In more than half of pediatric instances, comorbid conditions coexisted with ADHD, and it was difficult to determine if the symptoms impairing function were caused by the comorbid condition or the underlying ADHD. In addition, comorbid illnesses raise the risk of a more severe and protracted course and make treatment choices more difficult. To enhance research on ADHD using biorepository and other electronic record data, it was crucial to creating an algorithm that recognized ADHD and comorbidities. Using an electronic system to account for various mental diseases, it was possible to discriminate between ADHD alone and ADHD with comorbidities accurately. For effective future searches in sizable biorepositories, researchers intended to create an EHR phenotypic algorithm to more effectively distinguish cases of ADHD in isolation from cases of ADHD with comorbidities. Utilizing the Center for Applied Genomics (CAG) biobank at Children’s Hospital of Philadelphia (CHOP), they created a multi-source algorithm that provided a more comprehensive view of the patient’s EHR. Investigators mined EHRs from 2009 to 2016 using International Statistical Classification of Diseases and Related Health Problems (ICD) codes, medication history, keywords unique to ADHD, and comorbid mental illnesses to aid genotype-phenotype correlation efforts.

Chart abstracts and behavioral surveys further supported the psychiatric diagnosis. Most importantly, unlike many earlier algorithms, the one did not rule out other psychiatric diseases. Psychiatric and other neurological disorders were absent in the controls. Participants submitted a thoroughly informed consent form, including 1 authorizing prospective analyses of EHRs, before enrolling in several CAG studies at CHOP. They developed and validated an EHR-based algorithm to categorize ADHD and concomitant mental states in a pediatric healthcare network for use in upcoming genetic analyses and discovery-based investigations. In the retrospective case-control study, including data from 51,293 patients, 5,840 cases of ADHD were found, of which 46.1% had the disorder by itself, and 53.9% had it in conjunction with psychiatric comorbidities. For a study, study group sought to explore if the algorithm could detect and discriminate ADHD isolated patients from ADHD comorbid instances. According to the findings, searches for medications and ICD codes turned up the most cases. ADHD-related keywords, it turned out, did not increase yield. However, they discovered that incorporating drugs designed specifically for ADHD increased cases by 21%. For ADHD cases and controls, the positive predictive values (PPVs) were 95% and 93%, respectively. They developed a new algorithm to diagnose ADHD and comorbid illnesses effectively and showed that using an electronic algorithm technique was feasible. The work also confirmed the effectiveness of the vast biorepository for future genetic discovery-based research.

Source- jneurodevdisorders.biomedcentral.com/articles/10.1186/s11689-022-09447-9