Having a critical care specialist physically present in the intensive care unit overnight does not seem to improve a broad range of patient outcomes, according to research published May 20 in the New England Journal of Medicine (doi: 0.1056/NEMJoa1302854).
Dr. Meeta Prasad Kerlin and her colleagues at the University of Pennsylvania, Philadelphia, randomized admissions to a 24-bed medical ICU over a 1-year period; they offer several explanations as to why having a nighttime intensivist does not seem to make a difference. Their research was presented simultaneously at the annual meeting of the American Thoracic Society.
Photo Credit: Robert Press/Penn Medicine
Bringing in a new physician in the evening, rather than keeping the same one on call, could actually be a disruption to patient care, Dr. Meeta Prasad Kerlin's research found.
First, if the ICU is well staffed during the day and has adopted systems of care, there might not be a lot of room for patient improvement. Second, bringing on a new physician in the evening – rather than keeping the same intensivist on call – might disrupt continuity of care for some patients, they said.
All ICU patients admitted to the hospital at the university were randomized to the intervention group or the control group. The intervention group received care from an intensivist (a board-eligible or board-certified critical care specialist) who was on staff from 7 p.m. to 7 a.m., or from residents who normally were assigned to the ICU.
The control group received care from residents, who were able to reach two daytime intensivists or two critical care fellows at night by phone. Generally, the nursing ratio was one per two patients.
The nighttime intensivists were daytime staff who volunteered to take on a nighttime assignment. During the study, they covered one night a week.
Overall, 820 patients were assigned to the intervention (4 were later excluded due to missing data), and 778 were in the control group (7 ended up with missing data). In total, 175 nights were assigned to the intervention; 95% (166) of those nights were staffed by an intensivist.
The median age of patients was 60 years, and about half were men. Almost half were admitted from the emergency department, while just over a third were transferred from the hospital’s general floor. The median APACHE (Acute Physiology and Chronic Health Evaluation) III score was 67.
Sixty-one percent (970) of all patients were admitted to the ICU at night (5 p.m. to 5 a.m.). Overall, 381 patients (24%) died in the hospital, with 293 (18%) of them dying in the ICU.
The authors found no effect on length of stay in the ICU (median, 52.7 hours) for those in the intervention group. They calculated a "rate ratio," which was the rate of instantaneous discharge from the ICU in the intervention group divided by the rate of instantaneous discharge from the ICU in the control group. For ICU length of stay, the ratio was 0.98 (P = .72); a ratio of more than 1 would mean that the intervention shortened the time to discharge.
They found similar ratios for hospital length of stay (median, 174 hours in the intervention group vs. 166 hours in the control group) and mortality. A patient’s APACHE III score made no difference in the outcomes; neither did resident experience and training .
In a subanalysis, the authors determined that ratios were unchanged for patients admitted at night.
Having so many patients admitted at night was a strength of the study, the authors said. "If nighttime intensivists were effective, it is likely they would be particularly effective in an ICU with such a large nighttime workload," they wrote.
The researchers noted that further research is important because currently one-third of U.S. academic medical centers employ nighttime intensivists. On the surface, nighttime ICU staffing seems compelling, said the authors. But it also may be "one of several expensive medical practices that have been adopted without a supportive evidence base."
On Twitter @aliciaault