A Linkage Study of Adverse Birth Outcomes With Agricultural Land Use Practices in Missouri

Background: Missouri is an agriculturally intensive state, growing primarily corn, soybeans, and wheat, with additional high-intensity rice and cotton farming in the southeastern counties in the state. There is limited evidence that communities residing in close proximity to pesticide-treated fields may have increased exposure to pesticides as well as increased risks of adverse birth outcomes including certain birth defects and adverse birth outcomes such as low birth weight and preterm births. The study aims were (1) to evaluate the relationship between county-level measures of agricultural production and adverse birth outcomes in Missouri, and (2) to evaluate the importance of incorporating spatial and temporal information into the modeling of this data. Methods: Corn, soybean, wheat, rice, and cotton crop densities were evaluated for their relationship with both low birth weight and preterm births in Missouri between 20042006. The covariates considered as potential confounders and effect modifiers in this study included gender, mother’s race and ethnicity, mother’s age at birth, maternal smoking, access to prenatal care, quarter of birth, county median household income, and population density. Three statistical approaches were taken to evaluate the relationship between each measure of crop density and each outcome: (1) Poisson regression, (2) generalized estimating equations, and (3) distance decay random effects Poisson regression. Results: Strong positive associations were observed between rice and cotton density and both low birth weight and preterm births. Rice density was associated with significant increased risk of preterm births across all three models, as was cotton density with low birth weight births. Only 58 counties in Missouri produced rice or cotton between 2004 and 2006, with complete geographic overlap. Despite the geographic correlation between these counties, both rice and cotton densities remained significant predictors of low birth weight and preterm births in dual-exposure models with a slight weakening of the effect in the distance decay models. Conclusions: The associations between both rice and cotton and low birth weight and preterm births should be viewed cautiously because of the many limitations of the study including its ecological study design, limited data on other risk factors, and a lack of chemical-specific exposure estimates. Nonetheless these findings warrant further investigation.