Incorporating Unequal Exposure to Environmental Hazards in Sample Design: Capturing Children’s Proximity to Polluting Facilities

Alissa Cordner, Brown University
Seth Spielman, Brown University
Susan E. Short, Brown University
Rachel E. Goldberg, Brown University
John R. Logan, Brown University
Stephen Buka, Brown University

We describe the design of a prospective study investigating the association between the environment and children’s health and well-being in Bristol County, Massachusetts. Bristol County has a history of industrial pollution and we are particularly interested in the legacy of this activity for children’s health. While the literature in environmental justice makes a strong case for the need to connect exposure to environmental pollution and health outcomes, it is plagued by problems of endogeneity and selection bias, often due to the lack of suitable data. We describe a probability based two-stage clustering design that will yield a sample with variation in early life pollution exposures and socio-demographic factors. This strategy, developed for local implementation of the National Children’s Study, features development of a multi-dimensional measure of exposure to industrial activity at a fine spatial resolution and integration into a population-based sampling stratification scheme that assures diversity in socio-demographic characteristics.