What does the research actually show?
Four studies dominate the current evidence on semaglutide and drinking: one randomized trial in adults with alcohol use disorder and obesity, one large observational veteran cohort, and two small early-phase trials — one of which missed its primary endpoint. Every effect below is reported against its comparator, with sample size and limitations.
| Study | Design & sample | Findings vs comparator | Limitation |
|---|---|---|---|
| Klausen MK, et al. The Lancet, May 2, 2026. doi:10.1016/S0140-6736(26)00305-3 | Randomized, double-blind, 26 weeks, semaglutide 2.4 mg weekly vs placebo; every participant in every arm also received cognitive behavioural therapy. 108 treatment-seeking adults with alcohol use disorder and comorbid obesity (BMI 30+). 81% completed. | Heavy drinking days fell 41.1 percentage points from baseline versus 26.4 percentage points on placebo — a 13.7-percentage-point difference versus placebo (95% CI −22.0 to −5.4; P=0.0015). Weight loss 11.2 kg versus 2.2 kg on placebo. Number needed to treat 4.3. | Single-centre, obesity-only cohort, and not tested as a standalone therapy since every arm received CBT. |
| Cai M, Choi T, Xie Y, Al-Aly Z. BMJ. 2026;392:e086886. doi:10.1136/bmj-2025-086886 (PMID 41781010) | Observational cohort study of health records. 606,434 US veterans with type 2 diabetes. | GLP-1 use was associated with a 14% lower risk of developing any substance use disorder, and 18% lower risk for alcohol specifically, compared with non-users. | Observational — association, not causation. Cohort was largely older and male. VA-funded. |
| Schacht JP, Sakai JT, Raymond K, Shelton R. American Journal of Psychiatry, July 29, 2026. doi:10.1176/appi.ajp.20260003 | Phase 2 randomized trial, oral semaglutide, 8 weeks. 50 adults. | This trial MISSED ITS PRIMARY ENDPOINT: a laboratory measure of cue-elicited craving showed no significant difference from placebo. It also missed one of two key secondary endpoints. It did reduce heavy drinking days, drinks per drinking day, and real-world craving compared with placebo. | Primary endpoint not met, so the trial does not establish an effect on cue-elicited craving. Small and short. |
| Hendershot CS, Bremmer MP, Paladino MB, et al. JAMA Psychiatry. 2025;82(4):395-405. | Randomized trial, low-dose semaglutide, 9 weeks. 48 adults. | Reduced laboratory alcohol consumption and weekly craving compared with placebo. | Very small and very short. |
What is actually approved today
- Semaglutide is NOT FDA-approved to treat alcohol use disorder. It is approved for type 2 diabetes and chronic weight management only.
- There has been no newly FDA-approved medication for alcohol use disorder since 2006. Three medications are approved: naltrexone, acamprosate, and disulfiram.
- A Phase 3 trial exists precisely because the answer is not yet known. Randomized placebo-controlled trials require clinical equipoise — documented genuine uncertainty about which arm is better.
Sources
- ClinicalTrials.gov, study record NCT07218354
- National Institute on Alcohol Abuse and Alcoholism, alcohol use disorder overview
- Klausen MK, et al. The Lancet, May 2, 2026 (doi:10.1016/S0140-6736(26)00305-3)
- Cai M, et al. BMJ 2026;392:e086886 (doi:10.1136/bmj-2025-086886)
- Schacht JP, et al. American Journal of Psychiatry, July 29, 2026 (doi:10.1176/appi.ajp.20260003)
- Hendershot CS, et al. JAMA Psychiatry. 2025;82(4):395-405 (doi:10.1001/jamapsychiatry.2024.4789)