Government figures that have been cited to prove that autism is rapidly increasing in the United States are not reliable and thus unsuitable for tracking the disorder, a study said on Tuesday.

The U.S. Department of Education figures, based on the number of children receiving special education assistance, have internal “anomalies” and are in conflict with a number of studies on the prevalence of the condition, said the report from Portland State University in Oregon.

“Basically we don’t know what the true prevalence of autism is in this country,” the study’s author, physician James Laidler, said in an interview.

“The problem with the data is that it may be including kids who have problems other than autism - a less severe degree than was included even ... a year or five years ago - and it has some internal inconsistencies,” he added.

“That said, there may still be an autism epidemic in the United States,” but the figures most widely used to demonstrate that are not valid, he said.

The government figures in question estimate that autism as recorded in the U.S. public school population went from 5,415 cases in 1991-1992 to 118,602 in 2001-2002.

One figure used by the Autism Society of America and others is that about one out of every 250 babies in the United States is born with the disability. It usually appears in the first three years of childhood, permanently impairing development of those areas of the brain that control verbal and nonverbal communication as well as social interaction.

VACCINE LINK?

Its causes have not been determined, though several different theories have been suggested. Some have placed blame on childhood vaccines containing the preservative thimerosal, an organic compound that is 49 percent mercury and is no longer used in such vaccines in the United States.

The Institute of Medicine, an independent body that advises the federal government, said this year there is no evidence of any link between vaccines and autism.

The study said the government figures indicate there are as many children newly diagnosed as autistic at age 15 as there are at 8. However four recent studies found the majority of diagnoses are made before 8, it said.

The government figures also show a drop in the incidence of autism between age 11 and 12, a point at which increased diagnoses would be expected instead as children move from elementary school to a more unstructured middle school environment.

“These anomalies point to internal problems in the ... data that make them unsuitable for tracking autism prevalence,” concluded the report published in the July issue of Pediatrics, the journal of the American Academy of Pediatrics.

There are indications that the increasing awareness of autism among doctors and educators “may have led to a gradual shift in diagnosis to include less disabled individuals who would not previously have been described as autistic,” the study said.

In any case the government figures “are at odds with studies of autism prevalence, largely because the criteria used by the school districts to categorize children as autistic are neither rigorous nor consistent.

“They are inconsistent over time, as are the medical criteria, and are inconsistent from region to region (and thus) not reliable for tracking the prevalence of autism, and they in fact never were meant to fill this need,” it said.