A Landmark BMJ Study Unveils Promise of Obesity Drugs in Addiction Treatment
In a groundbreaking analysis published in the British Medical Journal (BMJ), researchers from Washington University School of Medicine and the VA St Louis Health Care System examined data from over 606,000 US veterans with type 2 diabetes. The study compared those starting glucagon-like peptide-1 receptor agonists (GLP-1RAs)—commonly known as obesity drugs like semaglutide (Ozempic, Wegovy) and tirzepatide (Mounjaro)—against those on sodium-glucose cotransporter-2 inhibitors (SGLT-2is), another diabetes medication class. The findings suggest GLP-1 agonists may significantly lower risks of developing substance use disorders (SUDs) and improve outcomes for those already struggling with addiction.
This research, highlighted in Nature, addresses a critical intersection of the US obesity epidemic—affecting about 42% of adults—and addiction, where roughly 14% of veterans report SUDs. Veterans, often facing higher obesity rates (up to 80% in some VA cohorts) and SUD prevalence (11-53% lifetime), represent a high-risk group. The study's scale and rigorous design position it as a pivotal moment for obesity drugs addiction treatment research.
Lead investigator Ziyad Al-Aly noted the consistency across substances as "quite a revelation," pointing to GLP-1 agonists' potential as broad-spectrum tools beyond weight management.
Study Design: Emulating Randomized Trials with Real-World Veteran Data
The researchers emulated eight target trials using VA electronic health records, ensuring new-user active-comparator design to minimize bias. Protocol 1 targeted incident SUDs in 524,817 veterans without prior history, followed for up to three years (median 3 years, 1.35 million person-years). Protocol 2 assessed adverse outcomes in 81,617 with pre-existing SUDs (205,000 person-years).
Initiators were balanced via inverse probability weighting for 219+ covariates (demographics, comorbidities, labs, time-varying factors). Outcomes used validated ICD-10 codes for alcohol, cannabis, cocaine, nicotine, opioids, and other SUDs. Adherence analyses (51% GLP-1 at three years) confirmed directionally similar results.
This methodology—cause-specific Cox models with standardized mortality ratio weighting—provides robust relative effect estimates, though observational nature limits causality claims.
Washington University experts like Al-Aly emphasize such large-scale pharmacoepidemiology as key to uncovering GLP-1 agonists substance use disorders links.
Striking Reductions in New Substance Use Disorders Among At-Risk Veterans
For veterans without SUD history, GLP-1RA initiation cut composite SUD risk by 14% (HR 0.86, 95% CI 0.83-0.88; net risk difference [NRD] -6.61 per 1000 over three years). Specifics included:
- Alcohol use disorder: 18% lower risk (HR 0.82, NRD -5.57/1000)
- Opioid use disorder: 25% lower (HR 0.75, NRD -0.86/1000)
- Cocaine: 20% lower (HR 0.80, NRD -0.97/1000)
- Cannabis: 14% lower (HR 0.86, NRD -2.25/1000)
- Nicotine: 20% lower (HR 0.80, NRD -1.64/1000)
Subgroups showed consistency across age, sex (though 90% male cohort), race (20% Black), BMI, and HbA1c. Semaglutide drove most effects, reflecting its prevalence.
These preventive signals challenge siloed addiction treatments, hinting obesity drugs addiction treatment could transform public health.
Improved Outcomes for Veterans Battling Existing Addictions
Among those with SUDs, GLP-1RAs slashed adverse events: 31% fewer SUD-related ED visits (HR 0.69, NRD -8.92/1000), 26% fewer hospitalizations (HR 0.74, NRD -6.23), 50% lower SUD mortality (HR 0.50, NRD -1.52), 39% fewer overdoses (HR 0.61), and 25% less suicidal ideation/attempts (HR 0.75). Composite adverse outcomes dropped 29% (HR 0.71).
By SUD type: alcohol (HR 0.72, NRD -22.49), opioids (HR 0.70, NRD -30.57). This suggests therapeutic potential in high-burden populations like veterans, where SUDs contribute to 2,421 alcohol-involved overdose deaths (2012-2018).
