Unscheduled hospitalization in advanced cancer patients: Time to rethink about the illness.

Abstract

e23007 Background: The aim of our study was to determine if an unscheduled hospitalization in advanced cancer patients is a good time for palliative care referrals, and if this indicator could be improved with some other prognostic score. Methods: We conducted a retrospective study, which included all patients with unscheduled hospitalization in the medical oncology service from January 2011 to December 2013. In order to improve the accuracy of the analysis we considered only first admissions of each patient. We excluded patients with scheduled and unindicated hospitalization. The main variable was survival after an unscheduled hospitalization. We collected the following variables: demographic (age, sex, date of birth, date of death and place of death), tumor type, clinical variables (reason for admission, date of hospitalization, assessment by the palliative care unit during hospitalization, date of last chemotherapy and ECOG), analytical variables (albumin, lymphocytes and lactate dehydrogenase) and a prognostic nomogram developed by our team. Results: Inclusion criteria were met by 1,180 patients. Table summarizes the main sociodemographic characteristics, the type of tumor, and the functional status of patients according to the ECOG scale. After two years of follow-up, 816 (69%) patients had died. The median overall survival was 188 days (161-214). Patients with breast and gynecological cancer had the longest survival, while patients with melanoma, gastric cancer, lung cancer, and pancreatic cancer had the poorest survival. 676 patients (58%) were readmitted during the following year. Readmissions meant a significant decrease of survival. The stage of the oncological disease and performance status measured by ECOG had a clear relationship with survival time. Prognostic nomogram showed high accuracy to predict the probability of survival. Median survival for different quartile was respectively: 348, 139, 68 and 31 days (p < 0´000). Conclusions: The median survival of patients after an unscheduled hospitalization was around 6 months. Been admitted seems a good indicator to introduce specific palliative care. The use of other prognostic scores, as a prognostic nomogram, could help to select patient’ groups with poorer survivals. [Table: see text]