Most families of children with autism can face long wait times for a diagnosis, and once it is obtained, it sometimes may not be definitive.
Now, two studies published Tuesday suggest that a newly developed eye-tracking tool could help doctors diagnose autism in minors as young as 16 months — and with greater certainty.
“This is not a tool that will replace expert doctors,” says Warren Jones, director of research at the Marcus Autism Center at Children’s Healthcare of Atlanta and Nien Chair in Autism at Emory University School of Medicine, one of the authors of both studies.
Rather, he said, the hope with this eye-tracking technology is that “by providing objective measurements that measure the same thing in every child,” it can help inform the diagnostic process.
The tool, called the EarliPoint Assessment, is cleared by the U.S. Food and Drug Administration (FDA) to help clinicians diagnose and evaluate autism, according to the researchers.
Traditionally, children are diagnosed with autism based on clinical assessment of their developmental history, behavior and parental reports. Assessments can take hours, and some subtle behaviors associated with autism can go unnoticed, especially among younger children.
“Typically, the way we diagnose autism is by assessing our impressions,” says Whitney Guthrie, a clinical psychologist and scientist at the Autism Research Center at the Children’s Hospital of Philadelphia. She was not involved in the new studies, but her research focuses on early diagnosis of autism.
It will always be important to take into account what the family and specialist observe in the child, but families may reflect a child’s behaviors differently based on his or her unique background or experiences, Guthrie said.
“Some of these new technologies are more objective. They don’t rely on humans, who are imperfect by nature,” he says. “They are also potentially more scalable and viable.”
How eye tracking can help
The technology works by tracking minors’ eye movements as they watch videos of other children’s social interactions.
A minor without autism may focus attention on the hand gestures of a child in the video who is pointing to something for another child in the video to look at. That child may also notice another child’s face that looks sad or smiling. But a child with autism might not pay attention at all to hand gestures or facial expressions, and this can be identified in his or her eye movements.
“We use these types of scenes to test whether or not children are paying attention to information that we expect typically developing children of the same age to pay attention to,” Jones explains.
In one of the new studies, published in the medical journal JAMA, 475 children between 16 and 30 months of age were assessed for autism at six specialty clinics in the United States. The children were enrolled in the study between April 2018 and May 2019, and the eye-tracking tool was included in the assessments.
The researchers found that, relative to only expert clinical diagnosis, the measures of the children’s eye tracking and social interaction with the videos had a sensitivity of 71%, meaning they helped accurately diagnose autism 71% of the time, and a specificity of 80.7%, meaning they helped designate a child without autism about 80% of the time.
Among the children, 335 had a diagnosis of autism that their physicians were “certain” of without using the eye-tracking tool. When the eye-tracking tool was used for this group alone, the researchers found that it showed a sensitivity of 78% and a specificity of 85.4% when expert clinicians were certain of their diagnoses.
The results suggest that using measurements of how minors watch and learn from social interactions in videos may offer a biomarker for diagnosing autism, helping to shorten the time needed to make a diagnosis and begin an intervention.
The other study, published in JAMA Network Open, found that among 719 children aged 16 to 30 months, eye-tracking-based measurements had a sensitivity of 81.9% and a specificity of 89.9%. In another group of 370 children aged 16 to 45 months, eye-tracking-based measurements had a sensitivity of 80.6% and a specificity of 82.3%.
In that study, the children were presented with 14 video scenes, each about 54 seconds long on average.
Overall, the data show that the technology has a sensitivity and specificity of 80%.
In the studies, the eye-tracking measurements, which collected data 120 times per second and within 12 minutes of video viewing, predicted expert physician assessments “with a high degree of accuracy,” according to a press release, and were consistent with expert clinical reference diagnoses.
“To me, there is no other return on investment in what we do in science and public health than giving families access to early diagnosis that makes possible the early intervention and support that are going to optimize children’s outcomes for the rest of their lives,” said Ami Klin, director of the Marcus Autism Center and division chief of Autism and Developmental Disabilities at Emory University School of Medicine, who authored both studies.
“In our healthcare system, those families come up with the concerns they have about their children, and they go to bed every night for months, if not years, waiting for an answer, knowing full well that the sooner they get that answer, the sooner their child will benefit from the kinds of interventions that can improve his or her life,” Klin said. “For the first time, we have an objective biomarker-based tool that can increase access. And my greatest hope is that this will be one of the tools we can use to decrease health care disparities that affect minority, low-income and rural families.”
“The equity issue is very significant.”
Medical experts have long known that autism screening, and subsequent diagnosis, has been “really, really suboptimal” for minors of color, as well as for those living in rural and underserved communities, including urban areas, said Dr. Kristin Sohl, a professor of pediatrics at the University of Missouri School of Medicine and chair of the American Academy of Pediatrics’ autism subcommittee.
Minors from these groups are diagnosed with autism later than those from other minority or rural areas.
“The issue of equity in autism is very significant,” Sohl says.
“When a young child shows characteristics of autism at a very young age, but is not even remotely identifiable, meaning no one has thought about it until he or she is in day care, there is a huge disadvantage for that child,” says Sohl, who was not involved in the new research.
Diagnostic tools such as eye-tracking technology and other clinical resources can help reduce those disparities, and improve early diagnosis rates overall, by allowing specialists to increase the capacity of their practices and helping to expand the pool of physicians who feel prepared to diagnose autism.
“There are many, many children who are very well served by physicians in their community. And yet we need to help our primary care physicians have a much better appreciation for autism,” Sohl says. “The age of diagnosis is often a reflection of how long parents have waited.”
Studies by the U.S. Centers for Disease Control and Prevention (CDC) have found that early detection of autism in children has improved, but disparities remain in practices for identifying and diagnosing the condition.
“Devices like this can be very useful in the hands of a qualified general practitioner, because they can take that information and use it to look at the other data they have and say, ‘Yes, I’m sure this is autism,’ or ‘No, this is not, and this is what I need to do,'” Sohl says.
However, Geraldine Dawson, of the Duke Center for Autism and Brain Development in North Carolina, wrote in an opinion published Tuesday alongside the new studies that more work needs to be done before eye-tracking tests become commonplace.
“There is work to be done before an eye-tracking test is used in clinical practice. Demonstrating that an eye-tracking test improves diagnostic certainty would require longitudinal follow-up of children whose diagnosis was uncertain to determine whether the test improves prediction of a later definitive diagnosis of autism,” he wrote.
“The intended use of the eye-tracking test is to help clinicians make a diagnosis of autism in young children who have been referred to a specialty clinic for evaluation,” Dawson wrote. “By integrating multiple sources of information, including the eye-tracking test, parent report, and clinical observations, the accuracy, certainty, and efficiency of the diagnostic evaluation of autism could potentially be improved, resulting in fewer missed cases and allowing more children to receive empirically validated early therapies from which they could benefit.”