Abstract
Following initial optimism regarding the potential for rapid vaccination, delays and shortages in vaccine supplies have occurred in many countries. Various strategies to counter this reality and speed up vaccination have been set forth, of which the most popular approach has been to delay the second vaccine dose for a longer period than originally recommended by the manufacturers. Using three different epidemic models calibrated to the COVID-19 epidemic in the Czech Republic, we explore when delaying the second vaccine dose from 21 days to 42 days is advantageous. We find that vaccine mode of action, epidemic severity, and vaccine supply rate are the primary determinants of the optimal strategy.
Introduction
In early 2021, global vaccine shortages forced public health officials to make a difficult choice: adhere to the manufacturer-recommended 21-day interval for maximum individual protection or delay the second dose to provide a "partial" first-dose shield to a larger portion of the population (dose-sparing). While early, simplified models often suggested dose-sparing was generally superior, this study argues that the decision depends on nuanced factors such as the vaccine's "mode of action"—whether it primarily prevents infection (leaky transmission) or prevents severe symptoms and death.
Methodology
To ensure robust results, the researchers employed a "triangulation" approach using three distinct SEIR-type models calibrated via Approximate Bayesian Computation (ABC) to Czech Republic data:
- Model H: A deterministic compartmental model focused on hospital dynamics (ICU and common bed capacity).
- Model M: A high-resolution agent-based model simulating 56,000 individuals across 30 layers of social contacts.
- Model F: A stochastic, discrete-time model accounting for both visible (tested) and hidden (asymptomatic) infections.
These models integrated real-world epidemiological data from the Institute of Health Information and Statistics (UZIS), sociological contact rate surveys from PAQ Research, and meta-analyses of BNT162b2 (Pfizer/BioNTech) efficacy.
The "Mode of Action" and Epidemic Context
A critical insight of this research is that vaccine effects are multiplicative. If a vaccine reduces infection risk and then further reduces the risk of death among those infected, the total benefit is the product of these reductions.
The study found that the optimal strategy pivots based on where the vaccine acts in the infection course:
- Early-stage action: If the vaccine is highly effective at preventing contagion, the standard 21-day interval is often better as it shuts down transmission chains more rapidly.
- Late-stage action: If the vaccine primarily prevents death rather than infection, delaying the second dose to protect more people from fatal outcomes becomes the winning strategy.
Key Results
Using "deaths averted in the 65+ age group" as the primary metric, the study identified clear thresholds for decision-making:
- When to Delay (42 days): This strategy is preferable when the vaccine primarily prevents severe symptoms/death, the epidemic is severe (high contact rates), and vaccine supply is low. In these high-pressure scenarios, the mathematical benefit of protecting the greatest number of people with at least one dose is overwhelming.
- When to Stick to Schedule (21 days): This is preferable during mild epidemics with sufficient vaccine supply, particularly if the vaccine is effective at stopping transmission.
- The 50% Efficiency Threshold: For the 42-day delay to be beneficial in most scenarios, the first dose must provide at least 50% efficacy.
Discussion and Limitations
While the models are robust, they assume that vaccine protection does not significantly wane between day 21 and day 42. Furthermore, the study does not account for "immune escape," where a large population with partial immunity might encourage the evolution of vaccine-resistant variants. Finally, while the findings provide a strong framework, they are calibrated to the specific population structure and contact patterns of the Czech Republic.
In conclusion, there is no universal "right" interval; the decision to delay the second dose must be continuously re-evaluated based on the current epidemic wave's severity and the specific clinical profile of the vaccines in use.
Summary updated