Attention deficit hyperactivity disorder, or ADHD, affects millions of children, but many go years without a diagnosis, missing the chance for early support.

Now, a new study from Duke Health, published April 27 in the journal Nature Mental Health, suggests artificial intelligence could help change that.

"We have this incredibly rich source of information sitting in electronic health records," said lead author Elliot Hill, a data scientist at Duke University School of Medicine in Durham, North Carolina. "The idea was to see whether patterns hidden in that data could help us predict which children might later be diagnosed with ADHD, well before that diagnosis usually happens."

Hill and his colleagues created an AI model that estimated a child's risk of developing ADHD — years before a typical diagnosis.

The AI model reviewed medical data from more than 140,000 children, with and without ADHD, from birth through early childhood. It learned to spot combinations of developmental, behavioral and clinical signs that often appear long before diagnosis.

The model proved highly accurate at estimating future ADHD in children age 5 and older. Results were consistent across sex, race, ethnicity and insurance status.

Researchers say the tool can flag kids for earlier evaluation, diagnosis and support. This is important because early intervention is linked to better academic, social and health outcomes.

"This is not an AI doctor," said senior author Dr. Matthew Engelhard, also from Duke University School of Medicine. "It's a tool to help clinicians focus their time and resources, so kids who need help don't fall through the cracks or wait years for answers."