197 - Early Identification of Autism in a High-risk Preterm Infant Population

Preterm infants are at increased risk for autism spectrum disorders (ASD) compared to term infants. Current screening tools yield high false negative and false positive rates with poor accuracy for detecting ASD in preterm children. The Communication and Symbolic Behavior Scales Developmental Profile Infant-Toddler Checklist (CSBS- ITC) has shown promise as a screener in preterm research populations, but data remains limited on use in a clinic-based preterm cohort.

Objective:

To assess use of the CSBS-ITC as an early screening tool in a high-risk infant follow-up program to identify preterm children later diagnosed with ASD.

Design/Methods:

Preterm infants < 34 weeks gestation seen in a high-risk infant follow-up program were screened at 9-18 months corrected age (CA). Infants underwent serial testing using the Bayley Scales of Infant Development, 3^rd edition (Bayley-III) and were invited to complete the Autism Diagnostic Observation Scale, 2^nd edition (ADOS-2). Given variable timing of developmental testing and loss to follow-up, infants were grouped based on CA at last Bayley-III testing. Results from the CSBS-ITC were compared to Bayley-III, ADOS-2, and clinical concern for ASD at last developmental testing. Kaplan-Meier survival analysis was used to compare infants with and without concern on the CSBS-ITC to identify differences in rates and time to ASD diagnosis. Data were obtained from patient electronic medical records.

 

Results:

A total of 54 preterm infants were screened using the CSBS-ITC with 22 (40.7%) demonstrating concern. Median CA at last Bayley-III administration was 10.8 months (group 1), 22 months (group 2), and 35.4 months (group 3) with no other significant differences between groups. Of infants with concern on the CSBS-ITC, 5 (22.7%) were later diagnosed with ASD based on ADOS-2 testing, whereas only 1 (3.1%) infant without concern was later diagnosed with ASD. Kaplan-Meier analysis of all infants revealed those with concern on the CSBS-ITC demonstrated a probability of 0.63 (or 63% risk) of ASD diagnosis by 36 months, compared to 0.20 (or 20% risk) among patients without concern (log-rank p-value: 0.02).

Conclusion(s):

Use of the CSBS-ITC in a high-risk infant follow-up program demonstrates promise as an early screener for ASD in preterm infants. Early identification of ASD risk using a routine screener can lead to earlier diagnosis and intervention, thereby improving long-term outcomes. Given our positive findings, future studies should evaluate the CSBS-ITC as an early screener for ASD, focusing on feasibility of use in larger, multi-center clinic-based preterm infant cohorts.