The risk of mortality associated with haloperidol compared with other antipsychotics: propensity-score-matched cohort study in Hong Kong

Abstract

Aim: Numerous studies reported an increased mortality risk associated with haloperidol compared with other antipsychotics. However, limited studies used propensity score method to adjust for confounding and explored the mechanism underpinning the deaths due to the lack of information on death causes. This study aimed to investigate the risk of mortality and the specific causes of death associated with haloperidol versus other antipsychotics to address these limitations. Methods: A cohort study was conducted using the data from Hong Kong Clinical Data Analysis and Report System. Patients with incident antipsychotic prescription (haloperidol, amisulpride, aripiprazole, chlorpromazine, olanzapine, quetiapine, risperidone, sulpiride, and trifluoperazine) in 2004-2014 were identified. Hazard ratios (HR) were estimated by comparing the risk of mortality for haloperidol with other individual antipsychotics using the Cox proportional hazards model stratified by propensity-score matched groups. Secondary analyses were conducted to further examine the specific causes of death including cardiovascular and pneumonia-related mortality. Results: A total of 136,593 antipsychotic users were included. In a mean follow-up of 3.2 person-years, lower risk of all-cause mortality were found for all non-haloperidol antipsychotic drugs versus haloperidol, with HRs from 0.68(95% CI:0.64-0.72) for chlorpromazine to 0.43(95% CI:0.36-0.53) for trifluoperazine. Reduced risk of cardiovascular mortality were observed for risperidone (HR:0.79;95% CI:0.66-0.93]), sulpiride(HR:0.78;95% CI:0.64-0.96), chlorpromazine(HR:0.76;95% CI:0.65-0.90) and quetiapine (HR:0.67;95% CI:0.57-0.78). Lower risk of pneumonia-related mortality were observed for all non-haloperidol antipsychotics except amisulpride and olanzapine, with HRs from 0.76(95% CI:0.68-0.85) for risperidone to 0.38(95% CI:0.24-0.61) for trifluoperazine. Conclusions: With robust control for confounding, haloperidol is found to be associated with an increased risk of mortality over other antipsychotics. The increased risk of cardiovascular and pneumonia-related mortality might account for some of the deaths associated with haloperidol. The cardiovascular and immunological profiles of the patients should be cautiously assessed before prescribing haloperidol. Other antipsychotics with lower observed risk could be considered as a preferred option.