INTRODUCTION
Population aging presents a significant societal challenge and has attracted substantial global attention [1]. According to the World Health Organization [2], the proportion of older adults worldwide is expected to increase to 22% between 2015 and 2050. In South Korea (hereafter Korea), the proportion of older adults aged ≥65 years in rural areas exceeded 14% in 2000, marking the transition to an aging society, and is expected to surpass 20%—defining a “super-aged society”—by 2025 [3]. The average life expectancy in Korea is 79.9 years for males and 85.6 years for females, exceeding the Organization for Economic Co-operation and Development (OECD) averages by 1.9 and 2.4 years, respectively [4]. As life expectancy and old age duration increase, older adults face diverse physical, psychological, and socioeconomic challenges.
Aging and chronic illnesses often result in diminished physical, cognitive, and social functioning, complicating independent living. Concurrently, reduced income and the loss of social roles after retirement contribute to shrinking social networks, while spousal bereavement can precipitate psychological distress, including loneliness and depression [5]. The 2021 Korea National Health and Nutrition Examination Survey (KNHANES) reported a 12.8% prevalence of depressive symptoms among older adults aged ≥65 years, exceeding the 10.8% observed in the general adult population aged ≥19 years [6].
Multiple factors influence depression in older adults, including economic status, health and functional capacity, social exclusion, and social support [7-9]. Among these, social exclusion is a critical determinant [9]. It encompasses multidimensional deprivation, including poverty arising from inadequate access to economic resources, housing, education, health care, and social services, affecting specific population groups [9,10]. As individuals age, access to resources and social participation decline, resulting in inequality and reduced quality of life [10]. In Korean society, concerns regarding older adults increasingly extend beyond poverty to encompass psychological issues, such as depression, loneliness, and suicide [5]. Therefore, examining social exclusion factors that affect older adults’ mental health is imperative.
With the extension of old age and intensification of associated health concerns, research has increasingly focused on social exclusion and its effects among older adults [7-9,11]. Prior studies have demonstrated that social exclusion elevates the risk of depression and suicidal ideation and exacerbates depressive symptoms, particularly among older adults living alone [7,11,12]. For example, Tong and Lai [8] reported that older adults in China experiencing social exclusion were twice as likely to suffer chronic diseases compared to those not excluded. These findings underscore the multifaceted nature of social exclusion and its pervasive impact on older adults’ lives.
For older adults experiencing depression linked to social exclusion, digital device use may mitigate loneliness [13,14]. Digital communication and information exchange have been proven to effectively reduce feelings of isolation and depression [13]. Online interactions enable older adults to prevent social isolation by maintaining and potentially strengthening social connections [15]. Conversely, limited digital literacy or access may exacerbate communication difficulties and social isolation, thereby increasing the risk of depression and other mental health disorders [13,16].
Differences in health status and digital device use have been observed among adults aged ≥65 years [17]. The old-old population tends to exhibit greater dependence due to deteriorating health and physical decline, compounded by economic hardship and reduced social networks following the loss of spouses or peers [18]. They are more vulnerable to social isolation and report higher levels of anxiety, depression, and feelings of loss than their young-old counterparts [17,19]. Domestic studies similarly report significantly higher depression levels among old-old adults, which increase with age and threaten quality of life [19]. These findings highlight the need to identify factors that can alleviate these adverse outcomes.
This study aims to examine the relationship between social exclusion and depression among young-old and old-old adults, investigating the moderating role of digital device use as a protective factor mitigating the effects of social exclusion on depression. It further seeks to provide a foundation for developing interventions and programs to reduce depression resulting from social exclusion in older adults. Specifically, this study intends to identify the degree of social exclusion—including economic exclusion, social activity exclusion, social relationship exclusion, health exclusion, and housing exclusion—and depression among young-old and old-old adults. Furthermore, it examines the moderating effect of digital device use on the relationship between social exclusion and depression.
METHODS
Ethics statement: This study was approved by the Institutional Review Board (IRB) of the Public Institutional Review Board (No. P01-202404-01-039). Informed consent was obtained from participants.
1. Study Design
This study is a secondary data analysis utilizing data from the 2020 National Survey of Older Koreans. It compares young-old and old-old adults and examines the moderating effect of digital device use on the relationship between social exclusion and depression. The study adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (https://www.strobe-statement.org).
