The One Percent Convergence
Today, the U.S. government updated its official estimate of autism prevalence to 1 out of 100 children, up from the 2007 estimate of 1 in 150. The figure is based both on the latest data collected by CDC’s intensive ADDM monitoring network and on a large, but less rigorous telephone survey of more than 70,000 parents conducted by the Health Resources and Services Administration (HRSA). The HRSA survey actually found a somewhat higher rate of 1 in 91, but phone surveys are bound to be less reliable than studies like the CDC’s, in which a considerable effort is made to confirm each diagnosis.
What’s striking is that so many different epidemiological surveys now converge on this one figure of 1 percent. This includes studies in England, Japan, Sweden and Canada. Most interesting is that a recent British study of autism in adults also came up with the 1 percent figure.
“It’s reassuring to scientists that no matter how you shake the bushes, you come up with this 1 percent,” Roy Richard Grinker of George Washington University told me this morning. “It provides what scientists call convergent validity.” Grinker himself is about to publish the first study of autism prevalence in South Korea and he says his findings are in line with these other studies.
It was interesting to see federal health officials scramble to manage the message conveyed by the new studies. They clearly did not want to raise new fears about an autism “epidemic.” In fact, the E word was completely avoided in an embargoed press conference held on Friday. Tom Insel of NIH, Ileana Arias of the CDC and Peter Van Dyke of HRSA made an obvious effort to say that scientists don’t know if the new numbers represent a genuine increase in autism or better detection, while also going out of their way to convey a sense of urgency about government efforts to address autism.
More details about the new studies can be found in my story for Time.com.