|
Activity Number:
|
56
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #305805 |
|
Title:
|
A Bayesian Pooled Analysis of Doubly Censored HIV Data Using the Hierarchical Cox Model
|
|
Author(s):
|
Wei Zhang*+ and Kathryn Chaloner and Ying Zhang and Mary K. Cowles
|
|
Companies:
|
Boehringer Ingelheim and The University of Iowa and The University of Iowa and The University of Iowa
|
|
Address:
|
900 Ridgebury Road, Ridgefield, CT, 06810,
|
|
Keywords:
|
doubly censored data ; hierarchical Cox model ; MCMC methods ; imputation
|
|
Abstract:
|
Two common statistical problems in pooling survival data from several studies are addressed here in the context of a real case study of HIV-infected individuals. The first problem is the data are collected from multiple studies, and it is likely that heterogeneity exists among the study populations. A random-effects hierarchical Cox proportional hazards model is therefore used to perform the pooled analysis. The second problem is the data are doubly censored in that the HIV infection time is interval-censored and the death event may be right-censored. Two approaches to incorporating the uncertainty of interval-censored HIV infection time are developed and compared to a more usual analysis using imputation. We recommend the use of the fully Bayesian approach, which adequately incorporates the uncertainty of interval-censored HIV infection times.
|
