학술논문
Cardiometabolic Pregnancy Complications in Association With Autism-Related Traits as Measured by the Social Responsiveness Scale in ECHO
Document Type
article
Author
Lyall, Kristen; Ning, Xuejuan; Aschner, Judy L; Avalos, Lyndsay A; Bennett, Deborah H; Bilder, Deborah A; Bush, Nicole R; Carroll, Kecia N; Chu, Su H; Croen, Lisa A; Dabelea, Dana; Daniels, Julie L; Duarte, Christiane; Elliott, Amy J; Fallin, M Daniele; Ferrara, Assiamira; Hertz-Picciotto, Irva; Hipwell, Alison E; Jensen, Elizabeth T; Johnson, Susan L; Joseph, Robert M; Karagas, Margaret; Kelly, Rachel S; Lester, Barry M; Margolis, Amy; McEvoy, Cindy T; Messinger, Daniel; Neiderhiser, Jenae M; O’Connor, Thomas G; Oken, Emily; Sathyanarayana, Sheela; Schmidt, Rebecca J; Sheinkopf, Stephen J; Talge, Nicole M; Turi, Kedir N; Wright, Rosalind J; Zhao, Qi; Newschaffer, Craig; Volk, Heather E; Ladd-Acosta, Christine; Outcomes, on behalf of program collaborators for Environmental Influences on Child Health
Source
American Journal of Epidemiology. 191(8)
Subject
Language
Abstract
Prior work has examined associations between cardiometabolic pregnancy complications and autism spectrum disorder (ASD) but not how these complications may relate to social communication traits more broadly. We addressed this question within the Environmental Influences on Child Health Outcomes program, with 6,778 participants from 40 cohorts conducted from 1998-2021 with information on ASD-related traits via the Social Responsiveness Scale. Four metabolic pregnancy complications were examined individually, and combined, in association with Social Responsiveness Scale scores, using crude and adjusted linear regression as well as quantile regression analyses. We also examined associations stratified by ASD diagnosis, and potential mediation by preterm birth and low birth weight, and modification by child sex and enriched risk of ASD. Increases in ASD-related traits were associated with obesity (β = 4.64, 95% confidence interval: 3.27, 6.01) and gestational diabetes (β = 5.21, 95% confidence interval: 2.41, 8.02), specifically, but not with hypertension or preeclampsia. Results among children without ASD were similar to main analyses, but weaker among ASD cases. There was not strong evidence for mediation or modification. Results suggest that common cardiometabolic pregnancy complications may influence child ASD-related traits, not only above a diagnostic threshold relevant to ASD but also across the population.