Evaluating targeted COVID-19 vaccination strategies with agent-based modeling

Area of Science:

• Epidemiology
• Public Health
• Computational Modeling

Background:

• COVID-19 vaccine distribution strategies significantly impact public health outcomes.
• Understanding the comparative effectiveness of different distribution approaches is crucial for pandemic response.

Purpose of the Study:

• To evaluate and compare the effectiveness of various COVID-19 vaccine distribution strategies.
• To identify optimal strategies for reducing severe outcomes like hospitalizations and deaths.

Main Methods:

• Utilized an agent-based model of human activity and COVID-19 transmission.
• Calibrated the model using detailed epidemiological data from Florida, USA (cases, hospitalizations, deaths, seroprevalence, breakthrough infections).
• Simulated four distribution strategies under varying vaccine supply levels from late 2020 to early 2022.

Main Results:

• Targeting individuals at high risk of severe disease emerged as the most effective strategy for reducing severe outcomes across all scenarios.
• Ring vaccination, while effective for infection reduction, was least effective for preventing deaths.
• Using age group as a proxy for disease risk also outperformed mass vaccination and ring vaccination.

Conclusions:

• Active targeting of high-risk populations is the optimal COVID-19 vaccine distribution strategy for minimizing severe outcomes.
• The relative benefit of targeting strategies varies with vaccine availability and specific viral waves (Alpha, Delta, Omicron).
• Despite variations in absolute benefit, the ranking of strategy performance in preventing severe outcomes remained consistent across different vaccine supply levels.