A place is more than just a geographical location because it represents a person's social identity or position within a socially intertwined system of possibility (Siegrist, 2000). People consider moving from one residence to another to suit their needs and preferences (Clark and Ledwith, 2006). Time or resources, essentially influenced by the life course and contextual factors, facilitate or limit residential mobility (Coulter and Van Ham, 2013). The current era is the most mobile in history, and cities are particularly prone to migration within or across their borders (Hermosilla and Rebello, 2019). The purpose of the study is to examine whether residential mobility during the COVID-19 pandemic is associated with the health status of residents of the largest metropolitan city in South Korea (hereafter referred to as Korea).

People's health status affects their decision to move their residence to another location (Boyle et al., 2002, 2009). For example, people with diabetes were more likely than non-diabetics to move to areas with high levels of deprivation (Cox et al., 2007). Another study also discovered that people in poorer health moved more frequently than others to higher-deprived areas (Holmager et al., 2021). Health-selective residential mobility (moving their residence to another location) also shows various patterns according to the life course. Chronic diseases reduce the likelihood of residential mobility among young people (Lu and Qin, 2014). However, older adults in poor health were more likely than those in good health to move to find medical facilities or family members who could care for them (Boyle et al., 2002). Health selection can occur directly through residential mobility based on health status or indirectly through residential mobility based on characteristics associated with health (Van Lenthe et al., 2007).

In previous studies, when researching health-related residential mobility, Limitations in Activities of Daily living (ADL), self-rated health, and the prevalence of chronic conditions were frequently used to measure health status, and each measure partially measures a multifaceted and complex element of health (Van Der Pers et al., 2018). Although degrees varied, all three health measures were related to people's residential mobility (Van Der Pers et al., 2018; Friedman et al., 2015; Wilmoth, 2010). ADL was shown to be more closely associated with residential mobility than self-rated health or the prevalence of chronic conditions in a study examining the association between health and community-based residential mobility of older adults (Van Der Pers et al., 2018). Comparing study results becomes challenging when distinct health measures are used in each study; for this reason, using all three types of measures was recommended (Van Der Pers et al., 2018).

Since the coronavirus disease 2019 (COVID-19) was initially identified in Wuhan, China, in December 2019, the disease has spread quickly worldwide. The World Health Organization (WHO) declared COVID-19 pandemic in March 2020. At the beginning of the COVID-19 pandemic, it was generally understood that high population densities, usually in urban settings, were associated with high rates of COVID-19 infection (Mcfarlane, 2023; Teller, 2021; Rowe et al., 2023). However, subsequent research have suggested that COVID-19 infection was influenced by several other factors, including connectivity, crowding, socioeconomic position (SEP), social capital, and COVID-19 public health measures rather than merely population density (Mcfarlane, 2023; Florida et al., 2023; Teller, 2021; Boterman, 2023; Jo et al., 2021; Jamshidi et al., 2020). A prior study divided density into four sub-categories (road, built-up, population, and aged people density) (Khavarian-Garmsir et al., 2021). During the early phases of the pandemic, district-level road density in Teheran, the capital of Iran, exhibited a positive association with COVID-19 infection. In contrast, built-up and population density were negatively correlated (Khavarian-Garmsir et al., 2021). In particular, the most influential variable was the density of aged people (Khavarian-Garmsir et al., 2021). Two opposing characteristics at high densities may explain the conflicting result depending on the type of density (Hamidi et al., 2020; Khavarian-Garmsir et al., 2021; Teller, 2021). Frequent social interaction due to high densities lead to increased COVID-19 infection; conversely, more accessible healthcare and adherence to social distancing lower COVID-19 infection. A Korean study also revealed that while COVID-19 infection rates were higher in densely populated areas, vulnerability and health risks were lower (Lee, 2023).

Research on residential mobility trends around the beginning of the COVID-19 pandemic revealed an upsurge in residential mobility from urban to rural areas (Rowe et al., 2023). It can be related to the general public's risk perception of the high number of COVID-19 infections in densely populated areas during the early phases of the pandemic. However, overall, significant changes like urban exodus are not anticipated to occur (Florida et al., 2023; Rowe et al., 2023). This is because the virulence of COVID-19 has declined as time passes, and more effective vaccinations and treatments have been developed (Rowe et al., 2023). There was also considerable residential mobility to suburban areas (Vogiazides and Kawalerowicz, 2023), and a significant portion of residential mobility was due to people moving between cities during the COVID-19 pandemic (Brown and Tousey, 2021; González-Leonardo et al., 2022a). Similar findings were found in research carried out in Korea. Even in the early phases of the COVID-19 pandemic, residential mobility patterns did not alter considerably from those before the pandemic, according to a previous study that examined monthly residential mobility patterns from 2015 to 2020 based on Korean administrative data (Shin et al., 2020). Ninety-three percent of the 1000 adult Korean respondents residing in three metropolitan cities in Korea answered that they had no intention of relocating their residence due to the COVID-19 risk in a separate research (Park et al., 2021).

While the macrogeographic trend remained consistent, variations were noted in residential mobility patterns based on the characteristics of the individuals moving. The availability of remote work contributed to a rise in the flow of migrants from urban areas to suburban or rural areas (Florida et al., 2023). Age-related variations in residential mobility were also observed. According to Norwegian research, those who left Oslo in 2020 were older than those who left between 2016 and 2019 (Tønnessen, 2021). Older adults in a prior Korean study expressed more concerns about living in urban settings than younger age groups during the COVID-19 pandemic (Kang et al., 2021). In addition, previous research has discovered differences based on SEP and foreign-born, even though age differences were not observed (González-Leonardo et al., 2022b). Regarding health, a study that used cross-sectional survey data from 2000 participants in three Korean cities in August 2020 found that the more concerned residents about their lives in the urban areas during the COVID-19 pandemic had worse health status (Kang et al., 2021). However, residential mobility patterns according to health status during the COVID-19 pandemic have rarely been revealed using longitudinal data at the individual level.

Seoul, the country's capital, is at the top of the social hierarchy among Korea's regions. Despite the recent downturn in population numbers, Seoul has the highest population density and the largest population (9.4 million as of 2022) in the country; about half of all Koreans live in the Seoul Capital Area (Seoul and its surroundings) (Korea Statistical Information System, 2023). Seoul is divided into 25 autonomous districts. As of 2022, the population of a district was 390,432 on the median; the interquartile range was 150,048, with 130,785 as the lowest and 664,514 as the highest value. During the pandemic, Seoul's high population density could have facilitated residential mobility to lower population-density areas (Lee, 2023). Meanwhile, Seoul has effectively controlled the COVID-19 outbreak using the testing-tracing-treatment strategy, much like other densely populated East Asian cities (Park et al., 2020). In addition, Seoul is among the healthiest regions in Korea and has the greatest concentration of health facilities (Korea Health Promotion Institue, 2018; Eun, 2022). Because housing prices are the highest and work opportunities are the most plentiful, it is also the concentration area for human and financial resources (Park and Jang, 2016). Seoul's contrasting features are considered to be push-pull factors influencing residents to either stay or leave the city, operating differently based on individual characteristics.

This study examines the health status of people who moved from their residences (movers) in Seoul during the COVID-19 pandemic. We estimated changes in the baseline-year health status of those who moved and those who did not (stayers) during the study period. We also assessed the association between baseline health status and residential mobility using the general estimation equation (GEE) method. We used various variables to examine health's direct and indirect effects on residential mobility.