Comparison of the spatiotemporal characteristics of the COVID-19 and SARS outbreaks in mainland China

Background Both coronavirus disease 2019 (COVID-19) and severe acute respiratory syndrome (SARS) are caused by coronaviruses and have infected people in China and worldwide. We aimed to investigate whether COVID-19 and SARS exhibited similar spatial and temporal features at the provincial level in mainland China. Methods The number of people infected by COVID-19 and SARS were extracted from daily briefings on newly confirmed cases during the epidemics, as of Mar. 4, 2020 and Aug. 3, 2003, respectively. We depicted the spatiotemporal patterns of the COVID-19 and SARS epidemics using spatial statistics such as Moran’s I and the local indicators of spatial association (LISA). Results Compared to SARS, COVID-19 had a higher incidence. We identified 3 clusters (predominantly located in south-central China, highest RR=135.08) for COVID-19 and 4 clusters (mainly in Northern China, highest RR=423.51) for SARS. Fewer secondary clusters were identified after the “Wuhan lockdown”. The LISA cluster map detected a significantly high-low (Hubei) and low-high spatial clustering (Anhui, Hunan, and Jiangxi, in Central China) for COVID-19. Two significant high-high (Beijing and Tianjin) and low-high (Hebei) clusters were detected for SARS, although the global Moran’s I value was not significant. Conclusions The different spatiotemporal clustering patterns between COVID-19 and SARS could point to changes in social and demographic factors, local government containment strategies or differences in transmission mechanisms between these coronaviruses.