|
Activity Number:
|
199
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #309887 |
|
Title:
|
Robust Assessment of Survival Differences Using Multiple Testing Procedures
|
|
Author(s):
|
Larry Ma*+ and Devan V. Mehrotra and Woollcott Smith
|
|
Companies:
|
Temple University and Merck & Co., Inc. and Temple University
|
|
Address:
|
129 Aileen Dr, Lansdale, PA, 19446,
|
|
Keywords:
|
Time-to-Event Data ; Survival Difference ; Weighted Log-Rank Statistics ; Multiple Testing Procedures ; Robustness
|
|
Abstract:
|
The log-rank test is optimal for assessing differences between two survival functions when the hazards are proportional to each other. However, if the proportional hazards assumption is not tenable, a weighted log-rank test is often used in which the pre-specified weight function assigns more weight to early or late events. Since the performance of a single weighted log-rank statistic is sensitive to the choice of the weight function, we suggest the simultaneous use of multiple weighted log-rank statistics when some prior information is available or there is no prior information at all. The equality of the two survival functions is tested using a pre-specified multiplicity adjustment. The MTPs that we consider include extensions of Fisher's and Simes' methods to the setting of correlated test statistics. We make recommendations based on simulation-based test size and power results.
|
