GLP-1 Receptor Agonists vs Alternatives for Alcohol Use Disorder: A Multi-Target Trial Emulation

BackgroundGlucagon-like peptide 1 receptor agonists (GLP-1 RAs) have shown promise for alcohol use disorder (AUD). ObjectiveTo evaluate the association between use of newer GLP-1 RAs (semaglutide, tirzepatide) and alcohol-related hospitalizations among adults with AUD and either type 2 diabetes (T2D) or obesity. MethodsThis retrospective target trial emulation study used electronic health record data from Truveta to identify adults with AUD and either T2D or obesity, who initiated a newer GLP-1 RA (semaglutide, tirzepatide) or relevant active comparator between 2018 and 2024. Four target trials were constructed to reflect clinically distinct populations and comparators: (1) ADM trial (patients with T2D and comparators of other anti-diabetic medications [ADM]), (2) AOM trial (patients with obesity but not T2D and comparators of other anti-obesity medications [AOM]), (3) MAUD-T2D trial (patients with T2D and markers of more severe AUD and comparators of medications for alcohol use disorder [MAUD]), (4) MAUD-obesity trial (patients with obesity, no T2D, and markers of more severe AUD and comparators of MAUD). The primary endpoint was time to alcohol-related hospitalization. Non-alcohol-related hospitalization served as a negative control outcome. Propensity score-based methods (weighting and matching) were used to control for confounding. Cox proportional hazards models were used to estimate the treatment effect of newer GLP-1 RA in four target trials. ResultsA total of 40,260 patients were identified, including 18,515 in the ADM trial, 9,256 in the AOM trial, 9,975 in the MAUD and T2D trial, and 11,039 in the MAUD and obesity trial. GLP-1 RAs were associated with a lower hazard of alcohol-related hospitalization in the ADM (HR [95% CI]: 0.70 [0.59 - 0.83] vs. sulfonylureas; 0.73 [0.62 - 0.86] vs. other ADMs), AOM (HR: 0.59 [0.48 - 0.74]), MAUD-T2D (HR: 0.36 [0.29 - 0.46]), and MAUD-obesity (HR: 0.32 [0.23 - 0.43]) trials. No significant differences were observed for non-alcohol-related hospitalizations for ADM and MAUD-T2D trials. ConclusionNewer GLP-1 RAs were associated with reduced risk of alcohol-related hospitalization across clinically distinct populations.