Catching and Studying Autism Early: The 1-Year Well-Baby Check-up Approach

Background: Early developmental screening is essential, not only to initiate immediate treatment and improve outcome for children with an autism spectrum disorder (ASD), but to allow prospective research designs. Here we discuss pilot results of a new prospective method, called the “1-Year Well-Baby Check-Up Approach” (1Yr-CU) that is both simple and practical, and is initiated by pediatricians as part of their regular routine. Using this method, infants in the general population are screened and tracked in an unbiased fashion thus allowing for studies of both fundamental behavioral (e.g., exploration) and biological features (e.g., patterns of brain growth) of autism at 12 months of age. Importantly, because the screening tool used (i.e., the CSBS Infant-Toddler-Checklist, Wetherby and Prizant, 2002) was originally designed to detect communication delays and not autism per se, the early bio-behavioral phenotype of key contrast groups, namely developmental delay (DD) and language delay (LD), can be studied as well. Objectives: (1) To determine the percentage of infants detected via the 1-Yr CU that test positive for an ASD at an older age (i.e., 30-36 months) and the unique pattern of screen scores that may distinguish infants at-risk for an ASD from those at-risk for a LD or DD; (2) To determine the ease of implementation of this program as indexed by pediatricians; (3) To identify biological and behavioral features that may capture a unique signature of infants at risk for an ASD in contrast to those that are DD, LD, or typically developing. Methods: This large-scale project is currently ongoing. To date, eighty-five pediatricians have screened 7,655 babies using the CSBS at the 1-Year Well-Baby Check-Up. Infants who failed the screening form were referred to our laboratory for further testing that included widespread biological (e.g., MRI) and behavioral testing (e.g., interaction patterns with mother, environmental exploration). Typically developing infants were referred as well. ANOVAs and other statistical tests were used as a first order examination of group differences on selected measures. Within the next six months as more infants reach an age of final diagnosis, multivariate regression techniques will be used to examine the single as well as combined behavioral and biological features that may distinguish infants at-risk for an ASD from those at-risk for a DD or LD. Results: Preliminary analyses indicate that of the infants that failed the screen at the 1-Yr CU, rates of ASD detected are generally consistent with published studies (Fombonne et al., 2005), taking into consideration that Asperger's, late onset and regression cases would not be expected to be detected at 12-months. Results also indicated that infants at-risk for an ASD have abnormal brain growth trajectories and display significantly different patterns of environmental exploration and interactions with caregivers. Additional analyses, such as multiple regression, will be performed to generate a “signature” of infants at-risk for an ASD. Conclusions: The 1-Yr CU Approach successfully detects a percentage of infants who will eventually manifest an ASD, is easily incorporated into pediatric practice, and allows for the ascertainment of behavioral and biological markers at 12-months.