|
Activity Number:
|
267
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 5, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #302012 |
|
Title:
|
Semiparametric Methods for the Analysis of Clustered Survival Data from Case-Cohort Studies
|
|
Author(s):
|
Hui Zhang*+ and Douglas E. Schaubel and John Kalbfleisch
|
|
Companies:
|
The University of Michigan and The University of Michigan and The University of Michigan
|
|
Address:
|
1420 Washington Heights, Ann Arbor, MI, 48105,
|
|
Keywords:
|
case-cohort study ; clustered data ; Cox model ; estimating equation ; robust variance ; survival analysis
|
|
Abstract:
|
Case-cohort sampling is often employed in large cohort studies. Although most existing methods apply to univariate survival data, subjects are often clustered in epidemiologic studies. We propose estimating equation based methods for three case-cohort designs that are applicable to clustered survival data. A marginal model is assumed, with a baseline hazard and regression coefficient that are common across clusters. The proposed regression parameter and cumulative hazard estimators are consistent and asymptotically normal, with variances that can be estimated empirically. The proposed procedures have increased efficiency relative to some existing methods. The proposed estimators are evaluated analytically, investigated through simulation, and then applied to a study of mortality among Canadian dialysis patients.
|
