Abstract
Relative maternal poverty is a suggested social determinant of autism spectrum disorders (ASDs) in offspring; however, this association may be confounded by the maternal broader autism phenotype (BAP). The biological mechanisms underlying this association are largely understudied. We examined the association between household income during pregnancy and ASDs in offspring, adjusting for confounders including maternal BAP, and explored whether maternal chronic stress, measured by allostatic load (AL) during pregnancy, mediates this association. Data on 59,998 mother-child dyads were obtained from the Japan Environment and Children's Study, a nationwide birth cohort. Household income was categorized into tertiles (< 4 million, 4-6 million, > 6 million JPY) and offspring ASD diagnosis by age four was assessed via guardian's report. Bayesian logistic regression models indicated that mothers from low- and middle-income households had a 58% (95% credible interval [CI]: 28%-98%) and a 37% (95% CI: 12%-70%) higher risk of offspring ASDs, respectively, compared to those from high-income households. AL, defined as three or more out of 10 biomarkers in the highest risk quartile, did not mediate these associations. Low and middle household income during pregnancy was associated with a higher risk of ASD diagnosis, and high AL did not mediate this association.