Skip Navigation
AHRQ--Agency for Healthcare Research and Quality: Advancing Excellence in Health Care
  • Home
  • Search for Research Summaries, Reviews, and Reports
 
 

EHC Component

  • DEcIDE Project

Topic Title

  • Comparative Safety of Conventional and Atypical Antipsychotic Medications: Risk of Death in British Columbia Seniors

Full Report



Abstract - Final

Comparative Safety of Conventional and Atypical Antipsychotic Medications: Risk of Death in British Columbia Seniors

Topic Abstract

Context: A Health Canada and Food and Drug Administration (FDA) Advisory has warned that atypical antipsychotic medications (APMs) increase mortality in older patients.

Objective: To assess the short-term mortality in a population-based cohort of all British Columbia seniors who initiated conventional vs. atypical APMs.

Design. Cohort study.

Setting. Population-based linked health care utilization data of all British Columbia residents 65 years and older from 1996 through 2004.

Patients. Senior BC residents who were initiated on antipsychotic medications and had no recorded cancer diagnosis.

Intervention. Conventional APMs vs. atypical APMs.

Main Outcome Measures: All cause mortality.

Results: 12,882 seniors initiated conventional APM therapy and 24,359 atypical APMs. Within the first 180 days of use, 1,822 patients (14.1%) who initiated treatment with conventional APMs died, compared with 2,337 patients (9.6%) who initiated treatment with atypical APMs (unadjusted mortality ratio = 1.47; 95% confidence interval: 1.39 – 1.56). Multivariable adjustment resulted in a 180-day mortality ratio (MR) of 1.32 (1.23-1.42). The increase in mortality was highest in users of haloperidol (MR = 2.14; 95% CI: 1.86 to 2.45) but lower for loxapine (MR = 1.29; 95% CI: 1.19 to 1.40).The greatest mortality increase occurred with use of higher (> median) conventional APM dosages (MR=1.67; 1.50-1.86) and during the first 40 days after initiation (MR=1.60; 1.42-1.80). Results were confirmed in propensity score analyses and instrumental variables estimation adjusting for unmeasured confounders.

Conclusions. Elderly patients using conventional APMs are at no lower risk of mortality than those using atypical APMs. The observed 32% increased mortality risk of conventional APMs is unlikely to be explained by confounding.