Risk of Serious Adverse Events Associated With Individual Cholinesterase Inhibitors Use in Older Adults With Dementia: A Population-Based Cohort Study

Abstract

Background and objective: Cholinesterase inhibitors (ChEIs) are used as first-line pharmacotherapy to manage dementia. However, there are limited data regarding their relative safety. This study evaluated the risk of serious adverse events (SAEs) associated with individual ChEIs in older adults with dementia and also examined sex-based and dose-based effects on this risk.

Methods: This was a retrospective cohort study using 2013-2015 US Medicare claims data involving Parts A, B, and D. Patients aged ≥ 65 years with a dementia diagnosis and incident use of the ChEIs, namely donepezil, galantamine, or rivastigmine, were included. The primary outcome of interest was SAEs defined as emergency department visits, inpatient hospitalizations, or death within 6 months of ChEI initiation. Multivariable Cox proportional hazards regression with propensity score (PS) as a covariate and inverse probability of treatment weighting generated using generalized boosted models was used to assess the risk of SAEs across individual ChEIs.

Results: The study included 767,684 older adults with dementia who were incident new users of ChEIs (donepezil 79.42%, rivastigmine 17.67%, galantamine 2.91%). SAEs were observed in 15.5% of the cohort within 6 months of ChEI prescription. Cox regression model with PS as covariate found that patients prescribed rivastigmine (adjusted hazard ratio [aHR] 1.12; 95% CI 1.03-1.33) and galantamine (aHR 1.51; 95% CI 1.24-1.84) were at increased risk of SAEs compared with patients on donepezil. Stratified analyses revealed that rivastigmine was associated with an 18% increased risk for SAEs in females (aHR 1.18; 95% CI 1.06-1.31), and galantamine was associated with a 71% increased risk in males (aHR 1.71; 95% CI 1.17-2.51) compared with donepezil. High and recommended index doses of rivastigmine and galantamine were associated with an increased risk of SAEs compared with donepezil. The findings were consistent in sensitivity analyses.

Conclusion: The study found that the risk of SAEs varied across individual ChEIs, with sex and dose moderating these effects. Therefore, these moderating effects should be carefully considered in personalizing dementia care.