boy with worried stressed face expression looking down with his hands pressed to either side of his face and with chaotic brain waves streaming out of his head to represent ADHD
Credit: SIphotography/Getty Images

Researchers at the Children’s Hospital of Philadelphia’s (CHOP) Center for Applied Genomics (CAG) have developed an algorithm using existing electronic health records (EHRs) is able to distinguish patients with attention deficit/hyperactivity disorder (ADHD) alone from other related conditions. The researchers hope the new tool can be applied in broader clinical settings to help eliminate the long diagnostic odyssey faced by patients with the condition and direct them to more personalized treatments.

“Our goal with this algorithm was to establish a tool that could be used to automate future genetic analyses and improve diagnostic yield and precision in future studies,” said Hakon Hakonarson, MD, PhD, director of the Center for Applied Genomics at CHOP and senior author of the study in a press release.

ADHD is a complex condition and one that can be difficult to definitively diagnose because it presents as three different types and the symptoms present on a spectrum. Also complicating diagnoses are comorbidities that affect about half of ADHD sufferers. These include learning, sleep and anxiety disorders. In all, the condition affects from 5% to 8% of school-age children and 2% to 4% of adults.

To multi-source algorithm developed by the team provides a more detailed view of data contained in each patient’s EHR. The team used its own data as well as data collected from 2009 to 2016 by the CHOP CAG. The retrospective case-control study included data from 51,293 patients which included 5,840 patients previously diagnosed with ADHD. Those with ADHD alone constituted 46.1% of that number while the remaining 53.9% had ADHD with comorbidities.

Testing the algorithm against these cases, the CHOP team found, had a positive predictive value of 95% for ADHD and 93% for controls, which, the researchers noted makes it accurate enough for use in future retrospective studies. Key to identifying ADHD patients from the medical record were the ADHD-specific medications noted rather than ADHD keywords.

In addition, the algorithm also had a positive predictive range of between 60% and 100% for other psychiatric conditions, with the higher number of patients with those comorbidities, such as anxiety and autism spectrum disorder, yielding more accurate results.

“With the high positive predictive values achieved by this algorithm, we believe we have developed a robust and useful tool for identifying appropriate datasets and successfully distinguishing between groups of patients,” Hakonarson noted. “It’s possible that these groups with or without comorbidities may respond differently to medication, which could help us design better and more effective methods for therapeutic intervention.”

CHOP operates a dedicated unit for ADHD patients for both treatment and research of the condition called the Center for Management of ADHD, which provides a team approach to the diagnosis and treatment of children 4 to 18. The goal of the center’s evaluation of patients “are to develop an understanding of your child’s strengths and challenges and to determine their primary treatment needs. This evaluation will include a comprehensive assessment of attention deficit/hyperactivity disorder (ADHD) as well as a screening for other emotional, behavioral and medical conditions,” it noted on its website.

The center also conducts ADHD research, with current areas of focus that include assessing ADHD in children and teens; developing and evaluating family and school interventions for children with ADHD or who are at risk of developing the condition; developing strategies to manage ADHD in primary care practices; factors contributing to driving risks for children with ADHD; and understanding the neurobiology of ADHD.

Also of Interest