A Family of Partial-Linear Single-Index Models for Analyzing Complex Environmental Exposures with Continuous, Categorical, Time-to-Event, and Longitudinal Health Outcomes

• Activity Number: 458 - Models for Spatial and Environmental Data
• Type: Contributed
• Date/Time: Thursday, August 6, 2020 : 10:00 AM to 2:00 PM
• Sponsor: Section on Statistics and the Environment
• Author(s): Yuyan Wang* and Yinxiang Wu and Myeonggyun Lee and Peng Jin and Leonardo Trasande and Mengling Liu and Melanie Jacobson
• Companies: NYU Langone Medical Center and NYU Langone Medical Center and NYU Langone Medical Center and New York University and NYU Langone Medical Center and NYU Langone Medical Center and NYU Langone Medical Center
• Keywords: Environmental mixtures; NHANES; Semiparametric model; Triglyceride

Abstract

Statistical methods to study the joint effects of environmental factors are of great importance to understand the impact of correlated exposures that may act synergistically or antagonistically on health outcomes. We present a unified family of partial-linear single-index (PLSI) models with a coherent algorithm proposed to assess the joint effects of environmental factors for continuous, categorical, time-to-event, and longitudinal outcomes. All PLSI models consist of a linear combination of exposures into a single index for practical interpretability of relative direction and importance and a nonparametric link function for modeling flexibility. We will present PLSI linear regression and PLSI quantile regression for the continuous outcome, PLSI generalized linear regression for the categorical outcome, PLSI Cox proportional hazards model for the time-to-event outcome, and PLSI mixed-effects model for the longitudinal outcome. Numerical results from applications to a real or simulated dataset of 800 subjects from NHANES 2003-2004 will be discussed to demonstrate the performance of our proposed method in estimating the joint effects of 10 environmental exposures on triglyceride.