Semaglutide and Alcohol Cravings: What the Research Shows
Yes—early trial evidence points that way. Three small randomized trials report fewer heavy drinking days on semaglutide, including a 41.1% drop in one 26-week trial, though craving results are mixed. It is not FDA-approved for alcohol use disorder, and larger trials are still needed.
Semaglutides.org provides information only, not medical advice. Talk to a licensed provider about your own care before making changes. Some telehealth links are affiliate links and are labeled where they appear.
Yes, early trial evidence points in that direction. Three randomized controlled trials have now reported that semaglutide reduced heavy drinking days, alcohol craving, or both in adults with alcohol use disorder, and the largest of them found a 41.1% reduction in heavy drinking days in people who also had obesity [2] [4] [6]. The interest started earlier, with anecdotal reports from patients describing a reduced desire to drink while on the drug for diabetes or weight loss [1] [2]. But semaglutide is FDA-approved only for type 2 diabetes (2017) and obesity (2021)—not for alcohol use disorder—and every trial team so far has concluded that larger and longer trials are warranted before anything changes [1] [2] [4] [6].
The strongest signal so far is on heavy drinking days. Craving results are more mixed, and one trial missed its main craving endpoint. Here is how the three trials stack up.
| Trial | Who was studied | Treatment | Length | Headline result |
|---|---|---|---|---|
| JAMA Psychiatry, February 2025 | 48 adults with AUD, not seeking treatment | Weekly semaglutide injection vs. placebo | 9 weeks | Lower lab drinking, craving, drinks per drinking day, and heavy drinking days vs. placebo [2] |
| The Lancet, 2026 | 108 treatment-seeking adults with AUD and obesity | Weekly semaglutide vs. placebo, added to cognitive behavioral therapy | 26 weeks | 41.1% reduction in heavy drinking days, 13.7% greater than placebo [4] |
| American Journal of Psychiatry, 2026 | 50 adults with moderate-to-severe AUD who wanted to cut down | Oral semaglutide 3 mg/day, then 7 mg/day, vs. placebo | 8 weeks | Fewer heavy drinking days and less everyday craving; missed its main lab-craving endpoint [6] [7] |
Why would a diabetes and weight-loss drug affect drinking?
Semaglutide is a GLP-1 receptor agonist [1]. In animal studies, it reduced alcohol intake and relapse-like drinking in male and female rats (Aranäs et al., EBioMedicine 2023), and it reduced alcohol drinking while modulating central GABA neurotransmission in rodents (Chuong et al., JCI Insight 2023) [1].
Klara Klein of UNC Chapel Hill, senior author of the first semaglutide alcohol trial, says preclinical work suggests these effects are likely mediated in the brain and involve changes in reward processing [2]. How that works in humans has not been established yet, and researchers say more work is needed to understand the mechanism [2].
How big is the problem these trials are aimed at?
An estimated 29.5 million Americans aged 12 and older—about 10.6% of that age group—had alcohol use disorder in 2021, according to the SAMHSA National Survey on Drug Use and Health [1]. The death toll is large, though estimates differ: one figure cited in the research puts alcohol use disorder at more than 80,000 US deaths per year [1], while the US Surgeon General has attributed an estimated 178,000 US deaths per year to alcohol [2].
Only three medications for alcohol use disorder are approved by the FDA (four by the European Medicines Agency), and their benefits are described as modest [1]. Those existing drugs are also under-utilized [2]. George Koob, director of the NIH’s National Institute on Alcohol Abuse and Alcoholism, said that very few medications are currently approved for alcohol use disorder, that these are vastly underutilized, and that a new option that is more accessible and more effective “could be a gamechanger for closing the treatment gap” [4].
What did the first randomized trial of injectable semaglutide find?
The first randomized trial, published in JAMA Psychiatry in February 2025, was led by first author Christian Hendershot of the USC Institute for Addiction Science and senior author Klara Klein of UNC Chapel Hill [2] [3]. It enrolled 48 adults with alcohol use disorder who were not actively seeking treatment [2].
Participants received weekly injections of semaglutide or placebo for 9 weeks [2]. Before and after treatment, they went through a lab session: a two-hour period in a comfortable setting where they could drink their preferred alcoholic beverages, with instructions to delay drinking if they wished, while researchers documented delays and drinks consumed [2].
