Elsevier

Archives of Gerontology and Geriatrics

Volume 47, Issue 3, November–December 2008, Pages 327-339
Archives of Gerontology and Geriatrics

Effects of social relationships on mortality of the elderly: How do the influences change with the passage of time?

https://doi.org/10.1016/j.archger.2007.08.015Get rights and content

Abstract

The purpose of this study was to determine the changes with time lapse in the influences of social relationship factors, in relation to mortality in the elderly. A baseline investigation was conducted in 1992 and survival conditions of 637 subjects aged 68–82 years were followed up for 12 years. The associations of social relationships with mortality were compared between the first and later half periods. The results showed that for men, close friends, group membership and finding life worth living were significantly associated with mortality in the first half period and the association disappeared in the later half; also in men, providing instrumental support was significantly associated in the 12-year period. In the first half, living arrangement was not significantly associated with mortality for men, and marital status and job were not for women, but in the later half, the association of those became apparent. The present study suggests that for men, the association of close friends, group membership and finding life worth living with mortality decreased with the passage of time while providing instrumental support had positive effects throughout the long term. The association of living arrangement increased for men, and marital status and job for women.

Introduction

Although in many previous studies dealing with social relationships the authors conducted long-term follow-ups (Avlund et al., 1998, Kushiro et al., 2002, Helmert, 2004), and tried to find characteristic aspects that could not be observed in short-term follow-ups, few studies focused on the alterations in their effects during a long term. Some aspects observed at the baseline conditions can be changed, and the intensity of the influence of each social relationship factor on mortality may change with age or the passage of time in a long term. To find out the changing or persisting aspects of various social relationship factors may help us to understand their roles in social relationships more clearly.

A number of studies have been carried out on social relationships with analysis of the aspects of social networks, i.e., the structural dimension, and social support, which represents the functional dimension (Berkman, 1983, Due et al., 1999). These social relationships directly or indirectly cause and result in active lifestyles of individuals to maintain social interactions. Active lifestyle is an important factor supporting the social relationships and the later also promotes it. In some studies, activity factors were examined in the analysis of social relationships (Walter-Ginzburg et al., 2002, Avlund et al., 2004, Fratiglioni et al., 2004). Therefore, in the present study, we examined three socially related aspects to analyze their influences on the mortality of the elderly, i.e., social network, social support and active lifestyle.

The aim of this study was to examine what changes occurred in the relationships of social factors with all-cause mortality and which aspect of social relationships persisted during the 12-year follow-up study.

Section snippets

Study population

In 1992, we selected all of 769 residents in the age range 68–82 years from the residential census data of a rural town in Hokkaido, Japan, where the population was about 7000 and the proportion of those aged 65 years and over was 18%, and distributed self-administering questionnaires to them. For 6 months after the selection of the subjects, we collected data: 652 (84.8%) persons answered, 44 refused to answer, 22 were hospitalized, 6 moved out, 3 died and 42 did not respond for unknown

Results

Table 2 shows the characteristics of the participants.

From the results of hazard proportionality test (Table 3), it was revealed that, for men, close friends, group membership, receiving emotional support, and finding life worth living significantly decreased their effects, where the hazard proportionalities were not satisfied (b > 0 and p < 0.05), and that, for women, group membership, finding life worth living, and leisure activity significantly decreased their effects, where the hazard

Discussion

We found that the effects of social relationships on the mortality of the elderly varied during the long-term observation and the increase or decrease of the effects differed from one factor to another factor. The consistency of HR that Cox's proportional hazard model postulated was not satisfied for the factors in which the effects decreased.

We postulated the following three patterns of the variables, considering the changes in the HR and their significance. The first pattern was characterized

Conclusions

In the present study we found alterations in the influences of social relationship factors through the 12 years of observation that might not be seen in short-term observation.

Acknowledgements

We appreciate the valuable assistance of Dr. Iwao Sugimura, and gratefully acknowledge municipal officials throughout the country who provided the residential data, and the elderly residents and their families in the town who generously made data available. This study was funded by Grants-in-Aid for Scientific Research from the Ministry of Health, Labour and Welfare, Grant-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology, and Grant-in-Aid from the Toyota

References (24)

  • G.A. Kaplan et al.

