Although many social and personal correlates of loneliness have been identified so far, a few number of studies have assessed the specific contribution of particular factors (or groups of factors) to the sample variance for loneliness. In the epidemiologic study conducted among the elderly population in Sweden, social and health-related factors were found to be more strongly associated with loneliness than the socio-demographic ones (6). In the study performed in the UK, although on the general population (including also the younger age groups), an increase in the level of physical comorbidity (measured by the number of diagnoses) was shown to increase the probability of having the feeling of loneliness (31). This was likely to be the case in individuals having at the same time also mental disorders, anxiety or depression. The mediating role of mental disorders in associations between higher levels of physical comorbidity and loneliness can be explained by the finding of another large epidemiologic study, where the probability of having mental health disorders was shown to increase in parallel with increasing number of physical comorbidities (32). The results of this study indicate towards the same, highlighting the major role of the health status in loneliness. In this study, too, the effect of poor health on loneliness is largest if mental disorders are part of comorbidities.
Mental disorders, in particular predisposition for depression or depressed mood, may enhance the feeling of loneliness by acting synergistically on creating negative emotions, such as pessimism and helplessness, negative judgement on self-efficacy, and negative beliefs in possibility of developing productive relationships with others (33, 34). In this regard, we found in this study that participants who had depression in early days (indicating a predisposition for depression) felt much lonely compared to participants who were not depressed. There is also an interaction effect between loneliness and depression in older individuals (35). That is, not only depression may cause loneliness, but also loneliness was found to be a strong and direct predictor (irrespective of the presence of confounding factors) of depression. Knowing that, is of the utmost importance for planning prevention in older individuals, considering the fact that depression has huge detrimental effects on health outcomes of older individulas, and the fact that loneliness is a controllable risk factor for depression (36). In this respect, it is important to identify the causal factors for loneliness.
The recently published systematic review emphasized that despite a range of factors examined in longitudinal studies, there were only a few ones with relatively consistent associations with loneliness (37). These factors included: not being married/partnered and partner loss, limited social network, low level of social activity, poor self-perceived health, and depression/depressed mood or a decrease in depression. Marital status, especially loss of the intimate partner in later life, has been highly rating in many studies as the factor strongly contributing to loneliness in older individuals (38). However, this situation may differently affect women and men, which may depend on cultural, situational, and internal psychological factors (39). In general, it was reported that an adjustment to the period of bereavement and problems faced by the widowhood, are more difficult for women than for men (40). There is a gender-dependent difference in solution searched for removing the focus from a preoccupation with bereavement to the rest of life: women find emotional attachment with children and grandchildren, and men are more likely to be socially engaged and involved in activities outside the home (38). In this study, although there were no differences in loneliness between men and women, difficulties in relations with family members and relatives, indicating emotional aspect of loneliness, were identified as to be a factor which can separate participants according to the level of loneliness. There is an increasing awareness among researches that the risk factors for emotional and social loneliness should be explored separately (37, 38, 41). When our results are viewed in this context, it was a wrong approach that we have made no clear distinction between living alone, being a single (not married), and experiencing the loss of the intimate partner.
The awareness on the distinction between emotional and social loneliness may help improve the future studies` design on loneliness. While some socio-demographic and social correlates of loneliness are consistent across the studies and settings, such as low socio-economic status, poor education, limited social network, and marital history, there are still uncertainties with the full range of potential social, and especially emotional and psychological factors, which can associate with loneliness (37, 39, 42). In this respect, although the full size prediction model in this study achieved a high level of explainability of the variance of loneliness (around 60%), there is still a room for improvement, by adding factors such as those associated with coping styles, personality characteristics, psychological resources (resilience), emotional responses, and cultural issues (10, 16, 39, 43). In addition, there is the need to more clearly identify social situations which in some surrounding may lead to loneliness in older individuals, or on the contrary, which may act protectively. In this study, we showed that some leisure and occupational activities may have a favorable effect against the feeling of loneliness. The results of this study support the findings of some other studies that personal religiosity, going to the church, or being engaged in religious organizations, may decrease emotional/social loneliness and improve the psychological well-being of older individuals, especially in the period of bereavement (44). However, there is an increasing awareness that loneliness in older individuals is a variable and multidimensional experience, which in the concrete person can be fully understood only by getting insights into the person`s life and environmental situation, which can be, as some authors stated, achieved only by including also qualitative methods and case descriptions in research on loneliness (45, 46).
