Factors associated with home versus institutional death among cancer patients in Connecticut

OBJECTIVE: To assess the relationships between home death and a set of demographic, disease-related, and health-resource factors among individuals who died of cancer. DESIGN: Prospective cohort study. SETTING: All adult deaths from cancer in Connecticut during 1994. PARTICIPANTS: Six thousand eight hundred and thirteen individuals who met all of the following criteria: died of a cancer-related cause in 1994, had previously been diagnosed with cancer in Connecticut, and were age 18 and older at the time of death. MEASUREMENT: Site of death. RESULTS: Twenty-nine percent of the study sample died at home, 42% died in a hospital, 17% died in a nursing home, and 11% died in an inpatient hospice facility. Multivariate analysis indicated that demographic characteristics (being married, female, white, and residing in a higher income area), disease-related factors (type of cancer, longer survival postdiagnosis), and health-resource factors (greater availability of hospice providers, less availability of hospital beds) were associated with dying at home rather than in a hospital or inpatient hospice. CONCLUSIONS: The implications of this study for clinical practice and health planning are considerable. The findings identify groups (men, unmarried individuals, and those living in lower income areas) at higher risk for institutionalized death-groups that may be targeted for possible interventions to promote home death when home death is preferred by patients and their families. Further, the findings suggest that site of death is influenced by available health-system resources. Thus, if home death is to be supported, the relative availability of hospital beds and hospice providers may be an effective policy tool for promoting home death.