Landmark Estimation of Survival Incorporating Intermediate Event Information in a Randomized Clinical Trial

• Activity Number: 573
• Type: Contributed
• Date/Time: Wednesday, August 1, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Layla Parast*+ and Tianxi Cai
• Companies: Harvard University and Harvard University
• Keywords: Efficiency Augmentation ; Landmark Prediction ; Semi-competing risks ; Survival analysis

Abstract

In many studies with a survival outcome, it is often too expensive, time consuming, or unethical to fully observe the primary event of interest in all subjects being studied. This often leads to heavily censored data and difficulty in efficiently estimating survival. In certain diseases, the event time of a non-fatal intermediate event may be associated with overall survival. In these settings, incorporating intermediate event information may lead to gains in efficiency in terms of estimation of survival and testing for a difference in survival between two treatment groups. If gains in efficiency can be achieved, it may then be possible to decrease the sample size of patients required for a study or decrease the duration of the study. We propose a semi-nonparametric two-stage procedure to estimate survival by incorporating intermediate event information observed before some landmark time. In addition, we present a testing procedure using these resulting estimates to test for a difference in survival between two treatment groups in a RCT setting. Simulation studies demonstrate substantial potential gains in efficiency in terms of estimation and power.