| |||||||||||||||||||||
| |||||||||||||||||||||
| |||||||||||||||||||||
 | |||||||||||||||||||||
| |||||||||||||||||||||
|
Clinical Operations Overview The Emergency Department (ED) of Northwestern University and Northwestern Memorial Hospital (NMH) has been a rapidly evolving clinical operation since the arrival of the current departmental chair in 2000. The same 30-bed emergency department which cared for 63,000 patients in 2000 saw well over 71,000 patient visits in 2005. Despite this brisk increase in patient volume, the ED has been able to achieve substantial improvements in the technical quality and safety of the care it delivers. The department has had year-over-year improvements in patient satisfaction. And, under the same pressures of decreasing accessibility of inpatient beds that has lead to ever escalating waits across major urban academic centers across the country, Northwestern has been able to steadily decrease both the time to care and the average patient’s length of stay.
The Emergency Department has made a very concerted effort to synchronize its strategic goals with the strategic thinking of Northwestern Memorial Hospital. To that end, the Emergency Department has focused its efforts to:
Each of our strategic goals outlined above also have associated quality improvement goals. The end result of these goals is specifically to improve clinical outcomes and processes with a focus on patient safety/reducing medical errors, enhanced pain management, infection control, and population-specific outcomes, with special attention to national quality standards, benchmarks, best practices, and variance reduction. Additionally, we strive to improve patient satisfaction to targeted levels as measured by surveys.
ESI Triage System In order to achieve these strategic goals, continued innovation has been required. One of the first such innovations focused on the modification of the intake process. Northwestern Memorial Hospital adopted the Emergency Severity Index (ESI) triage system back in 2000. ESI is a nursing triage system with proven reliability and validity. Paula Tanabe, RN, PhD, Research Assistant Professor in the emergency department, and a Fellow of the Northwestern University Center for Healthcare Studies, co-authored the original book on this triage tool and has been instrumental in the formulation of ensuing versions that have broadened its use across a wide variety of clinical settings and age distributions; across the nation and internationally. Process Improvement Using the Six-Sigma Methodology Process improvement efforts from 2001-2002 were enhanced by the efforts of General Electric consultants, whose Six Sigma methodology helped us critically analyze key processes. The goal was to optimize the system while minimizing variability. Toward this end, scientifically-informed industrial engineering practices were being applied to the ED. Significant strides were made in operational quality during this time and marked reductions in time to care in the ED and length of stay in the ED were achieved. Goals including the development of statistical databases and real-time process controls (connection to patient tracking systems created in-house in the department “Traxion” and “Monitor Man”) were also developed to facilitate operational design and management. With these accomplishments in place, operational overload, backlogs and delays could be recognized as they occur. Even better, the foundation was laid to begin to predict and prevent problems before they happen. Expanding Departmental Staffing and Improving Coordination Through the generosity of the Northwestern Memorial Hospital, the Northwestern Feinberg School of Medicine, the Northwestern Medical Faculty Foundation, and philanthropic contributions of members of the community the ED was able to expand its workforce. Physician coverage which started at 7 shifts per day (56 hours/day) in 2000 rose to 11 shifts per day (85 hours/day) by April of 2004. Similar increases were made in nursing staff. The residency, which has become the cornerstone of our clinical operations, expanded from 7 residents per year to 10 residents per year. At no point has the importance of teamwork been lost, the ED caregivers, physicians, nurses, technicians, administrative assistants, and registration personnel, have all been organized to work in a common physical area to facilitate teamwork and communication. Tighter Interdepartmental Integration Efforts to improve patient throughput depend upon multidisciplinary efforts involving the laboratory, radiology, inpatient services, and other key partners. The laboratory, in particular, has promoted innovative strategies such as incorporation of a laboratory technician into the ED. Such strategies have resulted in notable and measurable improvements. The Department of Radiology has appointed Steven Futterer, MD, as Director of Emergency Radiology. In this role, Dr. Futterer oversees the processes and quality of radiology support directly in the ED. As the number of requests for radiology studies increase, the close partnership assures that patient needs are met with expertise and efficiency. Fuller integration of laboratory and radiology processes will benefit all involved. Measures and monitors of success continue to be studied. Observation Unit The 23-bed observation unit, opened in June 2003, has enabled rapid movement of patients who require less than 24 hours of treatment and testing. This unit is comfortable for the patient yet designed for efficiency. It is managed by ED physician and nurse practitioner staff and supported by specialty services, such as cardiology, which ensure timely testing and expert consultation as needed. Physician Assisted Triage As volume increased in 2004 the department investigated the benefits of placing a physician in triage to assist with initiating patient work-ups while they waited. Unlike other hospitals, who have had some success with physician-assisted triage, we found this to have somewhat limited benefit. There were modest improvements in patient satisfaction with physicians in the waiting room. We did not, however, find significant improvements in patient throughput despite what ended up being a reasonable sizeable increase in physician coverage hours. It was clear, at least in our operational context that those physician hours could be far more productively