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

Yuke Wang; Peter F.M. Teunis

First, detailed data for 112 confirmed cases between 21 Jan 2020 and 12 Feb 2020 in Tianjin province was obtained from the website of Tianjin Health Commission, the website of Tianjin Centers for Disease Control and Prevention, and the official Weibo (China's Twitter equivalent) of a local newspaper [15, 16, 17] . Collected information included demographic information, travel history to Wuhan/Hubei, date of symptom onset, and any contact information that could be found. See Table 1 for a summary. Second, to evaluate the role of "the Market" in Wuhan and estimate the numbers of cases caused by contact with this source (its reproduction number) in early transmission, data for the first confirmed 425 COVID-19 cases with date of symptom onset and exposure information to "the Market" was extracted from a recently published report [6] . And in third place, we collected epidemic curve data: numbers of cases by province and date of confirmation from the Wikipedia page of COVID-19 cases in mainland China [18] and combined these data with detailed information, wherever available, about travel history and location (city and district) extracted from announcements of the Health Commissions of the provinces. These data were used to compare transmission of COVID-19 in different provinces and cities. Since the travel history to Wuhan is currently not available for all cases outside of Wuhan, the probability that any confirmed cases were linked with Wuhan was estimated using the 228 confirmed cases with known travel history between 20 January 2020 to 3 February 2020 in Beijing.