Summary Recombinant human erythropoietin (rHuEPO) has been advocated for the treatment of anaemia in patients submitted to cancer chemotherapy. We used decision analysis to compare the cost-effectiveness of rHuEPO supplemented with red blood cell (RBC) transfusions with conventional treatment with RBC transfusions alone. At Anaemia is a frequent complication in patients with cancer receiving cytotoxic chemotherapy. It impairs patients' physical capabilities and subjective sense of well-being, diminishing their quality of life. In controlled randomized studies, rHuEPO was associated with increased haemoglobin level, decreased transfusion requirements and improved quality of life (Abels, 1993;Case et al, 1993; Finelli et al, 1993;Cascinu et al, 1994;Henry and Abels, 1994;Cazzola et al, 1995;De Campos et al, 1995;Garton et al, 1995; Oesterborg et al, 1996;ten Bokkel-Huinink, 1996; Wurnig et al, 1996;Del Mastro et al, 1997). As the same results were obtained in a community oncology practice setting (Glaspy et al, 1997), recommendations for rHuEPO use and a treatment algorithm for its optimization were outlined (Glaspy et al, 1997).In contrast, because of increasing financial constraints throughout health care systems, the amount of money allotted for a new treatment has to be considered in relation to the magnitude of the benefit offered with respect to traditional therapy. As a consequence, cost-effectiveness analysis belongs to the development process aimed at establishing policy for drug use. The purpose of this work is to provide a cost-effectiveness analysis of rHuEPO in the chemotherapy-induced anaemia of cancer.Received 26 September 1997 Revised 30 January 1998 Accepted 16 March 1998 Correspondence to: G Barosi, Laboratorio di Informatica Medica, IRCCS Policlinico S. Matteo, Piazzale Golgi 2, 27100 Pavia, Italy
METHODS
Base caseAs the best strategy for administering rHuEPO to patients undergoing cancer chemotherapy has not yet been determined, at baseline we modelled the most resource-saving treatment strategy, i.e. the use of the drug for secondary prophylaxis, reserved for patients who had developed severe anaemia as a consequence of chemotherapy. The case upon which this analysis is based was derived from the results of a community-based US multicentre oncology study (Glaspy et al, 1997). As such, a hypothetical typical patient fulfilling the criteria for therapy receives treatment at a mean age of 65 years, does not carry any other cause for anaemia except cancer chemotherapy, always has an expected survival of more than 6 months and is treated with rHuEPO when haemoglobin falls below 10.7 g dl-'. To better reflect the health care system perspective, we assumed that both rHuEPO and RBC transfusions were administered in the hospital. We outlined our model on the basis of a 4-month course of rHuEPO with doses of 150 units kg-' three times per week, i.e. according to the protocol providing treatment guidance for patients in the community setting.