Examining associations between prenatal biomarkers of oxidative stress and ASD-related outcomes using quantile regression.

The children were siblings of autistic children, so the findings may not apply to families without a prior autistic child. The sample was small, and the number of children with autism diagnoses was small, so the study could not examine modification by factors such as maternal education. Maternal oxidative stress markers were measured only once, in mid to late pregnancy, so early pregnancy levels were not assessed. The study measured maternal markers, not child oxidative stress, so it cannot show whether maternal oxidative stress affected the child directly. The study did not measure all oxidative stress markers or placental samples, so other markers may be more relevant to child outcomes. The authors could not rule out chance, unmeasured confounding, measurement error, or mediating factors. Pregnancy complications were based on available report rather than medically confirmed information, so mediation or modification by complications could not be fully addressed.

The authors declared no relevant financial or non-financial interests.

Do not conclude that greater antioxidant balance prevents autism-related outcomes. The diagnosis-related association was a secondary analysis and was not statistically significant, while the main continuous trait outcomes showed no strong overall links.