Does semaglutide reduce drinking?
Early randomized trials suggest semaglutide may affect some alcohol-related outcomes, but results are mixed and do not establish a personal alcohol-treatment plan. A small 2025 injection trial found reductions in laboratory alcohol consumption and some drinking measures. A 2026 oral trial did not show a significant difference on its primary laboratory craving outcome, although a secondary heavy-drinking measure improved.[1] [2]
That is more informative than either “it works for alcohol” or “it does nothing.” The answer depends on the population, formulation, outcome and observation period. If drinking is a concern, discuss it directly with a qualified clinician rather than using a weight-treatment prescription as an experiment or changing the prescription to reproduce a trial.
This article is about interpreting research. It does not recommend semaglutide for alcohol use disorder, set a safe amount to drink or tell you to start or stop drinking while taking a medicine. Those decisions require an assessment of your situation, current treatment and the support available to you.
Read the 2025 injection study carefully
The 2025 phase 2 trial randomized 48 adults with alcohol use disorder who were not seeking alcohol treatment. Over nine weeks, semaglutide reduced alcohol consumed in a laboratory task, drinks per drinking day and weekly craving. Average drinks per calendar day and the number of drinking days did not significantly change. The small sample and short duration limit what can be concluded about lasting clinical benefit.[1]
A laboratory result and an everyday outcome are related questions, but they are not interchangeable. A controlled task helps researchers measure a response under specified conditions. It does not by itself establish how someone will function at home, sustain change over years or respond while receiving a different package of care.
The participant group matters too. People who are not seeking treatment may differ from people actively asking for help. A result in one group can justify further research without automatically predicting the other. This is one reason to read the later trial on its own terms rather than assuming that a similar medicine name means the studies should have identical findings.
For a broader view of trial interpretation, see semaglutide and weight plateaus.
What the newer oral study adds
The 2026 phase 2 trial included 50 treatment-seeking adults with moderate or severe alcohol use disorder over eight weeks. Its primary week-six cue-induced craving outcome was not significantly different between groups; drinks per day also did not differ significantly. Heavy-drinking days, a secondary outcome, declined more with semaglutide in the last four weeks. Three semaglutide participants withdrew, including one with a probably treatment-related allergic reaction.[2]
| The question | A useful next step |
|---|---|
| 2025 injection trial | Small non-treatment-seeking sample; laboratory consumption and selected drinking measures improved. |
| 2026 oral trial | Treatment-seeking sample; primary craving endpoint was not significant, with a secondary heavy-drinking signal. |
| Both studies | Short phase 2 trials; neither is a long-term comparison of routine alcohol-treatment services. |
A statistically significant secondary outcome should not be hidden, but it should not replace a negative primary outcome in the headline. Likewise, a primary result without a significant difference does not prove that every possible benefit is absent. The useful reading preserves both findings and asks what a larger, longer study would need to establish.
We have not used a dramatic responder percentage as a personal forecast. Small trials can be sensitive to missing observations and the definition of response. A decision about care should not rest on the most eye-catching number in an abstract while leaving out withdrawals, the main endpoint or the duration of follow-up.
A research question is not a reason to postpone help
NIAAA describes alcohol use disorder as a medical condition and outlines care that can include behavioral treatment, medications and mutual-support approaches. A clinician can help assess the problem and appropriate options. You do not need to wait for a newer trial or a weight-treatment appointment before asking for help with drinking.[3]
Tell the clinician what you want help with in straightforward language: drinking more than intended, difficulty cutting down, concern about interactions with medicines or uncertainty about where to seek care. You do not have to supply your own diagnosis. Ask which service is responsible for assessment and how follow-up will be arranged.
If you already use an online weight service, do not assume that its messaging team provides specialist alcohol treatment. Ask about its scope and referral process. A clear boundary can be useful: it tells you when another professional is needed and avoids letting an administrative enrollment route become a substitute for the right clinical conversation.
Keep three decisions separate
Do not turn the finding into a withdrawal plan
NIAAA warns that abruptly stopping after prolonged heavy drinking can cause potentially life-threatening withdrawal and advises medical help in planning recovery. A semaglutide trial does not provide a home withdrawal protocol. If you are concerned about withdrawal, seek medical advice; severe symptoms require urgent medical care.[4]
That warning is different from telling every reader that they have dependence or predicting what will happen. It identifies a reason that a general article cannot safely supply a universal instruction to stop. The right next step depends on assessment, not on matching your experience to an online anecdote.
Do not copy a study dose, add another medicine or change a prescription on your own. The trials had selection criteria and clinical oversight that a summary cannot reproduce. If a medicine is already prescribed, ask the prescribing clinician how your alcohol-related concerns fit the existing plan and whether another professional should be involved.
Use the evidence to ask a better question
A useful appointment question is: “I have seen research about semaglutide and drinking. What does it mean for my situation, and what established care should we consider now?” That invites a discussion of the evidence without making the answer depend on obtaining one particular prescription. Ask how benefit, safety and follow-up would be assessed for any option discussed.
Our question builder can organize general contact and follow-up questions without collecting alcohol-use details. Keep those details with the clinical team. The provider checklist helps clarify scope, records and referral arrangements. It cannot screen for alcohol use disorder or select a treatment.
The Ro review and DirectMeds review concern weight-care services; neither is a recommendation for alcohol treatment. Our GLP-1 comparison guide likewise compares documented care arrangements rather than addiction-treatment capability. For related reading about interpreting outcomes over time, see the semaglutide plateau explainer. The emerging evidence is a reason for careful research and an informed conversation, not a shortcut around appropriate care.
Sources
- Hendershot CS et al. Once-Weekly Semaglutide in Adults With Alcohol Use Disorder: A Randomized Clinical Trial. JAMA Psychiatry. 2025. 48 participants, nine weeks. Oral Semaglutide for Alcohol Use Disorder: randomized phase 2 trial, American Journal of Psychiatry, 2026; PMID 42522065. 50 treatment-seeking participants, eight weeks. Interpret primary and secondary outcomes separately. ↩
- 2026 oral semaglutide trial ↩
- NIAAA: understanding alcohol use disorder and care options ↩
- NIAAA: cutting down, quitting and withdrawal risk ↩