JMIR Publications

ABSTRACT

Social distancing behavior has been a critical non-pharmaceutical measure for mitigating the COVID-19 pandemic. For this reason, there has been widespread interest in the factors determining social distancing violations, with a particular focus on individual-based factors. In this paper, we examine an alternative, less appreciated, and scalable indicator of social distancing violations, the “situational opportunity” for keeping interpersonal distance in crowded settings. Data were a large body of video clips of public places recorded by municipal surveillance cameras throughout the first year of the pandemic. Using a computer vision algorithm to automatically recognize pedestrian presence and behavior, we recorded social distancing violations of more than half a million individuals in more or less crowded street contexts. Results showed a close positive association between crowding and social distancing violations. This relationship indicates that potential transmission situations can be identified by simply counting the number of people present in a location. Our findings thus provide a tool for epidemiologist to easily incorporate real-life behavior in predictive models of airborne contagious diseases. Our findings, furthermore, suggest that scholars and public health agencies should appreciate the situational basis of social distancing violations afforded by people crowding in public settings.