National Emergency Care Enterprise: Advancing Care Through Collaboration: Workshop Summary (2009)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

3 Quality and Patient Safety Chapters 3–6 capture the discussions from four panel sessions. Each focuses on a central theme in emergency care. The first of these sessions, on quality and patient safety, was chaired by Dr. Michael Rapp, director of the quality measurement and health assessment group at the Centers for Medicare & Medicaid Services (CMS). Rapp said that quality measurement is a tool the federal government has applied broadly to assess health care. It is also the basis for strategies involving pay for performance, including CMS’s plan to pay more for higher quality dialysis services within the next 2 years. In addi- tion, CMS has used quality measures to implement a five-star rating system for nursing homes. Rapp encouraged the audience to consider how this may be used to address additional quality challenges. Improving ED Crowding and Patient Flow Ron Anderson, president and chief executive officer (CEO) of Parkland Health & Hospital System in Dallas, began the panel discussion by describ- ing a comprehensive effort Parkland had undertaken to address emergency department (ED) crowding and patient flow. He observed that instigating a cultural change at the hospital had been the most important ingredient for success. Other floors of the hospital had had an “out of sight, out of mind” mindset regarding ED crowding, so Parkland implemented a multi- disciplinary solution that instituted shared ownership of ED outcomes. This involved revised performance indicators and quality management as well as direct financial incentives. The hospital gave its ED faculty the right to admit patients upstairs, 25 

26 NATIONAL EMERGENCY CARE ENTERPRISE and to “push people upstairs if necessary.” They also created a red and y ­ ellow alert system to expedite discharges. In addition, Parkland established 11 community clinics to provide medical homes for patients and encourage them to seek primary care in places other than the ED. This has resulted in decreased ED use and has saved hundreds of thousands of dollars in ­neonatal intensive care unit costs as a result of increased prenatal care, Anderson said. However, the clinics themselves still lose $2 million per year per clinic. To further reduce workload within the ED, the hospital incentivized radiology, pathology, consultative services, “and everybody else” to stop working up patients there. This had become common practice because it is easier to obtain diagnostic tests in the ED than in other parts of the hospital. Parkland also adopted a “pod system” from Detroit Receiving Hospital to increase productivity. Faculty M.D. supervisory time was increased from 68 to 96 hours/day. The hospital’s goal is to reduce dwell times for patients going home to 4 hours, and 8 hours for those who are admitted. However, Anderson said they would need additional capacity, probably 200 more beds, in order to achieve the latter objective. Anderson characterized Parkland’s reform effort as systematic and multi­ dimensional. It did not just involve the ED staff, because “[ED crowding] has got to be owned by everybody.” The key is to have the right expecta- tions, set them from the top, and then hold people accountable, including top executives. These tools have allowed them to change what they believed was a very good ED into an even better ED. Regionalization and Quality Health Care Susan Nedza, vice president of clinical quality and patient safety at the American Medical Association (AMA), said the purpose of regionalization is to improve the quality, safety, and efficiency of care provided in communi- ties. However, this approach has its trade-offs. Regionalization is typically framed in terms of specific disease states, such as stroke. However, focusing only on stroke patients—or on trauma patients—means that other types of patients, such as the elderly woman with undifferentiated disease sitting in the ED waiting room, will not receive the same level of attention. Nedza said we need to think more broadly in terms of community health priorities because these are sometimes different than acute health priorities. Nedza observed that acute care providers typically focus on individual patients when making decisions about resource allocation. But she said there are also broader considerations that involve the health of the whole community. For example, ambulance bypass protocols that direct heart patients to select hospitals may improve quality for individual patients (for those patients who reach the hospital), but have a negative impact on the 

