- Newer GLP-1 receptor agonists were associated with lower alcohol-related hospitalization risk among adults with alcohol use disorder (AUD) and either type 2 diabetes or obesity.
- Adults who initiated semaglutide or tirzepatide had a 22–32% lower risk of alcohol-related hospitalization than adults who initiated other diabetes or obesity medications.
- The study analyzed real-world data from more than 40,000 adults using a target trial emulation designed to reduce bias and strengthen confidence in the findings.
- Randomized clinical trials are still needed to determine whether GLP-1 medications directly reduce alcohol-related harm
Alcohol use disorder (AUD) is common, chronic, and often undertreated. While medications approved for AUD can be effective, real-world use remains low, and many patients continue to experience alcohol-related harm that leads to emergency department visits or hospitalizations.
A new peer-reviewed study published in BMJ Open evaluated whether newer glucagon-like peptide-1 receptor agonists (GLP-1 RAs)—specifically semaglutide and tirzepatide—were associated with lower risk of alcohol-related hospitalization among adults with AUD and either type 2 diabetes or obesity.
Using a target trial emulation framework, researchers from Truveta, Johns Hopkins Bloomberg School of Public Health, Duke University, Providence, and collaborating institutions analyzed de-identified electronic health record (EHR) data from 40,703 adults across US healthcare systems. The primary outcome was time to first alcohol-related emergency department visit or hospitalization within one year of treatment initiation.
Study snapshot
Researchers conducted a retrospective cohort study using Truveta Data. The study included adults with AUD and either type 2 diabetes or obesity who initiated a newer GLP-1 RA or an active comparator medication between January 1, 2018 and December 31, 2024.
The analysis included four target trial emulations: two evaluated GLP-1 RAs against other diabetes or obesity medications, and two evaluated GLP-1 RAs against medications approved for AUD.
Researchers also evaluated non-alcohol-related hospitalizations as a negative control outcome to assess whether observed associations appeared specific to alcohol-related events.
Newer GLP-1 RAs were associated with lower alcohol-related hospitalization risk
Across four comparisons, adults with AUD who initiated newer GLP-1 RAs had lower alcohol-related hospitalization risk than adults who initiated several comparator medications.
The clearest findings came from comparisons with other diabetes or obesity medications:
- Among adults with AUD and type 2 diabetes, newer GLP-1 RA initiation was associated with a 26% lower risk of alcohol-related hospitalization compared with sulfonylureas and a 22% lower risk compared with other anti-diabetic medications.
- Among adults with AUD and obesity, newer GLP-1 RA initiation was associated with a 32% lower hazard compared with other anti-obesity medications.
Researchers did not find a significant difference between newer and older GLP-1 RAs in the diabetes or obesity comparisons, suggesting the association may reflect a broader GLP-1 RA signal rather than an effect unique to semaglutide or tirzepatide.
The study also compared newer GLP-1 RAs with medications approved for AUD, including naltrexone, disulfiram, and acamprosate. These analyses showed larger associations: a 63% lower risk among adults with type 2 diabetes and a 65% lower risk among adults with obesity. However, these AUD-medication comparisons require more caution because newer GLP-1 RA users also had lower rates of non-alcohol-related hospitalization, suggesting the groups may have differed in ways the analysis could not fully adjust for.
Why the target trial framework matters
A key strength of this study was its target trial emulation design, which helps observational studies more closely mirror the structure of a randomized clinical trial.
Instead of comparing all GLP-1 RA users with all non-users, researchers evaluated new medication initiators within clinically distinct treatment contexts. They also used active comparator groups, propensity score methods, and negative control outcomes to reduce common sources of bias. These methods cannot prove causation, but they make the real-world signal more credible and easier to interpret.
What this study adds to the growing GLP-1 evidence base
This study adds to growing evidence that GLP-1 RAs may be associated with reduced alcohol-related harms. Its contribution is both clinical and methodological: it evaluated a serious outcome—alcohol-related emergency department visits and hospitalizations—across more than 40,000 adults, included newer GLP-1-based therapies, and found a consistent signal across adults with AUD and either type 2 diabetes or obesity.
