Prenatal perceived stress and urinary cortisol as risk factors for ASD and non-typical developmental outcomes in the MARBLES study

Abstract

Background

Findings for prenatal stress, previously measured through stressful life event surveys items or biomarkers, in association with ASD are inconsistent. We prospectively examined prenatal perceived stress and prenatal urinary cortisol (PUC) in association with ASD and other non-typical developmental (Non-TD) outcomes in the child in a high familial likelihood cohort.

Methods

The Perceived Stress Scale questionnaire was used to measure perceived stress in the longitudinal Markers of Autism Risk in Babies: Learning Early Signs (MARBLES) Study. Cortisol was measured through 24-hr urine collections. At three years of age, an algorithm consisting of scores from the Autism Diagnostic Observation Schedule (ADOS) and Mullen Scales of Early Learning (MSEL) was used to classify children with ASD, Non-TD, or typically developing (TD) outcomes. Relative risk ratios (RRR) with TD as the reference were estimated using multinomial logistic regression.

Results

Increased perceived stress was associated with Non-TD in trimester 1 (RRR 1.10; 95% CI: 1.00, 1.21) and ASD in trimesters 2 and 3 (RRR 1.08; 95% CI: 1.02, 1.14 and RRR 1.08; 95% CI: 1.03, 1.14, respectively). Results with PUC were non-significant but were in the direction consistent with previous findings suggesting that decreased cortisol levels are associated with higher likelihood of ASD.

Conclusions

Findings support the hypothesis that higher perceived stress is associated with increased likelihood of ASD and possibly Non-TD, relative to TD. This suggests that stress reduction interventions during pregnancy could serve as preventative measures that help optimize the child’s long-term health. Larger studies are needed to replicate these findings.