By Debra Wood, RN, contributor
May 31, 2013 - Early intervention for septic shock, respiratory failure and other conditions improves patient outcomes, but mortality risk significantly increases with delays in transferring declining patients to the intensive care unit (ICU), according to a new study from researchers at the University of Chicago Medical Center. But several things can be done to improve transfer times for these critically ill patients, experts point out.
“Our current study suggests that each hour matters once a patient is critically ill on the wards,” said lead author Matthew Churpek, MD, MPH, of the University of Chicago Medical Center. “Using an evidence-based risk score, such as the CART [Cardiac Arrest Risk Triage] score, may improve the identification of these patients.”
Churpek and colleagues developed the CART score, which is a vital sign-based early warning tool. It uses respiratory rate, heart rate, diastolic blood pressure and age, and a prior study showed it correctly identified patients who would have a cardiac arrest or need transfer 48 hours before the event with 90 percent specificity.
“Our CART score has been shown to detect cardiac arrest and ICU transfer in our hospital more accurately than the widely used MEWS, or Modified Early Warning Score,” Churpek said. “In particular, our current study suggested that patients with delayed transfer were older than those transferred earlier. The CART score takes age into account when calculating a patient's risk of deterioration, making it more likely to identify this vulnerable population.”
Transfer delays and mortality
In the current study, the researchers examined outcomes after delayed transfer to the ICU. It included 2,166 patients at an academic hospital transferred from medical–surgical wards to the ICU. Of these patients, 425, or 20 percent, died during admission.
In a subset of 260 patients transferred to the ICU within six hours of a first critical CART score value, 27 percent died during the hospital admission, compared with 45 percent of patients for whom ICU transfer was delayed for more than six hours.
Each one-hour increase in transfer delay was associated with a 7 percent increase in the odds of dying in the ICU, Churpek said. The odds of ICU mortality, with the chance of dying in the hospital, reached 52 percent among patients in whom transfer to the ICU was delayed 18-24 hours after reaching the critical CART value.
Abhay Padgaonkar indicated multiple steps must be taken to address all of the places where patient transfer delays can occur.
Abhay Padgaonkar, president of Innovative Solutions Consulting in Phoenix, said he was not surprised by the findings but questioned whether delays were a localized problem or if this is happening across the board.
Churpek said that delays occur in other hospitals as well.
“Studies looking at delayed transfers due to lack of beds have been published from various settings in and outside of the U.S,” Churpek said. “This is likely a widespread problem.”
Churpek’s research team has not yet looked at the specific reasons for the delays. However, he said, “We believe that there are many contributing factors. First, there is likely a delay in recognition of deteriorating patients. Second, once recognized, previous studies have shown that there is often a delay in calling for back up, both in activating systems such as rapid response teams and calling for critical care expertise. Finally, even when appropriately recognized and triaged, there can also be delays in transfer due to a lack of available beds.”
Prior research from the team found that lack of ICU beds was associated with an increase in the rate of cardiac arrest on the wards, he said.
Padgaonkar added that having a bed is not enough, it must be a staffed bed.
“You cannot put a patient in an empty bed if there is not an ICU nurse,” Padgaonkar said. “A staffed bed is key.”
Other causes of delay include the nurse not identifying the patient is failing, whether due to competency or if he or she is stretched too thin to properly monitor patients, Padgaonkar said.
“Part of the equation is a set of things that may be obstacles, including nursing ratios, if you have too many patients per nurse,” Padgaonkar added. “Nurses have to recognize the patient’s status has changed, and then the second thing is getting hold of the doctor.”
With hospitalists responsible for greater numbers of patients and not always on site, promptly reaching the physician may be a problem, he said. Additionally, someone covering for the admitting doctor may not know enough about that patient. They may want to come in, causing further delay.
Once the order is obtained, logistics, including physically transporting the patient can take hours, he said.
What can be done?
There is something that can be done to avoid stumbles at each of the steps and improve patient transfers: recognition, staffing ratios, communication with the physician, available staffed ICU beds and logistic efficiencies, Padgaonkar indicated.
“There are several things that can be done to speed up transfers,” Churpek agreed. “First, adopting an evidence-based risk score for patients on the wards can aid in the identification of at-risk patients.”
Additionally, Churpek said, “Efforts to improve nurse–physician communication regarding deteriorating patients can go a long way to ensuring patient safety.” Finally, he added, hospitals can evaluate their ICU bed availability and alter factors, such as OR schedules, to free up beds on those days with the most need.
“Research to improve patient identification, communication, triage decisions and bed availability is needed to improve the outcomes of these critically ill patients,” Churpek concluded.
© 2013. AMN Healthcare, Inc. All Rights Reserved.