New Autism Criteria Will Not Exclude Affected Kids

But Concerns Linger Among Autism Groups

Pam Harrison

October 05, 2012

October 5, 2012 — A redefinition of the diagnosis of autism will not exclude children with the disorder and render them ineligible for services, a new study shows. However, despite these findings, some in the autism community still have concerns.

Marisela Huerta, PhD, and colleagues from the Weill Cornell Medical College, New York City, are reporting that the new diagnostic criteria for what will now be called autism spectrum disorder (ASD) in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) identified 91% of children previously diagnosed with pervasive developmental disorder (PDD) on the basis of DSM-IV criteria.

Many of the remaining 9% would likely have the diagnosis reinstated with clinician input.

"I know that parents worry, but I don't believe there is any substantial reason to fear that children who need to be diagnosed with autism spectrum disorders and provided with vital services will not be included in the new criteria in this updated [DSM-5] manual," senior investigator Catherine Lord, PhD, director of the Center for Autism and Developing Brain, New York–Presbyterian Hospital's Westchester campus, in New York City, said in a press release on the study.

Dr. Catherine Lord

The study was published in the October 1 issue of the American Journal of Psychiatry.

High Sensitivity

Investigators used previously collected data from 3 different sources to evaluate current DSM-5 criteria in 4453 children with a DSM-IV clinical PDD diagnosis along with 690 others with non-PDD diagnoses.

All participants had previously undergone diagnostic testing that included the Autism Diagnostic Interview–Revised, the Autism Diagnostic Observation Schedule, and cognitive or developmental testing.

For the analyses, investigators relied primarily on the Autism Diagnostic Interview–Revised, a 96-item parent report measure.

"On the basis of parent-reported symptoms only...the sensitivity of the proposed DSM-5 criteria ranged from 0.89 to 0.93," the investigators write, "[whereas] on the basis of either parent report or clinical observation, DSM-5 sensitivity ranged from 0.97 to 0.99 for any PDD."

Again on the basis of parent report only, the DSM-5 sensitivity for the clinical diagnosis of Asperger's disorder or PDD–not otherwise specified ranged from 0.77 to 0.94.

The DSM-5 sensitivity for those with autism was very high at 0.93 to 0.95, the authors add.

Even when the researchers examined different PDD subgroups, they found that the new criteria had a high sensitivity for identifying girls with ASD as well as those in the high-functioning range of cognitive ability and those with a nonverbal IQ of 70 or less.

"Shifting a multicategorical diagnosis into 1 category gives people the impression that there is some culling of individuals involved," Dr. Hueta told Medscape Medical News.

"But it's more a reorganization of the symptoms as well as a way to assess the severity of the disorder and the support needed. The DSM-5 is really a guide that helps ensure we are using the same language to describe the disorder and that we are consistent in this process," she added.


Some autism groups have expressed concerns over the proposed collapse of the current diagnosis of PDD and its subtypes to a single diagnosis of ASD.

The overriding concern is that the new DSM-5 criteria would significantly reduce the number of new diagnoses for affected individuals, thus depriving them of important services.

Katie Weisman, director of communications and public policy at SafeMinds, told Medscape Medical News that the Weill Cornell study was "fair and honest," but she still had reservations about the ability of clinicians to generalize results from the study to the entire population of ASD individuals.

"Overall, the population they are reporting on is more severely impacted than the general autism population as reflected in the average IQ scores," she said.

Further, she said that 72% of the overall sample met the criteria for autistic disorder with only 1 of the data sets containing a reasonable sample of individuals with Asperger's disorder.

"The sensitivity or ability to pick up cases is similar between the 2 sets of criteria, but the DSM-5 criteria are more specific, which is exactly what you would expect using these populations that are weighted towards autism," Weisman observed. "[The criteria] also don't really address people on the higher-functioning end. I think the community's concerns still have validity, just maybe with somewhat less urgency."

SafeMinds and other concerned community groups are still inviting physicians to enter data on newly diagnosed ASD individuals to see how the proposed diagnostic changes might play out in a real-world setting.

Alycia Halladay, PhD, with Autism Speaks, a national patient advocacy group that also raises funds for autism research, told Medscape Medical News that those within the autism community are concerned about the proposed DSM-5 changes, because retrospective studies had suggested that from 23% to 47% of previously diagnosed cohorts might lose their diagnosis if the new DSM-5 criteria are applied.

"This study suggests that current criteria do capture almost everyone, although investigators are still using retrospective data," she noted. Because of this, Dr. Halladay emphasized that it is still critical to pursue ongoing field studies, which are prospectively evaluating children with face-to-face interviews.

Results from these prospective studies are the only valid way to ensure that current diagnostic criteria are as good as they seem to be, Dr. Halladay suggested.

"What we really want is for these children to maintain services," she added. "It's a bit tricky because some people with Asperger's disorder don't want to be labelled with autism, so we want people to keep their diagnosis and make sure they receive the care and the services they need."

"Promising" Diagnostic Rates

In a recent editorial in the Journal of Child Psychology and Psychiatry (2012;53:10:1092-1094), Sally Ozonoff, PhD, from the MIND Institute at the University of California, Davis, Medical Center in Sacramento, California, outlined the rationale for changing the current criteria.

"Since the DSM-IV came out, a lot of research has been published suggesting that there aren't a lot of empirical differences amongst the different subtypes of PDD and that we can't really differentiate these subtypes when we look at different variables," she told Medscape Medical News.

Certain diagnoses within the PDD subtypes are also less likely to receive services in different states than others, which skews the tendency for some diagnoses to be particularly uncommon in certain states.

"The net effect is that people use the different labels to try and secure services in various states, but to me, it's extremely unfair that services should be doled out or not on the basis of some diagnosis that science can't reliably differentiate," Dr. Ozonoff said.

In commenting specifically on the current study, Dr. Ozonoff reiterated that she was "generally enthusiastic" about the proposed changes to the DSM-5 because of all the inconsistencies and redundancies in DSM-IV diagnostic criteria.

Prior to this study, several other studies had reported the impact of DSM-5 changes on diagnostic rates, and the numbers were quite different from those reported by the Weill College group of investigators.

Because this study used current diagnostic criteria, however, "the diagnostic rates look very promising, as they had a 91% sensitivity [rate] in this study," Dr. Ozonoff said.

"Contrary to a field trial, this study still used previously collected data that was not designed for this purpose, so we are still waiting for the field trials," she said. "But this study is very reassuring."

Dr. Lord is a member of the DSM-5 Neurodevelopmental Disorders work group and reports receiving royalties for the Autism Diagnostic Interview–Revised and the Autism Diagnostic Observation Schedule. Dr. Lord donates all royalties to nonprofit autism charities. Dr. Ozonoff is not a member of the DSM-5 Neurodevelopmental Disorders work group and has disclosed no relevant financial relationships.

Am J Psychiatry. Published online October 1, 2012. Full article