Tolerability of Cholinesterase Inhibitors: A Population-Based Study of Persistence, Adherence, and Switching.

Abstract

Cholinesterase inhibitors (ChEIs) are prescribed to dementia patients despite their poor tolerance. Low tolerability potentially reduces persistence and adherence, while inducing switching between medications. Comparisons of these utilization measures contribute to knowledge of the relative tolerability of these medications. The aim was to compare persistence, adherence, and switching between donepezil, galantamine, oral rivastigmine, and rivastigmine patch. A population-based cohort study, using British Columbia claims data (2009-2013), assessed ChEI new users aged 40 and older. We conducted survival analysis to compare persistence and Poisson regression to estimate switching rates. Good adherence, defined as a medication possession ratio of ≥80%, was modeled using log-binomial regression. Analyses were adjusted using propensity scores. Patients on galantamine had longer mean persistence and better adherence compared with patients on donepezil, with a hazard ratio for discontinuation of 0.91 [95% confidence interval (CI) 0.87-0.96] and a relative risk for good adherence of 1.01 (95% CI 1.002-1.03). Rivastigmine was associated with the shortest mean persistence [3.6months (95% CI 3.0-4.2) and 5.0 (95% CI 4.7-5.3) for oral and patch, respectively] and the highest mean switching rates. Comparing the two rivastigmine preparations, the patch was associated with decreased discontinuation compared with oral [hazard ratio 0.79 (95% CI 0.71-0.89)] and decreased switching [relative risk 0.63 (95% CI 0.46-0.87) during the first 6months of treatment]. Paradoxically, the patch was also associated with poorer adherence [relative risk for good adherence 0.94 (95% CI 0.91-0.98)] than the oral formulation. Based on estimates of persistence, adherence, and switching, galantamine was the best tolerated ChEI and rivastigmine the least.