OBJECTIVES

Nonvalvular atrial fibrillation (NVAF) is highly prevalent and increases the risk of cardiovascular events significantly, leading to decreased quality of life and increased healthcare costs. In a recent subgroup analysis, treatment response was shown to vary for patients exhibiting worsening renal function (WRF) on-treatment. It is important to understand the cost-effectiveness of novel oral anticoagulant (NOAC) use in this population.

METHODS

A cost-effectiveness analysis (CEA) was conducted using a semi-Markov decision model with inputs from the medical literature. Input parameters were sourced from published clinical literature including a multicenter clinical trial and relevant subgroup analysis. Clinical assumptions were validated across other published CEA studies and with expert clinical opinion. The baseline population of interest was elderly US male patients at increased risk for stroke (CHADS score ≥2) undergoing treatment for electrocardiographically documented NVAF and exhibiting WRF. Rivaroxaban 20mg daily was compared to warfarin treatment with target international normalized ratio (INR) of 2.0-3.0. The main outcome measure of interest was incremental net monetary benefits (INMB) of rivaroxaban treatment versus warfarin.

RESULTS

Remaining lifetime use of rivaroxaban is associated with 5.69 QALYs at a cost of $66,075 per patient, while use of warfarin produced 5.22 QALYs with costs of $78,504 per patient. At a willingness-to-pay (WTP) of $150,000 per QALY, incremental net monetary benefits (INMB) per patient are $83,590. These results were shown to be highly robust through deterministic and probabilistic sensitivity analyses. In the male WRF population, treatment of NVAF with warfarin was dominated by treatment with rivaroxaban in 99.4% of 10,000 simulations.

CONCLUSIONS

Rivaroxaban is a dominant treatment over warfarin in elderly US male NVAF patients exhibiting WRF, providing increased QALYs at a decreased overall cost. Application of these findings may require healthcare providers to successfully predict which patients are likely to exhibit WRF prior to treatment initiation.