What is it about?
With two time-to-event outcomes, one has some options on how to combine them into a composite endpoint. However, when one of the outcomes is not continuous, there are fewer options left. One of these options is to use a rank test that combines the outcomes a very specific way: worst-rank composite endpoint. In this paper, we show how weights can be associated with the different components of worst-rank composite endpoint to evaluate the overall treatment effect. Then, we demonstrate how the related test which is based on the so-familiar idea of the Wilcoxon test (also referred to as Mann--Whitney test) can be conducted.
Featured Image
Why is it important?
Ignoring death or not accounting for the informative nature of truncation due to death leads spurious and untrustworthy results. This is the traditional survivor bias. To reduce such a survivor bias in the estimation of the treatment effect, appropriate measures need to be taken to guarantee a fair and efficient assessment of the overall treatment effect.l
Perspectives
There have been many papers highlighting the inadequacy of using the traditional "time-to-first-event" definition of composite endpoints since it may not capture the entire patients' experience to evaluate the overall treatment effect, especially in multi-faceted diseases where multiple outcomes are needed to run comparative of effectiveness analyses. This method steers away from the time-to-first event paradigm by combining mortality and non-fatal outcome while weighting appropriately their underlying contributions.
Dr. Roland A. Matsouaka
Duke University
Read the Original
This page is a summary of: An optimal Wilcoxon–Mann–Whitney test of mortality and a continuous outcome, Statistical Methods in Medical Research, December 2016, SAGE Publications,
DOI: 10.1177/0962280216680524.
You can read the full text:
Contributors
The following have contributed to this page
