So Sato, Yusuke Sasabuchi, Hayato Yamana, Takahide Kohro, Hideo Yasunaga
Among 8117 eligible patients, 6467 initiated oral formulations and 1650 initiated transdermal patches. Mean treatment duration was longer in the oral group than in the patch group (1.5 vs 0.6 years; log-rank P < .001). In adjusted Cox regression analysis, the patch group showed a higher risk of discontinuation than the oral group (hazard ratio, 11.3; 95% CI, 10.2-12.4). Hospitalization and mortality rates were lower in the oral group than in the patch group (236 vs 272 and 210 vs 272 per 1000 person-years, respectively).
OBJECTIVES: To examine treatment duration of oral and transdermal cholinesterase inhibitor formulations in older adults with severe Alzheimer's disease (AD) in routine clinical practice.
DESIGN: Retrospective population-based cohort study.
SETTING AND PARTICIPANTS: Linked medical and long-term care insurance claims databases from Tochigi Prefecture, Japan, between April 2019 and October 2023. We included patients aged ≥65 years with AD who were certified at care needs level 4 or 5 (indicating complete dependence in activities of daily living) or were institutionalized under the Japanese Long-Term Care Insurance system.
METHODS: Patients were classified according to the initial cholinesterase inhibitor formulation: oral formulations (donepezil or galantamine) or transdermal patches (rivastigmine or donepezil). The primary outcome was duration of treatment with the initial formulation. Continuous treatment was defined using a 30-day grace period. Kaplan-Meier analysis and Cox proportional hazards models were used to compare treatment discontinuation between groups. Hospitalization and mortality were evaluated as exploratory outcomes.
RESULTS: Among 8117 eligible patients, 6467 initiated oral formulations and 1650 initiated transdermal patches. Mean treatment duration was longer in the oral group than in the patch group (1.5 vs 0.6 years; log-rank P < .001). In adjusted Cox regression analysis, the patch group showed a higher risk of discontinuation than the oral group (hazard ratio, 11.3; 95% CI, 10.2-12.4). Hospitalization and mortality rates were lower in the oral group than in the patch group (236 vs 272 and 210 vs 272 per 1000 person-years, respectively).
CONCLUSIONS AND IMPLICATIONS: In older adults with severe AD, oral cholinesterase inhibitor formulations were continued substantially longer than transdermal patches. Given the limited evidence supporting prolonged cholinesterase inhibitor use in severe AD, these findings raise concerns regarding long-term prescribing practices in advanced dementia care.