ICU Transfer Delay Predicts Death, Prolongs Hospital Stay

Miriam E. Tucker

April 01, 2015

NATIONAL HARBOR, Maryland — For patients who become critically ill during hospitalization, the longer the delay in transfer to the intensive care unit (ICU), the longer the hospital stay and the greater the risk for mortality, a new study shows.

"Early recognition is the first step toward early intervention, which has been shown to improve patient outcomes for many conditions, including sepsis, myocardial infarction, and stroke," said Blair Wendlandt, MD, chief resident in the internal medicine residency program at the University of Chicago.

The electronic Cardiac Arrest Risk Triage (eCART) algorithm — which takes into account demographic, vital sign, and laboratory data — served as an objective marker of the onset of critical illness (Am J Respir Crit Care Med. 2014;190:649-655). eCART was developed at the University of Chicago, which holds a patent on it.

"Implementation of a real-time risk stratification tool such as the eCART has the potential to lead to earlier recognition of impending critical illness and improve outcomes for these patients," Dr Wendlandt told Medscape Medical News.

The findings were presented here at the Society of Hospital Medicine 2015 Annual Meeting, where the study was chosen as one of three Best of Research and Innovations in 2015.

"The thing that jumped out at me was the fairly strong and linear association between delay in ICU transfer and an outcome that we care about," said Margaret Fang, MD, who chaired the abstract selection committee and the conference session. "My question is, what is the mechanism behind that?"

Reasons for Delay

We definitely do not want to take provider opinion out of the equation. Bedside gestalt is irreplaceable.

Dr Fang said it is important to learn whether the delay is related to vital signs being missed or whether doctors and nurses are attempting to manage the patient on the floor. She wondered whether there might be an association between delays and ICU bed availability or nursing staff to patient ratio on the ward.

"I think the next step for these investigators is to try to figure out what is contributing to this," she said. "And now that they've demonstrated this association, can we be proactive about it?"

The retrospective cohort involved 3789 patients admitted to the medical or surgical ward of one of five hospitals from November 2008 to January 2013, who were subsequently transferred to the ICU. Dr Wendlandt and her team calculated eCART scores for each patient.

With eCART, the score is recalculated each time new data become available. If a patient's score trends toward the cutoff of 60, which is known to be predictive of adverse outcomes, that patient is flagged as being critically ill and a bedside assessment is automatically triggered.

A delayed transfer was defined a priori as a transfer more than 6 hours after an eCART indication of critical illness. For the entire group, the median time to ICU transfer was 5.4 hours.

Transfer was timely and occurred within 6 hours for 54% of patients; of these, 20% were transferred within the first hour. For 46% of the study cohort, however, transfer was delayed.

Patients who experienced a timely transfer were slightly older than those who experienced a delayed transfer (73 vs 71 years; P = .002), but there were no differences for sex, race, or surgical status.

There was a linear relation between transfer time and mortality. For patients transferred within 1 hour, in-hospital mortality was 23%; for those transferred 23 to 24 hours after the initial eCART indication, the rate was 38%.

Each additional hour of delay was associated with a significant 3% increase in mortality (P < .001), Dr Wendlandt reported.

The rate of in-hospital mortality was significantly lower in the timely group than in the delayed group (25% vs 33%; P < .001). For those who survived to discharge, hospital stays were 2 days shorter in the timely group than in the delayed group (11 vs 13 days; P < .001).

Dr Wendlandt acknowledged that not having information on the reasons for delay is a study limitation.

"The actual causes for delayed ICU transfer would be the subject of a whole separate investigation," she told Medscape Medical News. However, "we suspect that at least one of the major contributors is delayed provider recognition of increasing patient acuity, which is why our emphasis is on the real-time use of an objective marker to indicate the onset of critical illness."

Dr Fang pointed out that the study only looked at people who went to the ICU, not the overall number of patients who crossed the eCART critical illness threshold.

"How many had a high severity score, and of those, how many had something bad happen to them? It may be that a lot of people have high severity scores, but only a small proportion go to the ICU. If that's the case, maybe that surveillance system isn't the most appropriate. I think that would be the next thing to look at," she said.

Dr Wendlandt was asked by an audience member whether the eCART score could replace the judgment of nurses.

"We definitely do not want to take provider opinion out of the equation," she said. "Bedside gestalt is irreplaceable. I think the question is how to find the right balance between those subjective feelings toward patient acuity and the right objective markers. We're working on trying to find that balance."

Dr Wendlandt has no disclosures beyond her institution's eCART patent. Dr Fang has disclosed no relevant financial relationships.

Society of Hospital Medicine 2015 Annual Meeting. Presented March 31, 2015.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: