Beusterien_2004_CNS.Drugs_18_1143

INTRODUCTION: Although numerous studies have evaluated predictors of nursing home placement (NHP), few have focused on the effects of cholinesterase inhibitor (ChEI) use on NHP. The objective of this study was to compare the risk of NHP between rivastigmine patients versus no-ChEI patients (control group), and secondly, between rivastigmine versus donepezil patients.
METHODS: A retrospective analysis of a large US medical claims database was performed. Eligible subjects were identified from those who had continuous medical coverage from 1 April 2000 to 30 June 2002. Rivastigmine and donepezil subjects were new users, defined as having received no ChEI treatment during the initial 6 months of the study. Control subjects were diagnosed with Alzheimer's disease (AD) at some point after the initial 6-month period. All subjects were followed from baseline (initiation of ChEI therapy or initial AD diagnosis) to the date of NHP or 30 June 2002, whichever occurred first.
RESULTS: In the rivastigmine (n=1181), donepezil (n=3864), and control (n=517) groups, 3.7%, 4.4% and 11.0% of subjects, respectively, had an NHP (p <0.001 for rivastigmine versus control). A Cox proportional hazard model, controlling for age, gender, comorbidities and behavioural disorders, showed that the control subjects were almost 3-fold more likely to have NHP than rivastigmine subjects (hazard ratio [HR]=2.71; 95% CI 1.82, 4.03). The difference in the risk of NHP was not significant between the rivastigmine and donepezil groups (HR=1.23; 95% CI 0.89, 1.71). DISCUSSION: This study demonstrated that rivastigmine decreased the risk of NHP in a large insured population.