OBJECTIVES: Older people are often socially deprived, which may predispose them to frailty. The aim of the study was to identify the independent effects of social deficits and their combined effects with frailty ("social frailty") in predicting functional decline, institutionalization, and mortality in older adults living in the community.
DESIGN: Longitudinal study with a 3-year follow-up.
SETTING AND PARTICIPANTS: A nationally representative sample of community-dwelling older people aged 65 years or older in South Korea (N = 11,241).
METHODS: Social deficits were defined as deficiencies in socioeconomic status, living situation, social network size, contact with social networks, social support, and social activities. The number of social deficits was categorized as none (0), low (1-2), and high (3 or more), based on the summated score. Frailty status was defined using the Fried criteria. Cox proportional hazards model was used to analyze the effects of the increasing levels of social deficits on functional decline, institutionalization, and mortality. The combined effects of both social deficits and physical frailty were also assessed, controlling for covariates.
RESULTS: The prevalence of frailty was 7.7%, with 49.1% being prefrail at baseline. Overall, the presence of social deficits tended to elevate the risk of adverse outcomes, where acquiring 1 social deficit was associated with an 11% to 27% increase in their risks. In the multivariable analysis, those with a high level of social deficits demonstrated a significantly elevated mortality risk [hazard ratio (HR) = 1.45, 95% confidence interval (CI): 1.07, 1.97). There was a trend toward increasing hazards when social deficits were combined with physical frailty, with the group having a high level of social deficits and being frail exhibiting the highest mortality risk (HR = 3.14, 95% CI: 1.81, 5.46).
CONCLUSIONS AND IMPLICATIONS: Social adversity independently predicted mortality. When combined with physical frailty, an accumulation of social deficits increased the predictive value.