Review 1: "Prioritising COVID-19 Vaccination in Changing Social and Epidemiological Landscapes"

Jesus Gomez- Gardenes

doi:doi:10.1162/2e3983f5.4a6d11c2

Prioritising COVID-19 vaccination in changing social and epidemiological landscapes

by Peter Jentsch, Madhur Anand, and Chris T. Bauch

Show Description
dx.doi.org

Description

SummaryBackgroundDuring the COVID-19 pandemic, authorities must decide which groups to prioritise for vaccination. These decision will occur in a constantly shifting social-epidemiological landscape where the success of large-scale non-pharmaceutical interventions (NPIs) like physical distancing requires broad population acceptance.MethodsWe developed a coupled social-epidemiological model of SARS-CoV-2 transmission. Schools and workplaces are closed and re-opened based on reported cases. We used evolutionary game theory and mobility data to model individual adherence to NPIs. We explored the impact of vaccinating 60+ year-olds first; <20 year-olds first; uniformly by age; and a novel contact-based strategy. The last three strategies interrupt transmission while the first targets a vulnerable group. Vaccination rates ranged from 0.5% to 4.5% of the population per week, beginning in January or July 2021.FindingsCase notifications, NPI adherence, and lockdown periods undergo successive waves during the simulated pandemic. Vaccination reduces median deaths by 32% – 77% (22% – 63%) for January (July) availability, depending on the scenario. Vaccinating 60+ year-olds first prevents more deaths (up to 8% more) than transmission-interrupting strategies for January vaccine availability across most parameter regimes. In contrast, transmission-interrupting strategies prevent up to 33% more deaths than vaccinating 60+ year-olds first for July availability, due to higher levels of natural immunity by that time. Sensitivity analysis supports the findings.InterpretationFurther research is urgently needed to determine which populations can benefit from using SARS-CoV-2 vaccines to interrupt transmission.FundingOntario Ministry of Colleges and Universities.Research in contextEvidence before this studyWhether to vaccinate individuals who cause the most transmission or those who are at highest risk of death is relevant to prioritizing COVID-19 vaccination. We searched PubMed and medRxiv for the terms COVID19, vaccin*, model, and priorit* up to September 24, 2020, with no date or language restrictions. We identified 4 papers on mathematical models of COVID-19 vaccine prioritization that explored the conditions under which different age groups should be vaccinated first. We did not find any coupled social-epidemiological models that capture feedback between social dynamics and epidemic trajectories.Added value of this studyThe dynamic interaction between SARS-CoV-2 epidemics and the population response through scalable non-pharmaceutical interventions will continue to play a large role in the course of the pandemic, both before and after vaccines become available. Hence, social-epidemiological models may be useful. Our social-epidemiological model identifies the conditions under which COVID-19 deaths can be reduced most effectively by prioritizing older individuals first, versus other strategies designed to interrupt transmission. We explore how the best vaccination strategy varies depending on a wide range of socio-epidemiological and vaccine program parameters. We identify clear and interpretable conditions under which using COVID-19 vaccines to interrupt transmission can reduce mortality most effectively.Implications of all the available evidenceSeroprevalence surveys before the onset of vaccination could measure population-level SARS-CoV-2 immunity. In populations where seropositivity is high due to previous waves, vaccinating to interrupt transmission may reduce deaths more effectively than targeting older individuals. More research is urgently required to evaluate how to prioritise vaccination in populations that have experienced one or more waves of COVID-19.

RR:C19 Evidence Scale rating by reviewer:

Potentially informative. The main claims made are not strongly justified by the methods and data, but may yield some insight. The results and conclusions of the study may resemble those from the hypothetical ideal study, but there is substantial room for doubt. Decision-makers should consider this evidence only with a thorough understanding of its weaknesses, alongside other evidence and theory. Decision-makers should not consider this actionable, unless the weaknesses are clearly understood and there is other theory and evidence to further support it.

***************************************

Review:

Far from being extinct, COVID-19 pandemic keeps hitting the population and taking a huge humanitarian, social and economic toll worldwide. Recent promising advances in the design of vaccines shed a glimmer of hope to actively cut down its transmission via pharmaceutical interventions in 2021. However, the initial supply of vaccines will be limited and, in this scenario, an efficient vaccine deployment becomes indispensable in light of the foreseeable lack of resources.

In this manuscript, the authors analyze the performance of several vaccination strategies prioritising different age-groups by using an epidemiological model which accounts for the effect of both pharmaceutical and non-pharmaceutical interventions. Remarkably, they find that the optimal vaccine distribution across different age groups is not universal but strongly depends on when the vaccination campaign starts, thus highlighting the crucial role of previously acquired immunity over the different age groups. This finding stresses the need of determining the epidemiological seroprevalence existing in the population to decide the best vaccination strategy.

While the latter result is consistent throughout the manuscript, other important claims made by the authors should be taken with caution due to the limitations of the proposed model. Our main concern is that some of the results are strongly dependent on model assumptions, as revealed by the sensitivity analysis. For example, when removing seasonality from the equations or assuming uniform susceptibilities across different age groups, the authors find that prioritising young over elderly population is more efficient unlike the baseline scenario reported in the manuscript. Likewise, the large number of calibrated parameters in the model may give rise to overfitting issues. For example, the calibrated susceptibilities, reflecting that children are more susceptible to contracting the disease, contradict many recent studies suggesting the lower susceptibility of young population to SARS-CoV-2 [1,2].

Finally, although not very relevant during the first epidemic wave in most of the countries, test-trace and isolated policies are becoming more and more relevant as a tool to actively reduce the spread of COVID-19 by cutting possible transmission chains. This strategy can alter the results about the seroprevalence distribution across different age groups when vaccines become available.

In a nutshell, the manuscript presents a sophisticated and well-motivated socio-epidemiological model that captures most of the relevant factors determining the spread of SARS-CoV-2 and the relevance of non-pharmaceutical contention measures. Although the model could be improved and results can vary by removing some assumptions, it shows that an optimal vaccination strategy crucially depends on the precise time when this pharmaceutical intervention becomes available.

[1] Davies, N.G., Klepac, P., Liu, Y. et al. Age-dependent effects in the transmission and control of COVID-19 epidemics. Nat Med 26, 1205–1211 (2020). https://doi.org/10.1038/s41591-020-0962-9

[2] Viner RM, Mytton OT, Bonell C, et al. Susceptibility to SARS-CoV-2 Infection Among Children and Adolescents Compared With Adults: A Systematic Review and Meta-analysis. JAMA Pediatr. Published online September 25, 2020. doi:10.1001/jamapediatrics.2020.4573