Publication Date



Open access

Embargo Period


Degree Type


Degree Name

Doctor of Philosophy (PHD)


Psychology (Arts and Sciences)

Date of Defense


First Committee Member

Maria M. Llabre

Second Committee Member

Neil Schneiderman

Third Committee Member

Michael Antoni

Fourth Committee Member

Marc Gellman

Fifth Committee Member

Armando Mendez


Chronic stress can result in long-term changes that influence susceptibility to disease, including cardiovascular disease (CVD). Perceptions of stressors such as relationship struggles, financial burden, and job strain, differ based on availability of both tangible and intangible resources, such as level of education, emotional support, and health insurance coverage. The availability of these resources, which comprise an individual’s psychosocial reserve capacity, may moderate the relationship between chronic stress burden and cardiovascular health. Since chronic stress has been inconsistently operationalized in literature, the current study examined a latent variable model of chronic stress in relation to components of the metabolic syndrome (MetS) (waist circumference, HDL cholesterol, triglycerides, fasting glucose, systolic and diastolic blood pressure) and CVD. We then determined the extent to which certain resources in the psychosocial reserve capacity moderate the relationship between chronic stress and cardiovascular health. Exploratory analyses examined the moderating role of sex on health outcomes and the relationship between chronic stress and depressive symptoms. Analyses were conducted on previously collected data from 6685 White, Black, Hispanic, and Chinese American men and women between the ages of 45 - 84 enrolled in 2000 - 2002 in the Multi-Ethnic Study of Atherosclerosis. CVD outcome was collected up until nine years post-baseline. A latent variable model, comprised of four indicators of chronic stress (namely, money/financial, job/ability to work, relationship, and health problem (other) stress), fit the data (CFI = .988; RMSEA = .019 90% CI (.004 - .031); SRMR = .020). Additionally, measurement invariance was achieved across the four races/ethnicities, demonstrating that comparisons can be made on the same construct across groups. Results showed that latent chronic stress was positively associated with waist circumference (B = 1.00, 95% CI: 0.30 – 1.69, p < .01), and fasting glucose (B = 2.16, 95% CI: 0.75 – 3.56, p <.01) but not significantly related to HDL cholesterol, triglycerides, systolic, or diastolic blood pressure. Level of education moderated the relationship between chronic stress and HDL cholesterol, such that for less educated individuals, as chronic stress increases, HDL also increases by 1.62 mg/dL. Emotional support and health insurance coverage did not moderate the relationship between chronic stress and any MetS component. Surprisingly, chronic stress was not significantly related to CVD in any race/ethnicity, and none of the predicted moderators moderated the relationship between chronic stress and CVD. Exploratory analyses examined the moderating role of sex on MetS components and CVD. Sex moderated the relationship between chronic stress and waist circumference (B = -1.70, 95% CI: -2.88 - -0.53, p <.01). In both women and men, stress was positively related to waist circumference. However, the relationship was stronger in women than in men. Sex was also a significant moderator of the relationship between chronic stress and CVD. For three races/ethnicities, men had higher odds of CVD than women. Sex did not significantly moderate the relationship between stress and CVD in Chinese Americans. Additional exploratory analyses showed that chronic stress was positively related to depressive symptoms (B = 6.45, 95% CI: 5.61 - 7.29, p < .001). Higher levels of education and emotional support buffered the negative effects of chronic stress on depressive symptoms. For both levels of education, chronic stress was associated with an increase in depressive symptoms; however, this relationship was stronger in less educated individuals. Chronic stress was associated with more depressive symptoms at all levels of emotional support. However, this relationship was strongest for individuals at the lowest level of emotional support. Sex significantly moderated the relationship between chronic stress and depressive symptoms. Women started out with more depressive symptoms, than men; however, each additional increase in chronic stress was associated with a greater increase in depressive symptoms in men relative to women (p < .001). These findings enhance our understanding of the concept of chronic stress in a multi-ethnic sample. Results from this study suggests that future research examining the role of chronic stress in multiple populations could utilize a latent variable over a sum score or assigning individuals to groups of stress (high, medium, low), since all indicators loaded reliably onto a latent factor in an invariant manner across races/ethnicities. Future studies should test a second-order or bi-factor latent model to enhance the definition of latent chronic stress by including more information on what contributes to the individual sources of stress. Additional research is needed to identify moderators in the chronic stress-cardiovascular health pathway.


chronic stress; metabolic health; metabolic syndrome; cardiovascular disease; race/ethnicity