Publication Date



Open access

Embargo Period


Degree Type


Degree Name

Doctor of Philosophy (PHD)


Psychology (Arts and Sciences)

Date of Defense


First Committee Member

Youngmee Kim

Second Committee Member

Michael H. Antoni

Third Committee Member

Armando Mendez

Fourth Committee Member

Maria M. Llabre

Fifth Committee Member

Neil Schneiderman


Both cancer patients and their informal family caregivers develop chronic diseases earlier and more frequently than those who have not been affected by cancer. Elevated depressive symptoms, which are easily measureable and modifiable, have been linked to inflammation, dysregulated HPA axis functioning, and physical health decline in both cancer patients and healthy persons. Further, studies have shown patients’ and caregivers’ depressive symptoms to be correlated, and that patients’ distress relates to their caregivers’ poorer health, and vice versa. While it is known that one’s own depressive symptoms are a risk factor for one’s own inflammation and HPA axis dysregulation, and patients’ depressive symptoms are a risk factor for caregivers’ depressive symptoms (and vice versa), yet unknown is to what extent patients’ depressive symptoms may serve as a risk factor for their caregivers’ inflammation and HPA axis dysregulation, and vice versa. In this project, a dyadic biopsychosocial model accounting for interdependence between patients’ and caregivers’ psychological and physiological health was proposed to fill these gaps. Data were analyzed from 84 cancer patients and 86 caregivers (81 dyads) who participated in this study around three months following the patients’ cancer diagnosis. Participants reported depressive symptoms using the Center for Epidemiological Studies-Depression (CES-D) and provided blood and saliva samples for stress biomarkers. Blood samples were analyzed to measure levels of two pro-inflammatory markers: interleukin (IL)-6 and C-reactive protein (CRP). Three saliva samples per day were taken on two consecutive days for salivary cortisol slope values. Actor-Partner Interdependence Modeling (APIM) using Structural Equation Modeling (SEM) was used to test study hypotheses. It was hypothesized that cancer patients and their caregivers would show positively correlated depressive symptoms and stress biomarkers, which was supported for cortisol slope values only (Unstandardized estimate = 0.001, 95% CI [0 – 0.002], p = .03). It was further hypothesized that patients’ and caregivers’ depressive symptoms would be positively associated with both their own and their partners’ stress biomarkers, after controlling for covariates of age, BMI, and sex (and patients’ cancer treatment status and stage for patients only). These hypotheses were unsupported, but ability to test these hypotheses was limited by low power. Exploratory hypotheses posited that somatic depressive symptoms would be more strongly associated with own stress biomarkers, but these hypotheses were not supported. This study remains an important first test of the dyadic biopsychosocial model of psychological and physiological health of cancer patients and their caregivers. Interdependence among patients’ and caregivers’ physiology underscores the importance of not only studying patients and caregivers as a unit, but treating these partners as a unit as well, to ensure optimal health outcomes among both patients and their families.


family caregiver; cancer; psychoneuroimmunology; cortisol; Actor Partner Interdependence Modeling; dyadic analysis