Geri's DSM-5 Update from IMFAR

Monday, May 21, 2012 View Comments






May 18, 2012

Dear Friends,

I am pleased to provide another update on the DSM-5, after hearing a talk delivered by Sue Swedo, M.D., at the 2012 International Meeting on Autism Research (IMFAR), in Toronto.

As many of you are aware, Autism Speaks is playing a proactive role in monitoring the effect of the planned revisions to the definition of autism spectrum disorder (ASD) in the upcoming 5th edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), scheduled for publication in spring 2013. [See more of our DSM-5 coverage and comment here.]

Dr. Swedo chairs the committee responsible for recommending changes to the DSM criteria for autism. She began by saying that she wanted to make it clear that the committee had no intentions of reducing the number of people diagnosed on the autism spectrum and that it was the committee’s goal to ensure that the new criteria capture all people on the spectrum. She noted it is possible that the new criteria might capture people who had been previously missed because the new criteria include symptoms that were not part of the old criteria.

Dr. Swedo then reviewed the changes being proposed. Rather than three categories of symptoms (social, communication and repetitive behaviors), only two types of symptoms will be required (social communication and repetitive behaviors). “Pervasive developmental disorder” will now be simply called “autism spectrum disorder.” In fact, all subtypes will be collapsed under the one broad category of autism spectrum disorders.

Why? Because even expert clinicians cannot reliably distinguish among the different subtypes when they make their diagnoses. What one clinician calls high functioning autism, another calls Asperger syndrome. This lack of consistency occurs because the diagnosis of Asperger syndrome, as defined in the DSM-IV, is based on recall of early language acquisition. Such recall can prove difficult, especially when a diagnosis is being made later in life.

Another important difference in the DSM-5 criteria is that a much more detailed description – a set of “specifiers” – will be part of the diagnostic evaluation. It will be specified whether the child has intellectual disability or language delay. Similarly, the clinician will specify the pattern of onset (e.g. regression or early onset), any associated genetic cause and any accompanying medical conditions such as seizures and GI disorders. The intention is to capture all symptoms.

Dr. Swedo criticized the study, conducted by Fred Volkmar, that made headlines in the New York Times last spring. It was Volkmar’s report that raised concerns that a large percentage of individuals, particularly those who are higher functioning or have Asperger syndrome, would be excluded by the DSM-5 criteria. She pointed out that Volkmar’s study was based on an examination of diagnostic information collected as part of the DSM-IV field trials. The clinicians who collected that information wouldn’t have noted the new symptoms that are now part of the DSM-5. Thus, she argued, it is impossible to adequately assess the DSM-5 using retrospective information based on the DSM-IV.

She mentioned that all of the “hoopla” created from the New York Times article ended up being helpful. The feedback the committee has received since the article was published has led to clarifications in the way that the criteria are being written. For example, it became clear that many people didn’t understand that, in the DSM-5, symptoms can be based on either current functioning or past functioning. For example, an adolescent who had significant repetitive behaviors or social impairments early in life – but who no longer displays such behaviors – would still get “credit” for having such symptoms when a diagnosis is made. This expansion aims to capture people who may have shown some signs early on, but whose symptoms didn’t become evident until adolescence or adulthood.

Much of Dr. Swedo’s talk focused on the controversy surrounding the collapsing of autism subtypes into one category of autism spectrum disorder. She noted that the committee has now received over 6,000 comments about this decision. Approximately half of the comments supported the change; the other half did not. Some people expressed relief that they no longer had to try to explain the difference between autism and Asperger syndrome, whereas others were very distressed and felt that the diagnosis was an important distinction.

This became very evident during the Q and A that followed Dr. Swedo’s presentation. A person with Asperger syndrome came to the microphone and said, “You are basically erasing our identities, leaving us in limbo. Who are we?  We will feel lost. I implore you not to take it out. This will affect all Aspies.”

Dr. Swedo responded by saying, “Asperger syndrome will not disappear.” She stated that, in addition to the diagnosis of autism spectrum disorder, persons who have a diagnosis of Asperger syndrome can have that information noted as part of the diagnostic evaluation. They will still have Asperger syndrome. Dr. Swedo stressed that she “expected there to be continued growth in the rich scientific literature on Asperger syndrome.”

The results of the DSM-5 field trials were then presented. Overall, as I explained in my last update, the results suggest that the DSM-5 works pretty well. Out of the hundreds of children assessed, a handful of children who were previously diagnosed with an autism spectrum disorder based on DSM-IV no longer met criteria and were instead considered to have a diagnosis of “Social Communication Disorder.” In other words, they had difficulties with social communication but did not have repetitive behaviors. A handful of children who did not qualify for a diagnosis of ASD based on the DSM-IV criteria did receive a diagnosis of ASD based on DSM-5. So overall, the number of children diagnosed with ASD was similar based on the two sets of criteria, but the composition of children who received an ASD diagnosis was slightly different for each set of criteria. 

The DSM-5 includes changes that may indeed capture people who previously did not qualify for a diagnosis based on the DSM-IV. For one thing, the symptom of “sensory sensitivities” is included for the first time as a symptom under the category of repetitive behaviors. The age of onset is now defined as “early childhood” rather than “less than three years” to capture those children whose impairments might not be evident until they are in school and negotiating a complex social environment. Symptoms can be based on a history of their presence rather than requiring that the person currently have such symptoms.

Dr. Swedo emphasized that the committee was concerned that the changes “first do no harm,” especially in terms of excluding people from services they need.

How will the changes affect prevalence estimates?  This question was posed to Dr. Swedo during the Q and A. Her answer was, “I don’t know.” Her best guess – informed by the results of the field trials – is that the number will be similar. Fortunately, we will have a more definitive answer to this question based on research being funded by Autism Speaks. It will compare the two sets of criteria in a prospective study. I’ll have more to say about that study in my next update.

Warmest regards,




Geraldine Dawson