Cost-Effectiveness of Apixaban Compared With Aspirin for Stroke Prevention in Atrial Fibrillation Among Patients Unsuitable for Warfarin

Le résumé fourni par la source

BACKGROUND: Compared with aspirin, apixaban reduces stroke risk in atrial fibrillation (AF) patients unsuitable for warfarin by 63% but does not increase major bleeding. We sought to determine the cost-effectiveness of apixaban versus aspirin. METHODS AND RESULTS: Using the Apixaban versus Acetylsalicylic Acid to Prevent Stroke in Atrial Fibrillation Patients Who Have Failed or Are Unsuitable for Vitamin-K Antagonist Treatment (AVERROES) trial and other studies, we constructed a Markov model to evaluate the costs (2011US$), quality-adjusted life-years (QALYs), and incremental cost-effectiveness of apixaban versus aspirin from the Medicare perspective. Our base-case assumed a 70-year-old AF patient cohort with a CHADS(2) score=2 and a lower-risk of bleeding. We used a 1-month cycle-length and ran separate base-case analyses assuming a trial-length (1-year) and a longer-term (10-year) follow-up. Total costs/patient were $3454 and $1805 for apixaban and aspirin in the trial-length and $44 232 and $50 066 in the 10-year model. Corresponding QALYs were 0.96 and 0.96 in the trial-length and 6.87 and 6.51 in the 10-year model, making apixaban inferior in the first model but dominant in the latter. Conclusions were sensitive to baseline stroke rate in both models, and the monthly cost of major stroke, relative risk of stroke, and prior vitamin-K antagonist use in the life-time model. Probabilistic sensitivity analysis suggested apixaban would only be a cost-effective alternative (<$50 000/QALY) to aspirin 11% of the time in the trial-length model, but cost-effective or dominant 96.7% and 87.5% of iterations in the 10-year model. CONCLUSIONS: In our trial-length model, apixaban was more costly and no more effective than aspirin; however, as follow-up was extended, apixaban became cost-effective and eventually dominant.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.