applied elsewhere. Most alarmingly, we found what almost every other ED that has implemented this strategy has found, that an ever escalating percentage of patients received the entirety of their care in the waiting area – never reaching a patient care room optimally designed to assess and treat a patient. One very significant byproduct of the initiative however did come from the investigation -- a rich set of standing orders were formulated which allow the triage nurses to initiate care in the waiting room if placement in a care space was delayed due to overcrowding. Systems Engineering As 2005 approached it was clear that the pressures of increasing patient volume and decreasing access to inpatient beds due to hospital overcrowding were starting to erode the Departments successes in terms of timeliness of care and Emergency Department length of stay. Most alarmingly it became clear that erosions on this front had worrisome implications not only on patient satisfaction but also patient safety. A series of attempts were made at increasing resources, but ultimately realized only very small margins of return on these investments. Looking back at the historical data, it was clear that improvements in the outcomes of our clinical operations rarely corresponded directly with the addition of resources and was much more tightly correlated with process re-engineering initiatives. At that point, with the assistance of additions to the faculty, we undertook the task of completely re-engineering the flow of patients through the department. From Oct 2004 to Oct 2005 we fundamental restructured the movement of patients through the department and consequently changed the entire culture of our work force. Over that period of time we experienced an 8% increase in volume, a 30% increase in holding time per patient waiting for an inpatient bed, we gained no additional beds, no additional nursing FTEs, no expansion in resident coverage, and we ultimately were able to make a 10% reduction in attending coverage. Despite all that, we achieved a 35% reduction in time to care and set new department records for patient satisfaction. These successes were accomplished by focusing on the following operational agendas:
We ultimately realized we needed to modify organizational structure to achieve the desired culture change:
Organizational Transparency This website forms the foundation upon which we make public our operational successes and failures. It will provide public access to our operational goals and our operational strategies for achieving these goals. It will provide an archive of our daily and monthly operational performance summaries and of our physician productivity. This should provide a platform for sharing our experiences with the broader academic emergency medicine community. Monitoring Performance In order to make operational adjustments to improve the timeliness and quality of care, an emergency department must routinely monitor its own performance. To this end we’ve developed a daily Northwestern Emergency Department Performance Summary that is e-mailed throughout the organization each day to all stakeholders. All of this information provides the foundation to identify trends that require corrective action and allows us to formulate strategic plans for improvement. The following are parameters the department regularly tracks and reports:
Closing the Waiting Room The foundation for all our re-engineering initiatives revolved around one very simple principle, “no patient is better off in the waiting room then in a care area.” After our work with physicians at triage (or in the waiting room) it was clear to us that the right place for a patient was with a physician. What we ultimately came to realize was that the right place for these two to meet was not in the waiting room but in the care area. We changed our focus from trying to initiate and provide care in the waiting room to making sure we did everything possible to assure that every patient went straight back to a room. Patients who, based on their evaluation by primary nurse and physician, did not require extensive monitoring were then moved promptly to the hallway to assure that subsequent patients received the same timeliness of physician initiated care. In order to facilitate this we restructured our nursing assignments to place more nurses in direct patient care assignments, we validated the hallway spaces by labeling them and setting up routine monitoring capabilities. We made every attempt possible to make sure each of our care spaces was flexible so they could be utilized for more or less complex patients, as dictated by volume and distribution of patient acuity. Transforming Urgent Care to Fast Track The realization that low acuity patients are not patients who can wait for care but are in fact patients who should be expeditiously treated and released also lead to substantial operational improvements. The importance of this philosophy grows exponentially as an emergency department’s access to inpatient beds becomes more limited. In our case, with only minimal augmentation of additional resources to streamline the care of our low acuity, quick intervention patients, we have been able to substantially improve the time to care for not only these low acuity patients but high acuity patients as well. Incentive Systems for Improving Physician Productivity We are currently working on ways of increasing attending clinical productivity. Attending productivity unfortunately has remained one of the most variable factors in our clinical operations. Our work with General Electric has highlighted just how important reducing variability can be to achieving high quality. Creating incentives for increased clinical productivity in an academic/teaching institution can often be a far more complicated endeavor than it is in a community emergency department setting. We have begun publishing attending productivity to facilitate peer-wise comparisons and we are in the process of migrating to a clinical obligation model for our attendings that is based on required number of patients seen and RVUs generated rather than simply an hours-worked model. Below is a graph of the attending RVUs per clinical hour of work. Tightening this distribution is one of our primary focuses for the upcoming year. Emergency Department 2 (ED2) As we looked for ways to get an ever-increasing