Mortality has long been used as a primary end point for randomized controlled trials in critical care. Recently, a plurality of trials targeting mortality end points as their primary outcome has failed to detect a difference between study arms. While there are a number of reasons for the preponderance of such neutral trials, the use of mortality as an outcome is one important consideration. We explore some of the reasons why such trials may be biased toward a neutral result, as well as reasons to consider alternative end points that are better coupled to the expected therapeutic effect. We also discuss to what extent mortality as a binary outcome is patient-important in the ICU.
Keywords: critical care; mortality outcomes; neutral trials; patient-important outcomes; randomized controlled trials.
Copyright © 2019 American College of Chest Physicians. Published by Elsevier Inc. All rights reserved.