|
Activity Number:
|
213
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
ENAR
|
| Abstract - #305508 |
|
Title:
|
Joint Modeling of Longitudinal and Time-to-Event Data with Random Changepoints
|
|
Author(s):
|
Chengjie Xiong and Yuan Xu
|
|
Companies:
|
Washington University and Washington University
|
|
Address:
|
|
|
Keywords:
|
joint modeling ; mixed models ; survival models ; Alzheimer's disease
|
|
Abstract:
|
Longitudinally observed biomarkers often provide crucial information on the antecedent progression of many diseases such as Alzheimer's disease (AD) and HIV. We propose a joint model of longitudinal biomarker data and time to disease onset which allows a possible antecedent change on the rate of biomarker changes prior to the disease onset. The proposed model is based on the standard general linear mixed models for longitudinal data and the standard survival models for time to event data. We provide estimates to the changepoints as well as to the rate of biomarker changes. The proposed model is then demonstrated by using a real world study that seeks to understand the antecedent cognitive changes before the onset of Alzheimer's disease.
|