Introduction

Background

According to statistics from the World Health Organization, as of 2021, healthy life expectancy and average life expectancy in Korea are reported to be 72.5 years and 83.8 years, respectively [1]. With this increase in average life expectancy, in recent years, the number of elderly people admitted to long-term care facilities has also been continuously increasing [2]. However, it has been reported that a significant proportion of elderly residents in long-term care facilities suffers from chronic diseases, such as hypertension, diabetes, severe geriatric diseases, and heart disease, and 30% of them experience a decline in mental health, including cognitive function [3]. Specifically, levels of death anxiety and depression have been found to be high among elderly residents in long-term care facilities due to separation from the family, physical frailty, and social isolation [4]. These emotional problems, such as death anxiety and depression, can hinder the development of ego integrity, a positive psychological developmental process in old age [5,6], and eventually are likely to lead to a decline in life satisfaction and deterioration in mental health.

Admission to a long-term care facility is an experience that causes significant changes and psychological shock in the lives of the elderly, and has a significant impact on their mental health [2,3]. A number of elderly people regard admission to long-term care facilities as modern-day ‘Goryeojang’, which refers to the ancient practice of abandoning and leaving elderly parents to starve to death in a remote mountain or wilderness during the Goryeo Dynasty, and perceive it as a situation of social isolation and disconnection, which leads to depression and death anxiety among elderly residents in long-term care facilities [7]. Unlike general elderly people, who tend to show an increasing level of ego integrity with age, elderly residents in long-term care facilities have been shown to have difficulty in developing ego integrity due to factors such as death anxiety [7,8]. Therefore, there is a need to systematically investigate the physical and mental health of elderly people residing in long-term care facilities, which are a limited living environment, and develop evidence-based nursing intervention programs to manage their health in an integrated manner.

So far, various studies have been conducted to address mental health problems in elderly residents in long-term care facilities, and social support has been found to positively affect the alleviation of negative emotions such as depression and death anxiety in the elderly [4]. Social support has been evaluated as a key factor affecting mental health in the elderly, and depending on situations, it can act as a resource that buffers the negative effects of stress [6]. However, regarding the impacts of each type of subfactors of support, such as family support, friend support, and professional support, on psychological factors among elderly residents in long-term care facilities, previous studies reported conflicting results, and these findings have not been sufficiently verified in a systematic manner. It has been reported that residential satisfaction among elderly residents in long-term care facilities may vary depending on their relationships with the healthcare professionals, staff, and peers that live together with them [9]. These research results suggest that social interactions in the environment of long-term care facilities have a significant impact on the psychology and adaptation of elderly residents. In particular, considering the characteristics of elderly residents in long-term care facilities who experience psychological anxiety and loneliness due to isolation from the family, there is a need to systematically investigate and review the impact of social support in order to provide effective nursing services to them.

The ‘Stress Process Model’ is a theoretical model developed to explain the effects of stressors and coping mechanisms on mental health, and has been applied to various studies including research on mental health in the elderly [10-13]. According to this model, individuals’ social characteristics affect their stress experiences and levels of resources such as social support, and these factors have a significant impact on physical and mental health outcomes [10-13]. In particular, social and psychological resources can act as protective factors that can buffer the negative effects of stress [13]. In previous studies applying the Stress Process Model, it has been found that stress is closely related not only to negative mental health indicators such as depression and suicidal ideation but also to positive indicators such as happiness and parenting satisfaction [14-16]. In addition, studies of elderly people have shown that resources such as social support, ego integrity, and coping strategies can act as mediating or moderating variables in the relationship between stress and mental health [11,12]. As described above, the Stress Process Model can be presented as a theoretical framework that can be used to comprehensively explain the relationships between an individual’s social characteristics, resources, stress exposure, and psychological and physical health [17,18].

However, in previous studies applying the Stress Process Model, there has been a lack of research that sets death anxiety as a stressor and depression and ego integrity as outcome variables, and comprehensively verifies the moderating effect of social support among people residing in the special environment of long-term care facilities. A comprehensive analysis of the relationships between death anxiety, depression, ego integrity, and social support can provide important evidence or data for developing strategies to improve mental health and quality of life in the elderly. In particular, considering limited family and social networks of elderly residents in long-term care facilities, verifying whether social support acts as a protective factor has significant academic and practical implications.

Against this backdrop, this study set death anxiety as an independent variable (a stressor) and depression and ego integrity as dependent variables (outcomes), based on the Stress Process Model proposed by Turner [13] to examine the effects of death anxiety on depression and ego integrity among elderly residents in long-term care facilities. In addition, social support was set as a moderator to comprehensively investigate the relationships between the major variables when social support is provided. In particular, this study classified social support into the subdomains of family, friend, and professional support and analyzed how each subdomain of social support moderates the impact of death anxiety on mental health with the aim of providing foundational data for the development of mental health nursing services for elderly residents in long-term care facilities.

