Antiviral drugs are an effective means to improve the recovery rate and reduce mortalities of people with influenza if the virus is sensitive to the drug. Under widespread use for prophylaxis and treatment with the risk of resistant variants arising, the consequences and cost-effectiveness of antivirals are less certain. The objective of this study was to identify the antiviral strategies that remain effective in both life and cost terms, when the characteristics of an emerging pandemic are unknown.
Methods and Findings:
A numerical model incorporating antiviral treatment and post-exposure prophylaxis, resistance, and asymptomatic infections was used to investigate a wide variety of pandemics. Scenarios with various fitnesses (transmissibilities) of the resistant strain, as well as the rate at which the resistant strain arises and the case fatality ratio of the mutant strain were analysed to determine the antiviral strategy which results in the best, cost-effective average outcome under all scenarios. While more antiviral use for treatment and prophylaxis results in a greater reduction in the mortalities on average, the variance becomes quite large with the risk of the resistant strain dominating the sensitive strain and the cost-effectiveness is greatly reduced. However, high levels of treatment (greater than 50%) with moderate levels of prophylaxis (less than 30% of the population) result in up to
a 50% reduction in mortalities and QALYs lost while remaining cost-effective. The optimal policy was found to be 100% treatment with 5% of the population receiving post-exposure prophylaxis.
The consequences of antiviral resistance are generally outweighed by the benefits. The optimal policy of 100% treatment of those symptomatically infected with 5% of the population, results in significant cost-effective reduction in mortalities and QALYs lost. At lower levels of treatment, greater post-exposure prophylaxis use is required to achieve optimal outcomes.