Some people report drinking less after starting GLP-1 medicines, which copy a gut hormone involved in appetite and blood sugar. The new survey behind “does Ozempic make you drink less” headlines cannot prove those medicines caused the change.
Ozempic is a brand of semaglutide. The survey grouped medicines together, so it did not produce an Ozempic-specific result.
Does Ozempic make you drink less? What the survey found
As of November 5, 2026, the day this article was written, Harvard’s November 2 coverage highlights a new national survey. The research letter in JAMA Psychiatry was published online October 14, 2026.
Researchers surveyed RAND’s American Life Panel in March and April 2026. Among 2,009 adults, 368 reported current or past GLP-1 use. Of those, 242 were drinking alcohol when they started the medicine.
Those drinkers recalled whether their alcohol use changed during the first three months. About four in ten reported drinking less; just over half reported no change.
| Reported change | All 242 drinkers | 129 drinkers flagged for unhealthy use |
|---|---|---|
| Less alcohol | 44.2% | 54.1% |
| No change | 52.6% | 40.2% |
| More alcohol | 3.3% | 5.7% |
These are weighted percentages: answers were adjusted to better reflect the U.S. population. For example, 106 of the 242 drinkers reported drinking less, while the weighted estimate was 44.2%. The percentages are not simple divisions of the head counts.
A positive screen meant reporting at least one heavy-drinking day in the previous year. It did not diagnose alcohol use disorder, meaning difficulty controlling drinking despite harm.
For someone considering semaglutide for weight management, this is a finding about reported experience. It is not a prediction about what their prescription will do.
What a change in drinking can mean during treatment
Less interest in alcohol can be a noticeable change in an ordinary routine. The useful follow-up question is what actually changed: the urge to drink, the amount on each occasion, or how often drinking happened.
The survey’s headline cannot answer those questions for an individual. Nor does “44.2% reported less” mean people drank 44.2% less alcohol. The study did not establish the size of the reduction.
In Harvard’s account of the findings, the authors stressed that they were studying people’s perceptions, rather than testing a treatment. That distinction leaves room for someone’s experience without turning it into a promise for everyone else.
Only about one in ten respondents had heard that these medicines might reduce the urge to drink. That figure covered the whole survey sample, not only people taking GLP-1 medicines. Low awareness is interesting, but it does not establish why drinking changed.
What the survey does not prove about Ozempic or tirzepatide
There was no comparison group randomly assigned to go without the medicine. Without that comparison, researchers cannot separate a drug effect from other changes happening at the same time.
Someone starting weight treatment might also change meals, social habits, or goals around alcohol. Those are possible explanations, not reasons the survey measured. Remembering earlier drinking can also be imperfect, and people may report changes they feel good about.
People can notice a real change in their routine. Research still needs a comparison to show how much credit belongs to a medicine.
The paper also did not report drinking outcomes separately for semaglutide and tirzepatide, a different weight-management medicine. It cannot show whether either medicine changes drinking more than the other. The survey therefore supplies no basis for choosing between them to reduce alcohol use.
For someone already discussing tirzepatide for weight management, the question remains how that medicine fits their health history. The alcohol headline cannot make that choice.
Drinking less is separate from drinking safely
Reduced interest in alcohol does not establish that drinking during treatment is safe. The practical questions belong in the separate guide to semaglutide and alcohol.
Ozempic’s FDA prescribing information, “Revised: 05/2026,” warns about low blood sugar with insulin, a hormone that lowers blood sugar. The warning also covers a sulfonylurea, a diabetes medicine that increases insulin release. It also warns about pancreatitis, inflammation of the pancreas. The survey does not tell us how alcohol changes those risks.
Ozempic's label says the medicine should be stopped promptly and the prescriber contacted if pancreatitis is suspected. Its Medication Guide says to stop using it and call a healthcare provider right away for severe stomach-area pain that will not go away, with or without vomiting; the pain may be felt from the abdomen to the back. That is a label-based concern, separate from whether alcohol seems less appealing.
What this means for a semaglutide prescription
GLP-1 medicines are not FDA-approved for alcohol use disorder. Ozempic’s label lists diabetes and related heart and kidney indications, not alcohol treatment. NIAAA’s treatment guide describes counseling and medicines such as naltrexone, acamprosate, and disulfiram as existing alcohol-treatment options.
Semaglutide and tirzepatide prescribed through Promise are compounded medications, meaning a pharmacy prepares them for an individual prescription. A licensed U.S. compounding pharmacy dispenses them. They are different from an FDA-approved product: the formulations offered here are not FDA-approved. A licensed provider may still prescribe — that decision is between you and your doctor.
The survey did not compare compounded preparations with Ozempic or establish equivalent effects. A change in drinking is something a prescriber can discuss alongside the original reason for treatment.
What to watch next
Controlled trials ask a different question from a survey: what changes when people are assigned a medicine or placebo, an inactive comparison treatment?
A 48-person trial in JAMA Psychiatry, 2025, found differences in some drinking measures but not drinking frequency. The separate semaglutide alcohol use disorder trial explainer covers that study and later trials.
As of November 5, 2026, the day this article was written, the VA’s CRAVE trial record lists recruiting status and 622 planned participants. Its last update was posted September 21, 2026; no results are posted. Larger controlled studies can clarify who experiences a change, how large it is, and whether it lasts.
At Promise, a licensed provider reviews every request; not everyone qualifies. Drinking history belongs in that review, alongside current medicines and the reason for considering weight treatment.
