Phthalates, bisphenols, and childhood allergic Phenotypes: Findings from two birth cohort studies

Phthalates and bisphenols may contribute to childhood allergic outcomes, but whether these are differentially associated with atopic or non-atopic phenotypes is uncertain. We investigated whether early-life exposure to these chemicals differentially impacts atopic and non-atopic allergic outcomes. We used two birth cohorts to investigate late pregnancy and early childhood exposure windows. The Barwon Infant Study (n = 797) in Australia measured urinary phthalate and bisphenol metabolites at 36 weeks' gestation. The Canadian Healthy Infant Longitudinal Development Study (n = 993) measured phthalate metabolites at 3, 12, and 36 months. Atopy was assessed using skin prick tests at 4-5 years. Outcomes included asthma, wheeze, eczema, and rhinitis at 4-5 years. Models were stratified by atopy. We modelled exposure mixtures using quantile G-computation and Bayesian Kernel Machine Regression. Prenatal mono-carboxy-propyl phthalate was suggestively associated with non-atopic asthma (adjusted risk ratio [aRR] = 1.12; 95 % confidence interval [CI]: 0.99-1.27) with evidence of effect modification by atopy (p for interaction = 0.02). Prenatal bisphenol A was inversely associated with overall wheeze (aRR = 0.64; 95 % CI: 0.44-0.94). In the postnatal period, diethyl and dibutyl phthalates were associated with non-atopic asthma, but not with atopic asthma, though estimates did not differ substantially by atopic status. Prenatal phthalate mixtures were more strongly associated with non-atopic asthma (aRR = 1.83; 95 % CI: 1.10-3.04), with evidence of effect modification by atopy (p = 0.02 for interaction). Postnatal phthalate mixtures were associated with non-atopic asthma (aRR = 1.82, 95 % CI: 1.19-2.78), but not atopic asthma, though the association did not differ by phenotype (p for interaction = 0.45). Phthalate mixtures showed U-shaped (prenatal) and inverse U-shaped (postnatal) associations with atopic asthma, and linear positive associations with non-atopic asthma. There was little evidence of associations for other allergic outcomes. Early-life exposure to phthalates may differentially influence the risk of childhood atopic and non-atopic asthma.