Large Cohort Study in JAMA Psychiatry Links GLP-1 Drugs to Lower Mortality in People With Serious Mental Illness

A study published online August 26, 2026 in JAMA Psychiatry found lower all-cause mortality and fewer cardiovascular events among adults with serious mental illness who started a GLP-1 receptor agonist compared with those who started an SGLT2 inhibitor, in one of the largest real-world comparisons to date of the two drug classes in a psychiatric population. The paper, titled "Glucagon-Like Peptide 1 Receptor Agonists, Mortality, and Cardiovascular Outcomes in Serious Mental Illness," was authored by Roger S. McIntyre of the University of Toronto, Yanli Zhang-James of SUNY Upstate Medical University, and Angela T. H. Kwan of the University of Toronto.

The study is a retrospective target trial emulation conducted within the TriNetX Analytics Network, a multinational federated electronic health record database. Researchers used a new-user, active-comparator design with propensity score matching to compare adults who initiated a GLP-1 receptor agonist against adults who initiated an SGLT2 inhibitor, building separate matched cohorts for people with and without a serious mental illness, defined as bipolar disorder, major depressive disorder, or schizophrenia. Data were queried in March 2026, with prespecified follow-up intervals of 1, 4, and 10 years, and additional subgroup analyses were conducted following peer review in May 2026.

For the primary 4-year analysis, 1,528,230 adults were propensity score matched into 764,115 pairs, including 195,184 pairs with serious mental illness and 568,931 pairs without. Among participants with serious mental illness, 4-year all-cause mortality was lower with GLP-1 receptor agonist initiation than with SGLT2 inhibitor initiation: 9,585 of 195,184 people (4.91%) versus 12,584 of 195,184 people (6.45%), a hazard ratio of 0.76 (95% CI 0.74 to 0.78) and an absolute risk difference of -1.54 percentage points (95% CI -1.68 to -1.39). In a separately matched 1-year cohort of 198,065 pairs with serious mental illness, mortality was 2,898 of 198,065 (1.46%) with GLP-1 receptor agonists versus 5,624 of 198,065 (2.84%) with SGLT2 inhibitors, a relative risk of 0.52 (95% CI 0.49 to 0.54).

Among participants with serious mental illness who also had type 2 diabetes, semaglutide initiation was associated with a lower risk of a 3-point composite of major adverse cardiovascular events than SGLT2 inhibitor initiation (hazard ratio 0.77, 95% CI 0.76 to 0.79), along with lower risks of a broader 5-point composite, myocardial infarction, stroke, heart failure, and coronary artery bypass grafting. In exploratory 10-year analyses restricted to participants with type 2 diabetes, semaglutide initiation was associated with lower mortality than SGLT2 inhibitor initiation across all three serious-mental-illness subgroups: major depressive disorder (relative risk 0.55, 95% CI 0.53 to 0.56), bipolar disorder (relative risk 0.57, 95% CI 0.51 to 0.63), and schizophrenia (relative risk 0.67, 95% CI 0.59 to 0.75).

The authors write that the mortality association was evident within one year, held across prespecified sensitivity analyses that included diabetes stratification, exclusion of participants with baseline heart failure or chronic kidney disease, and censoring at treatment crossover, and was consistent across serious-mental-illness subgroups. They conclude that GLP-1 receptor agonist initiation was associated with lower mortality and cardiovascular events compared with SGLT2 inhibitor initiation, with greater absolute reductions among people with serious mental illness, and that the effect appeared to be driven largely by semaglutide and tirzepatide. The paper states plainly that "prospective randomized trials are warranted."

The study is observational and retrospective, drawing on electronic health record data rather than a randomized controlled trial, so it cannot establish that GLP-1 receptor agonists caused the lower mortality and cardiovascular event rates observed; people who are prescribed a GLP-1 receptor agonist may differ from those prescribed an SGLT2 inhibitor in ways that were not fully captured by propensity score matching. Semaglutide and tirzepatide remain approved by the FDA, the European Medicines Agency, and Australia's Therapeutic Goods Administration for their existing indications, and this study does not change that regulatory status or expand it to serious mental illness as an indication. Nothing in the study constitutes treatment or dosing guidance, and people with bipolar disorder, major depressive disorder, or schizophrenia who have questions about their medications should consult a health professional.