ارائه مدل ساختاری انزوای اجتماعی بر اساس سبک زندگی و خودکارآمدی با میانجی‌گری حمایت اجتماعی ادراک شده در سالمندان

Introduction

Aging is regarded as the final stage of life, marked by the cessation of work and the loss of typical daily activities. It is an important life phase and a biological-social phenomenon that occurs in human beings. This is a period during which an individual, having stepped away from the role they have played for many years, is expected to experience peace and comfort. However, due to unfamiliarity with this new role, individuals may experience anxiety and confusion. After years of professional life, this transition can have significant psychological and emotional effects. The experience of aging may vary widely among individuals, and in some cases, it may even be contradictory. Some view aging as the 'golden years,' offering individuals the chance to focus on other roles in their life and engage in personal interests or leisure activities. Thus, aging and retirement from the workforce can be seen as a release from mandatory work obligations, offering a period of rest, tranquility, and enjoyment in a new and golden phase of life. Understanding the internal relationships between feelings of loneliness, social isolation, lifestyle, and self-efficacy—mediated by perceived social support—can offer valuable insights for improving the mental health of older adults. Therefore, this study aims to present and test a structural model examining the impact of loneliness and social isolation on lifestyle and self-efficacy, with perceived social support as a mediator, among the elderly.

Method

This study is descriptive and correlational in design. In terms of its objectives, it is categorized as applied research. The study population consists of elderly individuals aged 60 and above, residing in nursing homes in Tehran during the years 2021 and 2022. Given the lack of precise statistics on the number of elderly residents in Tehran’s nursing homes, a multi-stage random sampling method was employed. Following the guidelines of Cramér's rule, which recommends a minimum of 40 participants per predictor variable for structural models, and taking into account the possibility of data distortion, a sample of 300 elderly individuals from various nursing homes in Tehran was selected for the study. The DiTommaso Loneliness Questionnaire (2004) was used to measure loneliness, the Lubben Social Network Scale for assessing social isolation in the elderly, the Iranian Elderly Healthy Lifestyle Questionnaire by Eshaghi et al. (2009) for lifestyle assessment, the General Self-Efficacy (GSE-17) Scale by Sherer et al. (1982) for self-efficacy measurement, and the Multidimensional Scale of Perceived Social Support (MSPSS) by Zimet, Dahlem, Zimet, and Farley (1998) for perceived social support. Data analysis was performed using structural equation modeling (SEM) with the AMOS software.

Results

In the measurement model of the study, 10 indicators were used to represent 3 latent variables. It was hypothesized that the latent variable of lifestyle would be measured by the indicators of prevention, nutrition, exercise, stress management, and social relationships. The latent variable of perceived social support was assessed through the indicators of support from friends, family, and others, while the latent variable of social isolation was measured using the indicators of friendships and family relationships. The model fit was evaluated using Confirmatory Factor Analysis (CFA) with AMOS 24.0 software and Maximum Likelihood (ML) estimation. Table 1 presents the goodness-of-fit indices for both the measurement and structural models.

Table 1. Goodness-of-fit indices for the measurement model and structural model

Goodness of fit index Measurement model Structural model ROC curve

Primary Model Refined Model

Chi-Squre 134.62 86.6 102.54 -

Degree of freedom of the model 32 31 38 -

df/2 4.21 2.78 2.70 Lesser than 3

GFI 0.926 0.951 0.947 < 0.90

AGFI 0.873 0.912 0.908 <0.850

CFI 0.916 0.955 0.952 < 0.90

RMSEA 0.100 0.074 0.073 > 0.08



Table 1 shows that, with the exception of two goodness-of-fit indices—RMSEA and χ²/df—all other goodness-of-fit indices obtained from the Confirmatory Factor Analysis (CFA) support the acceptable fit of the measurement model to the collected data. Given the importance of the two mentioned indices, further evaluation of the model was conducted by adding covariance between the errors of the two indicators, stress management and social relationships, within the lifestyle construct. This modification resulted in acceptable goodness-of-fit indices for the measurement model. In the measurement model, the highest factor loading was associated with the indicator of 'support from friends' (β = 0.839), and the lowest factor loading was associated with the indicator of 'exercise' (β = 0.444). Since the factor loadings for all indicators were greater than 0.32, it can be concluded that all indicators had sufficient power to measure the latent variables in this study

Structural Model: Following the evaluation of the measurement model fit, the goodness-of-fit indices for the structural model were estimated and assessed in the second phase. It was hypothesized in the structural model that lifestyle and self-efficacy, mediated by perceived social support, would predict social isolation in older adults. All the goodness-of-fit indices obtained from the analysis of the structural model indicate that the model has an acceptable fit with the collected data.

Conclusion

The present study was conducted with the aim of examining the fit of a structural model of loneliness and social isolation based on lifestyle and self-efficacy, with the mediation of perceived social support in older adults. The results of the statistical analyses assessing the structural model of loneliness and social isolation, based on lifestyle and self-efficacy with the mediation of perceived social support, showed that the collected data fit the model well. The results of the goodness-of-fit evaluation for the research indices indicated that, overall, the model was well-suited for explaining and fitting the data. Therefore, it can be concluded that loneliness and social isolation can be explained, analyzed, and fitted based on lifestyle and self-efficacy, with the mediation of perceived social support. In this way, a significant relationship was found between loneliness and social isolation and the variables of lifestyle, self-efficacy, and perceived social support. Regarding the recommendations for future research to enhance the level of studies in this area, it is suggested that dedicated spaces be created to strengthen the self-efficacy of older adults. These spaces should be designed with extensive networking to engage and activate the participation of elderly individuals, harnessing the diverse capabilities and potentials of this group. Furthermore, it is recommended that comprehensive educational programs be developed to promote healthy lifestyles among the elderly. These educational programs could focus on preventing various diseases, engaging in physical activities, exercise, recreation, healthy nutrition, stress management, and fostering social and interpersonal relationships within the elderly population.