Social relations and self-reported health: a prospective analysis of the French Gazel cohort
Introduction
Although over the past 20 years, evidence for the association between social relations and health has accumulated (Berkman, 1984; House, Landis, & Umbertson, 1988; Berkman, Glass, Brissette, & Seeman, 2000), the health impact of the social environment received little research attention in France (Antonucci, Fuhrer, & Dartigues, 1997; Briancon et al., 1985). Several studies have shown that social networks, the structural aspect of social relations, predict mortality (Berkman & Syme, 1979; Blazer, 1982; House, Robbins, & Metzner, 1982; Kawachi et al., 1996; Orth-Gomer & Johnson, 1987; Seeman et al., 1993a; Vogt, Mullooly, Ernst, Pope, & Hollis, 1992). Hypothesized to exert a very general effect upon health for a number of reasons, social networks are associated with cardiovascular mortality and morbidity (Kawachi et al., 1996; Olsen, 1993), with the recovery from myocardial infarction, cancer and stroke (Vogt et al., 1992; Berkman, 1995), the occurrence of psychiatric disorders (Fratiglioni, Wang, Ericsson, Maytan, & Winblad, 2000; Stansfeld, Rael, Head, Shipley, & Marmot, 1997; Stansfeld, Fuhrer, & Shipley, 1998b), quality of life (Achat et al., 1998), health functioning (Michael, Colditz, Coakley, & Kawachi, 1999; Stansfeld, Bosma, Hemingway, & Marmot, 1998a) and health status (Krantz & Orth, 2000; Litwin, 1998).
The association of social support with mortality and morbidity (Blazer, 1982; House et al., 1988; Fuhrer et al., 1999; Liang et al., 1999), as well as with self-assessments of health status (Minkler, Satariano, & Langhauser, 1983; Krause, 1987; Franks, Campbell & Shields, 1992; Baker & Taylor, 1997; Hillen, Schaub, Hiestermann, Kirshner, & Robra, 2000; Krantz & Orth, 2000; Lindstrom, Sundquist, & Ostergren, 2001) has also been consistently demonstrated. An important aim of research in this area is to clarify these paradigms and test their distinct association with health outcomes.
The causal relationship between social relations and health is not well understood, and social networks are likely to work through multiple pathways: behavioral and material, psychological and physiological (see Berkman et al. (2000) for an in-depth review). First, extensive social ties might increase the likelihood that individuals engage in health-promoting behaviors (physical activity, beneficial dietary practices) or refrain from health-damaging ones (cigarette smoking, excessive alcohol drinking). In addition, among individuals experiencing health problems, those with adequate social support will receive advice, services, material or financial aid from others, and thus will likely benefit from better medical and non-medical care (Berkman, 1984). Second, poor social relations affect psychological health. A lack of sufficient social support is associated with lower self-efficacy, which, in turn, has been shown to predict worse health outcomes. What is more, social networks and social support are also directly associated with the occurrence of psychiatric illness, and particularly depression (Blazer, Hughes, & George, 1992; Stansfeld et al., 1998b; Vilhjalmsson, 1993; Oxman & Hull, 2001). In addition, depression may also influence the quality of social relations. Another understudied psychological factor is the sense of meaning and purpose largely influenced by social interactions, which is thought to be a precondition for good health. Finally, there is increasing evidence about complex pathological mechanisms, which are likely to mediate the impact of social relations on well-being. Social isolation has been hypothesized to act as a chronic stressor, resulting in the accelerated aging of the human organism (Berkman, 1988; Seeman & McEwen, 1996). Animal as well as human studies have provided strong evidence for neuroendocrine changes, specifically in cortisol, epinephrine and norepinephrine levels, induced by social isolation (Seeman, Berkman, Blazer, & Row, 1993b). In addition, social relations are also hypothesized to alter the immune system, influencing the body's susceptibility to illness (Cohen, Doyle, Skoner, Rabin, & Gwaltney, 1997).
Broadly speaking, there are two ways in which social factors are thought to exert an impact on health (Cohen & Willis, 1985). First, social networks and social support might “buffer” the health effects of stressful life events. According to this hypothesis, there is an interaction between social relations and life events, and life events only have a detrimental effect on well being in the absence of an adequate social environment. The finding that perceived social support has been shown to buffer the risk of depression among individuals exposed to stressful life events seems to support this paradigm (Cohen & Willis, 1985; Vilhjalmsson, 1993). With regard to mental health outcomes, the evidence is strong that social support buffers life events, although several recent studies have also found support for the independent health effects of poor social ties (Liang et al., 1999; Stansfeld, Bosma, Hemingway, & Marmot (1998a), Stansfeld, Fuhrer, & Shipley (1998b), Stansfeld, Rael, Head, Shipley, & Marmot (1997)).
