|
Activity Number:
|
72
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #305482 |
|
Title:
|
A Method for Testing a Prespecified Subgroup in Clinical Trials
|
|
Author(s):
|
Yang Song*+ and George Chi
|
|
Companies:
|
Johnson & Johnson Pharmaceutical R&D and Johnson & Johnson Pharmaceutical R&D
|
|
Address:
|
920 Route 202, S., Raritan, NJ, 08869,
|
|
Keywords:
|
subgroup analysis ; type I error rate ; closed testing procedure ; alpha allocation ; adjusted p-values
|
|
Abstract:
|
In clinical trials, investigators often are interested in the effect of a given study treatment on a subgroup of patients with certain clinical or biological attributes in addition to its effect on the overall population. Such a subgroup analysis would become more important if an efficacy claim could be made for the subgroup when the test for the overall study population fails at a prespecified alpha level. In practice, such a claim is often dependent on prespecification of the subgroup and certain implicit or explicit requirements placed on the study results. By carefully considering these requirements, we propose a general statistical methodology for testing both the overall and subgroup hypotheses, which has optimal power and strongly controls the family-wise type I error rate.
|
