Impact of COVID-19 Infection and Related Social Concerns on Depressive Symptoms: Mediating Effects of Negative Changes in Daily Life and Moderating Effects of Age and Gender
Article information
Abstract
Objective
This study examined the mediating effect of negative changes in daily life due to the coronavirus disease-2019 (COVID-19) pandemic on depressive symptoms, considering COVID-19 infection and related social concerns. Additionally, comparisons of path coefficients between the groups were conducted based on age and gender.
Methods
A cross-sectional study design used data from the 2020 Korean Community Health Survey consisting of 229,269 individuals. This study used a self-reported questionnaire, including the Patient Health Questionnaire-9 and three items addressing social concerns related to COVID-19 infection. A single question assessed whether individuals had experienced COVID-19 infection within the last 3 months, and scores of negative changes in daily life due to the COVID-19 pandemic. Correlation analysis was performed on the variables. Structural equation model analysis was conducted to identify the mediating role of negative changes in daily life. Chi-square tests were also performed to compare the path coefficients based on age and gender.
Results
The structural equation models revealed that COVID-19 infection and related social concerns had both significant direct effects on depressive symptoms and indirect effects through negative changes in daily life. When comparing the path coefficients by age and gender, the coefficients related to depressive symptoms were highest in those under 65 years and in females.
Conclusion
Negative changes in daily life due to the COVID-19 pandemic serve as a partial mediator of the impact of COVID-19 infection and related social concerns on depressive symptoms. Special attention should be paid to depressive symptoms in those under 65 years of age and in females.
INTRODUCTION
Since the first case of coronavirus disease-2019 (COVID-19) was identified in 2019, the cumulative number of reported cases worldwide has reached nearly 197 million and the cumulative number of deaths has reached 4.2 million [1]. In Korea, >30 million people (more than 1/2 of the total population of Korea) have been diagnosed, and about 35,000 people died from COVID-19 by August 2023 [2].
As the COVID-19 outbreak continued, the emergence of associated depressive symptoms, often referred to as “corona blues” or “corona depression,” became a significant concern [3]. One meta-analysis reported a high prevalence of depressive symptoms reaching 33.7% during the COVID-19 pandemic [4]. Understanding the depressive symptoms in the context of the COVID-19 pandemic includes two aspects: COVID-19 infection itself, and social concerns related to COVID-19 infection. Several studies have reported that the prevalence of depressive symptoms following COVID-19 infection ranges from 11.5% to 31% [5-7]. In addition, as COVID-19 progressed into pandemic and profound social changes (such as economic recession, increased unemployment, and perceived social stigma) occurred, it is increasingly apparent that individuals’ depressive symptoms were related to social concerns arising from the pandemic [8,9].
Since the outbreak of the COVID-19 pandemic, many people have experienced changes in their daily lives due to quarantines and social distancing to reduce community spread [10]. These changes have included restrictions on public gatherings, distance education in schools, reduced work hours or closure of on-site work, and limited access to community support. Such disruptions have increased vulnerability to mental health problems [8,11]. Studies have shown that concerns about COVID-19 and COVID-19 infection also played a significant role in negative changes in daily life [6,9]. Subsequently, these negative changes in daily life are recognized as a factor that elevates the risk of developing depressive symptoms [8]. Therefore, it can be hypothesized that negative changes in daily life due to the COVID-19 pandemic may mediate the effects of COVID-19 infection and related social concerns regarding depressive symptoms. However, there is little evidence that supports whether negative changes in daily life due to the COVID-19 pandemic mediate the effects of COVID-19 infection and related social concerns on depressive symptoms.
