Sinha · Addiction science & clinical practice 2025 · systematic review and meta-analysis · n=9 studies (3 RCTs, N = 430; 6 observational studies, N = 2,740,207)

The effects of glucagon-like peptide-1 receptor agonists (GLP1-RAs) on alcohol-related outcomes: a systematic review and meta-analysis.

Cited 8 times in the scientific literature.

Level 1 - systematic review of randomized trials

Systematic review and meta-analysis including randomized controlled trials

PubMed 41350683 · doi:10.1186/s13722-025-00637-z · record verified 2026-08-26

What was done

A systematic review and meta-analysis (PRISMA-guided, searching PubMed, Embase, and Cochrane Library to May 2025) assessing glucagon-like peptide-1 receptor agonists (GLP-1RAs: semaglutide, liraglutide, exenatide, dulaglutide) versus placebo, no treatment, or active control on alcohol-related outcomes in adults with or without alcohol use disorder (AUD). Outcomes included alcohol consumption (standardised mean difference [SMD]), alcohol craving (SMD), and alcohol-related events (hazard ratio [HR]), synthesized via random-effects models with Restricted Maximum Likelihood and Hartung-Knapp adjustment.

What was found

Included 3 RCTs (N = 430) and 6 observational studies (N = 2,740,207). - In RCTs, GLP-1RAs showed non-significant reductions in alcohol consumption (SMD: -0.24, 95% CI: -0.70 to 0.23), drinks per drinking day (SMD: -0.23, 95% CI: -0.64 to 0.19), and craving (SMD: -0.14, 95% CI: -2.84 to 2.55; semaglutide showed greater craving reduction, p = 0.024). - In observational studies, GLP-1RAs were associated with reduced alcohol-related events (HR: 0.64, 95% CI: 0.59 to 0.69, p < 0.001), including in AUD cohorts (HR: 0.66, 95% CI: 0.63 to 0.70), SUD cohorts (HR: 0.66, 95% CI: 0.18 to 2.48), and intoxication (HR: 0.50).

Why it matters

Real-world data strongly link GLP-1RA therapy to fewer severe alcohol-related outcomes, but current randomized trial evidence is too small to confirm direct reductions in drinking or craving.

Limits

Randomized evidence is limited to only 3 small trials (N = 430 total) with very wide confidence intervals. Significant event reductions rely entirely on observational databases subject to residual confounding, prescription bias, and indirect outcome definitions.

Cited by