Title:Antipsychotic Medication Risk of Dementia and Death: A Propensity Matched Cohort Study
Volume: 14
Issue: 2
Author(s): Xiu R. Lowe*, Maqdooda Merchant and Rachel A. Whitmer
Affiliation:
- Department of Mental Health and Addiction Medicine, Kaiser Permanente Northern California, Union City, CA, USA
Keywords:
Antipsychotics, non-demented patients, long-term risk, psychiatric diagnosis, AP treatment dementia.
Abstract:
Objective: This study aimed to compare the incidence of dementia and all-cause mortality
up to 20 years post-treatment in an index non-demented cohort between antipsychotic (AP)
medication treatment and non-AP treatment groups.
Methods: All patients in Kaiser Permanente Northern California with a major psychiatric diagnosis
between 01/01/1996 and 12/31/2000, age ≥ 50 years, and without dementia diagnosis were
included. The study cohort was divided into a “user group”, patients treated with AP for ≥ 365
days (n = 1,829), and a “non-user group”, propensity score-matched on age, sex, and race (n =
9,145). The association between AP exposure and dementia or mortality during the follow-up period
(01/01/2001–12/31/2015) was evaluated using Cox proportional hazard models adjusted for
psychiatric diagnosis, comorbidities, and other medications.
Results: The user group had a hazard ratio (HR) of 2.2 (CI 1.8–2.7) for dementia and 1.3 (CI 1.2-
1.5) for death. The onset of dementia in the user group was significantly higher in patients aged ≤
65 years (p < 0.001). The user group was sub-grouped into atypical, typical, and both; HR for
dementia was 1.7 (CI 1.2-2.4), 2.5 (CI 1.9-3.1), and 1.8 (CI 1.4–2.4), respectively. Dementia and
mortality were significantly higher in patients concurrently treated with benzodiazepine (HR 1.3;
CI 1.2-1.5 and HR 1.4; CI 1.3–1.5) or tricyclic antidepressants (HR 1.2; CI 1.1–1.4 and HR 1.1; CI
1.0-1.2), respectively.
Conclusion: Our preliminary results reveal an association between AP treatment and increased
rates of both dementia and mortality. Future research is needed to substantiate our current findings.
