Author Response: Long-term Effects of Cholinesterase Inhibitors on Cognitive Decline and Mortality. (16th November 2021)

• Record Type: Journal Article
• Authors: Xu, Hong; Garcia-Ptacek, Sara; Eriksdotter, Maria
• Is Part Of: Neurology. Volume 97:Number 20(2021)
• Page Start: 965
• Page End: 965
• Publication Date: 2021-11-16
• Subjects: Neurology -- Periodicals; Neurology -- Periodicals; Neurologie -- Périodiques; 616.8
• DOI: 10.1212/WNL.0000000000012874
• Languages: English
• ISSNs: 0028-3878

Abstract

Abstract : We thank Dr. Kuschpel for the comment on our article. 1 We agree that people with Alzheimer disease (AD) who received cholinesterase inhibitors (ChEIs) and those who did not differed significantly, so we controlled for the unbalanced comorbidity and medication in our propensity score–matching cohort. The result was a well-matched cohort in the distribution of observed comorbidities. We also acknowledge the possibility of residual and unknown confounders, although we did not have more information about comorbidity severity. Our group has previously reported an association between ChEIs and reductions in myocardial infarction, stroke, and mortality. 2 In an RCT study by Hager et al., the 2-year mortality rate with the use of galantamine compared with the placebo group was similar to our findings, substantially reduced by 42%. The articles referenced by the commenter were meta-analyses of clinical trials, one of which addressed vascular dementia and not AD. 4, 5 Most clinical trials on the subject have been relatively short, from 6 months to 1 year. It is possible that short- and long-term mortality effects vary—a possibility that exceeds the scope of this study but is a possible research question for future studies. Given that this is a cohort study, it is difficult to completely exclude the possibility of bias, as we have acknowledged. The numbers in the abstract match the numbers for the propensity score-matched cohort from Table 2. We chose this as our main analysis and presented the imputed results as sensitivity analyses.