Dr. Xu et al. compared cognitive trajectories and the occurrence of severe dementia or death in 11,652 patients started on cholinesterase inhibitors (ChEIs) vs 5,826 untreated patients with Alzheimer dementia (AD) in the Swedish Dementia Registry. During an average of 5 years of follow-up, they found that ChEIs were associated with modest but persistent cognitive benefits with lower mortality. Galantamine was the only ChEI that was associated with a significantly lower risk of developing severe dementia. In response, Dr. Kuschpel noted that there were important baseline differences between patients who were and were not treated with ChEIs, with associated difficulties matching the 2 groups for analyses. Dr. Kuschpel also highlighted the importance of considering the stage and severity of comorbidities about survival estimates and is skeptical of the sizeable mortality reduction with ChEIs seen in the study, given the lack of precedence, the absence of a plausible molecular mechanism, and potential confounding by prescription bias. Concerns are raised about potential discrepancies in numbers between the abstract and the results section about observed differences in Mini-Mental State Examination (MMSE) scores. Responding to these comments, the authors argue that although there were significant baseline differences between the 2 groups, they were able to achieve a well-matched cohort, although they acknowledge the possibility of residual and unknown confounders, especially in the absence of data on comorbidity severity. They cite the previous work by their group observing reductions in myocardial infarction, stroke, and mortality with ChEI use, as well as a lower 2-year mortality with galantamine seen in a prior randomized-controlled trial. They note that previous trials of ChEIs have had considerably shorter follow-up durations. In summary, this exchange highlights the uncertain causal association between ChEI use and lower mortality in patients with AD, and the challenges of convincingly demonstrating such an association with observational data. Robust validation of this association would have important implications for the care of patients with AD.