Emily S. Barrett, J. Bruce Redmon, Christina Wang, Amy Sparks, Shanna H. Swan

Previous research has shown that prenatal exposure to endocrine-disrupting chemicals can alter children's neurodevelopment, including sex-typed behavior, and that it can do so in different ways in males and females. Non-chemical exposures, including psychosocial stress, may disrupt the prenatal hormonal milieu as well. To date, only one published study has prospectively examined the relationship between exposure to prenatal stress and gender-specific play behavior during childhood, finding masculinized play behavior in girls who experienced high prenatal life events stress, but no associations in boys. Here we examine this question in a second prospective cohort from the Study for Future Families. Pregnant women completed questionnaires on stressful life events during pregnancy, and those who reported one or more events were considered "stressed". Families were recontacted several years later (mean age of index child: 4.9 years), and mothers completed a questionnaire including the validated Preschool Activities Inventory (PSAI), which measures sexually dimorphic play behavior. In sex-stratified analyses, after adjusting for child's age, parental attitudes toward gender-atypical play, age and sex of siblings, and other relevant covariates, girls (n=72) exposed to prenatal life events stress had higher scores on the PSAI masculine sub-scale (β=3.48, p=0.006) and showed a trend toward higher (more masculine) composite scores (β=2.63, p=0.08). By contrast, in males (n=74), there was a trend toward an association between prenatal stress and higher PSAI feminine sub-scale scores (β=2.23, p=0.10), but no association with masculine or composite scores. These data confirm previous findings in humans and animal models suggesting that prenatal stress is a non-chemical endocrine disruptor that may have androgenic effects on female fetuses and anti-androgenic effects on male fetuses.