You are here:
An empirical assessment of exposure measurement errors and effect attenuation in bi-pollutant epidemiologic models
Citation:
Baxter, L., K. Dionisio, AND H. Chang. An empirical assessment of exposure measurement errors and effect attenuation in bi-pollutant epidemiologic models. Conference of the International Society of Environmental Epidemiology, Seattle, WA, August 24 - 28, 2014.
Impact/Purpose:
The National Exposure Research Laboratory (NERL) Human Exposure and Atmospheric Sciences Division (HEASD) conducts research in support of EPA mission to protect human health and the environment. HEASD research program supports Goal 1 (Clean Air) and Goal 4 (Healthy People) of EPA strategic plan. More specifically, our division conducts research to characterize the movement of pollutants from the source to contact with humans. Our multidisciplinary research program produces Methods, Measurements, and Models to identify relationships between and characterize processes that link source emissions, environmental concentrations, human exposures, and target-tissue dose. The impact of these tools is improved regulatory programs and policies for EPA.
Description:
Using multipollutant models to understand the combined health effects of exposure to multiple pollutants is becoming more common. However, the complex relationships between pollutants and differing degrees of exposure error across pollutants can make health effect estimates from multipollutant models difficult to interpret.Our objectives were to (1) quantify the relationships between multiple pollutants and their associated exposure errors across metrics of exposure; and (2) use empirical values to determine the potential attenuation of coefficients in epidemiologic models.Three daily exposure metrics (i.e., central-site measurements, air quality model estimates, and population exposure model estimates) for 193 ZIP codes in the Atlanta, GA metropolitan area, from 1999-2002, for fine particulate matter (PM2.5) and its components (elemental carbon (EC) and sulfate (SO4)), ozone (O3), carbon monoxide (CO), and nitrogen oxide (NOx)), were used to construct three types of exposure error: δspatial, δpopulation, and δtotal. We compare exposure metrics and exposure errors within and across pollutants, and present attenuation factors for single and bi-pollutant model coefficients.Pollutant concentrations and their exposure errors are moderately-highly correlated (typically >0.5), especially for CO, NOx, and EC (i.e., “local” pollutants), and differ across exposure metrics and types of exposure error (i.e. δspatial vs. δpopulation vs.δtotal). For local pollutants spatial variability exists, and variance of exposure error ranges from 0.25-0.83 (except δpopulation).The attenuation of model coefficients in single and bi-pollutant epidemiologic models differs across types of exposure error, pollutants, and zip code. Under a classical exposure error framework, attenuation may be substantial for local pollutants due to δspatial and δtotal, with true coefficients reduced by a factor typically <0.6 (results vary for δpopulation and regional pollutants).