The number of vaccine doses administered has been increasing on a daily basis since COVID-19 vaccine became available in December 2020. Vaccination rates have been boosted by community-based organisations, government partners, call centres, special events and advertising. New York City also implemented a large canvassing operation, the Test & Trace Corps, to reach people without access to vaccination sites.
Vaccine prioritisation strategies are essential for optimising the use of limited vaccine supplies in an age-balanced way. Several modelling studies (Bubar et al., 2021; Buckner et al., 2021; Guerstein et al., 2021; Matrajt et al., 2021) have shown that focusing on population groups with higher mortality or more interaction rates may reduce total fatalities.
Our work aims to evaluate and compare different prioritisation strategies using a SEIRD-type model segmented by population age groups that takes into account the impact of behavioural parameters such as mask-wearing and social distancing on the probability of infection. The model is used to simulate the distribution of vaccines under various scenarios where a certain amount of doses is available for each population group at different rates of vaccination rollout.
Results show that a prioritisation strategy targeting populations with higher mortality is more efficient than other strategies, particularly when vaccination rollout is fast and coverage high, as it leads to significant reductions in total fatalities. When vaccination rollout is slow prioritising groups with lower interaction rates leads to only modest reductions in total fatalities. Finally, a dynamic approach where each population group is called one time and then skipped until coverage has been reached yields similar results to all other strategies evaluated and performs closer to the best possible permutations of single-call strategies by age or interactions.