Quantile Regression for Mixed Models with Application to Recent Blood Pressure Trends in China

• Activity Number: 608
• Type: Contributed
• Date/Time: Wednesday, August 12, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Luke Smith* and Montserrat Fuentes and Penny Gordon-Larsen and Brian J. Reich
• Companies: Westat and North Carolina State University and The University of North Carolina at Chapel Hill and North Carolina State University
• Keywords: Quantile regression ; Longitudinal ; Multivariate ; Bayesian ; Blood pressure

Abstract

Cardiometabolic risk has substantially increased in China over the past 20 years and blood pressure is a primary modifiable risk factor. We use longitudinal data from 1991 to 2009 in the China Health and Nutrition Survey to examine recent blood pressure trends in China. Large values of blood pressure are of interest, so we model the quantile functions of systolic and diastolic blood pressure. We join the distributions of systolic and diastolic blood pressure using a copula. This permits the relationships between the covariates and the two responses to share information, preserves the population level marginal distributions, and accounts for within-subject dependence. Our regression effects change across quantile level, year, and blood pressure type, providing a rich environment for inference. To our knowledge, this is the first quantile function model to accommodate bivariate response, and is the first quantile function model to characterize within-subject autocorrelation over time. We find that the association between high blood pressure and living in an urban area has evolved from positive to negative, with the strongest changes occurring in the upper tail.