Evaluating strategies for spatial allocation of vaccines based on risk and centrality

Abstract

When vaccinating a large population in response to an invading pathogen, it is often necessary to prioritize some individuals to be vaccinated first. One way to do this is to choose individuals to vaccinate based on their location. Methods for this prioritization include strategies that target those regions most at risk of importing the pathogen, and strategies that target regions with high centrality on the travel network. We use a simple infectious disease epidemic model to compare a risk-targeting strategy to two different centrality-targeting strategies based on betweenness centrality and random walk percolation centrality, respectively. We find that the relative effectiveness of these strategies in reducing the total number of infections varies with the basic reproduction number of the pathogen, travel rates, structure of the travel network and vaccine availability. We conclude that when a pathogen has high spreading capacity, or when vaccine availability is limited, centrality-targeting strategies should be considered as an alternative to the more commonly used risk-targeting strategies.

Identifying highly connected sites for risk-based surveillance and control of cucurbit downy mildew in the eastern United States

“…The observation that BWC was more informative of node importance than other centrality measures emphasizes the need to generate centrality measures that are specific to the disease of interest ( Holme, 2018 ). Invariably, different centrality measures can result in a dissimilar ranking profiles of important nodes for diverse pathosystems, possibly due to the inherent differences in the underlying mechanisms of pathogen dispersal and disease spread, landscape connectivity, or other factors ( Dudkina et al, 2023 ; Holme, 2018 ; Singer, Thompson & Bonsall, 2022 ).…”

“…Where resources available for control are limited, targeting nodes with high BWC for treatment has also been found to be an effective strategy in impeding epidemics caused by a disease that spreads rapidly ( Singer, Thompson & Bonsall, 2022 ). The most central nodes identified as important based on BWC were sites in Michigan in the Great Lakes region, Ohio in the Midwest, and Maryland, North Carolina, South Carolina, and Virginia along the mid-Atlantic coast.…”

Identifying target areas for risk-based surveillance and control of transboundary animal diseases: a seasonal analysis of slaughter and live-trade cattle movements in Uganda

“…Our percolation analysis supports the use of centrality-targeting to reduce the size of the largest connected component and can inform district prioritisation for animal health resource allocation (e.g., personnel, vaccines, etc.). Centrality-targeting is one of the spatial strategies used for strategic implementation of disease control interventions and has shown to be superior to risk-targeting when seeking to reduce epidemic contagion, especially in resource-constrained areas affected by epidemics caused by highly transmissible pathogens 61 , 62 , as is the case of FMD and other TADs affecting livestock 63 . For instance, spatial vaccination strategies informed by centrality-based prioritisation have shown to be effective at halting epidemic spread by reducing the size of the largest connected component, total number of individuals infected and the maximum number of simultaneous infections during an outbreak 64 .…”

Implications of COVID-19 vaccination heterogeneity in mobility networks

“…Existing studies such as [11] also propose targeting central locations to substantially reduce transmission; however, they have examined this empirically at a resolution several orders of magnitude coarser than this work which covers over 200,000 CBGs across the US.…”

“…Our study investigates the effect of vaccination heterogeneity through largescale epidemic simulations on the US mobility network. Departing from highly aggregated models to understand vaccination performance [8][9][10][11], we employ a data-driven approach to study the impact of spatial vaccination heterogeneity. Specifically, we leverage fine-grained human mobility, vaccination, and census data in the US, along with an epidemiological model [12][13][14] to illustrate how different hypothetical vaccination distributions can lead to largely different country-wide outcomes.…”