|
Activity Number:
|
426
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #304768 |
|
Title:
|
Competing Risks Regression for Clustered Data
|
|
Author(s):
|
Bingqing Zhou*+
|
|
Companies:
|
The University of North Carolina at Chapel Hill
|
|
Address:
|
Department of Biostatistics, Chapel Hill, NC, 27599,
|
|
Keywords:
|
Competing Risk ; Hazard of subdistribution ; Partial likelihood ; Clustered
|
|
Abstract:
|
A clustered competing risks regression is proposed to assess the effect of covariates on the cumulative incidence function when there is clustering in competing risks setting. This method extends Fine-Gray proportional hazards model for subdistribution (1999) to accommodate situations where the failure times within a cluster might be correlated since the study subjects from the same cluster share common factors. Estimators of regression parameters in the Fine-Gray model under the independence working assumption are still consistent and asymptotically normal. However, the variance-covariance matrix and its consistent estimate cannot be obtained without addressing the correlation in the data. The comparisons of size and power are conducted to show the necessity of the proposed approach. The method is illustrated by data from European bone marrow transplantation registry.
|
