ABSTRACT & COMMENTARY
After-Hours ICU Discharge: A Potentially Modifiable Cause of Increased Hospital Mortality
By David J. Pierson, MD, Editor
SYNOPSIS: Examination of outcomes in 710,535 patients in relation to the timing of ICU discharge showed that being moved out to the ward between 1800 and 0600 hours was associated with increased risks of both in-hospital death and unplanned ICU readmission.
Gantner D, et al. Mortality related to after-hours discharge from intensive care in Australia and New Zealand, 2005-2012. Intensive Care Med 2014;40:1528-1535.
The Australian and New Zealand Intensive Care Society Adult Patient Database prospectively records data on patients cared for in 90% of Australian and 50% of New Zealand ICUs. Gantner and associates used this database to examine whether discharge from the ICU to the general ward after regular hours remained associated with increased in-hospital mortality and ICU readmission, as had been the case before various care and organizational improvements were implemented.
Over the 8-year study period, of 710,535 ICU patients who met study criteria and were discharged alive from the unit after an initial ICU stay, 28,507 (4.0%) subsequently died in the hospital. Approximately 85% of the patients left the unit between 0600 and 1800 hours and 15% were transferred out to the ward after hours. Patients transferred after hours were sicker (mean APACHE III score 50 vs 46, P < 0.001) and had higher predicted mortality (5.3% vs 3.4%, P < 0.001) and higher actual in-hospital mortality (6.4% vs 3.6%, P < 0.001) than their regular-hours counterparts. When the data were examined according to the actual hour of ICU discharge, the highest mortality risk occurred with those who left between 0200 and 0600 hours (odds ratio, 1.78-2.49) and the lowest between 0900 and 1100. Patients who left the ICU between 1800 and 0600 hours were also significantly more likely to be readmitted to the unit during their hospital stay (5.1% vs 4.5%, P < 0.001).
There were no detectable changes in the timing of ICU discharge during the 8 years of the study, and crude mortality rates were similarly unchanged. Post-ICU mortality rates for all patients decreased during the study period, but the risk associated with after-hours discharge remained elevated throughout (odds ratio, 1.34; 95% confidence interval, 1.30-1.38). The findings with respect to increased post-transfer mortality and unplanned readmission did not differ between surgical and medical patients.
COMMENTARY
Multiple studies from various parts of the world have consistently shown that after-hours discharge from the ICU is an independent risk factor for adverse patient outcomes, including increased mortality and unplanned ICU readmission. The proportion of ICU patients discharged after regular hours has been established as a national indicator of the quality of clinical care in Australia. Wide implementation of rapid response systems and numerous other organizational and clinical practice changes have taken place in part to address this problem. However, this carefully done study using a high-quality database that reflects patient outcomes nationwide documents the continued high rate of after-hours discharge and its strong association with adverse patient outcomes. This is the largest study of ICU discharge timing yet reported, and while comparable data from the United States are not available there is little reason to assume that the situation is different here.
Why should after-hours transfer from the ICU to the ward be a risk factor for death? It has been asserted that in many institutions whose busy ICUs tend to be full much of the time, the arrival of new, critically ill admissions may prompt the transfer out of patients who would not otherwise be discharged but are "least sick" at the moment. Over and above this possibility, Gantner et al suggest that in ICUs without 24/7 intensivist presence, the decision to transfer patients out during the evening and night shifts may be made by less experienced clinicians who are less able to assess clinical stability. Patient handovers may also be less complete during such times, and staffing levels on the wards are lower than in the ICU.
For this retrospective study, the authors did not have information on the reasons patients were transferred out of the ICU after hours. How full the units were at those times and whether the goals of care might have changed for the patients cannot be determined. Nonetheless, one must conclude that the factors leading to premature ICU discharge after hours have not been completely addressed in the institutions whose patients are included in this study. This example of incomplete transfer of evidence into practice in critical care clearly needs more attention if critically ill patients are to receive the full benefit of our knowledge.