Adverse Birth Outcomes and Contamination of Drinking Water by Arsenic, Atrazine, and Nitrate-Nitrite

Increasing concern exists about whether prenatal exposure to arsenic, atrazine, and nitrate-nitrite (NN) in drinking water is associated with adverse reproductive and developmental outcomes. The goals of this research were to examine the relationships between these three water contaminants and the following birth outcomes: small for gestational age (SGA), low birth weight among term infants (term LBW), very low birth weight (VLBW), preterm birth (PTB), and very preterm birth (VPTB). We used data from the Safe Drinking Water Information System (SDWIS) and Atrazine Monitoring Program (AMP) as well as birth certificate data from the state of Ohio. We examined the relationship between each contaminant and birth outcome through generalized estimating equation logistic regression models, adjusting for important covariates. For arsenic and NN models, we performed sensitivity analyses by restricting models to those counties with a low percentage (<20 or <10%) of the population using private well water. Atrazine analyses were limited to those communities served by AMP water systems. In counties with low private well usage, we found that the odds of VLBW and preterm birth increased with increasing levels of annual mean arsenic concentrations in drinking water. Our findings were unchanged when we restricted our analyses to only those infants with exposure below the current permissible level of arsenic in drinking water. Within rural counties in which <20 or <10% of the population derived drinking water from private wells, we observed a significant increase in the odds of VLBW births with increasing mean gestational NN concentrations. In communities receiving public drinking water from AMP water systems, we found a significant increase in the odds of term LBW births with increasing atrazine exposure in drinking water over the entire gestational period of the pregnancy, as well as within the first and second trimesters. The associations observed in these analyses were present at exposure levels that are below current Maximum Contaminant Levels (MCLs) for each contaminant examined. Our findings suggest that additional research should examine the effects of arsenic, atrazine, and NN on term LBW, VLBW, and PTB in areas of low contaminant exposure.