A mechanistic spatio-temporal modeling of COVID-19 data

Abstract

Understandingtheevolutionofanepidemicisessentialtoimplementtimelyandefficient preventive measures. The availability of epidemiological data at a finespatio-temporal scale is both novel and highly useful in this regard. Indeed, hav-ing geocoded data at the case level opens the door to analyze the spread of thedisease on an individual basis, allowing the detection of specific outbreaks or, ingeneral, of some interactions between cases that are not observable if aggregateddata are used. Point processes are the natural tool to perform such analyses. Weanalyze a spatio-temporal point pattern of Coronavirus disease 2019 (COVID-19)cases detected in Valencia (Spain) during the first 11 months (February 2020 toJanuary 2021) of the pandemic. In particular, we propose a mechanistic spatio-temporal model for the first-order intensity function of the point process. Thismodel includes separate estimates of the overall temporal and spatial intensitiesof the model and a spatio-temporal interaction term. For the latter, while similarstudies have considered different forms of this term solely based on the physicaldistances between the events, we have also incorporated mobility data to bettercapture the characteristics of human populations. The results suggest that therehas only been a mild level of spatio-temporal interaction between cases in thestudy area, which to a large extent corresponds to people living in the same res-idential location. Extending our proposed model to larger areas could help usgain knowledge on the propagation of COVID-19 across cities with high mobilitylevels.