Abstract: The goal of the MESA Air study is to determine whether air pollution exposure is associated with sub-clinical progression of atherosclerosis. It pairs state-of-the-art cardiovascular epidemiology with state-of-the-art exposure assessment. I will use the 2016 Lancet paper “Association between air pollution and coronary artery calcification within six metropolitan areas in the USA (the Multi-Ethnic Study of Atherosclerosis and Air Pollution): a longitudinal cohort study” to highlight key biostatistical contributions to MESA Air, focusing on exposure prediction modeling and health effect inference. I will describe advances to air pollution cohort study exposure assessment design and analysis, including the development of a new spatio-temporal prediction model that combines spatially rich exposure data collected over short time periods with long time series of ambient pollution measurements. Our goal was to obtain high quality spatially and temporally-resolved individual-level estimates of exposure. For health effect inference we adopted longitudinal data models. These focus on the primary target of inference: rate of change associated with air pollution exposure in a marker of sub-clinical cardiovascular disease. Because our exposures were predicted from a statistical model, we developed new exposure measurement error methods to improve the accuracy and nominal coverage of the parameter estimates. These methodological advances led to deeper understanding of priorities for exposure assessment design. Our experience suggests active biostatistical collaboration is essential to high-impact science.
Integrated Health Sciences Core of the Michigan Center on Lifestage Environmental Exposures and Disease (M-LEEaD)
Environmental Biostatistics and High-Impact Science: the MESA Air Study - Lianne Sheppard
A special event of the Integrated Health Sciences Core
March 14, 2018
3:00 PM - 5:00 PM
1690 SPH I
1415 Washington Heights
Ann Arbor, MI 48109-2029
Sponsored by: Integrated Health Sciences Core of the Michigan Center on Lifestage Environmental Exposures and Disease (M-LEEaD)
Contact Information: Meredith McGehee (mcgehee@umich.edu | 647-0819)
Abstract: The goal of the MESA Air study is to determine whether air pollution exposure is associated with sub-clinical progression of atherosclerosis. It pairs state-of-the-art cardiovascular epidemiology with state-of-the-art exposure assessment. I will use the 2016 Lancet paper “Association between air pollution and coronary artery calcification within six metropolitan areas in the USA (the Multi-Ethnic Study of Atherosclerosis and Air Pollution): a longitudinal cohort study” to highlight key biostatistical contributions to MESA Air, focusing on exposure prediction modeling and health effect inference. I will describe advances to air pollution cohort study exposure assessment design and analysis, including the development of a new spatio-temporal prediction model that combines spatially rich exposure data collected over short time periods with long time series of ambient pollution measurements. Our goal was to obtain high quality spatially and temporally-resolved individual-level estimates of exposure. For health effect inference we adopted longitudinal data models. These focus on the primary target of inference: rate of change associated with air pollution exposure in a marker of sub-clinical cardiovascular disease. Because our exposures were predicted from a statistical model, we developed new exposure measurement error methods to improve the accuracy and nominal coverage of the parameter estimates. These methodological advances led to deeper understanding of priorities for exposure assessment design. Our experience suggests active biostatistical collaboration is essential to high-impact science.