[Factors associated with behaviors to prevent COVID-19 infection during the declaration of emergency: A study among Tokyo residents].

Objectives　The COVID-19 pandemic spread rapidly across the globe during the first half of 2020. In Japan, a state of emergency was declared on April 7, 2020, which had a significant impact on the life of citizens. This study focused on behaviors like avoiding going out or coming in contact with others and frequent hand-washing to prevent the infection and the spread of COVID-19 among people living in Tokyo. We also examined the factors associated with these behaviors during the declaration of emergency.Methods　An online survey was conducted from April 26 to 29, 2020, approximately 20 days after the declaration of the emergency, among men and women aged 20-69 years living in Tokyo. The study framework was based on the protection motivation theory, which explains the risk-reducing behaviors, and focus theory of normative conduct, which explains the effect of others' behavior on one's own behavior. The frequency of behaviors like avoiding going out or coming in contact with others and frequent hand-washing, as well as the perception of the risk of COVID-19 during the week preceding the survey, were assessed. Each preventive action was evaluated based on the following factors: perceived effectiveness (response efficacy), perceived practicability (self-efficacy), necessary cost (response cost), and perceptions of how much should be done (injunctive norm) and how well others are doing it (descriptive norm). Hierarchical multiple regression analysis with these behaviors as outcomes were performed.Results　This study included 1,034 participants (50.3% male, mean age 44.82 years, standard deviation 14.00 years). The analyses of the frequency of avoiding going out or coming in contact with others showed that the injunctive norm was positively associated with the behavior (standardized partial regression coefficient (ß)=0.343, P<0.001), while the descriptive norm was negatively associated with the behavior (ß=-0.074, P=0.010). Furthermore, the two-way interaction between risk perception, response efficacy, and self-efficacy was significant (ß=0.129, P<0.001), indicating that risk perception was positively associated with the behavior only when either response efficacy or self-efficacy was low. A similar analysis conducted for hand-washing behavior revealed that injunctive norm (ß=0.256, P<0.001) and response efficacy (ß=0.132, P<0.001) were positively associated with the behavior, while the response cost (ß=-0.193, P<0.001) was negatively associated with the behavior.Conclusion　Some variables in the protection motivation theory and the focus theory of normative conduct were related to the behavior for the prevention of COVID-19. The results suggest that the application of these theories is useful in future studies.

COVID-19 Social Distancing Measures and Loneliness Among Older Adults.

OBJECTIVES: In response to the coronavirus disease 2019 (COVID-19) pandemic, older adults are advised to follow social distancing measures to prevent infection. However, such measures may increase the risk of loneliness. The current study aimed to investigate (a) whether social distancing measures, particularly limiting close social interactions, are associated with loneliness among older adults, and (b) whether the association between social distancing measures and loneliness is moderated by sociodemographic characteristics. METHOD: Data were from the fourth wave (April 29 to May 26, 2020) of the nationally representative Understanding America Study COVID-19 Survey. We used data on adults 50 years or older (N = 3,253). Logistic regression models of loneliness were performed. Five indicators of social distancing measures were considered: (a) avoiding public spaces, gatherings, or crowds; (b) canceling or postponing social activities; (c) social visits; (d) no close contact (within 6 feet) with people living together; and (e) with people not living together. RESULTS: Cancelling or postponing social activities and avoiding close contact with people living together were associated with 33% (odds ratio [OR] = 1.33, confidence interval [CI] = 1.06-1.68, p < .05) and 47% (OR = 1.47, CI = 1.09-1.99, p < .05) greater odds of loneliness, respectively. Furthermore, limiting close contact with coresidents increased the probability of loneliness more for males, non-Hispanic Whites, and those with higher levels of education and income. DISCUSSION: Efforts should be made to help older adults maintain social connectedness with close others by virtual communication methods. Our findings also call special attention to vulnerable groups at elevated risks of loneliness, emphasizing the need for tailored interventions.