The devil is in the headlines

The latest data from the Centers for Disease Control and Prevention (CDC) show that the rate of autism among U.S. children is increasing. The autism prevalence estimate was about 1 in 150 children between 2000 and 2002, but that figure rose to 1 in 88 children in 2008 an increase of 78 percent according to the CDC s most recent report.

The estimates come from the Autism and Developmental Disabilities Monitoring Network, where researchers from the CDC have counted, every two years since 2000, the number of 8-year-olds with autism in a dozen communities across the nation. Boys are still the most prone to developing autism spectrum disorders, outnumbering girls 5-to-1; currently, 1 in every 54 boys has the disorder.

CDC Director Dr. Thomas Frieden, however, attributes the latest increase to better and broader diagnoses. Doctors are getting better at diagnosing autism; communities are getting much better at [providing] services to children with autism, and CDC scientists are getting much better at tracking which kids in the communities we re studying have autism, he says. How much of that increase is a result of better tracking and how much of it is a result of an actual increase, we still don t know.

At least what we do know is that this increase has nothing to do with vaccines, says ACSH's Dr. Ruth Kava, noting that some parents are still concerned that vaccines are associated with autism, a fear generated by the infamous but entirely inaccurate 1998 study by Andrew Wakefield. Though the study has since been retracted and Wakefield lost his medical license, a fear of vaccines persists. But as Dr. Kava explains, the real source of the changing autism numbers is due to broader surveillance and broader definitions. "We re more aware of autism now than in the past, she says, which may explain why the incidence is increasing.

ACSH s Dr. Josh Bloom notes that headlines declaring that autism has increased by 78 percent can be wildly misleading. For instance, he says, When the PSA test was introduced, there was about an eight-fold increase in prostate cancer over the next decade. Were more people actually getting cancer? No. It was just being detected by a new technique. The reported increase in autism cases has to be due to a similar effect. Does anyone really believe there were actually 78 percent more cases of autism in the last decade?