After treatment, the semaglutide group consumed less alcohol in the lab, measured by grams of alcohol consumed and breath alcohol concentration [2]. Outside the lab, semaglutide was associated with reduced weekly alcohol craving, fewer average drinks on drinking days, and greater reductions in heavy drinking days compared with placebo [2]. Notably, the trial used only the lowest clinical doses of semaglutide, and the researchers reported that the magnitude of effects on several drinking outcomes appeared potentially greater than is often seen with existing AUD medications [2].
Among a small subgroup of participants who smoked cigarettes at baseline, those on semaglutide had significantly greater reductions in average cigarettes per day than the placebo group [2]. The study was funded by NIAAA grant R21AA026931 [2]. The authors described the data as preliminary, with long-term effects, ideal doses, and treatment duration still unknown [2].
What did the larger 26-week trial in people with AUD and obesity find?
A larger trial, led by Copenhagen University Hospital with NIH scientists and reported in an NIH news release on April 30, 2026, provided what NIH called the first randomized controlled trial evidence that a GLP-1 receptor agonist reduces heavy drinking days in patients with alcohol use disorder and obesity [4]. First author Mette Kruse Klausen and corresponding author Anders Fink-Jensen led the scientific team, and the study was published in The Lancet (DOI 10.1016/S0140-6736(26)00305-3) [4].
The trial enrolled 108 treatment-seeking patients with alcohol use disorder and comorbid obesity [4]. It was randomized, double-blind, and placebo-controlled: everyone received standard cognitive behavioral therapy, and participants got either weekly semaglutide or placebo for 26 weeks on top of it [4].
The semaglutide group experienced a 41.1% reduction in heavy drinking days—a 13.7% greater reduction than the placebo group [4]. Measurements of blood-alcohol biomarkers supported the self-reported drinking data [4]. The GLP-1 group also showed greater decreases in body weight, blood pressure, and other clinical measures [4]. Adverse effects included gastrointestinal symptoms, which the authors described as transient and mild [4].
One comparison drew attention: the number needed to treat in this study was 4.3, while the NNT of approved AUD medications is 7 or higher [4]. NIDA director Nora Volkow, a study co-author, said, “We’re beginning to see some of that potential for GLP-1s to treat drug addiction turn into reality. Questions remain but this is nonetheless very encouraging” [4].
There are limits to keep in mind. Everyone in the trial had obesity, and everyone received cognitive behavioral therapy, so semaglutide was tested as an add-on to therapy rather than as a standalone treatment [4]. The authors’ stated next step is studying the drug over a longer duration and in a larger population [4].
What about the pill version of semaglutide?
A phase 2 trial of oral (pill) semaglutide, published in the American Journal of Psychiatry and announced by the American Psychiatric Association on July 29, 2026, produced a mixed picture [6] [7].
The trial enrolled 50 adults with moderate-to-severe alcohol use disorder who wanted to reduce or stop drinking [6] [7]. It was double-blind, randomized, and parallel-arm: participants took oral semaglutide at 3 mg/day for 4 weeks, then 7 mg/day for 4 weeks—8 weeks total—or a placebo [6] [7]. Joseph Schacht, PhD, a professor of psychiatry at the University of Colorado Anschutz School of Medicine, said in the APA announcement that the study suggests oral semaglutide may help reduce heavy and harmful drinking, even in people who are not trying to quit alcohol entirely [6]. The study was funded by the NIAAA, the National Center for Advancing Translational Sciences, the Hewit Family Foundation, and an anonymous family donor [6].
The primary outcome—laboratory alcohol cue-elicited craving at week 6—was not significantly reduced, and drinks per day were not significantly reduced either [7]. But several preregistered and exploratory outcomes favored semaglutide: heavy drinking days were significantly reduced (b=-0.580, 95% CI -1.012 to -0.148), as were drinks per drinking day (b=-1.177, 95% CI -2.307 to -0.047) and naturalistic, everyday craving (b=-2.195, 95% CI -4.174 to -0.216) [7]. Alcohol-related negative consequences fell faster than placebo (b=-4.618, 95% CI -8.651 to -0.585) [7].
Participants started out drinking heavily—about six drinks per day and roughly five heavy drinking days per week, with nearly all at a high or very high WHO risk drinking level [6]. By the end of treatment, 81% of the semaglutide group had dropped at least one WHO risk drinking category, compared with about 54% of the placebo group [6]. Among a small subgroup of 11 participants who had used cannabis before the study, cannabis use days were also reduced (b=-1.434, 95% CI -2.568 to -0.301), though the researchers emphasized this finding was exploratory [6] [7].
