|
Activity Number:
|
480
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #309264 |
|
Title:
|
Semiparametric Methods for Estimating Cumulative Treatment Effects
|
|
Author(s):
|
Guanghui Wei*+ and Douglas Schaubel
|
|
Companies:
|
University of Michigan and University of Michigan
|
|
Address:
|
Department of Biostatistics, Ann Arbor, MI, 48109,
|
|
Keywords:
|
Confidence bands ; Empirical processes ; Integrated hazard ; Stratification ; Survival analysis ; Time-dependent effect
|
|
Abstract:
|
Often in biomedical studies of time to an event, the treatment effect is not constant over time. In such cases, the cumulative treatment effect is usually of much greater importance than the instantaneous effect. We propose semiparametric methods for estimating the cumulative treatment effect in the presence of non-proportional hazards. The proposed estimator is based on the treatment-specific cumulative hazards estimated under a stratified Cox model. No functional form for the non-proportionality need be assumed. Asymptotic properties of the proposed estimators are derived and assessed through simulation. Simultaneous confidence bands are developed. The proposed method is applied to national organ failure and registry data. Extensions of the proposed methods to the recurrent event setting are considered.
|
