Understanding the biological pathways in the associations between social position and mental health: an examination of allostatic load

Introduction Individuals from disadvantaged social positions experience poorer mental health than their peers from advantaged social positions. One biological mechanism by which these associations occur is theorised to be through psychosocial or ‘stress’ mediated processes. Allostatic load is hypothesised as a biological consequence of chronic stress composed of ‘primary’ biomarkers of stress and ‘secondary’ outcomes. Aims This thesis examines the social and biological factors underpinning inequality in mental health. Methods This study uses logistic and linear regression models and data from the English Longitudinal Study of Ageing (ELSA), the Understanding Society, the UK Household Longitudinal Study (UKHLS), the 1958 National Child Development Study (NCDS) and the 1970 British Cohort Study (BCS70) to examine the association between 1) social position indicators education and occupation and mental health; 2) allostatic load and mental health and 3) social position indicators and allostatic load. Given the inconsistencies in operationalising allostatic load identified in the literature review, the thesis used data from the above datasets and the MRC National Survey of Health and Development (NSHD), the Hertfordshire Cohort Study (HCS) and Avon Longitudinal Study of Parents and Children (ALSPAC) to understand the composition of allostatic load using factor analysis. The thesis investigated first whether variables representing each biological system affect the allostatic load factor structure; second, if including primary mediators will change how biomarker variables load onto factors; and finally, if the factor structures were similar across the seven datasets. Results Results suggest that 1) higher occupational class and higher educational attainment were linked with decreased odds of having poor mental health; however, only the link with occupational class persists when both variables are investigated concurrently. 2) Overall, five components of allostatic load, including metabolic, inflammatory, glucose metabolism, cardiovascular and neuroendocrine factors, were associated with mental health, although the direction of association varied in some studies. For example, in NCDS, each additional metabolic factor score decreased the odds of having poor mental health. However, some components of allostatic load, such as lipid and iron factors, were not associated with mental health. 3) Different variables representing specific biological systems did not affect the allostatic load factor structure. Including primary mediators did not change how biomarkers load onto factors; factor structures were similar across the seven datasets. Disadvantaged social position was associated with greater allostatic load, for example, neuroendocrine and inflammatory factors, but not all components were consistently patterned across all studies examined. Study contributions These findings support the literature on social position and mental health and provide better insights into how education is linked with mental health. The study suggests that lipid and iron factors were not associated with mental health and are unlikely to underpin any social differences in health. Conclusions In conclusion, higher occupational class and educational attainment are linked to better mental health. However, the association of educational attainment and mental health is working via occupational class. Individuals from disadvantaged social positions experience poor allostatic load, in particular neuroendocrine and inflammatory components. They are at higher risk of having poor mental health than their peers from advantaged social positions. This study suggests that the lipid and iron biomarker components of allostatic load do not contribute to inequalities in mental health.