|
Activity Number:
|
256
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #304118 |
|
Title:
|
Stratification in Design vs. Stratification in Analysis
|
|
Author(s):
|
Byron Jones*+ and Neal Thomas
|
|
Companies:
|
Pfizer Inc. and Pfizer Inc.
|
|
Address:
|
Ramsgate Road, Sandwich, International, CT13 9NJ, United Kingdom
|
|
Keywords:
|
stratification ; covariate-adjustment ; analysis-vs-design ; causal inference ; current practice
|
|
Abstract:
|
Debate on the role of covariate adjustment (stratification) in clinical trials has a long history with advocates for and against adjustment. We will review this history with emphasis on the distinction between stratification in the design versus stratification in the analysis, and differences in opinion arising due to differences in assumptions regarding the additivity of treatment effect. We will also survey recent methodological development in covariate adjustment, much of it coming from the causal inference literature. We conclude with comments on current practice as we have observed it, and some thoughts on potential changes to current practice.
|
