Effectiveness of physical activity interventions for improving depression, anxiety and distress: an overview of systematic reviews

Mental health disorders are a leading source of global disease burden, affecting around one in eight people worldwide and incurring substantial economic costs. Depression is the leading contributor to mental health-related disease burden and anxiety is the most prevalent disorder. The COVID-19 pandemic further increased psychological distress globally. Clinical guidelines vary in the positioning of lifestyle interventions such as physical activity relative to psychotherapy and pharmacotherapy, and despite growing evidence that physical activity can yield comparable benefits with added advantages (lower cost, fewer side-effects, broader health gains), its therapeutic adoption remains limited. Barriers include patient resistance, challenges prescribing and monitoring physical activity, and the large, heterogeneous evidence base. This umbrella review aims to comprehensively synthesise the effects of all modes of physical activity on symptoms of depression, anxiety and psychological distress among adults.

Hundreds of trials and several meta-reviews have examined physical activity for mental health outcomes. Prior meta-reviews often focused on specific populations, single conditions (e.g., depression only), or particular modes of physical activity, limiting generalisability. Existing evidence suggests physical activity may have effects similar to psychotherapy and pharmacotherapy for mild-to-moderate depression and anxiety, and observational studies indicate protective effects of physical activity against incident depression and anxiety. This review addresses gaps by providing a broad synthesis across all adult populations and physical activity modalities.

Design: Systematic umbrella review (meta-review) of systematic reviews with meta-analyses of randomized controlled trials (RCTs). Protocol: Prospectively registered on PROSPERO (CRD42021292710); reported according to PRISMA guidelines. Inclusion criteria (PICOS): Population: adults aged ≥18 years across any clinical or general population. Intervention: interventions designed to increase physical activity, defined as any bodily movement produced by skeletal muscles resulting in a substantial increase in energy expenditure; any modality, supervision, delivery method, or dose permitted. Comparator: ≥75% of component RCTs had usual care, waitlist, no intervention, equal-attention intervention, or a lower/lesser physical activity intervention; reviews were excluded if >25% of component RCTs compared physical activity to pharmaceuticals or compared equal-dose exercise types without a non-physical activity comparator. Outcomes: self-report or clinician-rated symptoms of depression, anxiety, or psychological distress. Study type: systematic reviews with meta-analyses of RCTs that included meta-analysis of the outcomes of interest. Search strategy: Twelve databases (CINAHL, Cochrane, Embase, MEDLINE, Emcare, ProQuest Health and Medical Complete, ProQuest Nursing and Allied Health Source, PsycINFO, Scopus, Sport Discus, EBSCOhost, Web of Science) from inception to 1 January 2022; English-language, peer-reviewed articles. Study management: Results deduplicated in EndNote and screened in Covidence. Title/abstract and full-text screening, data extraction, and risk of bias assessment were conducted in duplicate by independent reviewers, with disagreements resolved by discussion. Risk of bias: AMSTAR-2 tool (16 items; seven critical domains) used to rate review confidence as high, moderate, low, or critically low. Synthesis: Assessed overlap in component RCTs using Corrected Covered Area (CCA) with categories: slight (0–5%), moderate (6–10%), high (11–15%), very high (>15%). Publication bias assessed with funnel plots. Meta-analysis results extracted as standardized mean differences (SMD) and mean differences (MD), summarized via medians and IQRs, and presented in forest plots. Subgroup analyses examined clinical status and intervention characteristics (exercise intensity, duration of intervention, weekly duration, session frequency, session duration, and mode). Overall evidence graded using the Oxford Centre for Evidence-Based Medicine levels and grades (A–D).

