The latest edition of the diagnostic manual, DSM-5, which debuted last May, folds these categories into a single diagnosis of autism. Newschaffer says that even the next CDC report, which will use data from 2012, will be based on the DSM-IV. Eventually, however, the DSM-5 criteria may affect autism prevalence, possibly excluding some individuals who would be diagnosed with PDD-NOS under DSM-IV guidelines.
In the meantime, the new estimates of prevalence may signal another kind of shift. “There’s greater awareness in the community around autism, more training of clinicians, more early childhood educators — that whole effort has increased awareness,” Boyle says.
Variations in awareness may also underlie the differences in numbers from state to state. One reason for this, experts say, is the wide variability in the resources available for diagnosing and serving children in different communities.
For example, New Jersey, which has the highest reported prevalence, is known for providing extensive services for people with autism, notes Rebecca Landa, director of the Center for Autism and Related Disorders at Kennedy Krieger Institute in Baltimore, who was not involved in the study. “There’s a lot of awareness among professionals in that state,” she says.
In contrast, in largely rural states such as Alabama, “we don’t know how much the media is penetrating into small communities; we don’t know what kinds of hardships families in some of these communities are having,” she says.
What’s more, intellectual disability is more common in black and Hispanic children with autism than in white children with the disorder, suggesting that children in these groups with milder symptoms may be going undiagnosed.