OR in the ER

By David Eitel and Douglas A. Samuelson

No room at the ER: At many emergency departments, patients are “boarded” up to 24 hours on hallway gurneys.

Emergency medicine in the United States has reached a crisis state. Operations research has demonstrated a critical capability to address the problems, but most emergency care physicians and managers are unaware of this capability. As a few recent engagements have shown, O.R. analysts could contribute greatly to addressing this crisis if they change their communications approach in order to be more effective in supporting health care providers.

In late 2006, a number of the first author’s colleagues reported that emergency department (ED) conditions abruptly worsened all across the country. Apparently this was just due to congestion hitting a certain level, without a precipitating event, much like some highways at rush hour. Many patients who were “admitted” to hospitals remained in the ED up to 24 hours (and longer), lying on gurneys in ED hallways waiting for a hospital bed to become available. In Dr. Eitel’s own prototypical 40-bed ED, 20 to 35 patients (or more) were held or “boarded” – most days of the week, most hours of the day – until a hospital bed became available. (Note: “Emergency Department” and “Emergency Room” mean pretty much the same thing. “Department” has gained more widespread use lately since many EDs now comprise multiple rooms.)

The first author of this article, an experienced emergency physician who is also an MBA and a long-time member of INFORMS and its Health Applications Section, presented this situation at the annual INFORMS meeting in 2007, along with Murray Cote and Keith Willoughby. The presentation was intended to be a call for help. In the three and a half years since, the situation has gotten worse, and the O.R. profession’s help is needed more than ever.

The 2007 presentation featured photos that showed signs hanging from the ceiling in Dr. Eitel’s ED. These signs were a recent addition to the ED, and they were simply numbers. The numbers were part of a location-coding scheme that helped the ED keep track of which patients were “temporarily” stashed where – in the hallways on gurneys – because of the overcrowding.

The hospital had no plan to send extra resources to the ED to help care for these patients, who were often very ill, admitted for further treatment but then “boarded” in the ED. The ED caregivers were told that there simply were not resources available elsewhere in the hospital to send to the ED. This meant that ED care providers were required to deliver care to admitted but ED-boarded patients while simultaneously treating other patients who continued to arrive via walk-in triage and ambulance in ever-increasing numbers – an impossible situation. Compounding the problem, the ED was not physically set up to deliver the services and procedures many of the admitted, ED-boarded, ill patients required.

Most EDs in the country – for many hours of the day, most days of the week – are chaotic, frankly dangerous care-delivery settings. ED waiting rooms are packed with sick and injured patients, including those with potentially devastating presenting problems (such as abdominal pain, which can often be minor but can also be the first sign of a variety of very serious conditions). Many ED care providers, especially nurses asked to function in this untenable situation, can no longer tolerate it professionally or morally. As a result, the ED nursing turnover has become frighteningly high across the country. Meanwhile, upstairs in many hospitals, because of the pressure to get admitted patients out of the ED and into continuing care, an “any bed will do” philosophy permeates patient placement and thus hospital care delivery. Hence, admitted patients often go from the ED to the first available bed rather than to a unit with the skills mix and tools suitable for their presenting problem(s).

Hospital floor units are also understaffed. Nurses in particular struggle to find the time they need to provide the care they are trained to give to patients. As they become frustrated and leave, the “nursing shortage” deepens. In the November 2007 presentation, the first author said that the situation had gotten substantially worse over the preceding year. Now, in 2011, the situation is much worse, and that’s with no SARS or swine flu. What will next year look like? More signs hanging from the ceilings of EDs? This does not seem like the solution O.R. would recommend.

American health care providers often call this state of ED congestion “ED overcrowding.” Australian colleagues call it “access block.” O.R. analysts have another term for it: systems failure. The problem pervades the whole system, far beyond just the ED.

Most EDs in the country – for many hours of the day, most days of the week – are chaotic, frankly dangerous care delivery settings.

What can be done to improve the delivery of care in the ED and throughout the hospital setting?

The methods of O.R. and the analysts who practice it have much to contribute in the ED and hospital. For example, a few recent efforts have achieved dramatic improvement in the problem ED patients complain about most: the wait to be seen. The wait, in turn, depends critically on the accuracy and consistency of triage – the determination of who most urgently requires care. While many if not all EDs intensely review the care practices once patients are seen, far fewer look at the triage process.

