Do physicians correctly calculate thromboembolic risk scores? A comparison of concordance between manual and computer-based calculation of CHADS2 and CHA2DS2-VASc scores.

Background Clinical risk scores, CHADS₂ and CHA₂DS₂-VASc scores, are the established tools for assessing stroke risk in patients with atrial fibrillation ( AF). Aim The aim of this study is to assess concordance between manual and computer-based calculation of CHADS₂ and CHA₂DS₂-VASc scores, as well as to analyse the patient categories using CHADS₂ and the potential improvement on stroke risk stratification with CHA₂DS₂-VASc score. Methods We linked data from Atrial Fibrillation Spanish registry FANTASIIA. Between June 2013 and March 2014, 1318 consecutive outpatients were recruited. We explore the concordance between manual scoring and computer-based calculation. We compare the distribution of embolic risk of patients using both CHADS₂ and CHA₂DS₂-VASc scores Results The mean age was 73.8 ± 9.4 years, and 758 (57.5%) were male. For CHADS₂ score, concordance between manual scoring and computer-based calculation was 92.5%, whereas for CHA₂DS₂-VASc score was 96.4%. In CHADS₂ score, 6.37% of patients with AF changed indication on antithrombotic therapy (3.49% of patients with no treatment changed to need antithrombotic treatment and 2.88% of patients otherwise). Using CHA₂DS₂-VASc score, only 0.45% of patients with AF needed to change in the recommendation of antithrombotic therapy. Conclusion We have found a strong concordance between manual and computer-based score calculation of both CHADS₂ and CHA₂DS₂-VASc risk scores with minimal changes in anticoagulation recommendations. The use of CHA₂DS₂-VASc score significantly improves classification of AF patients at low and intermediate risk of stroke into higher grade of thromboembolic score. Moreover, CHA₂DS₂-VASc score could identify 'truly low risk' patients compared with CHADS₂ score. [ABSTRACT FROM AUTHOR]