Performance of Atrial Fibrillation Risk Prediction Models in Over Four Million Individuals.

Background - Atrial fibrillation (AF) is associated with increased risks of stroke and heart failure. Electronic health record (EHR) based AF risk prediction may facilitate efficient deployment of interventions to diagnose or prevent AF altogether. Methods - We externally validated an EHR atrial fibrillation (EHR-AF) score in IBM Explorys Life Sciences, a multi-institutional dataset containing statistically de-identified EHR data for over 21 million individuals ("Explorys Dataset"). We included individuals with complete AF risk data, ≥2 office visits within two years, and no prevalent AF. We compared EHR-AF to existing scores including CHARGE-AF, C2HEST, and CHA2DS2-VASc. We assessed association between AF risk scores and 5-year incident AF, stroke, and heart failure using Cox proportional hazards modeling, 5-year AF discrimination using c-indices, and calibration of predicted AF risk to observed AF incidence. Results - Of 21,825,853 individuals in the Explorys Dataset, 4,508,180 comprised the analysis (age 62.5, 56.3% female). AF risk scores were strongly associated with 5-year incident AF (hazard ratio [HR] per standard deviation [SD] increase 1.85 using CHA2DS2-VASc to 2.88 using EHR-AF), stroke (1.61 using C2HEST to 1.92 using CHARGE-AF), and heart failure (1.91 using CHA2DS2-VASc to 2.58 using EHR-AF). EHR-AF (c-index 0.808 [95%CI 0.807-0.809]) demonstrated favorable AF discrimination compared to CHARGE-AF (0.806 [0.805-0.807]), C2HEST (0.683 [0.682-0.684]), and CHA2DS2-VASc (0.720 [0.719-0.722]). Of the scores, EHR-AF demonstrated the best calibration to incident AF (calibration slope 1.002 [0.997-1.007]). In subgroup analyses, AF discrimination using EHR-AF was lower in individuals with stroke (c-index 0.696 [0.692-0.700]) and heart failure (0.621 [0.617-0.625]). Conclusions - EHR-AF demonstrates predictive accuracy for incident AF using readily ascertained EHR data. AF risk is associated with incident stroke and heart failure. Use of such risk scores may facilitate decision-support and population health management efforts focused on minimizing AF-related morbidity.

[1]  Richard J Stevens,et al.  Validation of clinical prediction models: what does the "calibration slope" really measure? , 2020, Journal of clinical epidemiology.

[2]  D. Stub,et al.  Alcohol Abstinence in Drinkers with Atrial Fibrillation. , 2020, The New England journal of medicine.

[3]  Jeffrey M. Ashburner,et al.  Development and Validation of a Prediction Model for Atrial Fibrillation Using Electronic Health Records. , 2019, JACC. Clinical electrophysiology.

[4]  A. Farcomeni,et al.  A Simple Clinical Risk Score (C2HEST) for Predicting Incident Atrial Fibrillation in Asian Subjects: Derivation in 471,446 Chinese Subjects, With Internal Validation and External Application in 451,199 Korean Subjects , 2019, Chest.

[5]  D. Mutlak,et al.  Risk Score for Prediction of 10-Year Atrial Fibrillation: A Community-Based Study , 2018, Thrombosis and Haemostasis.

[6]  R. Mahajan,et al.  PREVEntion and regReSsive Effect of weight-loss and risk factor modification on Atrial Fibrillation: the REVERSE-AF study , 2018, Europace : European pacing, arrhythmias, and cardiac electrophysiology : journal of the working groups on cardiac pacing, arrhythmias, and cardiac cellular electrophysiology of the European Society of Cardiology.

[7]  S. Nattel,et al.  Associations of Obstructive Sleep Apnea With Atrial Fibrillation and Continuous Positive Airway Pressure Treatment: A Review , 2018, JAMA cardiology.

[8]  S. Connolly,et al.  Rivaroxaban for Stroke Prevention after Embolic Stroke of Undetermined Source , 2018, The New England journal of medicine.

