Correlates of loneliness in older adults in Shanghai, China: does age matter?

Loneliness in older adults is a public health issue nowadays and deserves an increasing concern due to its negative effects in older adult population. Loneliness imposes a variety of health risks to older adults, such as cardiovascular health risk, declined cognitive function [1], increased morbidity and mortality [2‐4] and poor mental health, including depression, anxiety, and sleeping disorder [5, 6], and health care utilization [7]. Loneliness is a unique clinical entity that deserves special and separate attention compared to other mental health issues, such as depression and anxiety [8]. The current study aims to examine the correlates of loneliness in the context of China, with an ultimate goal of improving the design and the intervention of alleviating loneliness among older adults.

Despite a sizeable literature on loneliness in Western societies, much less is known about loneliness in older adults in China, a country where family system and collectivism are greatly emphasized and one where dramatic socio- and demographic- changes have taken place in the past 40 years [9]. The culture and the family in which an individual develops is an important factor that affects their experience of loneliness [10, 11]. For instance, Dykstra found that older adults in more familialistic countries tend to be more lonely than those in individualistic societies [12]. Yum argued that loneliness is more about personal romantic expectations in some individualistic cultures, but is more related to social approval and social network in collective cultures [13]. Embedded in a collective culture, older adults may experience a higher risk of loneliness due to great changes, including increasing migration, smaller households, the weakening of traditional filial piety, and a rising number of empty-nest older adults. Moreover, people may be reluctant to acknowledge loneliness due to the Chinese cultural influence (e.g., face culture). The worse scenario is the unidentified loneliness, and the invisible loneliness without timely interventions leads to even undesirable consequences, as seeking help for emotional problems is less socially acceptable than for physical problems [14, 15]. The cultural difference and the characteristics of the aging population in China deserve special attention and research to provide a better understanding of the phenomenon of loneliness.

Although the definitions of loneliness are inconsistent in the literature, loneliness is usually defined as a subjective, negative, and distressing feeling due to the deficits in social contacts [16‐20]. Due to the inconclusive definitions of loneliness, different measures have been used to assess loneliness, mainly including UCLA loneliness scale [19], De Jong Gierveld Loneliness scale [16, 21, 22], and a single item measure [9, 23]. The different measures make it difficult to compare the prevalence of loneliness of older adults in different countries or regions. Despite this, research suggests that there is an increase in the level of loneliness in older adults in China. For instance, Yang and Victor found that the prevalence of loneliness of older adults increased from 15.6% in 1992 to 29.6% in 2002 [9]. Similarly, Luo et al. found that there was a significant increase in the mean level of loneliness from 2002 to 2008 using a nation-wide survey [23]. Given the consequence of loneliness and the increase in the prevalence of loneliness in older adults in China, it is imperative to examine the loneliness level in older adults and the factors that are important to loneliness.

An overview of the current literature reveals that the correlates of loneliness in older adults include socio-demographics, health conditions, and psychosocial factors. The socio-demographic features associated with loneliness include being very old [18], gender [18], living arrangement [9, 24], and marital status [24, 25]. Regarding health factors, results imply that older adults with poor health in general tend to experience a higher level of loneliness, as poor health may hinder the maintenance of social contacts, and limit their daily activities that could be used to cope with loneliness [26]. Measures that have been used included self-rated health(SRH) [25, 27], chronic diseases, and ADL disabilities [27‐29]. In addition, social factors were also found to be associated with loneliness, including social contacts [30, 31], social relationship satisfaction and social participation [9, 32].

