The Role of Comprehensive Evaluation in the Differential Diagnosis of Autism in a Clinic Setting

Background:  With heightened public awareness, parents and professionals are increasingly likely to raise concerns specifically about autism spectrum disorders, and to seek the expertise of a variety of specialists.  Current guidelines from the Academy of Pediatrics (2010) recommend that children suspected of having ASDs should receive a comprehensive evaluation, but unfortunately these are not always available, and when they are, long waiting lists exist.  The present study examines findings from comprehensive evaluations conducted in a clinic setting at the Marcus Autism Center, in Atlanta, GA.  The center serves a diverse geographic and socioeconomic population and conducts more than 300 comprehensive autism diagnostic assessments within a year. 

Objectives:  The purpose of this study is to analyze diagnostic trends in a clinical population, specifically with regard to changes in diagnosis based on the use of a comprehensive assessment using standardized measures.   

Methods:  A record review was conducted of 326 diagnostic evaluations conducted between November 2010 and October 2011 at the Marcus Autism Center.  To be included in this study, the evaluation report had to include the following components: diagnostic interview, a developmental/cognitive measure (e.g. Bayley Scales, DAS-II), an adaptive measure (e.g. Vineland Scales), and the Autism Diagnostic Observation Scale (ADOS).  Information extracted from the reports included child’s age, gender, prior diagnosis, prior measures given, and primary diagnosis given after receiving the comprehensive assessment.  Of the 326 reports that were reviewed, 299 had sufficient data to be included in this study.

Results:  31% percent of the children in this sample had previously been seen by a specialist (developmental pediatrician, psychiatrist, psychologist, or neurologist) and 16% had already been given a diagnosis within the autism spectrum.  Of the children who had a prior ASD diagnosis, only 4(8%) were known to have been given an ADOS and/or ADI.   After receiving a full standardized assessment, 55% of the sample was found to meet DSM criteria for an ASD diagnosis and the average age of these children was 63 months. Differential diagnoses for the 45% of children who did not fall within the autism spectrum included the following:  Behavior disorder (11%), mood and anxiety disorders (8%), developmental delay/cognitive impairment (10%), and language disorders (8%).  Interestingly, of the 48 children who came into the clinic with a pre-existing diagnosis of autism, only 65% were found to meet diagnostic criteria for an ASD, based on the assessments given.   The remaining 35% were found to meet criteria for other diagnoses including behavior disorders, mood disorders, anxiety, developmental delay, and language disorders.  

Conclusions:  In a sizeable minority of cases (35%), a previous diagnosis of ASD was not confirmed once standardized measures of observable behavior were used.  This highlights the importance of comprehensive evaluations that include consideration of other childhood disorders and that measure functioning across multiple domains.  Furthermore, given the relatively high average age of diagnosis (5 years old), a continued emphasis on early screening and referral for comprehensive assessment in the toddler and preschool years is imperative so that the benefits of early intervention can be maximized.