Background COVID-19 was first detected in Wuhan, China, in 2019 and spread worldwide within a few weeks. The COVID-19 epidemic started to gain traction in France in March 2020. Subnational hospital admissions and deaths were then recorded daily and served as the main policy indicators. Concurrently, mobile phone positioning data have been curated to determine the frequency of users being colocalized within a given distance. Contrarily to individual tracking data, these can be a proxy for human contact networks between subnational administrative units. Methods Motivated by numerous studies correlating human mobility data and disease incidence, we developed predictive time series models of hospital incidence between July 2020 and April 2021. We added human contact network analytics, such as clustering coefficients, contact network strength, null links or curvature, as regressors. Findings We found that predictions can be improved substantially (by more than 50%) at both the national level and the subnational level for up to 2 weeks. Our subnational analysis also revealed the importance of spatial structure, as incidence in colocalized administrative units improved predictions. This original application of network analytics from colocalization data to epidemic spread opens new perspectives for epidemic forecasting and public health.
Keywords: human mobility; infectious disease; networks; time series.
Copyright © 2021 The Author(s). Published by Elsevier Ltd.. All rights reserved.