|
Activity Number:
|
429
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #304566 |
|
Title:
|
An Application of GEE Models in Assessing Noninferiority to an Aggregate Response
|
|
Author(s):
|
G. Frank Liu and Michael J. Dallas*+
|
|
Companies:
|
Merck Research Laboratories and Merck Research Laboratories
|
|
Address:
|
, North Wales, PA, 19454,
|
|
Keywords:
|
non-inferiority ; aggregate response ; GEE
|
|
Abstract:
|
Assessing non-inferiority to an aggregate response arises in the development of new vaccines that are comprised of some components that are identical to an existing vaccine and some components that are new. Among the new components, guidance states that each individual immune response from the new vaccine should be compared with the aggregate response from the existing vaccine to evaluate non-inferiority. This results in an interesting statistical problem. We focus on solving this problem using generalized estimating equations (GEE) to estimate the aggregate response. Consideration will be given to the choice of correlation matrix and link function. The estimated aggregate response is then compared with the response of each new component based on effective sample size. The methods will be evaluated using simulations and demonstrated with a real-life example.
|
