Exposure to ambient air pollution and early childhood behavior: A longitudinal cohort study

Background: Prenatal and early life air pollution exposure may impair healthy neurodevelopment, increasing risk of childhood behavioral disorders, but epidemiological evidence is inconsistent. Little is known about factors that determine susceptibility.

Methods: Participants were mother-child dyads from the CANDLE study, an ECHO PATHWAYS Consortium birth cohort set in the mid-South United States, who completed a preschool visit. We estimated prenatal and childhood exposures to nitrogen dioxide (NO₂) and particulate matter less than 10 μm (PM₁₀) at participants' residences using a national annual average universal kriging model (land-use regression with spatial smoothing). Distance to nearest major roadway was used as a proxy for traffic-related pollution. Primary outcomes were children's internalizing and externalizing behavior problems. Regression models were adjusted for individual- and neighborhood-level socioeconomic measures, maternal IQ, and multiple other potential confounders. We tested for effect modification by select maternal and child characteristics.

Results: The analytic sample (N = 975 of 1503 enrolled) was 64% African American and 53% had a household annual income below $35,000; child mean age was 4.3 years (SD: 0.4). Mean prenatal NO₂ and PM₁₀ exposures were 12.0 ppb (SD: 2.4) and 20.8 μg/m3 (SD: 2.0); postnatal exposures were lower. In fully adjusted models, 2 ppb higher prenatal NO₂ was positively associated with externalizing behavior (6%; 95% CI: 1, 11%). Associations with postnatal exposure were stronger (8% per 2 ppb NO₂; 95%CI: 0, 16%). Prenatal NO₂ exposure was also associated with an increased odds of clinically significant internalizing and externalizing behaviors. We found suggestive evidence that socioeconomic adversity and African American race increases susceptibility. PM₁₀ and road proximity were not associated with outcomes.

Conclusions: Findings showed that air pollution exposure is positively associated with child behavior problems and that African American and low SES children may be more susceptible. Importantly, associations were observed at exposures below current air quality standards.