GLP-1 and addiction research has moved from animal studies into real trials, and the clearest signal is for alcohol. In a 2026 Lancet trial of 108 adults with alcohol use disorder and obesity, weekly semaglutide 2.4 mg cut heavy drinking days by 41 percentage points versus 26 points on placebo over 26 weeks.[1] For opioids and nicotine, the evidence is mostly from health records and small pilot trials, and no GLP-1 drug is approved for any addiction.
This page covers the research. For everyday questions about drinking on these medicines, see our guide to GLP-1 drugs and alcohol. For the wider picture, start at the GLP-1 basics hub.
Bottom line
| Substance | Best evidence so far | Result | Strength |
|---|---|---|---|
| Alcohol | Randomized trials of semaglutide (3 trials, 48-108 people) | Fewer heavy drinking days or less drinking per occasion | Moderate, growing |
| Alcohol | Exenatide trial, 127 people | No overall effect on heavy drinking days | Mixed |
| Opioids | Health-record cohort, about 500,000 people | Lower overdose rates with GLP-1 prescriptions | Weak (observational) |
| Nicotine | Small trials of exenatide, dulaglutide and semaglutide | Mixed: less craving, but quit rates not clearly better | Weak to mixed |
Why researchers think GLP-1 drugs might affect addiction
GLP-1 is a gut hormone, but its receptors also sit in brain reward circuits. That is the same system that drives cravings for alcohol and drugs. In a 26-week exenatide trial, brain scans showed that the drug weakened responses to alcohol cues in the ventral striatum, a key reward area.[4] Many people taking semaglutide or tirzepatide for weight also describe less interest in drinking or snacking. That is part of why the same “quieter cravings” idea comes up in our page on GLP-1 drugs and emotional eating.
Alcohol: the key trials
SEMALCO (Lancet, 2026). This Danish trial enrolled 108 treatment-seeking adults with moderate to severe alcohol use disorder and obesity. Everyone got cognitive behavioral therapy plus weekly semaglutide 2.4 mg or placebo for 26 weeks. Heavy drinking days fell 41.1 percentage points from baseline on semaglutide versus 26.4 on placebo, a difference of 13.7 points. Side effects were mostly mild to moderate stomach upset, more common on semaglutide.[1]
Injectable semaglutide, low dose (JAMA Psychiatry, 2025). Forty-eight adults who were not seeking treatment took semaglutide up to 1 mg for 9 weeks. In a lab drinking session, they drank less alcohol than the placebo group. Day to day, drinks per drinking day and craving fell, but the number of drinking days did not change.[2]
Oral semaglutide (Am J Psychiatry, 2026). Fifty treatment-seeking adults took oral semaglutide (3 mg, then 7 mg daily) or placebo for 8 weeks. It did not reduce the primary outcome, lab-measured craving, or drinks per day. It did reduce heavy drinking days, drinks per drinking day and everyday craving.[3]
Exenatide (JCI Insight, 2022). In 127 patients, weekly exenatide 2 mg for 26 weeks did not reduce heavy drinking days overall. An exploratory analysis suggested a benefit in people with obesity, which fits with SEMALCO’s focus on that group.[4]
Observational studies. In a Swedish register of 227,866 people with alcohol use disorder, periods on semaglutide were linked to a 36% lower risk of hospitalization for alcohol problems (hazard ratio 0.64), and liraglutide also showed a lower risk.[5] A US health-record study of 83,825 people with obesity found semaglutide linked to a 50-56% lower risk of a new or repeat alcohol use disorder diagnosis compared with other weight-loss drugs.[6]
Opioids: mostly observational so far
A US study of health records from about 136 health systems included 503,747 people with opioid use disorder. Those with a GLP-1 or GIP/GLP-1 prescription had a 40% lower rate of opioid overdose (adjusted rate ratio 0.60).[7] This is a striking number, but people prescribed these drugs may differ in many ways from those who are not, such as access to care.
No completed large randomized trial has tested a GLP-1 drug for opioid use disorder. A trial of tirzepatide added to buprenorphine, with 310 planned participants, is recruiting.[13]
Nicotine: mixed results
- Health records: among 222,942 people with type 2 diabetes and tobacco use disorder, starting semaglutide was linked to fewer medical visits for tobacco use disorder than starting other diabetes drugs.[8]
- Exenatide pilot trial: in 84 smokers using a nicotine patch, adding exenatide raised 6-week abstinence to 46.3% versus 26.8% on placebo, and post-quit weight was 5.6 pounds lower.[9]
- Dulaglutide trial: in 255 smokers, dulaglutide did not improve quit rates. At 52 weeks, 32% were abstinent in both groups. It prevented weight gain during the 12 weeks of treatment, but that faded later.[10]
- Semaglutide trial: in 24 daily smokers not trying to quit, semaglutide did not change the main lab smoking measures, but it reduced craving over 9 weeks.[11]
The smoking-related reductions seen in the alcohol trial above were also small and exploratory.[2]
What’s established and what’s preliminary
Established:
- Semaglutide reduced heavy drinking in a randomized trial of people with alcohol use disorder and obesity.[1]
- Several observational studies, in different countries, link GLP-1 drugs with fewer alcohol-related problems.[5][6]
Preliminary:
- Whether the effect holds in people without obesity, at lower doses, or after the drug is stopped.
- Any benefit for opioid or nicotine addiction.
- Whether tirzepatide works better or worse than semaglutide.
Limitations
- Trials so far are small and short (8 to 26 weeks). Addiction is a long-term condition.
- Stomach side effects can reveal who is on the active drug, which can weaken blinding.
- Observational studies cannot rule out other explanations, such as healthier or better-insured people being more likely to get these drugs.[7]
- Several trials were funded in part by foundations or researchers with ties to drugmakers. The SEMALCO trial was funded partly by the Novo Nordisk Foundation.[1]
What’s next
ClinicalTrials.gov lists many active studies. The largest is a VA phase 3 trial (CRAVE) testing semaglutide up to 2.4 mg in 622 veterans with alcohol use disorder, with results expected around 2029.[12] Tirzepatide trials for alcohol, opioid and nicotine use are also underway.[13]
Questions to discuss with a clinician
- Do I qualify for a GLP-1 drug for diabetes, weight or heart risk, and would any effect on drinking be a side benefit?
- How does this compare with approved treatments for alcohol use disorder, such as naltrexone or acamprosate?
- What should I watch for if I drink while on the drug? Our GLP-1 and alcohol guide covers low blood sugar and pancreatitis risks.
- Could mood or appetite changes affect me? See our summary of GLP-1 drugs, depression and suicidal thoughts.
If you or someone close to you is struggling with alcohol or drugs in the US, the SAMHSA National Helpline (1-800-662-4357) is free and confidential.
