Predicting COVID-19 outbreak using open mobility data for minimal disruption on the country’s economy

Abstract

COVID-19 is an infectious disease caused by the SARS-COV-2 coronavirus, which was discovered in late 2019. Within a few months, COVID-19 was declared a global pandemic by the WHO. Several countries adopted social distancing measures, such as self-quarantine, workplace and mobility restrictions, reducing the probability of contact between non-infected and infected people. In general, these measures have a negative impact on low-income economies and small and medium businesses. During the outbreak, several predictive models have been proposed in order to assess the level of saturation that health services might have. Nevertheless, none of them considers information on the people’s mobility to assess the effectiveness of the social distancing policies. In this study, the authors propose a prediction method based on people’s open mobility data from Apple© and Google© databases to project potential scenarios and monitor case growth. The proposed method shows the importance of monitoring daily case increase for the first 4-6 weeks of the pandemic wave. Active monitoring is crucial to determine the reduction in mobility and proper actions. The results can contribute to health authorities for making timely decisions, preventing the spread of viruses while balancing the reduction of mobility with minimal disruption in people’s economies in future outbreaks.