Personality traits and health outcomes: an exploration into associations and potential mechanisms

Abstract

There were two main objectives of this thesis. First, given that personality traits have been linked to a number of diabetes risk factors and precursors such as lifestyle and the metabolic syndrome, our aim is to explore whether personality traits are associated with type 1 and type 2 diabetes mellitus. Second, we aim to investigate several potential mechanisms by which personality could influence diabetes, and other health outcomes such as cardiovascular disease and mortality. Chapter 1 provides an introductory overview of the history of personality-health research, and discusses strengths and limitations of different methodological frameworks. Chapters 2-4 focus on the associations between personality and diabetes. Two studies described in Chapter 2 examine cross-sectional and longitudinal associations between personality and type 1 and type 2 diabetes. We used a large national sample with ten years of follow-up. We detected positive associations between openness and neuroticism and type 1 diabetes prevalence, and negative associations between neuroticism and type 2 diabetes incidence. In Chapter 3, we examine relationships between personality and type 2 diabetes incidence using aggregated personality and diabetes data on a level of the U.S. counties and states. In a six-years follow-up study, we found no evidence that mean levels of personality traits were associated with diabetes incidence in the U.S. states. In the following chapter we explore whether a possible mechanism by which personality may influence diabetes is by moderating the expression of its genetic risk. The study described in Chapter 4 looks at interactions between personality domains and facets with polygenic risk score for type 2 diabetes in predicting glycated haemoglobin levels using a large community-dwelling sample. This study found a negative phenotypic correlation between openness and glycated haemoglobin levels, though this association was confounded by cognitive ability. Moreover, genetic risk for diabetes was more strongly associated with glycated haemoglobin levels in people with lower levels of either agreeableness or conscientiousness. In Chapter 5 we move away from diabetes to discuss previously reported contradictory results regarding the effects neuroticism has on mortality. Some of the previous studies reported higher neuroticism being associated with higher risk of mortality, whereas some reported that higher neuroticism was associated with lower risk of death. We tested whether the sign of the neuroticism effect was a function of the covariates included in the models. In a national sample with ten years of follow-up we found that neuroticism was a risk factor for death in the models that did not include objective and self-rated health variables. However, when these variables were included, neuroticism was related to lower risk of death. In the last empirical chapter, Chapter 6, we explore whether autonomic nervous system activity is a biomarker for personality traits. The first study tests whether openness is associated with measures of sympathetic and parasympathetic nervous system activation. We find that openness was associated with sympathetic nervous system activity under baseline but not in the stress conditions, and that it was not associated with measures of parasympathetic activation. The second study describes a model of associations between neuroticism and autonomic nervous system activation, while controlling for cardiovascular disease and depression and their mutual associations. We found that neuroticism has independent contributions to all measures of autonomic nervous system activity, and to heart disease, even when controlling for relevant clinical variables. Thus, autonomic nervous system activity may explain in part observed links between personality, and heart disease and mortality. Finally, in Chapter 7 we summarize the findings presented in the five empirical chapters, discuss the limitations of the current method, and offer suggestions for future research in the field.