By Anne Scheck
The Massachusetts experience, as it has been dubbed, suggests that diverting ambulances may not be necessary for maintaining a successful emergency safety net, at least at some centers.
In a regulation unprecedented by other states, Massachusetts legislators banned diversion two years ago. Many hospitals that relied on the practice to ease influx braced for the worst.
But at his institution, the worst just never happened, said Stephen Epstein, MD, an emergency physician at Beth Israel Deaconess Medical Center (BIDMC) and a member of the Massachusetts Department of Public Health's Boarding and Patient Flow Task Force.
In fact, there was no discernible impact at all. “When ambulance diversion stopped, there was no effect on length of stay,” Dr. Epstein noted.
This new and so far relatively circumstantial evidence supports what some in emergency medicine have long contended: Diversion is just not that effective for managing crowding. “Diversion is like a Band-Aid. It doesn’t relieve the root causes,” said Laura Burke, MD, an emergency physician at BIDMC and a co-investigator on the study.
Only months after the two Boston researchers announced their findings, hospitals across the country were reporting similar accounts, mostly anecdotally and with tepidity, but stating fairly optimistic findings nonetheless. This past spring, hospitals in New York revealed ambulance diversions by emergency departments in the Syracuse area were down about 75 percent, compared with the previous year, following a cross-city commitment to reduce the practice. (The Post-Standard, May 26, 2010; http://bit.ly/SyracuseED
Some metropolitan areas in California had similar if less dramatic outcomes, according to a multicenter study by the California ED Diversion Project (www.caeddiversionproject.com
), which showed declines of about 20 percent after a collaborative effort to diminish diversion. (J Emerg Med [abstract] September 2010; http://bit.ly/CalifDivert
How did they do it? The same way as BIDMC: by implementing best-practice approaches. Several such measures were instituted at BIDMC to improve patient flow within the hospital and to reduce ED crowding. An electronic "capacity dashboard" was created, which displays real-time ED and inpatient utilization rates, providing valuable information about available inpatient beds, explained Dr. Burke, a clinical instructor of medicine at Harvard Medical School.
Also reduced were preferences for elective or nonemergent admissions, which would otherwise compete with ED admissions when the hospital is crowded. In general, when the hospital was full and a hospital bed was reserved for an elective admission, admitted ED patients often remained in the ED until a hospital bed was available, Dr. Burke explained. Reducing these preferences minimized boarding admitted patients in the ED, she said.
At the annual meeting of the Society for Academic Emergency Medicine, where the results of this initial pilot study were first presented, “people were excited about this idea,” Dr. Burke noted.
If confirmed in a larger study, such a finding may prove reassuring to the public, who as a rule feel better served without diversion, and for good reason. “When patients are seen at a hospital where they have been before, there is greater likelihood for continuity of care, and in some cases, redundant testing can be avoided," Dr. Burke said.
All these observations may sound like just common sense, but they have been debated in the medical community since horsedrawn carriages first galloped through the streets of New York carrying suddenly sick inhabitants to hospitals, prompting physicians in the late 1800s to officially dub such lifesaving wagons “ambulances.” And such fast-wheel traditions have remained, in the form of three-wheeled scooters that now zip around events where ambulances cannot, at places like Madison Square Garden, helping hoist those taken ill among throngs of people.
The idea of fast transport has not been challenged, but as two parallel systems developed, the concept of the need in emergency care was transformed. The first change occurred with professionalization of prehospital care, as the first-responder teams employed better ways to stabilize patients. The second occurred right along with the prominence of emergency medicine, or more precisely, of the ED itself, which began attracting so many patients that it is common practice for physicians even outside the specialty to call emergency medicine a “victim of its own success.”
The Institute of Medicine, in reporting on the issue, observed an increase in boarding and crowding over the past decade, and it advised there was a link between the two, suggesting a reason for the dramatic rise in the number of ambulance diversions. This also was the conclusion of well known previous research into how crowding problems have been surmounted in emergency care. (Urgent Matters project; www.urgentmatters.org
Actual studies of the impact of diversion on patient health have been sparse, however. The reason: The practice of it was instituted to save lives, and if there is solid evidence showing they are imperiled instead, liability could be created. As it stands now, “EMTALA creates a hole in the financial dike of U.S. healthcare, and diverting ambulances provides a partial and intermittent plug,” according to a legal treatise on the issue. (Legislation and Public Policy 2010;13:175; http://bit.ly/DiversionSolutions
The scant research available — most of it from Europe — was equivocating, but a study from Australia proved more definitive, suggesting that patient safety, as measured by mortality is “positively enhanced” by ambulance diversion. (Med J Aust 2005;183[11-12]:672.) That finding has met with recent challenge in the United States. When deaths from acute myocardial infarction in New York City were investigated in light of ambulance diversion, AMI mortality was associated with high levels of diversion and simultaneous diversion across hospitals. (Inquiry 2010;47:81.)
But one study doesn’t change minds, particularly those from clinical medicine, nor should it, said Gretchen Chapman, PhD, a professor of psychology at Rutgers University in New Brunswick, NJ. Individual physicians tend to be influenced by their own caseloads and the personal experiences of their practices, she said. An “anchoring bias,” as social scientists sometimes call it, can be advantageous. Reluctance to embrace new data encourages solid scientific proof. “Physicians have their own private database,” Dr. Chapman said.
At Rutgers, the prevalence and incidence of ambulance diversion in New Jersey was examined by the Institute for Health, Health Care Policy and Aging Research. The study included seven hospitals, and Derek DeLia, PhD, an associate research professor at the Rutgers Center for State Health Policy, found that one such diversion occurred on average about every hour, with peak times in January and more likely at the beginning of the week rather than the end. (“Hospital Capacity, Patient Flow, and Emergency Department Use in New Jersey,” Report for the New Jersey Department of Health and Senior Services, September 2007; http://bit.ly/HospCapacity
One inference of the report is that increased diversion is somewhat predictable, and thus capable of being the focus of some preparation and planning. This is not a new concept; a few years ago, emergency medicine researchers wrote an entire paper to declare “management is prediction” in handling ED crowding. (Acad Emerg Med 2006;13:1095.)
It remains to be seen whether prohibiting ambulance diversion means that availability is sustained by EDs for potentially high patient loads, but Boston seems an ideal place to test the idea. With so many hospitals and so little time, those questions may first find answers in the city that arguably has the worst snarls of traffic, pedestrians, and bike riders in America.
Does enabling ambulance personnel to use their own discretion mean EDs across that city utilize facilities more efficiently? Does the Massachusetts ban on diversion affect some EDs more than others, and if so, what are those underlying reasons? The BIDMC pilot study was a first step, and now a larger, more comprehensive one is being launched involving multiple EDs, Dr. Epstein said. “There are questions that we hope a larger study might answer,” he added.