Theoretical framework

So far, various models have been proposed to examine the impact of stressors and stress coping strategies on mental health in the elderly. A representative model is the ‘Stress Process Model,’ which was developed by complementing the limitations of the physical stress model and the life stress model [18]. The Stress Process Model proposed by Pearlin et al. [18] is a model that explains how stressors and an individual’s coping strategies and resources interact to influence health or behavioral outcomes. This model is composed of stressors, outcomes resulting from stressors, background information influencing stress, and mediators between stressors and outcomes [18]. However, the original Stress Process Model proposed by Pearlin et al. [18] focuses on individuals’ stress experiences and coping processes, and although it includes some social contexts, this early model has limitations in that it fails to sufficiently explain the impacts of structural differences or inequalities among various population groups on stress experiences and outcomes. Therefore, various researchers applied a modified form of the Stress Process Model by modifying it to suit their own research objectives. In particular, Turner [13] expanded the model developed by Pearlin et al. [18] to present a more sophisticated analytical framework that comprehensively considers individuals’ social and economic factors as well as health-related variables. In particular, in the Stress Process Model proposed by Turner [13], relationships between stressors and health outcomes are explained through a structure in which social characteristics have a direct impact on social and personal resources as well as stress exposure, and this stress exposure in turn affects an individual’s physical and mental health as well as social and personal resources [19]. This model can be usefully utilized as a basis for developing programs to provide effective nursing services since it allows us to analyze the relationships between stressors and mental health among elderly people, and identify the factors related to social characteristics that influence the process [10,12].

The theoretical framework of this study was constructed based on the model proposed by Turner [13] by considering social background and contextual factors that can influence the overall life of the elderly, such as gender and subjective health status. Based on this model, in this study, death anxiety, a stressor, was set as an independent variable, and depression and ego integrity were set as dependent variables, and the present study attempted to investigate the moderating effect of social support in the relationships between the three variables in order to derive implications for providing nursing services to elderly residents in long-term care facilities.

Aim and objectives

The main purpose of this study was to derive implications that should be considered in providing nursing services to elderly residents in long-term care facilities. To achieve this aim, specific research objectives were set as follows:

1) To investigate the individual characteristics, the levels of death anxiety, depression, ego integrity, and social support among elderly residents in long-term care facilities through a survey;

2) to analyze correlations between anxiety, depression, ego integrity, and social support among elderly residents in long-term care facilities;

3) to investigate the impact of death anxiety on depression and ego integrity;

4) to verify whether social support has a moderating effect on the relationships between death anxiety, depression, and ego integrity, and to analyze the relationships between the variables using the Stress Process Model proposed by Turner [13] as a theoretical framework.

Based on these analysis results, this study aimed to comprehensively investigate the relationships between death anxiety, depression, ego integrity, and social support among elderly residents in long-term care facilities, and provide basic data for the development of nursing intervention programs.

Methods

Study design

This study is a cross-sectional study to examine the levels of depression and ego integrity among elderly residents in long-term care facilities, and to investigate the moderating effect of social support on the impact of death anxiety on depression and ego integrity. Data collection for this study was conducted after receiving approval from the Institutional Review Board of Keimyung University (IRB No: 40525-202206-HR-031). In addition, after obtaining prior consent regarding data collection from the directors of the relevant long-term care facilities, a survey was conducted among the participants who voluntarily agreed to participate.

Participants

In this study, to recruit participants, 7 facilities in P-si, Gyeongsangbuk-do were selected by convenient sampling. Specifically, 6 long-term care hospitals were selected among 28 long-term care hospitals located in P-si, Gyeongsangbuk-do, and 1 long-term care facility was selected among 27 long-term care facilities located in P-si that provide long-term care recipients with services such as protection of residents in the facility, physical activity support, and education and training for maintaining mental and physical functions. The participants of this study were elderly people aged 65 or older living in the selected care facilities who voluntarily agreed to participate in this study, and met the following inclusion criteria.

1) Elderly people who were able to understand the survey questions and answer them;

2) elderly people whose reason for admission to a long-term care hospital or facility was not mental illness or dementia;

3) elderly people who have been residing in a long-term care hospital or facility for 3 months or more.

Regarding the admission period, participants were limited to elderly people who have been residing in a long-term care facility for three months or more, because it has been reported that psychological instability is high during the initial stages of admission in the processes of adapting to a new environment and forming relationships, and this psychological instability can cause temporary fluctuations in emotional and stress levels [20]. In addition, according to the study by Lee and Rantz [21], since the level of activities of daily living (ADL) among elderly residents in long-term care facilities was found to become stable at 3 months after admission to a care facility, evaluations at least 3 months after admission can more consistently reflect the emotional and physical conditions of the participants. Therefore, in this study, elderly people who had been residing in a long-term care facility for 3 months or more were set as the target group in order to collect data at the time when the participants had psychologically adapted to living in a long-term care facility and had become psychologically stable.

