This cohort study used Medicare claims for enrolees with type 2 diabetes. GLP-1 RA users were propensity-matched 1:1 to users of other second-line glucose-lowering drugs, using stratified Cox regression. Aggregate psychiatric disorder risk was slightly higher (HR 1.04; 1.7 excess events per 1,000 person-years), as was depression (HR 1.11). Risks were lower for substance use disorders (HR 0.84), dementia (HR 0.87) and suicidal or self-harm events (HR 0.81). Monotherapy (HR 1.43) and escalation showed higher psychiatric risk, whereas switch and augmentation showed reduced or null risks. The aggregate increase is of uncertain clinical significance, and the monotherapy excess may reflect baseline clinical differences.
Why it is interesting: GLP-1 receptors sit in brain regions governing mood, reward and stress, yet the drugs' neuropsychiatric effects are poorly characterised. This study splits results by outcome, obesity status and treatment pattern rather than assuming one class effect.
Cite this item
BibTeX
Free sign-up. A digest of the new papers twice a week, grouped by domain.