|
Activity Number:
|
94
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #308784 |
|
Title:
|
Exploration of a Treatment-by-Region Interaction in a Clinical Trial Setting: Can the Interaction Be Explained by Clinical Factors?
|
|
Author(s):
|
Cristiana Gassmann-Mayer*+ and Rachel B. Weinstein and Jesse A. Berlin
|
|
Companies:
|
Johnson & Johnson PRD and Johnson & Johnson PRD and Johnson & Johnson PRD
|
|
Address:
|
1125 Trenton Harbourton Road, Titusville, NJ, 08560,
|
|
Keywords:
|
treatment-by-baseline factor interaction ; subgroup analyses ; clinical trial ; post-hoc analyses
|
|
Abstract:
|
An important clinical question in clinical trials is the evaluation of treatment-by-baseline factor interactions. Differences in treatment effects across subgroups are often difficult to anticipate a priori and post hoc differences in treatment effect across subgroups are difficult to interpret. When significant treatment-by-factor interactions are observed, further data investigations should be undertaken to understand the nature and sources of the interaction. Exploratory analyses including statistical modeling and subgroup analyses should be performed with a clear understanding of power and interpretation issues. The controversial aspect of post-hoc exploratory analysis will be illustrated by an example in a clinical trial setting with a treatment-by-region interaction. Modeling strategies will be discussed, along with effective graphical and tabular data presentations of results.
|
