|
Activity Number:
|
224
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #303935 |
|
Title:
|
Handling Baseline Responses in Repeated Measures Analyses with Data Missing at Random
|
|
Author(s):
|
Phillip Dinh*+ and Peiling Yang
|
|
Companies:
|
FDA and FDA
|
|
Address:
|
10903 New Hampshire Ave, Silver Spring, MD, 20993-0002,
|
|
Keywords:
|
Baseline ; Longitudinal Data ; Missing Data ; MMRM
|
|
Abstract:
|
The objective of this paper is to compare five strategies for handling baseline responses in an MMRM analysis when missing data are present. The five strategies are: 1) retain the baseline response as part of the outcome vector and make no assumptions about group differences in the mean response at baseline; 2) retain the baseline as part of the outcome vector and assume the group means are equal at baseline; 3) subtract the baseline from all of the remaining post-baseline responses, and analyze the differences from baseline; 4) Use the baseline as a covariate in the analysis of the post-baseline responses, assuming homogeneous regression slopes; 5) Use the baseline as a covariate in the analysis of the post-baseline responses, allowing different regression slopes. We will use simulations to evaluate the approaches based on the bias and the coverage accuracy of the confidence interval.
|
