International Vaccine Allocation: An Optimization Framework

Abstract

As observed during the COVID-19 pandemic, high-income countries, such as the U.S., may exhibit vaccine nationalism during a pandemic: stockpiling doses of vaccine for their own citizens and being reluctant to distribute doses of the vaccine to lower-income countries. While many cite moral objections to vaccine nationalism, vaccine inequity during a pandemic could possibly worsen the global effects of the pandemic, including in the high-income countries themselves, through the evolution of new variants of the virus. This paper uses the COVID-19 pandemic as a case study to identify scenarios under which it might be in a high-income nation's own interest to donate vaccine doses to another country before its own population has been fully vaccinated. We develop an extended SEIR (susceptible-exposed-infectious-recovered) epidemiological model embedded in an optimization framework and examine scenarios involving a single donor and multiple recipient (nondonor) geographic areas. We find that policies other than donor-first can delay the emergence of a more-contagious variant compared to donor-first, sometimes reducing donor-country deaths in addition to total deaths. Thus, vaccine distribution is not a zero-sum game between donor and nondonor countries: an optimization approach can achieve a dramatic reduction in total deaths with only a small increase in donor-country deaths. The iterative linear programming approximation approach we develop can help confirm those instances when a priority policy is optimal and, when not optimal, can identify superior policies. This optimization framework can be used to guide equitable vaccine distribution in future pandemics.