In terms of sociodemographic factors, demographic variables associated with depressive symptoms showed differences before and during the COVID-19 pandemic [12,13]. In a study comparing risk factors for depressive symptoms before and during the COVID-19 pandemic, the prevalence of depressive symptoms was approximately 1.5 times higher in females both before and during the pandemic [3]. Although the COVID-19 pandemic affected both males and females equally and led to a significant increase in prevalence of depressive symptoms, the male-to-female ratio was not significantly different from that observed before the pandemic [3,4]. Additionally, the risk of depressive symptoms was higher among younger age groups during the pandemic, differing from prepandemic trends [3]. Therefore, further stratification of the analysis by age and gender may help to better understand these risk factors during the COVID-19 pandemic [3]. Regarding gender differences, females are generally at higher risk for depression both before and during the COVID-19 pandemic [3]. Recent studies suggest that females may have had a higher prevalence of depression during the COVID-19 pandemic due to previously recognized hormonal effects [14] and psychosocial factors such as reduced social activity and increased caregiver burden [15]. Additionally, younger age groups, identified as a risk factor during the COVID-19 pandemic [3], who are more socially active, may have experienced a greater impact on their daily lives and perceived greater social role pressure in the pandemic [16,17].
From this perspective, this study used a structural equation model to investigate the impact of COVID-19 infection and related social concerns on depressive symptoms, as well as the mediating role of negative changes in daily life. In addition, we conducted subgroup analysis, examined moderation effects, and compared path coefficients based on age and gender.
METHODS
Participants
This study utilized data from the 2020 Korean Community Health Survey (KCHS) conducted by the Korea Centers for Disease Control and Prevention (KCDC) [18]. The KCHS dataset is publicly available from the Centers for Disease Control and Prevention website (https://chs.kdca.go.kr/chs/rawDta/rawDtaProvdMain.do). Since 2008, this survey has been a collaborative effort between survey teams from universities and public health centers in compliance with the Community Health Act and involving local communities. The data collection involved a two-step sampling process. Initially, sample regions within each district (Dong/Eup/Myeon) were selected based on the number of households categorized by housing type (apartment/detached house) using probability proportional sampling. Subsequently, households within the chosen sample regions were selected using systematic sampling. Approximately five households per region were included. The survey was conducted through face-to-face interviews with adults aged 19 years and older. It included various factors such as health behaviors, quality of life, socio-environmental factors, and healthcare utilization. From August 16 to October 31, 2020, trained investigators visited the selected households and conducted one-on-one interview surveys. The survey questionnaire consisted of a total of 142 items divided into 18 sections. This study included a substantial sample size of 226,765 participants. The KCHS protocol was approved by the Institutional Review Board of the KCDC (2016-10-01-P-A).
Assessment
Sociodemographic questionnaires
The general characteristics were gender (male and female), age group (<65 years and ≥65 years), marital status (never married, married, divorced, and widowed), education level (less than elementary school, middle school diploma, high school diploma, or college degree or higher), employment status (yes or no), physician-diagnosed hypertension (yes or no), and physician-diagnosed diabetes mellitus (yes or no).
Depressive symptoms
Depressive symptoms were measured using the Patient Health Questionnaire-9 (PHQ-9), which is a self-reported depression screening instrument that was developed by Spitzer et al. [19] It comprises the following nine items: anhedonia, depressed mood, sleep problems, fatigue, appetite change, feelings of worthlessness and self-deprecation, concentration difficulties, psychomotor agitation, thoughts of suicide and self-harm. Each item is rated on a 4-point Likert scale from 0 “not at all” to 3 “nearly every day”. The total score is a simple summation of all item scores, where a higher total score indicates a higher level of depressive symptoms. The Cronbach’s α coefficient for the depression items of PHQ-9 was 0.826 in this study.
Social concerns related to COVID-19 infection
The latent variable of social concerns related to COVID-19 infection consisted of the following three observed variables: “I’m concerned that if I get infected with COVID-19, I’ll face blame or harm from those around me;” “I’m concerned that if I get infected with COVID-19, vulnerable individuals in my family, such as the elderly, children, or those with existing health issues, might get infected with COVID-19;” and “ I’m concerned about the economic impact of the COVID-19 pandemic on both myself and my family, including concerns about job loss or financial difficulties if I get infected with COVID-19.” Each item was rated on a 5-point Likert scale from 1 “very much” to 5 “ not at all” and was reverse-scored. A higher score indicates a higher level of concern related to COVID-19 infection.
