Comparative mortality risk of antipsychotics in 41,695 patients with schizophrenia: an 11-year population-based cohort study in Hong Kong

Abstract

Antipsychotics are the mainstay treatment for schizophrenia, which is associated with excess mortality. Differential mortality-risk in relation to individual antipsychotics or various antipsychotic-regimens remains to be clarified. This population-based cohort study investigated the comparative mortality risk associated with antipsychotic-monotherapies (using perphenazine as reference-category) or antipsychotic regimens (using oral first-generation-antipsychotics [FGA] as reference-category) in treated-patients with schizophrenia, utilizing electronic-health-record of public healthcare-services in Hong-Kong within 2006–2016. Cox-regression analysis with antipsychotic-exposure as time-varying covariates was performed to examine all-cause, natural-cause, and unnatural-cause mortality-risks. In the overall-cohort (n = 41,695), antipsychotic-monotherapy analysis showed that clozapine-use was associated with the lowest risk for all-cause (adjusted-hazards-ratio, aHR: 0.41; 95 % confidence-interval (CI) [0.33–0.52]), natural-cause (0.52 [0.40–0.69]), and unnatural-cause mortality (0.16 [0.09–0.27]) among antipsychotic-monotherapies, compared with perphenazine. Among two long-acting-injectable (LAI) antipsychotics (paliperidone/risperidone), paliperidone-LAI demonstrated lower all-cause (0.51 [0.36–0.72]) and natural-cause (0.55 [0.37–0.83]) mortality-risk. Several commonly-used second-generation-antipsychotics (olanzapine (Zyprexa)/quetiapine/risperidone (Risperdal)/aripiprazole/amisulpride) were also associated with reduced mortality-risk relative to perphenazine. In antipsychotic-regimen analyses, reduced mortality-risk was noted for polypharmacy-regimens that included clozapine or LAI antipsychotics compared to FGA-oral monotherapy, while FGA-LAI monotherapy, any-antipsychotic polypharmacy and oral-antipsychotic polypharmacy without clozapine were associated with elevated all-cause and natural-cause mortality-risk. Generally consistent results were observed in the incident-cohort (n = 13,283). Our results highlight that mortality-risk is differentially associated with various antipsychotics and regimens, and indicate the critical role of clozapine and LAI antipsychotics in alleviating excess mortality-risk. Our findings underscore the importance of ensuring early access to clozapine and LAI antipsychotics to optimize psychiatric and physical outcomes in schizophrenia patients.