DSM-5 Proposed Autism Spectrum Criteria – Update - June 13, 2012 By Katie Weisman, Director of Communications
The autism community is extremely concerned about potential impacts of the proposed criteria for Autism Spectrum Disorders in the American Psychiatric Association’s Diagnostic and Statistical Manual – 5th Edition. The new criteria are supposed to be finalized by December 2012 for May 2013 publication.
In February, an article in the New York Times leaked data from a retrospective study indicating that the proposed DSM-5 criteria reduced the diagnosis in the sample population by 39.4% compared to the DSM-IV Revised criteria. The response from the DSM-5 workgroup on Neurodevelopmental Disorders that is writing the new criteria was that this was not an appropriate sample since the original diagnosis would not have captured some symptoms that are included in the new criteria. Since that study (McPartland/Volkmar) came out in April, there have been three more publications regarding the proposed criteria.
The Worley/Matson study showed a drop of 32.3% when using DSM-5 criteria vs. DSM-IVR criteria on a sample of 208 children and adolescents (Mean age of 8.28, SD 3.28). The study also found that those subjects who met the old criteria but not the new criteria “still exhibited significant overall symptoms of ASD.”
The Matson et al. study screened a population of 2721 toddlers (age 17-36 months) at risk for a developmental disability. 415 toddlers met the criteria for an autism spectrum disorder using the DSM-5 criteria and an additional 380 toddlers met the criteria for either Autistic Disorder or PDD-NOS based on the DSM-IVR criteria. The researchers assumed that any child who met DSM-5 criteria would also meet DSM-IVR criteria since it is less restrictive. Their conclusion was that, potentially, there could be a 47.79% decrease in diagnosis as a result of the proposed criteria. Children who met criteria for PDD-NOS were disproportionately impacted – 79.94% of them did not meet the proposed DSM-5 criteria for ASD. A weakness of this study is that it did not actually diagnose the children. The data was based on care-giver reports.
The Gibbs et al. study was just published. It was done in Australia and the team actually did new diagnoses on 132 referral cases. The subjects ranged from 2 to 16 years of age (Mean=6.06; SD=3.38). The ADOS and the ADI-R were administered to each child and their caregiver. 111 subjects were diagnosed with Autistic Disorder, PDD-NOS or Asperger’s Disorder using the DSM-IVR criteria. Of those, 26 did not meet the criteria for Autism Spectrum Disorder using the proposed DSM-5 criteria – a decrease of 23.4%. Again, the children diagnosed with PDD-NOS were most impacted – comprising about 2/3 of the decrease. This study identified the requirement for two criteria under the Restrictive and Repetitive Behavior domain as the most common reason for exclusion.
The Field Trials, as reported by Dr. Swedo at IMFAR, indicate that the decrease in the number of identified ASD cases using the proposed criteria would be in the single digits, but that this would be somewhat counter-balanced by the inclusion of some cases that had been missed by the DSM-IVR. I have not been able to get the exact numbers from Dr. Swedo, but I am still trying. The concern expressed in the Autism Speaks letter below is that the field trials included only 83 cases – all children.
The Bottom Line At This Point We don’t have enough data to be sure of the exact impact of the new criteria on autism spectrum disorders, but the information so far indicates that there will likely be a drop in cases and that the drop will likely be significant. The studies indicate that the children formerly diagnosed with PDD-NOS will be the most impacted. Also, at this point, there is absolutely no data on how the proposed criteria will impact adults on the spectrum.
The primary concerns of the community are all still valid at this point1. That a significant portion of the ASD population may be excluded by the proposed criteria – thereby impacting educational and Medicaid services and insurance. 2. That the tracking of incidence and prevalence numbers will be impacted making it very difficult to continue to follow the trends in ASD diagnosis. 3. That there is insufficient research on this issue and that much more research needs to be done.
1. Comment today or tomorrow on the APA website about your specific concerns. Go to . Please put Autism or ASD in your subject line. Public comments end on Friday, June 15th.
2. A parent workgroup lead by SafeMinds and the Holland Center is developing an online survey to collect our own data in the “real world”. Watch for announcements and please help us distribute it to as many clinicians as possible for their input. The more data we can collect, the better able we will be to take action to resolve this issue.
3. Read and consider signing the GRASP petition to the NDD workgroup within the blog post linked below.
4. Support state legislation in New York, Illinois, New Jersey and other states that are attempting to protect insurance coverage for those with ASD. Sign up for alerts from the Autism Action Network at