The sample size of the study was calculated using G power 3.1. The sample size for linear multiple regression was calculated using an effect size of .10, an α error of .05, a power (1-β) of .95, 3 independent variables (an independent variable, a moderator, an interaction variable), and a maximum of 10 control variables (based on control variables for ego integrity). As a result, the minimum sample size was determined as 277. Thus, considering a 10% dropout rate, a total of 308 copies of the questionnaire were distributed, and a total of 301 copies were used in the final analysis, excluding 7 copies that were thought to have insincere responses, such as uniform responses.

The participants of this study were elderly people aged 65 or older residing in long-term care facilities, and the majority of them were those aged 75 or older. Most of the participants had one or more chronic diseases, such as hypertension and diabetes, and many of them needed partial assistance with activities of daily living due to physical function decline. In addition, due to limited contact with the family and prolonged exposure to a living environment centered on medical services and care, they typically have a narrower social network and relatively fewer opportunities for emotional support, compared to community-dwelling elderly people. These characteristics of residents in long-term care facilities were considered as factors that may affect the levels of death anxiety and depression among the elderly residents.

Measures

1. Death anxiety

Death anxiety was measured using the Korean version of the Death Anxiety Scale used by Oh and Choi [22]. This scale was designed to be applicable to both elderly people living at home and those living in facilities. This tool consists of a total of 15 items, and each item is assessed on a 4-point Likert scale ranging from ‘Not at all’ to ‘Very much so.’ Higher scores indicate higher levels of death anxiety. The value of Cronbach’s α was reported as .82 by Oh and Choi [22], and it was calculated as .80 in this study.

2. Depression

The level of depression was assessed using the scale that Chon [23] applied to elderly people living in facilities. Each item is rated on a 4-point Likert scale ranging from 1 point (= ‘Hardly’) to 4 points (= ‘Always’), and a higher total score indicates a higher level of depression. This scale contains 11 items in total, and the total score ranges from 11 to 44 points. The value of Cronbach’s α was reported as .90 by Chon [23], and it was calculated as .92 in this study.

3. Ego integrity

Ego integrity was measured using an ego integrity scale developed for the elderly by Pai and Chandra [5]. Each item is rated on a 5-point Likert scale ranging from 1 point (= Not at all) to 5 points (= Very much so). This tool consists of five subcomponents and contains 30 items in total, including 6 items on attitude toward life, 6 items on living wisely, 8 items on satisfaction with current life, 4 items on acceptance of past life, and 6 items on acceptance of old age and death. A higher score indicates a higher level of ego integrity. The value of Cronbach’s α was reported as .58 by the authors of the scale [5], and it was calculated as .85 in this study.

4. Social support

Social support was assessed using a social support scale developed for elderly people living in facilities [24]. Each item is rated on a 5-point Likert scale ranging from 1 point (= Not at all) to 5 points (= Very much so), and a higher score indicates a higher level of social support. This tool contains a total of 12 items across 3 domains: family support, friend support, and professional support. The total scores range from 12 to 60 points. The value of Cronbach’s α was reported as .86 for parent support (family support), .66 for friend support, and .56 for teacher support (professional support) by the developer of the scale [24]. In this study, the value of Cronbach’s α was calculated as .88.

Data analysis

The collected data was analyzed using SPSS version 27.0 as follows.

First, the mean, standard deviation, and minimum and maximum values of the participants’ general characteristics and major variables (death anxiety, social support, depression, and ego integrity) were calculated and presented as descriptive statistics.

Second, differences in depression and ego integrity according to the general characteristics of the subjects were analyzed using the independent t-test and one-way ANOVA, and the Scheffé test was used for post-hoc tests.

Third, correlations between the major variables were analyzed using the Pearson correlation coefficient.

Fourth, hierarchical regression analysis was performed to test the research hypotheses, and mean centering was applied to test the moderating effect of social support. In Step 1 of the regression analysis, the control variable and independent variable were entered. In Step 2, the moderating variable was additionally entered, and in Step 3, the interaction term between the independent variable and the moderating variable was additionally entered.

To detect multicollinearity, tolerance and variance inflation factor (VIF) values were checked, and outliers were identified using Cook’s distance. The independence of the residuals was evaluated using the Durbin–Watson test. Finally, the moderating effect of social support was further verified using the PROCESS macro (Model 1).

