The global demand and potential public health impact of oral antiviral treatment stockpile for influenza pandemics

Antiviral drugs are among the few countermeasures available during the critical interval between the emergence of a novel influenza pandemic and vaccine availability. Antiviral stockpiling is a key pandemic preparedness measure, yet existing stockpiling estimates vary widely and rest on outdated assumptions about healthcare-seeking behaviour and drug-specific effectiveness – limitations that the COVID-19 pandemic and recent clinical trial evidence have made untenable. We developed a multi-scale transmission model incorporating heterogeneous healthcare-seeking behaviour and the first direct clinical estimates of antiviral transmission risk reduction to estimate country-specific demand and mortality impact across four pandemic scenarios in 186 countries. We find that baloxavir marboxil (BXM), due to its transmission-reducing potential, could avert 37– 68% of mean pandemic deaths in the first epidemic wave, approximately double the impact of oseltamivir, while requiring a mean stockpile approximately 5–10% smaller (7–34% of the population, compared to 28–36% for oseltamivir). Uncertainty in viral load dynamics and transmission reduction benefits from clinical trials means that BXM’s impact could vary, but sensitivity analyses consistently suggest that BXM is likely to be more effective than oseltamivir. Under limited drug availability, priority should be given to treatment over post-exposure prophylaxis. Although drug rationing for high-mortality populations (e.g. elderly) can substantially reduce BXM demand, doing so leads to greater total pandemic deaths. Critically, each week of delay in initiating antiviral distribution erodes impact by up to 3% of averted deaths, meaning that antiviral stockpiles must be accompanied by rapid deployment infrastructure to deliver their potential impact.