At Which Area Level Does COVID-19 Infection Matter Most for an Individual's Self-Rated Health? A Multilevel Fixed-Effects Model Analysis in Japan

Several studies have reported the adverse impacts of the COVID-19 pandemic on health outcomes. However, little is known about which area of COVID-19 infection matters most for an individual's subjective health outcomes. We addressed this issue in the present study. We used the longitudinal data of 2260 individuals obtained from a two-wave internet-based nationwide survey conducted in Japan. We estimated the multilevel regression models, which controlled for fixed effects at the individual and prefecture levels, to explain an individual's self-rated health (SRH) based on the reported number of new COVID-19 infection cases at different area levels: prefecture, group of neighboring prefectures, and regional bloc. We found that SRH was highly associated with the average and maximum number of new infection cases among neighboring prefectures or in the regional bloc, but not with those at the prefecture level, if used jointly as explanatory variables. The results suggest that inter-prefectural coordination is needed not only to contain COVID-19 but also to reduce its adverse impact on the subjective health outcomes of residents.