Pharmacoeconomics of antiviral therapy for influenza in a Japanese hospital setting

Abstract

Objectives: We constructed a cost-effectiveness decision model to determine a hypothetical 'best treatment' pathway for patients presenting at our institution with influenza virus infection when the choice of treatment was either oseltamivir, zanamivir or a control therapy not active against influenza.

Methods: The decision model was constructed using DATA 3.5 for evaluating the cost-effectiveness analysis of neuraminidase inhibitors from the perspective of the healthcare payer. The time horizon was set at 14 days based on the general duration of influenza infection in Japan. Clinical outcomes were mainly derived from reports and guidance published by the National Institute for Clinical Excellence in the UK. Japan-specific cost parameters incorporated into the decision model were taken from the Medical Fee Point Survey conducted at St Luke's International Hospital in accordance with medical fee receipts kept at our institution. The study included four professionals and a supporter who gathered information required for the analysis.

Results: In otherwise healthy adults, cost savings of yen831.6 (approximately $US6.72; 2002 values) in the oseltamivir group and an increment in cost of yen40.5 (approximately $US0.33) in the zanamivir group were achieved in comparison with the control group. In contrast, an incremental cost of yen288.4 (approximately $US2.33) was incurred in the oseltamivir group versus the control group when at-risk patients were assessed, but cost savings of yen159.8 (approximately $US1.29) were achieved in the zanamivir group. As a result of cost-effectiveness and cost-utility analyses in otherwise healthy adults, oseltamivir dominated the control therapy because cost savings in the oseltamivir group were made. In the zanamivir group the cost was incremental and the Incremental Cost-Utility Ratio (ICUR) compared with the control group was about yen13 000 (approximately $US107.34)/quality-adjusted life-year (QALY) gained. As a result of cost-effectiveness and cost-utility analyses in at-risk patients, in the oseltamivir group the cost was incremental and the ICUR compared with the control group was about yen230 000 (approximately $US2138.77)/QALY gained. As cost savings were made, zanamivir dominated the control therapy.

Conclusion: While the cost effectiveness (from the perspective of a healthcare payer) of the neuraminidase inhibitors was superior to that of the control group in the treatment of otherwise healthy adults with influenza in our study, it seemed necessary to take other factors into consideration before recommending one agent over the other as a first-line therapy. On the other hand, we suggest that zanamivir is the drug of choice for use in at-risk patients, and we recommend, in the light of our results, that if zanamivir is not available another therapy should be given rather than oseltamivir. Since with influenza infections deaths and hospitalisations of at-risk patients impact on the Japanese community, decision-making on the appropriate therapy should take into account the particular patient group involved.