(292) Atypical Antipsychotic Use and Mortality Risk in Parkinson's Disease

Publication Number: 623

Background: Pimavanserin is currently the only atypical antipsychotic (AAP) indicated for Parkinson disease (PD) psychosis in the US, although other AAPs are still prescribed off-label for this indication. There are inconsistent and limited data on the mortality risk associated with pimavanserin compared to preferred AAPs (i.e., quetiapine and clozapine) or non-preferred AAPs (i.e., other AAPs) in this population.

Objectives: To compare all-cause mortality risks of pimavanserin compared to preferred AAPs and non-preferred AAPs in PD psychosis

Methods: We conducted a retrospective cohort study consisting of individuals with PD who were new users of pimavanserin and other AAPs using 2016-2021 Optum’s de-identified Clinformatics® Data Mart. Pimavanserin users were matched to preferred AAP or non-preferred AAP users via propensity scores, and baseline characteristics before and after matching were compared using standardized differences. Cox proportional hazards models and Kaplan-Meier curves were used to compare all-cause mortality risks of pimavanserin and preferred AAPs or non-preferred AAPs and to illustrate survival rates, respectively.

Results: We identified 775, 4563, and 1297 eligible new users of pimavanserin, preferred AAPs, and non-preferred AAPs, respectively. Baseline characteristics were more comparable between groups after matching. After 1:1 matching, there was no difference in mortality risk among pimavanserin vs. preferred AAP users—adjusted hazard ratios (aHR) 0.99 (95% confidence interval 0.82-1.21) in the intention-to-treat analysis. After greedy matching with caliper, there was also no difference in mortality risk among pimavanserin vs. non-preferred AAP users—aHR 0.98 (0.79-1.22). We found similar results when matching each pimavanserin user to up to 5 users of preferred or non-preferred AAPs, aHRs 1.03 (0.88, 1.20) and 1.15 (0.97, 1.36), respectively.

Conclusions: Mortality risk among PD patients using AAPs did not differ by specific AAP drug or AAP category. If confirmed in other patient samples, research on the comparative morbidity of AAP drugs will be needed to guide clinical decision-making.