Cost-effectiveness of rivaroxaban versus warfarin in non-valvular atrial fibrillation patients with chronic kidney disease in China.

Abstract

In non-valvular atrial fibrillation (NVAF) patients with chronic kidney disease (CKD), rivaroxaban was not inferior to warfarin in preventing stroke and systemic embolism. However, a comparative evaluation of the cost-effectiveness of rivaroxaban and warfarin therapies for NVAF patients at different renal function levels has not yet been reported, and this study aimed to estimate the cost-effectiveness of rivaroxaban compared with warfarin in Chinese NVAF patients with CKD. A Markov model was constructed to estimate quality-adjusted life years (QALYs) and lifetime costs associated with the use of rivaroxaban relative to warfarin in patients with NVAF at different estimated glomerular filtration rate (eGFR) levels as follows: 30 to <50, 50 to <80 and ≥80mL/min. Input parameters were sourced from the clinical literature. Probabilistic sensitivity analyses were performed to assess model uncertainty. The incrementalQALYs with rivaroxaban was slightly increased by approximately 0.3 QALY as compared with that with warfarin in all the subgroups, resulting in an ICER of $9,736/QALY (eGFR, 30 to <50mL/min), $9,758/QALY (50 to <80mL/min) and $9,969/QALY (≥80mL/min). The probabilistic sensitivity analysis suggested a chance of >80% that the ICER would be lower than the willingness-to-pay threshold of three times the GDP of China in 2019 in all the subgroups. Results were consistent even under the assumption of anticoagulant discontinuation after major bleeding events. The model was most sensitive to event-free-related utility and survival rates. The existing evidence supports the cost-effectiveness of rivaroxaban therapy as an alternative anticoagulant to warfarin for patients with NVAF at different renal function levels.