GLP-1 drugs linked to lower addiction risk in large VA study

Diabetes medications including Ozempic and Mounjaro were associated with meaningfully lower risks of developing substance use disorders — and of dying from them — in a large analysis of Veterans Affairs health records published Wednesday in The BMJ. The study, drawing on data from more than 600,000 VA patients with diabetes over three years, found GLP-1 receptor agonists reduced addiction risk across alcohol, cannabis, cocaine, nicotine and opioids compared with other blood-sugar-lowering medications.

For people already struggling with addiction, starting a GLP-1 drug was linked to a 50% lower risk of death, a 39% lower risk of overdose and a 25% lower risk of suicidal thoughts or attempts, compared with patients on other diabetes medications, according to the study.

The findings suggest the same brain pathways that GLP-1 drugs suppress to curb food cravings may also dampen compulsive urges underlying substance use — raising the prospect of a new pharmacological tool against addiction, though researchers said randomized controlled trials are needed before the drugs could be prescribed for that purpose.

How the study was designed

Dr. Ziyad Al-Aly, the study’s lead author and a chief researcher at the VA St. Louis Health Care System, and his colleagues analyzed electronic health records from more than 600,000 VA patients with diabetes over three years. The researchers divided participants into seven parallel trials, comparing people who received GLP-1 drugs with those treated with other medications that lower blood sugar, and examined risks of developing addiction to alcohol, cannabis, cocaine, nicotine and opioids. A separate trial tracked specific harms among patients who already had substance use disorders.

Previous studies had suggested GLP-1 drugs could reduce addictive behavior by targeting the brain’s reward pathways, but those studies were small and often limited to a single substance, according to the report.

What the researchers found

Compared with patients on other diabetes medications, those taking GLP-1 drugs showed reduced risk of developing addiction across multiple substances: 18% lower for alcohol, 14% for cannabis, 20% for both cocaine and nicotine, and 25% for opioids, Al-Aly said.

Among patients who already had substance use disorders, starting a GLP-1 drug was linked to a 31% lower risk of emergency department visits, 26% lower risk of hospitalizations, 25% lower risk of suicidal thoughts or attempts, 39% lower risk of overdose and 50% lower risk of death.

Overall, GLP-1 drug use likely prevented about seven cases of substance use disorder and 12 incidents involving serious harm for every 1,000 users over three years, Al-Aly said.

“They’re actually working against the root cause of all these different addictions,” Al-Aly said.

What outside experts said

Dr. Lorenzo Leggio, a National Institute on Drug Abuse clinical director who was not involved in the study, said the results pointed to the GLP-1 system addressing addiction at a foundational level.

“Even though we don’t fully understand the mechanism, somehow the GLP-1 system is tackling addiction biology and the foundational system that underlies all these disorders,” Leggio said.

Previous research has shown GLP-1 drugs target hormones in the gut and brain that control appetite and feelings of fullness, cutting down on what is described as “food noise,” or intrusive thoughts of food. The new study indicates the drugs may similarly reduce “alcohol or drug noise,” Leggio said.

Dr. Anna Lembke, a Stanford University addiction medicine specialist, said the addiction treatment field has gone without new pharmacological options for a long time.

“We haven’t really had a new tool in our toolbox from a pharmacotherapy perspective to treat addiction in a long time,” Lembke said, noting that some addiction specialists are already prescribing GLP-1s off-label, especially when other treatments have failed.

Lembke cautioned that GLP-1 drugs do not work the same way for all users and carry risks that must be weighed against potential benefits.

Study limitations and what comes next

The analysis carries notable limitations. It was conducted within the VA health system, which serves a population that is predominantly older, white and male, though Al-Aly said results were consistent among more than 35,000 women in the dataset. The study drew only on patients with diabetes, not the general population, and researchers could not account for factors such as socioeconomic status or lifestyle choices that could affect results.

As an observational study, the analysis showed an association between GLP-1 drugs and reduced substance use disorder risk — not that the drugs themselves caused the reduction.

Al-Aly said the new findings do not, by themselves, justify prescribing GLP-1 drugs to prevent or treat substance use disorders. That evidence would need to come from randomized controlled clinical trials that directly compare the drugs against a placebo. Several such trials are pending, Leggio noted.

More than 48 million Americans have substance use disorders, according to the study. “The consequence in terms of chronic disease of these addictive drugs is actually gigantic in our society,” Leggio said.