Ozempic and Alcohol: What the First Clinical Trial Actually Found

The Study at a Glance

The trial was led by Christian Hendershot, PhD, at the USC Institute for Addiction Science, with Klara Klein, MD, PhD, at the UNC School of Medicine as senior author. It was published in JAMA Psychiatry — one of the highest-impact journals in the field — and funded by the National Institute on Alcohol Abuse and Alcoholism (NIAAA grant R21AA026931).

This wasn't a survey. It wasn't a Reddit scrape. It was a prospective, randomized, placebo-controlled Phase 2 clinical trial — the gold standard design for testing whether a drug actually does what people think it does.

How They Tested It

The researchers designed something unusual: a laboratory bar. One week before the first injection, participants came into a controlled lab setting, sat in a comfortable environment, and were given access to their preferred alcoholic beverages for two hours. They could drink or delay. Researchers tracked exactly how much they consumed using grams of alcohol and breath alcohol concentration readings.

Then participants were randomized — half received weekly semaglutide injections, half received placebo — for nine weeks. During those nine weeks, the research team also tracked weekly drinking patterns, cravings, and heavy drinking days through clinical assessments.

After the nine-week treatment period, everyone came back to the lab bar. Same setup. Same beverages. Same two-hour window. The question: would the semaglutide group drink less?

What They Found

The answer was yes — and the magnitude of the effect surprised the research team.

Critically, the trial used only the lowest clinical doses of semaglutide — 0.25 mg escalating to 0.5 mg. For context, the standard weight-loss dose of Wegovy is 2.4 mg, nearly five times higher. The researchers noted that the magnitude of semaglutide's effects on several drinking outcomes appeared potentially greater than what is typically seen in similar studies with existing FDA-approved AUD medications.

That's a remarkable signal for a drug at one-fifth its standard therapeutic dose.

The Smoking Bonus

Among participants who smoked cigarettes at the start of the trial, those in the semaglutide group showed significantly greater reductions in average cigarettes per day compared to placebo. This is particularly notable because no medication currently approved for AUD also helps with smoking cessation. If these dual effects hold up in larger trials, it would represent a genuinely novel treatment profile.

Why This Matters: The Treatment Gap

Alcohol kills an estimated 178,000 Americans per year, according to the CDC. It is linked to liver disease, cardiovascular disease, and is a confirmed carcinogen — a fact the U.S. Surgeon General underscored in a recent advisory. Alcohol use disorder affects roughly 29 million American adults.

Yet only three drugs are currently FDA-approved for AUD: naltrexone, acamprosate, and disulfiram. All three are significantly underutilized. Many patients don't respond. Many clinicians don't prescribe them. The field has been largely stagnant for decades.

The popularity of GLP-1 medications could change the calculus. Tens of millions of people already take semaglutide for diabetes or weight management. If even a fraction of them experience reduced drinking as a secondary benefit — and the clinical data says many do — it creates a natural pathway for broader adoption in AUD treatment.

What This Trial Doesn't Tell Us

The researchers themselves are careful about the limitations, and it's important to be honest about what we still don't know.

It was small. 48 participants is enough to detect a signal, not to establish definitive treatment guidelines. Phase 3 trials with hundreds or thousands of participants are needed.

It was short. Nine weeks of treatment doesn't tell us about long-term effects. Does the craving reduction persist? Does tolerance develop? What happens when patients stop taking the drug?

Participants weren't seeking treatment. The trial deliberately enrolled people who weren't actively trying to quit drinking. That's useful for isolating the pharmacological effect, but it means we don't yet know how semaglutide performs in patients who are motivated to change their relationship with alcohol.

The mechanism isn't fully understood. Preclinical studies suggest GLP-1 receptors in the brain's reward circuitry — particularly the ventral tegmental area (VTA) and nucleus accumbens — play a role. Semaglutide appears to modulate dopamine signaling in these areas, dampening the neurological "want" signal. But the exact pathway is still being mapped.

The Bigger Picture: Supporting Evidence

The Hendershot trial didn't emerge in isolation. It confirmed what several large observational studies had already suggested.

A 2024 study by Wang, Volkow, and colleagues published in Nature Communications analyzed over 682,000 patient records and found that semaglutide was associated with approximately 50% lower risk of new alcohol use disorder diagnoses — and 50% lower recurrence rates in patients with prior AUD.

Separately, Qeadan and colleagues analyzed 1.3 million electronic health records from 136 U.S. health systems and found a 50% lower incidence of alcohol intoxication events among patients prescribed GLP-1 medications. That study was published in Addiction in early 2025.

And a Finnish study by Lähteenvuo and colleagues (2025) found semaglutide was associated with a 36% lower risk of alcohol-related hospitalizations in a cohort of 228,000 individuals.

None of these observational studies can prove causation — that's what the Hendershot RCT was designed to do. But when an RCT confirms what multiple large-scale observational datasets already suggested, the convergence is significant.

What Happens Next

Phase 3 trials evaluating semaglutide specifically for alcohol use disorder are now underway. Eli Lilly's CEO stated in late 2024 that the company would begin large-scale studies of its GLP-1 drugs (including tirzepatide) in alcohol and drug abuse. Novo Nordisk has an ongoing Phase 2 trial studying semaglutide in patients with alcoholic liver disease, with alcohol consumption as a secondary endpoint.

The Hendershot team and other researchers have published trial protocols for larger, longer studies that will test optimal dosing, treatment duration, and long-term outcomes. The FDA has not yet approved any GLP-1 medication for treating alcohol use disorder — but the clinical momentum is undeniable.