Where a Child Is Tested May Affect Autism Diagnosis
MONDAY Nov. 7, 2011 -- Whether health professionals diagnose a child with autism or with a milder form of the neurodevelopmental disorder such as Asperger syndrome isn't determined only by the child's symptoms, but by where the child is assessed, a new study finds.
Researchers found wide variability at autism centers across the nation in the criteria health professionals used to determine whether a child should be diagnosed with "autistic disorder" or instead receive a diagnosis of a subtype of autism, including pervasive developmental disorder-not otherwise specified (PDD-NOS) and Asperger syndrome.
The study authors said the findings support the growing movement among some autism experts to do away with distinctions such as Asperger and instead put everyone who meets certain criteria under the same umbrella, "autism spectrum disorder."
In the study, researchers collected data on about 2,100 children aged 4 to 18 who met the criteria for an autism spectrum disorder and had been seen at one of 12 university-based autism centers across the United States. The children had all undergone a battery of tests commonly used to measure language, thinking and communications skills, and behavior. A psychologist, psychiatrist or other health professional with autism expertise used the results of the tests and an examination to make the final diagnosis.
At one site, every child was diagnosed with autism; no child was diagnosed with Asperger or PDD-NOS. Conversely, another site labeled about 40 percent of kids as having Asperger and 40 percent with PDD-NOS, leaving only 20 percent with autism.
The reason for the inconsistency was that each site appeared to be using its own cutoffs and criteria for autism versus Asperger and PDD-NOS. For example, at one site, kids with an IQ of over 70 were given an Asperger diagnosis, while at another site, kids had to have an IQ of 115 to be given an Asperger diagnosis, said lead study author Catherine Lord, director of the Institute for Brain Development at Weill Cornell Medical College and New York-Presbyterian Hospital.
For the 61 percent of children with severe language, communications and social deficits, and clearly repetitive behavior -- all classic signs of autism -- there was good consistency among the various centers in the autism diagnosis, according to the report in the Nov. 7 online issue of the Archives of General Psychiatry.
But for the 39 percent of children with milder problems, there were large inconsistencies in whether the child was labeled with autism, Asperger or PDD-NOS, the investigators found.
"The second most important predictor of which diagnosis the clinician made was where they worked, not any characteristic of the child, which is not good," said Lord.
Geraldine Dawson, chief science officer for Autism Speaks, agreed that it may be time do away with diagnosing specific autism subtypes and instead use the broader, and more reliable, autism spectrum disorder.
"Given the subtle distinctions among the subtypes of autism spectrum disorder, it is not surprising that clinicians were not consistent in how they applied these diagnoses," Dawson said. "In the future, it is likely that we will no longer attempt to classify individuals in the autism spectrum into different subtypes."
Furthermore, whether a child is diagnosed with Asperger or autism doesn't change the treatment.
There are several possible reasons for the discrepancies among autism centers. Parents may want to hear that their child has Asperger rather than autism, and health professionals may be influenced by that, Lord said.
On the other hand, in some regions, kids can only get certain services if they have autism, so those centers may apply that term more broadly.
Generally speaking, children with significant impairments in language, communications and social skills, who exhibit repetitive behaviors, and who have intellectual disabilities are diagnosed with autism.
Children who have some of the characteristics of autism, such as language delays and social problems, but who have only very subtle or not obvious repetitive behaviors, may be diagnosed with PDD-NOS. And those who have some characteristics of autism but who have good verbal skills, or who are highly intelligent, or who have some less obvious social impairment, are often said to have Asperger, Lord noted.
Within individual centers, health professionals were consistent on the criteria they used to diagnose children. The discrepancies emerged when comparing one center's diagnostic criteria to another's, she explained.
"It's really much more valid to talk about autism spectrum disorder," Lord said, with the acknowledgment that the people who fall within it can have a very wide range of abilities.
"Using one broad category and then characterizing the specific strengths and challenges of each child may be clinically more useful," added Dawson. "That said, people who have come to identify with a specific diagnosis, such as Asperger syndrome, may want to continue to use that diagnostic label. We need to be sensitive to the fact that some people with autism spectrum disorder may not embrace these diagnostic changes."
The U.S. Centers for Disease Control and Prevention has more about autism spectrum disorders.
Posted: November 2011