A National Study of Autistic Symptoms in the General Population of School Age Children and Those Diagnosed with Autism Spectrum Disorder (ASD)

Background: Autism spectrum disorders (ASD) are associated with myriad developmental difficulties. However, many of its current diagnostic symptoms were historically arrived at by consensus rather than research.  As part of the development of the Autism Spectrum Rating Scales (ASRS; Goldstein & Naglieri, 2010), parent and teacher ratings of clinical and non-clinical children were obtained, affording the opportunity to study autistic symptoms in a nationally representative sample as well as those diagnosed with ASD and other disorders.

Objectives: The purpose of the present paper is to identify the key symptoms of ASDs as measured by the ASRS, using a large representative sample of children and youth.  Results would provide empirical support and direction for the accurate diagnosis of ASD.

Methods: Participants included two samples of children aged 6- to 18 years from 70 data collection sites in the U.S. and Canada. Parents rated 1,881 youth (54.4% male; mean age 11.0 years, SD = 3.7 years; 65.9% White).  Most cases (76.9%) were without a formal clinical diagnosis of any psychological disorder.  Of those cases with a diagnosis, the most common were ASD (43.4%) and ADHD (23.9%). Teachers rated 2,171 participants (51.4% male; mean age = 11.9 years, SD = 3.7 years; 58.5% White).  Most cases (76.6%) did not have a formal clinical diagnosis of any psychological disorder and of those that did, the most common were ASD (41.9%) and ADHD (25.2%). The ASRS was designed to assess ASD-relevant behaviors as rated by parents and/or teachers, and shows strong reliability and validity (Goldstein & Naglieri, 2010).  The 71 Likert-style items are based on a comprehensive review of current theory and literature on the assessment of ASDs, DSM-IV-TR and ICD-10 diagnostic criteria, and the authors’ clinical and research experiences.  Exploratory factor analyses (EFA) and factor congruence analyses were conducted separately on the parent and teacher data to identify the ASRS factor structure and its replicability across demographic groups.

Results:  Various EFA criteria suggested the presence of three correlated factors:  (a) Unusual Behaviors (stereotypical and repetitive behaviors), (b) Self-Regulation (inattention, impulsivity, non-compliance), and (c) Social/Communication (difficulties with social interactions, verbal and non-verbal communication).  Factor congruence analyses showed replication of the factors between sexes (males vs. females), ethnic groups (White vs. non-White), age groups (6-11 vs. 12-18 years), and clinical status (clinical vs. general population).

Conclusions: Results from this study provide empirical support for a three-factor interpretation of ASD: Unusual Behaviors, Self-Regulation, and Social/Communication.  There are several important implications of this factor structure.  First, the Self-Regulation factor that emerged is novel in major symptomatic descriptions of ASD.  Its strong correlation with the other two ASRS factors implies it is a core feature of ASD, not a simple co-morbidity.  Second, social and communication symptoms merged into a single factor, contradictory to many current conceptualizations of ASDs (e.g., DSM-IV-TR).  Finally, the large nationally representative sample of general population and clinical children, combined with the consistency of the factor structure across various demographic groups and in both parent and teacher ratings, suggests these results are highly generalizable.