What is the relative cost-effectiveness of available anticoagulation therapies for nonvalvular atrial fibrillation?
The authors used data from completed clinical trials to construct a Markov decision-analysis model for the purpose of evaluating lifetime costs and quality-adjusted life-years (QALYs), for the novel oral anticoagulants compared with warfarin. The model used a hypothetical cohort of 70-year-old patients with nonvalvular atrial fibrillation, with at least one of the CHADS2
risk factors, no contraindications to anticoagulation, and a creatinine clearance of at least 50 ml/min. The cost estimates for endpoints (ischemic stroke, intracranial hemorrhage, myocardial infarction, gastrointestinal bleed), where estimated from the 2009 mean costs published by the Agency for Healthcare Research and Quality from the Healthcare Cost and Utilization Project data, based on International Classification of Diseases (ICD)-9 codes. Sensitivity analyses were performed of the model variables to determine the relative impact of each. A Monte Carlo probabilistic sensitivity analysis was performed. A standard threshold of willingness-to-pay of $50,000 per QALYs was used.
The authors reported that warfarin had the lowest lifetime cost of $77,813 (standard deviation [SD], $2,223). Rivaroxaban 20 mg had a cost of $78,738 (± $1,852), dabigatran 150 mg had a cost of $82,719 (± $1,959), and apixaban 5 mg had a cost of $85,326 (± $1,512). The estimated QALYs gained were 8.47 (SD, 0.06) for apixaban 5 mg, 8.41 (± 0.07) for dabigatran 150 mg, 8.26 (± 0.06) for rivaroxaban 20 mg, and 7.97 (± 0.04) for warfarin. In a Monte Carlo probabilistic analysis, apixaban 5 mg, dabigatran 150 mg, rivaroxaban 20 mg, and warfarin were cost-effective in 45.1%, 40%, 14.9%, and 0% of the simulations, respectively.
The authors concluded that in patients with nonvalvular atrial fibrillation and an increased risk of stroke, apixaban 5 mg, dabigatran 150 mg, and rivaroxaban 20 mg were all cost-effective alternatives to warfarin. The authors further opined that the cost-effectiveness of novel oral anticoagulants was dependent on pricing in the United States and neurological events associated with rivaroxaban 20 mg.
This rigorous decision-analysis modeling study uses data from the clinical trials to estimate potential costs of different anticoagulants. When interpreting these results, it is important to remember that these data are based on clinical trial data, not real-world observations. It is also important to note that this analysis was limited to a hypothetical cohort of subjects older than age 70, with at least one risk factor for stroke. Also, the resulting cost-effectiveness estimates are extremely sensitive to drug price, as well as exquisitely sensitive to patient age. Age had an even more potent effect on the cost-effectiveness estimates than drug price, reflecting the fact that age is an extremely potent predictor of risk of stroke without anticoagulation, as well as risk of intracranial hemorrhage with anticoagulation. It is also important to note that no consideration for international normalized ratio (INR) results, or quality of warfarin therapy, was included in this model study. And while the models studied did include cost of INR testing, they did not appear to include manpower costs for managing warfarin. For the time being, and with all of the potential limitations considered, these data provide the most reasonable guide for clinicians in considering overall relative cost-effectiveness of the currently available anticoagulation options for patients with nonvalvular atrial fibrillation.
James B. Froehlich, M.D., M.P.H., F.A.C.C. (Disclosure)
Arrhythmias, Cardiac Rhythm Management, Afib, General Cardiology, Prevention/Vascular, Anticoagulation Management