One O.R. response to this problem is the Emergency Severity Index (ESI), developed in the late 1990s at seven hospitals in the United States. ESI is an ED triage tool that categorizes patients by acuity of patient (how threatening the symptoms are) and the resource(s) needed to deliver care, yielding a rapid, reproducible, operationally and clinically relevant patient stratification from 1 (most urgent) to 5 (least resource intensive). Assessments and resultant training of nurses sharply reduced inconsistency in triage, producing higher reliability and hence fewer truly urgent patients suffering preventable complications because of long waits for care. This, in turn, also produced overall reductions in wait times, as fewer patients required the extra and time-consuming efforts that become necessary when high-acuity patients wait too long. Finally, because fewer of the most seriously ill patients were lying around waiting – vulnerable to various hazards (including, in at least one unfortunate well-publicized case in Philadelphia, robbery by other patients) – patient safety also improved.

O.R. concepts can be taken much further. A more recent innovation is Door-to-Doc, an ED service delivery system for assigning all ED resources in the context of service demand with the goal of virtually eliminating waits for appropriate care. A key assumption in most EDs is that patients must be assigned a bed before they can be treated, but beds are scarcer than physicians and not always needed. With reliable “quick look” triage and very early assessment of patients by an emergency physician to set up patient care streams, all patients can be treated quickly. Door-to-Doc was developed by Dr. Jeff Cochran at Arizona State University in partnership with the Banner Health Care under a grant from the Agency for Healthcare Research and Quality (AHRQ). It has been tried with considerable success at hospitals in the Banner Health System. The Door-to-Doc toolkit is available free of charge, but significant training is needed for effective use.

Already moving on his own path but building on the success of Door-to-Doc, Dr. Joseph Guarisco, chair of Emergency Medicine at the Ochsner Health System in New Orleans, developed qTrack. qTrack is a complete service system approach to delivering Emergency Department care. The first author describes his offered training solutions at his Web site, www.nowaited.com. He and others are also working to develop a spreadsheet-based, scenario-driven predictive analytics tool that will provide many of the “what-if” benefits of simulation to people who do not know and are not about to learn what simulation is.

So what is keeping such successes from rapid dissemination and productivity in the ED community? The answer is that within the context of existing EDs and hospitals, clinicians and managers are largely unfamiliar with the discipline of O.R. and its capabilities. The first author of this article first learned of O.R. as part of an executive MBA program in 1993 after many years of clinical practice. Until then, he had no idea that such methods existed.

With new patients continuously arriving via ambulances and as walk-ins, accuracy and consistency of triage is critical in EDs.

Another obstacle is the O.R. community’s focus on tools and techniques, while savvy managers in any application area recognize that correct understanding of the problem is much more important. A seminal work on quality improvement, “The Goal,” includes this quote: “Let me explain. When I was a physicist, someone would come to me from time to time with problems in mathematics he could not solve. He wanted me to check his numbers for him. But after a while I learned not to waste my time checking the numbers, because the numbers were almost always right. However, if I checked the assumptions, they were almost always wrong.”

The reality:

• The vast majority of ED and hospital clinicians and managers have no idea that O.R. even exists, let alone what it can do to address some of these health care systems and operational problems.
• Medical professionals have gotten by, forever, “by getting by.” They have become very good at “workarounds” versus methods O.R. analysts might reflexively think about. Medical providers have a few tools of quality that they use on occasion, but such behaviors are not part of the medical culture. And culture will eat strategy for lunch every time.
• Hospital clinicians and managers are very bright people, but they are generally not “mathematical.” In fact, most are “quantitatively hesitant,” although they can do dosage-dilution-response calculations in their heads and expect nurses to be able to do so as well. The second author has succeeded in lowering resistance in some cases by pointing out this capability and its implications.
• The terminologies O.R. analysts use for their concepts and tools are incomprehensible and hence absolutely frightening to most clinicians and managers, so they just run away.

