The Socioeconomic Determinants of Pandemics: A Spatial Methodological Approach with Evidence from COVID-19 in Nice, France

During the period from January 4 to February 14, 2021 the spread of the COVID epidemic peaked in the city of Nice, with a worrying number of infected cases. The spatial dynamics of the pandemic revealed explicit geographical patterns. This article focuses on analyzing the spatial pattern of virus spread and assessing the geographical factors influencing this distribution. Thus, in this article, spatial modeling was carried out to examine geographical disparities in terms of distribution, incidence and prevalence of the virus, while taking socio-economic factors into account. A multiple linear regression model was used to identify the key socio-economic variables affecting the spread of COVID-19 in Nice. Global and local spatial autocorrelation was measured using Moran and LISA indices, followed by spatial autocorrelation analysis of the residuals. Similarly, we used a global regression model and local models (the Geographically Weighted Regression (GWR) model and the Multiscale Geographically Weighted Regression (MGWR) model), to assess the influence of socio-economic factors that vary on a global and local scale, in order to adopt the most appropriate model explaining the spread of the disease. The results confirm that covid-19 is strongly spatially correlated, and that spatial analysis is an essential step in implementing effective preventive measures. The various global and local models identified four significant variables with regard to vulnerability to COVID disease in Nice. Our results reveal a marked geographical polarization, with affluent areas in the southeast contrasting sharply with disadvantaged neighborhoods in the northwest. Neighborhoods with low LHDI, low levels of education, social housing and immigrant populations. These latter factors all point to worrying values. On the other hand, people who use public transport are significantly negatively correlated with contamination by the virus. These results underline the importance of geographically predicting COVID-19 distribution patterns to guide targeted interventions and health policies in Nice. Understanding these spatial patterns using models such as MGWR can help guide public health interventions and inform future health policies, particularly in the context of pandemics.