A large clinical trial of a heralded early intervention therapy for autism shows the treatment might not be as advantageous as previously believed.

Young children with autism receiving the therapy known as Early Start Denver Model, or ESDM, did show larger gains in language skills compared to children given other behavior therapies, according to the study published recently in the Journal of the American Academy of Child and Adolescent Psychiatry.

But all of the children in the study showed improvements in language, cognitive and social skills regardless of the type of treatment, signaling a benefit to early intervention generally rather than an overall benefit to any particular method, the researchers said.

“There was statistically little difference between conventional and ESDM therapies for the children,” said Richard O. Williams, a Colorado geneticist who studies autism, but was not involved in the study. “The take home here is that most any intervention methodology will have some benefit. Parents should try to identify programs that attempt to address their concerns, location to services and budget.”

ESDM is based on the theory of applied behavior analysis. Children as young as 12 months are encouraged by therapists and parents to develop their language and social skills through play. The method can be offered in a clinic but is more commonly woven into the child’s natural routines in the home or day care.

The study involved 118 children with autism ages 14 months to 2 years old at three sites across the U.S. The children were followed through their experiences either in ESDM therapy or other clinic-based interventions for more than two years. Those in the ESDM group received about 15 hours a week of play therapy in their homes or day-care centers, while those in the clinic group received varying lengths of therapy depending on the study site.

Researchers found that children in the ESDM group made significantly greater improvements in language skills compared to children in the clinic or community interventions when averaged across the three sites.

“We were very pleased with the results of the study,” said Geraldine Dawson, a professor of psychiatry and director of the Duke Center for Autism and Brain Development. Dawson helped develop ESDM in the 1980s and is a co-author of the study.

Although there were language advantages for the ESDM cohort, children in the study’s control group who received community-based therapies also showed significant and similar gains in their cognitive and social skills, Dawson said.

Most of the children in the study increased their IQ scores by an average of about 17 points over the course of the research.

“We think that’s good news,” Dawson said. “Children in the community are now receiving more intensive interventions.”

Some critics of the ESDM research said it did not fully replicate the success of earlier published findings. The significant language gains in the ESDM group were seen in only two out of three of the study sites, prompting further questions.

But the study’s authors and other advocates said the research reveals more about the overall progress in early interventions.

“It is key to note that at all three sites, the control group showed significant improvement. This may mean that community treatment approaches are getting better,” said Thomas Frazier, chief science officer for Autism Speaks, which provided some grant funding for the research.

The Association for Science in Autism Treatment categorizes ESDM under “needs more research” for effectiveness. Frazier agrees that future studies are needed to determine the best treatments for each child. For now, researchers agree that early intervention should continue to be recommended in every case.

“I don’t think that ESDM is the only model,” Dawson said. “The most important message is to fight for access for every child to good early intervention services.”

Read more stories like this one. Sign up for Disability Scoop's free email newsletter to get the latest developmental disability news sent straight to your inbox.