COVID-19 infection
Being quarantined or hospitalized due to COVID-19 infection was represented by a positive response to the question, “Have you ever been quarantined or hospitalized for COVID-19 infection within the last 3 months?”
Negative changes in daily life due to the COVID-19 pandemic
Negative changes in daily life due to the COVID-19 pandemic were assessed through direct scoring, as follows: “In comparison to your daily life before the COVID-19 pandemic, where 100 points represents the pre-pandemic state and 0 points indicates a complete standstill of daily activities, what would you rate your current situation?” The scale employed 10-point intervals, with reverse scoring applied. A score of 0 denoted no change in daily life compared to that of the pre-COVID-19 pandemic, while 100 signified a complete cessation of daily activities. Higher scores indicated a more significant negative impact on daily life due to the COVID-19 pandemic.
Statistical analysis
All analyses were conducted in R software (version 4.3.1). Descriptive statistical analyses were conducted to identify the means and standard deviations of the clinical characteristics. A correlation analysis was performed to explore whether the variables included in our models were correlated after controlling for age and gender. p values <0.05 were considered significant. We performed structural equation modeling using the lavaan package to examine whether COVID-19 infection within the preceding three months and social concerns about COVID-19 infection impacted PHQ-9 scores through mediation of negative changes in daily life due to the COVID-19 pandemic. The structural equation model was controlled for age, gender, marital status, and employment status. In addition to the previously mentioned risk factors, individuals with preexisting conditions of hypertension or diabetes mellitus, who are at higher risk for severe COVID-19 symptoms, may experience increased depression due to lifestyle changes, such as spending more time at home and reduced social interaction due to fears of infection [3,20,21]. Therefore, physician-diagnosed hypertension and diabetes mellitus, as preexisting medical conditions that could influence depression during the COVID-19 pandemic, were included as covariates in the analysis. The fitness of the model was assessed using the Root Mean Square Error of Approximation (RMSEA), Comparative Fit Index (CFI), Goodness Fit of Index (GFI), and Tucker-Lewis Index (TLI). RMSEA values of 0.05 or lower indicated a close fit. The RMSEA should ideally be between 0.02 and 0.07, and the CFI and TLI should be >0.90 [22]. The significance of the indirect effects was tested using the bootstrapping method with 500 re-samples.
Two-subgroup analyses with Satorra-Bentler scaled chisquared tests and two-sample Z-tests were also conducted by categorizing the sample into age groups (under 65 years and 65 years or older) and gender groups (male and female). These subgroup analyses were used for the following reasons: to verify whether the model is statistically significant in each of the subgroups, to explore potential differences in the effects of COVID-19 infection and social concerns about COVID-19 infection on depressive mood and negative changes in daily life, and to compare the estimated coefficients of latent variables and paths coefficients between the groups. If a Z-score was >1.96 or <-1.96, it was considered statistically significant at the 5% significance level [23].
RESULTS
Demographic statistics and correlation analysis
Descriptive statistics were performed using frequency and proportion for basic characteristics in statistical analysis. The demographic characteristics and clinical questionnaire scores are presented in Table 1 as female (54.7% of the sample), age under 65 years (68.2%), married (65.6%), employed (60.6%), physician-diagnosed hypertension (27.9%), physician-diagnosed diabetes mellitus (11.7%), and COVID-19 infection within the preceding 3 months (0.5%). The prevalence of depressive symptoms, defined as PHQ-9 score of 10 or higher, was 3.1%. The results of the correlation matrices of clinical and controlled variables are presented in Table 2. Moderate correlations were observed only among the three variables related to social concerns about COVID-19 infection. Regarding the social concerns about COVID-19 infection, concern about surrounding blame positively correlated with concern about infection of vulnerable individuals (r=0.4721, p<0.001) and concern about economic impact (r=0.4120, p< 0.001). Concern about infection of vulnerable individuals positively correlated with concern about economic impact (r=0.4888, p<0.001). Scores of negative changes in daily life due to the COVID-19 pandemic (COVID-NCDL) were weakly positively correlated with concerns about surrounding blame (r=0.0981, p<0.001), concern about infection of vulnerable individuals (r=0.1021, p<0.001), and concern about economic impact (r=0.1233, p<0.001).