Scope: 97 systematic reviews included, comprising 1039 unique RCTs and 128,119 participants; mean ages ranged 29–86 years (median 55). Overlap: CCA = 0.6% (slight overlap). Quality: Most reviews (n=77) rated critically low by AMSTAR-2; 10 low; 10 high. Publication bias: Funnel plots showed no evidence of publication bias. Effects on depression: 72 meta-analyses (n=875 component RCTs; >62,040 participants) showed a medium effect favoring physical activity (median SMD = −0.43; IQR = −0.66 to −0.27). Instrument-specific MDs included Beck Depression Inventory: −5.53 (IQR −6.24 to −4.81), Edinburgh Postnatal Depression Scale: −2.97 (IQR −3.49 to −2.44), Hospital Anxiety and Depression Scale: −1.26 (IQR −1.41 to −1.18), among others. Oxford grade: A (consistent level 1 evidence). Effects on anxiety: 28 meta-analyses (171 component RCTs; >10,952 participants) showed a medium effect (median SMD = −0.42; IQR = −0.66 to −0.26). Instrument-specific MDs included State-Trait Anxiety Inventory: −3.61 (IQR −6.01 to −1.66), Hospital Anxiety and Depression Scale: −1.26 (IQR −1.26 to −0.79). Oxford grade: A. Effects on psychological distress: One review reporting SMD (6 RCTs; 508 participants) showed medium effect (SMD = −0.60; 95% CI −0.78 to −0.42). One review reporting MD (1 RCT; 39 participants) showed no significant effect (MD = −0.30; 95% CI −5.55 to 4.95). Oxford grade: B. Subgroups by clinical condition: Depression benefits were broadly observed; largest median SMDs in kidney disease (−0.85), HIV/AIDS (−0.84), COPD (−0.72), general population adults (−0.62 to −0.69), and diagnosed depression (median −0.62; diagnosed subgroup −0.66). Anxiety benefits ranged from HIV (−1.23) and general population (−0.85) to multiple sclerosis (−0.16). Exercise intensity: For depression, median SMDs were −0.22 (low), −0.56 (moderate), −0.70 (high). For anxiety, low −0.26; moderate −0.47; high −0.44 (IQR −0.49 to −0.13). Intervention duration: For depression, short (≤12 weeks) −0.84 (IQR −1.50 to −0.48), medium (12–23 weeks) −0.46 (IQR −0.53 to −0.25), long (≥24 weeks) −0.28 (IQR −1.15 to −0.17). For anxiety, short −0.55 (IQR −0.83 to −0.27), medium −0.47 (IQR −0.72 to −0.08), long −0.15. Weekly duration: For depression, ≤150 min/week −0.58 (IQR −0.77 to −0.30) vs >150 min/week −0.29 (IQR −0.40 to −0.07). For anxiety, <150 min/week −1.23 vs ≥150 min/week −0.99. Session frequency: For depression, high (5–7/week) −0.76 (IQR −1.20 to −0.32), moderate (4–5/week) −1.12 (IQR −1.39 to −0.85), low (<4/week) −0.47 (IQR −0.59 to −0.35). For anxiety, 2–3/week −0.50; 4–5/week −0.96; 6–7/week −0.52. Session duration: For depression, ≥60 min −0.57 (IQR −0.85 to −0.35), 30–60 min −0.60 (IQR −0.78 to −0.41), <30 min 0.01. Exercise mode: For depression, strength −0.64 (IQR −0.86 to −0.19), mixed −0.47 (IQR −0.64 to −0.29), mind-body (e.g., yoga) −0.46 (IQR −0.77 to −0.33), aerobic −0.45 (IQR −0.79 to −0.37). For anxiety, strength −0.23 (IQR −0.37 to −0.08), mixed −0.35 (IQR −0.86 to −0.23), mind-body −0.42 (IQR −0.78 to −0.16), aerobic −0.29 (IQR −0.54 to −0.16). Overall, higher intensity and shorter interventions tended to yield larger improvements.

The umbrella review demonstrates that physical activity interventions produce consistent, medium-sized reductions in depressive and anxiety symptoms across diverse adult populations, addressing the research aim of comprehensively synthesising the evidence base. Benefits were observed in clinical and general populations, with especially large effects in diagnosed depression, kidney disease, HIV, COPD, and pregnant/postpartum women. All modes of physical activity were beneficial, with resistance modes tending to have stronger effects on depression and mind-body modalities (e.g., yoga) particularly effective for anxiety. Greater intensities were associated with larger benefits, plausibly due to stronger neurobiological and psychosocial mechanisms (e.g., increased neurotrophic factors, monoamine availability, HPA axis regulation, reduced inflammation). Smaller effects for longer-duration interventions may reflect adherence decline, diminishing expectancy effects over time, or insufficient progression of dose; shorter weekly durations may be more feasible and still effective, offering pragmatic guidance for service delivery. The findings support public health recommendations for multimodal, moderate-to-vigorous physical activity and suggest that physical activity should be integrated as a mainstay treatment option for depression, anxiety, and distress.

Physical activity is effective in improving symptoms of depression, anxiety, and psychological distress across a wide range of adult populations and clinical conditions. All modes of physical activity are beneficial, with moderate-to-high intensity generally yielding greater improvements. Shorter-duration interventions can achieve meaningful benefits, highlighting practical implications for health services. Given effect sizes comparable to or exceeding those of psychotherapy and pharmacotherapy reported in prior literature, physical activity should be central to the management of mild-to-moderate depression and anxiety. Future research should directly compare the relative and combined effectiveness of physical activity with psychotherapy and pharmacotherapy, optimize intervention dose and progression over longer durations, and explore strategies to enhance adherence and sustainability.

The evidence base was more extensive for mild-to-moderate depression than for anxiety and psychological distress, limiting the strength of subgroup conclusions for the latter outcomes. Most included reviews (n=77) were rated as critically low quality by AMSTAR-2, which may affect confidence in detailed subgroup estimates. Potential issues inherent to exercise trials, such as lack of participant blinding and adherence challenges, may influence observed effects, particularly over longer interventions.