|
Activity Number:
|
472
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 9, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #307419 |
|
Title:
|
Assessing Learning Effect and Nonrandom Dropout in a Contraceptive Device Trial
|
|
Author(s):
|
Pai-Lien Chen*+
|
|
Companies:
|
Family Health International
|
|
Address:
|
P.O. Box 13950, RTP, NC, 27709,
|
|
Keywords:
|
learning effect ; nonrandom dropout ; transition model
|
|
Abstract:
|
Participants in the new device group of a randomized control device trial might need time to become skilled at using the device. In addition, some participants who are not used to the new device might consequently drop out from the study. As a result, the learning effect and the nonrandom dropout can substantially influence the evaluation of treatment effects. Without considering those factors, the conclusion of the trail is vulnerable to potential bias. In this study we propose a transition model to assess the effect of learning and of nonrandom dropout from a randomized control device trial. Data from a randomized condom contraceptive trail is used to illustrate the proposed method. The impact of learning effect and nonrandom dropout of this condom trail on its surrogate endpoints and on the study conclusion regarding the primary endpoint are discussed.
|
