PROJECT SUMMARY/ABSTRACT
Evidence suggests that the prevalence of certain developmental disorders, specifically autism spectrum
disorders (ASD) and attention deficit hyperactivity disorder (ADHD), has increased sharply in recent decades.
Recent studies have shown a moderate heritability for neurodevelopmental delays (NDDs), suggesting a
nongenetic component contributing to its etiology. The estimated annual cost of environmentally mediated NDDs
in US children is $18.3 billion. Epidemiologic studies examined prenatal and early childhood exposure to air
pollutants and NDDs, with non-conclusive evidence on the effects of different pollutants, interactions, sensitive
windows of exposure and chemical components of PM2.5. In addition, the role of socioeconomic status (SES) in
these associations is complex. Our overarching goal is to comprehensively investigate the independent and joint
prenatal and early childhood exposure to PM2.5, NO2 and O3, PM2.5 chemical components and interactions with
temperature on NDDs as well as effect modification by SES. The specific aims of this application are: 1) Quantify
the association prenatal and early childhood exposure to PM2.5, NO2, O3, temperature and NDDs, 2) Examine the
association between prenatal and early childhood exposure to PM2.5 chemical speciation and NDDs, and 3)
Examine pathways and synergistic effects, among environmental exposures and SES on NDDs. We have
established a birth cohort of more than 4 million pregnancies from two population-based databases—the
Medicaid Analytic eXtract and IBM MarketScan—that include rich information on location and individual-level
characteristics. By 8 years of age, 23.9% of publicly insured children and 11.0% of privately insured children
received a diagnosis of 1 or more NDDs. For all pollutants, PM2.5 chemical components and temperature, we
have validated highly resolved daily predictions for the entire continental US. The unique composition of these
two cohorts, allows us to determine the extent to which SES modifies these associations. For our analyses, we
will adjust for a wide range of individual-level characteristics and spatiotemporal covariates. We will estimate
hazard ratios and their 95% confidence intervals using cox regression, distributed-linear/non-linear lag models,
and we will examine effect modification by infant sex. We will use advanced statistical methods to estimate the
effects of mixtures. Our prior work and data support both the hypothesis for and the feasibility of the proposed
study. The proposed research is innovative because it will be the first study to: 1) investigate associations
between air pollutants, temperature and NDDs on a national scale; 2) use state-of-the-art spatiotemporal models
to assess individual exposures to PM2.5, NO2, O3 and temperature; 3) employ innovative resources of clinical data
by leveraging two large pregnancy cohorts. The proposed research is of great significance because it will guide
decisions and pollutant regulations to protect the health of pregnant women and children, inform heat wave
warning systems, potentially reducing rates of NDDs and subsequent social, financial, and public health burden.
