Population anxiety and positive behaviour change during the COVID-19 epidemic: Cross-sectional surveys in Singapore, China and Italy

Jane M Lim; Zaw Myo Tun; Vishakha Kumar; Sharon Quaye; Vittoria Offeddu; Alex R Cook; May Oo Lwin; Shaohai Jiang; Clarence C Tam

To measure anxiety, we adapted items from the State-Trait Anxiety Inventory 5 to assess respondents' level of worry in relation to the epidemic using a series of questions with a 5-point Likert response scale. To measure behavioural responses, we asked participants if they had modified or engaged in specific behaviours to reduce the risk of infection to themselves or others. These included personal hygiene behaviours and other protective measures, avoidance of workplaces, public spaces, social engagements or public transportation, changes to work-related or personal travel plans, and stockpiling of groceries, medications or personal protective equipment. Additionally, we included questions related to confidence in government and epidemic control measures, and acceptance of quarantine measures from previous published surveys. 6, 7 The basic questionnaire (Appendix A) was the same in all countries, but was modified to include countryspecific response options or epidemic control measures. In China, it was not possible to include questions on confidence in government control measures. In addition, we omitted the majority of questions on knowledge from the Italian questionnaire as it was apparent by the time the survey was administered that the vast majority of the population would be aware of how COVID-19 was acquired, what the signs and symptoms were and what control measures had been implemented. The survey was administered in English in Singapore, in Chinese in China, and in Italian for respondents in Italy.