This means that to succeed, O.R. analysts must move away from the language of tools to the language of solutions. Instead of “variability and interdependency,” analysts need to talk about “the way the world really works, uncertainty and all.” Analysts can talk about “effectively forecasting service loads” without discussing multiple linear regression, exponential smoothing or other such technique-specific terms. Analysts can talk about “capacity planning” without queueing theory and its various Greek letters that characterize how rapidly people arrive and how quickly they get served. Analysts can offer to help “schedule staff, taking business realities into account,” without trying to explain integer programming and constraints.

Based in part on a brainstorming exercise that was part of the first author’s presentation at the INFORMS Health Applications Conference in Montreal in June 2011, following are the key recommendations to help analysts broaden O.R.’s role in the U.S. health care system:

• Partner with others with similar interest, such as the Health Care Division of the American Society for Quality and the HIMSS and Health Systems subdivisions of the Institute of Industrial Engineers. In addition, attend meetings of health care professional societies and learn the terminology and concerns of the health care community.
• Aggressively promote “analytics” in health care settings. O.R. analysts have to let ED and hospital clinicians and managers know that such analysts exist and what analysis can do for them “to improve the delivery of care in the ED and throughout the hospital setting.” Select those solutions likely to be most helpful to them. (Recall that many health care environments are often chaotic.) Use terms that are neither frightening nor off-putting. Use terms that are easily understood. Remember that health care providers and managers have little awareness of the existence of O.R. tools, and few if any have seen O.R. applications within an ED or hospital.
• Present O.R. as a uniquely productive way of looking at the world rather than as a tool kit. Medical providers do get the idea that system problems require system solutions. “Curing Health Care” is now 20 years old, and later reports by the National Institute of Medicine made this clear. But medical training promotes specialization rather than systems thinking and focuses on the patient rather than on the system of care. O.R. analysts can help with the systems approach, if they can communicate that capability in the system managers’ terms.

Major opportunities abound in this area for O.R. analysts who are willing to get informed and get involved.

David Eitel, M.D., M.B.A., is a physician advisor in case management for Wellspan Health System, York, Pa. He has been an emergency physician for 35 years. Douglas A. Samuelson, D.Sc., is president and chief scientist of InfoLogix, Inc., a consulting and R&D firm in Annandale, Va. He has applied operations research to a wide variety of problems, with health care as one of his major interests. He is a contributing editor of OR/MS Today.

References

1. Donald Berwick, E. Blanton Godfrey, and Jane Roessner, “Curing Health Care,” Jossey-Bass, 1990
2. Committee on Quality of Health Care, Institute of Medicine, “To Err Is Human: Building a Safer Health Care System,” Washington, D.C.: National Academy Press, 2000.
3. Committee on Quality of Health Care, Institute of Medicine, “Crossing the Quality Chasm: A New Health System for the 21st Century,” Washington, D.C.: National Academy Press, 2001.
4. David Eitel, Murray Cote and Keith Willoughby, “The Congested Emergency Department: A Physician’s Perspective,” INFORMS Annual Meeting, 2007.
5. David Eitel, training to support qTrack and other suggested innovations, http://www.nowaited.com.
6. Eliyahu M. Goldratt and Jeff Cox, “The Goal: A Process of Ongoing Improvement,” North River Press, 2004 (3rd ed.).
7. Joseph Guarisco, “Door-to-Doc,” AHRQ 10th Annual Conference, September 2010.
8. Ed Popovich and David Eitel, “Emergency Department Triage Reliability and Impact Upon Patient Flow,” WCBF 10th Lean Six Sigma and Process Improvement in Health Care Summit, New Orleans, May 2011.
9. Douglas A. Samuelson, “A Dose of OR: Analysts Explore Ways to Improve Quality and Productivity in Health Care,” OR/MS Today, December 1991.
10. Douglas A. Samuelson, “Diagnosing the Real Health Care Villain,” OR/MS Today, February 1995
11. Douglas A. Samuelson, “A New Frontier? Health Services Research and Medical Informatics,” OR/MS Today, February 2000.
12. Richard Wuerz and David Eitel, “Emergency Severity Index (ESI): http://www.ahrq.gov/research/esi: Emergency Severity Index Version 4 Implementation Handbook and DVDs.