Previous study reported that patients of institutions with religious background are more likely to achieve ‘good death’ than patients in non-religious institutions, although the reasons are unclear. This study aims to examine the reasons for this difference using a national survey of religious and non-religious institutions. We sent a self-administered questionnaire to 10,715 bereaved family caregiver and 133 hospice and palliative care units which cared for their family members prior to death. The questionnaire for bereaved families included the Good Death Inventory (GDI). The questionnaire for institutions included their religious affiliation and their subjective and objective provided care. In total, 7,286 bereaved families (68%) and 127 (95%) institutions including 23 religious end-of-life care institutions responded. We performed t-test and χ-square test to compare the GDI scores and provided care for religious and non-religious end-of-life care institutions. The GDI total score was significantly higher (p=0.01) in religious institutions compared to non-religious institutions. Items such as “providing many activities for seasonal events to the patients”, “providing more bereavement care”, “having facilities to practice their religion”, and “having regular visits from religious leaders”, were all reported more frequently in the religious institutions, and significantly associated with achievement of ‘good death’ (p<0.05 for all items). This study suggests that factors for achieving ‘good death’ are not only limited to providing religious care but also providing other forms of care, services, and activities that can enhance the patients’ ‘good death’.