Robust Estimators for the Evaluation of Multiple Surrogate Markers with Censored Data

• Activity Number: 634 - Dynamic Modeling for Timely Health Care Decisions
• Type: Invited
• Date/Time: Thursday, August 3, 2017 : 10:30 AM to 12:20 PM
• Sponsor: WNAR
• Author(s): Layla Parast* and Tianxi Cai and Lu Tian
• Companies: RAND Corporation and Harvard University and Stanford University
• Keywords: survival ; surrogate ; landmark ; double robust

Abstract

The utilization of surrogate markers offers the opportunity to reduce the length of required follow-up time and/or costs of a randomized trial examining the effectiveness of an intervention or treatment. There are many available methods for evaluating the utility of a single surrogate marker including both parametric and nonparametric approaches. However, when the dimension of the surrogate marker is greater than two, a completely nonparametric procedure is infeasible due to the curse of dimensionality. In this paper, we define a quantity to assess the value of multiple surrogate markers in a time-to-event outcome setting and propose three different robust estimation approaches for censored data. The first approach is based on a dimension reduction procedure while the second approach relies on a weighted estimator. The third approach combines components of the first two approaches yielding a double robust estimator. We examine the finite sample performance of all three estimators under various scenarios using a simulation study. We illustrate the estimation and inference procedures using data from the Diabetes Prevention Program (DPP).