Delay in Transfer to the ICU Leads to Poorer Clinical Outcomes
Abstract & Commentary
Synopsis: A time lapse of > 4 hours in ICU transfer after the development of 1 or more physiologic threshold criteria was associated with greater mortality, longer hospital length of stay, and higher costs.
Source: Young MP, et al. Inpatient transfers to the intensive care unit. Delays are associated with increased mortality and morbidity. J Gen Intern Med. 2003;18(2):77-83.
For hospitalized patients, the timing of transfer to the ICU may be an important determinant of outcome. To determine whether delayed transfer after the development of physiologic deterioration was associated with adverse outcomes, Young and colleagues retrospectively examined the medical records of 91 consecutive noncardiac patients who were transferred to the ICU from medical and surgical units in a community hospital. Using lab values and physician and nurse notes, all cases were identified wherein the patient met any 1 of 11 physiologic threshold criteria. The criteria were selected based on their inclusion in APACHE II or SAPS and clinical judgment. Examples included: respiratory rate > 35 breaths/min for > 30 minutes; pH < 7.25 or PaCO2 > 60 mm Hg; SaO2 < 95% on non-rebreathing oxygen mask for > 30 minutes; systolic blood pressure < 85 mm Hg for 30 minutes; and heart rate < 40 or > 140 beats/min. Transfer to the ICU was classified as "slow" if more than 4 hours elapsed between the time the patient first met any criterion and ICU admission. Young et al also recorded the time the patient’s physician was first notified, time of the first documented bedside visit, and APACHE II scores on the ward and at ICU admission.
At the time the first criterion was met, slow and rapid transfer patients were similar in regard to age, gender, diagnosis, hospital days, and severity of illness. At ICU admission, slow-transfer patients had higher APACHE II scores (21.7 vs 16.2; P = .002) and were more likely to die in the hospital (41% vs 11%; relative risk, 3.5; 95% CI, 1.4-9.5). Median hospital length of stay was longer (14 vs 9 days; P = .03) and hospital costs were greater ($34,000 vs $21,000; P = .01) for slow transfers. These differences remained after adjusting for potential confounders. Slow-transfer patients were less likely to have had their physician notified of their deterioration within 2 hours of meeting criteria (59% vs 31%; P = .001) or to have had a bedside evaluation within 3 hours after meeting criteria (23% vs 83%; P = .001).
Comment by Leslie A. Hoffman, PhD, RN
Findings of this study provide an interesting perspective on the hazards of delaying prompt assessment and transfer to the ICU when patients experience physiologic deterioration. After patients met 1 or more defined physiologic threshold criteria, a delay in transfer to the ICU resulted in a nearly 5-fold higher adjusted risk of death, as compared to outcomes when patients were transferred earlier. Given trends toward early discharge, patients who require an inpatient stay are likely to be older and have complex problems. While physiologic deterioration can occur suddenly, the more likely scenario is a gradual downward spiral preceded by subtle and then more obvious signs and symptoms. Timely evaluation and titration of therapy—such as prompt diagnostic testing, receipt of sufficient intravenous fluids and antibiotics, and provision of adequate respiratory support—may avert further problems. While it may be possible to provide this support on the clinical unit, resources may be insufficient to provide the level of care required.
In this study, slow-transfer patients were less likely to have their physician promptly notified or to have a prompt bedside evaluation. This finding has potential implications in regard to ability of the nursing staff to recognize and appropriately identify the need to intervene, physician availability, patterns of communication between physicians and nurses, and unit staffing. Given the current nursing shortage and lack of experienced nursing staff on some units, it is possible that staff nurses may not have promptly recognized and communicated changes in a timely manner. It is also possible that physicians delayed visiting the patient due to other duties or misinterpretation of the seriousness of events.
Several strategies might be used to overcome this problem. Some practices use a hospitalist to manage inpatient care and ensure quick response. Clinical nurse specialists and nurse educators are needed to provide needed education and periodic reinforcement. Nurse practitioners and physician assistants can assist by providing more prompt initial evaluation. Many of the physiologic markers used in the study are recorded in the electronic medical record and could be used to generate computerized alerts.
To more fully evaluate the physiologic threshold criteria used in this study, Young et al carried out a 2-month prospective analysis of their sensitivity and specificity for predicting ICU transfer. The criteria had a high sensitivity (88%). However, they lacked specificity (13%) and had a low positive predictive value (8%). While disappointing, they still retain the ability to alert their physician to an acute change in status and prompt immediate evaluation. Further study is needed to identify better predictors. Although problems resulting from delays in transport for acute myocardial infarction and stroke have received substantial attention, little attention has focused on how to ensure that hospitalized patients are promptly evaluated and transferred to the ICU when their condition merits this.
Dr. Hoffman is Professor Medical-Surgical Nursing Chair, Department of Acute/Tertiary Care University of Pittsburgh School of Nursing.