Background: Caregiver satisfaction with hospice care is a quality measure for hospice organizations and predicts caregiver and patient health, mortality and resource use. Yet contributing factors to caregiver satisfaction are poorly understood and there are no tested models of caregiver satisfaction in the literature to date. To fill this gap, our aim was to develop and test a model of caregiver satisfaction with hospice care in the cancer population. Methods: This was a retrospective cohort study of national data collected by the National Hospice and Palliative Care Organization (NHPCO) using the Family Evaluation of Hospice Care (FEHC) survey in 2011. We utilized a random sample of 1000 cancer caregivers (70,782 available). An initial theoretical model was developed which posited a relationship between demand (as measured by reported patient symptoms) and satisfaction, modified by caregiver internal and external resources. Structural equation modeling was used to test the model and adjustments made to improve model fit. Results: The final model fit the data well (LR=189.08, df=78; RMSEA=0.04; CFI=0.91; SRMR=0.03). Caregiver satisfaction was predicted by demand – higher perception of demand (more patient symptoms) reduces caregiver satisfaction. Patient age, caregiver race, education, relationship to the patient and length of hospice stay modify this relationship directly – black, non-spousal and more highly educated caregivers are less satisfied, along with caregivers of younger patients and those with shorter lengths of stay. Patient and caregiver age, caregiver sex and length of stay influence this relationship by altering caregiver perception of demand. Younger caregiver and patient age, being a female caregiver and longer length of stay increases caregiver demand, decreasing satisfaction. Note that length of stay works both for and against satisfaction. Conclusions: Caregiver satisfaction is a complex perception that depends on caregiver and patient characteristics, as well as hospice care received. Hospice care providers should consider screening caregiver-patient dyads for “high-risk” characteristics and providing additional supportive measures.