Demographics characteristics of participants and survey scores
Correlation matrix among variables (social concerns related to COVID-19 infection, COVID-NCDL, PHQ-9 and COVID-19 infection within the last 3 months) after controlling for gender and age
Mediation model between COVID-19 infection, social concerns about COVID-19 infection, negative changes in daily life due to the COVID-19 pandemic, and depressive mood
The results of the mediating effects of negative changes in daily life due to the COVID-19 pandemic on the association between COVID-19 infection, social concerns about COVID-19 infection, and depressive mood are shown in Table 3 and Figure 1. Table 3 presents the unstandardized coefficients, standard errors, the statistical significance of the direct and indirect effects, and standardized coefficients. The paths are presented in Figure 1, including the corresponding standardized coefficients. The model demonstrated a good fit (χ2=7,069.380, df=25.000, p<0.001, CFI=0.952, TLI=0.913, RMSEA=0.037, GFI=0.987) with controlling age, gender, marital status, employment status, physician-diagnosed hypertension, and physician-diagnosed diabetes mellitus. The structural equation model showed significant direct effects of the two independent variables (social concern of COVID-19 infection, COVID-19 infection within the last 3 months) on the outcome (PHQ-9 scores). The direct effects were statistically significant (all p values<0.01), including the effect of negative changes in daily life due to the COVID-19 pandemic on PHQ-9 (β=0.063, p<0.001).
Direct and bootstrap indirect effects in the multiple mediational models for PHQ-9 scores
Structural equation model for PHQ-9 controlled by age, gender, marital status, employment status, physician-diagnosed hypertension, and physician-diagnosed diabetes mellitus. All coefficients are standardized. *p<0.05; **p<0.01; ***p<0.001. COVID-19, coronavirus disease-2019; PHQ-9, Patient Health Questionnaire-9.
Moderation by age in the mediation model
For analysis by age, we first divided individuals younger than 65 years into two groups: those younger than 45 years (young adults) and those aged 45 years and older (mid adults). We then compared the path coefficients between these two age groups. Since there were few significant differences between the groups, we combined them into a single group (under 65 years old), performed subgroup analysis, and compared the path coefficients with those of individuals aged 65 years or older. As a result, the model fit for the under 65 years group showed χ2=4,453.702, df=22.000, p<0.001, CFI=0.948, TLI=0.905, RMSEA=0.038, and GFI=0.988. For the 65 years or old-er group, the model fit resulted in χ2=1,288.856, df=22.000, p<0.001, CFI=0.978, TLI=0.960, RMSEA=0.029, and GFI=0.992, demonstrating a good model fit in both age groups. Structural equation models for each age group (under 65 years and 65 years or older) are shown in Figure 2.
Structural equation models for age (A) under 65 years old (B) 65 years and older controlled by gender, marital status, employment status, physician-diagnosed hypertension, and physician-diagnosed diabetes mellitus. *p<0.05; **p<0.01; ***p<0.001. COVID-19, coronavirus disease-2019; PHQ-9, Patient Health Questionnaire-9.
The constrained model for measurement coefficients demonstrated a good fit (χ2=5,859.562, df=51.000, p<0.001, CFI=0.959, TLI=0.936, RMSEA=0.033, GFI=0.999). There were significant differences in the coefficient estimates of the latent variables between the under 65 years and 65 years or older groups (χ2=7.315, p=0.026). The constrained model for structural coefficients, which examined whether the path coefficients between the variables across groups are equal, exhibited a good fit (χ2=6,300.015, df=61.000, p<0.001, CFI=0.956, TLI=0.943, RMSEA=0.031, GFI=0.999). Additionally, it was confirmed that there is a moderating effects of gender among path coefficients between the two groups (χ2=440.450, p<0.001).
