ASD Diagnosed and Then Ruled out by Community Professionals in a Population-Based Study

Background: Autism spectrum disorders (ASDs) are developmental disabilities that affect social, communication, and behavioral development and are often associated with intellectual disabilities (ID). Although ASDs are generally accepted as life-long conditions, recent studies indicate that 10-20% of children once diagnosed with an ASD no longer meet diagnostic criteria when evaluated several years later. However, it is uncertain whether the likelihood of ruling out an ASD diagnosis after confirming an ASD diagnosis is increasing over time and what factors predict a change in ASD classification.

Objectives: We used a population-based surveillance dataset to examine changes in ASD diagnoses, specifically examining whether a community professional excluded an ASD diagnosis that had previously been confirmed by the same or another community professional during surveillance years 2000 to 2006. We also examined whether cognitive functioning or the original ASD subtype noted predicted a change in ASD classification and alternative diagnoses given when an ASD was ruled out.

Methods: Participants were identified from the Autism and Developmental Disabilities Monitoring (ADDM) Network. Study clinicians applied a standardized coding scheme based on DSM-IV-TR to information contained in health and educational records of children with conditions associated with ASDs (e.g., language and cognitive delays) to determine surveillance case status. Clinicians also recorded diagnostic impressions of community professionals who authored each report; which included confirming ASD diagnoses, ruling out ASD diagnoses, and giving alternative diagnoses such as language delay.

Results: A total of 4,958 children had surveillance records reviewed; 1,394 of these children had a known age of first ASD diagnosis. Of these 1,394 children, only 63 (5%) had an ASD diagnosis ruled out after an ASD diagnosis was confirmed. Children without ID were more likely to have an ASD ruled out after an ASD was confirmed than children with ID (p=.012, OR=2.84; 95%CI=1.26-6.45). The likelihood of having an ASD ruled out after an ASD was confirmed increased from 2000 to 2006 (p=.009; OR=2.42; 95%CI=1.23-4.67); this increase was limited to children without ID (defined as IQ>70; p=.005; OR=7.75; 95%CI=1.84-32.54). Original ASD subtypes among children who later had an ASD ruled out were autistic disorder (37%), general ASD or PDD-NOS (33%), subtype not specified (25%), and Asperger’s disorder (5%). The most common diagnoses given when an ASD was ruled out were language delay or disorder (38%) and ADHD (22%).

Conclusions: We found fewer children with an ASD ruled out after an ASD was confirmed than recent reports, indicating stability in ASD diagnoses in this population. However, the likelihood of rejecting an ASD after confirming an ASD increased over time for children without ID. The subtype of autistic disorder was just as common as the subtype of general ASD or PDD-NOS in children who later had an ASD ruled out; suggesting subtype is less influential in predicting changes in ASD classification than cognitive functioning. Our results also suggest that ASDs may sometimes be difficult to accurately distinguish from language delay or disorder and ADHD since these diagnosed were commonly confirmed when an ASD was ruled out.