number of patients the emergency care they needed, we started looking at ways of using our department’s space more flexibly. One of our greatest opportunities came in the utilization of bed hours freed by an astoundingly efficient observation unit. Unlike most observations units where the average length of stay approaches 24 hours, our observation unit’s average length of stay is in the 14-15 hour range. This frees up an incredible number of bed hours. What was also interesting to note was that these open bed hours occurred at periods of peak emergency department census. We have been able to successfully fill the bed hours in the observation unit during the middle of the day with young health emergency care patients with medical issues which will take a moderate amount of time to diagnose and treat. Just as the main room of our ED continually takes the sickest patient from the waiting area into its care space, ED2 is designed to take the least sick patient from the waiting room into its care space. This has markedly reduced the tail of our distribution of time care for patients. We piloted this in the early summer of 2005 and have successfully provided primary emergency care to over 1500 patients in our observation unit on the second floor. It has been a huge patient satisfier. By taking the young healthy patients with work-ups/treatments that require a protracted period of time immediately up to our observation unit, we have been able to take the patients who were routinely the most dissatisfied with their emergency care experience and transformed them into the subset of patients most satisfied with their emergency care experience. What we have also learned is that for most patients in the ESI level 3 category, this is a much more appropriate care context than the typical emergency room setting. This discovery has formed the foundation of our expansion plans for our department in the future. What was also interesting to note when looking at our data on the operational performance of the emergency room, as we opened our observation unit, was how little impact an observation unit had on time to care and length of stay in the emergency room. Observation units clearly have benefit to the hospital, but their impact on time-to-care and length-of-stay is relatively small for the relative number of beds allocated. SepsisA landmark paper published in New England Journal of Medicine in 2001 showed that early intensive resuscitation and treatment had an absolute mortality benefit of 16%. Despite compelling data supporting the initiation of Early Goal Directed Therapy Protocols for sepsis, most hospitals have been unable to implement such protocols because of the huge operational infrastructure and interdepartmental collaboration required. With the support of Northwestern Memorial Hospital’s Excellence in Academic Medicine Grant Program, under the leadership of key hospital administration, collaboration with MICU, and Pharmacy, an ED sepsis protocol was launched on Feb 14, 2005 after intensive planning, organizational re-design, and education for nurses, residents, and staff physicians. Initial analysis based on ICD-9 coding data has shown a relative mortality reduction of over 50% for all patients' diagnosed with severe sepsis or septic shock admitted through the ED. Stroke NMH was the first hospital in Chicago to be credentialed by the Joint Commission on Accreditation of Healthcare Organizations as a stroke center. This rigorous evaluation determined that NMH has all the operations features necessary to assure the highest quality of care for stroke patients (both hemorrhagic and ischemic) throughout their course of care. The hospital has established a stroke pager and the Neurology Department’s time-to-bedside in the emergency department for stroke patients has averaged approximately five minutes with no significant outliers. Chest Pain Current efforts are focusing on patients with chest pain in order to reliably assure that 100% of those with acute ischemia are in the cardiac catheterization laboratory in less than 90 minutes. All chest pain patients are benefiting from efforts to streamline the intake and evaluation process. Pneumonia Another critical quality indicator is measuring the time elapsed from patient presentation to administration of antibiotics in those patients presenting with pneumonia. Currently the benchmark is for administration of antibiotics within 4 hours. This ongoing project is meant to assure that the process is free from barriers so the benchmark is reliably met 100% of the time.
Expansion of the Northwestern Emergency Department We are currently in the process of formulating an operational design for our department with a new 23-bed expansion to our existing 23-bed observation unit. With this augmentation of space we should be able to get away from optimizing emergency care in an extremely suboptimal setting (inadequate beds for current volume – requiring caring for spaces in a hallway rather then a room and excessively long waiting for inpatient beds). With these additional 23 beds we will have more care space upstairs in the department than downstairs in the primary emergency care area. This will enable us to generalize the ED2 concept across the entire distribution of ESI level-3 patients. ESI level-4s and level-5s will go immediately over to Fast track for rapid evaluation, treatment and discharge. ESI level-1s and level-2s will be rushed without delay to the main room of the emergency department which will now be freed to provide highly focused emergency care for the critically ill patients. It will no longer be encumbered with the burden of simultaneously attempting to move the masses of intermediately ill patients. The ESI level-3 patients will go upstairs to receive there care in a setting environmentally much more similar to an inpatient setting than an emergency room. They will transition without notice to observation status if their condition dictates or as historical data has shown for 40% of the ESI level-3 patients the department sees, will be converted to an inpatient holding status awaiting an inpatient bed. The inpatient service can, in the intervening placement-time, come down, evaluate and write orders on the patient so that their hospital care can be initiated in a timely fashion.
| | |||||||||||||||||||
| |||||||||||||||||||||
| |||||||||||||||||||||
