Cost-Effectiveness of Preference-Based Antithrombotic Therapy for Patients With Nonvalvular Atrial Fibrillation

B. F. Gage, A. B. Cardinalli, D. K. Owens
1998 Stroke  
and Purpose-Recent atrial fibrillation guidelines recommend the incorporation of patient preferences into the selection of antithrombotic therapy. However, no trial has examined how incorporating such preferences would affect quality-adjusted survival or medical expenditure. We compared 10-year projections of quality-adjusted survival and medical expenditure associated with two atrial fibrillation treatment strategies: warfarin-for-all therapy versus preference-based therapy. The
more » ... d strategy prescribed whichever antithrombotic therapy, warfarin or aspirin, had the greater projected quality-adjusted survival. Methods-We used decision analysis stratified by the number of stroke risk factors (history of stroke, transient ischemic attack, hypertension, diabetes, or heart disease). The base case focused on compliant 65-year-old patients who had nonvalvular atrial fibrillation and no contraindications to antithrombotic therapy. Results-In patients whose only risk factor for stroke was atrial fibrillation, preference-based therapy improved projected quality-adjusted survival by 0.05 quality-adjusted life year (QALY) and saved $670. For patients who had atrial fibrillation and one additional risk factor for stroke, preference-based therapy improved quality-adjusted survival by 0.02 QALY and saved $90. In patients who had atrial fibrillation and multiple additional risk factors for stroke, preference-based therapy increased medical expenditures and did not improve quality-adjusted survival substantially. The benefits of preference-flexible therapy arose from the minority of patients who would have had a longer quality-adjusted survival if they had been prescribed aspirin rather than warfarin. Conclusions-As do risks of stroke and of hemorrhage, patients' preferences help to determine which antithrombotic therapy is optimal. Preference-based treatment should improve quality-adjusted survival and reduce medical expenditure in patients who have nonvalvular atrial fibrillation and not more than one additional risk factor for stroke. (Stroke. 1998;29:1083-1091.)
doi:10.1161/01.str.29.6.1083 pmid:9626276 fatcat:sfdeqgqzy5dplf7c2puhtb3tta