"(Autism) is a devastating diagnosis to most families," he added. "Then there is a big question mark above what happens next."
Autism is thought to affect one in 68 children in the United States and one in 100 children in the UK, with half estimated to have intellectual disabilities and about 10% never learning to speak, according to the researchers. It is a spectrum disorder, with children experiencing a range of symptoms and severity. The causes are not fully understood, and there is currently no pharmaceutical treatment or cure.
Previous studies and therapies
have involved having children work directly with therapists to improve their communication skills and reduce symptoms, such as repetition. By contrast, the new trial "is a parent-mediated intervention," said Green, who believes this enables a 24/7 therapeutic space for children with autism in their own homes.
During the trial, children with core symptoms of autism were divided into two groups: one in which children and parents received standard treatment and one in which parents worked with therapists to better understand how they can communicate with their children effectively, as well as pick up on their child's cues for interaction.
The immediate outcome of this trial was reported in a previous study in 2010
, soon after it ended. It found that parents showed improvement in their ability to interact with their children, improvement in children communicating back with their parents and reduced severity of symptoms among children whose parents received this training.
"When we change the parental interactive behavior ... that leads to child initiation with the parent, and that change in child initiation with the parent is related to the child changing symptoms," Green said.
The new study analyzed the long-term impact on children and their parents more than six years later. Although the parents' synchrony -- how much they were attuned to and responsive with their child's communication -- had fallen to equal that of the control group, the researchers saw sustained improvement in children's communication with their parents as well as a greater reduction in the severity of symptoms among children whose parents had received training.
"That difference had persisted," said Andrew Pickles, professor of biostatistics and psychological methods
at Kings College London, who co-led the research. "Getting in early when there is more chance of general change."
The researchers described three severity levels associated with autism: mild, moderate and severe. Severe means the child is unaware of people around them, may move their arms vigorously and ignores their parents when they call their name. In contrast, moderate symptoms would involve a child seeming younger than their age and making requests in a loud voice, while mild symptoms may be undetectable.
More than six years after the trial, 46% of the children whose parents had received training were considered to have severe symptoms, but 63% of children whose parents had received standard care were considered to have severe symptoms. At the start of the trial, the two groups has been almost equal.
The team stresses that it's a potential treatment, not a cure.
"(This) clearly doesn't cure autism," Green said, but he pointed out that it does have a "sustained decrease in severity."
Some symptoms did not differ between the two groups, including measures of child anxiety, challenging behaviors and language scores.
Learning to communicate
The original trial involved 152 children with autism, between 2 and 4 years old, and their parents. During the followup, 121 children remained, 59 of whom received the Pre-school Autism Communication Trial intervention.
As part of the therapy, parents had two sessions with a therapist each month for six months, combined with homework to interact with their child for 30 minutes per day. For the next six months, the sessions were reduced to monthly, but the homework remained.
During the therapy sessions, parents watched videos of themselves interacting with their child to learn how they could improve and learn to spot when, and how, their child is interacting with them. The aim was "to help adults self-learn and become self-aware," Green said.
The video feedback was also used to show parents what actions get better responses from their child, "to make the parent more confident in how they respond back," said Catherine Aldred, a consultant speech and language therapist with Stockport NHS Trust and the University of Manchester in the UK, who led this part of the trial. The aim was also to "draw the child into becoming an active communicator," she said.
Having seen such promising results, the team now hopes to take this approach into schools in hope of running parallel sessions, alongside the training of parents, to get "parents and schools talking to each other better," Green said. This would also "get the child functioning in their everyday world."
They also hope to further investigate why this intervention had the effect it did. "We want to look at the mechanism for this sustained effect," Green said. "We could estimate that if it worked with a severe group, it might also work with a mild group."
But the researchers highlighted that the intervention was done soon after diagnosis, between the ages of 2 and 5 years, and that their intervention may not be effective in older children. "Early after diagnosis, when families are ready, that's the time to start," said Green.
"I can see why these researchers are excited," said Dorothy Bishop, professor of developmental neuropsychology
at the University of Oxford, who was not involved in the trial. "The initial results in their previous report were a bit ambiguous, but these results at follow-up are pretty consistent in showing the benefit of this early intervention for autism across a range of measures. ... For parents of children with autism, even a modest reduction would be worthwhile."
"It is a remarkably positive story, because the intervention itself was neither intensive nor invasive," said Uta Frith, emeritus professor of cognitive development at University College London
, who also highlights that more work needs to be done. "There were improvements on a range of measures, but the effects were not dramatic, and they were very variable across the group. We are still a long way from individualized programs that might produce lasting behavioral improvements in all autistic children."