Prenatal exposure to endocrine disrupting chemicals and the association with behavioural difficulties in 7-year-old children in the SELMA study

BACKGROUND: Endocrine disrupting chemicals (EDCs) can cross the placenta and thereby expose the fetus, which may lead to developmental consequences. It is still unclear which chemicals are of concern regarding neurodevelopment and specifically behaviour, when being exposed to a mixture. OBJECTIVE: The objective is to determine associations between prenatal exposure to EDCs and behavioural difficulties. Furthermore, we investigated sex-specific associations and determined chemicals of concern in significant regressions. METHODS: Associations between prenatal exposure to EDCs (both as single compounds and their mixtures) and behavioural outcomes using the Strengths and Difficulties Questionnaire (SDQ) were estimated in 607 mother-child pairs in the Swedish Environmental Longitudinal, Mother and Child, Asthma and Allergy (SELMA) study. Levels for chemical compounds were measured in either urine or serum (median of 10 weeks of gestation). Associations were estimated for the total SDQ score (quasipoisson regression) and a 90th percentile cut-off (logistic regression). Exposure for EDC mixtures (phenols, phthalates, PFAS and persistent chlorinated) was studied using weighted quantile sum (WQS) regression with deciles and with and without repeated holdout validation techniques. The models were adjusted for selected covariates. RESULTS: The odds for behavioural difficulties increased in girls with higher chemical exposures (OR 1.77, 95% CI 1.67, 1.87) using the full sample and borderline for the validation set (OR 1.31, 95% CI 0.93, 1.85) with 94/100 positive betas in the 100 repeated holdout validations. Chemicals of concern for girls are mostly short-lived chemicals and more specifically plasticizers. No pattern of significant associations was detected for boys. SIGNIFICANCE: There is an indication of increased behavioural difficulties for girls in the SELMA population with higher exposure to mixtures of EDCs. Using the repeated holdout validation techniques, the inference is more stable, reproducible and generalisable. Prenatal exposure to mixtures of environmental chemicals should be considered when assessing the safety of chemicals. IMPACT: Growing evidence points towards a "mixture effect" where different environmental chemicals might act jointly where individual compounds may be below a level of concern, but the combination may have an effect on human health. We are constantly exposed to a complicated mixture pattern that is individual for every person as this mixture depends on personal choices of lifestyle, diet and housing to name a few. Our study suggests that prenatal exposure to EDCs might adversely affect the behaviour of children and especially girls. Hence, risk assessment needs to improve and sex-specific mechanisms should be included in assessments.