Impact of Different Electronic Cohort Definitions to Identify Patients With Atrial Fibrillation From the Electronic Medical Record

Abstract

Background Electronic medical records ( EMR s) allow identification of disease‐specific patient populations, but varying electronic cohort definitions could result in different populations. We compared the characteristics of an electronic medical record –derived atrial fibrillation ( AF ) patient population using 5 different electronic cohort definitions. Methods and Results Adult patients with at least 1 AF billing code from January 1, 2010, to December 31, 2017, were included. Based on different electronic cohort definitions, we trained 5 different logistic regression models using a labeled training data set (n=786). Each model yielded a predicted probability; patients were classified as having AF if the probability was higher than a specified cut point. Test characteristics were calculated for each model. These models were then applied to the full cohort and resulting characteristics were compared. In the training set, the comprehensive model (including demographics, billing codes, and natural language processing results) performed best, with an area under the curve of 0.89, sensitivity of 0.90, and specificity of 0.87. Among a candidate population (n=22 000), the proportion of patients identified as having AF varied from 61% in the model using diagnosis or procedure International Classification of Diseases ( ICD ) billing codes to 83% in the model using natural language processing of clinical notes. Among identified AF patients, the proportion of patients with a CHA 2 DS 2 ‐ VAS c score ≥2 varied from 69% to 85%; oral anticoagulant treatment rates varied from 50% to 66% depending on the model. Conclusions Different electronic cohort definitions result in substantially different AF study samples. This difference threatens the quality and reproducibility of electronic medical record–based research and quality initiatives.

Comparison of International Classification of Diseases and Related Health Problems, Tenth Revision Codes With Electronic Medical Records Among Patients With Symptoms of Coronavirus Disease 2019

“…For example, ICD-10 codes perform poorly to identify patients with atrial fibrillation, with a sensitivity of 88% and a specificity of 42%. 4 Similar inaccuracies have been reported for other conditions, such as stroke and acute kidney injury. 13 , 14 Our work represents clinician documentation of symptoms, and clinicians may not document all symptoms for all patients, particularly when patient volume is high or in drive-through testing scenarios.…”

Oral Anticoagulation Use in High-Risk Patients Is Improved by Elimination of False-Positive and Inactive Atrial Fibrillation Cases

“…The main novel finding of our study is that current registries, insurance databases, and health system records, with unadjudicated electronic medical record diagnoses, overreport the number of high-risk atrial fibrillation patients not taking an oral anticoagulant and thus, overestimate the treatment gap in such patients. Our 30% false-positive diagnosis of atrial fibrillation was higher than previous studies, [21][22][23][24] however, we included inactive atrial fibrillation patients with no documented atrial fibrillation in the last 5 years. Eliminating this cohort results in a false-positive atrial fibrillation of 18%, consistent with multiple previous reports.…”

“…18,20 In reviewing patients diagnosed as having atrial fibrillation as part of a quality-of-care project using our Cerner electronic medical record, we observed that active disease was not present in a large number of patients labeled as having atrial fibrillation, either because of wrong diagnosis or no documented arrhythmia recurrences in the last 5 years. Several prior reports have suggested that the misdiagnosis of atrial fibrillation in the electronic medical record is common, [21][22][23][24][25] and varies by the criteria used. We undertook a systematic retrospective electronic medical record review to determine the presence or absence of active atrial fibrillation to accurately identify high-risk patients who should be treated with oral anticoagulants.…”

Comparison between the International Classification of Primary Care and the International Classification of Diseases Classifications in Primary Care in Korea

“…[1][2][3][4][5] The classification and coding of medical care content are important for accumulating patients' medical information and supplying appropriate and continuous medical services during primary care. 6) The 10th Revision of the International Statistical Classification of Diseases and Related Health Problems (International Classification of Diseases-10 [ICD-10]), developed by the World Health Organization, is an international classification and diagnostic tool that provides information on the causes of death and diagnoses of morbidity and mortality statistics. 7) In South Korea, the translated version of ICD-10 has been adapted for medical classification under the national health insurance (NHI) system, and medical fee reimbursement is performed based on ICD-10.…”