Semaglutide and Alcohol: Why Ozempic Reduces Drinking
GLP-1 and Addiction
40%
Participants taking semaglutide drank approximately 40% less alcohol than placebo in a lab-based self-administration test after 9 weeks of treatment.
Hendershot et al., JAMA Psychiatry, 2025
Hendershot et al., JAMA Psychiatry, 2025
View as imagePeople taking semaglutide for weight loss or diabetes started reporting something unexpected: they no longer wanted to drink. Not a decision to quit. Not willpower. A genuine loss of interest in alcohol, as if the craving had been quietly switched off. This anecdotal pattern, repeated across thousands of patient reports, prompted researchers to test whether the effect was real. The results from both clinical trials and large-scale real-world data are now available, and they point to a biological mechanism rooted in how GLP-1 receptors in the brain's reward center modulate the dopamine system that drives addictive behavior.
Key Takeaways
- In a 48-patient randomized trial, semaglutide reduced alcohol consumption by roughly 40% in a lab self-administration test and lowered weekly cravings compared to placebo[1]
- A real-world analysis of 83,825 patients found semaglutide was associated with 50-56% lower risk of alcohol use disorder incidence and recurrence compared to other anti-obesity medications[2]
- Semaglutide reduced alcohol intake and relapse-like drinking in both male and female rats, with effects on dopamine release in the nucleus accumbens[3]
- In 20 alcohol-preferring vervet monkeys, semaglutide at 0.05 mg/kg twice weekly reduced voluntary alcohol intake over 4 weeks[4]
- A systematic review of observational data found GLP-1 RAs associated with a 36% lower hazard of alcohol-related events (HR 0.64, p<0.001)[5]
- Both semaglutide and tirzepatide reduced alcohol consumption in individuals with obesity, with effects reported across multiple GLP-1 drugs[6]
The First Randomized Trial
Hendershot et al. published the first randomized, placebo-controlled trial of semaglutide for alcohol use disorder (AUD) in JAMA Psychiatry in 2025.[1] The study enrolled 48 adults with moderate AUD who received either low-dose semaglutide or placebo for 9 weeks.
The primary endpoint was alcohol consumption during a post-treatment laboratory self-administration task, where participants had access to their preferred alcoholic beverage under controlled conditions. Those who had received semaglutide consumed approximately 40% less alcohol than placebo participants. The effect sizes for grams of alcohol consumed and peak breath alcohol concentration were medium to large.
Semaglutide also reduced drinks per drinking day and weekly alcohol craving scores. The craving reduction was statistically meaningful (p=0.024) and predicted greater reductions in heavy drinking over time relative to placebo.
The trial did not find differences in average drinks per calendar day or total number of drinking days. This distinction matters: semaglutide appears to reduce how much people drink when they do drink, rather than eliminating drinking occasions entirely. The pattern suggests modulation of the rewarding properties of alcohol rather than a blanket behavioral suppression.
Among a small subgroup of participants who smoked cigarettes, those taking semaglutide also reduced their daily cigarette count, hinting at a broader reward-modulation effect that extends beyond alcohol.
Real-World Evidence at Scale
Wang et al. (2024) analyzed electronic health records from 83,825 patients with obesity and compared outcomes between those prescribed semaglutide and those on other anti-obesity medications.[2]
Semaglutide was associated with a 50-56% lower risk for both the incidence (new cases) and recurrence of alcohol use disorder over a 12-month follow-up period. This association held across demographic subgroups and was replicated in a separate cohort of patients with type 2 diabetes, suggesting the effect is not limited to obesity populations.
King et al. (2025) reported on 14 patients with overweight/obesity and hazardous drinking (AUDIT score 8 or above) who were prescribed GLP-1 RAs (8 on semaglutide, 5 on tirzepatide, 1 on liraglutide). After a mean of 9.6 months of treatment, mean AUDIT scores dropped from 11.9 to 4.1, a 66% reduction.[7]
Richards et al. (2023) documented 6 patients with positive AUDIT screenings who showed significant reductions in AUD symptoms after starting semaglutide for weight loss. The mean AUDIT score decreased by 9.5 points.[8]
Sinha et al. (2025) conducted a systematic review and meta-analysis combining observational and trial data. The observational studies showed GLP-1 RAs were associated with a hazard ratio of 0.64 for alcohol-related events (p<0.001), a 0.66 hazard ratio for AUD specifically, and a 0.50 hazard ratio for intoxication events. The randomized trial data, drawn from 430 participants, showed trends toward reduced consumption and craving but did not reach statistical significance for all endpoints.[5]
Nasrollahizadeh et al. (2026) published a systematic review and meta-analysis specifically focused on repurposing GLP-1 RAs for AUD, further consolidating the evidence for alcohol-reduction effects across multiple study designs.[9]
The Brain Mechanism
GLP-1 receptors are not limited to the gut and pancreas. They are expressed throughout the brain, including in regions central to reward processing: the ventral tegmental area (VTA), nucleus accumbens (NAc), lateral septum, and hypothalamus. This distribution is what connects a diabetes drug to addiction neurobiology, a topic explored in depth in the pillar article on GLP-1 and the addiction connection.
Dopamine modulation
Alcohol produces its rewarding effects partly by triggering dopamine release in the nucleus accumbens. Chuong et al. (2023) showed that semaglutide dose-dependently reduced binge-like alcohol drinking in mice and also reduced intake of other caloric and noncaloric solutions, suggesting a general dampening of reward-driven consumption. The study identified effects on central monoamine signaling, indicating that semaglutide modulates the dopamine pathways that make alcohol feel rewarding.[10]
Aranas et al. (2023) demonstrated that semaglutide reduced alcohol intake and relapse-like drinking in both male and female rats, with measurable effects on dopamine release in the nucleus accumbens.[3] The relapse reduction is particularly relevant: preventing return to drinking after a period of abstinence is one of the hardest problems in AUD treatment.
