Impact of a palliative care service on in-hospital mortality in a comprehensive cancer center

Background: Palliative care services provide symptom control and psychosocial support for dying patients and their families. These services are not available in many cancer centers and tertiary hospitals. The purpose of this study was to review the impact of a palliative care program, established in 1999, on overall in-hospital mortality. Methods: We reviewed the M. D. Anderson Cancer Center computerized database to determine the total number of deaths and discharges and the place of death for each fiscal year from 1999 to 2004. The median length of stay for patients who died in different locations within the hospital was calculated. Annual palliative care consultations for patients who subsequently died in the hospital were retrieved. The annual mortality rate for the cancer center was calculated. Results: The overall in-hospital mortality rates were 3.6, 3.7, 3.6, 3.5, 3.6, and 3.7% of all discharges for the period 1999-2004 respectively (p > 0.2). The number of deaths in the medical intensive care unit (MICU) dropped from 252 in 671 (38%) in 1999 to 213 in 764 (28%) in 2004 (p < 0.0001). Involvement of the palliative care service in the care of patients dying in the hospital grew from 8 in 583 (1%) in 1999 to 264 in 764 (35%) in 2004 (p < 0.0001). The median length of hospital stay (MLOS) for patients who subsequently died in-hospital was significantly longer than that for patients who were discharged alive. Conclusions: Increased involvement by the palliative care service in the care of decedent patients was associated with a decreased MICU mortality and no change in overall hospital mortality rate or inpatient length of hospital stay.