Glucagon-like peptide-1 receptor agonists and risk of substance use disorders among US veterans with type 2 diabetes: cohort study.
Level 3 - non-randomized controlled study
Retrospective active-comparator cohort study using target trial emulation
PubMed 41781010 · doi:10.1136/bmj-2025-086886
What was done
Target trial emulation using electronic health records from the US Department of Veterans Affairs evaluated 606,434 veterans with type 2 diabetes followed for up to 3 years. Protocol 1 included 524,817 participants without a history of substance use disorders (SUDs) initiating GLP-1 receptor agonists (n=124,001) or SGLT-2 inhibitors (n=400,816) to assess incident alcohol, cannabis, cocaine, nicotine, opioid, other SUDs, and a composite outcome. Protocol 2 included 81,617 participants with pre-existing SUDs initiating GLP-1 receptor agonists (n=16,768) or SGLT-2 inhibitors (n=64,849) to evaluate SUD-related emergency visits, hospital admissions, mortality, overdose, and suicidal ideation/attempts. Cause-specific Cox models used inverse probability weighting.
What was found
Compared with SGLT-2 inhibitor initiators, GLP-1 receptor agonist initiators had lower risk of incident alcohol use disorder (HR 0.82, 95% CI 0.78 to 0.85; net 3-year risk difference [NRD] per 1,000 people -5.57, 95% CI -6.61 to -4.53), cannabis (HR 0.86, 95% CI 0.81 to 0.90; NRD -2.25), cocaine (HR 0.80, 95% CI 0.72 to 0.88; NRD -0.97), nicotine (HR 0.80, 95% CI 0.74 to 0.87; NRD -1.64), opioid (HR 0.75, 95% CI 0.67 to 0.85; NRD -0.86), other SUDs (HR 0.87, 95% CI 0.81 to 0.94; NRD -1.12), and composite SUD (HR 0.86, 95% CI 0.83 to 0.88; NRD -6.61). Among those with pre-existing SUDs, GLP-1 receptor agonist use was associated with reduced risk of SUD-related ED visits (HR 0.69, 95% CI 0.61 to 0.78; NRD -8.92), hospital admissions (HR 0.74, 95% CI 0.65 to 0.85; NRD -6.23), SUD-related mortality (HR 0.50, 95% CI 0.32 to 0.79; NRD -1.52), drug overdose (HR 0.61, 95% CI 0.42 to 0.88; NRD -1.49), and suicidal ideation or attempt (HR 0.75, 95% CI 0.67 to 0.83; NRD -9.95).
Why it matters
This large observational study indicates that GLP-1 receptor agonists are associated with broad reductions in both incident substance use disorders across multiple drug classes and severe clinical complications among patients with established addiction. These findings provide real-world support for ongoing randomized trials exploring GLP-1 agonists as treatments for substance use disorders.
Limits
The study is observational and restricted to US veterans with type 2 diabetes, a population predominantly older and male, limiting generalizability to women, younger cohorts, and non-diabetic populations. Electronic health records rely on clinical diagnostic coding, which may underdetect substance use or miss events occurring outside the Veterans Affairs health system.
Cited by
- supports GLP-1 receptor agonists suppress non-food cravings and compulsive behaviors, including cannabis use, video gaming, and alcohol consumption.