Integrating real-time clinical information to provide estimates of net clinical benefit of antithrombotic therapy for patients with atrial fibrillation

Abstract

Background: Guidelines for anticoagulant therapy in patients with atrial fibrillation are based on stroke risk as calculated by either the CHADS2 or the CHA2DS2VASc scores and do not integrate bleeding risk in an explicit, quantitative manner. Our objective was to quantify the net clinical benefit resulting from improved decision making about antithrombotic therapy.

Methods and results: This study is a retrospective cohort study of 1876 adults with nonvalvular atrial fibrillation or flutter seen in primary care settings of an integrated healthcare delivery system between December 2012 and January 2014. Projections for quality-adjusted life expectancy reported as quality-adjusted life-years were calculated by a decision analytic model that integrates patient-specific risk factors for stroke and hemorrhage and examines strategies of no antithrombotic therapy, aspirin, or oral anticoagulation with warfarin. Net clinical benefit was defined by the gain or loss in quality-adjusted life expectancy between current treatment and treatment recommended by an Atrial Fibrillation Decision Support Tool. Current treatment was discordant from treatment recommended by the Atrial Fibrillation Decision Support Tool in 931 patients. A clinically significant gain in quality-adjusted life expectancy (defined as ≥0.1 quality-adjusted life-years) was projected in 832 patients. Subgroups were examined. For example, oral anticoagulant therapy was recommended for 188 who currently were receiving no antithrombotic therapy. For the entire cohort, a total of 736 quality-adjusted life-years could be gained were treatment changed to that recommended by the Atrial Fibrillation Decision Support Tool.

Conclusions: Use of a decision support tool that integrates patient-specific stroke and bleeding risk could result in significant gains in quality-adjusted life expectancy for a primary care population of patients with atrial fibrillation.

Keywords: atrial fibrillation; decision support techniques; medical informatics.