How does DSM-5 change the diagnosis of autism?

Both patients and professionals have raised concern that the new DSM-5 criteria will prevent many patients from meeting the diagnosis for autism. Numerous reviews have debated the advantages and concerns of the new criteria for autism spectrum disorders (ASD). The prevalence of autism has increased dramatically in the past 10 years in part due to increased public awareness but also as a result of diagnostic practices. In response, DSM-5 was developed with the goal of a clearer, simpler, more reliable diagnosis, which recognized the “essential shared feature of the autism spectrum.”

DSM-5 presents some changes for diagnosing autism. The diagnoses of autism, Asperger’s disorder and pervasive developmental disorder have been replaced by one diagnosis of ASD. This move was prompted because experts don’t agree on which patients would be diagnosed with Asperger’s disorder and which patients would be diagnosed with autism using DSM-IV. Also, research has not shown that the outcome for Asperger’s disorder is different from that of high-functioning autism.

DSM-5 combines criteria for social and language symptoms into one domain of social, communicative and interactive problems, and now all three criteria must be met. The manual continues with the same set of criteria for repetitive, restricted behaviors, but also includes symptoms for sensory problems.

Studies comparing DSM-IV and DSM-5 criteria for autism show that DSM-5 will diagnose ASD less often. For example, a well-publicized study by McPartland and colleagues showed that 61% of cases with autism disorders would not meet DSM-5 diagnoses. A more recent study with a larger patient group showed that DSM-5 would diagnose 91% of those children with a DSM-IV diagnosis of an autism disorder. It's worth noting that this study was co-authored by a member of the committee tasked by the APA with developing DSM-5 criteria for autism.

However, studies show DSM-5 is more specific than its predecessor; if a clinician using DSM-5 diagnoses a patient with autism, it is more likely the patient actually has the disorder. Various authors have encouraged modifications to DSM-5 to require fewer symptoms to be present to meet the diagnosis. However, the committee that created the new criteria did not accept these suggestions. Instead, DSM-5 allows symptoms to be present either by history or at present and features more descriptive examples to meet the diagnosis.

In summary, DSM-5 is an evolutionary step for the diagnosis of autism based on consensus opinion of leading experts in the field. Ongoing, prospective studies with large samples comparing DSM-IV and DSM-5 criteria will answer many of the remaining questions. Until research-based biologic diagnoses can be made, as proposed by the NIMH with the Research Domain Criteria project, it is the best method we have for reliable and valid diagnosis.

For more information:
McPartland JC. J Am Acad Child Adolesc Psychiatry. 2012;51:368-383.
Huerta, M. Am J Psychiatry. 2012; 169: 1056-1 064.