Abstract
BACKGROUND: In 2018, France withdrew reimbursement for cholinesterase inhibitors (ChEIs) in Alzheimer's disease, citing modest efficacy, lack of long-term benefit, and safety concerns. This policy shift provided a unique opportunity to assess ChEI effectiveness in real-world settings, by evaluating, among patients treated with ChEI between 01/08/2017 and 01/08/2018, the impact of treatment discontinuation on cognitive decline (MMSE score) and survival. METHODS: Using the French National Alzheimer's Database (BNA) and Meotis databases, we emulated a pragmatic, intention-to-treat trial comparing patients who discontinued ChEIs after delisting to those who continued treatment, under quasi-experimental conditions. To model cognitive trajectories, we used the inverse probability treatment-weighted (IPTW) cohort and applied mixed-effects models with random intercepts across all follow-up visits. Survival was estimated with a pooled logistic regression including treatment group, follow-up time, and an interaction between treatment and time. FINDINGS: The mean difference in MMSE decline between discontinuers and continuers after one year was 0·97 points (95% CI 0·68-1·27, p < 0·001), reaching 1·81 points (0·91-2·71, p < 0·001), after four years. No significant difference in mortality (RR 1·10, 95% CI [0·95-1·29]) was observed over a five-year period. INTERPRETATION: Our findings confirm and extend prior trials by demonstrating the sustained cognitive benefits of ChEIs in a real-world setting. While acknowledging the limitations associated with its retrospective nature, our study argues for reconsidering the 2018 delisting decision, as ChEIs remain safe and clinically relevant for mild-to-moderate Alzheimer's disease. FUNDING: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.