Due to high contagion rate and rapid mutations of the SARS-CoV-2 virus, Covid-19 spread spatially throughout the Brazilian territory in 2020 and 2021. Factors related to population mobility may also have contributed to the increase in the number of cases. In this sense, this study aimed to evaluate the correlation between network centrality measures (NCMs) and the incidence rate of Covid-19 (INC) in cities in the region of São José do Rio Preto, located in the southeastern Brazil. Data on INC accumulated through July 10, 2021, were used (Brasil, 2021). The nodes (cities) of the network and the road connection (edges) between the nodes data were imported into Gephi software (Bastian et al., 2009) and then the following NCMs were calculated: degree (DEG), betweenness (BET), closeness (CLO) and eigenvector (EIN). Then, the correlation coefficients between the NCMs and INC were estimated. The results showed that the EIN measure presented the highest positive correlation with the INC of Covid-19 (r=0.377; p<0.001), followed by the DEG (r=0.284; p=0.003) and BET (r=0.190; p=0.049) measures. Cities with high EIN values are usually close to the best-connected cities in the network. This may explain the occurrence of higher INC values in these cities. Even though these cities are smaller, their proximity to well-connected cities can make them vulnerable to virus importation. Cities with high DEG values have a large number of roads that lead to them. For this reason, these cities may be the travel destination of people who are infected with the virus, increasing the possibility of introducing the contagion to the noninfected local population. Higher BET values are associated with cities that are crossed by a large number of short paths. Therefore, they can be intermediate bridges between larger cities, where high incidence rates occur. These preliminary results thus far indicate the importance of the network-based spatial approach to modeling the spread of the virus.
Mots clés : Network Analysis|Covid-19|Centrality Measures|Brazil|Spatial Analysis
A103460MF