Objective To estimate potential costs and effects of building an oseltamivir stockpile both for treatment and prophylaxis for influenza.

Methods Costs and potential effects of oseltamivir were calculated for several different scenarios in a decision analytic model. In the least severe scenario, the number and of influenza cases was about the same as during a “normal” winter-epidemic with a case fatality rate of 0.1% and attack-rate of 15%. In the worst scenario, the number of cases was 2.7 million (in a population of 4.5 mill) and the fatality rate close to what has been observed for the recent H5N1 avian influenza cases in SE Asia (50%). We assumed that future pandemics would be as frequent as they were in the 20th century (3 per 100 year). The cost of 5 days of treatment with 2x75 mg of oseltamivir was assumed to be NOK 99 while the cost of 6 weeks prophylaxis with 1x75 mg oseltamivir would be NOK 415.

Results Treatment for every Norwegian being infected by the pandemic virus was estimated to imply a cost below NOK 50,000 (€6,000) per life-year gained, regardless of attack-rate and fatality rate. Prophylaxis to the entire Norwegian population would imply a cost per life year gained of less than NOK 305,000 (€ 36,500) per life-year gained if the attack-rate were above 35% and the fatality rate above 0.1%.

Estimated average longevity gain per Norwegian inhabitant was up to 4 years three months for treatment and 6 years 10 months for prophylaxis.

Discussion Oseltamivir was considered to be cost-effective under a range of possible scenarios, both as treatment and prophylaxis. Based on this analysis the Norwegian government has increased the Norwegian stockpile of oseltamivir to cover 30% of the population with 5 days of treatment..