Dynamic Vaccine Allocation for Control of Human-Transmissible Disease

Area of Science:

• Epidemiology
• Mathematical Modeling
• Public Health Policy

Background:

• Vaccine supply limitations during pandemics, like COVID-19, necessitate strategic allocation.
• Effective vaccine distribution is crucial for mitigating disease spread and mortality.
• Existing allocation strategies may not fully optimize outcomes based on dynamic epidemiological factors.

Purpose of the Study:

• To develop and apply a dynamic, age-structured compartmental model for optimizing COVID-19 vaccine allocation in the U.S.
• To minimize a weighted sum of deaths and cases through a data-driven allocation strategy.
• To compare model-based optimal allocation with actual vaccine distribution patterns.

Main Methods:

• Utilized a time-sensitive, age-structured compartmental model for disease dynamics.
• Employed Sequential Least Squares Quadratic Programming to determine optimal allocation strategies.
• Segmented allocation by region (50 U.S. states), age groups, and timeframe (December 2020 - June 2021).

Main Results:

• An optimized vaccine allocation could have averted an estimated 1.8 million cases and 9,000 deaths in the U.S.
• Prioritizing case reduction suggests allocating vaccines to younger individuals (17 and under) due to their higher transmission potential.
• Prioritizing death reduction indicates allocating vaccines to older individuals, who are more susceptible to severe disease.

Conclusions:

• A dynamic, optimized vaccine allocation strategy can significantly improve public health outcomes during pandemics.
• The optimal allocation strategy depends on the specific public health objective (case vs. death reduction).
• The developed methodology is generalizable to other infectious diseases, offering a framework for future epidemic preparedness.