OBJECTIVES: This study aims to estimates cost and effectiveness of itraconazole and posaconazole in prophylaxis of invasive fungal infections (IFI) patients with neutropenia in Spain. METHODS: From the payer’s perspective, a Markov model was developed to represent the transition of a neutropenic cohort of patients with prophylaxis of IFI through different health states: patient in prophylaxis of IFI, IFI, no-IFI, death by IFI and death by some other causes. Efficacy data on incidence of proven or probable IFI, survival data and health-state utilities were obtained from published studies. Deterministic results were estimated and a probabilistic sensitivity analysis was conducted using statistical distributions in order to capture parameter uncertainty in the decision model. Treatment costs were obtained from a panel of clinical experts. Costs were referred to year 2009 and a time horizon of 1 year was chosen. Results were presented as expected cost per quality adjusted life years (QALYs) gained and represented in cost-effectiveness acceptability curves (CEACs). RESULTS: In the deterministic analysis, the expected cost per patient was greater in the posaconazole cohort (€19,522) in comparison with itraconazole cohort (€19,508). The estimated effectiveness was the same (0.42 QALY gained) in both cohorts. In the probabilistic analysis CEACs showed that the probability that the treatment with itraconazole was more cost-effective than the treatment with posaconazole using alternative values for the maximum value that the health service would be willing to pay for an additional QALY in prophylaxis of IFI was greater in the itraconazole cohort. CONCLUSIONS: When itraconazole has been compared to posaconazole we may conclude that the first one is a dominant strategy. Results from probabilistic sensibility analysis show that the choice of optimal strategy is independent on the maximum that the health service is prepared to pay per additional QALY gained because itraconazole has a greater probability of being cost-effective for all threshold values.