Development and validation of a heart failure with preserved ejection fraction cohort using electronic medical records

• Published in: BMC cardiovascular disorders 2018-06, Vol.18 (1), p.128-128, Article 128
• Main Authors: Patel, Yash R, Robbins, Jeremy M, Kurgansky, Katherine E, Imran, Tasnim, Orkaby, Ariela R, McLean, Robert R, Ho, Yuk-Lam, Cho, Kelly, Michael Gaziano, J, Djousse, Luc, Gagnon, David R, Joseph, Jacob
• Format: Article
• Language: English
• Subjects: Aged; Aged, 80 and over; Algorithms; Analysis; Arthritis; Biomarkers - blood; Brain natriuretic peptide; Cardiac output; Clinical trials; Data Mining - methods; Databases, Factual; Development and progression; Diagnosis; Diuretics - therapeutic use; Dyspnea; Echocardiography; Ejection fraction; Electronic Health Records; Electronic medical records; Electronic records; Epidemiology; Female; Genomes; Heart diseases; Heart failure; Heart Failure - classification; Heart Failure - diagnosis; Heart Failure - epidemiology; Heart Failure - physiopathology; Humans; International Classification of Diseases; Male; Medical records; Middle Aged; Mortality; Natriuretic Peptide, Brain - blood; Natural Language Processing; Peptide Fragments - blood; Preserved ejection fraction; Primary care; Reproducibility of Results; Stroke Volume; Ultrasonic imaging; United States - epidemiology; United States Department of Veterans Affairs; Validation; Ventricle; Ventricular Function, Left; Veterans
• ISSN: 1471-2261; 1471-2261
• DOI: 10.1186/s12872-018-0866-5

Heart failure (HF) with preserved ejection fraction (HFpEF) comprises nearly half of prevalent HF, yet is challenging to curate in a large database of electronic medical records (EMR) since it requires both accurate HF diagnosis and left ventricular ejection fraction (EF) values to be consistently ≥50%. We used the national Veterans Affairs EMR to curate a cohort of HFpEF patients from 2002 to 2014. EF values were extracted from clinical documents utilizing natural language processing and an iterative approach was used to refine the algorithm for verification of clinical HFpEF. The final algorithm utilized the following inclusion criteria: any International Classification of Diseases-9 (ICD-9) code of HF (428.xx); all recorded EF ≥50%; and either B-type natriuretic peptide (BNP) or aminoterminal pro-BNP (NT-proBNP) values recorded OR diuretic use within one month of diagnosis of HF. Validation of the algorithm was performed by 3 independent reviewers doing manual chart review of 100 HFpEF cases and 100 controls. We established a HFpEF cohort of 80,248 patients (out of a total 1,155,376 patients with the ICD-9 diagnosis of HF). Mean age was 72 years; 96% were males and 12% were African-Americans. Validation analysis of the HFpEF algorithm had a sensitivity of 88%, specificity of 96%, positive predictive value of 96%, and a negative predictive value of 87% to identify HFpEF cases. We developed a sensitive, highly specific algorithm for detecting HFpEF in a large national database. This approach may be applicable to other large EMR databases to identify HFpEF patients.