The need for the more comprehensive approach in research of loneliness is also indicated, although indirectly, by the results of this study on associations of the diagnoses of chronic health conditions with loneliness. As it can be seen from differences in selected disorders, more important than listing the diagnoses of diseases could be knowing information on the level of disability and dependency on others, which a disease can provoke (as indicated, e.g., by the fact that cerebrovascular disease did matter, and cardiovascular disease did not). A disability associated with sensory organ impairment does also highly matter when evaluating the functional status of older individuals (47). It is also important to know comorbidities and complications associated with the indexed disease; in this study, e.g., the selected diagnoses, type 2 diabetes and severe osteoarthritis, are both known as disorders that are accompanied with a high level of comorbidity and disability (48, 49). In addition, as indicated by the knowledge gained so far, the level of adaptation to chronic disease (in association with providing an adequate social and service support) could be of the utmost importance for avoiding negative emotions and achieving a satisfactory level of psychological resilience and well-being (13, 16, 50, 51). Education of older individuals with chronic diseases in changing unproductive into productive coping with stress mechanisms, is considered the right intervention approach to improving adaptation to chronic diseases (51, 52).
And finally, considering that chronic diseases tend to appear in one person together, as coexisting diseases (which is termed multimorbidity), and that both, the composition and the severity of chronic diseases, determine their associations with functional impairments, there is a trend, which we have also promoted in our previous work, to represent chronic health conditions, found to dominate in a particular population of older individuals, as the subgroups of disorders which cluster together (14, 53, 54). This approach could also help us getting more insights into understanding of associations between low mental health and somatic comorbidities in older individuals, due to the fact that mental health impairments in older individuals are less considered as psychiatric disorders per se, and more as a somatization reaction to chronic disease (an extreme focus on symptoms, which causes emotional distress and aggravates low functioning) (55). This, in our study, is indicated by disorders such as chronic pain and constipation, which were found to be more expressed in participants with higher levels of loneliness. Somatization can be a consequence of low coping strategies with unfavorable social situations or with the presence of chronic diseases, or may be an expression of the common neurobiological pathways which underlie both, somatic disorders and low mood (56, 57). Important to mention, in this integrated view of associations between somatic comorbidities and mental health disorders in older individuals, is the role of frailty – the condition characterized with homeostatic mechanisms exhaustion and increased vulnerability to poor health-related outcomes, which develops as a result of the integrated action of multiple comorbid disorders in the combination with advanced age (58). Sensory organ impairments, and chronic pain, indicated in this study as conditions which may discern between those with lower vs. those with higher level of loneliness, can be considered as signs of frailty (24, 59). The characteristics of frailty, such as low level activity and chronic fatigue, may in a part explain associations between low self-rated health, limited functioning, self-imported social isolation, and loneliness (9, 10, 60).
Since physical resilience and psychological resilience in older individuals are known to mutually interact, interventions made at either axis is expected to lead to the improvement of health-related outcomes (16). As suggested by our results, interventions aimed at improving emotional regulation and alleviating negative moods, as well as interventions aimed at securing better social support and enhancing functional independence, would be efficient ways to decrease loneliness among older individuals with comorbidities and emotional stress associate with it. Similar to our result, the large-scale study performed in USA, in a comprehensive mail-in survey of successful aging, assessing physical, cognitive, and psychological domains, emphasized resilience and depression as factors that are significantly associated with self-rated successful aging, with effects comparable in size to that for physical health (61). Regarding the health-related factors, evidence suggests that maintenance of healthy lifestyles in later life could have a crucial role in promoting resilient brain during aging, which is then capable to efficiently cope with daily stressors (62).