Investigating prenatal and early-life environmental causes of neurodevelopment, using genetically informative study designs

Neurodevelopmental conditions such as autism and attention-deficit hyperactivity disorder (ADHD) emerge in childhood and persist throughout life, impacting multiple domains of functioning. Although a substantial body of research highlights the strong genetic basis of these conditions, environmental influences remain an area of investigation. In this work we explored associations between adverse life events (ALEs), psychosocial adversity during pregnancy and early childhood, neonatal jaundice, and the likelihood of autism and ADHD in children. To address potential biases from unmeasured familial confounding, we applied family-based study designs.

Study I examined the association between psychosocial adversity in the family in the first year of life on autism and ADHD diagnoses in the child. While we observed an increased likelihood of autism and a dose-response relationship for ADHD in the general population, these associations weakened as the level of familial relatedness increased (cousin, half-sibling, and sibling comparisons), suggesting that the initial findings were influenced by unmeasured familial confounding.

Study II investigated the association between ALEs during pregnancy on autism diagnoses and mother-rated autistic traits in offspring. The results indicated a higher likelihood of autism diagnosis and more pronounced autistic traits in the full sample. However, sibling comparisons revealed that these associations were largely explained by unmeasured familial confounding rather than a direct causal relationship.

Study III assessed the association between severe, repeated exposure to ALEs in early childhood on autism diagnoses and mother-rated autistic traits at age 3. We found a dose-response relationship in the general population, but the associations attenuated in sibling comparisons, suggesting familial confounding. While we also observed an increased likelihood of an autism diagnosis, sibling comparisons lacked statistical power, therefore, this result should be interpreted with caution.

Study IV evaluated the association between neonatal jaundice and autism diagnosis. Our findings indicated that this relationship was primarily confounded by perinatal factors, particularly gestational age, with little evidence for additional familial confounding.

These results underscore the importance of accounting for familial confounding.

List of scientific papers

I. Kanina A, Larsson H, Sjölander A, Butwicka A, Taylor MJ, Martini MI, et al. Association between cumulative psychosocial adversity in the family and ADHD and autism: a family-based cohort study. Transl Psychiatry. 2023;13(1):282. https://doi.org/10.1038/s41398-023-02571-7

II. Kanina A, Sjölander A, Martini MI, Butwicka A, Larsson H, Hughes A, Rádo MK, Taylor MJ, Havdahl A, Ask H, Rosenqvist M. Prenatal exposure to adverse life events and autism and autistic-like traits in children in the 1 Norwegian Mother, Father and Child Cohort Study (MoBa). 2024. [Accepted]

III. Kanina A, Sjölander A, Martini MI, Butwicka A, Ronald A, Rádo MK, Larsson H, Ask H, Rosenqvist M, Taylor MJ. Early life exposure to severe adverse life events in the family and diagnosed autism and autistic traits at age 3 in children: a genetically informative study. [Manuscript]

IV. Kanina A, Li Z, Rosenqvist M, Butwicka A, Larsson H, Johansson S, Rádo MK, Martini MI, Taylor MJ. The association between clinically diagnosed neonatal jaundice and autism: a Swedish register-based cohort study. [Manuscript]