GLP-1 drugs and addiction evidence

A 2026 systematic review found signals that GLP-1 receptor agonists, including semaglutide, might be associated with lower substance-related harm. Most of the evidence was observational, and the randomized trials did not give one consistent answer^[1]. These medicines are not established treatments for substance use disorders on the strength of this review.

What the review included

The authors followed PRISMA methods and searched five research databases through March 2026. From 1,641 records, they included 11 publications that examined effectiveness, safety, or both. The set mixed randomized controlled trials with observational analyses, so its results do not all carry the same ability to test cause and effect^[2].

A randomized trial assigns an intervention and can reduce baseline differences between groups. An observational study follows care or records that already exist. It can reveal a useful association in a large population, yet treatment selection and health differences may explain part of that association even after statistical adjustment.

The review covered several substances and several kinds of outcome. Some analyses asked whether a diagnosis was recorded. Others tracked hospitalization, overdose, mortality, or changes in use. Those are not interchangeable endpoints. A reduction in a billing-code diagnosis may reflect care patterns, while an overdose event has a different clinical meaning. The authors therefore synthesized a small and varied evidence base rather than one repeatable treatment experiment^[1].

The signals differed by substance

For alcohol use disorder, the review reported lower hospitalization risk in observational data, with adjusted hazard ratios between 0.64 and 0.72. Estimates for incident alcohol use disorder also favored GLP-1 treatment in some analyses. The randomized evidence was mixed: one result was not statistically significant at p = 0.37, while another reported a 29% reduction with p = 0.04^[1]. Mixed findings call for restraint, not an average-sounding promise.

The review also reported that semaglutide was associated with a lower recorded risk of tobacco use disorder, with a hazard ratio of 0.68 and a 95% confidence interval from 0.63 to 0.74. Observational studies of opioid outcomes reported lower overdose risk, with hazard ratios from 0.32 to 0.58^[2]. A hazard ratio compares event rates over time between groups. It does not mean that a medicine reduced an individual's risk by a fixed percentage.

For broader substance use disorder outcomes, the abstract reports associations with lower incidence, mortality, and suicidal ideation. Those outcomes are clinically different and may be recorded differently across health systems. Combining them into one narrative can make the evidence look more settled than it is^[1]. Each needs confirmation in a study designed around that outcome.

Why association is not treatment proof

People prescribed GLP-1 medicines may differ from comparison groups in diabetes care, weight management, healthcare contact, or socioeconomic circumstances. Some may receive more frequent follow-up. Others may have different diagnoses recorded. Statistical adjustment helps only when the relevant factor is measured well enough to include.

The reviewers rated the randomized studies as low risk of bias and the observational studies as moderate risk. They still concluded that stronger randomized trials are needed before causality or clinical use can be confirmed^[2]. That conclusion is the main boundary around the promising estimates.

Safety needs disorder-specific study

The review found no pattern of increasing serious adverse events across the included studies^[1]. That does not establish safety for every substance use disorder population. Trials may be small or short, and people with unstable medical or psychiatric illness can be excluded. Interactions with intoxication, withdrawal, poor nutrition, or other prescribed medicines require direct assessment.

GLP-1 receptor agonists already have approved metabolic indications, with product-specific warnings and adverse effects. The semaglutide evidence overview provides that background. A signal in addiction research does not create a new approved indication and should not be treated as a reason to use someone else's prescription.

What research must answer next

Future trials need a defined substance use disorder, a clear clinical endpoint, and enough follow-up to measure relapse as well as harm. They should report current treatments and psychiatric care rather than treating GLP-1 exposure as the only meaningful difference. Enrollment also needs to reflect the populations who would receive addiction care outside specialist research centres.

For now, the review supports investigation. It does not support replacing medications already shown to reduce harm in alcohol or opioid use disorders, and it does not show that semaglutide treats addiction. Anyone seeking help with substance use should contact a qualified clinician or addiction service that can discuss evidence-based care and urgent safety needs.

Medical disclaimer: This article is for educational and journalistic purposes only and does not constitute medical advice. Peptides discussed may be classified as prescription medicines or research chemicals depending on your jurisdiction. Always consult a qualified healthcare professional before using any peptide product. PeptideMethods.com does not sell, distribute, or facilitate the sale of any peptide product.