Voriconazole treatment of invasive aspergillosis: real-world versus health-economic model results.

objectiveThe objective of this study was to assess, in a real-world setting, the predictive validity (in terms of clinical outcome and treatment cost) of the voriconazole arm of a health-economic model applied in the Belgian reimbursement submission for use of voriconazole in the treatment of invasive aspergillosis.

methodsA non-interventional study was designed to prospectively collect clinical response and direct costs data related to the treatment of invasive aspergillosis with voriconazole (oral, intravenous) in real-world practice. The outcomes of this study were compared with the inputs and outputs of the health-economic model. For the analysis, unit costs of 2002 from the public payer's perspective, as used in the model, were applied.

resultsData from 116 patients with invasive aspergillosis starting treatment with voriconazole were collected. At 12 weeks, there were similar rates of satisfactory clinical response for the observational study and the model, the latter based on the results of a clinical study (50% vs 53%, respectively). Overall mortality rates at 84 days were 42% in the observational study and 29% in the model. Average total healthcare cost associated with voriconazole treatment was lower in the observational study compared with the model for all patients. When the cost for all hospitalization days from the start until the end of the fungal infection was included in the analysis, the average total cost was euro19,674. When the cost for only those hospitalization days solely related to the fungal infection was included in the analysis, the average total cost was euro12,376. These costs are below the cost predicted by the model of euro21,298.

conclusionsThis analysis demonstrates that the results provided in the voriconazole arm of the health-economic model were valid estimates with regard to real-world outcomes but with a slightly better survival rate and higher costs than in real life.