precision and implementation, future trials should standardize supervision/adherence reporting (e.g., TIDieR/CERT) and include preregistered follow-ups (3-12 months) to assess durability, while training-prescription guidance remains preliminary pending better-reported, preregistered studies. Systematic review registration: Unique Identifier: CRD42024583413, https://www.crd.york.ac.uk/PROSPERO/view/CRD42024583413.