GLP-1s and addiction: how Ozempic, Wegovy, and the next generation could remap alcohol treatment

A drug class developed to treat diabetes and obesity has spent the last three years quietly rewriting how researchers think about substance use. The drugs are GLP-1 receptor agonists — semaglutide (sold as Ozempic and Wegovy), tirzepatide, and a deepening pipeline at Eli Lilly, Novo Nordisk, and several smaller players. The most-cited finding came in 2024, when a retrospective cohort of more than 600,000 U.S. veterans showed that people who started a GLP-1 for diabetes were 15–20% less likely to develop or relapse to alcohol, opioid, or stimulant use disorders over a three-year window (Science / AAAS). The next twelve months will turn that signal into evidence — or rule it out.

Why this story matters more for alcohol than anything else

Alcohol kills roughly 178,000 Americans every year — a figure CDC analysts published in MMWR after a 29% increase between the 2016–2017 and 2020–2021 averages. That is more than twice the projected drug overdose total for the year ending November 2025. Yet only a small fraction of people with an alcohol use disorder (AUD) ever receive an FDA-approved medication. The three approved drugs — naltrexone, acamprosate, and disulfiram — have been on the market for decades and are written for fewer than 10% of patients who would qualify. AUD is the substance use disorder where the gap between “drugs we have” and “drugs people are getting” is most painful, and where a new mechanism would do the most population-level work.

That gap is what makes the GLP-1 story consequential. A randomized trial of low-dose semaglutide in adults with AUD, published in early 2025 by the National Institute on Alcohol Abuse and Alcoholism, found reductions in heavy drinking days and in self-reported cravings over a nine-week trial — small, but in the same direction as the larger observational signal. A second, larger NIDA trial in Baltimore is underway, with Eli Lilly’s own AUD readout expected later in 2026 (NPR).

The mechanism, in plain terms

For people in recovery, the “why” is often the most useful part. GLP-1 receptor agonists work, in part, by dampening signaling in the mesolimbic dopamine pathway — the same reward circuit that food, alcohol, opioids, nicotine, cocaine, and cannabis all converge on (WashU Medicine). Patients in qualitative studies don’t describe newfound willpower. They describe the “pull” toward a drink, a cigarette, or a slice of cake going quiet — what researchers call a reduction in reward salience. That is a fundamentally different intervention than the existing FDA-approved AUD medications, which work by blocking opioid receptors (naltrexone) or producing aversive symptoms (disulfiram).

Nora Volkow, the longtime director of the National Institute on Drug Abuse, co-authored an April 2026 review that catalogues these signals across alcohol-related events, opioid overdose risk, nicotine, cocaine, cannabis, and other substances. Her conclusion: the cross-substance breadth is the most striking thing about the data — and the strongest reason to keep the experimental enthusiasm in check.

What the trials cannot yet tell us

A retrospective cohort, even one of 600,000 veterans, cannot establish causality the way a randomized trial can. The veterans who chose to start a GLP-1 were probably different in dozens of unmeasured ways from those who did not. The published RCTs so far are small and short. We do not yet know whether the effect on heavy-drinking days persists at twelve months, whether people regain craving when they stop the drug, or how it interacts with the medications people already take for AUD. We also do not know what it costs to put a 30-year-old in early recovery on a $1,000-a-month injectable for a decade — or who pays for that.

There is also a wider-aperture concern. GLP-1 supply remains constrained for the people for whom these drugs were first approved — patients with diabetes and severe obesity. Expanding indications to addiction would compete for that supply. Any clinical recommendation will need to grapple with that, not just the pharmacology.

GLP-1 supply remains constrained for the people for whom these drugs were first approved — patients with diabetes and severe obesity.

Where this fits in someone’s recovery decision

If you are weighing options for AUD treatment today, GLP-1s are not yet an FDA-approved indication, and most clinicians cannot prescribe them off-label for addiction without significant friction. The evidence-based options remain naltrexone, acamprosate, disulfiram, and behavioral therapies like cognitive behavioral therapy and motivational enhancement therapy — all available through MAT and outpatient programs across Arizona. What is changing is the horizon. By late 2027, depending on how the Eli Lilly readout and the NIDA Baltimore trial perform, it is plausible that a new tool will join the list. That would be the first new mechanism in alcohol pharmacotherapy in twenty years.

Why this matters for people in recovery

Recovery is rarely a single-medication story. The most reliable evidence — for alcohol, opioids, and stimulants alike — points to combinations: medication plus counseling plus peer or community support plus, very often, address-the-rest-of-your-life work like housing and stable income. GLP-1s, if they pan out, will fit inside that mix, not replace it. The hopeful read is that a generation of new biology is finally meeting a treatment gap that has been allowed to stay open for a generation.

If you or someone you love is looking for AUD treatment in Arizona, Rize can help you compare options near you — including the medications and therapies that have evidence behind them today, not just the ones that may arrive next year.

---

This article reports on emerging research. It is not medical advice. Decisions about any medication should be made with a clinician who knows your full history.

If you are in crisis, call 988 (Suicide & Crisis Lifeline) or text HOME to 741741. SAMHSA’s National Helpline: 1-800-662-HELP.