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

Yuke Wang; Peter F.M. Teunis

In December 2019, several cases of severe pneumonia appeared in Wuhan, the capital city of Hubei province in China. The outbreak was caused by a novel coronavirus: Severe Acute Respiratory Syndrome Coronavirus 2, or SARS-CoV-2 [1] and the disease (COVID- 19) started to spread rapidly [2] in Wuhan. As of 19 February 2020, globally 75,204 cases have been confirmed with 2,009 deaths in 26 countries [3] . In China, 83·2% (62,031/74,576) of the confirmed cases were located in Hubei province and 60·4% (45,027/74,576) were located in the city of Wuhan, where the outbreak originated [4, 5] . The "pneumonia of unknown etiology" appeared in Wuhan from 8 December 2019. Many early cases have been reported to be linked to the Huanan (Southern China) Seafood Wholesale Market (hereafter referred to as "the Market") [6] . By 2 January 2020, 41 initial cases were confirmed as COVID-19 [7] . As the novel coronavirus epidemic was spreading within Wuhan, the Chinese New Year/Spring Festival (25 January 2020), the most important holiday in China, was approaching. In 2019, 2.99 billion of people traveled by bus, train, and plane during 40 days around Chinese New Year [8] . Wuhan, with a population size of 11 million, is one of the four most important railway hubs in China. With billions of people travelling and lots of family and friends gathering, there was greatly increased risk of rapidly spreading this newly emerging infectious disease, nationally, and even globally. On 19 January 2020, the first confirmed COVID-19 case outside of Wuhan appeared in Shenzhen, Guangdong [9] . As of 23 January 2020, confirmed COVID-19 cases have been reported in 29 provinces of mainland China and nine countries and areas outside of mainland China [10, 11] . On the same day, the central government of the People's Republic of China initiated a lockdown in Wuhan and two nearby prefectural cities, Huanggang and Ezhou, in Hubei province, to prevent spreading of the COVID-19 outbreak [12, 13] . However, as many cases had "escaped" from Wuhan before the lockdown, COVID-19 has spread to most provinces. The numbers of cases imported from Wuhan for different provinces heavily depend on their connectivities with Wuhan. At the same time, municipalities responded differently, regarding timeliness and adequacy of measures (including declarations of public health emergency, holiday extension, event cancelation, and surveillance using infrared thermometers in public spaces. In this study, we used a method for analysing transmission patterns based on the serial interval between clinical cases of COVID-19 [14] . Based on dates of symptom onset and contact information for confirmed COVID-19 cases in Tianjin province, it was possible to estimate the serial interval distribution. With this information, it was possible to estimate numbers of new cases caused by any subject infectious with COVID-19, or their reproduction numbers during early stages of the outbreak in Wuhan. More importantly, we examined the heterogeneity in transmission among 30 provinces of mainland China and also among 20 cities in Guangdong province. Table 1 here.