Abstract 5977
Background
Acute palliative care units (APCUs) admit patients with cancer for symptom control, transition to community palliative care unit/hospice or end-of-life care. Prognostication is crucial for decision-making. We evaluated factors associated with patients’ length of stay (LOS) on an APCU in a cancer centre.
Methods
We analyzed demographic, administrative and clinical data for patients admitted to the APCU in 2015. Clinical data included cancer diagnosis, palliative performance scale (PPS) on admission, delirium screening using the short Confusion Assessment Method (CAM), and Edmonton Symptom Assessment System (ESAS) symptoms. ESAS distress score (EDS; sum of all 9 symptoms) and FDSA sub-score (fatigue, drowsiness, shortness of breath, appetite) were calculated. We conducted univariable (UVA) and multivariable (MVA) regression analyses of factors associated with LOS of patients who died on the APCU and of those who were discharged.
Results
Among 280 patients, 156 (56%) died on the unit and 124 (44%) were discharged. Median LOS was 14 days for discharged patients and 8 days for those who died (p < 0.001). Discharged patients were older (median age 68 vs 64, p = 0.003) and had higher functional status (median PPS 50 vs 40, p < 0.001) than those who died. Patients who died had higher symptom burden (median EDS 44 vs 38, p < 0.009), were more likely to be admitted from an inpatient unit (p < 0.001) and for terminal care (p < 0.001), and were more likely to develop delirium (p = 0.04). On MVA of patients who died on the APCU, reason for admission (p = 0.007), delirium (p = 0.02) and FDSA score (p = 0.002) were associated with LOS. Shorter LOS was associated with admission for terminal care (p = 0.05) and missing FDSA (patients were too ill to complete) (p < 0.001); longer LOS was associated with delirium (p = 0.02). For patients who were discharged from the APCU, delirium was associated with longer LOS (p = 0.02).
Conclusions
In cancer patients admitted to an APCU, development of delirium was associated with longer LOS in patients who died on the unit as well as in patients who were discharged home or to PCU/hospice.
Clinical trial identification
Legal entity responsible for the study
Camilla Zimmermann.
Funding
Canadian Institutes of Health Research.
Editorial Acknowledgement
Disclosure
All authors have declared no conflicts of interest.