Admission to intensive care unit at the end-of-life: is it an informed decision?

UNLABELLED: There is a major deficiency in the end-of-life care offered to patients dying in the intensive care unit (ICU). HYPOTHESIS: Hospitalized dying patients had informed discussions on end-of-life and palliative care options before admission to ICU. PATIENTS AND METHODS: A descriptive non-interventional study was performed at a teaching hospital to examine if patients who died in hospital had informed discussions on end-of-life care before admission to ICU. The impact of these discussions on subsequent patient care: aggressive therapy in the ICU, the quality of palliation, use of hospice care services and utilization of hospital resources were examined. Data were collected from medical records for all hospital deaths over 24 months.
RESULTS: Of 252 hospital deaths, 196 (78%) were treated and subsequently 165 (65%) died in the ICU. Patients treated either in the ICU or general hospital wards had similar frequency of ultimately or rapidly fatal pre-existing disease (47% versus 62%, P: ns) and readmission to hospital within one year before death (43% versus 57%, P, ns). The median age (10-90% percentile) was slightly younger for the ICU than hospital wards patients: 73 (45-85) versus 76 (55-91) years, P < 0.01. Of the 156 patients who were transferred to ICU from hospital wards: 136 (87%) were managed by house staff on teaching services and 20 (13%) were managed by attending staff hospitalists, P < 0.01. None of those transferred to the ICU who subsequently died had discussion of palliation or end-of-life care as an alternative treatment. Of those who died who were treated on general wards, 14 (25%) patients had discussion of palliation as an alternative treatment option before death. Do-not-resuscitate decisions were made in 48% of cases two days before death. Patients who were treated in the ICU had more invasive tests performed on them and were less likely to have adequate pain control or referral to hospice care services than on a general ward. Median hospital charge was much higher for patients who received ICU versus general ward care (33,252 dollars versus 8549 dollars, P < 0.001).
CONCLUSIONS: Patients who died in the ICU did not have informed discussions of end-of-life or palliative care as an alternative treatment option before admission. The quality of end-of-life care was disrupted for patients with fatal pre-existing chronic disease who were admitted to the ICU before death. Lack of clinical experience, knowledge and competency with end-of-life care influenced admission of patients to ICU regardless of poor prognosis. Decisions regarding the pursuit of aggressive therapy versus palliative care must be addressed with patients by physicians who are competent and experienced in end-of-life care as this will have a profound impact on both the quality of care delivered and effective use of limited hospital resources.