The comparison of path coefficients of the structural model between the two groups, along with the Z-test results, estimates, and standard errors, are presented in Table 4. Among the group aged 65 years or above, the only path that showed a higher estimate compared to those under 65 years was COVID-19 infection → COVID-NCDL (Z score=-3.2021). Other paths such as COVID-NCDL → PHQ-9 (Z score=13.3250), COVID-19 infection → PHQ-9 (Z score=6.5253), social concerns → PHQ-9 (Z score=5.4280), and social concerns → COVID-NCDL (Z score=8.9900) all indicated higher estimates for those under 65 years old than in those 65 years and older.
Comparison of path coefficients in structural equation models based on age (under 65 years old, 65 years and older) and gender
Moderation by gender in the mediation model
The results of subgroup analysis by gender show that for males, the model fit yielded χ2=2,855.159, df=22.000, p<0.001, CFI=0.955, TLI=0.917, RMSEA=0.037, GFI=0.988. For females, the model fit resulted in χ2=2,021.258, df=22.000, p< 0.001, CFI=0.974, TLI=0.952, RMSEA=0.028, and GFI=0.993, indicating a good model fit in both groups. Structural equation models for each gender group are shown in Figure 3.
Structural equation models for (A) males and (B) females controlled by age, marital status, employment status, physician-diagnosed hypertension, and physician-diagnosed diabetes mellitus. *p<0.05; **p<0.01; ***p<0.001. COVID-19, coronavirus disease-2019; PHQ-9, Patient Health Questionnaire-9.
The constrained model for measurement coefficients demonstrated a good fit (χ2=6,013.767, df=51.000, p<0.001, CFI=0.957, TLI=0.933, RMSEA=0.033, GFI=0.999). There were significant differences in the coefficient estimates of latent variables between males and females (χ2=10.624, p=0.005). The constrained model for structural coefficients exhibited a good fit (χ2=6,324.052, df=61.000, p<0.001, CFI=0.955, TLI=0.941, RMSEA=0.031, GFI=0.999). In addition, we found that there is a moderating effect of age among path coefficients between the two groups (χ2=310.29, p<0.001).
The comparison of path coefficients of the structural model between the two groups, along with the Z-test results, estimates, and standard errors, are presented in Table 4. When differences in path coefficients between male and female groups were examined, the estimate for the social concerns → COVID-NCDL path was significantly larger in males than it was in females (Z score=5.2173). In contrast, the estimates for the paths of COVID-NCDL → PHQ-9 (Z score=-11.2580) and social concerns → PHQ-9 (Z score=-5.8327) were significantly larger in females than they were in males.
DISCUSSION
In this study, both COVID-19 infection and related social concerns were significantly positively associated with depressive symptoms, the prevalence of which was 3.1%. Previous research on the prevalence of depressive symptoms during the early COVID-19 pandemic has reported figures ranging from 11.2% to 27.8% [3,11,24]. The variation in prevalence may be attributed to differences in infection rates, regulations, and perceptions of mental health problems across countries [3]. However, given that a 2018 KCHS conducted by the KCDC reported a prevalence of depressive symptoms of 2.8%, the observed increase in prevalence is consistent with previous research findings [5-7,25]. Our findings are also consistent with previous research indicating that higher COVID-19-infection-related social concerns, including economic impact, blame, and transmission to vulnerable family members, are linked to increased levels of depressive symptoms [26-28].
Using a structural equation model, we discovered that both COVID-19 infection and related social concerns indirectly influenced depressive symptoms through negative changes in daily life caused by the COVID-19 pandemic.
Each of the three social concerns may result in depressive symptoms through negative changes in daily life. For instance, concern about economic loss if one is infected with COVID-19 is likely to be associated with job absence and inability to complete one’s tasks [29,30]. In this situation, individuals may have an economic concern regarding the potential financial instability caused by COVID-19 infection. Job instability resulting from COVID-19 infection can lead to financial stress and disruption of routines and social relationships, potentially contributing to depressive symptoms [31].
