Background: The aim of this study was to clarify the decadal trends in the structure and usage status of palliative care units in Japan and the association with length of stay. Methods: We conducted a secondary analysis of data from nationwide surveys of palliative care units conducted by Hospice Palliative Care Japan from 2002 to 2012. Length of stay was divided into three categories (within 30 days, 31 to 60 days, and 61 days or more). Results: Significant decadal trends in usage status were observed: the mean annual number of admitted patients (119±55 persons in 2001, 163.0±77.7 in 2011, p＜0.001), and deaths (99±44, 136±58, p＜0.001) increased and the mean length of stay in a palliative care unit (8±15 days, 39±15, p＜0.001) decreased. Palliative care units with a shorter mean length of stay had significantly more admitted patients, and more discharged patients, and a significantly lower mean ratio of death to discharge, and a lower bed availability rate. In contrast, length of stay was not significantly associated with the structure of palliative care units, including the type of units, number of inpatient beds, percentage of private rooms, and number of staff. Conclusion: The features of decadal trends in the structure and usage status of palliative care units in Japan were a shortening of length of stay, and an increase in the number of patients. Palliative care units with a shorter mean length of stay cared for a larger number of terminally ill patients.

Purpose: The aim of this study was to clarify the factors contributing to the evaluation of perceived care in inpatient palliative care services from the aspect of informal caregiver after their loss of the patient to cancer. Method: A questionnaire was mailed to 9,684 bereaved subjects who had lost family members at one of 103 palliative care units in Japan to evaluate their sociodemographic characteristics and the evaluation of perceived care. An institution survey was performed to collect organization-related variables. The evaluation of perceived care was rated by the Care Evalutaion Scale (CES), the Good Death Inventory (GDI) and single item of overall satisfation. Result: A total of 5,810 responses were analyzed (response rate＝60％). Uni-variate and multivariate analysis was performed to clarify the determinants of each scale. Significant determinants of the score in the evaluation of care identified were: the rate of private room (100％), independent facility, palliative care physician being night-time duty, the number of nurses at night (＞0.1 per bed), sending a letter to every bereaved family, holding a memorial services for every bereaved family and having a religious background. Conclusion: In conclusion, the evaluation of palliative care from the aspect of informal caregiver was influenced by various organization-related variables.