Comparison of algorithms for identifying people with HIV from electronic medical records in a large, multi-site database. Issue 2 (17th May 2022)

• Record Type: Journal Article
• Authors: Ridgway, Jessica P; Mason, Joseph A; Friedman, Eleanor E; Devlin, Samantha; Zhou, Junlan; Meltzer, David; Schneider, John
• Is Part Of: JAMIA open. Volume 5:Issue 2(2022)
• Publication Date: 2022-05-17
• Subjects: HIV -- clinical phenotyping -- EMR -- clinical informatics -- diagnostic algorithm; Medical informatics -- Periodicals; 610.285
• DOI: 10.1093/jamiaopen/ooac033
• Languages: English
• ISSNs: 2574-2531

Abstract

Abstract: Objective: As electronic medical record (EMR) data are increasingly used in HIV clinical and epidemiologic research, accurately identifying people with HIV (PWH) from EMR data is paramount. We sought to evaluate EMR data types and compare EMR algorithms for identifying PWH in a multicenter EMR database. Materials and Methods: We collected EMR data from 7 healthcare systems in the Chicago Area Patient-Centered Outcomes Research Network (CAPriCORN) including diagnosis codes, anti-retroviral therapy (ART), and laboratory test results. Results: In total, 13 935 patients had a positive laboratory test for HIV; 33 412 patients had a diagnosis code for HIV; and 17 725 patients were on ART. Only 8576 patients had evidence of HIV-positive status for all 3 data types (laboratory results, diagnosis code, and ART). A previously validated combination algorithm identified 22 411 patients as PWH. Conclusion: EMR algorithms that combine laboratory results, administrative data, and ART can be applied to multicenter EMR data to identify PWH. Lay Summary: Electronic medical record (EMR) data are increasingly utilized for HIV-related research. Therefore, it is important to accurately identify people who are HIV-positive from data present in EMRs. We evaluated different types of EMR data and compared EMR algorithms for identifying people with HIV (PWH) in a multicenter EMR database. Our data source was the Chicago Area Patient-Centered Outcomes Research Network (CAPriCORN), which contains EMR data from diverse healthcare systems in Chicago. We collected different EMR data types from CAPriCORN, including diagnosis codes, HIV medication data, and laboratory test results, to determine which data types were most helpful for determining if patients were HIV-positive. In the database, 13 935 patients had a positive laboratory test for HIV; 33 412 patients had a diagnosis code for HIV; and 17 725 patients were prescribed HIV-specific medication. Only 8576 patients were identified as HIV-positive in all 3 data types (laboratory results, diagnosis code, and medications). We applied an algorithm that utilized combinations of different data types, and it identified 22 411 patients as PWH. In conclusion, we found that EMR algorithms that combine laboratory results, diagnosis codes, and medications can be applied to multicenter EMR data to identify PWH.