OBJECTIVES: To systematically evaluate the effect of exercise interventions on mental health outcomes (encompassing anxiety, depression and stress) and alcohol consumption (encompassing daily or weekly standardised alcohol consumption) in individuals with alcohol use disorder (AUD).

DESIGN: We performed a systematic review and meta-analysis of eligible randomised controlled trials (RCTs).

DATA SOURCES: Web of Science, PubMed and the Cochrane Library were searched from database inception to February 2025 using both manual and software-assisted methods (NoteExpress).

ELIGIBILITY CRITERIA: RCTs examining the effects of exercise interventions on individuals diagnosed with AUD were included. Studies were required to report outcomes related to mental health (anxiety, depression and stress) and/or alcohol consumption. Studies without a control group or a before-after intervention design lacking a comparison were excluded. Language was restricted to English.

DATA EXTRACTION AND SYNTHESIS: Two researchers independently extracted data. A random-effects model was used to synthesise results, reported as standardised mean differences (SMDs) with 95% CIs. Risk of bias, heterogeneity (I²) and publication bias were assessed using Stata version 14. Sensitivity and subgroup analyses were conducted to evaluate robustness and explore sources of heterogeneity.

RESULTS: A total of 15 studies met the inclusion criteria and were included in this systematic review and meta-analysis. Exercise interventions significantly improved mental health outcomes in individuals with AUD, reducing self-reported symptoms of anxiety (SMD -0.53, 95% CI -0.87 to -0.18, p=0.003), depression (SMD -0.84, 95% CI -1.20 to -0.48, p<0.001) and stress (SMD -1.14, 95% CI -1.91 to -0.3, p=0.004). However, no significant reductions were observed in daily or weekly alcohol consumption, although a non-significant decreasing trend was noted, particularly in European populations. Heterogeneity was moderate to high across anxiety (I² = 64.5%), depression (I² = 75.6%), stress (I² = 81.8%) and daily alcohol use (I² = 55.8%). Subgroup analyses revealed that short-term interventions (<14 weeks) were more effective for anxiety and depression. Yoga demonstrated superior effects on depression compared with aerobic exercise, particularly when session durations were <60 min.

CONCLUSIONS: Exercise-based interventions are effective in improving mental health in individuals with AUD, particularly in alleviating anxiety, depression and stress. Although effects on alcohol consumption were not statistically significant, consistent trends suggest potential benefits, influenced by geography, intervention duration and exercise modality. Tailored exercise programmes may enhance AUD treatment outcomes.

PROSPERO REGISTRATION NUMBER: CRD420251122437.