Abstract 17117: Prediction of Ischemic Stroke or Transient Ischemic Attack in Atrial Fibrillation Using the CHA2DS2-VASc Score in Whites, Blacks, and Hispanics
Introduction: The CHA2DS2-VASc score—derived and validated in mostly white populations—is widely used to predict the risk of ischemic stroke (IS) in patients with atrial fibrillation (AF). Data on its predictive ability in other races or ethnicities are scarce.
Hypothesis: The association of the CHA2DS2-VASc score and its predictive ability for IS or transient ischemic attack (TIA) will be comparable in whites, blacks, and Hispanics. We tested our hypothesis in Optum Clinformatics, a large administrative claims database.
Methods: We included patients enrolled in Optum Clinformatics between 2009 and 2014 with 1 inpatient or 2 outpatient claims for AF and >180 days of enrollment before AF diagnosis. Patients who were on anticoagulants before AF diagnosis were excluded. IS and TIA were ascertained using hospital discharge diagnosis codes in the first position. Stratified by race/ethnicity, we computed the (a) incidence rates of IS or TIA by CHA2DS2-VASc, and (b) c-statistic of CHA2DS2-VASc for 1-year risk of IS or TIA.
Results: Data from 231,073 patients (mean age, 73±13 years; 47% women; mean CHA2DS2-VASc, 4.1±2.1; 194,789 white; 20,103 black; 16,181 Hispanic) were studied. After a mean follow-up of 19 months, there were 5,209 IS or TIA events. Incidence rates of IS or TIA increased with increasing CHA2DS2-VASc; there were no race/ethnicity-based differences (P for interaction = 0.20) (Figure). The c-statistic (95% CI) of the CHA2DS2-VASc and CHA2DS2-VASc + race/ethnicity were 0.683 (0.674-0.692) and 0.685 (0.676-0.693), respectively. The C-statistic of the CHA2DS2-VASc score in whites, blacks, and Hispanics were 0.681 (0.672-0.691), 0.694 (0.671-0.712), and 0.665 (0.636-0.695), respectively.
Conclusions: The predictive ability of the CHA2DS2-VASc score for IS or TIA is comparable among whites, blacks, and Hispanics; hence, it can be used reliably in the latter 2 groups. Addition of race/ethnicity to the CHA2DS2-VASc score does not improve its predictive ability.
Author Disclosures: L.Y. Chen: None. F.L. Norby: None. A.M. Chamberlain: None. R.F. MacLehose: None. L.G. Bengtson: None. P.L. Lutsey: None. A. Alonso: None.
- © 2016 by American Heart Association, Inc.