# Prenatal exposure to per- and polyfluoroalkyl substances: Association with child behavior in the environmental influences on child health outcomes (ECHO) Cohort

## TL;DR

This study finds limited evidence that prenatal exposure to certain chemicals called PFAS affects child behavior, based on data from a large U.S. cohort.

## Contribution

The study leverages a large, pooled dataset from nine ECHO Cohort sites to examine PFAS exposure and child behavior with age-stratified models.

## Key findings

- Most PFAS concentrations in maternal serum were low (<5 ng/mL), with PFOS, PFOA, PFHxS, and PFNA being the most frequently detected.
- Associations between PFAS and child behavioral problems were mostly null, with only some suggestive findings for PFHxS in preschool-age children.
- No consistent sex differences or significant associations for PFAS mixtures were observed.

## Abstract

Prenatal exposure to per- and polyfluoroalkyl substances (PFAS) may adversely impact child neurodevelopment; however, epidemiologic findings remain inconclusive because of small sample sizes, limited exposure variability, and differing neurodevelopmental measures.

We aimed to investigate the relationship between prenatal PFAS exposure and child behavior.

We pooled data from nine study sites in the nationwide Environmental influences on Child Health Outcomes (ECHO) Cohort. PFAS were quantified in maternal serum samples collected between 2 and 42 weeks’ gestation. Behavioral and emotional problems were assessed via the Child Behavior Checklist (CBCL) in preschool-age (n = 1,723) and school-age (n = 627) children. We used age-stratified, covariate-adjusted linear mixed-effects models to estimate differences in CBCL scores by PFAS quartile, focusing on analytes detected at >75 %. We also fit quantile g-computation models to examine associations for PFAS mixtures and tested for effect modification by child sex.

Perfluorooctane sulfonate (PFOS), perfluorooctanoic acid (PFOA), perfluorohexane sulfonate (PFHxS), and perfluorononanoic acid (PFNA) were the most frequently detected analytes in maternal prenatal serum, although concentrations were generally low (<5 ng/mL). Associations between PFAS concentrations and CBCL scores were mostly null, except for some suggestive findings for PFHxS in the preschool-age subset. No consistent sex differences were observed, and associations for PFAS mixtures were statistically insignificant.

We found little evidence of associations between prenatal PFAS exposures and child behavioral problems in the ECHO Cohort. Future studies should consider PFAS exposure during the postnatal period, which may be a more sensitive window.