[9]  L. Jordaens,et al.  Catheter Ablation for Atrial Fibrillation with Heart Failure , 2018, The New England journal of medicine.

[10]  Kenney Ng,et al.  The MELD-Plus: A generalizable prediction risk score in cirrhosis , 2017, PloS one.

[11]  P. Wolf,et al.  Stroke as the Initial Manifestation of Atrial Fibrillation: The Framingham Heart Study , 2017, Stroke.

[12]  Pedro L. Teixeira,et al.  Evaluation of a Prediction Model for the Development of Atrial Fibrillation in a Repository of Electronic Medical Records. , 2016, JAMA cardiology.

[13]  D. McManus,et al.  A comparison of the CHARGE-AF and the CHA2DS2-VASc risk scores for prediction of atrial fibrillation in the Framingham Heart Study. , 2016, American heart journal.

[14]  N. Gronich,et al.  Usefulness of CHADS2 and CHA2DS2-VASc Scores in the Prediction of New-Onset Atrial Fibrillation: A Population-Based Study. , 2016, The American journal of medicine.

[15]  N. Roetker,et al.  Prediction of Atrial Fibrillation in a Racially Diverse Cohort: The Multi‐Ethnic Study of Atherosclerosis (MESA) , 2016, Journal of the American Heart Association.

[16]  P. Ellinor,et al.  A Simple and Portable Algorithm for Identifying Atrial Fibrillation in the Electronic Medical Record. , 2016, The American journal of cardiology.

[17]  Olga V. Demler,et al.  Tests of calibration and goodness‐of‐fit in the survival setting , 2015, Statistics in medicine.

[18]  B. Lo.  Sharing clinical trial data: maximizing benefits, minimizing risk. , 2015, JAMA.

[19]  P. Okin,et al.  Atrial fibrillation incidence and risk factors in relation to race-ethnicity and the population attributable fraction of atrial fibrillation risk factors: the Multi-Ethnic Study of Atherosclerosis. , 2015, Annals of epidemiology.

[20]  M. O'Neill,et al.  Catheter ablation of atrial fibrillation in heart failure. , 2013, Heart failure clinics.

[21]  David D. McManus,et al.  Simple Risk Model Predicts Incidence of Atrial Fibrillation in a Racially and Geographically Diverse Population: the CHARGE‐AF Consortium , 2013, Journal of the American Heart Association.

[22]  Gregory Y H Lip,et al.  Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation. , 2010, Chest.

[23]  Chin-Tsang Chiang,et al.  Estimation methods for time‐dependent AUC models with survival data , 2009 .

[24]  M. Pencina,et al.  Evaluating the added predictive ability of a new marker: From area under the ROC curve to reclassification and beyond , 2008, Statistics in medicine.

[25]  Tianxi Cai,et al.  Evaluating Prediction Rules for t-Year Survivors With Censored Regression Models , 2007 .

[26]  M. Domanski,et al.  Relationships Between Sinus Rhythm, Treatment, and Survival in the Atrial Fibrillation Follow-Up Investigation of Rhythm Management (AFFIRM) Study , 2004, Circulation.

[27]  J. Riancho,et al.  Atrial fibrillation as a risk factor for stroke recurrence. , 2003, The American journal of medicine.

[28]  Lippincott Williams Wilkins,et al.  Stroke Prevention in Atrial Fibrillation Study: Final Results , 1991, Circulation.

[29]  P. Wolf,et al.  Atrial fibrillation: a major contributor to stroke in the elderly. The Framingham Study. , 1987, Archives of internal medicine.

[30]  D. Cox.  Two further applications of a model for binary regression , 1958 .

[31]  L. Di Biase,et al.  Validation of the Framingham Heart Study and CHARGE-AF Risk Scores for Atrial Fibrillation in Hispanics, African-Americans, and Non-Hispanic Whites. , 2016, The American journal of cardiology.

[32]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[33]  Maria Trusz Gluza.  [Atrial fibrillation and hypertension]. , 2012, Kardiologia polska.