Cholinesterase inhibitors and memantine are associated with a reduced mortality in nursing home residents with dementia: a longitudinal observational study

Abstract

A large proportion of nursing home (NH) residents suffer from dementia and effects of conventional anti-dementia drugs on their health is poorly known. We aimed to investigate the associations between exposure to anti-dementia drugs and mortality among NH residents. This retrospective longitudinal observational study involved 329 French NH and the residents admitted in these facilities since 2014 and having major neurocognitive disorder. From their electronic health records, we obtained their age, sex, level of dependency, Charlson comorbidity index, and Mini mental examination score at admission. Exposure to anti-dementia drugs was determined using their prescription into 4 categories: none, exposure to acetylcholinesterase inhibitors (AChEI) alone, exposure to memantine alone, exposure to AChEI and memantine. Survival until the end of 2019 was studied in the entire cohort by Cox proportional hazards. To alleviate bias related to prescription of anti-dementia drugs, we formed propensity-score matched cohorts for each type of anti-dementia drug exposure, and studied survival by the same method. We studied 25,358 NH residents with major neurocognitive disorder. Their age at admission was 87.1 + 7.1 years and 69.8% of them were women. Exposure to anti-dementia drugs occurred in 2,550 (10.1%) for AChEI alone, in 2,055 (8.1%) for memantine alone, in 460 (0.2%) for AChEI plus memantine, whereas 20,293 (80.0%) had no exposure to anti-dementia drugs. Adjusted hazard ratios for mortality were significantly reduced for these three groups exposed to anti-dementia drugs, as compared to reference group: HR: 0.826, 95%CI 0.769 to 0.888 for AChEI; 0.857, 95%CI 0.795 to 0.923 for memantine; 0.742, 95%CI 0.640 to 0.861 for AChEI plus memantine. Results were consistent in propensity-score matched cohorts. The use of conventional anti-dementia drugs is associated with a lower mortality in nursing home residents with dementia and should be widely used in this population.