Neuropsychiatric Outcomes Associated With GLP ‐1 Receptor Agonist Use in Medicare Beneficiaries With Type 2 Diabetes: Variation by Treatment Pattern and Obesity Status

Aims Glucagon‐like peptide‐1 (GLP‐1) receptors are expressed in brain regions governing mood, reward and stress, yet the neuropsychiatric effects of GLP‐1 receptor agonist (GLP‐1 RA) therapy remain poorly characterized. We estimated associations between GLP‐1 RA use and neuropsychiatric disorders and variation by obesity status and treatment pattern. Material and Methods This population‐based cohort study used claims data from Medicare enrolees with Type 2 diabetes mellitus. GLP‐1 RAs were compared with other second‐line glucose‐lowering medications and evaluated across four treatment patterns: monotherapy, switch, escalation and augmentation. Eight neuropsychiatric outcomes were examined using 1:1 propensity score‐matched, covariate‐adjusted stratified Cox regression with time‐varying covariates. Results GLP‐1 RA use was associated with a small increase in aggregate psychiatric disorder risk (Hazard Ratio [HR] = 1.04; 95% Confidence Interval: 1.03–1.06; 1.7 excess events/1000 person‐years), of uncertain clinical significance given the large cohort size. Risk was elevated for depression (HR = 1.11; 3.0 excess events/1000 person‐years) but reduced for substance use disorders (HR = 0.84; 4.1 fewer events/1000 person‐years), dementia (HR = 0.87; 2.8 fewer) and suicidal/self‐harm (HR = 0.81; 0.4 fewer events/1000 person‐years). Associations varied by treatment pattern: monotherapy showed elevated psychiatric (HR = 1.43) and elevated emotional symptom (HR = 1.32) risks; escalation also exhibited elevated psychiatric risk, whereas switch and augmentation demonstrated reduced or null risks. Conclusions Rather than a uniform class effect, the neuropsychiatric profile of GLP‐1 RA appears context‐dependent, differing by treatment pattern, obesity status and outcome examined. Elevated risk with monotherapy or escalation may reflect baseline clinical differences rather than pharmacological effect. Reduced risks of substance use disorders, dementia and suicidal/self‐harm are reassuring.