OBJECTIVES: Apixaban was identified with the highest probability of being the most cost-effective first-line treatment for patients diagnosed with non-valvular atrial fibrillation (NVAF). However, there is limited data evaluating the cost-effectiveness of oral anticoagulants in high-risk NVAF subpopulations. Subgroup analyses of ARISTOPHANES, a large real-world evidence study, evaluated the safety and effectiveness of oral anticoagulants in high-risk NVAF patient subgroups. Our analysis aimed to estimate the cost-effectiveness of apixaban compared to rivaroxaban, dabigatran and warfarin from a UK perspective over a lifetime horizon using data from ARISTOPHANES subpopulations (obesity, polypharmacy, stage 3-5 chronic kidney disease (CKD), cancer, diabetes, frailty, high risk of gastrointestinal bleed, multimorbidity, prior bleed and elderly).

METHODS: A cohort-level Markov model based on the methodology of Lopez-Lopez et al. and Sterne et al. was developed and adapted to the UK NHS setting. Effectiveness/safety and baseline risk inputs were based on the ARISTOPHANES subgroup analyses. Cost and utility data were sourced from published literature, databases and datasets. Cost (reported in 2020 GBP) and quality-adjusted life years (QALY) were discounted by 3.5% annually. Cost-effectiveness was based on a willingness-to-pay threshold of £30,000/QALY. Uncertainty was explored through deterministic one-way and probabilistic sensitivity analyses for each subgroup.

RESULTS: Apixaban was cost-effective (£2,098 per QALY) in the main ARISTOPHANES population and dominant (greater QALYs and cost saving) across all subgroups versus warfarin. Apixaban was dominant versus dabigatran and rivaroxaban in the main ARISTOPHANES population and across all subgroups. The probabilistic sensitivity analysis trends were consistent with the base-case deterministic results for each subgroup.

CONCLUSIONS: Across all high-risk patient subgroups assessed in the ARISTOPHANES study, apixaban was dominant versus warfarin, dabigatran and rivaroxaban.