Increased off-hours mortality forces hospitalists to learn new skills, forge new partnerships, increase staffing
by Larry Beresford
Recent research published in the Journal of the American Medical Association (JAMA) highlights a continued hospital care issue identified in previous studies: worse patient outcomes from in-hospital cardiac arrests occur at night or on the weekend.1
An analysis of 86,748 adult cardiac events between January 2000 and February 2007 in 507 hospitals participating in the National Registry of Cardiopulmonary Resuscitation (NRCPR) compared outcomes at night (from 11 p.m. to 6:59 a.m.) and weekends (from 11 p.m. Friday to 6:59 a.m. Monday) with day/evening shifts. The primary measure of survival to discharge and secondary outcomes from in-hospital cardiac arrests were significantly worse during nights and weekends. In essence, heart attack patients were 41% more likely to survive if treated during daytime weekday hours.
“This was the first comprehensive, large-scale study in a cross section of hospitals across the country of heart attack survival differences between shifts,” says lead author Mary Ann Peberdy, MD, of Virginia Commonwealth University in Richmond. “We adjusted for a variety of potentially confounding factors and patient characteristics, none of which explained the worse outcomes nights and weekends.”
The national database was not designed to provide an explanation for its findings, which may be due to multiple patient, event or hospital factors. “We can’t exclude physiological factors of patients or of staff working on the night shift,” Dr. Peberdy explains. “But I think we need to focus on process issues. We know that hospitals simply do not run the same way at night. Things are different—more errors, more accidents, more needle sticks, less people around. Those who work the night shift may also be less experienced,” and early identification of heart attacks is critical to positive outcome.
The JAMA results confirm previous research documenting worse outcomes on nights and weekends. Single-site studies and a smaller study of heart attack survival in New Jersey hospitals for weekend versus weekday admissions found similar trends.2
Stroke patients who enter the hospital at night or over the weekend are more likely to die in the hospital than those admitted during daytime hours (7 a.m. to 6 p.m.) on weekdays, according to two studies presented at the American Heart Association’s International Stroke Conference in New Orleans in February 2008.3 Those differences were particularly striking for hemorrhagic strokes. Similar outcomes also have been reported for pulmonary embolisms.
Different studies have approached this issue in different ways, comparing business hours (e.g., 7 a.m. to 6 p.m.) with evenings, nights, and weekends, or days and evenings up to 11 p.m. with nights. Weekends are compared to weekdays but also to weekend nights.
The size of the hospital did not explain the shift differences found in Dr. Peberdy’s study. For hospitalists trying to address the underlying problems of after-hours quality, the size of the hospital is relevant. Only larger hospitals can afford hospitalist groups large enough to cover night and weekend shifts. For those that can, are all members of the group taking their turns at night, does this duty fall to the junior members, or is the group lucky enough to employ nocturnists who want to work at night? (See The Hospitalist, May 2006, p. 27, for an article on nocturnists in hospital medicine.) Whether the hospital has an academic emphasis also can influence who responds to crises after hours—attendings or house staff (in other words, sleep-deprived residents).
If hospitalists work nights, they are more likely to notice what isn’t available or what doesn’t work as well as what contributes to nocturnal quality problems and what might help to compensate for these differences. Even if hospitalists are not present in the facility at night, technology can help guide appropriate response to cardiac crises, suggests David Grace, MD, area medical officer for the Schumacher Group’s Hospital Medicine Division and a hospitalist at Southwest Medical Center in Lafayette, La.
“Several weeks ago a patient in the hospital was having chest pains,” he recalls. “A nurse called me at home and I ordered an electrocardiogram.” The electrocardiogram’s (EKG) computer program indicated “nonspecific changes” in the patient’s cardiac function, but Dr. Grace asked the nurse to scan the printout and send it to his Web-based fax number.
“I looked at the EKG on my PDA,” he continues. “It was subtle, but it seemed to me that this patient was having a myocardial infarction[MI]. I told the nurse to do the blood work for a suspected MI, give the patient an aspirin and take another EKG, which more clearly showed the MI. If I had not had the ability to look at the printout, I would have had to trust the nurse’s observation or the EKG computer program. As it was, we caught it early and the patient did well.”
“I have lived it. I certainly understand the research showing different outcomes from MIs at night,” Dr. Grace observes. “At night, patients are usually asleep, so processes that begin with early warning signals, such as chest pains, may go further down the path before they are identified, especially if the patient has taken a sleeping pill. Often, nurse-to-patient staffing ratios are dramatically different at night—and somewhat reduced on weekends. I’ve also worked in hospitals where on weekends, unless it was a true life-threatening emergency, you could not get an MRI. So if you ordered one on a Saturday, it wouldn’t happen until Monday. There are things you pick up on the MRI that you miss on the CT scan; for example, bleeding, which can affect your management of the patient.”
“I would not be surprised to hear of worse survival for any of these acute decompensations—it goes for acute GI bleeds, stroke, and hemorrhage, as well as MIs,” adds Steven Liu, MD, of Emory University Eastside Medical Center in Atlanta. “We dealt with the problem in this hospital five years ago and addressed a lot of these quality issues by partnering with hospital administration and specialists.”
On weekends, many services may be less available or not available at all in the hospital, from interventional radiology to physical therapy. If patients have fewer contacts with different hospital personnel, it is more likely that subtle early signs of acute problems will go unnoticed until later.
“At my hospital, the interventional radiology [IR] department is not open 24 hours,” says Erica Grabscheid, MD, associate director of the hospitalist program at Beth Israel Hospital in New York City. If a peripherally inserted central catheter needs to be placed on the weekend but IR is not available, the patient may have to wait until Monday morning. Alternatives for the hospitalist, Dr. Grabscheid says, are to become skilled at line placement or to collect data for the hospital’s administrators on the costs of not having 24-hour IR.
Jeffrey Robinson, MD, hospitalist group leader for Intermountain Medical Center in Salt Lake City, Utah, believes staffing is an essential part of the equation. He says Intermountain Health has made a commitment to do what’s necessary for quality patient care—including adequate staffing. “Every time we feel we’re stretched, we add more hospitalist shifts, including weekends and holidays. I feel we give good patient care from the physician side. Obstacles have more to do with ancillaries, but we’ve made great progress in addressing these, as well.”
At one time, interventional radiology only was available in the hospital five days a week, Dr. Robinson notes. “You couldn’t get feeding tubes placed or certain other interventional procedures. But with the commitment of radiology and the hiring power of Intermountain Health, we now have 24-hour availability. MRIs have been harder to cover after hours, but there is now an on-call team for MRIs.”
In each case, hospitalists were important advocates for expanding the hours of availability, Dr. Robinson says: “We needed to gather data and look at results for patients staying over the weekend. We also got expanded case management, so that nursing home placements could happen on Saturdays and Sundays.” TH
Larry Beresford is a medical writer based in California.
The Hospitalist newsmagazine reports on issues and trends in hospital medicine. The Hospitalist reaches more than 25,000 hospitalists, physician assistants, nurse practitioners, residents, and medical administrators interested in the practice and business of hospital medicine.