Background: The bioavailability of warfarin is an important factor affecting the achievement of therapeutic anticoagulation. It is uncertain whether less expensive generic preparations of warfarin would compromise prevention of thromboembolism or increase bleeding risk in patients with atrial fibrillation.
Objective: To compare the cost effectiveness of strategies using warfarin products with variable bioavailability in patients with a prior stroke or transient ischaemic attack related to atrial fibrillation.
Design: A Markov decision-analytic model simulating health and economic outcomes over 1 year using the perspective of a government provincial payer was created. Four strategies were compared (where F = 1 is the assumed bioavailability of the branded/reference product): (i) warfarin F = 1; (ii) warfarin F = 1.25; (iii) warfarin F = 0.80; and (iv) alternating warfarin F = 1.25 and 0.80 every other month. Direct medical costs for drugs, physician fees, laboratory testing and hospitalisation for morbid events were obtained from a government payer, a local accounting system and the medical literature. The cost of warfarin F = 1 was equivalent to the cost of the brand name warfarin and the cost of warfarin F not equal 1 was equivalent to generic warfarin.
Results: In our institution, warfarin F = 1 was similar in cost to the other three strategies (Can dollars 1361 vs Can dollars 1334-1613) and may be more effective than switching between generic preparations which have bioavailabilities at the extremes of acceptable limits (thromboembolism and bleeds 7. 1% vs 9.3%).
Conclusions: In patients with atrial fibrillation and a prior ischaemic stroke or transient ischaemic attack, the use of one warfarin agent within the range of acceptable bioavailability can be considered economically attractive from the healthcare perspective.