Mapping the changing Internet attention to the spread of coronavirus disease 2019 in China

The incessant spread of the coronavirus disease 2019 (COVID-19) has been a great threat to human health. By 17 March 2020, the number of laboratory-confirmed cases had exceeded 179,000, with more than 7000 deaths across at least 150 countries. Due to the extremely contagious nature of COVID-19, the Chinese government has made broad and aggressive responses to restrict movement, transportation, and business for six to eleven weeks. Wuhan, a city in Hubei (HB) province from which COVID-19 emanated, has been quarantined since 12 January 2020, and many other cities have been placed under travel restrictions. Citizens have been strongly encouraged to stay home and limit face-to-face contact; as a result, people’s daily lives are dominated by the Internet as never before. Here, we visualize the spread of COVID-19 and people’s Internet attention in China.

Six special dates were chosen for data collection: (a) 3 January 2020, when China reported COVID-19 to the World Health Organization (WHO); (b) 20 January 2020, when cases were reported outside HB province; (c) 24 January 2020, when WHO considered declaring COVID-19 as a Public Health Emergency of International Concern (PHEIC); (d) 31 January 2020, when WHO declared COVID-19 to be a PHEIC; (e) 7 February 2020, which is one week after the declaration of PHEIC; and (f) 12 March 2020, a recent day (at time of writing). We collected two datasets for these dates. One dataset is the number of confirmed cases reported by each province of China, and the other is the number of Baidu—the most widely used search engine in China—searches by province containing any of the following keywords in Chinese: “novel coronavirus,” “coronavirus,” “nCoV,” “CoV,” “pneumonia,” “SARS,” “fever,” “isolation,” and “face mask.”

The data were visualized in the form of cartograms (Dorling, 1996, 2007; Gao et al., 2019) using the diffusion-based method proposed by Gastner and Newman (2004) as it is effective in the representation of spatial data (Sun and Li, 2010). The following findings can be drawn from Figure 1:

OPEN IN VIEWER

Figure 1.

Numbers of confirmed COVID-19 cases (a) and that of Chinese searches relating to COVID-19 (b) by province in China. The two letters in each polygon denote the standard abbreviation of the corresponding province, which is available at https://www.cottongen.org/data/nomenclatures/China_provinces

HB is the most severely afflicted area in China in terms of the number of confirmed cases. Subsequently, there was a rapid increase in COVID-19 cases in Zhejiang (ZJ) and Guangdong (GD) provinces which have developed economies and very close economic cooperation with HB. For instance, more than 200,000 people from Wenzhou city of ZJ have invested in Wuhan city of Hubei (HB). Finally, the pandemic expanded significantly in provinces adjacent to HB. Therefore, the spread of COVID-19 is influenced by both geographical proximity and economy.

Contrarily, the number of searches shows a different pattern from that of confirmed cases. First, the number of searches from HB is not outstanding among all provinces. This is probably because people from HB are busy fighting COVID-19. Second, more provinces paid intense attention to COVID-19 after the declaration of PHEIC (31 January), indicating that WHO is vital in leading people in the fight against COVID-19. Third, the number of searches decreased eventually in most provinces, suggesting reduced anxiety and increased confidence among the people. The different pattern between COVID-19 cases and the number of searches may be because people not only refer to the Internet but also use other social media to share disease information.