Could GLP-1 drugs ease alcohol cravings? The evidence is promising but mixed

Heavy-drinking days declined more in the semaglutide group. The reduction from baseline was 41.1 percentage points with semaglutide and 26.4 points with placebo, for an estimated 13.7-point difference between the groups. Those figures describe changes in the share of heavy-drinking days, not the percentage of participants who stopped drinking.

The placebo group improved, too. That makes the difference between the groups more informative than the larger before-and-after change in the semaglutide group alone. Gastrointestinal side effects were more frequent with semaglutide, and the trial’s results concern people who had both alcohol use disorder and obesity—not everyone seeking addiction care.

A 2025 randomized trial of 48 adults, which lasted nine weeks, adds a smaller signal. Participants receiving semaglutide reported reduced weekly craving and fewer drinks per drinking day, but the study found no significant difference in average drinks per calendar day or in the number of drinking days. The PBS report on GLP-1 drugs and addiction research describes the wider interest in these medicines, but the results themselves do not yet make a consistent case across drinking measures.

That is meaningful progress from anecdotes or medical-record associations: randomized comparisons can test whether outcomes differ between people assigned the drug and those assigned a placebo. But a reduction in heavy drinking is not the same as abstinence, and the studies do not show that semaglutide works for everyone.

Craving results depend on what researchers measure

A separate eight-week trial tested oral semaglutide in 50 adults seeking treatment for moderate-to-severe alcohol use disorder; 47 completed treatment. Its preregistered primary measure was craving after participants encountered an alcohol cue in a laboratory. On that key measure, semaglutide did not differ significantly from placebo.

Some secondary results—including heavy-drinking days and craving reported in everyday life—favored semaglutide. Those findings are worth following, but they do not erase the negative primary result. A laboratory response to an alcohol cue and a person’s reported cravings in daily life are different measures; neither alone can establish how a treatment affects recovery over time.

The evidence beyond alcohol is less direct. A 2026 study of 606,434 veterans with Type 2 diabetes compared people starting GLP-1 drugs with people starting SGLT-2 inhibitors, another class of diabetes medicines. The GLP-1 group had lower rates of new diagnoses of several substance use disorders, including alcohol, cannabis, cocaine, nicotine and opioid disorders. Because the study was observational, it shows an association, not that GLP-1 drugs prevented those disorders or reduced cravings.

The oral-semaglutide trial also found fewer cannabis-use days among 11 participants who used cannabis at the start of the study. That exploratory result—drawn from five people in the placebo group and six in the semaglutide group—is far too small to show that the drug treats cannabis use disorder.

Researchers are investigating whether GLP-1 drugs, which affect brain pathways involved in appetite and reward, might also influence alcohol use. The mechanism remains unsettled. Dr. Lorenzo Leggio, clinical director of the National Institute on Drug Abuse, told PBS News: “We don't fully understand how they work, yes, actually, not just for addiction. We don't fully understand how they work for obesity either”.

Addiction care needs evidence, not a weight-loss story

The prospect matters because people seeking help need options that are tested against the problems they want treated. Reports of people experiencing fewer cravings, and treatment programs’ observations, can help explain why researchers and patients are interested. They cannot establish effectiveness: reports from a program using off-label GLP-1 prescriptions, for example, are not the same as a controlled trial.

The clinical question is not whether a medication changes weight. It is whether it safely and reliably changes drinking or other substance use, for whom, and for how long. The public conversation about GLP-1 medicines can readily turn toward weight and body image, as PBS has reported. In addiction research, body size should not become a proxy for recovery or distract from the outcomes people seeking care need addressed.

The strongest alcohol-use trial tested semaglutide alongside cognitive behavioral therapy. That matters: the result points to a possible addition to care, not a reason to replace support or established treatment. The trials also leave a central question open: whether results in people with obesity apply to people with alcohol use disorder who do not have obesity.

The next step is larger, longer randomized research that can test whether the alcohol-use signal holds up, which patients benefit, and whether changes last. Until then, GLP-1 drugs are a promising research question—not an evidence-based recommendation for addiction treatment.