Impact Factors of COVID-19 Epidemic Spread in Hubei Province Based on Multi-Source Data

With more than 26 million confirmed cases and over two million case-fatalities worldwide, the coronavirus disease (COVID-19) pandemic has transformed the dynamics of human lives globally. It has been designated as a pandemic by the World Health Organization. The COVID-19 virus can be transmitted through droplets, aerosols, or direct contact. It possesses evident characteristics of human-to-human transmission. Additionally, COVID-19 is a highly pathogenic new coronavirus, and people are prone to serious respiratory diseases resulting in high mortality after becoming infected. It has posed a great security threat to the entire human society and caused hundreds of billions of economic losses. The novel coronavirus disease 2019 (COVID-19) epidemic spread from Wuhan to all other cities in China before Spring Festival, causing serious public health issues and preventing the growth of the social economy. Analyzing the spatial-temporal spread pattern of COVID-19 can support the prevention of the epidemic. Thus, this study aims to analyze the temporal-spatial spread characteristics of COVID-19 in Hubei Province. First, a regression model with variables of migration big data (mobility scale index (MSI) and traffic intensity) is employed to explore the temporal pattern of the spread of the epidemic. Second, the spatial spread characteristics of COVID-19 are analyzed using a regression model comprising transportation information (primary and secondary road transportation networks) and social economic information (2018 GDP data). The results illustrate the following. First, the regression model based on population migration data and daily COVID-19 cases in each city was significant (Sig.=0.00), with R2 up to 0.715, indicating that the independent variable could explain the dependent variable. As indicated by the standardized coefficient results, MSI (0.85) has a greater impact on the daily new cases in each city. Second, the cumulative infection rate per 10000 people was positively correlated with the number of medical institutions and GDP with correlation coefficients of 0.689 and 0.774, respectively, Sig. was less than 0.05. However, it was not correlated with the number of beds (Sig. > 0.05). Third, the spatial regression model based on the traffic network, socio-economic data and cumulative infection rate of ten thousand people in each city of Hubei was also significant. The independent variables in the model can explain the variability of 67.2% of the dependent variables. The results of the standardized coefficient show that the GDP ratio of each city has a greater impact on the model. The results of the study are expected to provide scientific data support for the government and epidemic prevention workers to formulate efficient epidemic prevention and policy decisions. In conclusion, the model fit of multiple regression on the time scale is better than that on the spatial scale. Population migration has the greatest impact on the spread of the epidemic. That is, population mobility has a greater effect on the prevention and control of epidemic situations. The results of the study are expected to provide scientific data support for the government on formulating epidemic prevention policies