LAUSR

Background The association between acrylamide exposure and the odds of developmental disabilities (DDs) is unclear. We conducted this analysis to explore whether acrylamide exposure is related to DDs. Methods We analyzed a sample of 1,140 children aged 6–17 years old from the US National Health and Nutrition Examination Survey 2013–2014 to 2015–2016. DDs were determined by reports of parents. Acrylamide exposure was evaluated by the hemoglobin adducts of acrylamide (HbAA) and its major metabolite glycidamide (HbGA). We investigated the association using binomial logistic regression analysis by taking HbAA and HbGA as continuous or quartile variables. Restricted cubic splines (RCS) were used to explore the non-linear relationship between HbAA or HbGA and the odds of DDs. Interaction analysis and propensity score matching (PSM) were used to validate the results. Results A total of 134 participants were reported to have DDs. The median level of HbAA and HbGA was 41.6 and 40.5 pmol/g Hb, respectively. HbAA and HbGA were not associated with the odds of DDs when taken as continuous variables. When divided into quartiles, there was no evidence for a linear trend for HbAA and HbGA. RCS showed that there was a J-shaped association between HbGA and the odds of DDs (P for non-linearity, 0.023). The results were consistent in interaction analysis by age, gender, and race, and after PSM. Conclusion HbGA level was associated with the odds of DDs in a J-shaped manner among children. Further investigation is warranted to determine the causality and underlying mechanisms.