2. Study Participants
Data were derived from the 2020 National Survey of Older Koreans [20], conducted triennially since 2007 under Article 5 of the Welfare of Senior Citizens Act. It was first administered to adults aged ≥65 years in 2008 to assess the health status of older Koreans. The 2020 survey—the fifth round—was administered by 169 trained interviewers between September 14 and November 20, 2020, using a tablet-personal computer (PC) Assisted Personal Interview (TAPI) one-on-one direct interview method based on a researcher-developed questionnaire. The target population included adults aged ≥65 years residing in general housing across 17 metropolitan and provincial regions. Participants were selected via stratified two-stage cluster sampling method. Of 10,097 participants, 9,930 with complete data were included in the final analysis. Young-old adults were defined as individuals aged 65~75 years, and old-old adults as those aged ≥75 years, according to Neugarten’s criteria [21]. Accordingly, the sample included 5,656 young-old and 4,274 old-old adults.
3. Study Variables
1) Sociodemographic Characteristics
Five sociodemographic variables were assessed: sex (male, female), age group (young-old [65~74 years], old-old [≥75 years]), marital status (having a spouse, not having a spouse), family type (single-person, non-single-person, or other household [e.g., other older adult households or long-term care facility]), and education level (elementary school or below, middle school, high school, college or higher).
2) Independent Variable: Social Exclusion
Social exclusion was measured across five domains: economic exclusion, social activity exclusion, social relationship exclusion, health exclusion, and housing exclusion [22]. Each domain was coded dichotomously (presence=1, absence=0), and the total social exclusion score (range, 0~5) was computed by summing the domains.
Economic exclusion was evaluated with income and satisfaction with economic status. Household income was adjusted for household size to calculate equivalized income. Participants with equivalized income below 60% of the median were classified as “excluded” (1 point), while those at or above this threshold were classified as “not excluded” (0 points). Economic satisfaction was evaluated by the question, “How satisfied are you with your current economic status?” Responses of “not satisfied” or “not satisfied at all” were coded as “excluded” (1 point), whereas responses of “very satisfied,” “satisfied,” or “neutral” were coded as “not excluded” (0 points). Participants meeting the exclusion criteria on either measure were classified as economically excluded.
Social activity exclusion was determined based on participation in leisure and social activities, including cultural, educational, group, or religious activities. Participants who responded “no” to the question, “In the past year, have you engaged in a leisure or cultural activity (excluding watching television [TV], listening to the radio, and traveling)?” were coded as “excluded” (1 point), and those who answered “yes” were coded as “not excluded” (0 points).
Social relationship exclusion was evaluated by frequency of contact with children, siblings/relatives, and friends/neighbors/acquaintances. Participants without co-residing children who communicated rarely (less than 1~2 times over 3 months) with non-co-residing children were coded as “excluded” (1 point), whereas those co-residing with children or communicating at least four times per week were coded as “not excluded” (0 points). Similar criteria were applied to contact with siblings/relatives and friends/neighbors/acquaintances. Participants meeting exclusion criteria in any of these groups were classified as socially excluded.
Health exclusion was assessed through subjective health status and functional limitations in Activities of Daily Living (ADLs) and Instrumental Activities of Daily Living (IADLs). Participants who rated their general health as “poor” or “very poor” were coded as “excluded” (1 point), while those reporting “very good,” “good,” or “fair” health were coded as “not excluded” (0 points). Participants meeting the exclusion criteria on any of these indicators were classified as experiencing health exclusion. Functional limitations were assessed via assistance needed in ADLs and IADLs over the past week. Participants requiring “complete” or “partial assistance” with tasks, such as dressing, washing face/hair and brushing teeth, bathing/showering, eating, getting up and going outside the room, and continence were considered “excluded” (1 point), while those fully independent were coded as “not excluded” (0 points). Similarly, assistance needed for IADLs—including personal grooming, housework, meal preparation, medication management, financial tasks, communication, and transportation—was coded as exclusion if participants required any assistance, and full independence was coded as not excluded. Meeting exclusion criteria in either ADLs or IADLs constituted health exclusion.
Housing exclusion was defined by home ownership status and satisfaction with current housing. Participants who did not own their residence (either individually or jointly with a spouse) or were leasing were coded as “excluded” (1 point), whereas homeowners were coded as “not excluded” (0 points). Housing satisfaction was assessed by the question, “How satisfied are you with your current housing overall?” Reponses of “dissatisfied” or “very dissatisfied” were coded as “excluded” (1 point), while responses of “very satisfied,” “satisfied,” or “neutral” were coded as “not excluded” (0 points). Participants meeting exclusion criteria on either ownership or satisfaction were classified as housing excluded.
