|Title||Dimensions of Social Isolation and Adverse Health Outcomes among Older Men and Women in the United States: How Aging, Living Alone, and Obesity Contribute to Mortality|
|Year of Publication||2022|
|University||University of Massachusetts Amherst|
|Keywords||health outcomes, Living Alone, Mortality, Obesity, social isolation|
This dissertation examines the interaction of a series of potential social and biological mechanisms that may affect inequalities in mortality among those 65 or older. The mechanisms are social isolation and its relationships with obesity and diabetes. Although many prior studies have confirmed social isolation as a risk factor for adverse health outcomes such as obesity, diabetes, and even premature death, how and why social isolation works to generate adverse health outcomes remains mostly unknown. To extend the scholarship on social isolation and health, each chapter of this dissertation tests distinctive aspects of social isolation: living alone, rurality, and retirement, respectively, while using various datasets from the Health and Retirement Study. First, this dissertation finds that older women’s living alone status is related to a lower likelihood of sarcopenic obesity, whereas their widowhood status showed a higher possibility of sarcopenic obesity. Second, while the rural-urban classification complicates results, older men experiencing severe food insecurity, accounting for rural-urban distinctions, tended to have a greater likelihood of obesity than not being obese. Last, older adults who were partly retired or had a pension had a lower mortality risk. However, their significance disappeared after adjusting for confounders. Yet, in-person contact emerged to be associated with a decreased mortality risk. The dissertation contributes three novel ideas for scholars with expertise in social determinants of health. First, it bears keeping in mind that although a composite construct such as social isolation can be effectively used to examine health outcomes, the relationship between each component of the variable and outcomes should also be carefully examined. Second, this thesis constructed a geographic metabolic risk partition measure in order to establish high-metabolic-risk states vs. low-risk states. This measure reflects the reality that the prevalence of obesity and diabetes is deeply rooted in an environmental context such as a state. Third, clinical research already focuses on sarcopenic obesity, but social factors for the disease are rarely studied. Since this dissertation found younger cohorts’ greater likelihood of sarcopenic obesity among older Americans, social scientists should pursue new frontiers in research focused on the social aspects of sarcopenic obesity.