Patients with diabetes on GLP-1 drugs landed in the hospital for drinking about half as often
STUDY: Tang H et al, Diabetes, Obesity and Metabolism 2026
STUDY TYPE: Cohort study
FUNDING: American Heart Association; National Institutes of Health
Background
GLP-1 drugs treat diabetes and obesity, but they also affect the brain’s reward circuits. They reduce alcohol use in animals, but so far the only clear signal from a well-designed trial has been for people with obesity and alcohol use disorder. This new study is not randomized, so only gives a hint of the effects.
The Study
- About 300,000 US adults with type 2 diabetes and no prior alcohol use disorder.
- New users of semaglutide or tirzepatide were matched to new users of SGLT2 inhibitors and to each other.
- Mean age was about 60; follow-up averaged one year.
- Main outcome: a new hospitalization or emergency visit for alcohol use disorder.
Results
Visits for alcohol problems were rare, and less common on GLP-1 drugs:
- Semaglutide: 2.4 vs 4.5 per 1,000 patient-years on SGLT2 inhibitors (HR 0.54)
- Tirzepatide: 2.3 vs 4.0 per 1,000 patient-years (HR 0.61)
- Semaglutide vs tirzepatide: about the same (HR 0.98)
Those differences were small. To prevent one alcohol-related hospital visit, around 475 patients would need a year on semaglutide. Turning to prevention of new alcohol use, we see a bigger shift.
New alcohol use disorder diagnoses were about 30% less common on both drugs, and alcohol-related liver disease about 45% less common. Results held across subgroups, including depression and other substance use.
Limitations
Observational, not randomized, all had diabetes and simply treating that may reduce alcohol problems. Using another diabetes drug as the comparison reduces bias. Still, the records couldn’t measure how much anyone drank, and one author consults for Lilly, which makes tirzepatide.
Practice Implications
- This is a hint, not a confirmation.
—Chris Aiken, MD
Director, Psych Partners
Editor in Chief, Carlat Psychiatry Report
