|
Activity Number:
|
355
- Contributed Poster Presentations: Section on Bayesian Statistical Science
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 1, 2017 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
|
Abstract #324172
|
|
|
Title:
|
Joint Modeling of Multiple Longitudinal Ordinal Outcomes and Competing Risks: An Application to Parkinson's Disease Study
|
|
Author(s):
|
Jun Zhang* and Shen Luo
|
|
Companies:
|
UT Health Science Cener and UT School of Public Health
|
|
Keywords:
|
piece-wise constant ;
MCMC ;
Non-parametric
|
|
Abstract:
|
In Parkinson's disease(PD) studies, disease progression is usually monitored using repeated measurements of various rating scales consisting of multiple ordinal items. The follow-up of PD patients are subject to competing terminal events (e.g., worsening of disease, dropout or death) which may be outcome dependent. Current state-of-art joint models can only handle a few ordinal outcomes and run into severe computational difficulty when the number of ordinal outcomes is large (>4). We develop a joint modeling framework that consists of a multilevel latent variable submodel for the multiple longitudinal ordinal measurements and cause-specific hazard submodels for the competing terminal events. Two submodels are linked by the latent variables denoting the unobserved disease severity. Our joint model helps researchers to make inference for event failure times and multiple longitudinal ordinal outcomes which can incorporate many ordinal items (>10). The model inference is conducted using a Bayesian framework via Markov Chain Monte Carlo simulation implemented in STAN language. Our proposed model is evaluated by simulation studies and is applied to the Parkinson's Progression Ma
|
Authors who are presenting talks have a * after their name.
