(A-087) Effect of Glucagon-Like Peptide-1 Receptor Agonists (GLP-1 RAs) in Alcohol Use Disorder: A Retrospective Cohort Study Using Electronic Health Records

Background: Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) such as semaglutide and tirzepatide may have beneficial effects on alcohol use disorder (AUD) by reducing alcohol’s stimulatory and dopamine-inducing properties.

Objectives: To determine the association between GLP-1 RAs and AUD-related emergency department (ED) visits and hospitalizations.

Methods: This retrospective, new user cohort study used Truveta Data, which contains electronic health records for 30 US healthcare systems. We included individuals with encounters from January 2020 to February 6, 2025 and a history of usual care and created two cohorts: adults with AUD and type 2 diabetes mellitus (T2DM) (T2DM cohort) and adults with AUD and obesity without T2DM (obesity cohort). In both cohorts, we compared the effects of incident use of newer GLP-1 RAs (semaglutide and tirzepatide) to older GLP-1 RAs (dulaglutide, exenatide, and liraglutide) and AUD medications (acamprosate, disulfiram, and naltrexone). The primary outcome was an ED visit or hospitalization with AUD. We employed an on-treatment approach, following patients from the first medication initiation, until the outcome of interest, 60 days following their last drug dispensing or administration, or the last encounter. We used a Cox regression model to estimate the association of newer GLP-1 RA use with AUD outcomes, controlling for age, sex, race, history of AUD hospitalization in the previous year, duration of AUD, and calendar year.

Results: The study included 29,465 individuals, with 14,193 in the T2DM cohort (new GLP-1 RA: 6,263, older GLP-1 RA: 4,204, AUD: 3,726) and 15,272 in the obesity cohort (new GLP-1 RA: 3,336, older GLP-1 RA: 756, AUD: 11,180). The time to an AUD-associated ED visit or hospitalization significantly differed across the three exposure groups in both cohorts (logrank p value < 0.01). Semaglutide or tirzepatide use was associated with a lower risk of AUD events compared to AUD medication use in both the T2DM cohort (adjusted hazard ratio [aHR] 0.23, 95% confidence intervals [CI] 0.18-0.28) and the obesity cohort (aHR 0.21, CI 0.15-0.28). Compared to older GLP-1 RA, semaglutide or tirzepatide use was significantly associated with a lower risk of AUD events in the T2DM cohort (aHR 0.73, CI 0.58-0.92), but not in the obesity cohort (aHR 0.75, CI 0.39-1.42).

Conclusions: In a large, diverse cohort of individuals with AUD, the use of semaglutide or tirzepatide was associated with a reduced risk of an AUD event compared to AUD medication use. Our analysis adjusted for only a partial list of confounders. Future analyses should include additional control for confounding and selection bias.