I read thoroughly the article written by Lee et al, which reported that rivaroxoban would be a cost–effective alternative to warfarin even with the lowest price of rivaroxoban in the market. However, in the sensitivity analysis, the authors did not mention if the lowest price of warfarin (i.e., $2/month) had any impact on the cost–effectiveness of rivaroxoban. The assumed monthly cost of warfarin in this study was $32 irrespective of the monthly cost required for coagulation monitoring. With an easy calculation, using the lowest price of warfarin would save an additional cost of $12,600 within the time horizon of 35 years ([32 − 2] × 12 ^months × 35 ^years ). When this number is added to $5,912, the extra cost of rivaroxoban compared to warfarin calculated by Lee et al , using warfarin would save $18,512. Furthermore, they reported that rivaroxoban might gain a further 0.215 quality-adjusted life years . If the additional cost of $18,512 is divided by 0.215, the resulting incremental cost–effectiveness ratio would be $86,102 per quality-adjusted life year, surpassing the cost–effectiveness threshold of $50,000. Even when we consider the monthly cost of $15 for warfarin, as did Shah and Gage for the evaluation of dabigatran cost–effectiveness, the incremental cost–effectiveness ratio would be $56,800, higher than the cost–effectiveness threshold.

Lee et al mathematically extrapolated the efficacy and safety outcomes of rivaroxoban within a 2-year follow up, achieved in the Rivaroxaban Once Daily Oral Direct Factor Xa Inhibition Compared with Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation (ROCKET-AF), to a time horizon of 35 years. It is not clear that such a mathematical extrapolation could reliably reflect the behavior of rivaroxoban in the real world. Indeed, if rivaroxoban were compared to warfarin within longer follow-ups, the rate of efficacy and safety outcomes would change, altering the cost–effectiveness of rivaroxoban.