3) Dependent Variable: Depression
Depression was measured using the short version of the Geriatric Depression Scale, which consists of 15 items that assess aspects, such as current life satisfaction, feelings of emptiness, lack of motivation, boredom, anxiety, and cheerfulness. Response options were either “yes” or “no.” Five positively-worded items (“Are you generally satisfied with your life?,” “Do you normally feel refreshed?,” “Do you usually feel cheerful?,” “Do you currently find being alive enjoyable?,” and “Do you feel you have good energy?”) were reverse coded. Total scores ranged from 0 to 15, with higher scores indicating more severe depressive symptoms.
4) Moderating Variable: Digital Device Use
Digital device use was assessed with the question, “Do you use a PC, mobile phone, tablet PC to do any of the following activities: receive or send messages (text, KakaoTalk, Telegram, etc.), browse online (news, weather, etc.), listen to music (MP3, radio, etc.), take photos or videos, play games, watch videos (movies, TV programs, YouTube, etc.), use social networking services (blogs, Facebook, Instagram, etc.), engage in e-commerce (online shopping, reservations, etc.), complete financial transactions (online banking, securities, etc.), search for and install applications, and others (total of 12 items)?” Responses were coded as either “yes” or “no.” Each “yes” was converted into a count variable and then added together in the final analysis, with higher scores indicating greater digital device use.
4. Data Collection and Ethical Considerations
The data from the 2020 National Survey of Older Koreans are publicly available. Approval to use the microdata was obtained from the Korea Institute for Health and Social Affairs (https://data.kihasa.re.kr). Data were collected anonymously via TAPI, with unique identifiers assigned to protect confidentiality. This study received approval from the Public Institutional Review Board (IRB No. P01-202404-01-039).
5. Statistical Analysis
Weighted complex sample analyses were conducted using SPSS/WIN 26.0 (IBM Corp.) to account for the complex sampling design. Descriptive statistics (frequency, percentage, mean, standard deviation) was employed to characterize participants’ sociodemographic variables, social exclusion, and depression. Group differences in depression by sociodemographic variables were examined using independent t-tests, chi-square tests, and one-way analysis of variance. Hierarchical regression analysis was performed to evaluate the effect of social exclusion on depression and the moderating effect of digital device use.
RESULTS
1. Differences in Sociodemographic Characteristics and Social Exclusion Between Young-Old and Old-Old Adults
A comparison of sociodemographic characteristics and degrees of social exclusion between young-old and old-old adults revealed significant differences across all variables (p<.05). The old-old group had higher proportions of females, individuals without a spouse, those living alone, and those with lower education level than the young-old group. Additionally, they exhibited greater degrees of social exclusion across all five subdomains (Table 1).
Descriptive statistics summarized social exclusion, depression, and digital device use for both groups (Table 1). Data normality was confirmed, with skewness values below |2| and kurtosis values below |7| [23]. The mean number of excluded indicators across the social exclusion subdomains was 1.56±1.16 points. By subdomain, the mean scores were as follows: 0.62±0.49 points for economic exclusion, 0.19±0.39 points for social activity exclusion, 0.29±0.45 points for social relationship exclusion, 0.24±0.43 points for health exclusion, and 0.22±0.42 points for house exclusion. The mean scores for depression and digital device use were 3.48±3.37 points and 3.34±3.19 points, respectively.
2. Depression by Sociodemographic Characteristics and Social Exclusion in Young-Old and Old-Old Adults
Depression scores for both groups differed significantly according to sex, marital status, family type, education level, and social exclusion (p<.05). Females, those without a spouse or co-habiting partner, and those with lower education level showed higher levels of depression. Moreover, older adults who experienced economic, social activity, social relationship, health, or house exclusion had significantly higher scores than those who did not (Table 2).
3. Effects of Social Exclusion on Depression and the Moderating Role of Digital Device Use in Young-Old and Old-Old Adults
The relationship between social exclusion and depression between young-old and old-old adults as well as the moderating role of digital device use were examined (Table 3). Among the control variables, sex was categorized with males as the reference group, and educational attainment with college graduates or higher as the reference group. Marital status and family type were highly similar, and thus “living alone” was selected as the reference group to avoid multicollinearity. Model 1 included control variables, such as sex, family type, and education level in addition to social exclusion and digital device use. Meanwhile, an interaction term between social exclusion and digital device use was added in Model 2 to test the moderating effect. The Durbin–Watson statistic was close to 2 for both groups, confirming the independence of the residuals. The variance inflation factor was below 10 for all variables, indicating no multicollinearity issues.
