Maternal Dietary Patterns during Pregnancy and Child Autism-Related Traits: Results from Two US Cohorts.

Researchers looked at whether mothers' eating patterns during pregnancy might be linked to autism traits in their children. They studied over 800 mother-child pairs and examined five different diet types. They found a weak suggestion that Western-style diets (typically high in processed foods) might be associated with more autism traits, but this connection became less clear when they accounted for total calories eaten. Overall, no strong links were found between maternal diet and autism.

This study examined maternal dietary patterns during pregnancy and their relationship to autism-related traits in children across two US cohorts (EARLI, n=154; NHSII, n=727). Researchers analyzed five dietary patterns including inflammatory, healthy eating, Western, prudent, and Mediterranean diets using food frequency questionnaires. The primary outcome was autism-related traits measured by the Social Responsiveness Scale, with autism spectrum disorder diagnosis as a secondary outcome. Results showed a suggestion of positive association between Western dietary patterns and autism traits in one cohort, though this association weakened after adjusting for total energy intake.

No clear associations were found with other dietary patterns or ASD diagnosis.

Findings are preliminary and insufficient to support specific dietary recommendations for pregnant mothers regarding autism prevention. The attenuation of associations after energy adjustment suggests overall caloric intake may be more relevant than specific dietary patterns. Further research with larger samples is needed.

Small sample size in EARLI cohort (n=154), observational design preventing causal conclusions, reliance on food frequency questionnaires which may have measurement error, and inconsistent findings across the two cohorts studied.

We examined the relationship between maternal intake of established dietary patterns and child autism-related outcomes in two prospective cohorts in the United States. Participants were drawn from the Early Autism Risk Longitudinal Investigation (EARLI,= 154) and the Nurses' Health Study II (NHSII,= 727). Dietary information was collected via food frequency questionnaires (FFQs) and used to calculate the empirical dietary inflammatory pattern (EDIP), Alternative Healthy Eating Index (AHEI), Western and Prudent dietary patterns, and the alternative Mediterranean Diet (aMED) score. Primary analyses examined associations with continuous autism-related traits as measured by the Social Responsiveness Scale (SRS), and secondary analyses with autism spectrum disorder (ASD) diagnosis.

We used crude and multivariable quantile regression fixed at the 50th percentile to examine associations between quartiles of dietary patterns and SRS scores, and logistic regression to examine associations with ASD diagnosis. There was suggestion of a positive association with the Western diet (Q4 vs. Q1, ß = 11.19, 95% CI: 3.30, 19.90) in EARLI, though the association was attenuated with adjustment for total energy intake, and no clear associations were observed with other dietary patterns and ASD diagnosis or SRS scores. Further work is needed to better understand the role of maternal dietary patterns in ASD and related outcomes.