Stratified Win Ratio and Handling of Ties

• 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
• Author(s): Gaohong Dong and Junshan Qiu and Roland A. Matsouaka and Victoria Chang and Jiuzhou Wang and David C. Hoaglin* and Marc Vandemeulebroecke
• Companies: iStats Inc. and FDA/CDER and Duke University School of Medicine and Abbvie and ImmunoGen Inc. and University of Massachusetts Medical School and Novartis Phama AG
• Keywords: Win ratio; Win odds; Mantel-Haenszel method; Prioritized pairwise comparisons; Prioritized multiple outcomes; Composite endpoint

Abstract

A composite endpoint combining multiple outcomes is frequently used for clinical trials. Conventional time-to-first-event analyses have several limitations as the first event analyzed is often a less important, but more frequent outcome. The win ratio (Pocock et al., 2012) addresses these limitations; it considers the clinical importance order and the relative time sequence of the outcomes. In this talk, we consider stratified designs and the stratified win ratio (Dong, Qiu, Wang and Vandemeulebroecke, 2018), constructed similarly to the Mantel-Haenszel stratified odds ratio. As win ratio analyses usually involve many more ties than wins, we present the win odds (Dong et al., 2019), which assigns 50% of ties to the numerator and 50% to the denominator. We illustrate this statistic and compare it with the win ratio using real data from clinical studies.