    Risk factors and the study of prevention in the elderly: methodological issues

  • S. Katz

    Assessing self-maintenance: activities of daily living, mobility, and instrumental activities of daily living

    J. Am. Geriatr. Soc.

    (1983)
  • Cited by (21)

    • The strength of family ties: A meta-analysis and meta-regression of self-reported social support and mortality

      2013, Social Networks
      Citation Excerpt :

      Indeed, the positive effects of support were often found to be stronger in men than in women (House et al., 1982; Kaplan et al., 1988; Wilkins, 2003), though the opposite has also been reported (Lyyra and Heikkinen, 2006). Age: Many of the previous studies on social support and mortality have focused on older-age persons (e.g., Andre-Petersson et al., 2006; Rodriguez-Artalejo et al., 2006; Rodriguez-Laso et al., 2007; Sato et al., 2007; Zhang et al., 2007). The (often implicit) assumption behind this choice is that the benefits of social support are especially pronounced in older populations.

    • The role of different predictors in 20-year mortality among Krakow older citizens

      2013, Archives of Gerontology and Geriatrics
      Citation Excerpt :

      Some researchers suggest that functional impairment is not necessarily a predictor of mortality by itself, because it can be a marker of disease and general health status (Cooper et al., 2010; Nybo et al., 2003). Studies looking for predictors of mortality cover different follow-up periods from one up to over twenty years (Bassuk et al., 2002; Bath, 2003; Beckett et al., 1996; Carmel et al., 2007; Cesari et al., 2008; Deeg & Kriegsman, 2003; Elo & Preston, 1996; Ernstsen et al., 2011; Jaffe et al., 2005; Knoops et al., 2004; Lyyra, Heikkinen, Lyyra, & Jylha, 2006; Manderbacka et al., 2003; Manor et al., 2000; Martelin, 1994; Millunpalo, Vuori, Oja, Pasanen, & Urponen, 1997; Murata et al., 2006; Nybo et al., 2003; Sato et al., 2008; Tobiasz-Adamczyk et al., 2007; van Dam et al., 2008). The results are mainly based on the assumption that it is enough to measure predictors only once at the baseline study and that their predictive ability will be stable over time (Benyamini, Blumstein, Lusky, & Modan, 2003).

    • Gender differences on the impacts of social exclusion on mortality among older Japanese: AGES cohort study

      2012, Social Science and Medicine
      Citation Excerpt :

      La Veist, Sellers, Brown, and Nickerson (1997) reported that extremely socially isolated older African American women, who were living alone and had no contact with family or friends in the 2 weeks prior to the survey, were 3 times more likely to die prematurely than non-isolated women. Similar evidence was provided in a 12-year follow-up study of 637 older Japanese people, Having fewer close friends (HR = 2.22) and group memberships (HR = 1.89) were associated with increased mortality in older men, but not in women (Sato et al., 2008). Moreover, there is abundant evidence for the link of social inactivity to poor health outcomes such as dementia (Takeda, Kondo, Hirai, & Murata, 2007; Wang, Karp, Winblad, & Fratiglinoi, 2002) and functional disability (Haga, Shibata, & Ueno, 1991; Kondo, Kazama, Suzuki, & Yamagata, 2008; Kondo, Minai, Imai, & Yamagata, 2007) in Japan.

    • Factors relating to social isolation in urban Japanese older people: A 2-year prospective cohort study

      2020, Archives of Gerontology and Geriatrics
      Citation Excerpt :

      For example, cross-sectional studies have reported that social isolation was associated with physical disorders (Saito et al., 2010; Simonsick, Kasper, & Phillips, 2015; Weinstein & Ventry, 1982), poor nutritional status (Locher et al., 2005; Walker & Beauchene, 1991), and low self-assessment of health (Litwin, 1998; Wenger, 1984). Moreover, longitudinal studies have reported social isolation to be significantly associated with early death (Berkman & Glass, 2000; Sato et al., 2008; Schoenbach, Kaplan, Fredman, & Kleinbaum, 1986), declining physical function (Seeman, Bruce, & McAvay, 1996), transition to long-term care (Saito, Kondo, Ojima, Kondo, & Hirai, 2013), and levels of health risk similar to those posed by smoking (Holt-Lunstad, Smith, & Layton, 2010; cf. Saito, Kondo, Ojima, Hirai, & the JAGES group, 2015). Several definitions have been proposed for social isolation.

    View all citing articles on Scopus
    View full text