This study preliminarily investigated on the public’s anxiety level during the outbreak of COVID-19 in China, which is one of the biggest infectious disease outbreaks in Chinese history. One notable strength of the present study was that the data were collected at the peak of COVID-19 outbreak in China. Thus, the public responses to COVID-19 were more accurately recorded than retrospective reporting. Generally, it was found that the anxiety levels of most respondents were normal during the outbreak of COVID-19. Even though the data were collected at the peak of COVID-19 outbreak, a majority of respondents reported low to medium level of perceived risks. One possible explanation for these findings is that most people were mandated to be quarantined at home, which might lower the risk perceptions and anxiety over COVID-19 infection. Most respondents considered that they themselves knew the COVID-19 fairly well.
Media played a particular important role in affecting the public responses to COVID-19 as most Chinese people were quarantined at home to prevent the possible virus spreading during the outbreak. As a result, the people spent a large amount of time on media. Previous research has generally found that the increased exposure to pandemic-related information leads to a higher level of anxiety . These studies commonly measured media exposure by assessing the amount of information that people received from the media during the pandemic, while little is known about what the specific type of information that may increase or decrease anxiety. The present study focused on the effect of media exposure to information about COVID-19 on the anxiety level. National media, social media, the China CDC, and government authorities, disseminate information in a timely manner to make people stay informed during the pandemic. It is thus virtually impossible that people are not exposed themselves to any COVID-19 information at all. The main finding of the present study is that media exposure to different COVID-19 information influenced people’s self-rated anxiety in distinct ways. The results showed that some COVID-19 information increased public anxiety, while others decreased it. For example, it was found that donation information and the life of ordinary people were positively with self-rated anxiety. One possible explanation was that donation information was a signal that the hospitals were in shortage of protective equipment, which might therefore increase the public anxiety level. During the pandemic, most news coverage about the life of ordinary people were negatively-valanced, which focused on the stories of people who were infected with or died of the novel coronavirus. Therefore, exposure to information about the life of ordinary people increased people’s anxiety level. Information about returning to work/school was negatively associated with the anxiety level, because it suggested that the pandemic was under control and quarantine was soon be over. Interestingly, information about the number of reported COVID-19 cases and the analysis of the COVID-19 pandemic were negatively associated with anxiety. It was speculated that information regarding the numbers of reported cases and deaths actually lowered the uncertainty level of the public, because uncertainty often arises with lack of information and “uncertainty is experienced subjectively as anxiety” . Compared with those who do not know the real situation of the pandemic, people are likely to feel less anxious when they get to know more information about the seriousness of the pandemic through media. Not surprisingly, information about how to prevent COVID-19 was negatively associated with the level of anxiety in this study, because knowing more about the prevention of COVID-19 could lower people’s anxiety over the novel coronavirus infection.
It was hard to draw a strong causal conclusion about the relationship between media exposure and anxiety based on results in this study. Yet it seemed to be reasonable to assume that different media contents had different impacts on people’s anxiety level, since a majority of participants (72.1%) didn’t actively search for COVID-19 information using web search engine. Despite that social media afford users some level of flexibility in choosing the content, in general, most people are the passive social media consumers who only acquire information on their timeline. Therefore, it is not easy for people to get themselves exposed to the content-specific COVID-19 information because most participants receive information in a more passive manner.
Overall, the more educated and younger respondents experienced a relatively higher level of anxiety. Such results were generally consistent with the previous research suggesting that young people were more likely to be anxious than the older adults . Previous studies have found that disease knowledge is a significant predictor of anxiety level  . This may be ascribed to the fact that the young people and people who are more educated receive more information about COVID-19 through different media outlets than older adults and people who are less educated, so they are more likely know the severity of contracting COVID-19. Therefore, younger and more educated people who have more knowledge about COVID-19 are more likely to feel anxious. Gender and income did not have significant associations with anxiety level.
Another important finding generated from the present study was that, both social proximity and geographical proximity to COVID-19 were positively associated with anxiety. Due to the high infectiousness of COVID-19, people who personally knew someone infected with COVID-19 were more susceptible to anxious thoughts than others. Living in an area that had reported cases was also positively associated with anxiety. The results were consistent with the previous study showing that a shorter distance to the risk resulted in higher risk perception . Moreover, individuals who perceived more risks reported a higher level of anxiety than those who perceived less risks. It is possible that people who have higher level of risk perception overestimate their risk of COVID-19 infection and therefore feel more anxious. Nevertheless, to better understand the causal relationship between anxiety and risk perception, longitudinal studies are recommended.
Our study has several limitations that are worth noting. Firstly, our results might suffer from generalizability problem. Besides, the online panel inherently came with sample selection bias. The respondents were generally younger and more educated in our sample, and may therefore not be representative of rural populations in China. At present, a majority of Chinese citizens can get access to the Internet, but there are still people who are unable to access to Internet, especially for the older adults or those living in the economically disadvantaged areas. Moreover, this study was unable to explore how the level of anxiety fluctuated from the beginning of the pandemic to the time of data collection, therefore we cannot make strong causal inference due to the cross-sectional nature of this study.
This study raises important implications. Government and public health practitioners are recommended to take prevention-focused approach to promote citizens’ mental health during the pandemic. If the public’s psychological well-being cannot be ensured, heightened emotions such as anxiety could lead to detrimental social effects. For example, one study found that increased anxiety may lead to shortage of medical supplies such as face masks .