|
Title
|
|
* ! Adaptive Sample-Size Considerations in the Presence of Unknown Treatment Effect and Nuisance Parameters
|
|
Date / Time / Room
|
Sponsor
|
Type
|
08/12/2002
2:00 PM -
3:50 PM
Room: H-West Ballroom
|
Biopharmaceutical Section*, International Indian Statisticial Association, WNAR
|
Invited
|
|
Organizer:
|
Cyrus Mehta, Cytel Software Corporation
|
|
Chair:
|
Anastasios A. Tsiatis, North Carolina State University
|
|
Discussant:
|
3:25 PM - Kuang-Kuo G. Lan, Pfizer, Inc.
|
|
Floor Discussion
|
3:40 PM
|
Description
There has been considerable recent interest in achieving greater flexibility in sample size calculations at the design stage of a
study, either because the sponsor is very uncertain about the underlying effect size or because of uncertainty about nuisance parameters like variance, baseline response rate or baseline hazard rate. Adaptive sample size calculations that preserve the type-1 error and power are possible in both settings, especially in conjunction with group sequential clinical trials where the data are monitored routinely anyway. However, such adaptive designs raise important technical and philosophical questions that warrant discussion.This session will present some of the latest research on this topic by researchers from industry, academia, the NIH and the FDA. The discussant is an acknowledged leader in this field.
|
|