Results

Basic statistics and results of normality tests

The participants of this study were 301 people in total. 37.2% of the participants were male (n=112) and 62.8% were female (n=189). Regarding age, the 65-70 age group took up 26.2%, the 71-75 age group, 34.2%, and the ≥76 age group, 39.5%. The average age of the participants was 74.2 years (SD=5.1). Regarding presence of the spouse, people with the spouse accounted for 39.2%(n=118). In education level, middle school or lower took up 56.8%(n=171), accounting for the largest proportion. The average number of children was 1.9 people (SD=0.8). As for subjective health status, people with moderate subjective health status took up the largest proportion at 48.8%(n=147).

As a result of examining the distribution characteristics of major variables, it was found that each variable exhibited the shape of a normal distribution, as shown in Tables S1 and S2, indicating that they are suitable for subsequent analyses.

Differences in the levels of major variables according to general characteristics

Analysis of the scores of depression according to the general characteristics of the participants showed that there were significant differences in the level of depression according to presence of the spouse, number of children, and subjective health status, but there were no significant differences in the depression level according to other variables. More specifically, the group without the spouse, the group without children, and the group with lower subjective health status showed a higher score for depression (Table 1).

Regarding ego integrity, there were significant differences in the level of ego integrity according to gender, presence of the spouse, number of children, subjective health status, and monthly income, but there were no significant differences according to other variables. More specifically, males, the group with the spouse, the group with children, and the group with high subjective health status showed a higher level of ego integrity. In the case of monthly income, a post-hoc analysis showed that there were no significant differences in the level of ego integrity according to the monthly income level (Table 1).

Correlations between major variables

Pearson’s correlation analysis for correlations between major variables was conducted to determine the correlations between variables and the possibility of multicollinearity. As shown in Table 2, analysis results showed that there was a statistically significant negative correlation between depression and ego integrity (r=−.53, p<.001), indicating that a higher level of depression is associated with a lower level of ego integrity. Further, depression was positively correlated with death anxiety (r=.22, p<.001), indicating that a higher level of death anxiety is associated with the increase of emotional distress and depression levels. These results indicate that death anxiety can be a major stressor negatively affecting mental health among the elderly.

In addition, depression showed a statistically significant negative correlation with the total score of social support, which includes all subdomains of social support (r=−.19, p<.01), so it was found that the level of depression tends to decrease with the increase of the level of social support. Among the subdomains of social support, family support (r=−.16, p<.01) and friend support (r=−.14, p<.05) showed a significant negative correlation with depression, but professional support did not have a statistically significant correlation with depression (r=−.11, p=.053). In addition, death anxiety showed a significant negative correlation with ego integrity (r=−.22, p<.001), while the total score of social support had a significant positive correlation with ego integrity (r=.35, p<.001). The correlation coefficients between the major variables were generally at a moderate or lower level, so it was determined that there was no serious multicollinearity problem in the regression analysis.

Moderating effect of social support in the relationship between death anxiety and depression

Hierarchical regression analysis was conducted to evaluate the moderating effect of social support on the impact of death anxiety on depression (Table 3). As a result of analyzing the moderating effect of the total score of social support in Panel A, in Step 1, gender, subjective health status, presence of the spouse, number of children, and monthly income were entered as control variables. As a result, subjective health status and number of children were found to have a significant effect on depression, and the model was statistically significant (F=6.08, p<.001). The model had an explanatory power of 16% (Adjusted R²=.16). In Step 2, death anxiety was additionally entered as an independent variable, and as a result, death anxiety was found to have a significant positive effect on depression (β=.23, p<.001). This result means that the level of depression is increased with the increase of the level of death anxiety. The explanatory power of the model increased significantly to 17% (Δ Adjusted R²=.02). In Step 3, the total score of social support was additionally entered, and it was found that social support had a significant negative effect on depression (β=−.14, p=.014). In other words, the level of depression tended to be decreased with the increase of social support.

In Step 4, as a result of additionally entering the interaction term between death anxiety and social support, the interaction effect was found to be statistically significant (β=.13, p=.014). This result means that the size of the impact of death anxiety on depression varies depending on the level of social support. In particular, the interaction coefficient appeared in a positive direction, so it was found that as the level of social support increases, the positive relationship between death anxiety and depression becomes stronger. The final model had an explanatory power of 19% (Adjusted R²=.19), and the overall model was statistically significant (F=6.28, p<.001).

In addition, as a result of testing the moderating effects of the subfactors of social support in Panel B, family support (β=.03, p=.512) and friend support (β=.02, p=.649) did not show a statistically significant moderating effect. However, professional support was found to significantly moderate the relationship between death anxiety and depression (β=.13, p=.014). These results indicate that the relationship (slope) between death anxiety and depression varies depending on the level of professional support (Figure 1). In addition, the interaction observed regarding the total score of social support can be interpreted as a result primarily attributable to the effect of professional support in view of the analysis results of the subdomains of social support.

