1. Introduction
The aging of the population is one of the most important problems globally. Aging leads to an increased incidence of chronic disease, especially musculoskeletal disease (MSD). As the prevalence of MSD increases with age, 20% to 30% of people across the globe are living with painful musculoskeletal impairments [
1]. The prevalence of MSD in the Korean elderly aged 65 years or over increases yearly, and has been estimated to reach 70.2% by 2017, with a of 2.3-fold increase expected by 2040 [
2,
3]. MSD often occurs in the joints and most commonly in the knee, shoulder, lumbar, and cervical areas [
4]. One major MSD symptom in the elderly is chronic pain, which may be persistent or recurrent, due to the disease directly affecting bones, joints, muscles, or related soft tissues [
5]. Chronic musculoskeletal pain (CMP) is a major health problem for the elderly, experienced by 83% of the elderly in the community [
6].
The prevalence, symptoms, and responses to symptoms differ between the sexes. According to the World Health Organization (WHO) nationwide study on aging (SAGE), a comparison of MSD prevalence between men and women showed that general chronic back pain and joint pain was more prevalent among women than men [
7]. Women exhibit a lower threshold for musculoskeletal pain than men, making them more sensitive to pain [
8,
9]. Pain-related worries were more common in women than men, with subjective pain and discomfort being higher in women [
10]. In addition, CMP-related depression was significantly higher in women than men [
11]. In a large longitudinal study of the elderly community in Singapore, differences between the sexes in daily performance and perceived health status variables were observed [
12].
Pain coping strategies are the specific set of behaviors that individuals use to control and emotionally respond to their pain [
13]. How a person copes with pain regulates the physical and mental factors that ultimately affect their quality of life (QOL) [
14]. However, a systematic literature review of the analysis of sex differences in coping styles for chronic pain showed that women are passive and maladaptive, whereas men are more adaptive [
15]. The absence of a social support can be a leading cause of the deterioration of psychological as well as physical health [
16,
17] while strong social support exhibits a positive effect on the health-related QOL of the elderly [
18]. Compared to men, women tend to underestimate their ability to cope, and obtain social support from their surroundings [
12].
Older people with CMP exhibit lower QOLs compared to those without [
9], and those with MSD report significantly lower QOLs compared to those with cardiovascular or endocrine chronic diseases [
19]. Since the health-related QOL of elderly people who experience CMP is an important issue for health promotion in the elderly, identifying the various factors influencing improvements in QOL is necessary.
Wilson and Cleary proposed a conceptual model for health-related QOL, including the clinical and social science paradigm [
20]. Based on Wilson and Cleary’s model, some previous studies have attempted to explain the QOL of elderly people with CMP using some dimensions or variables, such as pain, capability to perform the tasks of daily living, and physical activity limitation [
21], as well as psychological aspects, such as depression, anxiety, and powerlessness [
22]. In a prior study, the authors examined the aspects that affect the QOL of elderly with CMP, including demographic characteristics, pain characteristics, functional limitations and perceived health status, pain coping, and social support [
23]. The results showed that high levels of education, low pain, low functional limitations, use of accommodative coping with pain, a high perceived health status, and strong family and friend support are factors related to improvements in QOL. However, the study used multiple regression analyses, which cannot evaluate the direct and indirect effects of the variable associated with the QOL and, therefore, cannot identify the structural relationships between predictors. In other words, the authors found that pain, functional limitations, pain management, perceived health status, and social support all affect QOL. However, the regression coefficients of each variable only indicate the influence when the other variables are adjusted, so the relationship between the variables and the effect of the relationship cannot be identified. Only the explanatory power explained by the total independent variables included in the multiple regression model can be determined; the specific relationships between the variables and the direct and indirect effects on the dependent variables cannot be calculated.
The structural equation model (SEM) overcomes these limitations of multiple regression. One principal difference with SEM is that a construct that acts as an independent variable in one relationship can be the dependent variable in another relationship [
24]. For instance, the perceived health status in multiple regression analysis is an independent variable for QOL; in the hypothetical model in this study, it is a dependent variable of pain coping and social support, and also a mediating variable in the relationship between pain coping and social support and QOL. Hence, both the direct and indirect effects of perceived health status on QOL could be examined.
A review of the literature did not reveal the structure of the variables affecting QOL of the elderly who experience CMP, and it was not clear whether there were any differences between sexes. The purpose of this study was to empirically test the theory-based hypothetical model and to identify sex differences to explain the relationship between variables related to the QOL of elderly people who experience CMP using multigroup structural equation model analysis.
4. Discussion
In this study, we aimed to construct an SEM based on Wilson and Cleary’s QOL model to analyze the influence of various factors on QOL after establishing measurement invariance across sexes in a sample of Korean elderly aged 65 years or over and experiencing CMP. The results of the SEM were partially consistent with Wilson and Cleary’s model. First, a causal relationship was found between adjacent variables in terms of pain, functional limitations, perceived health status, and QOL in both sexes. Wilson and Cleary’s model empirically examined the causal relationship between adjacent concepts through various subjects with chronic disease [
39]. However, the model had not been previously tested in older patients with CMP; in this study, we tested the model in the real world with community-dwelling elderly experiencing CMP and demonstrated its applicability.
Wilson and Cleary’s model empirically examined the causal relationship between adjacent concepts through various subjects with chronic disease [
39]. However, the model had not been previously tested in older patients with CMP; in this study, we tested the model in the real world with community-dwelling elderly experiencing CMP and demonstrated its applicability. The results of the relationships in this study were found to be inconsistent with the Wilson and Cleary model. First, we added pathways to Wilson and Cleary’s model, where pain and functional limitations directly affect QOL. Second, pain coping and social support were found to have a direct effect on QOL, but the effects on pain, functional limitation, and perceived health status were inconsistent. In multigroup analysis, we observed differences in the paths between the sexes.