Biological Mechanisms: Targeting the Brain's Reward Pathways
GLP-1RAs cross the blood-brain barrier, binding receptors in the mesolimbic dopamine system—key to reward and craving. They dampen drug-induced dopamine surges, reduce reinstatement (relapse model), potentiate GABA inhibition, modulate serotonin/glutamate, and offer neuroprotection via anti-inflammation.
Preclinical rodent studies confirm reduced seeking for alcohol, cocaine, opioids, nicotine. Human fMRI shows blunted reward responses. Rapid onset aligns with patient reports of quelled "food noise" extending to substances.
Endocrinologist Daniel Drucker (University of Toronto) affirms patient anecdotes: "I don’t feel like I want to smoke anymore." This multitarget action explains broad efficacy.
Photo by Frankie Cordoba on Unsplash
Building on Prior Research: From Small Trials to Veteran Cohorts
Earlier work: Swedish study (227,000 AUD patients) showed 36% lower alcohol hospitalization with GLP-1s; semaglutide cut cannabis/opioid relapse (Nature Mental Health). Small RCTs like Hendershot's (JAMA Psych) reduced AUD drinking; Leggio's ongoing semaglutide AUD trial.
WashU's prior mapping (Nat Med 2025) linked GLP-1s to lower dementia/addiction risks. Anecdotes from clinics fueled interest, now validated at scale.
Link to academic career advice for pharmacoepidemiologists driving such discoveries.
Ongoing Trials: Seeking Causal Proof for GLP-1 in Addiction
Over a dozen RCTs underway: Eli Lilly's tirzepatide for AUD (2026 results); semaglutide for opioids/AUD (NCT05891587, NCT07218354); exenatide extensions. Brown University's Haass-Koffler calls for trials confirming responders.
NIDA's Leggio: GLP-1s could reach underserved via diabetes pathways, but not replace naltrexone yet. Off-label use grows; FDA approval pending robust data.
Explore ClinicalTrials.gov for updates.Higher Education's Role: Universities Fueling Addiction Research Breakthroughs
Washington University leads, with Al-Aly's team exemplifying interdisciplinary pharmacoepidemiology. Ties to VA foster real-world data science. Such studies attract funding, training postdocs in big data analytics for SUD.
Universities like U Toronto (Drucker) pioneer GLP-1 neurobiology. Aspiring researchers: Explore research assistant jobs or postdoc positions in addiction neuroscience.
Implications for med schools: Curriculum shifts toward GLP-1 agonists substance use disorders integration.
Caveats: Observational Limits and Path to Confirmation
Strengths: Scale, emulation, weighting. Limitations: VA cohort (older males), residual confounding, underdiagnosis, no absolute vs placebo. No differential GLP-1/SGLT-2 SUD coding bias assumed.
Experts urge RCTs; Al-Aly: Not for addiction alone yet. Side effects (GI, pancreatitis) balance needed. Generalizability beyond veterans/T2D unclear.
US Context: Dual Epidemics of Obesity and Veteran SUDs
Obesity hits 42% US adults, 80% veterans; SUDs affect 14% veterans (higher than civilians). Opioid crisis: 80k deaths/year; alcohol: 140k. Veterans: 11% first VA visit SUD; co-morbidities amplify risks.
GLP-1s, prescribed to 85% diabetics vs <2% AUD, offer accessible intervention. Ties to professor salaries in public health driving policy.
Photo by Frankie Cordoba on Unsplash
Expert Views and Road Ahead for Obesity Drugs in Addiction
Al-Aly: "Nothing like that in our armamentarium." Leggio (NIDA): Reach via diabetes care. Haass-Koffler: Exciting, destigmatizing. Drucker: Real responders, need RCTs.
Future: Trials, combo therapies, FDA nods. Academia: More clinical research jobs.
Insights for Researchers, Clinicians, and Future Careers
This positions GLP-1 agonists as preventive/therapeutic pillars. Clinicians: Discuss with comorbid patients. Researchers: Pursue mechanisms, equity.
Explore opportunities at higher-ed-jobs, university-jobs, rate-my-professor for mentors, or higher-ed-career-advice. Check recruitment for roles advancing obesity drugs addiction treatment.
