Hotspot model shows how location-based superspreading accelerates and reshapes epidemics

During outbreaks of many diseases, a small number of infected individuals are responsible for a disproportionately large number of new infections in what are called superspreading events (SSEs). SSEs broadly fall into four categories: (1) a single individual is more infectious due to biological differences in their infection or (2) their greater degree of social connection; or (3) the disease spreads more readily in certain high-risk facilities or (4) “opportunistic” situations such as large gatherings. Existing modeling approaches work well to understand the first two of these, but are not well suited to describe the dynamics in the latter two. Here we introduce a simple agent-based model which captures the essential features of disease spreading more readily at high-risk locations or gatherings, which we call “hotspots”. In our model, disease spreads and people recover as in a standard Susceptible, Infected, Recovered (SIR) model, but agents are also characterized by individual probability of visiting the hotspot where disease spreads much more readily, providing an additional risk structure to the population. We use this model to investigate how an outbreak's probability, peak, and final size all vary under different risk heterogeneity assumptions. We show how some particular distributions of risk-taking behavior across the population heighten these effects. We complement our simulations with analytic results that provide theoretical bases for all of our numerical results and allow for robust interpretation and prediction.