I read the article by Xu et al.1 with interest. However, I would like to raise a few questions about the study. The published literature shows a strong difference between patients who received a prescription for a cholinesterase inhibitor (ChEI) and those who did not.2 The authors also mention that they had trouble matching the 2 groups because their groups differed significantly. Although the authors tried to correct for this, it does not seem that the stage or severity of comorbidities was considered in this study, despite its critical clinical relevance to survival estimates. The mortality reduction found by the authors was immense and has not been found in previous prospective studies.3,4 At this time, there is no known molecular mechanism that could explain such a strong effect. Could this result be a simple case of reverse causality, based on an inadequately control for prescription bias?