Semaglutide and Alcohol: What the 2026 Lancet Trial Really Found

A larger clinical trial has moved the conversation about semaglutide and alcohol beyond unusual patient stories. Published in The Lancet in 2026, the study found that adults receiving weekly semaglutide had fewer heavy-drinking days than those receiving placebo. The finding is important—but it is also easier to exaggerate than it first appears.

The result in plain English: 108 treatment-seeking adults with obesity and moderate-to-severe alcohol-use disorder were assigned to semaglutide or placebo for 26 weeks. Both groups also received cognitive behavioural therapy. Heavy-drinking days fell in both groups, but they fell significantly more with semaglutide. This strengthens the case for further research; it does not make Wegovy or Ozempic an approved alcohol-dependence treatment.

The semaglutide alcohol trial 2026 matters because it was longer and more clinically relevant than earlier experimental work. Participants were not merely curious volunteers completing a laboratory drinking session. They had sought treatment for alcohol-use disorder and received psychological therapy alongside the injection.

For a patient in Cardiff, Birmingham or London reading headlines about a “weight-loss jab for alcoholism”, however, several questions remain. How large was the difference? Did semaglutide work independently of therapy? Would it help someone without obesity? What happens after the injections stop? And can it be prescribed for alcohol cravings on the NHS?

What was the 2026 semaglutide alcohol trial?

The study was a phase 2 randomised, double-blind, placebo-controlled trial conducted at the Mental Health Centre Copenhagen in Denmark. It was led by Mette Kruse Klausen and colleagues and published in the 2 May 2026 issue of The Lancet.

Researchers enrolled 108 adults between June 2023 and February 2025. Participants were assigned equally to one of two groups:

54 received once-weekly subcutaneous semaglutide, increasing towards 2.4 mg; and
54 received a placebo injection containing saline.

Neither the participants nor the clinical research team knew which treatment an individual had received during the trial. This double-blind design helps reduce the risk that expectations influence reporting or care.

Crucially, both groups were offered standard cognitive behavioural therapy. The comparison was therefore not semaglutide versus doing nothing. It was semaglutide plus CBT versus placebo plus CBT.

108participants randomised

26 weekstreatment period

88completed the full intervention

Who took part in the study?

Participants were aged between 18 and 70 and had moderate-to-severe alcohol-use disorder. They also had obesity, defined for the trial as a body mass index of at least 30 kg/m². Importantly, they wanted treatment for their drinking.

The group was almost evenly divided by sex: 53 women and 55 men. Their average age was about 52. At the beginning of the study, participants reported approximately 17 heavy-drinking days during the previous 30 days.

These details define who the findings directly apply to. The study does not prove that semaglutide has the same effect in:

people who drink heavily but do not meet criteria for alcohol-use disorder;
people with alcohol dependence who do not have obesity;
people who are not seeking treatment;
younger or much older populations; or
patients using the medicine without psychological support.

That does not make the findings unhelpful. It prevents a carefully selected trial population from being mistaken for every person who drinks alcohol.

What did the trial find?

The primary outcome was the reduction in heavy-drinking days after 26 weeks. Heavy-drinking days fell by 41.1 percentage points from baseline in the semaglutide group and by 26.4 percentage points in the placebo group.

After statistical adjustment, the estimated difference between the groups was 13.7 percentage points. The 95% confidence interval ranged from 5.4 to 22.0 percentage points, and the p-value was 0.0015. In statistical terms, the result was unlikely to be explained by random variation alone.

A more intuitive version: participants began with roughly 17 heavy-drinking days per month. By six months, reports accompanying the study described an average of approximately five heavy-drinking days in the semaglutide group and nine in the placebo group. Both groups improved substantially; the semaglutide group improved more.

Semaglutide was also associated with improvements across several secondary outcomes, including total alcohol consumption, drinks per drinking day, alcohol craving and World Health Organization drinking-risk levels. Biomarkers and metabolic measures provided additional evidence that the change was not limited to questionnaire responses.

Outcome	What happened	How to interpret it
Heavy-drinking days	Fell more with semaglutide than placebo.	This was the trial’s primary result and provides the strongest evidence of benefit.
Total alcohol consumption	Declined in both groups, with a greater reduction in the semaglutide group.	CBT and treatment-seeking probably helped both groups; semaglutide appeared to add an effect.
Drinks per drinking day	Improved more with semaglutide.	The medicine may have affected how much people consumed once drinking began.
Alcohol craving	Reduced with semaglutide.	This supports a possible effect on reward or motivational pathways.
Body weight	Fell substantially more with semaglutide.	Expected in people with obesity, but it complicates separating metabolic and addiction effects.
Side effects	Gastrointestinal effects occurred more often with semaglutide.	The benefit was not free of treatment burden, although events were generally temporary and mild to moderate.

Did semaglutide cut heavy drinking by 41%?

