Abstract
BACKGROUND: Mothers' experience of adverse life events (ALEs, e.g., divorce, bereavement, injury) during pregnancy has been linked with neurodevelopmental conditions like autism, and related traits like social communication difficulties and repetitive behavior in children. However, both the cumulative association and the underlying mechanism are unclear, and these associations might be confounded by unmeasured genetic or other early environmental factors shared within families. METHOD: This longitudinal population-based cohort study included 114,247 children, born in Norway between 1999 and 2009, who participated in the Norwegian Mother, Father and Child Cohort Study. During week 30 of pregnancy, mothers of 51,940 children (of whom 12,597 were siblings) reported whether they had experienced ALEs. We estimated associations between mothers' cumulative exposure to and perception of ALE and their children's clinical diagnosis of autism, and maternal reports on their children's autistic traits at ages 3 and 8 years through the Social Communication Questionnaire (SCQ). Sibling comparisons were conducted to account for unmeasured familial confounding. RESULTS: Each additional prenatal ALE was associated with increased adjusted hazard ratios [HR: 1.23, 95% CI: 1.16-1.30] of autism diagnosis, compared to unexposed children. Adjusting for unmeasured familial confounding in sibling comparisons, the association attenuated: HR = 0.53, 95% CI [0.31-0.90]. ALEs perceived as more painful were associated with a 12% elevated likelihood of autism diagnosis [95% CI: 7%-16%], but this association attenuated after sibling comparisons. SCQ scores in children exposed to cumulative prenatal ALE compared to unexposed children were higher at age 3 (β-coefficient: 0.24 (95%CI [0.21-0.27])), but only slightly at age 8 (β-coefficient: 0.07 [95% CI: 0.04-0.10]) with differences nullified in the sibling comparison analysis. CONCLUSION: The association between maternal prenatal exposure to cumulative ALEs and diagnosis of autism and autism-associated traits is likely due to unmeasured familial confounding rather than a direct causal relationship.