Rethinking the need for ADOS testing to diagnose autism in young children

The Autism Diagnostic Observation Schedule, or ADOS testing, was developed in the 1980s as a tool for autism research. Through a series of semi-structured observations, trained evaluators assess children’s communication skills, social interaction, and imaginative use of materials.

But over time, the ADOS has come to be considered the gold standard for a clinical diagnosis of autism spectrum disorder (ASD). And that has unintentionally delayed care for many children. The time-consuming test adds cost to the diagnostic process, and there is a shortage of people trained to administer it.

“The ADOS was never designed to be used in the clinic,” says William Barbaresi, MD, chief of the Division of Developmental Medicine at Boston Children’s Hospital. “Young children can wait months or even years for an assessment to diagnose ASD. That makes it difficult for them to access intensive early intervention services when they are most effective, ideally starting at around 24 months of age. If the ADOS is required for diagnosis, it becomes an additional barrier due to the time and training required.”

Now, a multicenter study led by Boston Children’s through the national Developmental and Behavioral Pediatrics Research Network (DBPNet) is pushing back. It finds that trained developmental-behavioral pediatricians can almost always diagnose ASD in young children without ADOS testing. Barbaresi, the study’s principal investigator, hopes it will convince intervention agencies, schools, and insurers to drop their requirement for the ADOS when a diagnosis is made by a trained expert.

Putting ADOS testing to the test

The study, published last month in JAMA Pediatrics, was sponsored by DBPNet and funded by the U.S. Maternal and Child Health Bureau. It involved 349 children 18 months to 5 years old seen at one of nine academic pediatric centers.

Developmental-behavioral pediatricians (DBPs) first diagnosed the children based on their clinical assessment. A specially trained clinician then administered the ADOS. Results were shared with the DBP, who then could revise their diagnosis.

The ADOS was never designed to be used in the clinic. If the ADOS is required for diagnosis, it becomes an additional barrier due to the time and training required.

In 90 percent of cases, the diagnosis, including the ADOS, was consistent with the original clinical diagnosis. Consistency was greatest when the clinician felt highly certain of their original diagnosis.

“Overall, this study is good news,” says Barbaresi. “We believe it has the potential to change practice by reducing wait times for diagnostic evaluations so children can receive early, intensive treatment for ASD.”

Removing a barrier to early intervention

The investigators hope their findings will spur a national effort to persuade insurers and education agencies to change the requirements for ASD assessments. In the meantime, pediatric practices should feel comfortable relying on DBPs’ clinical judgment to refer young children for autism interventions unless the ADOS is required for insurance or other reasons, Barbaresi says.

The study’s other participating DBPNet centers were the Children’s Hospital of Philadelphia, Children’s Hospital Colorado, University of Arkansas for Medical Sciences, University of California-Davis, Children’s Hospital Los Angeles, Hospital of St. John of God (Linz, Austria), Rainbow Babies and Children’s Hospital (Cleveland, Ohio), and the Children’s Hospital at Montefiore (Bronx, NY).

Learn more about the Developmental Research Program and Laboratories of Cognitive Neuroscience at Boston Children’s.