Even if the clinical efficacy of recombinant human erythropoietin on chemotherapy-induced anemia was demonstrated, most economic studies have given unfavorable results, whatever the design and the outcome considered. OBJECTIVES: To calculate the incremental cost-effectiveness ratio (ICER) of darbepoetin alfa (Aranesp) as compared to standard palliative care in a cohort of patients treated by chemotherapy for lung cancer in clinical practice. METHODS: A Markov model was constructed to evaluate the cost effectiveness ratio of one weekly injection darbepoetin (Aranesp) compared with palliative standard care (red blood transfusion if hg<8g/dl) in the correction of chemotherapy-induced anemia. Baseline probabilities and consumed resources were calculated on the basis of a two-year retrospective study, comparing two cohorts of patients treated by chemotherapy who received (n=94) or did not receive (n=89) Aranesp. The incremental cost-effectiveness ratio (ICER) was calculated as the difference in direct costs from a healthcare perspective (transfusion requirement and anemia management costs) divided by the difference in effect (changes in haemoglobin levels). Sensitivity analysis was used to test uncertain data. RESULTS: The use of Aranesp significantly reduced the proportion of patients needing transfusions (from 33.6% to 19.1%, p<0.05) and the number of red cell units used by transfusion (from 2.97 ± 1.47 to 2.11 ± 0.47, p <0.01). Markov modeling showed that the Aranesp strategy significantly increased the mean Hb level (13 ± 0.5 vs 11.9 ± 1 g/dl, p<0.001), at the price of an increase in the main cost (respectively 1732 ± 897 and 996 ± 643 €; p<0.01). The incremental cost-effectiveness ratio was estimated to be 202 € per haemoglobin gram gained. Sensitivity analysis showed that the Aranesp strategy remained dominant in most situations. CONCLUSION: Routine use of Aranesp appears to be cost-effective in patients receiving chemotherapy for lung cancer.