Second, an individual may have concerns about the blame associated with contracting COVID-19. When the risk of a disease increases due to non-compliance with government infection prevention policies, there is a tendency to blame the victim’s morality [30]. During the COVID-19 pandemic, South Korea assessed people’s morality based on adherence to social distancing rules, such as wearing face masks, avoiding crowded places, and openly sharing their travel history [32]. The result showed that such concerns could lead to the potentially mistaken belief that non-compliance with government infection prevention policies, perceived as immoral behavior, was the cause of the infection [33]. In addition, East Asians with a background in Confucianism may be more concerned about criticism from those around them due to work absence from COVID-19 infection [34]. Therefore, in the process of keeping social distancing for avoiding blame with COVID-19 infection, people may view others as asymptomatic carriers or as potentially contaminated. Diminished personal and social trust can prompt defensive behaviors, resulting in restrictions in daily life and a reduction in social interactions [35]. Consequently, this may give rise to emotional challenges such as frustration, boredom, lethargy, and loneliness [36]. On a physical level, daily life restrictions may result in reduced physical activity or changes in circadian rhythm, which can also lead to depressive symptoms [8].
Lastly, individuals may have concerns about transmission of COVID-19 infection to vulnerable family members, including the elderly, children, and patients with disease, who may have a higher risk of severe symptoms, morbidity, and mortality with COVID-19 infection [37]. This problem may be more pronounced in South Korea, where the proportion of the elderly population has been steadily increasing since 2001 [38]. In an effort to prevent the infection of vulnerable populations within families, restrictions on social interactions and daily activities are imposed on both the vulnerable individuals and their families, which potentially makes vulnerable people more dependent on their families [39]. Especially during the early stages of the COVID-19 pandemic in 2020, the lack of remote systems for social support and access to health services exacerbated caregivers’ physical and mental health, which contributed to caregiver burden and burnout [40]. This situation may ultimately lead to depressive symptoms among caregivers of vulnerable populations [39,40].
Our results showed that negative changes in daily life are closely associated with depressive symptoms. During the COVID-19 infection period, individuals undergoing isolation may have experienced heightened psychological distress and depressive symptoms [41]. Moreover, constraints on visitation and reduced social support may have exacerbated these symptoms [42]. Workers, especially those in transportation, food, personal care, and service occupations, may have experienced heightened stress and economic issues during the infection period [29]. These stressors may lead to increased negative changes in daily life and potentially contribute to depressive symptoms. COVID-19 survivors often suffer reduced muscle strength, joint mobility, and respiratory capacity, along with somatic symptoms like pain and dyspnea [43]. All of these side effects can significantly impair both basic activities of daily living and instrumental activities of daily living [43,44]. Such decreases in activities of daily living can diminish autonomy, independence, self-esteem, and quality of life, potentially contributing to depressive symptoms [44].
In this study, the total effect of social concerns (including economic problems, fear of infecting vulnerable individuals, and social blame for COVID-19 infection) had a greater impact on depressive symptoms than did COVID-19 infection itself. Our data represent the early stages of the COVID-19 pandemic, which was characterized by high mortality and morbidity but relatively low infection rates [45,46], as well as the absence of vaccines or treatments, strict infection prevention policies [47,48], and a rapid economic recession [49]. Therefore, these results may indicate that fear of COVID-19 infection during the early stages of the pandemic may be significantly associated with depressive symptoms.
Furthermore, in this model, subgroup analyses were conducted based on gender and age. When comparing path coefficients by age, the impact of COVID-19 infection on negative changes in daily life was more significant in those aged 65 years or older than it was in younger individuals. These findings are consistent with the results of a previous study, indicating that elderly patients experience a higher severity of COVID-19 infection, resulting in a greater prevalence of long-term sequelae and an impact on their quality of life [5,50]. Although older adults are more vulnerable to the virus, the path coefficients associated with depressive symptoms and those from social concerns about COVID-19 infection to negative changes in daily life were notably larger in the group under 65 years than they were in the group aged 65 years or older. This finding is consistent with recent research results suggesting that young people may experience higher rates of depressive symptoms than older individuals [16,17,51], possibly due to social role pressures resulting from factors such as career and academic demands [16,17]. Additionally, young adults may be more vulnerable to the effects of the COVID-19 pandemic due to factors such as greater media exposure, the impact of financial crises, managing workload responsibilities, and less effective coping strategies compared to older adults [11].