Moderating effect of social support in the relationship between death anxiety and ego integrity

Next, hierarchical regression analysis was performed to statistically determine whether social support had a significant moderating effect on the impact of death anxiety on ego integrity. Gender, subjective health status, presence of the spouse, number of children, and monthly income were entered as control variables. In Step 1, as a result of entering general characteristics, as shown in Table 4, the model was statistically significant (F=6.02, p<.001), and had an explanatory power of 15% (Adjusted R²=.15). In Step 1, subjective health status was found to have a statistically significant positive effect on ego integrity. In Step 2, death anxiety was additionally entered as an independent variable, and as a result, death anxiety was found to have a statistically significant negative effect on ego integrity (β=−.18, p=.002). These results mean that the level of ego integrity was decreased with the increase of the level of death anxiety, and the explanatory power of the model significantly increased to 17% (Δ Adjusted R² = .02). Next, in Step 3, as a result of additionally entering social support, social support was found to have a statistically significant positive effect on ego integrity. In other words, it was shown that a higher level of social support was linked to a higher level of ego integrity.

Next, in Step 4, the interaction term between death anxiety and social support was additionally entered. As a result, it was found that the interaction effect was not statistically significant (β=.04, p=.381), and there was no increase in the explanatory power of the model (Δ Adjusted R²=0). In addition, the analysis of the subfactors of social support revealed that the p-values for family support (β=.03, p=.512), friend support (β=.02, p=.649), and professional support (β=.05, p=.341) were all greater than 0.5, indicating that each subdomain of social support had no statistically significant moderating effect, and the final model of the analysis was statistically significant (F=6.05, p<.001).

Discussion

This study comprehensively analyze the relationships between death anxiety, depression, ego integrity, and social support, and derived implications for providing nursing services to elderly residents in long-term care facilities based on the analysis results. In this study, the Stress Process Model proposed by Turner [13] was used as a theoretical framework. According to the Stress Process Model, stress directly affects individuals’ mental health, and social or psychological resources have been reported to moderate this impact of stress on mental health [13]. Based on this Stress Process Model, this study set death anxiety as a stressor and examined its effects on depression and ego integrity in elderly residents in long-term care facilities, a special living environment. In addition, this study statistically analyzed whether social support and each of its subdomains (friend support, family support, and professional support) moderate the impact of death anxiety as a stressor on depression and ego integrity.

First, analysis of the general characteristics of the participants revealed that among elderly residents in long-term care facilities, the depression score was statistically significantly higher in the group without the spouse, and it was also found to be higher in the group without children compared to the group with two children or the group with three or more children (Table 1). In addition, it was found that the average score for depression was higher in the group with lower subjective health status, compared to the group with moderate subjective health status or the group with high subjective health status. These results are consistent with previous studies of elderly people reporting that there were differences in the level of depression among the elderly according to the presence of the spouse, the number of children, and subjective health status [4,25,26]. The level of depression in the elderly has been reported to be closely related to their relationship with the family. In particular, it has been reported that among the elderly living in facilities, those with children or the spouse have a relatively higher frequency of family visits, and that continuous communication with the family have a positive effect on alleviating loneliness and reducing depression levels [27,28]. In addition, elderly people with good health status can participate more actively in the activities within facilities, and their freedom of movement is likely to increase opportunities for social interactions. It is presumed that these characteristics contributed to the reduction of social isolation as well as emotional stability among elderly residents in long-term care facilities in this study, thereby eventually affecting the reduction of the level of depression.

The score for ego integrity was statistically significantly higher in men than women, and it was also higher in the group with the spouse than the group without the spouse. Additionally, the group with two or more children showed a higher score for ego integrity, compared to the group without children, and the group with lower subjective health status showed a significantly lower score for ego integrity. In addition, the group with high monthly income showed a relatively higher level of ego integrity (Table 1). These results suggest that although ego integrity in the elderly may be viewed as individuals’ internal characteristic, it interacts closely with life resources such as family relationships, economic stability, and health status. The findings regarding the impact of general characteristics on ego integrity are consistent with previous studies that reported differences in mental health according to gender, presence of children, presence of the spouse, health status, and income among elderly residents in long-term care facilities [25]. As depression levels are influenced by family relationships and health status, the level of ego integrity tends to be higher in elderly people who maintain a close relationship with the family and have good health status [4,25]. In addition, it has been reported that higher income levels are linked to more active participation in leisure activities [25]. Thus, it is thought that these factors influenced positive psychological factors, such as ego integrity, among elderly residents in long-term care facilities in this study as well.

Hierarchical regression analysis was conducted to analyze the effects of death anxiety on depression and ego integrity among elderly residents in long-term care facilities, and to examine the moderating effect of social support. Death anxiety was entered as an independent variable after considering control variables. As a result, death anxiety was found to have a positive effect on depression, but have a negative effect on ego integrity (Table 3, 4).

