To investigate factors affecting the place of death of patients receiving home care services, we performed a retrospective cohort study using 252 elderly Japanese patients. During a 3-year period, 40 patients died at home (78.4 ± 12 years olds), and 59 patients died at hospitals (77.6 ± 13.4 years olds). Patient profiles, including their demographic characteristics, comorbidities, as well as formal and informal care levels were evaluated. Patients who died at home received a shorter term of home care, suffered malignancies more often and received visiting nurse services more often than those at hospitals. Age, gender, comorbidities, laboratory data, independency of activity of daily living (ADL), number of family looking after patients, number and dose of prescriptions and number of medical treatments such as decubitus were not different between these groups. Multivariate logistic regression analyses revealed that patients who died at home had an increased likelihood of suffering from malignancy (odds ratio = OR = 2.18, HR: 1.04-3.98, p = 0.049) and an increased likelihood of receiving visiting nurse services (OR = 3.13, HR: 1.08-6.21, p = 0.029) compared to those who died at hospitals. Thus, dying at home may be associated with cases of malignancy compared to other diseases in Japan, and it may be associated with a greater need for home visiting nurses. In conclusion, the nature of the patient's disease and the presence of visiting nurses influenced the decision regarding the patient's place of death. More study is necessary to better understand the end stages of death at home.
Copyright © 2010 Elsevier Ireland Ltd. All rights reserved.