Social network heterogeneity is essential for contact tracing

Contact tracing is suggested as an effective strategy for controlling an epidemic without severely limiting personal mobility. Here, we explore how social structure affects contact tracing of COVID-19. Using smartphone proximity data, we simulate the spread of COVID-19 and find that heterogeneity in the social network and activity levels of individuals decreases the severity of an epidemic and improves the effectiveness of contact tracing. As a mitigation strategy, contact tracing depends strongly on social structure and can be remarkably effective, even if only frequent contacts are traced. In perspective, this highlights the necessity of incorporating social heterogeneity into models of mitigation strategies. O_TEXTBOXSignificance StatementThe COVID-19 epidemic has put severe limitations on individual mobility in the form of lockdowns and closed national borders. Mitigation strategies permitting individual mobility while limiting disease spreading are needed, and contact tracing is a potentially effective example of such a strategy. We use smartphone proximity data to monitor contacts between people, and find that contact tracing is highly dependent on social structure, being very effective on real contact networks. This shows that mitigation of COVID-19 may be possible with contact tracing, and that epidemiological models must incorporate social network structure. C_TEXTBOX