A large-scale “All of Us” analysis suggests that people currently prescribed GLP-1 drink alcohol less frequently, but observations do not prove that the drug reduces alcohol use.
Study: Association between GLP-1 receptor agonist prescription and alcohol consumption in the National Institutes of Health All of Us cohort. Image credit: N Universe / Shutterstock
In a recent study published in the journal Alcohol: Clinical and experimental studiesResearchers investigated whether recorded prescriptions for glucagon-like peptide-1 receptor agonists (GLP-1RAs) were associated with self-reported alcohol intake among participants in the National Institutes of Health (NIH) All of Us research program.
background
Approximately 178,000 people die each year in the United States from excessive alcohol consumption, highlighting the urgent need for more effective treatment options. Although there are drugs to treat alcohol use disorder (AUD), only three medications are approved by the U.S. Food and Drug Administration, and they are not widely used or effective for all patients.
GLP-1RA is approved for the treatment of type 2 diabetes, obesity, and obstructive sleep apnea, and is currently being studied for its effects on processes such as appetite and satiety, reward processing, stress regulation, cognition, and neuroinflammation.
Although early animal and human studies have reported promising findings, evidence from large and diverse populations remains limited. Further research is needed to confirm whether these drugs can reduce alcohol consumption and to understand the biological mechanisms and drinking behavior involved.
About research
The researchers conducted an observational cohort study that combined one-time cross-sectional measurements of alcohol consumption with longitudinal electronic health record data from the NIH’s All of Us Research Program Curated Data Repository version 8.
They analyzed survey responses, electronic health records, and physical measurements collected from 1981 to October 2023, although most participants’ electronic health record histories began between 2014 and 2019.
Eligible adults were to complete a lifestyle survey, have a documented or calculable BMI, have had at least one inpatient, outpatient, or emergency department visit in the past 2 years, and no history of medullary thyroid cancer, multiple endocrine tumor syndrome type 2, or end-stage renal disease.
Participants were excluded if they had a record of a pregnancy in the previous year, a history of bariatric surgery, a record of a prescription for naltrexone, acamprosate, or disulfiram, or a lifetime history of alcohol use.
Participants with at least two GLP-1RA records on separate days, including at least one record in the 365 days prior to survey completion, were categorized into current or previous prescribing groups according to the timing of these records relative to completion of the lifestyle survey.
The prospective prescription group included individuals whose first GLP-1RA exposure was recorded at or after completion of the lifestyle survey and was considered the comparison group.
The term “prescription” included clinician prescriptions, dispensing, billing, administration, medication list completion, self-reporting, and records whose source was not identified and therefore did not necessarily confirm medication use or compliance.
The Alcohol Use Disorders Identification Test Consumption (AUDIT-C) questionnaire was used to measure frequency of alcohol consumption, number of drinks consumed on a typical drinking day, and frequency of consuming six or more drinks on a single occasion.
To calculate the incidence rate ratio (IRR), the researchers performed a multivariable weighted negative binomial regression analysis. They used propensity score weighting and adjusted for demographic, clinical, and health care utilization variables.
Additional analyzes included each of the three AUDIT-C questions separately and compared the results of the matched sample and the unweighted model. The Benjamini-Hochberg procedure was used to adjust for multiple comparisons.
Research results
Of the 393,596 NIH All of Us Research Program participants with available electronic health records, 20,768 participants had at least one record of GLP-1RA exposure, and 15,447 participants had two or more records on separate days.
After applying the inclusion criteria, participants were divided into three groups. 3,650 people in the current prescription group, 544 people in the previous prescription group, and 5,642 people in the future prescription group, which served as the main comparison group. An additional 270,324 eligible participants were available for propensity score matching, allowing for a closely matched comparison group.
After weighting, the absolute standardized mean differences were less than 0.1 for all measured traits. After matching, it was less than 0.1 for all characteristics except previous tobacco use.
The primary analysis assessed the association between GLP-1RA prescriptions and AUDIT-C scores. Analysis showed that individuals currently prescribed GLP-1RAs had lower AUDIT-C scores than participants in the future prescribed group. The mean AUDIT-C score for the current prescription group was approximately 5% lower than for the future prescription group, a statistically significant difference.
Participants who were previously prescribed a GLP-1RA had an average AUDIT-C score approximately 8% lower than the prospectively prescribed group, but the difference was not statistically significant. Additionally, gender-stratified analyzes showed similar patterns for both men and women in the current prescription group, but statistical significance was observed only in women before adjustment for multiple comparisons, and no significance remained after adjustment.
The researchers were unable to perform a gender-stratified analysis of previous prescription groups due to the small sample size.
The results of the secondary analyzes were generally consistent with those of the primary analyses. In propensity score matched analyses, participants with current or previous GLP-1RA records were compared to matched participants without GLP-1RA records at the time of study completion.
Paired analyzes also found that mean AUDIT-C scores were significantly lower among participants with a current prescription, whereas no statistically significant differences were observed among participants with a previously recorded prescription.
When each question on the AUDIT-C was analyzed separately, both the currently prescribed and previously prescribed groups reported drinking less frequently than the future prescribed group. Reported drinking frequency was approximately 4% lower in the currently prescribed group and 10% lower in the previously prescribed group.
No statistically significant differences were observed in reported alcohol intake on a typical drinking day or in the frequency of drinking six or more drinks at a time.
conclusion
The results showed that current GLP-1RA regimens were associated with slightly lower self-reported AUDIT-C scores, a difference that appears to reflect lower frequency of reported drinking rather than fewer drinks per session or less binge drinking. Participants were typically prescribed GLP-1RAs for non-alcohol-related indications and were not specifically recruited for AUD treatment.
Results were generally similar across weighted and propensity score matched analyses. The observed associations were modest and this study could not establish causality. Alcohol intake was assessed only once, prescription records did not confirm compliance, there is still potential for residual confounding, and the All of Us cohort is not nationally representative.
Additionally, the AUDIT-C study used a threshold of six or more drinks for all participants, which may underestimate clinically meaningful alcohol consumption in women and adults aged 65 and older. However, this result supports further investigation of GLP-1RA as a potential treatment for AUD, and larger randomized controlled trials are needed to elucidate its mechanism and assess its potential efficacy in AUD patients.
Reference magazines:
- Tindall, B., Gasdaska, A., Brannock, M.D., Preble, E., McPheeters, M., Marchal, L., Huda, A., Egan, J., Litwin, T.R., Ajemian, J., Sastry, C., Faroknia, M., and Reggio, L. (2026). Association between GLP-1 receptor agonist prescription and alcohol intake in the National Institutes of Health All of Us cohort. Alcohol: Clinical and experimental studies. 50, e70357. Doi: 10.1111/acer.70357, https://onlinelibrary.wiley.com/doi/10.1111/acer.70357
