The Analysis of Recurrent Event Data in the Presence of Competing Risks and Informative Censoring

• Activity Number: 117
• Type: Topic Contributed
• Date/Time: Monday, August 5, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Lee-Jen Wei*+
• Companies: Harvard University

Abstract

Comparative effectiveness in the presence of competing risks for multiple event time observations L.J. Wei for B. Claggett, L. Tian, and D. Castagno Abstract In this talk, we will consider the case that each study subject potentially may have several distinct outcomes (events) during the study period. Moreover, the times to the events may not be observed completely due to competing risks. We will review statistical methods which deal with such complex data structure, especially for cases with recurrent event time analysis. We will show the inadequacy of the conventional procedures, for example, the typical analysis using the first event time observation or its so-called component analysis. We then propose a new approach to handle multiple outcomes in the presence of competing risks. Furthermore, we show how to systematically identify a subgroup of subjects which may benefit from the new treatment using data from a randomized, comparative clinical trial with the patient's baseline covariates. All the procedures are illustrated with the data from a cardiovascular study for evaluating a beta blocker treating chronic heart failure.