A Prospective Birth Cohort Study on the Independent and Joint Effect of Maternal Preconception Obesity/Diabetes, and Gestational Diabetes in the Development of ASD, ID and Other DD

Background: Autism Spectrum Disorder (ASD) affects 1 in 68 U.S. children today and is frequently accompanied by other developmental disorders. Despite the growing burden of obesity and diabetes in preconception and pregnant women, few prospective birth cohort studies have examined their role in the development of ASD. Such study is particularly lacking among the US low-income minority populations, where the burden of obesity and diabetes is high.

Objectives: This study aims to examine the independent and joint effect of maternal preconception and prenatal obesity and diabetes in the development of ASD, intellectual disability (ID) and other developmental disorders (DD) controlling for other important risk factors in a US low-income minority prospective birth cohort.

Methods: The study includes 2734 children (a subset of the Boston Birth Cohort) who were enrolled at birth and followed onward at the Boston Medical Center (BMC) between 1998 and 2014. Data on demographic characteristics, maternal preconception health, pregnancy complications and health behaviors, and birth outcomes were obtained by maternal questionnaire interview after delivery, and from maternal electronic medical records (EMRs). Children with physician diagnoses of ASD, ID and/or other DD were identified through ICD 9 codes in their EMRs up to their last contact with the BMC. Sequential Cox proportional hazard regression was used to examine the independent and joint effect of maternal preconception and prenatal obesity and diabetes on the risk of incident ASD, without and with adjustment for other preconception, prenatal and perinatal factors significantly associated with the risk of ASD. Similar models were fit on the risk of ID and other DD. The adjusted population attributable fraction (PFA) of the conditions for ASD is calculated based on the final model. 

Results: 102 and 125 children were ever diagnosed with ASD and ID, respectively; 864 children without ASD, ID or other psychiatric conditions were ever diagnosed with another DD. Maternal preconception obesity alone, pregestational diabetes alone, and the two conditions combined are associated with 1.72 (p-value 0.03), 2.35 (p-value 0.25) and 3.16 (p-value 0.06) times the risk of incident ASD in children respectively, comparing those with neither condition. Associations of similar magnitudes hold for ID, but much weaker for other DD. The PAF of the two conditions combined for ASD is estimated 15.0% in this population. 

Conclusions: This is the first prospective birth cohort study in a US urban low-income minority population to demonstrate that maternal preconception obesity is independently associated with higher risk of ASD in children, while presence of both obesity and pregestational diabetes confers an even higher risk, together, with a substantial PAF. Similar findings in ID but not other DD may suggest possible etiologic link between ASD and ID.  More studies are needed to confirm our findings and to elucidate the biological mechanisms underlying the obesity/diabetes and ASD associations.