Artificial intelligence can help clinicians spot children at risk for attention-deficit/hyperactivity disorder (ADHD) much earlier, a new study suggests. By analyzing data collected during routine care, AI can identify at-risk children years before a diagnosis is typically made.
A Duke Health study found that AI models analyzing electronic health records can estimate ADHD risk in children before a formal diagnosis, suggesting that earlier intervention is possible with no additional screening.
AI tools can analyze routine pediatric health records to flag children at higher risk for Attention-deficit/hyperactivity disorder, helping clinicians prioritize earlier screening, referral, and follow-up care.
Turning Routine Data into Early Insight
ADHD affects millions of children, yet diagnosis is often delayed even when early signs are present. That gap can limit access to interventions that improve long-term academic, social, and health outcomes.
Researchers at Duke University School of Medicine examined whether patterns in routine medical data could help identify risk earlier in a child’s development.
“We have this incredibly rich source of information sitting in electronic health records,” said Elliot Hill, lead author of the study and data scientist in the Department of Biostatistics & Bioinformatics at Duke University School of Medicine. “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.”
Matthew Engelhard, M.D., Ph.D. (left) and Elliot Hill (right) Credit: Duke Health
How the Model Works
The research team analyzed electronic health records from more than 140,000 children with and without ADHD, training an AI model on data from birth through early childhood.
The model learned to recognize combinations of developmental, behavioral, and clinical patterns that often precede diagnosis. It demonstrated strong accuracy in estimating risk for children aged 5 and older, with consistent performance across sex, race, ethnicity, and insurance status.
For nurses and primary care teams, the approach reflects a growing use of existing data to support earlier clinical awareness without adding new workflows.
Supporting Clinical Judgment, Not Replacing It
The tool is designed to support decision-making, not to diagnose ADHD. Instead, it highlights children who may benefit from closer monitoring or earlier referral for evaluation.
“This is not an AI doctor,” said Matthew Engelhard, M.D., Ph.D., senior author of the study and faculty member in Duke’s Department of Biostatistics & Bioinformatics. “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.”
That distinction is critical in practice, where nurses are often the first to notice developmental concerns and play a central role in coordinating follow-up care.
Why Earlier Action Matters
Earlier identification can lead to earlier evaluation and intervention, which is associated with better outcomes for children with ADHD.
“Children with ADHD can really struggle when their needs aren’t understood, and adequate supports are not in place,” said Naomi Davis, Ph.D., associate professor in the Department of Psychiatry and Behavioral Sciences. “Connecting families with timely, evidence-based interventions is essential for helping them achieve their goals and laying a foundation for future success.”
Earlier insight may also help care teams engage families sooner and tailor support before challenges escalate in school or at home.
What Comes Next
While the findings are promising, researchers say additional study is needed before the tool can be implemented in clinical settings. Ongoing work is exploring how similar AI models may help predict mental health risks in children and adolescents.