This headline is too crude. The study reported a 41.1 percentage-point reduction from baseline within the semaglutide group. It did not show that the medicine alone reduced heavy drinking by 41% compared with receiving no treatment.

The placebo group—whose participants also wanted help and received CBT—improved by 26.4 percentage points. The estimated treatment difference attributable to assignment was 13.7 percentage points.

There are at least three reasons why both groups may have improved:

Participants had actively sought treatment and were motivated to change.
Everyone was offered cognitive behavioural therapy.
Regular appointments, monitoring and the structure of a trial can alter behaviour.

This does not weaken the semaglutide result. A treatment showing an additional benefit on top of active psychological care is clinically interesting. It simply gives the reader an honest comparison.

Why cognitive behavioural therapy matters

CBT helps people recognise thoughts, emotions and situations linked with drinking and develop alternative responses. A participant might notice that Friday evenings, loneliness or conflict reliably precede alcohol use. Therapy then turns that observation into a practical plan.

If semaglutide makes cravings quieter, CBT may help a person use that quieter period to change entrenched routines. Conversely, a medicine cannot resolve every reason someone drinks. Work stress in Leeds, social isolation in rural Wales or a long-established pub culture within a friendship group does not disappear because reward signalling changes.

The trial therefore tested something closer to real addiction care: medication alongside psychological treatment. It did not support the idea of obtaining Wegovy online and using it as a solitary cure.

How might semaglutide affect drinking?

Semaglutide imitates some actions of glucagon-like peptide-1, or GLP-1. The hormone is involved in blood-glucose control, digestion and satiety. GLP-1 receptors are also present in areas of the brain involved in motivation and reward.

Preclinical studies suggest GLP-1 receptor stimulation can reduce alcohol intake and relapse-like behaviour. Several mechanisms may contribute in humans:

alcohol may feel less rewarding;
cues associated with drinking may provoke less craving;
a wider feeling of satiety may reduce the urge to continue;
slower gastric emptying may change the experience of drinking; and
nausea or reflux may make alcohol less appealing.

The Copenhagen trial shows a clinical effect but does not settle precisely how it arose. It is possible that brain reward effects, weight loss, physical side effects and behavioural treatment interacted.

Could the results simply be caused by nausea?

Gastrointestinal effects were more common with semaglutide. They were described as temporary and generally mild to moderate. Five participants discontinued because of adverse effects, four of whom were in the semaglutide group.

Feeling sick could certainly discourage drinking. Yet the overall research picture—including animal studies, craving measures and previous human work—suggests that nausea is unlikely to explain everything. A future treatment would still need to demonstrate that reduced drinking is not merely the product of persistent discomfort.

The NHS advises that it is best not to drink alcohol while using semaglutide because alcohol may increase side effects such as nausea and vomiting. Anyone taking semaglutide should follow advice from their own prescriber rather than using the trial to experiment with alcohol.

What happened to participants’ weight?

Participants receiving semaglutide lost substantially more weight than those receiving placebo—approximately 11 kg on average in reports of the full trial results. Waist circumference, BMI and blood-glucose measures also improved.

This creates both an opportunity and a scientific complication. Alcohol-use disorder and obesity can coexist, and alcohol supplies considerable energy. One medicine potentially addressing drinking, weight and metabolic risk would be valuable.

At the same time, the trial cannot tell us whether reduced alcohol use was independent of weight loss. The investigators reported a relationship between weight reduction and the fall in heavy-drinking days within the semaglutide group. That may reflect shared biological mechanisms, reduced alcohol calories, or several changes occurring together.

What are the trial’s main limitations?

No single phase 2 study should determine clinical practice. The main limitations include:

Only 108 people participated. That is larger than the earlier 48-person semaglutide trial, but still small for establishing routine treatment.
It was conducted at one centre. Results from Copenhagen need replication in other healthcare systems and populations.
All participants had obesity. We cannot assume equal benefit for people with a BMI below 30.
There was no extended post-treatment follow-up. We do not know whether improvements continued after semaglutide stopped.
Twenty participants did not complete the full intervention. The researchers used intention-to-treat analysis and statistical methods for missing data, but attrition remains relevant.
Side effects may reveal treatment assignment. Recognisable nausea or appetite changes can make perfect blinding difficult.
CBT was part of treatment. This is clinically appropriate but means the results should not be applied to unsupervised medication use.

The most important unknown: we do not know whether alcohol cravings and heavy drinking return after semaglutide is discontinued. The trial ended after 26 weeks without a longer off-treatment comparison.

Would semaglutide work without obesity?

The trial cannot answer that question. Every participant had a BMI of at least 30. Earlier research suggests GLP-1 medicines may affect alcohol reward directly, but some studies have found clearer benefits in participants with obesity.

People without obesity could experience a different balance of benefit and harm. Unwanted weight loss would matter, and the most suitable addiction dose may not be the dose used for weight management. A future trial would need to recruit people across a wider BMI range and monitor nutrition carefully.

How does this compare with the 2025 trial?

