Letter to the editor Self-management of Vitamin K Antagonists Is More Cost-effective Than Dabigatran for Stroke Prevention in Non-valvular Atrial Fibrillation in Spain. Response

To the Editor,

We feel that, in their letter, Souto et al. have not taken into account a series of highly relevant points concerning the results of our article and the topics raised.¹

First, the argument that treatment with vitamin K antagonists is more cost-effective is not supported by evidence, since there are no economic evaluations that compare the 2 options. In addition, the authors interpret the results cited of the indirect comparison of the efficacy of self-management vs dabigatran as a “clear trend favoring self-management”, whereas the original study found no statistically significant differences in any of the variables analyzed.

Second, in our study, the key factor indicating the advantage of dabigatran over warfarin in terms of the cost-effectiveness ratio was not the cost of monitoring the international normalized ratio, but the difference in the efficacy of the two agents, especially with respect to the risk of ischemic stroke (see Table 7 in Juanatey et al.¹).

Third, self-management of oral anticoagulation therapy is known to be very infrequent in the Spanish population (1%-2%).

Fourth, and highly relevant, as explained in our Table 3, our estimate of the cost of international normalized ratio monitoring was obtained by weighting the different modalities of oral anticoagulation therapy monitoring employed in Spain (primary care, in-hospital, home monitoring, self-monitoring) based on expert opinion and the best publicly-available evidence.²

To respond to the letter of Souto et al., we believe that the cost proposed by these authors, and the difference between the 2 sources still does not affect our results because self-management is performed by only 1% of all patients. To be precise, self-management would lead to a 1% cost reduction in monitoring in well controlled patients (382.2 euros vs 378.2 euros) and to a 2% reduction in poorly controlled patients (472.7 euros vs 464.7 euros). A related issue is that the authors fail to consider that our study included a sensitivity analysis of this variable, which examined the extent to which the incremental cost-effectiveness ratio differed if we varied the cost of international normalized ratio monitoring by ±30%; this analysis showed that the costs obtained were also under euros 30 000/quality-adjusted life-year gained. Finally, the authors’ estimate of 420 euros (private service fee for supervised self-management) does not take into account periodic visits to the physician, funded by the Spanish National Health System, whose perspective is the relevant issue in this analysis.

Given all of the above, we consider that the conclusion of Souto et al.—“The self-management model will therefore be the dominant pharmacoeconomic model (net savings) regardless of the type of analysis used”—to only be a hypothesis. Thus, the aforementioned statement and the title of the letter should, at the very least, be expressed in the conditional tense.

CONFLICTS OF INTEREST

All of the authors of the original article have contributed to this response. Two of the authors (N. González-Rojas and V. Becerra) are employees of Boehringer Ingelheim. Another author (I. Oyagüez) is a member of Pharmacoeconomics & Outcomes Research (PORIB), a consultancy that specializes in financial analysis, which assisted Boehringer Ingelheim in an advisory capacity in relation to this study.

However, in no case did this situation influence the results presented.

.