According to the “main effects” model, positive as well as negative aspects of social relations exert an independent effect on health. Thus, poor social networks and a lack of social support act as stressors and contribute to poor health in and of themselves, while extended social networks and satisfactory social support enhance well being. This hypothesis is illustrated by the consistent main effect associations between a lack of social relations and all-cause mortality (Berkman & Glass, 2000, p. 151–153). The scientific debate regarding the definition and optimal measurement of social relations adds to the lack of consensus on the causal pathways between the social environment and health and has been the cause of inconsistent results in the field. A greater concern about the development of theoretically based instruments (Berkman & Seeman, 1986) has led to the demonstration of the complex relationship between social networks and social support (Seeman & Berkman, 1988). A psychometric analysis of the Berkman Syme Social Network Index (SNI) in an elderly population, has argued for the presence of four dimensions of social networks, concurring with four groups of social counterparts: children, friends, relatives and a confidant (Glass, Mendes de Leon, Seeman, & Berkman, 1997). This finding confirms prior research, which indicated that the size of the social network, and not the identity of its members, has an effect on health (Berkman, 1987).
The purpose of the current study is to determine whether social networks and social support are associated with self-reported health status in a cross-sectional analysis, and whether they predict self-reported health over a 1-year follow-up in a population of middle-aged men and women who are part of an ongoing occupational cohort of gas and electricity workers in France. A single-item measure of self-rated health status has been shown to be an accurate predictor of mortality and chronic morbidity (Idler & Benyamini, 1997; Moller, Kristensen, & Hollnagel, 1996). As such, it is a useful indicator of health status in young and middle-aged populations. This study also offers an opportunity to examine the SNI questionnaire in its French version. It allows us to add to the existing cross-cultural evidence for the importance of the association between social relations and health.
Section snippets
Study population
The Gazel cohort was recruited in 1989 among employees of France's national gas and electricity company: Eléctricité de France-Gaz de France (EDF-GDF). At that time, 20 624 employees gave consent to participate in this study of a variety of health-related risk factors, referred to as an “epidemiological laboratory” (Goldberg et al., 1994). At baseline, in 1989, the cohort comprised men aged 40–50 and women aged 35–50 (Goldberg et al. (1990a), Goldberg et al. (1990b)). The volunteers are followed
Cross-sectional study
Among the 11 105 men and 4070 women who responded to the 1994 questionnaire, respectively, 18.5% and 22.9% reported being in poor health. Among men and women, the social support scale correlated with the SRS (respectively rho=0.41 and rho=0.44), as well as with the SNI (respectively rho=0.26 and rho=0.28). There was no association between the SNI and SRS (among men rho=0.04, among women rho=0.03). Table 1 shows the distribution of demographic, behavioral and occupational characteristics in
Discussion
We studied the relationship between social networks, satisfaction with social relations and social support and self-reported health among middle-aged men and women participating in the Gazel cohort study. In the cross-sectional part of our study, for men social isolation, poor satisfaction with social relations and inadequate social support were associated with poor health, while for women social support was the most significant predictor. In the prospective part of our study these results were
Conclusion
Our study found that social support and overall satisfaction with social relations, as well as occupational status, are predictive of self-reported health. The role of social relations in affecting specific health outcomes, over longer periods of follow-up, requires further study. Nevertheless, these results confirm the importance of social factors in determining the well-being of a cohort of French volunteers and indicate that both clinicians in their practice and researchers may do well to
Acknowledgements
The authors wish to express their gratitude to all the volunteers of the Gazel cohort, EDF-GDF, especially the Service des Etudes Médicales and the members of the Gazel study team, in particular Isabelle Bugel, Sébastien Bonenfant and Jean-François Chastang for their technical assistance. Our thanks go to Steven Gortmaker and Maria Glymour for their thoughtful comments on earlier versions of this manuscript.
References (57)
- et al.
The relationship between condition-specific morbidity, social support and material deprivation in pregnancy and early motherhood
Social Science & Medicine
(1997) - et al.
From social integration of healthDurkheim in the new millenium
Social Science & Medicine
(2000) - et al.