Young-old adults with spouse had lower levels of depression compared to those who were single (β=-.07, p<.001), while those with lower levels of education (high school or below) had higher levels of depression than those with a college degree or higher (β=.04, p=.004). Furthermore, social exclusion exhibited a significant positive effect on depression (β=.23, p<.001) while digital device use had a significant negative effect (β=-.12, p<.001). This indicates that young-old adults who experienced a greater degree of social exclusion and used digital devices less reported higher levels of depression. Meanwhile, in Model 2, explanatory power did not increase, and the added interaction term was not significant, suggesting that digital device use did not moderate the effect of social exclusion on depression in young-old adults.
Among old-old adults, those without a spouse (but not necessarily single) exhibited higher levels of depression compared to those living alone (β=.08, p<.001). Social exclusion had a significant positive effect on depression (β=.37, p<.001), whereas digital device use had a significant negative effect (β=-.05, p=.002), indicating that higher social exclusion and lower digital device use correspond to increased depression levels among old-old adults. In Model 2, explanatory power increased slightly from 16.0% to 16.3% (adjusted R2= 15.9%→16.1%), and the interaction term was significantly negative (β=-.06, p=.001), indicating that digital device use moderates and attenuates the effect of social exclusion on depression. Figure 1 illustrates this moderation: depression rises more steeply with social exclusion among individuals with low digital device use, whereas those with high usage show a comparatively flatter increase despite similar levels of social exclusion. This suggests that the impact of social exclusion on depression in older adults varies according to their level of digital device use.
4. Effects of Social Exclusion Subdomains on Depression and the Moderating Role of Digital Device Use in Young-Old and Old-Old Adults
Analysis of the five social exclusion subdomains revealed that digital device use significantly moderated housing exclusion in young-old adults (β=-.04, p=.010), while it moderated economic (β=-.05, p=.004), social relationship (β=-.05, p=.008), and house (β=-.04, p=.031) exclusion in old-old adults (Table 4).
DISCUSSION
This study examined the moderating effect of digital device use on the relationship between social exclusion and depression among young-old and old-old adults, utilizing data from the 2020 National Survey of Older Koreans. It aimed to deepen understanding of mental health disparities between these age groups and to highlight the necessity for targeted interventions in preventing and detecting depression based on digital device use. Findings indicated that depression levels were higher in old-old adults than in their young-old counterparts, and that social exclusion was positively associated with depression in both groups. Moreover, digital device use significantly attenuated the impact of social exclusion on depression in old-old adults, confirming its moderating role.
Consistent with prior research, depression was more severe in the old-old adults than in the young-old adults [16,17-19]. Social exclusion was significantly associated with elevated depression in both cohorts, corroborating previous studies that identify social exclusion as a critical factor contributing to depressive symptoms [5,11]. Older adults who experience restricted access to rights and opportunities in employment, healthcare, and social participation not only face economic hardship but also suffer adverse physical and mental health outcomes [10,11]. As social roles diminish with age, experiences of alienation and discrimination may increase, further exacerbating depression and suicidal ideation [11]. Therefore, preventive policies aimed at enhancing societal perceptions of older adults and providing them with meaningful social roles are essential to mitigate depression related to social exclusion.
The moderating effect of digital device use on the social exclusion–depression relationship was observed exclusively among old-old adults. Specifically, digital device use buffered the adverse effect of social exclusion on depression in this group, aligning with evidence that digital engagement can alleviate depressive symptoms and improve life satisfaction in older adults [14,16]. Given the greater extent of social exclusion often experienced by old-old adults, digital literacy may be pivotal in mitigating and preventing depression. Conversely, no such moderating effect was evident in young-old adults, possibly reflecting their generally higher levels of physical and social activity, greater independence, and more opportunities for forming new social connections or engaging in activities [24]. While health perception contributes to life satisfaction, participation in social activities exerts an even stronger influence in the young-old adults [24]. Their relatively favorable health and activity levels may reduce reliance on digital devices for social connectivity [25].
Digital literacy—the capacity to effectively utilize technology—has become increasingly important in promoting active aging within a digitally driven society [15-17]. Digital device use is no longer optional but essential for maintaining independence and expanding social networks online [14,15]. Additionally, digital access increasingly reflects social inequalities, as digital vulnerability can exacerbate disparities by limiting information access and social participation [14,15,26]. Treating older adults as a homogeneous group in digital inequality research neglects the fact that subgroups, particularly old-old adults, are disproportionately vulnerable to social exclusion [15,27]. Many old-old adults retired before widespread technology adoption, limiting opportunities to acquire digital skills and likely contributing to greater underlying difficulties [27].
