Objective. To highlight the significant impact of social relationships on health and illness and suggest implications of these effects for health promotion efforts among older adults. Data Sources. Published studies on social relationships and health (or health behaviors) for the period 1970-1998 were identified through MEDLINE by using the key words social relationships, social support, and health, as well as review of health-related journals such as the American Journal of Epidemiology, Annals of Epidemiology, American Journal of Public Health, Journal of Health and Social Behavior, Social Science and Medicine, and the Journals of Gerontology. Study Selection. Major published original research was considered. Where published research was too extensive for full discussion of all studies, preference was given to studies focusing on older adults and those using stronger methodology (i.e., representative samples, longitudinal data, or multivariate analyses controlling for potential confounders). Data Extraction. Reported findings were organized in terms of three major categories: (1) results related to major health outcomes such as mortality, CHD and depression; (2) findings related to health behaviors; and (3) findings related to potential biological pathways for observed health effects of social relationships. Data Synthesis. Protective effects of social integration with respect to mortality risk among older adults are the most thoroughly documented, although protective effects have also been documented with respect to risks for mental and physical health outcomes and for better recovery after disease onset. There is also now a growing awareness of the potential for negative health effects from social relationships that are characterized by more negative patterns of critical and/or demanding interactions, including increased risks for depression and angina. Biological pathways are suggested by evidence that more negative social interactions are associated with physiological profiles characterized by elevated stress hormones, increased cardiovascular activity, and depressed immune function, whereas more positive, supportive social interactions are associated with the opposite profile. Conclusions. Available data clearly indicate that social relationships have the potential for both health promoting and health damaging effects in older adults, and that there are biologically plausible pathways for these effects. Such evidence suggests that aspects of the social environment could play an important role in future health promotion efforts for older adults, although careful consideration of bath potentially positive as well as negative! social influences is needed.