The authors framed the results as confirming earlier findings among non-treatment-seekers with less severe AUD, and concluded that continued development of semaglutide for AUD is warranted [7].
What did the earlier GLP-1 trial with exenatide show?
Before semaglutide was tested, a smaller randomized trial (n=127) evaluated exenatide, an older GLP-1 receptor agonist, once weekly versus placebo as an adjunct to cognitive behavioral therapy [1]. It reported significant reductions in heavy drinking days and total alcohol intake only in the subgroup of patients with obesity [1]. The BMJ summarized the same trial as showing no significant reduction in heavy drinking except in people with a BMI over 30 [8], and NIH framed it as a trial that found no effect overall but a strong response in the obesity subset [4]. Klein noted that the first human trial of an older GLP-1 receptor agonist on alcohol use was inconclusive [2].
What do real-world medical records show?
Beyond trials, database studies of electronic health records point the same direction, though they cannot prove cause and effect.
A retrospective cohort study of 83,825 patients with obesity, published in Nature Communications in 2024, found that semaglutide was associated with a 50%-56% lower risk of both incident and recurrent alcohol use disorder over 12 months compared with other anti-obesity medications [1]. The findings were consistent across gender, age group, race, and in patients with and without type 2 diabetes, and were replicated in a separate population of 598,803 patients with type 2 diabetes [1]. In one matched comparison, AUD diagnoses occurred in 0.37% of the semaglutide group versus 0.73% of those on non-GLP-1 anti-obesity medications (HR 0.50, 95% CI 0.39-0.63); in another, the figures were 0.35% versus 0.78% when compared with naltrexone or topiramate (HR 0.44, 95% CI 0.32-0.61) [1]. The authors called for randomized clinical trials [1].
A target-trial emulation published in The BMJ on March 4, 2026, drew on US Department of Veterans Affairs records with a base population of 606,434 veterans with type 2 diabetes [8]. Compared with starting an SGLT-2 inhibitor, starting a GLP-1 receptor agonist was associated with lower risk of alcohol use disorder (HR 0.82, 95% CI 0.78-0.85), along with lower risk of cannabis (0.86), cocaine (0.80), nicotine (0.80), and opioid (0.75) use disorders and a composite of all substance use disorders (0.86) [8]. Among people with a pre-existing substance use disorder, GLP-1 initiation was associated with lower risk of SUD-related emergency department visits (HR 0.69), hospital admissions (0.74), SUD-related mortality (0.50), drug overdose (0.61), and suicidal ideation or attempt (0.75) [8].
These are observational findings, and the BMJ authors themselves describe them as a “potential role” warranting further evaluation—not proof [8].
What about smoking and other substances?
The UNC/USC research group also ran a randomized trial of once-weekly semaglutide in adults with daily cigarette use, published in JAMA Network Open on May 26, 2026 [10] [3].
The BMJ paper’s background notes that small GLP-1 smoking-cessation trials have shown mixed results—one found no effect on abstinence, another found improved cigarette abstinence [8]. Prior observational work has linked GLP-1 use to reduced risk of opioid overdose and alcohol intoxication, and GLP-1 effects on cannabis use disorder have also been reported in real-world cohorts [8].
What is semaglutide approved for right now?
GLP-1 receptor agonists, including semaglutide, are FDA-approved for weight loss and management of type 2 diabetes only [6]. As of the latest published trials in July 2026, no GLP-1 has an FDA-approved indication for alcohol use disorder [6] [7].
The three FDA-approved AUD medications still exist and remain under-used [1] [2]. The semaglutide trials were conducted under medical supervision, with monitoring and defined dosing schedules set by researchers—not by everyday prescribing practice [2] [6]. Anyone considering questions about drinking and medication should talk with a clinician.
What is still unknown?
Plenty. The long-term effect on alcohol consumption, the optimal dose, and the ideal treatment duration are all open questions—and the right dosing for AUD may differ from what is recommended for diabetes and obesity [2]. Larger and longer studies in broader populations are needed to establish safety and efficacy in people with alcohol use disorder [2].
The Copenhagen team plans longer-duration, larger-population studies next [4]. Whether the results generalize to people with AUD who do not have obesity was not tested in that trial, since everyone enrolled had obesity [4].
