Higher level of maternal autistic traits associated with increased risk of preterm birth

By Vijay Kumar MalesuJan 24 2024Reviewed by Danielle Ellis, B.Sc.

In a recent study published in JAMA Network Open, a group of researchers investigated the association between maternal autistic traits and the risk of adverse birth outcomes in the general population.

Background 

Autism spectrum disorder (ASD) is a lifelong neurodevelopmental disorder marked by distinct social and communication challenges and repetitive behaviors. Recent research highlights significant health disparities for pregnant women with ASD, including limited healthcare access, heightened antenatal stress, chronic health issues, and increased pregnancy complications. These factors contribute to a higher risk of adverse birth outcomes like preterm delivery and babies born small for gestational age (SGA). While some studies link ASD diagnosis in women to premature births, they often group ASD with other developmental disorders, lacking specificity.

ASD is now viewed as a spectrum with autistic traits present in varying degrees across the population. Women, less frequently diagnosed than men, may exhibit high autistic traits without a formal diagnosis, necessitating further research into how these traits in the general population relate to birth outcomes. 

About the study 

The study, part of the Japan Environmental Children's Study, examined children born from pregnancies between January 2011 and March 2014, focusing on those with singleton live births and excluding cases with implausible birth weights, gestational ages over 42 weeks, or chromosomal abnormalities. It adhered to ethical guidelines, including the Declaration of Helsinki, and obtained written consent from participants. The study's methods followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline.

Maternal autistic traits were assessed during the second or third trimesters using the Autism-Spectrum Quotient Japanese version (AQ-J10), a self-reported questionnaire adapted from the 50-item AQ, to measure autistic traits, mainly social, in the general population. The AQ-J10's reliability was confirmed in this study. Birth outcomes analyzed included preterm birth and children SGA, with preterm births further categorized based on gestational weeks.

Covariates such as maternal age, smoking status during pregnancy, education, pre-pregnancy body mass index (BMI), preexisting health conditions, child sex, and pregnancy complications were considered. Data on these covariates were collected from medical records and maternal self-reports.

The study employed generalized linear models to explore the link between maternal autistic traits and birth outcomes, adjusting for covariates and pregnancy complications. It assessed risks in women with heightened autistic traits and conducted sensitivity analyses for psychological distress and psychiatric history, using robust statistical methods for significance.

Study results 

In the present study, which included 92,944 women with singleton live births, a rigorous selection process excluded those with children having implausible birth weights, gestational ages exceeding 42 weeks, or chromosomal abnormalities, as well as those missing critical data, resulting in 87,687 women for analysis. The average age of these women was 31.2 years, with 43.8% being first-time mothers. Notably, 20.9% had preexisting physical health conditions, and 5.3% smoked during pregnancy. Maternal autistic traits were measured using the AQ-J10, where a score of 7 or above indicated a clinical threshold. The mean score in the study was 2.8, and only 2.7% of women scored above the clinical threshold, with a tiny fraction diagnosed with ASD.

The study revealed that women in the clinical range of the AQ-J10 tended to be younger, first-time mothers with lower education levels, a history of smoking during pregnancy, and higher antenatal psychological distress. The analysis also highlighted a potential sample bias, as women with higher AQ-J10 scores or certain demographic characteristics were more likely to be excluded.

The main analysis found that higher maternal autistic traits correlated with increased risks of preterm birth, including moderate-to-late and very preterm births, as well as children born SGA. These associations persisted across different models, adjusting for various covariates and pregnancy complications. Women scoring in the AQ-J10's clinical range particularly showed higher risks for these adverse birth outcomes.

To deepen the understanding, postestimation analysis was conducted, further confirming the increased risk of adverse outcomes for women with higher AQ-J10 scores. Sensitivity analyses adjusting for factors like antenatal psychological distress and excluding women with psychiatric histories or those on psychotropic medication during pregnancy did not significantly alter the results, underscoring the robustness of the findings. 

This study highlights the relationship between maternal autistic traits and birth outcomes, emphasizing the need for targeted support and interventions for pregnant women exhibiting higher autistic traits. The findings are pivotal for healthcare providers and policymakers in tailoring prenatal care to mitigate risks associated with ASD traits.