Accuracy of Physician Estimation of Thromboembolic Risk and Its Association with Oral Anticoagulant Utilization for the Management of Atrial Fibrillation.
Academic Article
Overview
abstract
Clinical practice guidelines for atrial fibrillation (AF) recommend thromboembolic risk stratification using empirical risk scores. We aimed to investigate how appropriately physicians recognize patients' actual thromboembolic risk and how this correlates with the use of oral anticoagulants (OACs). This prospective cohort study enrolled consecutive patients with initial AF treatment. Physicians were asked to numerically estimate each patient's thromboembolic risk, which was consecutively categorized as low (<1.0%/year), intermediate (1.0%≤ to <2.0%/year), or high risk (2.0%≤/year). The empirical thromboembolic risk was defined by the CHA2DS2-VASc score and categorized into three groups: low (0 or 1 point), intermediate (2 points), and high risk (3≤ points). Overestimation and underestimation were defined as physicians assigning a higher or lower risk category than the empirical risk. A multivariable logistic regression model was constructed to investigate the association between physicians' underestimation and OAC use. Among the 285 patients (68±12 years, female 27%), the mean CHA2DS2-VASc score was 2.3±1.6. The thromboembolic risk was accurately estimated by treating physicians in 51.6% of cases, while 39.6% were overestimated and 8.8% were underestimated. OACs were used in 89.8%, 84.1%, and 72.0% of the correctly, over, and underestimated group, respectively. After multivariable adjustment, physicians' underestimation of thromboembolic risk was independently associated with less use of OACs (adjusted OR 0.17, 95% CI 0.05-0.54, P=0.003). In conclusion, physicians frequently misrecognized thromboembolic risk in AF patients. Risk underestimation was independently associated with lower OAC use, suggesting the importance of integrating validated risk stratification tools into clinical practice to enhance shared decision-making.
