Online Program Home
My Program

Abstract Details

Activity Number: 277 - Statistical Methods for Composite Time-To-Event Endpoints
Type: Invited
Date/Time: Tuesday, July 30, 2019 : 8:30 AM to 10:20 AM
Sponsor: Lifetime Data Science Section
Abstract #300163 Presentation
Title: Some Meaningful Weighted Win Loss Statistics
Author(s): Xiaodong Luo* and Hui Quan
Companies: and Sanofi US
Keywords: win ratio; RMST; multiple testing; composite endpoints

It is often difficult to summarize treatment effect in survival trials in the presence of non-proportional hazards. It is even harder to do so when there is more than one endpoints (PFS and OS, for example) in the trial. In this talk, I will introduce some meaningful treatment effect measurements that are derived from weighted log-rank and weighted win loss statistics. I will illustrate that the use of such statistics will provide a coherent way to conduct hypothesis testing and treatment effect estimation in survival trials. Such statistics can also alleviate the multiple testing issues of the commonly used weighted log-rank statistics and the recently proposed (un-weighted) win ratio statistic in the presence of non-proportional hazards and/or multiple endpoints. I will also demonstrate some potential power loss in hypothesis testing as a trade-off for having a more interpretable test statistic.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program