Optimal vaccine allocation for COVID-19 in the Netherlands: A data-driven prioritization

Area of Science:

• Epidemiology
• Mathematical Biology
• Public Health

Background:

• Vaccination programs are essential for COVID-19 control but face vaccine supply limitations.
• Determining optimal vaccine allocation strategies is critical for maximizing public health impact.
• Existing mathematical modeling approaches can be computationally intensive due to numerous allocation schemes.

Purpose of the Study:

• To develop an algorithm for near-optimal vaccine allocation strategies.
• To identify allocation schemes that minimize specific outcomes like infections, hospitalizations, or deaths.
• To evaluate the performance of optimized strategies against other allocation methods.

Main Methods:

• Reconstruction of age-specific transmission intensity using an approximation method (next-generation matrix).
• Calculation of expected impact of subgroup vaccination based on incidence and force of infection.
• Evaluation of the proposed algorithm using simulated epidemics and real-world COVID-19 data from the Netherlands.

Main Results:

• Optimal vaccine allocation strategies are highly dependent on the specific public health objective (e.g., minimizing deaths vs. infections).
• A strategy optimized for minimizing deaths may not be efficient for minimizing infections.
• Optimized allocation strategies outperformed random, young-to-old, and old-to-young strategies in simulated epidemics.
• The Netherlands' old-to-young vaccination policy aligned with the strategy that minimizes deaths.

Conclusions:

• The proposed algorithm provides an effective method for determining near-optimal vaccine allocation strategies using surveillance data.
• This approach supports robust decision-making for intervention strategies by offering plausible simulation scenarios.
• Tailoring vaccine allocation to specific objectives is essential for efficient pandemic control.