A large observational study found lower hospitalization rates among adults with alcohol-use disorder who started semaglutide or tirzepatide, but it cannot show that the drugs reduced drinking or treated alcohol dependence.
Adults with alcohol-use disorder who started newer GLP-1 drugs for obesity or type 2 diabetes were less likely to be hospitalized for alcohol-related reasons than similar patients taking other medications, according to an observational study involving more than 40,000 people.
The findings add to growing interest in whether medications such as semaglutide and tirzepatide may affect urges involving more than food. However, the study did not measure how much alcohol participants drank, establish why hospitalization rates differed or prove that GLP-1 drugs can treat alcohol-use disorder. The research was funded by Truveta, the health data company that provided the records used in the analysis, and several authors were current or former Truveta employees.
GLP-1 receptor agonists were developed to help manage blood sugar and are also widely used for weight management. Semaglutide is sold under brand names including Ozempic and Wegovy, while tirzepatide is sold as Mounjaro and Zepbound.
Reports from patients and earlier research have raised the possibility that these medications may also reduce interest in alcohol. Researchers conducted the new study to examine whether that apparent change could translate into fewer serious alcohol-related health events.
The study, published in BMJ Open, included 40,703 adults who had alcohol-use disorder along with either obesity or type 2 diabetes. Participants started semaglutide, tirzepatide or a comparison medication between January 2018 and December 2024.
Researchers used electronic health records to create four analyses designed to resemble clinical trials. This approach, called target trial emulation, attempts to compare groups of patients who are similar in relevant ways. It remains observational, however, because patients were not randomly assigned to treatments.
Two analyses compared GLP-1 drugs with other medications used for diabetes or obesity. Among adults with type 2 diabetes, starting semaglutide or tirzepatide was associated with a 26% lower risk of an alcohol-related hospitalization than starting another diabetes medication.
Among adults with obesity, GLP-1 use was associated with a 32% lower risk than use of another weight management medication.
The other two analyses compared people taking GLP-1 drugs with those taking medications used specifically for alcohol-use disorder, including acamprosate, disulfiram and naltrexone.
In those comparisons, GLP-1 use was associated with a 63% lower risk of alcohol-related hospitalization among people with type 2 diabetes and a 65% lower risk among people with obesity.
Those larger differences should be interpreted especially cautiously. People prescribed newer GLP-1 drugs may have differed from those receiving alcohol-use disorder medications in ways the health records could not fully capture, including the severity of their alcohol dependence, their overall health, socioeconomic circumstances and their engagement with medical care.
GLP-1 users in those analyses also had fewer hospitalizations for reasons unrelated to alcohol. That suggests the groups may have differed in their general health or access to care, rather than the medications alone explaining the lower alcohol-related hospitalization rates.
Treatment discontinuation was also common, creating another opportunity for bias. Alcohol-use disorder may have been missed or recorded inconsistently in medical records, particularly because stigma can discourage people from discussing their drinking with health care professionals.
The outcome itself requires careful interpretation. Researchers tracked alcohol-related hospitalizations using diagnosis codes and laboratory evidence of alcohol exposure. They did not directly measure participants’ drinking, cravings, treatment goals or recovery.
A reduction in hospitalizations therefore would not necessarily mean that patients stopped drinking or consumed less alcohol. It could reflect differences in drinking patterns, health status, medical supervision or the likelihood of seeking hospital care.
The study also cannot determine how GLP-1 drugs might influence alcohol-related behavior. One possibility is that their effects on appetite and the brain’s reward systems could extend to alcohol cravings, but this analysis did not test that mechanism.
“Initiation of newer GLP-1 receptor agonists among patients with alcohol-use disorder was associated with a lower observed risk of alcohol-related hospitalization, with similar associations across populations with type 2 diabetes and obesity,” the researchers wrote.
They added that the findings “may suggest a potential role for GLP-1 receptor agonists in the context of alcohol-use disorder.”
For now, semaglutide and tirzepatide are not approved treatments for alcohol-use disorder, and this study does not show that they should replace established medications or behavioral treatment. Randomized clinical trials that directly track drinking, cravings, alcohol-related harm and treatment retention will be needed to determine whether GLP-1 drugs offer a meaningful therapeutic benefit.
The research was funded by Truveta, which also provided the health record data used in the study. The authors reported that the funder had no role in the study’s design, data collection, analysis, interpretation, manuscript preparation or publication decision. The study authors conducted the research while employed by Truveta.
Several authors are current or former Truveta employees. One former Truveta employee is now employed by Eli Lilly, which manufactures tirzepatide. The authors stated that the study was conducted and submitted before that employment and that Eli Lilly had no role in the research. Other disclosed relationships included unpaid committee service for the American Heart Association and research funding from the American Heart Association and the National Institute on Aging.