Cardiovascular disease (CVD, including heart disease and stroke) is the leading cause of death in the United States and worldwide, making CVD a key priority in epidemiological and public health research. The utility of current exposure and risk assessment models utilizing traditional risk factors such as sociodemographic characteristics and health behaviors in CVD prevention efforts has plateaued, with a large proportion of excess CVD morbidity and mortality risk remaining unexplained. In recent years, psychosocial stressors have gained prominence as important drivers of CVD etiology, with stressful exposures such as adverse childhood experiences (ACEs), social isolation, and job strain emerging as novel CVD risk factors. The objective of this project was to conduct a systematic epidemiological investigation of ACEs, social isolation, job strain, cardiometabolic biomarkers implicated in allostatic load, and their associations with risk of CVD mortality using data from three large, population-based, nationally representative cohort studies characterizing the health and disease profiles of childhood, mid-life, and old age in the United States. This project implemented a secondary data analysis of the based National Longitudinal Study of Adolescent to Adult Health (Add Health), Mid-life in the United States (MIDUS), and the Health and Retirement Study (HRS) cohort studies, examining the contribution of psychosocial stressors at different life stages (childhood and adulthood), in different life domains (working life and non-working life) to the development and progression of CVD.
The primary aim of the study was to examine the associations of ACEs, social isolation, and job strain with CVD mortality risk in the Add Health, MIDUS, and HRS cohorts, implementing a comprehensive exposure assessment of work and nonwork related psychosocial factors across the life-course. The secondary aim of the study was to identify psycho-physiological pathways reflecting allostatic load, linking ACEs, social isolation, and job strain with CVD, offering insight into potential mechanisms underlying associations of psychosocial exposures with CVD etiology and pathogenesis.
The sample population was large and included workers from diverse occupational categories and broad sociodemographic strata, with detailed data on characteristics such as household income, educational attainment, smoking, alcohol consumption, and physical exercise. ACE measures across the three cohorts included financial stress, household dysfunction, and abuse, with the Add Health dataset providing prospective data from childhood surveys. Job strain measures were based on Karasek’s well-validated demand-control model of job strain, while the extensively used Berkman-Syme Social Network Index provided an assessment of social isolation. The national cohorts also implemented a rigorous biomarker data collection process and allowed for the analysis of seven cardiometabolic and inflammatory biomarkers related to CVD pathogenesis, enabling the construction of an allostatic load index as an indicator of general “wear and tear” and chronic stress on the body. Furthermore, the MIDUS dataset provided autonomic measures allowing for analyses of heart-rate variability, an index of overall cardiovascular functioning and reactivity. Multivariate Cox proportional hazard regression models were used to conduct survival analyses of associations of ACEs, social isolation, and job strain with CVD mortality risk, and multivariate Poisson regression models were implemented to estimate associations of ACES, social isolation, and job strain with allostatic load index. Statistical models included iterative adjustment for sociodemographic characteristics and lifestyle behaviors, and further stratified analyses by ACE exposure level were performed.
The results of the study indicated significantly increased risk of CVD mortality among participants with a high level of ACEs in the HRS study, and among participants with higher levels of social isolation and job strain in the Add Health and HRS studies. While findings for associations of psychosocial exposures with allostatic load and heart-rate variability were not statistically significant, a stable and consistent trend of increased effect size estimates with increasing exposures was observed in the MIDUS cohort.
This project conducted a comprehensive exposure assessment of work and nonwork related factors and their associations with CVD mortality and allostatic load biomarkers clinically implicated in CVD pathology. This novel and innovative approach to exposure modelling addressed the limitations of prior work in the field of occupational and cardiovascular epidemiology by highlighting the impact of cumulative exposures. The successful implementation of the analytic strategy and the detection of promising preliminary findings for associations of ACEs and APDs with CVD mortality and allostatic load offer an empirical foundation for future investigational and translational work targeting psychosocial exposures.
Collectively, these findings suggest that high exposure to ACEs and APDs are closely related to allostatic load and multimorbid disease conditions in adulthood, as well as increased risk of CVD mortality. While the findings of the present study were not always statistically significant across the three cohorts of the Add Health, MIDUS, and HRS studies, the general trend of results indicate increased risk of CVD mortality and potentially increased allostatic load due to high exposure to ACEs and APDs. In context of the rapidly expanding literature on ACEs, social isolation, job strain, and allostatic load and HRV, these results suggest cumulative psychosocial exposures may increase the risk of chronic cardiometabolic and inflammatory risk conditions implicated in CVD pathology and mortality.
