Discontinuation of antipsychotics among residents of saskatchewan long-term care facilities
Date
2014-08-19
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
ORCID
Type
Degree Level
Masters
Abstract
Background and Objectives: Antipsychotic medications (APMs) are used for the treatment of behavioural symptoms of dementia. The use of APMs among residents of long-term care facilities (LTCFs), who have a high probability of dementia, is correspondingly high, and has been linked to adverse patient outcomes. The study objectives were to: (a) describe facility variation in APM discontinuation rates, (b) test the association between time to APM discontinuation and patient and facility explanatory variables, and (c) conduct a sensitivity analysis about the effect of changes in the measurement of APM discontinuation on variable associations. Methods: The study used a population-based retrospective cohort design. Saskatchewan’s (SK) administrative health databases for the period from April 1, 2004 to March 31, 2011 were the data sources. The study cohort included all seniors (≥65 years of age) with a first admission to a SK LTCF and an APM dispensation on or after the admission date. Discontinuation was defined as a 70-day gap after the last APM dispensation. Patient-level explanatory variables included socio-demographics, comorbidity, prior medication exposure, behavioural and cognitive status, and health services utilization. Facility-level explanatory variables included size, location, licensing status, and type. Percentage discontinuation across facilities was descriptively analyzed. Cox proportional hazards regression models with adjustment for clustering of patients within LCTFs were used to test associations with time to discontinuation. A sensitivity analysis of APM discontinuation was conducted by shortening (35 days) and lengthening (105 days) the time from last dispensation. Results: Among all residents eligible to be cohort members 35.7% were dispensed an APM. A total of 19.5% of the 8358 cohort members discontinued APMs in the observation period. The Kaplan-Meier estimate of the median time to discontinuation was 6.5 years. Demographic, comorbidity, behavioural, and drug exposure variables were most strongly associated with APM discontinuation. Discontinuation was not associated with facility characteristics. Variable associations were insensitive to the definition of APM discontinuation, but changed over time. Conclusion: Discontinuation of APMs is low, despite high rates of utilization over long periods of time. Patient characteristics are associated with APM discontinuation, but not facility characteristics, suggesting that LCTFs are applying consistent approaches to patient management. However, low levels of discontinuation suggest that there may be a need for health care providers to regularly review the prescribing, dispensing, and administration of APMs to LTCF residents in order to ensure appropriate use of these pharmaceuticals.
Description
Keywords
long-term care, antipsychotic, epidemiology, health services research, drug utilization, pharmacology
Citation
Degree
Master of Science (M.Sc.)
Department
Community Health and Epidemiology
Program
Community and Population Health Science