Although the evidence supporting claims for an "epidemic" of autism is weak, the "subsequent controversy has put autism on the public agenda," notes Dr. Eric Fombonne, who was not involved in the study.

In the new study, Dr. Marshalyn Yeargin-Allsopp and colleagues looked at the prevalence of autism in 289,456 children aged 3 to 10 living in Atlanta in 1996. They found that 34 of every 10,000 children had symptoms of autism, a disorder characterized by social isolation, difficulty communicating, repetitive behaviors and delayed and unusual speech.

The findings are published in the January 1 issue of the Journal of the American Medical Association.

"The rate of autism found in this study was higher than the rates from studies conducted in the United States during the 1980s and early 1990s, but it was consistent with those of more recent studies," the authors conclude.

Only four similar studies have been conducted in the US, they note. Three done in the 1980s and early 1990s found that about 4 per 10,000 children had autism. The new results concur with those from a 1998 study conducted in Brick Township, New Jersey that found that between 40 per 10,000 children have autistic disorder and 67 per 10,000 do if all types of autism-like behavior are included.

The investigators suggest a variety of factors may have contributed to the increase in autism rates. For one, the symptoms used to diagnose the illness have been expanded over the years and there is heightened public awareness of the condition. This is "due in large part to efforts of parent and advocacy groups, availability of more medical and educational resources, increased media coverage of affected children and families, and more training and information for physicians, psychologists and other service providers," they write.

And in 1991, the US Department of Education included autism as category for special education services, which may have also increased diagnoses.

"The combined influence of these factors has probably contributed to the identification of more individuals with autism," the authors write. "However, it remains unclear whether specific environmental, immunologic, genetic or unidentified factors also may have contributed to these higher reported prevalence rates," they conclude.

In other findings, the researchers found that the prevalence for autism appeared to be similar for black and white children and about 40% of children with autism were identified through the school system.

The study is the first to get a good population-based estimate of autism in black children, showing that the rates are similar to those found in other youngsters, noted Fombonne in an editorial accompanying the study.

Fombonne, of the McGill University and Montreal Children's Hospital in Canada, notes that the CDC has "recently funded a surveillance network across several states."

"This and other initiatives should help address more directly hypotheses about...changes in the incidence of autism spectrum disorders," he writes.

SOURCE: Journal of the American Medical Association 2003;289:49-55.