Neuropsychiatric association of tirzepatide and semaglutide in obesity with and without type 2 diabetes.

BACKGROUND: Neuropsychiatric outcomes after starting newer anti-obesity medicines remain uncertain in routine care. We evaluate associations between initiation of tirzepatide or semaglutide and incident neuropsychiatric diagnoses in people with obesity, stratified by type 2 diabetes (T2D). METHODS: Using a United States electronic health record network (2020-2025), we identified adults with obesity who newly started tirzepatide or semaglutide and matched them one-to-one to new users of naltrexone-bupropion, phentermine, or phentermine-topiramate. We modeled the time to first recorded diagnosis over 24 months; bariatric surgery recipients were examined as a contextual comparator. Confidence intervals are 95%. RESULTS: Here we show that, versus naltrexone-bupropion, semaglutide is associated with lower hazards of anxiety disorder in adults with T2D (hazard ratio 0.67, 0.51-0.87) and without T2D (0.73, 0.64-0.83), and lower hazards of depression in both strata (with T2D: 0.69, 0.50-0.95; without T2D: 0.53, 0.45-0.61). Tirzepatide versus naltrexone-bupropion is associated with lower hazard ratios for anxiety disorder (with T2D: 0.55, 0.40-0.76; without T2D: 0.78, 0.68-0.89) and depression (with T2D: 0.49, 0.33-0.72; without T2D: 0.57, 0.49-0.66). In adults without T2D, tirzepatide versus semaglutide is associated with higher hazards of anxiety disorder (1.12, 1.06-1.18) and insomnia (1.13, 1.05-1.21). CONCLUSIONS: Initiation of tirzepatide and semaglutide is associated with differing hazard patterns across prescription comparators, with bariatric surgery providing additional context. These findings inform monitoring and shared decisions, while remaining subject to residual confounding and diagnosis misclassification.