Educational content, not medical advice. This article reports published research on GLP-1 medicines and substance use, including its size and its limits. Nothing here is a recommendation, and no medicine in this class has an approved use for any substance use condition. Treatment for alcohol or substance use belongs with the clinicians who specialise in it.

Few questions about this class of medicines travel faster than whether they quiet cravings beyond food. The interest is enormous, the anecdotes are vivid, and the randomised evidence would fit in a small room. This subject also needs one boundary drawn at the start: drinking while taking one of these medicines, the practical question, is covered at Alcohol on a GLP-1: What to Know and What to Track. Whether the medicine treats drinking is a different question, and it is the one with the thin file.

What the labels say, precisely

Across the ten current US labels for this class, checked on November 4, 2026, there is no indication, no warning and no mention at all of addiction, craving, alcohol use disorder, smoking, nicotine or tobacco. The word alcohol does appear, and its appearances are worth listing because they show how complete the absence is: alcohol swabs in the injection instructions, benzyl alcohol and polyvinyl alcohol as inactive ingredients, and the phrase nonalcoholic steatohepatitis in the Wegovy indication. That is every occurrence. A reader who arrives asking whether these medicines are approved for any of this has a clean answer, and it is no.

The semaglutide trial in alcohol use disorder

The study most often cited is a phase 2, double-blind, randomised trial published in JAMA Psychiatry in 2025, testing once-weekly semaglutide in adults with alcohol use disorder. Its shape matters as much as its result, so here is the shape: 48 participants were randomised, out of 504 assessed, at a single US academic medical centre, over 9 weeks of outpatient treatment, at a low dose of 0.5 mg a week, in people who were not seeking treatment for their drinking. Every one of those details narrows what the numbers can carry.

The primary outcome was alcohol self-administration in a laboratory, not drinking in daily life. On that task, semaglutide reduced the grams of alcohol consumed and peak breath alcohol concentration, with medium to large effect sizes. In everyday drinking, the results were mixed, and the mixture is the story: semaglutide did not affect average drinks per calendar day or the number of drinking days, while it did reduce drinks per drinking day and weekly alcohol craving, and predicted greater reductions in heavy drinking over time. An exploratory finding in a subsample of participants who smoked showed a greater relative reduction in cigarettes per day. The authors' own conclusion is the right note to end any summary on: initial prospective evidence that justifies larger clinical trials.

GLP 1 Tracker AppDoses and side effects in one dated, private record, on your phone and nowhere else. Free, no account, nothing leaves your device. Get the app

The wider base: five randomised trials

Zoom out and the file gets no thicker. A 2024 systematic review of randomised trials in Drug and Alcohol Dependence screened 1,218 studies and found only five that met its criteria, covering 630 participants in total, using exenatide in three and dulaglutide in two. Three of the five showed a significant decrease in substance use, covering alcohol and nicotine. Five randomised trials is the entire randomised evidence base that review found. That number deserves to travel with every headline this subject generates, because it is the difference between a field with early findings and a field with answers.

It is also worth noticing which molecules were studied. Most of the randomised work used exenatide and dulaglutide, the older members of the class, not the medicines most people now take. Reading a headline about a GLP-1 trial and assuming it tested semaglutide or tirzepatide is one of the quiet ways this literature gets inflated, and it is exactly the kind of detail the reading guide at How to Read a GLP-1 Study Headline: Relative Risk, Subgroups and Funding trains you to check first.

What the appetite connection does and does not mean

The reason this research feels plausible to so many people is an experience, not a mechanism. Many people taking these medicines report that intrusive thoughts about food quiet down, a patient-reported phenomenon covered at Food Noise: What It Is and How GLP-1s Change It. It is a short step from there to assuming the medicines must act on cravings of every kind, but appetite and addiction are different constructs, the research above tested specific outcomes in specific populations, and nothing in the trial file establishes that the two share one process. Plausibility is where research starts, not where it ends.

Where this leaves a curious reader

The honest summary is short. A handful of small randomised trials, mostly with older molecules, with mixed results. One 48-person phase 2 trial of semaglutide whose primary outcome was measured in a laboratory, with mixed everyday results. No approved use for any substance use condition, anywhere, and no label that mentions the subject at all. Interesting, genuinely early, and not a treatment. And the clinical field that actually treats alcohol and substance use has its own professionals and its own evidence base; anyone with a concern about their own drinking or smoking deserves that care specifically, not a weight-loss medicine justified by a phase 2 trial. What any individual takes, for any reason, is a prescriber's decision.

Frequently asked questions

Is semaglutide approved to treat alcohol use disorder?

No. Across the ten current US labels for this class, checked on November 4, 2026, there is no indication, warning or mention of alcohol use disorder, addiction, craving, smoking, nicotine or tobacco. The only appearances of the word alcohol in any of these labels are alcohol swabs in the injection instructions, benzyl or polyvinyl alcohol among inactive ingredients, and the phrase nonalcoholic steatohepatitis in the Wegovy indication. No medicine in this class is approved for any substance use condition.

What did the semaglutide drinking study actually find?

A phase 2, double-blind, randomised trial published in JAMA Psychiatry in 2025 tested once-weekly semaglutide at a low dose of 0.5 mg in 48 adults with alcohol use disorder who were not seeking treatment, over 9 weeks at a single US academic medical centre. The primary outcome was alcohol self-administration in a laboratory, on which semaglutide reduced grams of alcohol consumed and peak breath alcohol concentration. In everyday drinking it did not affect average drinks per calendar day or number of drinking days, while it reduced drinks per drinking day and weekly alcohol craving.

How much research exists on GLP-1s for addiction overall?

Very little in randomised form. A 2024 systematic review in Drug and Alcohol Dependence screened 1,218 studies and found only five randomised trials meeting its criteria, covering 630 participants in total, using exenatide in three and dulaglutide in two. Three of the five showed a significant decrease in substance use, covering alcohol and nicotine. Five randomised trials is the entire randomised evidence base that review found, which is the single most useful number on the subject.

Do GLP-1s help people stop smoking?

The evidence is one exploratory finding and part of a very small base. In the 48-person semaglutide trial, a subsample of participants who smoked showed a greater relative reduction in cigarettes per day, and in the 2024 review, three of its five randomised trials showed a significant decrease in substance use, covering alcohol and nicotine between them, using exenatide and dulaglutide. Nothing in that is an approved use, and no label mentions smoking or tobacco at all.

Can I get a GLP-1 prescribed to help me drink less?

That is not what these medicines are approved for, and this site recommends no such use. Treatment for alcohol or substance use is a clinical field with its own professionals, and someone with a concern about their drinking deserves that care specifically rather than a weight-loss medicine repurposed on the strength of a 48-person trial. What any individual should take, for any reason, is a conversation with a prescriber.