This study utilizes global and local spatial modeling techniques to comprehend the factors that may contribute to the expansion of the SARS-CoV-2 virus at the commune level in Chile. The results show that the global regression model (Ordinary Least Squares) presents a poor fit with an adjusted R2 of 0.33. However, the local modeling regressions (Multiscale Geographically Weighted Regression, MGWR) obtain a better fit in the explanation of the factors that affect the incidence rate of the disease COVID-19 with a R2 of 0.90 and presents an AICc lower than the models obtained by the Geographically Weighted Regression. Therefore, MGWR significantly improves the performance in comparison with the common global regression model. The mobility indices, meteorological factors such as wind speed and precipitation, and the environmental pollution variables proved to be related with the incidence rate of COVID-19. The use of local regression models yields relevant information to health professionals and legislators to help effectively control this virus and other similar viruses with similar spread characteristics in Chile.