It has long been believed that the increase in autism rates across New Jersey and the rest of the country is due, in part, to the improvements in diagnosis and classification.  A new study at Penn State University suggests as much.

(Devonyu, ThinkStock)

Researchers looked at special education enrollment figures and found that 97 percent of the increase in autism seen between 2000 and 2010 could be accounted for by reclassification. However, some professionals are skeptical.

"We have known for a long time that part of the increase in autism prevalence over time is due to the shift in diagnosis and changes in the way we diagnose autism. It certainly accounts for part of the increase, but doesn't account for the entire increase," said Michael Rosanoff, director of public health research at Autism Speaks. "This study only looked at education records and special education at that, so the question is what about all those high functioning or less affected individuals who are not enrolled in special education?"

Even the broadening of diagnostic criteria and the better recognition of autism cannot explain the rapid increase in cases over the past two decades, according to Rosanoff.

"The increase is in part due to environmental factors that we have yet to identify and that is where research is going," he said.

According to research reported in the American Journal of Medical Genetics, two-thirds of the increase could be due to reclassification.

"We have seen a number of shifts over the years in the way autism is diagnosed. One shift resulted in an increase in autism prevalence because the criteria was broadened. More recently, we've shifted to different criteria which actually constricts the diagnosis of autism and we're seeing an artificial decrease because of it," Rosanoff said. "We need to be very cautious in interpreting how changes in diagnostic criteria effect the reported prevalence numbers because that may not be the true number of individuals that are affected by autism and that can benefit from services."