Lee_2024_JAMA.Netw.Open_7_e2445878

IMPORTANCE: Discontinuing cholinesterase inhibitors when initiating memantine in patients with dementia may be reasonable to reduce treatment burden, costs, and the risk of adverse drug events. OBJECTIVE: To assess the association of cholinesterase inhibitor discontinuation on long-term care institutionalization among older adults with dementia who initiate memantine. DESIGN, SETTING, AND PARTICIPANTS: This retrospective propensity score-matched cohort study used Medicare claims data from January 2014 to December 2019. Participants included fee-for-service Medicare beneficiaries with dementia. Data were analyzed from September 2021 to August 2024. EXPOSURES: Discontinuation vs continuation of cholinesterase inhibitor. MAIN OUTCOMES AND MEASURES: The primary outcome was 1-year long-term care institutionalization-free days. Secondary outcomes include all-cause death and adverse drug events over 1 year. We performed subgroup analyses based on age, sex, dementia type (Alzheimer disease vs other), frailty, and dementia severity (mild vs moderate or severe) based on claims-based algorithms. The primary outcome was analyzed using nonparametric restricted mean survival time analysis. RESULTS: Among 16 292 beneficiaries who initiated memantine, 1820 (11.2%) discontinued cholinesterase inhibitors. In the propensity score-matched cohort of 3612 beneficiaries, the mean (SD) age was 80.7 (6.7) years, 2261 (62.6%) were female, and 1989 (55.0%) had a diagnosis of Alzheimer disease. Over 1 year, long-term care institutionalization occurred in 51 of 1806 beneficiaries (2.8%) who discontinued cholinesterase inhibitors (3.4 per 100 person-years) and 62 of 1806 beneficiaries (3.4%) who continued (4.1 per 100 person-years). There was no statistically significant difference in the 1-year mean institutionalization-free days between discontinuation and continuation groups (360.6 [95% CI, 359.3 to 362.0] days vs 359.1 [95% CI, 357.5 to 360.6] days; mean difference, 1.5 [95% CI,-0.5 to 3.6] days). The mean difference in the long-term care institutionalization-free days did not differ by age category, sex, dementia type, frailty, or dementia stage. Individuals who discontinued had a lower rate of fall-related injury (0.9 vs 2.0 per 100 person-years; hazard ratio [HR], 0.47 [95% CI, 0.25 to 0.88]). There was no difference between the discontinuation and continuation groups in all-cause death (10.4 vs 11.6 per 100 person-years; HR, 0.89 [95% CI, 0.72 to 1.10]). CONCLUSIONS AND RELEVANCE: In this study, discontinuing cholinesterase inhibitors upon memantine initiation was not associated with an increased risk of long-term care institutionalization but with a lower risk of fall-related injury among older adults with dementia. These findings offer valuable insights for clinicians aiming to reduce treatment burden in this population.