Strongly heterogeneous transmission of COVID-19 in mainland China: local and regional variation

Yuke Wang; Peter F.M. Teunis

The copyright holder for this preprint . https://doi.org/10.1101/2020.03. 10.20033852 doi: medRxiv preprint Early transmission of a newly emerging infectious disease in a population lacking immunity could reveal the basic reproduction number as long as there is no intervention. Currently, the source of COVID-19 is still unclear but many of the early cases were reported to have had contact with the Huanan Seafood Wholesale Market. A recent announcement released by the Chinese Academy of Sciences indicated that the source may not have come from this Seafood Market [25] . Any patient zero would have occurred before the initial 425 confirmed cases, but lacking any specific information about early spread of COVID-19, it can be assumed that a reservoir of infectious subjects could function as a super-spreading node causing new infections upon contact at any time within the outbreak. Assuming such a source for the initial cluster in Wuhan ("the Market") led to an estimated basic reproduction number 13·6. It must be noted that this environmental "super-spreading node" is hypothetical, and represents the production of new cases by the joint presence of infectious subjects connected to the initial source. There was strong heterogeneity in disease transmission in different provinces of mainland China. The present analysis produces two relevant characteristics: the numbers of cases imported from Wuhan, and the average reproduction number for the first five days after any confirmed imported case in each province. When there are many imported cases from Wuhan, these constitute a large base number of cases to start local spreading of the disease. Provinces geographically close to Hubei province are expected to experience mass migration from Wuhan, especially near the holiday season. Provinces with big tier-1 cities like Zhejiang and Guangdong both have very high estimated numbers of cases imported from Wuhan due to their economic connection with Hubei. When early transmission has a high reproduction number, local growth rate in a province will be high. Provinces geographically distant from Hubei, like Hebei, Guizhou, and Shandong, appear to have relatively high reproduction numbers during early local transmission. In provinces distant from Wuhan, the central hub of the outbreak, inhabitants and local governments may have been less cautious (with few public and personal prevention measures) considering high costs of interventions and perceived low risk. Since Guangdong province has both large numbers of imported cases and high reproduction numbers for early transmission, we further examined the transmission inside of Guangdong by city. Shenzhen and Guangzhou, two of the biggest cities in China, had many imported cases. This is expected considering intense economic connectivity: transportation links, for goods and people including enormous numbers of migrant workers working in those two cities. The numbers of cases imported from Wuhan into other cities in Guangdong province was smaller. This pattern matches the flow of people in public transport (by train and airplane) converging onto the main transportation hubs (Shenzen and Guangzhou) and diverging from there towards other destinations in Guangdong province. For early transmission following import from Wuhan, the reproduction number in most cities in Guangdong had an average reproduction number around 1, during the first five days. With appropriate and sustained disease prevention and control measures