The research objective of this K22 Career Development Award (CDA) is to ascertain if agricultural pesticide exposure confounds or modifies the relation between ambient air pollution exposure and pulmonary function in children with asthma. The central hypothesis is that early-life exposure to agricultural pesticides in conjunction with exposure to high levels of ambient air pollution is associated significantly with decreased lung function. The respiratory effects of chronic exposure to agricultural pesticides in childhood remain unknown. As the State of California has required full reporting of agricultural pesticide use for over two decades, retrospective exposure assessment of community-level pesticide application can be conducted effectively for hypothesized critical windows for asthma etiology. In addition to pesticide exposure, the land use patterns and topography of the San Joaquin Valley, the state's major agricultural region, result in high levels of exposure to ambient air pollution (AAP). The Fresno Asthmatic Children's Environment Study (FACES) offers a well-characterized longitudinal cohort to investigate the simultaneous effects of pesticides and AAP on changes in lung function and development of asthma. The research approach will consist of three Specific Aims.
(Aim 1) : Implementation of three pesticide exposure assessment strategies (geographic-based exposure, collection of dust samples, and collection of urinary biomarkers) to estimate associations with current lung function in the FACES cohort. We hypothesize that the relation between lung function and pesticides will vary by exposure assessment method and by class of pesticide. We anticipate that these short-term exposure measures can be used to calibrate the long-term pesticide exposures estimates.
(Aim 2) : Comparison of two Bayesian statistical methods to understand the relation between agricultural pesticide exposure and lung function.
This aim will consist of statistical simulations that will be conducted with pesticide exposure estimates informed by Aim 1 and retrospective air pollution data based on residential histories from the FACES cohort (n = 315). The relation between lung function, environmental exposures and demographic covariates (sex and race) will be analyzed using two recently developed statistical methods, nonparametric Bayes shrinkage and profile regression. The objective of this aim is to understand the statistical model performance of these Bayesian approaches for research of complex environmental mixtures and demographic factors. (3). Assessment of the independent and joint effects of pesticide exposure in combination with ambient air pollution on lung function. We will estimate the effect of early childhood (ages 0 - 1) exposure to community-level pesticide and ambient air pollutants on baseline lung function from the FACES cohort (n=315) in a model derived in Aim 2. We hypothesize that early childhood (0 - 1 year) exposure to three classes of agricultural pesticides will modify the relation between air pollution and FEV1 measured at study enrollment (ages 6 - 11). The CDA Candidate is a tenure-track Assistant Professor of Epidemiology in the Department of Environmental and Radiological Health Sciences at Colorado State University (CSU) with training in environmental and social epidemiology. Her short-term career goal is to expand her current research in asthma epidemiology and develop a research program to address multi-pollutant exposure analyses in vulnerable communities, with specific focus on agricultural pesticide exposures and ambient air pollution. Her long-term career goal is to understand the dynamic interaction of environmental pollutants, biogenic exposures, and host factors on asthma incidence in children. This CDA will support her didactic and experiential training in toxicology and Bayesian statistics. Her statisticl training for this grant application is necessitated by the inferential limitations of conventional statistical methods (e.g. generalized linear models) for multi-pollutant analyses (i.e. regarding assessment of toxic compounds given application of mixtures or combinations of pesticides) and multi-domain analyses (i.e. ascertainment of the independent contribution of pesticides among combinations of correlated factors, both anthropogenic and biogenic in nature.) These statistical approaches must be informed by an understanding of compound interaction on a toxicological level. Her biostatistics training will include data simulation and Bayesian approaches for epidemiologic studies; toxicology training will include course work, field work in exposure assessment, and laboratory work in environmental and biomarker sample analysis. Her department has expertise in environmental epidemiology, exposure assessment and toxicology; the campus offers comprehensive training in applied statistical approaches and methods. These experiences will inform the conduct of this CDA, as well as future work transdisciplinary research environmental health. This CDA supports the mission of the NIEHS via investigation of the health effects of low-level chronic exposures.
As epidemiologic models largely focus on the health effect of one exposure, it is important that we are able to develop statistical models that try to capture th health effects of the range of environmental factors to which we are exposed. This is especially critical for children who live in communities with high levels of various exposures, such as ambient air pollution and pesticides.
| Magzamen, Sheryl; Moore, Brianna F; Yost, Michael G et al. (2017) Ozone-Related Respiratory Morbidity in a Low-Pollution Region. J Occup Environ Med 59:624-630 |
