Systematic review and meta-analysis shows exercise reduces anxiety symptoms in adults with high anxiety or anxiety disorders

OBJECTIVE: This study aimed to evaluate the anxiolytic effects of exercise in individuals with high anxiety or diagnosed anxiety disorders and to identify optimal exercise prescription parameters. A key methodological innovation of this work is the application of a three-level meta-analytic model, which accounts for effect-size dependency arising from multiple outcomes within studies, thereby improving the robustness and precision of the pooled estimates. METHODS: A comprehensive literature search was conducted across five databases: Embase, PubMed, Cochrane Library, Web of Science, and Scopus. INCLUSION CRITERIA: (1) randomized controlled trial (RCT) design; (2) outcomes assessed using validated scales; (3) participants diagnosed with an anxiety disorder or identified as high-anxiety through validated scales; and (4) inclusion of both an intervention group and a control group, with the sole distinction between them being the addition of exercise in the intervention group. Analyses were performed using a three-level restricted maximum likelihood random-effects model. RESULTS: Twenty-seven randomized controlled trials involving 1447 participants were included. Exercise demonstrated a significant anxiolytic effect (Cohen's d = -0.42, p < 0.01). Subgroup analysis indicated that comorbid chronic conditions attenuated treatment efficacy (Cohen's d = -0.29, p < 0.01). Optimal outcomes were observed with exercise programs lasting < 12 weeks, sessions lasting 45-60 min, frequencies of > 3 times per week, and total weekly exercise durations of ≥ 180 min. In contrast, individuals without chronic conditions showed more favorable results. The most beneficial exercise regimens for this group included programs lasting < 12 weeks, sessions lasting > 60 min, frequencies of 3 times per week, and total weekly durations of > 180 min. CONCLUSION: Exercise appears to be an effective non-pharmacological intervention for alleviating anxiety symptoms. While chronic conditions may diminish the magnitude of these benefits, tailored exercise prescriptions remain promising. Nevertheless, given the heterogeneity of interventions and limited long-term follow-up across included studies, further high-quality research is needed to confirm the sustainability and generalizability of these findings. TRIAL REGISTRATION: The protocol was registered in PROSPERO with the registration number CRD420251014897.