The COVID-19 pandemic has affected every aspect of human society, from the economy and politics on the macro level to livelihoods and socializing on the individual level. Regarding its impact on population health, in addition to the millions of COVID-19 cases and deaths worldwide (Dong et al., 2020), the pandemic also triggered a mental health crisis. As estimated, the pandemic caused a 27.6% increase in cases of major depressive disorder globally in 2020 (COVID-19 Mental Disorders Collaborators, 2021).

Social capital, broadly referring to resources available to members of social groups and social networks (Villalonga-Olives and Kawachi, 2015), was considered a social determinant of mental health (Lund et al., 2018) and, therefore, may protect mental health against the pandemic. As a concept imported from social science, social capital was conceptualized in two approaches in public health studies: the network approach and the cohesion approach (Moore and Kawachi, 2017). The network approach mainly follows Lin (1999) and conceptualizes social capital as social resources embedded in individuals’ social networks. On the other hand, the cohesion approach mainly follows Putnam (2000) and Coleman (1988) and emphasizes structural and cognitive dimensions of social capital. Structural social capital refers to the degree to which individuals participate in social networks, associations, and civic events and emphasizes what people do; in contrast, cognitive social capital refers to perceptions of trust and reciprocity and emphasizes what people feel (Kawachi et al., 2008). A systematic review showed that cognitive social capital was protective against developing common mental disorders (CMD), while results about structural social capital and CMD were less consistent (Ehsan and De Silva, 2015). A previous study even found that greater structural social capital was associated with worse mental health in some specific circumstance (Thuy and Berry, 2013).

Social capital could also protect mental health following natural disasters, such as the 2005 Hurricane Katrina in the US (Beaudoin, 2007), the 2008 Wenchuan earthquake in China (Zhao, 2013), and the 2011 Great East Japan Earthquake and Tsunami (Kawachi et al., 2020). Moreover, pre-disaster social capital was conducive to disaster resilience. As shown previously, pre-disaster social capital could facilitate survivors' housing reconstruction after the 2008 Wenchuan earthquake (Wei and Han, 2018) and lower older survivors' risk of developing posttraumatic stress disorder (PTSD) after the 2011 Great East Japan Earthquake and Tsunami (Hikichi et al., 2016). Furthermore, social capital was more important in protecting mental health for survivors with more severe housing damage than for those with less severe housing damage (Hikichi et al., 2017, 2018), suggesting that disaster experiences moderated the association between social capital and survivors’ mental health.

Although the COVID-19 pandemic differs from natural disasters as it requires physical distancing and lasts for months, even years, it still shares some similarities with natural disasters regarding the impact on human life. Previous research showed that during the COVID-19 pandemic, social capital was also associated with better mental health (Chan et al., 2021). Moreover, pre-COVID-19 cognitive social capital could reduce older people's risk of developing depressive symptoms during COVID-19 (Sato et al., 2022). Additionally, neighborhood social support could buffer the negative effect of COVID-19-related stressors on psychological distress (Chen et al., 2021), indicating that COVID-19 experiences might also moderate the association between social capital and mental health.

However, few studies have compared the longitudinal association between social capital and mental health in a disaster/pandemic scenario with that in a pre-disaster/pre-pandemic scenario. Therefore, it is unclear whether social capital is more or less important in prospectively protecting mental health when facing a disaster or a pandemic as compared with facing no such event. Moreover, within a disaster/pandemic scenario, ecological-level severity of the situation may serve as a contextual effect threatening mental health. Nevertheless, previous studies mainly focused on whether social capital was more important in protecting mental health as individual-level disaster-related damage (e.g., housing damage) increased, but few studies examined how the strength of the association between social capital and mental health varied as ecological-level disaster/pandemic situation escalated.

We aim to fill the above-mentioned research gap with a focus on the early stage of the COVID-19 pandemic in China. China experienced its first large-scale COVID-19 outbreak in early 2020. To contain the first COVID-19 outbreak, China implemented a strict lockdown of Wuhan, the epicenter of China's first COVID-19 outbreak, on January 23rd, 2020. Following the lockdown of Wuhan, provinces around the country issued Tier 1 public health emergency responses between January 23rd and January 29th, 2020 (State Council, 2020). The first outbreak was generally under control since mid-March 2020, and China entered a “normalization” stage of COVID-19 prevention and control defined by the Chinese government at the end of April (State Council, 2020). In this study, we measured peri-COVID-19 depression in 2020 during the “normalization” stage of COVID-19 prevention and control (State Council, 2020). During the time of data collection in 2020, the Chinese government had not proposed the “Dynamic Zero-COVID-19” policy.

Specifically, using a prospective cohort survey in China, which included two waves of data before COVID-19 and one after the end of the first COVID-19 outbreak in China, we examined whether the longitudinal association between social capital and depression in the peri-COVID-19 context was stronger compared with the pre-COVID-19 context. Furthermore, we examined within the COVID-19 context (i.e., 2020) whether the association between pre-COVID-19 social capital (in 2016 and 2018) and peri-COVID-19 depression became stronger in provinces with worse COVID-19 situation. We focused on cognitive social capital in this study. By examining the association between pre-COVID-19 cognitive social capital and peri-COVID-19 depression, this study can contribute to understanding the role of cognitive social capital in pandemic preparedness.