Before autism strikes

CAN autism be prevented? It sounds like a very long shot. But that is the focus of innovative research at the University of Washington that will use behavioral techniques with infants genetically at risk for the condition to try to stave off the symptoms of this baffling neurological disorder.

If the approach proves beneficial, it could save thousands of children from a life of social isolation and permanent disability.

"This is a very exciting and potentially revolutionary study because it is the first to focus on infants," says Alice Kau, an autism expert at the National Institute of Child Health and Human Development in Bethesda, Md. "Diagnosis and intervention at such a young age could prevent the development of full-blown autism."

Once an exceedingly rare diagnosis, autism is thought now to afflict 1 in 150 newborns, more than Down's syndrome or childhood cancers, and the steep rise in these numbers has lent urgency to the search for better ways to reverse or even halt the development of the disorder. Experts don't know what causes autism, but they do know that it runs in families. When infants have an autistic older brother or sister, the odds that they will develop the disorder are five to 10 times higher than in the general population.

The federally funded $11-million study, which enrolled its first patients last week and is expected to last four years, will eventually involve 200 Seattle-area infants 6 months or younger upon enrollment, each of whom has an older sibling diagnosed with autism.

Half of the babies will be monitored by specialists and referred for community treatment. Mothers and infants in the other group will participate once a week for 12 weeks in social interaction workshops that promote the formation of emotional bonds. All the children will be evaluated at 12 months.

Next, those infants in the treatment group will undergo an intensive intervention program called applied behavior analysis (see box), and parents will be taught how to encourage play and communication. At 24 months, the children will be assessed again.

Although the researchers cannot know which of the 200 babies would have gone on to receive a diagnosis of autism, they can see whether the number was reduced from the expected autism rate in children with autistic siblings, which is 1 in 20.