Pain did not directly affect QOL in structural models for men and though a direct influence pathway was established for women; however, we found no statistically significant differences between sexes. The path coefficient from pain to functional limitation, however, was larger in women than men and this difference was considered statistically significant. In this study, pain was more severe in women than in men, which is consistent with the studies of European elderly with CMP [
40]. This is related to differences in the perception of pain between the sexes [
41]. Therefore, in considering the indirect effects of pain on QOL through functional limitation, pain management needs to be improved in elderly women with CMP.
Functional limitation directly affected QOL more in men than in women, with a statistically significant difference between sexes. Limitations with respect to the activities of daily life considerably impacted the QOL of people with musculoskeletal disorders [
42]. In elderly people with hallux valgus, the greater the degree of deformity, the greater the functional limitation, which ultimately affects the quality of life [
43]. As these limitations also impact family and social relationships [
42], the aim of long-term management to improve the quality of life for patients with musculoskeletal disease should focus on rehabilitation strategies that minimize these functional limitations. In particular, men in Korea have actively played social roles through professional activities for a long time compared with women, which can result in a greater effect on the quality of life through increased negative attitudes due to physical limitations. Therefore, further research is needed to confirm how men and women perceive social and physical functions, and how this is related to quality of life. We found that older women were more restricted in function than older men, which is consistent with the results of a study of CMP in older European adults [
40]. Given the better evaluation of physical health by men than women [
44], we suggest that an approach that takes into account the differences in the basic functional levels of elderly men and women would be effective.
Pain coping affected pain and functional limitation in men, but in women, pain coping directly affected QOL. However, only the path from pain coping to pain was significantly different between the sexes. Pain coping, an individual characteristic, was found to directly affect the QOL and to affect QOL through pain and functional limitation. The greater the pain coping in patients with spinal stenosis, the higher their subjective walking ability and social function [
45]. The decrease in coping strategies in patients with irritable bowel disease (IBD) was a predictor of decreasing QOL [
46]. Pain coping is an emotional response that can change through mediation. According to the pain coping styles of patients suffering from chronic low back pain, different beliefs about mood disorders and pain control are reported, suggesting the need for pain coping programs in terms of personal behavior and cognition [
47]. Intervention to reinforce strategies for coping with pain in patients with chronic pain resulted in a reduction in the intensity and interval of pain, and a positive change in psychological function [
48]. Pain response training programs, including exercise interventions, were conducted in adults over 50 years of age living in the community, resulting in reduced functional disability and pain [
49]. Therefore, the implementation of pain response skills training for the elderly with CMP is a major intervention that leads to pain and function improvement, ultimately improving QOL.
According to a previous study, women show passive and maladaptive coping types, whereas men use more adaptive coping patterns [
15]. The more active the pain coping strategy, the higher the QOL; patients using accommodative coping exhibit less depression and disability [
50]. Conversely, passive pain coping can lead to a negative perception of pain and cause symptoms, such as depression and anxiety, and depression in elderly patients with chronic pain results in decreased QOL [
51]. However, in this study, passive pain coping as a sub-factor of PRI was eliminated in the measurement model due to a low path coefficient and the poor fit of the model. The subjects in this study were able to function independently. For instance, their visits to a welfare center implies that their social activities are active to an extent. Korean seniors use the least amount of avoidance pain treatment [
52]; hence, we suggest that the avoidance coping strategy in this study was not significant in the construction of the PRI. Carefully testing the cultural validity of the PRI in the Korean population is necessary. The cultural validity of the tools needs to be closely examined. Therefore, to improve the QOL of elderly people with CMP, reinforcing their active or accommodative coping strategies is necessary. Also, further research is needed to identify which types of pain management strategies are more effective according to sex.
Overall, the SEM showed that social support affected QOL but differed structurally between sexes. For women, it was found that social support affects QOL directly as well as indirectly through perceived health status, though the effect was not statistically significant. The examination of gender differences in the QOL of elderly people living alone showed that women visit senior citizens’ welfare centers and actively engage in religious activities more than older men [
44]. This means that older women receiving more peer support through these activities compared to older men. In contrast to women, older men resort to their own experiences rather than receive social support. This may enhance their psychological well-being and productivity and reduce depression [
53]. In this sense, exploring the differences between men and women in terms of whether the experience of receiving or providing social support affects QOL is necessary.
This study is the first to provide a rationale for customized intervention strategies to improve QOL among elderly Koreans with CMP. However, this study has some limitations that affect the interpretation of the results. Because twice as many women than men were included, the social support in the overall model may reflect female characteristics. Further research is needed using a different sample size and examining sociodemographic factors that can affect social support. The subjects of this study were collected by convenience sampling, so bias may exist in the relationship between study variables due to the inability to control sociodemographic variables. Therefore, in future research, the bias between groups should be controlled using methods, such as the propensity matching score. In this study, variables in terms of biological factors were not included; therefore, we could not sufficiently validate the theoretical model as a whole. We recommend that future studies measure more specific variables associated with musculoskeletal disorders, such as biomarkers or radiological images. Future studies should test the model using different measurement variables reflecting the environmental and personal factors described in the Wilson and Cleary model also. In this study, the passive coping construct was removed from the original version of the pain coping measurement. Thus, further validation is required to establish the external validity and reliability of the Korean version measurement of PRI. Although the causal relationship between concepts was verified in the model, this study is limited by its cross-sectional nature, so in-depth longitudinal studies are needed. Lastly, since cultural differences can affect environmental factors, demographic characteristics, and personal preferences and values, further studies should be conducted with data collected from different countries.