(A-320) Evaluating the Impact of GLP-1RAs on Mental Health and Antidepressant Use: A Target Trial Emulation with Distributional NCO Calibration

Background: Glucagon-like peptide-1 receptor agonists (GLP-1RAs) are widely prescribed for type 2 diabetes and obesity, yet their impact on mental health disorders and antidepressant use remains unclear. While some evidence suggests potential neuropsychiatric effects, findings are inconsistent, warranting further investigation.

Objectives: To evaluate whether GLP-1RAs influence the risk of mental health disorders and antidepressant use compared to sodium-glucose cotransporter-2 inhibitors (SGLT2is) or dipeptidyl peptidase-4 inhibitors (DPP4is).

Methods: We conducted a target trial emulation using electronic health record (EHR) data from Penn Medicine (January 2019–September 2024). Patients prescribed GLP-1RAs were compared to those receiving DPP4is or SGLT2is. We applied propensity score matching to balance demographics (age, race/ethnicity, sex), medical history, site index, and cohort entry timing (categorized to one month). Mental health disorders and antidepressant use were assessed following treatment initiation. To address residual confounding, we incorporated distributional negative control outcome (NCO) calibration.

Results: Compared to SGLT2is, GLP-1RAs were associated with an increased risk of nonorganic sleep disorders (RR=1.41, 95% CI: 1.16–1.70) and somnolence, stupor, and coma (RR=1.20, 95% CI: 1.03–1.39) before NCO calibration, but these associations were not significant after calibration. The risk of overall antidepressant use was slightly lower with GLP-1RAs (RR=0.93, 95% CI: 0.88–1.00) before calibration, with no significant change after calibration. Compared to DPP4is, GLP-1RAs showed an increased risk of several mental health disorders before NCO calibration, including recurrent depressive disorder (RR=1.71, 95% CI: 1.49–1.96), nonorganic sleep disorders (RR=1.69, 95% CI: 1.38–2.07), and bipolar affective disorder (RR=1.42, 95% CI: 1.31–1.53), but none remained significant after calibration. Antidepressant use was also higher before calibration, particularly for venlafaxine (RR=1.54, 95% CI: 1.25–1.90) and duloxetine (RR=1.49, 95% CI: 1.31–1.70), with no significant differences post-calibration.

Conclusions: GLP-1RAs may be associated with an increased risk of mental health disorders and changes in antidepressant use, but these effects did not persist after addressing residual confounding.