In addition to these well-studied variables in the literature, new media use in older adults has been a hot topic in the communication research. Most studies have examined the characteristics of older adults who use new media or not, and the reasons why older adults do not use new media [33, 34]. Another line of research is about the impact of new media use on older adults’ well-being [35, 36],andquality of life [37]. Previous research suggests that the use of new media could be used to reduce loneliness in the older adults [38]. For instance, social media is a perfect platform for older adults and their family to communicate, especially when older adults usually do not share the same household with their adult children nowadays. Nevertheless, there is also research suggesting that the Internet use doesn’t necessarily benefit older adults in terms of loneliness [39]. In China, the ownership rate of smart-phones in middle-aged and older adults reached 95.6% [40], and more specifically the Internet penetration rate in Shanghai reached 74.1% [41]. New media has greatly impacted older adults’ way of living, such as reading news, entertainment, and social contact. Given the prevalence of new media in older adults, how it will impact older adults’ loneliness needs more research.

Despite a large number of literature on the correlates of loneliness, to the best of our knowledge, little is known about the age-group differences in the correlates of loneliness of older adults. Gerontological research suggests that the older adult population is not homogeneous, and there are two phases in late adulthood that are characterized by various qualities--namely the young old and the old old [42‐44]. The young old (60–79), who are in their post-employment stage, are generally well functioning both physically and cognitively, and are actively socially engaged [42‐44]. However, the old old (80+) often face increasing limitations to their independence due to more chronic conditions and health restrictions, and have a less-desirable psychological profile, and more death-related [42‐44]. The different internal and external conditions associated with the two age groups may play a significant and differential role when older adults adapt to the age-related changes, such as social losses, or loneliness. Previous research on correlates of well-being and life attitudes actually indicates an age-differential pattern. For instance, Isaacowitz and Smith found that cognition and social relationships were significant predictors of positive affect only for the young old, but not for the old old [45]. Similarly, Jopp et al. supported age-differential correlates of valuation of life, such that health factors was especially stronger for the young old than for the old old [46]. These findings demonstrated that the correlates of loneliness might operate differently for the young old and the old old. For instance, the young old, who are relatively healthier, may compensate for being lonely by socially engaging in different activities, such as volunteering or square dancing. However, this might not be the case for the less healthier old old. Further empirical research is necessary to provide a more comprehensive picture of the extent to which the importance of the correlates of loneliness would vary in the two stages of later life.

We focused on older adults in Shanghai in the current study. China represents the largest aging population around the world. The population aged 60 and above accounts for 16.7% of the total population by 2016 [47]. Among the cities, Shanghai is the first aging city of China and has been an aging city since 1979, and the percentage of its aging population (60+) has reached 33.2% by the end of 2017 [48], which is almost twice higher than that of the whole nation. In addition to the high percentage of older adults, the aging population in Shanghai is also characterized by a high percentage of the oldest old and those living in “empty nest” households. The percentage of the oldest old (80+) has reached 16.4% among the older adults. The number of older adults living in “empty nest” households is more than 1 million, among of whom, 0.3 million are 80 years old and above. In this sense, Shanghai becomes an ideal place to examine older adults’ loneliness and its correlates. We believe such a study in Shanghai would shed light on the loneliness in other cities in China and would provide statistics to compare with other countries.

The mean age of the participants in this study was 69.72 years old (SD = 8.07). The majority of participants were current married (81.0%), with a highest education of junior high or lower (65%), and living with others (87.6%). In addition, 11.2% volunteered in the previous year (Table 1).

We also compared the differences of the key variables in the young old and in the old old (Table 1). Moreover, there were significant age-group differences with respect to marital status, education, living arrangement, volunteering experience, objective health indicators, contact with neighbors, playing mahjong, and new media use. Compared to the young old, the old old had a higher percentage of being widowed, were generally less educated, and were more likely to live with others. About 13% of the young old reported volunteering, but only 2.3% of the old old did. In addition, the old old reported more IADL and chronic diseases, and lower level of chatting with neighbors, square dancing, and new media use compared to the young old.

In terms of loneliness, the mean score was 18.48 (SD = 5.77). The loneliness level in the old old (M = 19.27, SD = 5.30) was significantly higher than that in the young old (M = 18.35, SD = 5.84).

