More Documented Diagnoses of Autism Likely Responsible for Increased Identification of the Autism Spectrum Disorders

Background:  

The number of children with an Autism Spectrum Disorder (ASD) has increased substantially over the last few decades, which has lead to the hypothesis that increased ASD identification is due to the recognition of more children with typically milder ASDs. Yet no study has examined changes in the proportional distribution of children in differing diagnostic subgroups (e.g., Autistic Disorder versus other ASD) identified from a variety of providers within the community.

Objectives:  

Our goals were to examine 1) whether children identified by a population-based surveillance system were receiving more diagnoses of an ASD and fewer notes of ASD characteristics without a confirmatory diagnosis over time and 2) changes in the proportional distribution of diagnostic subgroups and how these changes differed based on the presence or absence of intellectual disability (ID).

Methods:  

Participants were 1,497 children who met the Centers for Disease Control and Prevention (CDC) ASD surveillance definition in Atlanta, GA in surveillance years (SYs) 2000, 2002, 2004, and 2006. Children were placed in one of five groups based on ASD diagnostic specificity identified by a community professional and stratified by ID: 1) a 299.0 billing code or diagnosis of Autistic Disorder, 2) a 299.8 billing code or diagnosis of ASD without a diagnosis of Autistic Disorder, 3) autism eligibility at a public school without a noted clinical diagnosis, 4) a documented suspicion of an ASD without a clinical diagnosis or autism eligibility at a public school, and 5) none of the above. Changes in the distribution of these diagnostic groups were evaluated using Bayesian multinomial logistic regression.

Results:  

The relative risk of being diagnosed with Autistic Disorder increased substantially from 2000 to 2002 (RR=1.7) and then stabilized in 2004 (RR=1.7) and 2006 (RR=1.8) for children with and without an ID (Table 1). There were no substantial changes in the proportion of children diagnosed with other ASD or children who received school services for an ASD who did not have a diagnosis of Autistic Disorder.  Fewer children identified by the surveillance system were classified as having no ASD diagnosis or school services or suspicion of an ASD noted in service records; suggesting the proportion of children in this group was decreasing while the proportion of children with a noted Autistic Disorder was increasing over time.

Conclusions:  

We found substantial changes in documented diagnoses of Autistic Disorder, which could be due to growing awareness of ASD symptoms and less reluctance to diagnose autism when enough behavioral symptoms are noted. The major shift in providing more diagnoses of Autistic Disorder occurred after the introduction of the DSM-IV in 1994 (i.e., SY 2002). Our findings do not support the conclusion that expanded diagnostic criteria has a major impact on the increased use of ASD diagnoses by community professionals since there were no changes in the proportion of children diagnosed with an ASD. However, autism is being diagnosed more now than in the past; which highlights the need for more services for children with ASD and continued support for their families.