Altogether, 1477 pieces of authorized information on COVID-19 were retrieved from the national and local health commissions websites (n = 180, 12%), People’s Daily Online News (n = 806, 55%) and CCTV news (n = 491, 33%) via the Parsehub tool. Over the same period, 297,000 pieces of short videos were found using the keyword “new coronavirus” and “suspected pneumonia,” which were released by mainstream Chinese social media and had been played over 21.8 billion times [9]. These channels were also selected as primary information sources by the 1311 respondents of our online survey, who were between the age of 25–60 years. 42% of these participants were males, and 68% had a bachelor degree or above (as shown in Table 1).
The survey results show that only 24% of responses chose to “obtain epidemic information” through TikTok, and only 4.8% believed that transmitted high credibility information. On the other hand, 85.76% of people choose to “obtain epidemic information” through “official news,” and 58.8% of participants tend to obtain the disease data through the “National Health Commission, Centers for Disease Control, and the official websites of hospitals.” Among them, 88.9% of the masses believe “official news” as a channel for obtaining highly credible information. Despite the high trust in authorized information released by official channels, the contrasting amount of information released by official channels versus social media heightens the possibility of information overload to the public.
We further analyzed the content and release time of authorized information by official channels against the public’s perceived risk and actions. During the window period, as a kind of special and new disease, the public’s knowledge about COVID-19 is low. These had less information publicly available to announce and promote the disease, so only two disease-related announcements were released by the Wuhan Municipal Health Commission. The rising period was during Chunyun, the World’s largest annual human migration. A gradual rise in COVID-19 disease-related information was most evident during this period, when the suspected and confirmed cases of COVID-19 had already accumulated to 830 and 1072 respectively, scattered over 29 cities across China. Information exploded over the white-hot period: 1338 pieces of data were released over 18 days; with 57 pieces of disease-related, 5 pieces of public prevention-related, and 13 pieces of social welfare-related information were released per day.
The public’s risk perception towards COVID-19 was dominated by “I do not know” (45%) and “be alert, but not enough attention” (73%) during the window period (Fig. 1, bar chart). Even at the rising period, 37% of respondents still responded, “be alert, but not enough attention” and only half of them started to “take it seriously”. Not until the white-hot period did the public start to pay attention to the epidemic. The public’s responses of “take it seriously” and “fear and panic” rose to 80% and 4%, while “be alert, but not enough attention” dropped to 13%, yet still ranked as the top two most prevalent responses. A similar lag in the public’s protective actions was also apparent (Fig. 2, bubble chart). Associated with the public’s low-risk perceptions, over two- thirds of the respondents chose to “take no actions” during the window period, which was about 2-2.5 times higher than those starting to take “multiple actions” (25%) (i.e. personal protection, self-isolation and family education). This ratio reversed in the rising period, where 65% of respondents took “multiple actions.” and “take no actions” responses dropped to 24%. By the white-hot period, respondents predominately took “multiple actions”, increased by 61% as compared to the initial period.