Activity Number:
|
478
- Advanced Data Analysis with Bayesian Latent Variable Modeling
|
Type:
|
Contributed
|
Date/Time:
|
Thursday, August 6, 2020 : 10:00 AM to 2:00 PM
|
Sponsor:
|
Section on Bayesian Statistical Science
|
Abstract #312946
|
|
Title:
|
Bayesian Models for Longitudinal Clustered Data with Applications to Dental Fluorosis Study
|
Author(s):
|
Tong Kang* and Jeremy Thomas Gaskins and Steven Levy and Somnath Datta
|
Companies:
|
Department of Biostatistics, University of Florida and University of Louisville and Department of Preventive and Community Dentistry, University of Iowa and University of Florida
|
Keywords:
|
Bayesian analysis;
Longitudinal ordinal data;
Mixed effects model;
Dental fluorosis;
Clustered data
|
Abstract:
|
We propose a Bayesian hierarchical model to analyze longitudinal ordinal data with complex multi-level clustered structure. This research was motivated by the fluorosis data gathered from the Iowa Fluoride Study (IFS). Dental fluorosis is characterized by spots on tooth enamel and caused by ingestion of excessive fluoride intake during enamel formation. The hierarchical structure in the data stems from the fact that observations are collected from multiple surfaces on each tooth, and on all available teeth of a subject’s mouth, which are longitudinally clustered across three measurement occasions (ages 9, 13, and 17). We evaluate the performance of our method in studying the effects of various risk and protective factors using a comprehensive simulation study and undertake a secondary analysis of the IFS data.
|
Authors who are presenting talks have a * after their name.