Regarding gender, the path coefficient for social concerns about COVID-19 infection influencing negative changes in daily life due to the COVID-19 pandemic was larger in males than in females. While males tend to experience greater daily life changes because of their more frequent engagement in social activity compared to females [52], it is noteworthy that the path coefficient related to depressive symptoms was larger in females than in males. This suggests that depression in females may be more significantly affected by concerns and negative life changes associated with COVID-19, explaining for the observed gender difference in COVID-19-related depression. These findings are consistent with previous research showing that females are significantly affected psychologically by the COVID-19 pandemic [16,17,53]. Although objective risks such as COVID-19 morbidity and mortality are higher in males, emotional responses are more pronounced in females [54]. This suggests that factors beyond the severity of COVID-19 infection may influence emotional reactions. During the COVID-19 pandemic, females were more concerned about family health and well-being than were males [15]; therefore, females’ predominant role as family caregivers may increase their vulnerability to social isolation and distress [14,55]. Additionally, from a biological standpoint, females tend to have larger endocrine, affective, and arousal responses to stress than males, which may lead to a greater susceptibility to social isolation [14].
This study has several limitations. First, this study is the cross-sectional design; therefore, the results are exploratory and should be interpreted with caution. While there were differences in lifestyle and prevalence of depressive symptom between the pre-pandemic and pandemic periods, there were also shifts in these patterns during the pandemic due to variations in COVID-19 policy intensity and perceptions of the disease [3,56]. Therefore, it may be difficult to generalize the findings of this study, as the survey was conducted during a period of few COVID-19 infections and, as a cross-sectional study, may not fully capture these changes. Second, only a self-reported evaluation scale was used to evaluate psychological symptoms. Objective and structured interviews were not conducted, making it difficult to ensure the objectivity and validity of the psychiatric conditions reported. Use of a self-rated 0–100 scale to measure negative changes in daily life may not fully reflect all types of negative changes experienced by individuals. Third, despite adherence to personal hygiene rules by interviewers and participants (such as hand disinfection, mask-wearing, and social distancing), many people did not voluntarily participate in the survey, indicating selection bias.
Overall, this study is the first to examine the mediating effect of negative changes in daily life on the impact of COVID-19 infection and related social concerns on depressive symptoms in South Korea using structural equation modeling. Our findings demonstrate that social concerns about COVID-19 infection have a more significant impact on depressive symptoms than does COVID-19 infection itself. The large sample size of this study enhances its generalizability, and subgroup analyses based on age and gender underscore the importance of tailored interventions to address depressive symptoms during the COVID-19 pandemic. Lastly, another strength of this study is its reflection of the general population, especially considering that individuals aged 65 years and older represented 16.4% of the total population in 2020 [57].
Notes
Availability of Data and Material
The datasets generated or analyzed during the study are available from the corresponding author on reasonable request.
Conflicts of Interest
The authors have no potential conflicts of interest to disclose.
Author Contributions
Conceptualization: Weon-Jeong Lim. Data curation: Weon-Jeong Lim. Funding acquisition: Weon-Jeong Lim. Investigation: Dham Ho, Sun-Young Kim. Methodology: Dham Ho, Sun-Young Kim, Hye Ah Lee, Hyunsun Cho. Project administration: Weon-Jeong Lim. Software: Hyunsun Cho. Supervision: Sun-Young Kim. Writing—original draft: Dham Ho. Writing—review & editing: all authors.
Funding Statement
This work was supported by the Ewha Womans University Research Grant of 2024.
Acknowledgements
We acknowledge the fundings that support our study and the participants involved in this study.