Spatiotemporal characteristics of coronavirus disease 2019 in 258 cities in China

  Objective  To explore the spatiotemporal characteristics of coronavirus disease 2019 (COVID-19) in China and related factors, and provide evidence for COVID-19 prevention and control.

  Methods  Based on the information about COVID-19 incidence, population migration from Wuhan of Hubei, socio-demography and geography in 258 prefecture-level cities in China from January 21 to March 23, 2020, a spatial autocorrelation analysis and a hot spot analysis were conducted to explore the spatial heterogeneity and identify the hot spots of COVID-19 epidemic. The geographically weighted regression (GWR) model combined with linear regression model was used to identify the related factors for spatial heterogeneity.

  Results  During January 21 to March 23, 2020, a total of 29 789 COVID-19 cases were reported in 258 cities. The overall incidence of COVID-19 showed spatial clustering (Moran's I=0.436, Z=25.363, P<0.001). The Baidu migration index (from Wuhan) was statistically related to COVID-19 incidence at city level (t=14.550, P<0.001), showing positive effect ( β: 0.564–0.565).

  Conclusion  Spatial heterogeneity was noted in COVID-19 epidemic in China. Cities with more migration from Wuhan were more likely to report higher COVID-19 incidence. Understanding spatiotemporal characteristics of emerging infectious diseases is helpful to the early warning and control of the epidemic.