Online Program Home
My Program

Abstract Details

Activity Number: 594 - Recent Advances in Statistical Modeling for Multivariate/Correlated/Time-Varying Longitudinal Data
Type: Invited
Date/Time: Thursday, August 1, 2019 : 8:30 AM to 10:20 AM
Sponsor: WNAR
Abstract #300135
Title: Quantile Regression Based Methods for Characterizing Highly Correlated Behavioral Data in Relation to Longitudinal Biomarkers with Censored Values
Author(s): MinJae Lee* and Michelle Vidoni and Belinda Reininger
Companies: University of Texas McGovern Medical School and Univ. of Texas Health Science Center at Houston and Univ. of Texas School of Public Health
Keywords: Weighted Index; Quantile Regression; Correlated Behavioral Data; Biomarkers; Longitudinal data; Left-censoring

There are a large number of research studies aimed at promoting positive lifestyle changes, such as increased healthy food consumption, to attenuate lifetime risk of non-communicable disease. However, most commonly used measurement tools for behavioral assessments rely on participant self-report. Due to gaps and biases in participant reporting, statistical modeling to assess these measurements is challenging. Biomarkers of chronic disease may provide a proximal measure of dietary intake, but the lack of specific dietary biomarkers has been recognized as an area requiring future research. Moreover, the measurement of biomarkers is often subject to left-censoring due to detection limits. We propose a statistical method that constructs a quantile-specific weighted index of multiple behavioral components. Under the quantile regression framework, the proposed method characterizes highly correlated multiple dietary intake and evaluates their relative effects on varying levels of longitudinal biomarker data in the presence of left-censoring issues. We also demonstrate application of our method to data from Tu Salud ¡Sí Cuenta! (Your Health Matters) Intervention study.

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

Back to the full JSM 2019 program