Glucagon-Like Peptide-1 Receptor Agonists and Alcohol Use: A Real-Word Observational Study in a Large, Integrated Health Care System.

BACKGROUND

Growing research suggests that glucagon-like peptide-1 receptor agonists (GLP-1RAs) reduce alcohol consumption, positioning them as new potential pharmacotherapies for alcohol use disorder. However, human studies examining alcohol consumption measures in generalizable samples have been limited.

METHODS

In this cohort study, we analyzed electronic health records from a large, integrated health care system (Kaiser Permanente Northern California) to examine the association between new prescriptions of GLP-1RAs and change in alcohol use among adults. Propensity score matching was utilized to account for differences in baseline characteristics between GLP-1RA-treated (n = 1214) and untreated (n = 1063) individuals. Changes in average drinks consumed per week (drinks/week) from baseline to follow-up (up to 1 year) were compared between groups using difference-in-differences (D-I-D) analysis. Stratified analyses examined treatment effect variation by sex, obesity, and baseline alcohol use risk level.

RESULTS

Both GLP1-RA-treated and untreated groups reduced their drinks/week from baseline to follow-up (mean change [95% CI] = -1.81 [-2.11 to -1.51] and -1.38 [-1.70 to -1.06], respectively); the group difference did not reach statistical significance (D-I-D [95% CI] = -0.43 [-0.87 to 0.01]). Among individuals with low-risk baseline alcohol use, including 1126 (92.8%) GLP-1RA-treated and 996 (93.7%) untreated individuals, receipt of GLP-1RAs was associated with significantly greater reductions in drinks/week (D-I-D [95% CI] = -0.32 [-0.64 to -0.01]). Treatment effects did not vary by sex or obesity.

CONCLUSIONS

GLP-1RAs may be effective in reducing average weekly alcohol consumption, even in individuals with low-risk use. The small subsample of individuals with high-risk use limited our ability to estimate associations in this group.