Among the subdomains of social exclusion, digital device use moderated the relationship between house exclusion and depression in both young-old and old-old adults. Low levels of digital device use were associated with a stronger adverse effect of house exclusion on depression, whereas higher usage attenuated this effect. This suggests that digital device use may help alleviate feelings of social isolation or psychological withdrawal linked to inadequate housing. A similar moderating effect was observed for the relationship between economic and social relationship exclusion and depression, but only among old-old adults. When digital device use was low, economic and social relational exclusion exerted stronger negative effects on depression; conversely, higher digital use mitigated these effects. Given the critical role of economic resources in older adult welfare and the impact of social relationships on mental health, digital device use appears to function as a significant protective factor [11].
This study holds academic significance by extending prior research on the association between social exclusion and depression, demonstrating the moderating role of digital device use [11]. It also provides an empirical foundation for interventions aimed at enhancing digital literacy among old-old adults. Based on these findings, several practical recommendations are proposed, including strengthening mental health monitoring through early intervention and targeted screening of high-risk groups, alongside offering diverse psychological support services. Remote counseling and digital depression prevention programs may also offer emotional support and facilitate the development of online social networks.
Supporting old-old adults in adapting to the digital age and reducing daily inconveniences requires educational programs and policies that facilitate effective digital device use. In rapidly aging societies, limited digital literacy among older adults exacerbates the intergenerational digital divide, leading to digital exclusion [27]. Contrary to common assumptions that older adults are indifferent to technology, this indifference often results from a lack of technology designed specifically for their needs rather than inherent resistance [28,29]. Such stereotypes risk perpetuating digital exclusion unless older adults are actively educated and engaged in the development and dissemination of digital technologies [14,28,29]. Accordingly, digital programs should be tailored to this population, considering prevalent physical limitations such as visual impairment and reduced fine motor skills. These adaptations can reduce apprehension toward unfamiliar devices and prevent experiences of inconvenience or discrimination. Governments should also promote voluntary corporate initiatives supporting digital inclusion through mechanisms such as digital participation pledges.
The absence of digital education and policies addressing older adults’ specific needs further contributes to social exclusion [18]. Having exited formal education and workplace learning environments, older adults have fewer opportunities for acquiring new skills relative to younger cohorts [29]. Community-based educational services offering digital literacy programs as part of social consumer education are therefore recommended to facilitate intergenerational communication [28,29]. Such services could be expanded by deploying staff at community and welfare centers to provide digital skills training, including communication, online transactions, and cyber awareness. Level-based instruction in small groups or one-on-one mentoring with younger facilitators may foster mutual understanding and reduce intergenerational conflict while enhancing knowledge transfer. Digital education should equip old-old adults not only with technical proficiency but also with critical evaluation skills to discern reliable information sources [14]. To safely navigate the digital environment, emphasis should be placed on privacy protection, online safety, security measures, and awareness of cybersecurity threats such as phishing and fraud [30]. Reliable internet access and digital competence empower old-old adults to manage finances effectively, engage confidently in digital contexts, and mitigate feelings of exclusion. These efforts are essential to bridging the digital divide and fostering an inclusive digital society [27]. Furthermore, integrating digital tools into healthcare services—such as remote health monitoring, online appointment scheduling, and health-related mobile applications—holds promise for enhancing both the physical and mental well-being of old-old adults.
CONCLUSION
This study examined the moderating effect of digital device use on the relationship between social exclusion and depression among young-old and old-old adults. Findings indicated that social exclusion was positively associated with depression in both groups, while digital device use moderated this effect only in old-old adults. These results underscore the potential role of digital engagement in mitigating depression related to social exclusion in older adults. Empirical evidence supports digital device use as a valuable component in future welfare policies targeting the mental health of the old-old population.
Several limitations warrant consideration. First, although social exclusion was conceptualized as a multidimensional construct encompassing employment, economy, education, public services, and healthcare, not all domains were exhaustively examined. Future studies should apply comprehensive classification criteria across these areas. Second, given the cross-sectional design and use of large-scale survey data, causal inferences are limited; longitudinal research with time-lagged measurements is needed to clarify causal pathways. Lastly, variations in digital literacy and skills among older adults were not analyzed in detail. Future research should investigate barriers to digital access and usage, as well as the broader physical and mental health effects of digital engagement.
Further investigation into the long-term impact of social exclusion subdomains on older adults’ lives, alongside mixed-methods research focusing on those experiencing social exclusion, may deepen understanding of its multidimensional nature.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation 
Print 
