|
Activity Number:
|
504
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #304201 |
|
Title:
|
Principal Stratification for Evaluation of Surrogate Endpoints with Variable Control-Group Response
|
|
Author(s):
|
Corwin M. Zigler*+ and Thomas R. Belin
|
|
Companies:
|
University of California, Los Angeles and University of California, Los Angeles
|
|
Address:
|
Department of Biostatistics, Los Angeles, CA, 90024,
|
|
Keywords:
|
principal stratification ; potential outcomes ; surrogate marker ; biomarker
|
|
Abstract:
|
The principal stratification framework has contributed the definition of a "principal surrogate" to the ongoing research regarding evaluation of surrogate markers that can be used in place of clinical outcomes. Work by Gilbert and Hudgens (Biometrics, 2008) proposes evaluation of potential surrogates by their "Causal Effect Predictiveness (CEP)" surface in a context that lends itself to the assumption of constant surrogate response in the absence of treatment. Such an assumption alleviates the challenge of identifying the CEP surface through imputation of potential surrogate outcomes under treatments not given. We propose a flexible Bayesian modeling approach that accommodates less restrictive assumptions about missing potential outcomes, potentially extending practicality of estimation of the CEP surface to more general settings that involve variable control-group response.
|
