Arsenic contamination of drinking water in California and risk of spontaneous preterm birth

Abstract

Arsenic is among the most common groundwater contaminants worldwide. Studies have identified elevated risk of adverse birth outcomes, including preterm birth, among populations with high concentrations of arsenic in drinking water. However, the population impact of low-level arsenic exposure on preterm birth remains uncertain. We evaluated the association between arsenic in drinking water and risk of spontaneous preterm birth among nearly one million singleton live births in California from 2000-2011. We used public water system monitoring records to estimate arsenic concentrations in drinking water for each maternal residence during gestation. Our statistical approach emulated a crossover design: We applied mixed effects logistic regression to evaluate preterm birth risk in women who “crossed over” into different arsenic exposure between consecutive pregnancies. We did not observe an association between odds of late preterm birth (32-36 weeks of gestation) and tap water arsenic concentration greater than 5 micrograms per liter during pregnancy. We found a borderline significant increase of approximately 15% in adjusted odds of early preterm birth (20-31 weeks of gestation) associated with the same exposure among women who did not move water systems between consecutive pregnancies. However, this finding was attenuated in several sensitivity analyses. This study does not suggest strong associations between low-level arsenic in drinking water during pregnancy and spontaneous preterm birth, though modest increased risk for gestational length 20-31 weeks was observed with arsenic concentrations above 5 micrograms per liter. Future research will evaluate the cumulative impact of arsenic with other inorganic contaminants in drinking water on preterm birth.