Online Program Home
My Program

Abstract Details

Activity Number: 262
Type: Contributed
Date/Time: Monday, August 1, 2016 : 2:00 PM to 3:50 PM
Sponsor: Biometrics Section
Abstract #320781
Title: Use Nonlinear Mixed Effects Model to Predict Unobservable AD Symptom Duration
Author(s): Ho-Lan Peng* and Wenyaw Chan and Rachelle S. Doody
Companies: and The University of Texas Health Science Center at Houston and Baylor College of Medicine
Keywords: Nonlinear Mixed Model ; EM Algorithm

Alzheimer's disease (AD), a common form of dementia, is one of the costliest chronic diseases to society. Knowing the pre-clinic symptom duration defined as symptom onset until the first clinic visit, may help understanding the disease progression and planning for patient care. Researchers use this duration to calculate AD onset age, a key but not easily measured predictor of AD progression. In this research, we propose a statistical method using longitudinal neuropsychological measurements to predict the pre-clinic symptom duration and thus help understand the profile of AD severity. Several studies reported that neuropsychological measurements and symptom duration are nonlinearly related. We will apply a joint non-linear mixed effect model using each neuropsychological measurement as the outcome to predict the symptom duration and estimate other parameters in the model by using EM algorithm. This model is also adjusted by some demographic characteristics and baseline clinical variables. Simulation study has also been conducted to verify our estimation method.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

Copyright © American Statistical Association