The Relationship of Early Childhood Characteristics to Autism Severity at School Age

Saturday, May 13, 2017: 12:00 PM-1:40 PM
Golden Gate Ballroom (Marriott Marquis Hotel)
P. Towle1 and N. Turygin2, (1)Westchester Insitute for Human Development, Valhalla, NY, (2)Behavioral Psychology, Westchester Institute for Human Development, Valhalla, NY
Background: Children with autism spectrum disorder (ASD) are diagnosed at increasingly early ages (prior to age 3). It is widely known that outcomes vary widely; some toddlers diagnosed with ASD improve greatly as they age, while others continue to experience severe challenges. However, the symptoms most predictive of the future presence and severity of the disorder in later childhood are not well understood. The present longitudinal study explored early predictors of later autism severity and adaptive behavior.

Objectives:  The present study explored the relationship of early symptoms and developmental level to the later presentation of the disorder. Specifically, we examined the relationship of ASD symptom severity and adaptive behavior scores at diagnosis (prior to age 3), gender, and age of diagnosis, to symptom severity and adaptive behavior scores at follow-up (school age).

Methods:  This longitudinal study examined 70 children who were diagnosed with ASD prior to age 3 (Time 1) and who were 7 years or older (Time 2) (range 7-16 years). The participants were identified from review of early intervention clinical charts at a University Center for Excellence in Disabilities. Approximately half the children were diagnosed before age 24 months. Time 1 data were extracted from early intervention charts, and included: age at diagnosis, sex, and scores from the Vineland Adaptive Behavior Scales (VABS I & II) and the Childhood Autism Rating Scale (CARS). Parents who agreed to participate were mailed a packet containing a consent form, the Gilliam Autism Rating Scale (GARS-3), the Social Communication Questionnaire-Current (SCQ-C) and an extensive parent questionnaire. Once the packet was returned, a phone call was made to administer the VABS- II. Based on the parental responses, GARS-3, SCQ-C, and the VABS-II Daily Living Skills score, participants were divided into one of the three diagnostic/disability categories: ASD with moderate to severe disability, ASD with mild disability, and no ASD diagnosis (but may have had other diagnoses such as learning disabilities and ADHD).

Results:  Data were analyzed using ANOVA, ANCOVA, and correlations. In terms of the three diagnostic categories at school-age, 52.2% retained a diagnosis of ASD with moderate to severe developmental disability, 29.0% continued to exhibit ASD with milder disability, and 18.8% no longer met diagnostic criteria for ASD. Autism symptoms at Time 1 were significantly associated symptoms at Time 2 for measures of core ASD symptoms and adaptive behavior. However, early socialization scores were most predictive of both later adaptive behavior scores (Adaptive Behavior Composite, Communication, Socialization Vineland Subscales) as well as later autism severity (GARS total, GARS social, and GARS RRB). As well, the CARS Hyperactivity item was differentially strongly associated with later symptom severity.

Conclusions: Several variables predict later functioning in children with early diagnoses of ASD. Socialization scores were predictive of both later adaptive functioning and autism severity at follow up, as was the single CARS item of hyperactivity, demonstrating continuity in child functioning features over several years. The significant relationships and continuity is striking, given that in this data set, from 4 to over ten years have lapsed since the measurements were taken.