Lagged Kernel Machine Regression for Identifying Time Windows of Susceptibility to Exposures of Complex Metal Mixtures

• Activity Number: 501 - Biometrics Student Paper Awards 1
• Type: Topic Contributed
• Date/Time: Wednesday, August 2, 2017 : 10:30 AM to 12:15 PM
• Sponsor: Biometrics Section
• Author(s): Shelley Liu* and Jennifer Bobb and Kyu Ha Lee and Chris Gennings and Birgit Claus Henn and David Bellinger and Christine Austin and Lourdes Schnaas and Martha Tellez-Rojo and Howard Hu and Robert Wright and Manish Arora and Brent Coull
• Companies: Icahn School of Medicine at Mount Sinai and Group Health Research Institute and The Forsyth Institute and Icahn School of Medicine at Mount Sinai and Boston University School of Public Health and Harvard T.H. Chan School of Public Health and Icahn School of Medicine at Mount Sinai and National Institute of Perinatology, Mexico and National Institute of Public Health, Mexico and University of Toronto Dalla Lana School of Public Health and Icahn School of Medicine at Mount Sinai and Icahn School of Medicine at Mount Sinai and Harvard T.H. Chan School of Public Health
• Keywords: Bayesian analysis ; Environmental epidemiology ; Hierarchical models

Abstract

The impact of neurotoxic chemical mixtures on children's health is a critical public health concern. It is well known that during early life, toxic exposures may impact cognitive function during critical time intervals of increased vulnerability, known as windows of susceptibility. We develop a flexible statistical method, called lagged kernel machine regression (LKMR). To our knowledge, LKMR is the first statistical method to identify critical exposure windows of chemical mixtures, and account for complex nonlinear and non-additive effects of the mixture at any given exposure window. LKMR estimates how the effects of a mixture of exposures change with the exposure time window using a Bayesian formulation of a grouped, fused lasso penalty within a kernel machine regression framework. A simulation study demonstrates the performance of LKMR under realistic exposure-response scenarios, and demonstrates large gains over approaches that consider each time window separately, particularly when serial correlation among the time-varying exposures is high. We apply LKMR to estimate associations between neurodevelopment and metal mixtures in ELEMENT, a prospective birth cohort study.