The 2025 JAMA Psychiatry trial was an important first step. It randomised 48 adults with alcohol-use disorder who were not seeking treatment and used lower doses of semaglutide over nine weeks. Semaglutide reduced laboratory alcohol consumption, weekly craving and some—but not all—real-world drinking measures.

The 2026 Lancet trial moved closer to clinical practice:

it recruited treatment-seeking patients;
it lasted 26 rather than nine weeks;
it included 108 rather than 48 participants;
it tested the 2.4 mg target dose; and
it provided CBT to both study groups.

A separate eight-week trial reported in July 2026 also found that oral semaglutide reduced several real-world alcohol outcomes in 50 treatment-seeking adults. Taken together, the studies are beginning to form a consistent signal, although none yet establishes long-term NHS treatment.

Can Wegovy or Ozempic now be prescribed for alcohol dependence?

No. Semaglutide is not currently licensed or routinely recommended by NICE specifically for alcohol-use disorder. Wegovy has authorised uses connected with weight management and certain other eligible groups; Ozempic is used in type 2 diabetes. Sharing an active ingredient does not make the brands interchangeable or turn either product into an approved addiction medicine.

A clinician can sometimes prescribe a medicine off-label, but that requires individual professional judgement and responsibility. Publication of one promising phase 2 trial does not create a standard NHS prescribing pathway.

British research is continuing. A four-year project involving Plymouth and Nottingham is studying semaglutide in people with alcohol dependence, obesity and alcohol-related liver disease. This should provide evidence that is more directly relevant to NHS care in the UK.

Related guides:

What alcohol treatment is available in England and Wales now?

Current evidence-based care can include psychological treatment, a medically assisted withdrawal when needed, and relapse-prevention medicines. NICE recommends considering acamprosate or oral naltrexone with an individual psychological intervention for suitable patients following withdrawal. Disulfiram may be considered in selected situations.

In England, a GP or local community alcohol service can provide an assessment or referral. Many services allow self-referral. In Wales, DAN 24/7 provides confidential information and help finding local support on 0808 808 2234, at any time.

Alcohol-withdrawal warning: do not suddenly stop or sharply reduce alcohol if you may be physically dependent. Morning shaking, sweating, retching, marked anxiety relieved by drinking, hallucinations or a previous withdrawal seizure require medical advice. Call 999 for a seizure, severe confusion, hallucinations or severe shaking.

Frequently asked questions

Did the 2026 trial prove semaglutide treats alcoholism?

No. It provided strong phase 2 evidence that semaglutide can reduce heavy-drinking days in treatment-seeking adults with obesity and alcohol-use disorder. Larger, multicentre and longer-term trials are still needed.

How many heavy-drinking days did participants have?

Participants began with around 17 heavy-drinking days in the previous month. By 26 weeks, accompanying reports described averages of roughly five days with semaglutide and nine with placebo.

Did everyone in the semaglutide group stop drinking?

No. The main finding was a reduction in heavy-drinking days, not universal abstinence. Reducing harmful drinking can still be clinically meaningful.

Was semaglutide used on its own?

No. Both the semaglutide and placebo groups were offered cognitive behavioural therapy. The trial tested whether semaglutide added benefit to structured treatment.

Were participants taking Ozempic or Wegovy?

The study tested once-weekly subcutaneous semaglutide titrated towards 2.4 mg, rather than evaluating branded products as consumer treatments. It should not be interpreted as permission to use Ozempic or Wegovy off-label without clinical supervision.

Did semaglutide reduce alcohol cravings?

Yes, craving was among the secondary alcohol-related outcomes that improved. The primary result, however, was the change in heavy-drinking days.

What were the common side effects?

Gastrointestinal effects such as nausea, vomiting and reflux were more common with semaglutide. They were generally temporary and mild to moderate, although several participants stopped treatment because of adverse effects.

Will the benefits last after stopping semaglutide?

We do not know. The absence of longer off-treatment follow-up is one of the study’s most important limitations.

Can someone without obesity use semaglutide for alcohol cravings?

This trial provides no direct evidence for that group. Every participant had obesity, and semaglutide is not an approved alcohol-dependence treatment.

The honest conclusion

The 2026 Lancet trial is one of the strongest pieces of human evidence so far that GLP-1 medicines may have a future in addiction treatment. Semaglutide did more than help people lose weight: on top of CBT and regular clinical contact, it produced a statistically and potentially clinically meaningful reduction in heavy-drinking days.

The result deserves attention without being turned into a miracle-drug story. The study was small, single-centre and limited to people with obesity. Side effects were more common, and nobody knows whether reduced drinking persists after the injections stop.

For researchers, it is a clear signal to proceed. For NHS policymakers, it is evidence worth watching. For patients and families, it is a reason for cautious hope—not a reason to buy semaglutide online or replace established alcohol treatment.

References

This article provides general health information and does not replace an assessment, diagnosis or treatment from a qualified healthcare professional.