Apilot study conducted by our research group in 2019 showed that adecision made in advance about the transfer of palliative care patients documented within the emergency information was associated with the actual place of death, thus realizing the patient's wishes regarding the place of death. There were also details on how the scope, duration, and frequency of home visits by mobile palliative care teams should be designed in order to fulfill patients' wishes regarding the place of death. Systematic reviews point to adiscrepancy between preferred and actual places of death and alack of prospective studies and data to support patients and their caregivers.The current observational study lasted one year, included all mobile palliative care teams for adults working in Styria and collected data from atotal of 1425 patients (52.6% men). The median age at the time of death was 78.1 years (IQR = 17.4). 76.9% of the sample were cancer patients. Other diseases included cardiovascular (6.3%), neurological (4.1%) and respiratory diseases (2.3%).The emergency information used in this study was an optimized version of the emergency information used in the 2019 pilot, based on input from all palliative care teams who participated in 2019.Atotal of 109 patients (7.6%) of the entire sample took part in the intervention (emergency information). Of these, 85individuals (78.9% of the intervention group) indicated apreference for on-side treatment, i.e., without transfer. Only 8patients (7.3%) indicated apreference for transfer and 16 (14.7%) did not indicate apreference. Of the 85patients who did not wish to be transferred, 75 (88.2%) died at home, i.e., in around 9out of 10cases the patient's wishes were complied with. In contrast, of the 8people who requested transfer, only 3 (37.5%) died at home, i.e., the majority (5or 62.5%) died in hospital or apalliative care unit, indicating that the patient's wishes were complied with in the majority of cases.The preference expressed in the emergency information for on-site treatment and against transfer was associated with amore than five times higher odds (p < 0.001) of actually dying at home. The preference expressed in the emergency information increased the probability of dying at home from 63% in the control group to 89% in the intervention group.There was also asignificant difference according to the number of medical contacts, i.e. patients who took part in the intervention and did not wish to be transferred had twice as many contacts with physicians (p = 0.031).The results of the logistic regression model also suggested an influence of age, disease and intensity of care by physicians on the odds of dying at home: Older people and the minority of palliative care patients who did not suffer from cancer had higher odds of dying at home than those with cancer. While there was no correlation between the duration of care and the total number of care contacts, the odds of dying at home increased with the number of medical contacts (+7% per contact).
Read full abstract