|
Activity Number:
|
24
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #306011 |
|
Title:
|
Assessing Treatment Differences Adjusted by Possible Carryover Effects in Crossover Clinical Trials
|
|
Author(s):
|
Ling Chen*+
|
|
Companies:
|
U.S. Food and Drug Administration
|
|
Address:
|
10903 New Hampshire Ave., Silver Spring, MD, 20993,
|
|
Keywords:
|
crossover design ; carryover effects ; multiple treatments ; Williams design
|
|
Abstract:
|
In crossover clinical trials the study model often includes the first-order carryover as a fixed effect. Using conventional methodology we are not able to accurately detect treatment differences if carryover effects exist. A test for no first-order carryover effect is usually performed before testing for the treatment difference. The test is likely to have low power to detect a significant difference. In addition, Freeman (1989) showed that this two-stage procedure of testing for a carryover difference in the first stage and then for a direct treatment difference in the second stage not only inflated the probability of making a type I error, but also produced a biased estimate of the direct treatment difference. In this article, I propose a new test for treatment differences adjusted by possible first-order carryover effects in crossover designed studies for three or more treatments.
|
