S735 Psychiatric Risk in Obese vs Diabetic Non-Obese Patients on GLP-1 Receptor Agonists: A Retrospective Cohort Study

Introduction: Glucagon-like peptide-1 receptor agonists (GLP-1RAs) have revolutionized the treatment of obesity and type 2 diabetes, offering both metabolic and cardiovascular benefits. However, emerging data have raised concerns about potential neuropsychiatric effects, including mood disturbances and suicidality. While these risks have drawn increasing scrutiny, it remains unclear whether psychiatric outcomes differ based on underlying metabolic phenotype. Methods: We conducted a retrospective cohort study using the TriNetX research platform. Adult patients (≥18 years) prescribed GLP-1RAs (liraglutide or semaglutide) were stratified into 2 cohorts: obese patients (body mass index [BMI] ≥30) and diabetic non-obese patients (BMI <30). Individuals with prior psychiatric diagnoses, substance use, or relevant comorbidities were excluded. Propensity score matching (1:1) was used to balance baseline demographics and clinical covariates. The primary outcomes were new-onset major depressive disorder (MDD), anxiety, and suicidal ideation or attempts. Relative risk (RR), 95% confidence intervals (CI), P-values, and risk differences were reported. Results: Compared to diabetic non-obese patients, obese patients on GLP-1RAs demonstrated significantly higher risk of psychiatric disorders. For MDD, the relative risk was 0.573 (95% CI: 0.461–0.714; P < 0.0001) with a risk difference of -2.454%. For anxiety, the relative risk was 0.552 (95% CI: 0.465–0.655; P < 0.0001) and the risk difference was -4.2%. There was no statistically significant difference in suicidal ideation or attempts between the 2 cohorts (P = 1), with both the relative risk and risk difference indicating no effect. Conclusion: Obese patients treated with GLP-1RAs had a significantly higher risk of developing major depressive disorder and anxiety compared to diabetic non-obese patients on the same therapy. These results suggest that obesity itself may confer increased psychiatric vulnerability in the context of GLP-1RA use. However, no difference in suicidal ideation or behavior was observed. These findings warrant further investigation and support the need for psychiatric screening in high-risk populations prior to GLP-1RA initiation.