Activity Number:
|
335
|
Type:
|
Luncheons
|
Date/Time:
|
Wednesday, August 14, 2002 : 12:30 PM to 1:50 PM
|
Sponsor:
|
Biopharmaceutical Section*
|
Abstract - #302048 |
Title:
|
Accounting for Sites in the Analysis of Data from Multicenter Clinical Trials
|
Author(s):
|
Paul Gallo*+
|
Affiliation(s):
|
Novartis Pharmaceuticals
|
Address:
|
One Health Plaza, East Hanover, New Jersey, 07936,
|
Keywords:
|
|
Abstract:
|
Questions regarding how clinical sites should be accounted for in analysis models for data from multicenter trials comprise an interesting, complex, and frequently controversial issue. Specific questions may include: whether center or treatment-by-center interaction should be routinely accounted for in primary analysis models; whether these should be considered fixed or random effects; whether artificial composite centers should be defined; how analysis results on treatment effects should be interpreted in the presence of apparent or possible interaction with site. These issues will be discussed not only in the linear model framework, which has been the basis for much of the historical discussion on this topic, but also for different types of analyses and more general data structures.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2002 program |