LAUSR

In Reply: We would like to thank Dr Chen and Dr You for their interest in our published article prospectively comparing the performance of the AIMS65 score and the Glasgow-Blatchford risk score in predicting important outcomes in upper gastrointestinal hemorrhage (UGIH).1 The authors correctly state that previous studies have identified multiple factors (ie, ethnicity, age, etc.) that affect UGIH prognosis. However, unlike what the authors suggest, because the 2 scores were compared using the same patient population, there is actually no need to adjust for the influence of those factors. The distribution of those factors is identical for both the AIMS65 score and Glasgow-Blatchford risk score, and thus those factors do not affect the outcomes compared. This is the same statistical principle that makes adjustment for confounders in large randomized controlled trial unnecessary: The randomization process results in an equal distribution of measured and unmeasured confounders among cases and controls and, as a result, those confounders no longer affect the outcomes. The authors also correctly state that our study population was from 3 centers located in Massachusetts. Thus, it is possible that the findings of our study may not be directly generalizable to other patient populations. This is a limitation we mention in the discussion section of our article. We welcome and support Dr Chen and Dr You’s call for more comparative studies among the existing scores, as evidence-based practice is only as valid as the evidence that supports it. However, we also would like to point out that at least 2 other articles published since ours was published support our findings in very different patient populations (Australia and Spain).2,3 The national and international UGIH guidelines recommend the use of validated risk stratification scores to help in early patient triage. The AIMS65 score has gained increasing attention in the literature as a valid and easy to remember and calculate tool to achieve this goal. Our understanding of its performance and generalizability compared with other UGIH scores has been evolving with the increasing number of comparative studies from different locations. We look forward to additional prospective comparative studies that help us understand how best to use these scores and better manage our patients with UGIH.