|
Activity Number:
|
653
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 4, 2016 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
|
Abstract #321243
|
View Presentation
|
|
Title:
|
Joint Modeling of Longitudinal Cognitive Responses and Survival Time in an Alzheimer's Disease Cohort Study
|
|
Author(s):
|
Eveleen Darby* and Wenyaw Chan and Elaine Symanski and Rachelle S. Doody
|
|
Companies:
|
Baylor College of Medicine and The University of Texas Health Science Center at Houston and The University of Texas Health Science Center at Houston and Baylor College of Medicine
|
|
Keywords:
|
Joint model ;
Longitudinal process ;
Survival analysis ;
Alzheimer's disease ;
Dementia
|
|
Abstract:
|
Predictors of survival among persons with Alzheimer's disease (AD) and dementia as well as estimates of predicted survival have been studied extensively. Many observational studies also examined factors that influence survival and progression of cognitive measurements. Joint modeling of event time data and longitudinal data has been shown to be more efficient on making these predictions. In this study, we applied joint modeling approaches to analyze mortality among persons with AD adjusting for baseline characteristics and time-dependent cognitive function measurements: (1) Joint model with baseline hazard modeled as a piecewise-constant function (2) Joint model with baseline hazard modeled as Weibull survival function and (3) time-dependent last value carry forward Cox model. All joint modeling methods show a significant association of the repeated measures with the survival process. Education in years, which did not have significant association in the time-dependent Cox model prior to joint model analysis, was found to be associated with the risk of death in the joint models. A time-dependent Cox model showed different result from that in the joint model analysis.
|
Authors who are presenting talks have a * after their name.
