Add To Favorites

Treatments for Autism Provide Mixed Results

By: Benedict Carey, The New York Times

Desperate parents of autistic children have tried almost everything -- hormone injections, exotic diets, faith healing -- in the hope of finding a cure.

But more than 60 years after it was first identified, autism remains mystifying and stubbornly difficult to treat. About the only thing parents, doctors and policy-makers agree on is that the best chance for autistic children to develop social and language skills is to enroll them in some type of intensive behavioral therapy.

A government-appointed panel has endorsed such therapies, which can cost from $40,000 to more than $60,000 per year. Parents fight to get their children placed in behavioral programs, encouraged by the claims of some therapists that they can produce astonishing improvement in up to 50 percent of cases. An estimated 141,000 children with autism receive special education services, in many cases including behavioral therapies, through public schools.

Yet the science behind behavioral treatments is modest at best. Researchers have published very few rigorously controlled studies of the therapies, and the results of those studies have been mixed. While some children thrive, even joining regular classrooms, the studies have found that most show moderate or little improvement.

And researchers say most parents now experiment with so many alternative treatments -- including vitamins, diets, sensory therapies, computer games and various combinations -- that they muddy the results of behavior treatment, making it hard to say what is causing a child to gain skills, or to decline.

The most recent analysis of treatment research, financed by the National Institutes of Health and scheduled to be published next year, concludes that although behavior treatments benefit many children, there is no evidence that any particular treatment leads to recovery. Doctors do not yet know how to predict which children will improve in the treatments, or even how treatable the condition is, the report concludes.

"If so many kids are being cured, then where are they? Who are they? Show me 10 percent," said Dr. Bryna Siegel, director of the autism clinic at the University of California, San Francisco. "The reason practitioners can't show you all these kids is because there simply aren't that many of them out there."

Scientists also have not answered perhaps the most fundamental question: Why does one child thrive in treatment while another, equally affected, does not?

Some researchers think the response could have to do with structural properties of the brain, which would show up on brain scans. Others suspect that children who do not do well in treatment have a subtle language-processing problem in addition to autism.

Each of these ideas is under investigation. "I expect with more research we may find that there are two groups of children: one group that does well in directed teaching and another that needs a biomedical treatment," said Dr. Geraldine Dawson, director of the autism program at the University of Washington in Seattle. "And we would be able to tell them apart."