787 - Developmental Trajectories and Autism Risk: A Longitudinal Study of Children Born Low Birthweight, with Autistic Siblings, or Typical Birth Peers

Publication Number: 787.4518

Kelsey J. Dommer, Seattle Children's, Seattle, WA, United States; Megan C. Reninger, Seattle Children's Research Institute, Seattle, WA, United States; Hope Willis, University of Washington School of Medicine, Seattle, WA, United States; Jenny AF. Skytta, Seattle Children's, Shoreline, WA, United States; Sarah Corrigan, Seattle Children’s, Seattle, WA, United States; Kristen E. Michener, Seattle Childrens Research Institute, Seattle, WA, United States; Sandra E. Juul, University of Washington, Seattle, WA, United States; Dennis E. Mayock, UWMC, Lake Forest Park, WA, United States; Dimitri Christakis, University of Washington School of Medicine, seattle, WA, United States; Sara Jane Webb, Seattle Children's Research Institute, Seattle, WA, United States; Frederick Shic, University of Washington School of Medicine, Seattle, WA, United States

Background: Children born preterm, very low birthweight (LBW) or with autistic siblings may show atypical developmental trajectories, including higher rates of autism spectrum disorder (ASD) (Caravale et al, 2005; Wolfe et al, 2002; Lampi et al, 2012). Parsing out developmental differences in these groups could help hone screening tools and lead to more precision interventions.

Objective: Evaluate developmental trajectories of children at elevated likelihood for ASD and investigate whether trajectories differ for autistic and non-autistic children based on their likelihood group at birth.

Design/Methods: Data were collected from 257 children in a longitudinal study at approximately 7, 14, 28, and 39 months of age. Participants were categorized as low likelihood (LL) or elevated likelihood (EL) due to LBW ( < 2400g, LBW) or having an older autistic sibling (Sib). We analyzed diagnostic outcomes and developmental assessments using the Brief Observation of Symptoms of Autism (BOSA), Mullen Scales of Early Learning (Receptive Language and Visual Reception), and Vineland Adaptive Behavior Scales – 3rd Edition (Social, Communication, Daily Living Skills, and Motor subscales). Linear mixed models predicted scores by likelihood group, diagnostic status (ASD, non-ASD), and age with random participant intercept.

Results: Chi-square tests showed significantly higher ASD rates in Sibs compared to both LBW and LL (p <.001). No difference in rate of diagnosis of ASD between LBW and LL was identified. LBW trended towards higher atypical non-ASD diagnoses than LL (p=.057) (Table 1). All models showed significant effects of age and interactions between diagnosis*age (p <.05) (Figure 1). Amid other results, BOSA scores indicated fewer autism symptoms in LBW non-ASD compared to LL non-ASD at 28-months, while LBW-ASD had higher symptomology than LL-ASD at 39-months (p <.05). Broad trends suggested LBW non-ASD tested worse than LL non-ASD, especially at older ages. The Sib-ASD group also scored lower than LL-ASD in various domains of the Vineland, while LBW-ASD scored higher in the Social subscale at 14-months (p <.05).

Conclusion(s): Autistic children with an autistic sibling exhibited greater developmental deficits than their LBW- or LL-ASD counterparts. Non-autistic LBW children exhibited vulnerabilities to developmental deficits by 28-months, yet demonstrated fewer autism symptoms than LL non-ASD peers at two years, possibly exhibiting a relative strength in sociality. LBW-ASD also showed early selective social strengths compared to LL-ASD peers, despite higher autism symptoms by age three.

Table 1
Diagnostic outcome by recruitment group

Figure 1
Linear mixed models evaluating Outcomes predicted by group, diagnosis, and age (Outcome ~ group * diagnosis * age + (1|id)). Groups shown are Elevated Likelihood due to being born low birthweight (EL-LBW) or having an older autistic sibling (EL-Sib) or Low Likelihood (LL) of developmental delay and/or autism.