|
Activity Number:
|
539
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 2, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #310325 |
|
Title:
|
Evaluating Subgroup Risks in Trials with Many Subgroups: Issues and Potential Solutions
|
|
Author(s):
|
Xiaojiang Zhan*+ and Amarjot Kaur
|
|
Companies:
|
Merck & Co., Inc. and Merck & Co., Inc.
|
|
Address:
|
126 E Lincoln Ave, Rahway, NJ, 07065,
|
|
Keywords:
|
multiplicity ; subgroup analysis ; bias-correction ; hazard ratio ; estimation ; clinical trials
|
|
Abstract:
|
In the analysis of clinical trials it is not uncommon to examine response to therapy in various patient subgroups in order to understand any differential in treatment response. ICH guidance for the common technical documents and many individual therapeutic area regulatory guidance also ask for subgroup analyses. In a typical subgroup analysis of risk, the estimates of risk are provided for multiple subgroups without accounting for multiplicity. Although the interest in examining multiple subgroups is valid to advance medical research, there is inherent cost (increase in Type I or Type II errors) associated with examining multiple subgroups. In this investigation the multiplicity issue is examined using time-to-event data. We characterize the magnitude of problem in terms of observed bias in estimating the hazard ratios for multiple subgroups and propose a potential correction.
|
