• Sakiko Fukui, Hiromi Kawagoe, Sakai Masako, Nishikido Noriko, Nagae Hiroko, and Miyazaki Toshie.
• Department of Community Health Nursing, School of Nursing, Tokyo Metropolitan University of Health Sciences, Tokyo, Japan. sfukui@post.metro-hs.ac.jp
• Palliat Med. 2003 Jul 1;17(5):445-53.

AbstractAlthough the place of death of patients with terminal cancer is influenced by multiple factors, few studies have systematically investigated its determinants. The purpose of this study was to examine the influence of the patients' sociodemographic, clinical and support network variables on the place of death of terminally ill cancer patients under the care of home care agencies in Japan. Among 528 patients from 259 home care agencies, 342 (65%) died at home and 186 (35%) died at a hospital. From the multivariate logistic regression model, patients who expressed the desire for receiving home care at referral [odds ratio (OR), 95% confidence interval (CI): 2.19, 1.09-4.40] in addition to the family caregiver's desire for the same (OR, 95%CI: 3.19, 1.75-5.81), who had more than one family caregiver (OR, 95%CI: 2.28, 1.05-4.94), who had the support of their family physician (OR, 95%CI: 2.23, 1.21-4.08), who were never rehospitalized (OR, 95%CI: 0.04, 0.02-0.07), who received more home visits by the home hospice nurse during the stable phase under home hospice care (OR, 95%CI: 1.25, 1.02-1.53), and who were in the greatest functionally dependent status during the last week prior to death (OR, 95%CI: 8.60, 4.97-14.89) were more likely to die at home. Overall, this model could accurately classify 95% of the places of death, which is higher than other published studies. A clearer understanding of factors that might influence the place of death of terminally ill cancer patients would allow healthcare professionals to modify healthcare systems and tailor effective interventions to help patients die at their place of preference.

26 keywords...

hide…