Flow and access: Driving forces of COVID-19 spreading in the first stage around Hubei, China

Abstract

Background This research takes the six provinces around Hubei Province where the Corona virus disease 2019 (COVID-19) outbreak as the research area, collected the number of cumulative confirmed cases (NCCC) in the first four weeks after the lockdown to explore the spatiotemporal characteristics, and to identify its influencing factors by correlation and regression analysis, finally providing reference for epidemic prevention and control policy. Methods The analysis of variance was used to test the spatiotemporal variability of the NCCC in the six provinces, the Pearson coefficient was taken to find the correlation relationship between the NCCC and multiple factor data in socio-economic, geography and transportation, and the following regression equation was obtained based on regression analysis. Results This study found that there is significant spatial variability in the NCCC among the six provinces and the significant influencing factors are changing along the four weeks. The NCCC in Shaanxi and Chongqing in the West was less than that in the other four provinces, especially in Shaanxi in the northwest, which was significantly different from the four provinces in the East, and has the largest difference with adjacent Henan province (792 cases). Correlation analysis shows that the correlation coefficient of the number of main pass is the largest in the first week, the correlation coefficient of the length of road networks is the largest in the second week, and the NCCC in the third and fourth week is significantly correlated with the average elevation. For all four weeks, the highest correlation coefficient belongs to the average elevation in the third week (r = 0.943, P = 0.005). Regression analysis shows that there is a multiple linear regression relationship between the average elevation, the number of main pass and the NCCC in the first week, there is no multiple linear regression relationship in the second week. The following univariate regression analysis shows that the regression equations of various factors are different. And, there is a multiple linear regression relationship between the average elevation, the length of road networks and the NCCC in the third and fourth week, as well as a multiple linear regression relationship between the average elevation, population and the confirmed cases in the fourth week. Conclusion There are significant spatial differences in the NCCC among the six provinces and the influencing factors varied in different weeks. The average elevation, population, the number of main pass and the length of road networks are significantly correlated with the NCCC. The average elevation, as a geographical variable, affects the two traffic factors: the number of main pass and the length of road networks. Therefore, the NCCC is mainly related to the factor categories of flow and access.