Retrospective Analysis of Equity-Based Optimization for COVID-19 Vaccine Allocation

Abstract

Marginalized racial and ethnic groups in the United States were disproportionally affected by the COVID-19 pandemic. To study these disparities, we construct an age-and-race-stratified mathematical model of SARS-CoV-2 transmission fitted to age-and-race-stratified data from 2020 in Oregon and analyze counterfactual vaccination strategies in early 2021. We consider two racial groups: non-Hispanic White persons and persons belonging to BIPOC groups (including non-Hispanic Black persons, non-Hispanic Asian persons, non-Hispanic American Indian or Alaska Native persons, and Hispanic or Latino persons). We allocate a limited amount of vaccine to minimize overall disease burden (deaths or years of life lost), inequity in disease outcomes between racial groups (measured with five different metrics), or both. We find that, when allocating small amounts of vaccine (10% coverage), there is a trade-off between minimizing disease burden and minimizing inequity. Older age groups, who are at a greater risk of severe disease and death, are prioritized when minimizing measures of disease burden, and younger BIPOC groups, who face the most inequities, are prioritized when minimizing measures of inequity. The allocation strategies that minimize combinations of measures can produce middle-ground solutions that similarly improve both disease burden and inequity, but the trade-off can only be mitigated by increasing the vaccine supply. With enough resources to vaccinate 20% of the population the trade-off lessens, and with 30% coverage, we can optimize both equity and mortality. Our goal is to provide a race-conscious framework to quantify and minimize inequity that can be used for future pandemics and other public health interventions.