A specific brain circuit
Edvardsson et al. (2025) identified a GLP-1 circuit in the lateral septum that modulates reward processing and alcohol intake in rats. This finding moves beyond the general observation that GLP-1 receptors exist in reward regions and toward mapping the specific neural architecture through which semaglutide reduces alcohol consumption.[11]
Cross-species consistency
Fink-Jensen et al. (2025) tested semaglutide in 20 alcohol-preferring male vervet monkeys, a primate model with natural variation in alcohol preference that is closer to human drinking behavior than rodent models. Semaglutide at 0.05 mg/kg twice weekly subcutaneously reduced voluntary alcohol intake during a 4-week alcohol access period.[4] The consistency across mice, rats, monkeys, and now human clinical data strengthens the case that GLP-1 receptor activation reduces alcohol's rewarding properties through a conserved biological mechanism.
Windram et al. (2025) showed that semaglutide, tirzepatide, and retatrutide all attenuated the interoceptive effects of alcohol (the internal sensations that signal intoxication) in both male and female rats, suggesting the effect extends across the GLP-1 agonist drug class and is not unique to semaglutide.[12]
Beyond Alcohol: A Broader Reward Effect
The alcohol findings fit within a larger pattern. People on semaglutide report reduced interest in multiple reward-driven behaviors, not just drinking. Quddos et al. (2023) showed that both semaglutide and tirzepatide reduced alcohol consumption in individuals with obesity.[6] The smoking reduction noted in the Hendershot trial adds another data point. This broader pattern is explored in the cluster article on GLP-1 agonists and compulsive behavior.
Jerlhag (2025) reviewed the evidence for GLP-1 RAs as therapeutic targets for AUD and concluded that the reward-modulating mechanism is consistent with effects on multiple substances of abuse, not just alcohol.[13] The connection between neuropeptides and alcohol dependence extends well beyond GLP-1, with NPY, opioid peptides, and oxytocin all playing documented roles.
Aranas et al. (2025) took this a step further by combining GLP-1 and amylin agonists, demonstrating synergistic-like decreases in alcohol intake that exceeded either drug alone.[14] This combination approach mirrors the multi-receptor agonist strategy already transforming obesity treatment (tirzepatide targets both GLP-1 and GIP receptors) and suggests that addiction may eventually be treated with similarly engineered multi-target peptides.
The genetic evidence adds another layer. Reitz et al. (2025) used genetically modeled GLP-1R and GIPR agonism to show that both receptor pathways reduce binge drinking and alcohol-associated phenotypes, providing Mendelian randomization evidence that GLP-1 receptor activation causally influences alcohol behavior rather than merely correlating with it.[15]
What the Evidence Does Not Show
The enthusiasm around semaglutide and alcohol needs to be tempered by what remains unknown.
The randomized evidence is small. The Hendershot trial enrolled 48 patients. Larger trials are underway, but definitive evidence of efficacy for AUD does not yet exist. The meta-analysis by Sinha et al. found that randomized trial data did not reach significance for all consumption endpoints, even though observational data showed strong associations.[5]
Observational studies cannot prove causation. The Wang et al. analysis of 83,825 patients is the largest dataset, but patients who received semaglutide may differ from those who received other medications in ways that independently affect drinking behavior. Confounding by indication, healthy user bias, and surveillance bias are all possible.
Optimal dose for AUD is unknown. The Hendershot trial used low-dose semaglutide (0.24 mg escalating to 1.0 mg weekly), not the full 2.4 mg dose used for weight loss. Whether higher doses produce stronger alcohol effects, or whether the dose-response curve differs for addiction versus metabolic endpoints, has not been established.
Duration of effect is unclear. If people stop taking semaglutide, does alcohol interest return? The current data do not answer this question. The parallel with weight regain after GLP-1 discontinuation raises the possibility that alcohol reduction may also be drug-dependent rather than representing a permanent neurological reset.
Semaglutide is not approved for AUD. Any use for alcohol reduction is off-label. The gastrointestinal side effects (nausea, vomiting, diarrhea) that accompany semaglutide initiation add a practical barrier for patients whose primary concern is drinking rather than weight or blood sugar.
The mechanism may not be alcohol-specific. Chuong et al. (2023) showed semaglutide reduced intake of caloric and noncaloric solutions generally, not just alcohol.[10] The reduced drinking may partly reflect a global decrease in reward-seeking or caloric intake rather than a targeted effect on alcohol neurobiology. Whether this distinction matters clinically (the drinking still decreases) or mechanistically (the drug may not address alcohol-specific craving circuitry) remains debated.
The Bottom Line
Converging evidence from a randomized clinical trial, large-scale real-world data, and preclinical studies across mice, rats, and monkeys shows that semaglutide reduces alcohol consumption and cravings. The mechanism involves GLP-1 receptor activation in brain reward circuits that modulate dopamine signaling. The effect appears to extend across GLP-1 agonist drugs and beyond alcohol to other reward-driven behaviors. The clinical evidence, while promising, consists of one small randomized trial and observational data that cannot establish causation. Larger trials, optimal dosing studies, and long-term follow-up data are needed before semaglutide can be considered a validated treatment for alcohol use disorder.