First, the results of this study showed that death anxiety increases the level of depression among elderly residents in long-term care facilities. This finding is consistent with previous studies showing that elderly people placed in the special environment of long-term care facilities experience high levels of death anxiety and depression due to physical frailty and isolation from the family and social relationships [4,22]. To examine the moderating effect of social support on the impact of death anxiety on depression, hierarchical regression analysis was conducted for each of the subfactors of social support, including family support, friend support, and professional support. As a result, professional support was found to significantly moderate the relationship between death anxiety and depression, but family support or friend support did not show a significant moderating effect (Table 3). In particular, the interaction effect appeared in a positive direction. Thus, it was found that as the level of professional support increased, the positive relationship between death anxiety and depression became stronger.

These results could be interpreted as indicating that professional support directly moderates the relationship between death anxiety and depression, but they may also be interpreted as reflecting the possibility that professionals’ intervention or contact with professionals was relatively more frequently given to elderly people with a higher level of death anxiety and greater emotional vulnerability. In other words, it is difficult to rule out the possibility that study results may have statistically reflected the structural characteristic of providing more professional support to elderly people at high risk of depression in long-term care facilities. In addition, studies by Kim [29] and by Lee and Yang [6] reported that social support has a protective role, showing that social support contributed to emotional stability among elderly people. The present study is partially consistent with these prior studies in that the total score of social support also showed a significant negative main effect on depression. However, in the interaction between social support and death anxiety, only professional support was found to have a significant effect. In this regard, there is the possibility that the specific environmental context of long-term care facilities influenced the way social support has an impact.

Elderly residents in long-term care facilities are placed in an environment where contact with family and friends is limited, and their support resources that are actually accessible and consistently provided are likely to be professional support. Therefore, it is reasonable to think that the results of this study did not imply that a specific form of support is better but suggest that the functions of support resources may vary depending on the environmental characteristics of elderly residents in long-term care facilities. In other words, rather than implying a causal relationship between professional support and the increased level of depression, the findings of this study suggest that the allocation structure of professional support, where contact with professionals is concentrated on at-risk groups within facilities, or the possibility of reverse causation should also be considered. In follow-up research, it is necessary to more clearly investigate the causal direction between professional support and depression through a longitudinal study design, and interpret the effects of the subdomains of social support in a more generalized context through comparative analysis with community-dwelling elderly people.

Next, this study found that death anxiety negatively affected ego integrity, and this finding is consistent with several previous studies showing that a higher level of death anxiety in elderly people is associated with greater difficulty in positively accepting one’s life and a lower level of ego integrity [5]. The results of hierarchical regression analysis showed that the moderating effect of social support in the negative association between death anxiety and ego integrity was not statistically significant (Table 4). This lack of a significant moderating effect of social support can be interpreted in relation to the above-discussed special characteristics of the environment of long-term care facilities. Elderly people admitted to long-term care facilities are placed in an environment where continuous and reciprocal interactions with the family and friends are limited, and their lives tend to be restructured around relationships with professionals and caregivers. However, it has been reported that ego integrity is a psychological concept closely related to the processes of a review of one’s life and meaning integration in old age, and that long-term relational resources, such as an emotional bond with the family, play a more important role. In fact, the study by Lee [30] found that family support had a significant mediating effect on ego integrity. In addition, the study by Jeon [31] reported that various subfactors of social support, such as instrumental support, affectionate support, positive social interactions, emotional support, and informational support, showed a partial mediating effect in relation to ego integrity. However, it should be noted that there is an environmental difference between these prior studies and the present study in that these prior studies were primarily conducted with community-dwelling elderly people, the participants of this study were elderly residents in facilities. Thus, the results of this study suggest that in the process where death anxiety affects ego integrity, the function of social support may vary depending on the residential environment of the elderly. From the perspective of the Stress Process Model, these results may be interpreted as showing that even the same type of social resource may have a limited buffering function depending on the particular environmental conditions and contexts where it is actually operating. Therefore, follow-up research is needed to more meticulously investigate the relationship between social support and ego integrity by including elderly people in various environments and distinguishing qualitative levels of social support, the frequency of social interactions, and types of support (emotional, informational, instrumental, etc.).

Additionally, in this study, social support showed a significant main effect on ego integrity, but it did not have a significant moderating effect in the relationship between death anxiety and ego integrity (Table 4). This means that although social support acted as a protective factor that directly enhances ego integrity, it did not have a sufficient impact as a moderating resource that buffers the impact of death anxiety, the specific stressor considered in the study. These results are partially consistent with previous studies showing that the effects of social support may be different depending on the types and functions of social support [32]. Although this study analyzed social support by classifying it into the three subdomains of family, friend, and professional support, it has limitations in that it did not divide the qualitative characteristics of social support or functional types of social support such as informational and instrumental support. Therefore, in follow-up research, there is a need to more specifically distinguish types and qualitative levels of social support in order to more closely investigate the mechanisms through which social support influences ego integrity.

