Matters of the heart: exploring the association between loneliness, social isolation and all-cause mortality among individuals living with cardiovascular disease.

Introduction: This thesis aimed to explore the associations between subjective (loneliness) and objective (social isolation) aspects of social health on mortality risk among individuals living with cardiovascular disease (CVD). More specifically, this research aimed to systematically identify and synthesise empirical evidence regarding these associations (Chapter 2). To extend the evidence base, the empirical study presented in Chapter 3, sought to investigate whether loneliness and social isolation are associated with an increased risk of mortality among individuals with established CVD. More specifically, whether the associations differ across global, emotional and social loneliness.

Methods: Regarding the systematic review and meta-analysis presented in Chapter 2, five electronic databases were searched from inception to 25th November 2021. In all, 35 studies were included in a narrative synthesis and where appropriate, a meta-analytic evaluation using a random-effects model. The sample included in the empirical paper (Chapter 3), were individuals with established CVD drawn from two longitudinal prospective studies: the Health and Retirement Study [HRS, N = 5762, M (SD) = 70.57 (8.95) years, 42.8% male] and the Berlin Aging Study [BASE, N = 465, M (SD) = 84.94 (8.58) years, 50% male]. Separate cox proportional hazard regression models were used to examine the association between loneliness and social isolation and risk of all-cause mortality for both studies.

Results: The findings presented in Chapter 2 indicated subjective and objective aspects of social health were associated with an increased risk of all-cause mortality. However, meta-analysis was not feasible for loneliness as exposure. In Chapter 3, the adjusted models in the HRS indicated, loneliness: global [HR = 1.14; 95% CI (1.08– 1.20)]; emotional [HR = 1.09; 95% CI (1.03 – 1.14)] and social [HR = 1.07; 95% CI (1.01 – 1.13)], were significant predictors of all-cause mortality. However, these findings were not replicated in BASE. Social isolation was not a significant predictor of mortality in either study.

Conclusions: Limitations were noted in both studies. In Chapter 2, there was evidence of publication bias, large methodological differences across studies and meta-analytic data could not be drawn for loneliness. In Chapter 3, the possibility of survivor effects as well as methodological differences may explain why findings were not replicated across studies. However, both studies make valuable contributions to knowledge and clinical practice and support public health concerns regarding the deleterious impacts of these social health factors on health outcomes. Further research examining loneliness among those living with CVD is required so further comparisons between these aspects of social health can be made.