The present study represents a first attempt to investigate loneliness in older adults of Shanghai, China, with a particular attention on the similarities and differences in the correlates between the young old and the old old in the backdrop of emerging socio-demographic features. We found that the old old reported a higher level of loneliness, and health status and social factors were important correlates of loneliness. In addition, analyses revealed an age-differential pattern, such that SRH and family functioning were important for both the young old and the old old, whereas marital status, education, social participation, and new media use were significant only for the young old.

We found that 4 to 39% of the participants agreed with at least one of the six items of the De Jong Gierveld loneliness scale, which suggests that the level of loneliness is not very high as expected, but still a small proportion of the sample acknowledge the feelings of loneliness. This finding challenges the often-held notion that older adults are lonely. There are several possible reasons. First, older adults may be reluctant to acknowledge the feeling of being lonely, as direct admission possibly indicates a lack of a satisfactory family relationship and in turn compromise their self-worth. Second, some older adults who live alone may have the necessary resources to choose such a living arrangement to pursue a sense of autonomy [58, 59]. Especially for the young old, living alone is not a significant risk factor of loneliness and this finding implied that living alone may not necessarily cause loneliness, instead it might be a personal choice out of autonomy and an indication of independence. More research is needed to explore the reasons behind such a phenomenon. More importantly, we noticed that those who were older, with rural hukou, single and widowed, less educated, living alone, in poorer health, with poorer family functioning, less socially engaged, and less frequent in using new media tend to experience a higher level of loneliness, and this suggests a “disadvantage accumulation” phenomenon. It is imperative to understand the meaning, reasons, and level of suffering implied on those feelings of loneliness, and to design multi-factorial interventions that target the deficits for older adults at high risk of loneliness.

Consistent with previous research, the mean level of loneliness was higher in the old old than in the young old [12, 18]. The old old experience more physical limitations and more social losses, and their expectations for social contacts may be difficult to fulfill, thus they experience higher loneliness than their young old counterparts. Regarding gender differences, we did not find significant results. This is in line with previous research implying that gender alone may not be directly related to loneliness, but rather interacts with other factors, such as marital status and living arrangement [18]. In addition, compared to married older adults, those who were single and divorced reported a significantly higher level of loneliness, whereas the widowed older adults were not significantly different from their married counterparts. Most of older adults’ support needs are satisfied by their spouses, being single or divorced implies an absence of such an intimate relationship, thus leading to more loneliness [60]. However, widowhood might be more of a normative state than being single or divorced for older adults [61], and this state imposes a less impact on loneliness, especially when they have a satisfactory family relationship. Moreover, more educated older adults tend to experience less loneliness, because they generally have better coping abilities and more resources to enrich their life and combat loneliness [62]. Compared to other living arrangements, those living alone reported a higher level of loneliness, and this is consistent with previous findings [63].

In terms of health factors, the three health indicators in the study were all significant, such that better health is associated with a lower level of loneliness. The loss of social contacts is part of the aging process, and poorer health further imposes obstacles for older adults to socialize, to effectively communicate with others, and to participate in social activities, all of which could be used to cope with loneliness [26, 64]. Less healthy older adults may also possess less resilience and lower self-efficacy to establish and maintain social contacts. Thus older adults in poorer health are at higher risk of loneliness.

In terms of social factors, humans are social creatures and a higher level of social integration and connections are protective factors of loneliness. Helping those in need and maintaining engagement with others is two important feature of volunteering, and it is a powerful antidote to loneliness because it gives people the opportunities to be needed [65]. In Shanghai, there are about 2.6 million registered volunteers, among whom 34% are over 60 years old by 2016 [66], and being a volunteer serves as a meaningful buffer for older adults’ loneliness. In addition, as there is an increase in the number of empty-nest older adults, neighbors are more easily available due to the proximity, and would be especially important for older adults’ daily contacts, thus associating with a lower level of loneliness [18].