The limitations of this study derived from the study results and discussion of them are as follows. First, the participants of this study were healthy elderly residents in long-term care facilities who were able to respond to the survey. Therefore, frail elderly individuals who were unable to respond due to physical or mental issues were excluded from analysis. Consequently, it is another limitation of this study that its findings cannot be generalized to a broader population. Second, additional factors that may influence professional support were not sufficiently discussed in this study. In particular, considering that professional support, a subdomain of social support, was found to reinforce the positive relationship between death anxiety and depression, there is a need for a more in-depth analysis of the characteristics of groups that receive a higher level of professional support. Third, this study did not give in-depth consideration to diverse types of professionals working in long-term care facilities as well as differences in environmental factors such as programs and facilities within the institutions. Therefore, further research is needed to reexamine the effect of social support by considering special situations as well.

Conclusions

In this study, Turner’s Stress Process Model was used to statistically investigate the moderating effect of social support in the relationships between death anxiety, depression, and ego integrity among elderly residents in long-term care facilities. Analysis results showed that death anxiety was a significant predictor for the increase of depression and the decrease of ego integrity [18]. In addition, among the subdomains of social support, professional support was found to have a significant moderating effect on the positive relationship between death anxiety and depression, while the total social support did not show a statistically significant moderating effect on the relationship between death anxiety and ego integrity. The results of this study showed that social support resources may operate differently in the special environmental context of long-term care facilities from the way they operate in the local community. In particular, regarding professional support, which is a formal support system structurally provided in long-term care facilities, provision of professional support tends to be concentrated on elderly people with a high level of death anxiety, and the results of this study suggest that this characteristic of professional support may have been statistically reflected. Therefore, the results of this study suggest that rather than simply strengthening professional support, there is a need to develop a systematic nursing intervention program to identify elderly people with a high level of death anxiety at an early stage, and to ensure that the content and methods of professional interventions will effectively contribute to reducing the level of depression. In addition, longitudinal studies should be conducted with elderly people in various residential environments to more clearly elucidate the functions and causal direction of social support.

Supplementary Materials

Figure 1.

Moderating effects of social support on the relationship between death anxiety and psychological outcomes. (A) Professional support significantly moderated the association between death anxiety and depression; (B) Social support did not significantly moderate the association between death anxiety and ego-integrity (interaction p=.381).

