|
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 - #309352 |
|
Title:
|
Comparison of Adjusted Survival Curves with Different Reference Populations
|
|
Author(s):
|
Xu Zhang*+
|
|
Companies:
|
Georgia State University
|
|
Address:
|
30 Pryor Street 7th floor, Atlanta, GA, 30309,
|
|
Keywords:
|
Observational study ; Selection bias ; Cox model
|
|
Abstract:
|
The direct adjustment method is often used to construct the adjusted survival probabilities of different treatments, when imbalance of prognostic factors is present in the failure time data. The whole study cohort is the commonly selected reference population on which one applies direct adjustment. In this paper, we consider the problem of reference population selection. We apply the proposed procedures to real data and have the following conclusions. 1. In observational studies, the treatment groups may have distinct survival outcomes due to imbalance of prognostic factors. 2. When one wishes to compare the survival probabilities between two treatments, the comparison result is robust to various reference populations. 3. The group with inferior treatment can be used as the reference population to calculate the adjusted survival probabilities.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2007 program |
