Social heterogeneity drives complex patterns of the COVID-19 pandemic: insights from a novel Stochastic Heterogeneous Epidemic Model

Maltsev, Alexander V; Stern, Michael D

As of August 2020, it has become evident that regional infection curves of COVID-19 exhibit complex patterns which often differ from curves predicted by forecasting models. We hypothesized that this may be due to social heterogeneity not accounted for by regional models. Here we present a new Stochastic Heterogeneous Epidemic Model (SHEM) to investigate the role of heterogenous societal structure. SHEM is intended to be a general tool with which to explore scenarios and determine the expected consequences of various interventions. We represent a society by an arbitrary network of sub-populations that could represent social as well as geographical strata. We created several scenarios with large clusters of people with R0 of COVID-19 interacting with multiple smaller local clusters that have larger internal R0. We find that isolation or embedding of these vulnerable sub-clusters generate complex infection patterns which includes multiple peaks and growth periods, an extended plateau, a prolonged tail, or a delayed second wave of infection, which may or may not form due to stochasticity. We also show that local clusters can either be driving or driven forces in infection progression. Embedded vulnerable groups become hotspots that drive infection despite efforts of the main population to socially distance, while isolated areas suffer delayed but intense infection. Social heterogeneity is a key factor in the formation of complex infection curves. Vulnerable subgroups that cannot implement mitigation strategies can spread infection to socially distanced populations, defeating mitigations. This implies that mitigation of vulnerable groups is essential to control the epidemic.