How many lives would vaccine sharing have saved?

By Sam HancockReviewed by Aimee MolineuxJan 31 2022

There are extremely wide disparities in worldwide vaccine coverage for coronavirus disease 2019 (COVID-19), with developing countries tending to lag significantly behind developed countries.

As many governments are unwilling to share vaccines before their entire population is vaccinated, many vaccines have gone to less at-risk individuals due to their location, with some countries even providing nationwide 'booster' vaccinations before some countries have reached 5% vaccination.

Researchers from the University of Warwick have been investigating different vaccine sharing simulations, and how they would affect global mortality.

This news article was a review of a preliminary scientific report that had not undergone peer-review at the time of publication. Since its initial publication, the scientific report has now been peer reviewed and accepted for publication in a Scientific Journal. Links to the preliminary and peer-reviewed reports are available in the Sources section at the bottom of this article. View Sources

A preprint version of the researchers’ study can be found on the medRxiv* preprint server as the paper undergoes peer review.

The study

The researchers developed a model of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) transmission and the outcome of infection, and applied it to 152 different countries with different parameters to reflect changing demographics. Countries were stimulated using an age-structured compartmental infection model coupled with global evolution of new variants and sharing of vaccines when national conditions are met.

The countries were partitioned into four income classifications given by the world bank, and age-based heterogonies in disease outcome were captured by stratification of populations into age groups of 5 years each, based on country-level data. Each of these groups had their own parameters for risk of severe disease, symptoms, and susceptibility.

Epidemiologically relevant contacts within each country are varied to allow for different levels of non-pharmaceutical interventions. Parameters for national control measures and death rates were fitted using daily estimates for infections and deaths. The researchers were worried about inaccuracies of official death rates, and so based their models on estimates using excess mortality statistics.

The researchers created several different scenarios for vaccine sharing but applied the same assumptions to all of them. They assumed that all countries were aiming to achieve a 90% uptake of vaccine in over 60s, and an 80% uptake in all over 12s.

Vaccination protection assumptions were based on efficacy characteristics similar to those observed for the AstraZeneca vaccine, with protection against susceptibility, onward transmission, probability of symptoms and severity of disease as measured by hospitalizations and deaths. The likelihood of vaccine protection waning over time was also included in the analysis.

The first scenario that was created reflected the historical delivery of vaccines, with little to no sharing. This was followed with a similar scenario named the 2-dose threshold, in which the current pattern continued until every country vaccinated the declared threshold of the population with two doses, and then shared every following vaccine delivery between other countries proportioned based on the number of unvaccinated individuals remained.

The next scenario, 40+ threshold was similar, except vaccine sharing began once a 2-dose vaccination was completed for all over 40s. The 65+ threshold scenario followed the same pattern for over 65s, while the full sharing scenario began at the start of 2021 with every vaccine delivery pooled and equally distributed based on the number of unvaccinated individuals remaining in each country.

Emerging variants were also taken into account, with the scientists considering different rates of transmission for different variants and the global proportion of said variants over time. Older variants have historical data which can help with the accuracy of this, but for variants more recently emerged than Omicron the researchers were forced to estimate the transmission. They then used the relationship between historic infections and reproductive number to realize the impact of different infection levels on variant emergence.

At the time the paper was initially written, 50% of the global population was vaccinated, with the vaccination rate in individual countries varying wildly from 75-2%. In their simulations, the researchers observed that increased vaccine sharing would likely have had beneficial effects in all but the highest income countries, with infections reduced by 11.5, 5.0 and 2.4% in low, lower-middle, and higher middle-income countries, respectively in the full sharing scenario. High-income countries, if they had continued to reduce non-pharmaceutical interventions, would likely have seen 17.2% more infections.

However, when the researchers included extended behavioral changes due to reduced vaccination into the plan, infection and mortality were cut drastically worldwide. The changes in mortality, however, would have been significantly lower in high-income countries as the vast majority of at-risk individuals were already vaccinated - the scientists estimate that globally there would have been 22 fewer deaths per 100,000 individuals. In the other sharing scenarios, these benefits were significantly curtailed, as the increased time before vaccine sharing led to more infections in lower-income countries.

Conclusion

The authors have successfully shown that increased vaccine sharing would have reduced the level of global mortality due to COVID-19, and their findings have been supported by multiple other studies with similar conclusions. Alongside this, many have suggested that increased vaccination in developing countries could help cut the risk of new variants emerging- many of which show some ability to evade vaccine-induced immunity. Hopefully, governments will begin to take note of these findings - although there has been little evidence of that so far.

This news article was a review of a preliminary scientific report that had not undergone peer-review at the time of publication. Since its initial publication, the scientific report has now been peer reviewed and accepted for publication in a Scientific Journal. Links to the preliminary and peer-reviewed reports are available in the Sources section at the bottom of this article. View Sources