Authors: David Lyver, Mihai Nica, Corentin Cot, Giacomo Cacciapaglia, Zahra Mohammadi, Edward W. Thommes, Monica-Gabriela Cojocaru

Abstract: The epidemiology of pandemics is classically viewed using geographical and political borders; however, these viewpoints can result in a misunderstanding of the epidemiological state within a given region. To bring a fresh new viewpoint into focus, we develop a clustering algorithm which is capable of recasting geographical regions (such as states or provinces) into subpopulation clusters with a high level of inter-cluster mobility, while minimizing the population flows towards other clusters. In order to demonstrate the capabilities of this algorithm, we use counties of the USA and cellular mobility data to recast the USA into well-mixed clusters. Herein, we show a more granular spread of SARS-CoV-2 throughout the first weeks/months of the pandemic. Two insights arise directly from this method: 1) the absolute importance of importation cases via flights; 2) the clear higher weight of urban vs. rural areas; 3) that state level incidence information can be misleading in the early days/weeks of a pandemic spread.

From a policy perspective, we are able to identify areas (groups of counties) that were experiencing above average levels of transmission within a given state, as well as pan-state areas (clusters overlapping more than one state) with very similar disease spread. Therefore, our method enables policymakers to make more informed decisions on the use of public health interventions within their jurisdiction, as well as guide collaboration with surrounding regions to benefit the general population in controlling the spread of communicable diseases.