Just 1 in 4 children with autism is diagnosed before age 3, but a new eye-tracking technology may allow for earlier diagnosis and intervention, according to three clinical studies of more than 1,500 kids.

Autism is a disorder marked by difficulties with communication and social interaction. In the United States, it affects about 1 child in 36, according to the U.S. Centers for Disease Control and Prevention. Treatment includes behavioral, educational and family therapies and is most effective when started early and tailored to the individual child.

The new eye-tracking technology provides automated measures of children’s looking behavior and can help spot signs of autism as early as 16 months of age, researchers said. It may also help predict kids' strengths and vulnerabilities.

“Objective measurements can help speed the time to diagnosis and speed the start of individualized treatment plans for newly diagnosed children at a younger age, which has been shown to lead to better outcomes for children with autism,” said study author Warren Jones, director of research at the Marcus Autism Center and a chair in autism at Emory University School of Medicine, both in Atlanta.

“Our hope is that this tool, which is now U.S. Food and Drug Administration-authorized for use in children between 16 and 30 months of age, can help alleviate this enormous public health challenge with earlier diagnoses and treatment," he said.

Autism spectrum disorder, or ASD, is a developmental disability caused by differences in the brain, according to the CDC. People with ASD often have problems with social interaction and different ways of learning, moving or paying attention.

No single test can diagnose autism. Instead, doctors look at the child's developmental history and behavior to make a diagnosis.

"[The new tool] will be used to supplement informed and experienced clinical judgment,” Jones said.

The EarliPoint Evaluation is currently available at Marcus Autism Center and select autism centers in the United States. Developers hope it will be available elsewhere soon.

Children’s Healthcare of Atlanta, of which the Marcus Autism Center is a subsidiary, and the authors have a financial interest in the company.

The eye-tracking tool measures a child’s eye movements while they watch a 10-minute video. Hundreds of important social cues are presented during the video, and the technology captures around 120 measurements per second.

Researchers compared measurements from children with suspected autism to those of typically developing peers.

"Typically developing children pay attention to these cues, adjusting their looking on a moment-by-moment basis," Jones said. “The tool quantifies the number, degree and timing of any missed cues, which are essentially missed opportunities for social learning."

These differences emerge early in infancy, past research has shown.

The new technology can also help determine autism severity, said study author Ami Klin, director of the Marcus Autism Center and division chief of autism and related disorders at Emory University School of Medicine.

"Because looking behavior is foundational to the way children learn how to speak and acquire language, and the way they learn how to solve things in their world, looking behavior … also allowed us to show how severe is their autism,” Klin said.

The more children deviate from the way typically developing kids look at things in the world, the more severe their autism, he said.

“Additional clinical trials are underway to develop indications for younger infants and for older children and to also measure change so that we can see if children are responding to treatment -- and how,” Klin said.

In the Sept. 5 issue of JAMA Network Open, researchers shared initial results from more than 1,080 kids that led to development of the new tool.

The tool was further tested in 475 children aged 16 to 30 months who were evaluated at six U.S. autism specialty clinics. Results from this study appeared Sept. 5 in the Journal of the American Medical Association.

The eye-tracking technology predicted expert diagnoses of autism with high specificity and sensitivity, researchers reported. (Tests with high specificity correctly identify kids without a condition, while a test with a high sensitivity can correctly identify kids who do have the condition.)

The test was even more sensitive when cases in which the clinician was uncertain about the diagnosis were removed from the analysis, the study showed.

Geraldine Dawson, director of the Duke Center for Autism and Brain Development in Durham, N.C., wrote an editorial that accompanied the findings.

“These studies suggest that an eye-tracking test could help an autism specialist be more confident in making an autism diagnosis,” she said. “The eye-tracking test is not an autism screening tool, but rather is designed to provide additional, objective information when an autism specialist is performing a diagnostic assessment.”

As it stands, doctors diagnose autism using subjective clinical judgment.

“Having additional objective information could be useful in making a more confident diagnosis,” Dawson added.