Social relationships and healthThe relative roles of family functioning and social support
Social Science & Medicine
(1992) - et al.
Influence of social network on occurrence of dementiaA community-based longitudinal study
Lancet
(2000) - et al.
Beyond single indicators of social networksA LISREL analysis of social ties among the elderly
Social Science & Medicine
(1997) - et al.
Global self-ratings of health and mortality hazard in the North Carolina Piedmont
Journal of Clinical Epidemiology
(1996) - et al.
Stress, social relations, and old age mortality in Taiwan
Journal of Clinical Epidemiology
(1999) Social network type and health status in a national sample of elderly Israelis
Social Science & Medicine
(1998)- et al.
Do risk factors and health behaviors contribute to self-ratings of health?
Social Science & Medicine
(1999) - et al.
Social network interaction and mortalityA six-year follow-up of a random sample of the Swedish population
Journal of Chronic Disease
(1987)
Structural characteristics of social networks and their relationship with social support in the elderlyWho provides support
Social Science & Medicine
Intercommunity variations in the association between social ties and mortality in the elderlyA comparative analysis of the three communities
Annals of Epidemiology
Life stress, social support and clinical depressionA reanalysis of the literature
Social Science & Medicine
Social networks as predictors of ischemic heart disease, cancer, stroke and hypertensionIncidence, survival and mortality
Journal of Clinical Epidemiology
Social networks, stress and health-related quality of life
Quality of Life Research
Social relations and depressive symptomatology in a sample of community-dwelling French older adults
Psychology and Aging
Widowhood and illnessA comparison of social network characteristics in France, Germany, Japan and the United States
Psychology and Aging
Assessing the physical health effects of social networks and social support
Annual Review of Public Health
Assessing social networks and social support in epidemiologic studies
Revue Annuelle d’Epidémiologie et de Santé Publique
The changing and heterogeneous nature of aging and longevityA social and biomedical perspective
Annual Review of Gerontology and Geriatrics
The role of social relations in health-promotion
Psychosomatic Medicine
The influence of social relationships on aging and the development of cardiovascular disease—a review
Postgraduate Medical Journal
Social networks, host resistance and mortalityA nine-year follow-up study of Alameda country residents
American Journal of Epidemiology
Social support and mortality in an elderly community population
American Journal of Epidemiology
Age and impaired subjective support. Predictors of depressive symptoms at one-year follow-up
Journal of Nervous and Mental Disease
Isolation, social support, life events and health status
Revue Annuelle d’Epidémiologie et de Santé Publique
Social ties and susceptibility to the common cold
Journal of the American Medical Association
Psychological disorder and mortality in French older adultsDo social relations modify the association?
American Journal of Epidemiology
Cited by (141)
Decent work and physical health: A multi-wave investigation
2021, Journal of Vocational BehaviorSocial isolation and depression onset among middle-aged and older adults in China: Moderating effects of education and gender differences
2021, Journal of Affective DisordersCitation Excerpt :Education was chosen as a measure of SES in this study as it is thought to be a better SES indicator than occupation or income for those who are no longer working (Nguyen et al., 2019). Building on previous evidence indicating possible gender differences in health inequalities as well as in the health impact of social relationships (Kawachi and Berkman, 2001; Liu and Umberson, 2008; Melchior et al., 2003), the moderating effect of gender differences was also examined. Accordingly, two major hypotheses (H1, H2) were tested in this study.
Alcohol consumption, life satisfaction and mental health among Norwegian college and university students
2019, Addictive Behaviors ReportsCitation Excerpt :Similarly, adults who do not drink in cultures where consumption is the norm, have been found to have lower social support (Rodgers et al., 2000b) and fewer close friends (Skogen et al., 2009), as well as lower social well-being, sociability and social participation (Peele & Brodsky, 2000). In general, social relationships and social support are associated with better physical and mental health (Berkman et al., 2000; Melchior et al., 2003; Seeman, 1996), and indeed, studies have found poorer mental health among individuals who do not drink (Rodgers et al., 2000a; Rodgers et al., 2000b; Skogen et al., 2009). Investigating mental and social wellbeing among students who abstain from alcohol is therefore of particular importance.
Social networks and patterns of health risk behaviours over two decades: A multi-cohort study
2017, Journal of Psychosomatic ResearchPositive Psychosocial Factors and Cognition in Ethnically Diverse Older Adults
2018, Journal of the International Neuropsychological Society