On page 539 of the issue there is a brief explanation of how monitoring of patients under warfarin was costed.

An expert panel with proven clinical experience was consulted to identify and quantify the resources used for monitoring patients under warfarin. The panel consisted of seven experts selected on the basis of geographic location, specialty, and number of patients treated as a percentage of the national total. The Delbecq method was adopted as providing more robust estimates.

The costs of monitoring patients under warfarin presented in the article refer to 2011 and were rounded up. An inflation rate of 3% per year was applied to all costs taken from sources prior to 2011. The resources identified for monitoring warfarin included INR measurement and medical consultations to adjust dosage on the assumption that the patient would not be present for 50% of such consultations. According to the expert panel, 35% of patients were managed in a primary care setting and 65% in a specialist setting (50% by internal medicine specialists and 50% by cardiologists).

The source of data on costs of patients followed in primary care was the table of state-funded health services published by the Central Administration of the Portuguese national health system (ACSS) on October 6, 2011. The costs of services paid directly by patients were not included in the analysis, since a large proportion of this population are exempt. The source of the costs of primary care consultations (without the patient present), and the costs incurred by patients managed by specialists, was Order in Council 132/2009, modified by Order in Council 839-A/2009. The costs of specialist consultations were based on the Cost Accounting Report of National Health Service hospitals for 2006, using the total unit cost. The costs for monitoring patients under warfarin are presented in Table 1.

There is indeed an error in Table 4 in our article: the costs of the individual components (drug acquisition, drug monitoring and event treatment) are incorrect. The total costs and incremental cost-effectiveness ratios are correct. The corrected costs are presented in Table 2.

On behalf of the authors, I appreciate the interest taken in our article and the comments made. It is not common practice for experts and specialists to engage in debate in the Journal using this important form of communication, but I am sure that it helps raise the profile of the Journal, of the subject-matter it publishes, and hence of its authors.

The author has no conflicts of interest to declare.