Association Between Antipsychotics and All-Cause Mortality Among Community-Dwelling Older Adults

Abstract

Background: Antipsychotics are prescribed to treat various symptoms in older adults, however, their safety in this context has not been fully evaluated. The objective was to evaluate mortality risks associated with off-label use of antipsychotics among older adults with no preexisting mental illness or dementia relative to those with diagnosis of dementia.

Methods: Data (2007-2015) were derived from Department of Veterans Affairs registries for 730,226 patients (≥65 years) with no baseline serious mental illness, dementia). We estimated the cumulative incidence of antipsychotics prescription and 10-year all-cause mortality. The extended Cox models were used to estimate Hazard Ratios (HRs) associated with antipsychotics prescription, adjusted for time-varying covariates, dementia diagnosis, comorbidity index score, and age at time of first exposure to antipsychotics.

Results: The study included 98% males, 13% African Americans, and 81% Caucasian. Patients with dementia and antipsychotics had the highest risk of mortality (78.0%), followed by (73.0%) for patients with dementia alone and compared with patients without dementia or antipsychotics exposure who had the lowest mortality risk (42.0%). Exposure to typical antipsychotics was associated with (HR: 2.1, confidence interval [CI] 2.0-2.2) compared with atypical antipsychotics (HR: 1.5, CI 1.4-1.5, p = <.0001).

Conclusion: In a large cohort of older adults, antipsychotics were associated with an increased risk of all-cause mortality. While significant increase in mortality was attributable to the diagnosis of dementia, the addition of antipsychotics resulted in added mortality risk among all patients. Antipsychotic medications should be used cautiously in all older adults, not only those with dementia.

Keywords: Cognition; Drug related; Primary care.

Figure 1.

Kaplan–Meier estimates of the cumulative risk of death by 10 years. Shown is the probability of death among patients who no exposure to antipsychotics or diagnosis of dementia (black curve), among patients who were exposed to antipsychotics with no dementia diagnosis (green curve), among patients who were not exposed to antipsychotics with dementia diagnosis (blue curve), and among patients who were exposed to antipsychotics with dementia diagnosis (red curve). Rates were compared with the use of the log-rank test.