|
Activity Number:
|
296
|
|
Type:
|
Luncheons
|
|
Date/Time:
|
Tuesday, August 8, 2006 : 12:30 PM to 1:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #305432 |
|
Title:
|
Retrospective Testing for Baseline Variables Comparability and Subsequent Covariate Adjustments for Significant Imbalance: Good Clinical Practice or Bad Statistical Practice?
|
|
Author(s):
|
Abdul Sankoh*+
|
|
Companies:
|
sanofi-aventis
|
|
Address:
|
200 Crossing Blvd., Bridgewater, NJ, 08807,
|
|
Keywords:
|
baseline variables ; covariate adjustment ; significant imbalance
|
|
Abstract:
|
Baseline covariate adjustment analyses often are carried out in randomized clinical trials when there is evidence that the treatment difference depends on certain patient characteristics for two main reasons. The first of these is to adjust for imbalance in baseline variables related to the outcome; such adjustment helps correct for the groups' predisposition to behave differently from the outset. The second reason is to reduce the underlying variability in the data so more precise comparisons can be made. It is, however, well-known that post-randomization testing for baseline imbalance and subsequent covariate adjustments for such imbalance has no theoretical justification, notwithstanding its practical appeal.
|
- 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 2006 program |
