Essential for modern agriculture, nitrogen pollution through nitrate contaminated water has emerged as a public health concern. Research examining nitrate exposure and cancer risk across the life course has yielded inconsistent conclusions. However, despite biological plausibility and heightened vulnerability during gestation, the link between nitrogen pollution and Brain/Central Nervous System (CNS) cancer in infants remains underexplored. We conducted a population-based ecological analysis linking county-level nitrogen fertilizer application data (USGS) to 0-year-old infant Brain/CNS cancer incidence data (SEER-12, 1992–2012) and mortality data (NCHS, 1990–2012) in the United States. We constructed county-level panel datasets and estimated Generalized Estimating Equation probit models with population offsets, state-year fixed effects, and robust variance estimators. Nitrogen application was standardized, lagged, and modeled separately for incidence and mortality. Sensitivity analyses added a twice lagged nitrogen measure as an instrumental variable as a generalized structural equation model. We found that a county-level one–standard deviation increase in prior-year nitrogen application was associated with a 0.87 percentage-point increase in the probability of an infant Brain/CNS cancer diagnosis (p < 0.001), corresponding to a 13% relative increase, and a 0.53 percentage-point increase in the probability of infant Brain/CNS cancer death (p < 0.001), a 36% relative increase. Instrumental variable estimates were consistent. These findings reiterate that gestation remains a critical window of susceptibility, suggesting that nitrogen fertilizer pollution may contribute to rare but severe pediatric cancer outcomes. Future research establishing or quantifying causality, and understanding etiologic heterogeneity could inform policymaking and targeted public health interventions.
Joshi et al. (Sun,) studied this question.