|
Activity Number:
|
409
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #308328 |
|
Title:
|
Optimal Crossover Designs: Investigating the Role of Baseline Measurements
|
|
Author(s):
|
Yuanyuan Liang*+ and Kimmie C. Carriere
|
|
Companies:
|
University of Alberta and University of Alberta
|
|
Address:
|
CAB 632, Edmonton, AB, T6G 2G1, Canada
|
|
Keywords:
|
optimal design ; baseline measurements ; self and mixed carryover effects model ; random subject effects ; Lagrange multiplier
|
|
Abstract:
|
In this paper, we apply the Lagrange multiplier method to solve the optimal design problem under a more general model: self and mixed carryover effects model with random subject effects. Generally, crossover designs lose the practical appeal when required to allow more than the traditional first-order carryover effects. This is when utilizing the baseline measurements clearly saves the design efficiency. We explore the effect of baseline measurements on optimal design results. We strongly recommend to use the baseline measurements as they improve the efficiency by 2-3 folds. We also find that there is a dramatic reduction in variability for estimating the direct treatment effect contrast when extending two-period designs to three-period or four-period designs. Such a reduction is much more pronounced when baseline measurements are incorporated in each period.
|
