Problems With Clinical Application of Low-Dose Vasopressin for Traumatic Hemorrhagic Shock —Reply

In Reply Thank you for your interest in our article, “Effect of Low-Dose Supplementation of Arginine Vasopressin on Need for Blood Product Transfusions in Patients With Trauma and Hemorrhagic Shock.” We believe Gauss et al have misread our statistical approach and are incorrectly interpreting the absolute standardized differences (ASD) as P values . There is a growing momentum to use P value alternatives, such as ASDs, in clinical research. Absolute standardized differences assess the magnitude of differences between groups (specifically, the absolute value of the difference in means, mean ranks, or proportions divided by the pooled standard deviation). We used ASD rather than standard significance tests because using multiple significant tests to evaluate baseline variables can be misleading. Each test carries a 5% probability of type 1 error; with multiple statistical tests, the type I error rate becomes exaggerated, resulting in an i nflated probability of a significant difference between groups based on random chance alone. The use of ASD mitigates the risk of amplifying type 1 errors. Groups were considered imbalanced for any variable that had absolute standardized differences greater than 0.392. All baseline variables had an ASD less than 0.392, suggesting that groups were not imbalanced for these variables.
Source: JAMA Surgery - Category: Sports Medicine Source Type: research