An important question we examined in the current study is the age-group differences in the correlates of loneliness. Results revealed that SRH and family functioning are two very important correlates for both the young old and the old old. SRH is a global measure of health status [67], and has been found to predict a variety of outcomes, e.g., mental health, mortality [68], and loneliness [69]. Even though SRH is influenced by objective health, there is evidence suggesting that the correspondence between chronic diseases and SRH generally declines with age [70]. The old old could still have a positive SRH despite their poor health, and this opens up the opportunities for the intervention to improve older adults’ SRH [71]. In addition, family functioning is a very strong protector of loneliness. In the context of China where the culture greatly emphasized family system, individuals experience more intimate connections from family members, have more emotional communication, have higher mutual trust, and are better able to adapt to changes or hardships in a well-functioning family, all of which could minimize loneliness [3].

It is interesting that some of the significant factors for the young old lost their significance for the old old. Variables, including marital status, education, IADL, contact with neighbors, volunteering, and square dancing were no longer significant for the old old. These differences might be largely due to the differing qualities in these two age groups. The demographic variables are no longer significant for the old old, perhaps loneliness in the old old is more of a result of internal factors, such as how they view their health and family relationship [46]. The worse health status might be more of a normative phenomenon for the old old, whereas worse health status is an “off-time” event for the young old, thus leading to a significant effect of IADL and chronic illnesses only for the young old [72]. Healthier young old might benefit more from the social participation, whereas poorer health attenuates the beneficial effect of social engagement for the old old. It is also possible that the level of social participation in the old old does not reach the minimal level, beyond which social participation can be effective for alleviating loneliness. More importantly, new media use was associated with lower loneliness for the young old, but not for the old old. Perhaps the young old who are relatively healthier are more affected by the pervasive new media and are more capable of using the new media to communicate with others, to read what they are interested in, to entertain on the new media than the old old [33, 34].

This study has meaningful implications for interventions of alleviating loneliness. First, our study challenges the negative stereotype that older adults are lonely, actually the majority of the older adults reported being not lonely. Nevertheless, those who reported being lonely and those with higher risk of loneliness deserve more concern. Second, age-group differences were found for the correlates of loneliness, and age-tailored interventions are needed to better help older adults prevent or combat loneliness. For instance, interventions addressing SRH and family functioning are appreciated for both age groups, and those creating more opportunities for the young old to volunteer also could be helpful. Third, in an era of new technology, we could take advantage of new media to help them enrich their life and better communicate with their family or friends, such as teaching older adults how to use new media (e.g., Wechat) to chat with family or friends. Study suggested that contemporary information and communication technologies have the potential to prevent or reduce the social isolation of older adults [73]. Lastly, future interventions would be beneficial using a life-course perspective. Targeting risk factors at earlier stages of life could also prevent loneliness at later life stages.

There are several limitations that should be noted. First, the study is based on a cross-sectional dataset, thus the causal relationship among the variables cannot be established in the study. Future research is needed to examine what factors predict the onset and the alleviation of loneliness using longitudinal data, thus providing more robust evidence for interventions. Second, other factors that could possibly influence loneliness are not considered given the current information of the data, such as personality and community-level factors. It would benefit if future research could incorporate multi-level factors to provide an overarching picture of loneliness of older adults. Third, participants included in the study were only community-dwelling older adults. The results may not be generalized to those who live in the institutions.

Our study examined correlates of loneliness, and as far as we know, and was the first attempt to explore age-group differences in loneliness of older adults in Shanghai, China, where population aging displays its unique features. Results revealed that loneliness in older adults does not prevail as the negative aging stereotype suggests. Older adults who are living alone, single and divorced, less educated, in poorer health, less socially engaged, less satisfied with family relationship, and less frequent in using new media are at a higher risk for loneliness. In addition, different important correlates of loneliness in the young old and in the old old are identified, such that some essential factors in the young old lost their significance in the old old. Future interventions should be tailored to the young old and the old old to better help older adults alleviate the loneliness and enhance their well-being.