August 22, 2022
A new electronic health records (EHR) algorithm, which can differentiate attention deficit hyperactivity disorder (ADHD) from comorbid conditions in children, may lead to more accurate diagnoses and treatment interventions. According to findings published in the journal Neurodevelopmental Disorders, more than half of children with ADHD have one comorbidity or more. This has confounded clinicians who’ve had trouble discerning whether a patient’s symptoms were caused by ADHD or comorbidity.1
Given the prevalence of ADHD comorbidities, researchers from the Center for Applied Genomics (CAG) at Children’s Hospital of Philadelphia (CHOP) developed a multi-source EHR rule-based algorithm with natural language processing (NLP) test mining to provide a comprehensive view of a patient’s medical record. Using electronic health records and data from CHOP, and data from the CAG between 2009 and 2016, the research team performed a retrospective case-control study on a total of 51,293 patients aged eight and older. Of those, 5,840 were diagnosed with ADHD; among those cases, 46.1% had ADHD alone, and 53.9% had ADHD along with at least one comorbidity.
The algorithm had a positive predictive value of 95% for ADHD and 93% for controls, and it had a positive predictive value ranging from 60% to 100% for comorbid conditions. The higher number of patients with comorbidities, such as anxiety (27.1% of ADHD cases) and autism spectrum disorder (15.1% of cases), yielded more accurate results. Other comorbidities observed in the cohort included learning disorders (11.8%), conduct disorder (10.1%), and oppositional defiant disorder (9.1%).1
ADHD keywords did not significantly help distinguish patients. However, ADHD-specific medications on EHRs did — it increased identified cases by 21%.
Although the algorithm is in the early stages of development, researchers recommend implementing it in genomics and discovery-based studies. “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,” said Hakon Hakonarson, M.D., Ph.D., director of the Center for Applied Genomics at CHOP and senior author of the study. “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 interventions.”
1Slaby, I., Hain, H. S., Abrams, D., Mentch, F. D., Glessner, J. T., Sleiman, P., & Hakonarson, H. (2022). An electronic health record (EHR) phenotype algorithm to identify patients with attention deficit hyperactivity disorders (ADHD) and psychiatric comorbidities. Journal of neurodevelopmental disorders, 14(1), 37.