Optimizing global COVID-19 vaccine allocation: An agent-based computational model of 148 countries

Area of Science:

• Epidemiology
• Computational modeling
• Public health policy

Background:

• The World Health Organization (WHO) and COVAX utilize mathematical optimization for vaccine allocation, prioritizing equity and effectiveness.
• This study addresses the challenge of optimizing global vaccine distribution strategies.

Purpose of the Study:

• To solve the vaccine allocation optimization problem using agent-based simulations.
• To evaluate the impact of different vaccine allocation strategies on pandemic outcomes.

Main Methods:

• Developed open-sourced agent-based models simulating virus transmission across 198 million people in 148 countries.
• Compared a baseline scenario (current vaccine progress) with scenarios involving minimum vaccination rates and allocation based on pandemic levels.
• Fitted simulation models using global pandemic data from January 2020 to June 2021.

Main Results:

• Under the baseline, projections indicated 24.36 million new cases and 468,945 deaths in three months.
• Achieving 26% population vaccination required 5 million additional daily doses, reducing cases by 3.32 million.
• Allocating 5 million doses based on projected cases averted over six times more cases (9.20 million) compared to current global distribution (1.45 million).

Conclusions:

• Global COVID-19 vaccine distribution can be optimized for enhanced equity and effectiveness.
• Alternative allocation strategies, particularly those based on pandemic levels, can avert significantly more cases and deaths.
• COVAX can implement alternative allocation strategies to reduce cross-country inequity and improve global health outcomes.