Neonatal Markers of Infant Behavior and Quantitative Traits Related to Autism Spectrum Disorders in a Longitudinal Birth Cohort Study

Background: Identification of antecedent markers of neurodevelopmental disorders like autism spectrum disorders (ASD) can help to target interventions at earlier ages. Prospective cohort studies that assess children in infancy are uniquely positioned to examine early markers in relation to quantitative traits related to ASD later in childhood.

Objectives: To investigate associations of infant behavior, including the capacity to regulate internal state and respond to environmental stimuli, on quantitative traits related to ASD, measured in adolescence.

Methods: The CHAMACOS (Center for Health Assessment of Mothers and Children of Salinas) longitudinal birth cohort study investigates the health of pregnant women and their children living in a predominantly Mexican-American agricultural community. We recruited pregnant women ≥18 years old and <20 weeks gestation between October 1999 and October 2000 at prenatal clinics serving the Salinas Valley. Of the 528 live births in CHAMACOS, we administered the Brazelton Neonatal Behavioral Assessment Scale (NBAS) to 419 infants (median (25^th-75%ile) age: 9 (1-25) days). The NBAS measures 27 behavioral and 18 reflex items; we used the Lester scoring method (Lester et al. 1982) to distill these items into seven clusters: Habituation, Orientation, Motor, Range of State, Regulation of State, Autonomic Stability, and Reflex. When the CHAMACOS children were age 14 years, parents of 212 children with an NBAS were administered the Social Responsiveness Scale, Second Edition (SRS-2), a rating scale that quantifies the frequency of traits in their children related to social behavior and stereotypic behavior/restricted interests. We fit multivariable regression models to estimate associations of NBAS cluster scores with SRS-2 scores, adjusting for infant’s age (days) at NBAS, child’s sex, number of years the mother lived in the U.S., language the mother used to complete the SRS-2, maternal depression and a measure of the home environment (HOME score).

Results: Mean (standard deviation) SRS-2 Total T-score for 212 children with an NBAS was 55.5 (7.4), with 23 children scoring at or above the cut-off at which screening for ASD is recommended (raw score of 70). We found associations of poorer regulation in infancy (e.g., less self-quieting, poorer consolability) with higher odds of maternal-reported adolescent ASD traits in the range recommended for ASD screening (OR=1.60; 95% CI: 1.02, 2.51). In contrast, infants with poorer motor performance (e.g. less physical maturity, lower tone) had lower odds of meeting the threshold for ASD screening in adolescence (OR=0.43; 95% CI: 0.19, 0.99).

Conclusions: We found that regulation of state and motor function in infancy were associated with quantitative traits related to ASD in adolescence. Data from this longitudinal cohort contribute to a growing literature suggesting that early life markers may help to identify children that will later develop symptoms consistent with a diagnosis of ASD. This can have important implications for targeting intervention earlier with the goal of improving the lives of individuals living with ASD.