References

• 1. World Health Organization. WHO country data [Internet]. Geneva: World Health Organization; c2025 [cited 2025 Oct 31]. Available from: https://data.who.int/countries/410
• 2. Cho Y, Chung H, Joo H, Park HJ, Joh HK, Kim JW, et al. Comparison of patient perceptions of primary care quality across healthcare facilities in Korea: a cross-sectional study. PLoS One. 2020;15(3):e0230034. https://doi.org/10.1371/journal.pone.0230034ArticlePubMedPMC
• 3. Leung MY, Wang C, Famakin IO. Integrated model for indoor built environment and cognitive functional ability of older residents with dementia in care and attention homes. Building and Environment. 2021;195:107734. https://doi.org/10.1016/j.buildenv.2021.107734Article
• 4. Lim YA. Relationship between social supports and depression symptoms among the elderlies at care facilities. Journal of the Korea Academia-Industrial Cooperation Society. 2019;20(11):335–47. https://doi.org/10.5762/KAIS.2019.20.11.335Article
• 5. Pai S, Chandra S. Life satisfaction and ego integrity among older adults: a comparative study. Indian Journal of Gerontology. 2021;35(1):111–125.
• 6. Lee GM, Yang Y. Relationship between social support and depression in elderly people: a systematic review. Journal of the Korean Public Health Association. 2021;47(4):31–55.
• 7. Gullette MM. American eldercide: How it happened, how to prevent it. Chicago: University of Chicago Press; 2024. 328 p.
• 8. Lee SY, Hung L, Jung HW, Chaudhury H. Role of physical environment in quality of life among residents in dementia care facilities through a longitudinal observational study-for facilities in Canada and South Korea. Journal of Korean Housing Association. 2021;32(6):23–32. https://doi.org/10.6107/JKHA.2021.32.6.023Article
• 9. Mitchell JM, Kemp BJ. Quality of life in assisted living homes: a multidimensional analysis. Journal of Gerontology: Series B, Psychological Sciences and Social Sciences. 2000;55(2):P117–P127. https://doi.org/10.1093/geronb/55.2.P117Article
• 10. Kim Y. Factors associated with care stress of families with elderly dementia patients at long-term care: compared with families with non-dementia [dissertation]. [Seoul]: Seoul National University; 2024. 75 p.
• 11. Kang D. Path analysis of the elderly’s deprivation experience on suicidal ideation. Journal of Social Science. 2019;58(1):197–245.
• 12. Jeon J, Jeon H. Effects of stress on depression among older adults living alone: the mediating effects of social support and resilience. Korean Journal of Gerontological Social Welfare. 2022;77(1):41–67.Article
• 13. Turner RJ. Understanding health disparities: the relevance of the stress process model. Social Mental Health. 2013;3(3):170–186. https://doi.org/10.1177/21568693134881Article
• 14. Park R, Lee MK. The moderating effect of family strength in the relationship between daily life restriction and depressive symptoms of elderly with disability. Journal of the Korean Gerontological Society. 2025;45(1):39–54.
• 15. Lee C. The effects of multidimensional health on suicidal ideation among older adults [dissertation]. [Seoul]: Seoul National University; 2024. 172 p.
• 16. Yoo H, Kim K, Kim J, Boerner K, Han G. Care experiences among very old parents and their old children. Journal of the Korean Gerontological Society. 2024;44(5):623–44. https://doi.org/10.31888/JKGS.2024.44.5.623Article
• 17. Pearlin LI, Mullan JT, Semple SJ, Skaff MM. Caregiving and the stress process: an overview of concepts and their measures. The Gerontologist. 1990;30(5):583–594.Article
• 18. Pearlin LI, Menaghan EG, Lieberman MA, Mullan JT. The stress process. Journal of Health and Social Behavior. 1981;22(4):337–56. https://doi.org/10.2307/2136676ArticlePubMed
• 19. Aneshensel CS, Avison WR. The stress process: an appreciation of Leonard I. Pearlin. Society and Mental Health. 2015;5(2):67–85. https://doi.org/10.1177/2156869315585388Article
• 20. Goh J, Hwang I, Oh H. The experiences of newly admitted elderly residents in nursing homes. Journal of the Korean Gerontological Society. 2009;29(2):477–488.
• 21. Lee J, Rantz MJ. Admission factors influencing activities of daily living in nursing home residents admitted from acute care hospitals. Journal of Korean Gerontological Nursing. 2008;10(1):68–76.
• 22. Oh MN, Choi WS. A study on self-esteem, death anxiety and depression of the elderly in home and facilities. Human Ecology Research. 2005;43(3):105–118.
• 23. Chon SJ. Reminiscence: content analysis and its nursing implication II. Journal of Red Cross College Nursing. 1989;14(1):81–101.
• 24. Cho AM. The mediating role of social support in the relationship between adolescents' stress and academic achievement. Korean Journal of Educational Research. 2005;43(3):137–155.
• 25. Lee JH, Un KG. The relationship of subjective health status, sarcopenia and health-related quality of life of the elderly in long-term care facilities. Journal of Korean Society of Wellness. 2020;15(3):417–429. https://doi.org/10.21097/ksw.2020.08.15.3.417Article
• 26. Lee EH. Environmental factors for making residents feel ‘at home’ in long-term care facilities in South Korea. Social Science Research Review. 2019;35(3):131–159. https://doi.org/10.18859/ssrr.2019.8.35.3.131Article
• 27. Lim JS, Chung SM, Moon CS. The multiple additive moderation effects of family solidarity and resilience on the relationship between social isolation and depression of the elderly. Humanities and Social Sciences 21. 2020;11(6):2527–2541. https://doi.org/10.22143/HSS21.11.6.178Article
• 28. Jeon MS, Tae MO. Impact of socio-economic factors on the elderly's satisfaction with life: mediating effect of the elderly's leisure participation. The Journal of the Korea Contents Association. 2016;16(6):323–333. https://doi.org/10.5392/JKCA.2016.16.06.323Article
• 29. Kim BG. The effect of stress on depression of the elderly: focusing on the moderating effect of social support and mediating effect of sense of self-respect. The Journal of the Korea Contents Association. 2019;19(6):281–291. https://doi.org/10.5392/JKCA.2019.19.06.281Article
• 30. Lee M. Mediated effect of family support in the influence route of health condition and self-integration of older adults: Focused on differences by age group. The Journal of the Korea Contents Association. 2012;12(10):280–290. https://doi.org/10.5392/JKCA.2012.12.10.280Article
• 31. Jeon HJ. Mediating effects of social support on the relationship between self-esteem and self-integrity of the elderly [master’s thesis]. [Naju]: Honam University; 2019. 60 p.
• 32. Shin SH, Lee JH, Kwon WA, Kweon YH, Kim CS. The effect of social support on elderly people for rehabilitation at nursing homes. Journal of the Korea Academia-Industrial Cooperation Society. 2011;12(6):2616–2623.Article

Figure & Data

References

Citations

Citations to this article as recorded by