Bisphenol A (BPA) is a ubiquitous endocrine disrupting compound. Several experimental and epidemiological studies suggest that gestational BPA exposure can lead to neurodevelopmental and behavioral problems in early-life, but results have been inconsistent. We previously reported that prenatal BPA exposure may affect child behavior and differently among boys and girls at ages 3-5 years.
We investigated the association of prenatal and early childhood BPA exposure with behavioral outcomes in 7-9 year old minority children and hypothesized that we would observe the same sex-specific pattern observed at earlier ages.
African-American and Dominican women enrolled in an inner-city prospective cohort study and their children were followed from mother's pregnancy through children's age 7-9 years. Women during the third trimester of pregnancy and children at ages 3 and 5 years provided spot urine samples. BPA exposure was categorized by tertiles of BPA urinary concentrations. The Child Behavioral Checklist (CBCL) was administered at ages 7 and 9 to assess multiple child behavior domains. Associations between behavior and prenatal (maternal) BPA concentrations and behavior and postnatal (child) BPA concentration were assessed via Poisson regression in models stratified by sex. These models accounted for potential confounders including prenatal or postnatal urinary BPA concentrations, child age at CBCL assessment, ethnicity, gestational age, maternal intelligence, maternal education and demoralization, quality of child's home environment, prenatal environmental tobacco smoke exposure, and prenatal mono-n-butyl phthalate concentration.
The direction of the associations differed between boys and girls. Among boys (n=115), high prenatal BPA concentration (upper tertile vs. lower two tertiles) was associated with increased internalizing (β=0.41, p<0.0001) and externalizing composite scores (β=0.40, p<0.0001) and with their corresponding individual syndrome scales. There was a general decrease in scores among girls that was significant for the internalizing composite score (β=-0.17, p=0.04) (n=135). After accounting for possible selection bias, the results remained consistent for boys. Conversely, high postnatal BPA concentration was associated with increased behaviors on both the internalizing composite (β=0.30, p=0.0002) and externalizing composite scores (β=0.33, p<0.0001) and individual subscores in girls but fewer symptoms in boys. These results remained significant in girls after accounting for selection bias.
These results suggest BPA exposure may affect childhood behavioral outcomes in a sex-specific manner and differently depending on timing of exposure.