A phenotyping algorithm to identify acute ischemic stroke accurately from a national biobank: the Million Veteran Program

• Type of Material: Article
• Authors: Tasnim F Imran, VA Boston Healthcare System; Daniel Posner, VA Boston Healthcare System; Jacqueline Honerlaw, VA Boston Healthcare System; Jason L Vassy, VA Boston Healthcare System; Rebecca J Song, VA Boston Healthcare System; Yuk-Lam Ho, VA Boston Healthcare System; Steven J Kittner, Baltimore VA Medical Center; Katherine P Liao, VA Boston Healthcare System; Tianxi Cai, VA Boston Healthcare System; Christopher J O'Donnell, VA Boston Healthcare System; Luc Djousse, VA Boston Healthcare System; David R Gagnon, VA Boston Healthcare System; J Michael Gaziano, VA Boston Healthcare System; Peter Wilson, Emory University; Kelly Cho, Harvard Medical School
• Language: English
• Date: 2018-01-01
• Publisher: Dove Medical Press
• Publication Version: Final Published Version
• Copyright Statement: © 2018 Imran et al.
• License: Creative Commons Attribution-Noncommercial 3.0 Unported
• Final Published Version (URL): http://dx.doi.org/10.2147/CLEP.S160764
• Title of Journal or Parent Work: Clinical Epidemiology
• ISSN: 1179-1349
• Volume: 10
• Start Page: 1509
• End Page: 1521
• Grant/Funding Information: The Cardiovascular Health Study is funded under VA Merit Award I01-CX001025. The Million Veteran Program is funded by the Office of Research and Development, Department of Veterans Affairs, supported by grant CSPG002.
• Keywords: Science & Technology; algorithm; ATTACK; ADMINISTRATIVE DATA; REGRESSION; ADAPTIVE LASSO; administrative health data; Life Sciences & Biomedicine; big data; acute ischemic stroke; HEALTH-CARE PROFESSIONALS; ASSOCIATION; DIAGNOSIS; Public, Environmental & Occupational Health; large databases; cerebrovascular accident
• Research Categories: Health Sciences, Public Health; Health Sciences, Epidemiology

Abstract

Background: Large databases provide an efficient way to analyze patient data. A challenge with these databases is the inconsistency of ICD codes and a potential for inaccurate ascertainment of cases. The purpose of this study was to develop and validate a reliable protocol to identify cases of acute ischemic stroke (AIS) from a large national database. Methods: Using the national Veterans Affairs electronic health-record system, Center for Medicare and Medicaid Services, and National Death Index data, we developed an algorithm to identify cases of AIS. Using a combination of inpatient and outpatient ICD9 codes, we selected cases of AIS and controls from 1992 to 2014. Diagnoses determined after medical-chart review were considered the gold standard. We used a machine-learning algorithm and a neural network approach to identify AIS from ICD9 codes and electronic health-record information and compared it with a previous rule-based stroke-classification algorithm. Results: We reviewed administrative hospital data, ICD9 codes, and medical records of 268 patients in detail. Compared with the gold standard, this AIS algorithm had a sensitivity of 91%, specificity of 95%, and positive predictive value of 88%. A total of 80,508 highly likely cases of AIS were identified using the algorithm in the Veterans Affairs national cardiovascular disease-risk cohort (n=2,114,458). Conclusion: Our algorithm had high specificity for identifying AIS in a nationwide electronic health-record system. This approach may be utilized in other electronic health databases to accurately identify patients with AIS.