Impacts of detection and contact tracing on the epidemic spread in time-varying networks

Abstract

In the situation of insufficient vaccines and rapid mutation of the virus, detection and contact tracing have been argued to be effective interventions in the containment of emergent epidemics. However, most of previous studies are devoted to data-driven, leading to insufficient understanding of quantifying their effectiveness, especially when individuals’ interactions evolve with time. Here, we aim at quantifying the effectiveness of detection and contact tracing interventions in suppressing the epidemic in time-varying networks. We propose the Susceptible-Exposed-Infected-Removed-Dead-Hospitalized (SEIRDH) model with detection and contact tracing. Under the framework of time-varying networks and with a mean-field approach, we analyze the epidemic thresholds under different situations. Experimental results show that detection can effectively suppress the epidemic spread with an increased epidemic threshold, while the role of tracing depends on the characteristics of the epidemic. When an epidemic is infectious in the incubation period, contact tracing has an obvious effect in suppressing the epidemic spread, but not when the epidemic is not infectious in the incubation. Thus, we apply this framework in real networks to explore possible contact tracing measures by taking use of individuals’ properties. We find that contact tracing based on activity and historical information is more efficient than random contact tracing. Moreover, individuals’ attractiveness and aging effects also affect the efficiency of detection and contact tracing. In conclusion, making full use of individuals’ properties can remarkably improve the effectiveness of detection and contact tracing. The proposed method is expected to provide theoretical guidance for coping with the COVID-19 or other emergent epidemics.