Sources
- Nature Communications. Associations of semaglutide with incidence and recurrence of alcohol use disorder in real-world population.
- UNC Health Newsroom. Semaglutide shows promise in reducing cravings for alcohol, heavy drinking. February 2025.
- JAMA Psychiatry. Once-weekly semaglutide in adults with alcohol use disorder: a randomized clinical trial.
- National Institutes of Health. Adding weekly GLP-1 to cognitive behavioral therapy further reduces heavy drinking.
- The Lancet. Once-weekly semaglutide versus placebo in patients with alcohol use disorder and comorbid obesity: a randomised, double-blind, placebo-controlled trial.
- American Psychiatric Association. New research: oral semaglutide reduced heavy drinking in adults with alcohol use disorder.
- PubMed. Oral semaglutide for alcohol use disorder: a randomized clinical trial.
- The BMJ. Glucagon-like peptide-1 receptor agonists and risk of substance use disorders among US veterans with type 2 diabetes: cohort study. March 4, 2026.
- The Lancet Psychiatry. Association between GLP-1 receptor agonist use and worsening mental illness in people with depression and anxiety in Sweden: a national cohort study.
- JAMA Network Open. Once-weekly semaglutide in adults with daily cigarette use: a randomized clinical trial.
- Addiction Science & Clinical Practice. Efficacy of semaglutide in abstinence from illicit and nonprescribed opioids in an outpatient population with OUD: trial protocol.
- ClinicalTrials.gov. Effects of NNC0194-0499, cagrilintide, and semaglutide alone or in combinations on liver damage and alcohol use in people with alcohol-related liver disease.
Frequently asked questions
Does semaglutide reduce alcohol cravings or just how much you drink?
Both, depending on the measure. The injection trial reported reduced weekly alcohol craving, and the oral trial found significant reductions in naturalistic, everyday craving. However, the oral trial's primary outcome—laboratory cue-elicited craving at week 6—was not significantly reduced, so craving evidence is not uniform across settings.
How much did drinking actually fall in the trials?
In the 26-week trial of 108 patients with AUD and obesity, the semaglutide group cut heavy drinking days by 41.1%, a 13.7% greater reduction than placebo. In the oral trial, 81% of the semaglutide group dropped at least one WHO risk drinking category by the end of treatment, versus about 54% on placebo.
Does the pill version of semaglutide work for drinking?
Partially, based on one phase 2 trial of 50 adults. Oral semaglutide at 3 mg/day then 7 mg/day did not significantly reduce laboratory craving or drinks per day, but did significantly reduce heavy drinking days, drinks per drinking day, everyday craving, and alcohol-related consequences.
Is semaglutide approved by the FDA for alcohol use disorder?
No. GLP-1 receptor agonists, including semaglutide, are FDA-approved only for weight loss and type 2 diabetes management. As of the latest published trials in July 2026, there is no FDA-approved GLP-1 indication for alcohol use disorder.
How does semaglutide compare with existing alcohol use disorder medications?
In the 26-week trial, semaglutide's number needed to treat was 4.3, compared with 7 or higher for approved AUD medications. The UNC team also noted the effect magnitude on several drinking outcomes appeared potentially greater than typically seen with existing AUD medications, even at the lowest clinical doses. Only three AUD medications are FDA-approved, and their benefits are modest.
Do you need to have obesity for semaglutide to help with drinking?
It is not yet known. The 26-week trial enrolled only people with AUD plus obesity, so whether the results apply to people without obesity was not tested. An earlier exenatide trial found significant reductions in heavy drinking days and total alcohol intake only in the subgroup of patients with obesity.
Was therapy involved in the trials, or was it the drug alone?
In the 26-week trial of 108 patients, everyone received standard cognitive behavioral therapy, and semaglutide or placebo was added on top, so the drug was tested as an add-on to therapy rather than a standalone treatment. The trial results therefore reflect semaglutide combined with cognitive behavioral therapy, not the drug by itself.
What side effects were seen in the alcohol trials?
In the 26-week trial of 108 patients with AUD and obesity, the researchers noted adverse effects including gastrointestinal symptoms, which they described as transient and mild. That trial also recorded greater decreases in body weight, blood pressure, and other clinical measures in the GLP-1 group. Anyone with concerns about side effects should raise them with a clinician.
This article summarizes FDA labeling, published research and company information current as of September 14, 2026. It is not medical advice and does not replace a conversation with your own healthcare provider. How we research and verify.