QUALITY AND PATIENT SAFETY 27 surrounding hospitals, which may see a drop in admissions and revenue. As volume drops in these facilities, patients in the future may not have access to adequate infrastructure to meet their needs. As measurement organizations have created—and CMS has adopted— performance indicators to measure door-to-needle times in select tertiary care hospitals, community hospitals with limited resources have less incen- tive to handle patients who walk in with an acute myocardial infarction and other measured conditions. Nedza observed that patients are sometimes transferred because there is no on-call M.D. to help them at the transferring hospital. During her tenure as a regional chief medical officer for CMS, she encountered instances when on-call services were tenuous. In downstate Illinois, for example, she visited an area where two neurologists, both nearing retirement age, were covering an extended geographic area. At one facility in Northern Wisconsin, one invasive cardiologist working on a visa was serving a multicounty area. In addition, the physician workforce is making career choices to become more subspecialized. Ultimately, this will decrease the availability of needed spe- cialties in many communities (see Chapter 5). To improve access, issues of physician supply and demand must be recognized and addressed, she said, adding that these must be reflected in quality measures and incentives. Trauma System Evaluations John Fildes, chair of the American College of Surgeons (ACS) Commit- tee on Trauma, discussed the U.S. trauma system. He said trauma systems follow a disease-specific model that includes everything from public educa- tion and prevention to access to emergency and resuscitative care, operative care, critical care, in-patient care, and on through either long-term care or rehabilitation, then return to home, work, and family. The trauma model, he said, has been a successful example of regionalized care. It has become a model for treatment of other diseases, such as stroke and acute myocardial infarction. Fildes said that the vast majority of Americans believe that if they are severely injured, they will be taken to a designated Level One trauma center within 20 minutes of their home. However, only 80 percent of the United States is within an hour of a trauma system. The ACS Committee on Trauma has helped to foster the development of trauma systems across the country. Although there are examples of exceptional systems, great dispari- ties among states remain with respect to the laws on the books, what kind of staffing is present, and what kind of authority is in place. Strong leadership at the state level is essential, he asserted. Fildes reported that an ACS-designated trauma systems committee is starting to map the country by visiting states and conducting trauma systems evaluations. Evaluation teams assess whether facilities have the resources 

28 NATIONAL EMERGENCY CARE ENTERPRISE they need to provide adequate care, and whether patients receive the care they should. If they do not, facilities do not get verified, which means they are not allowed to be part of the trauma system. Fildes said trauma systems do not have to undergo evaluation every 3 or 4 years, like most other care processes do. In some states, verification is a one-time event. In recent years a sophisticated metrics-based approach to monitoring care has been under development. Fildes helped to create the system that allows the National EMS Information System (NEMSIS) and the national trauma data standard to interact. Commercial products using wireless technology can now transmit emergency medical services (EMS) data to trauma centers, allowing trauma center staff to begin populating an acute care record before the patient arrives. The system also allows EMS providers to learn their patient’s outcome. The data are aggregated at the national level using a 50-element data- set, Fildes said. This system has collected millions of patient records for those treated in U.S. trauma centers. The dataset captures injuries, services received, and cost of care. This information has been used to support the trauma quality improvement project, which provides risk-adjusted com- parisons of patient outcomes at individual centers and within various cities, counties, states, and the country as a whole. Standardizing Quality in EMS Dia Gainor, bureau chief of EMS for the state of Idaho and past presi- dent of the National Association of State Emergency Medical Services Offi- cials, said there is no universal culture of quality in EMS. She acknowledged there have been various projects and initiatives in this area, most notably the NEMSIS project which “catapults us into a new way of thinking about capturing and using data.” But, she said, the fact is that “when we move from one state to the next, or from one local EMS agency to the next, there is really no standard of quality that you can find that is pervasive throughout the system.” Adopting private-sector techniques to improve quality and pro- ductivity, such as those used in the manufacturing sector, would be helpful, Gainor said. She said that these strategies should be adopted on a wholesale basis by EMS, and perhaps by emergency care as a whole. Federal initiatives to assist states and localities have been important, helpful, well timed, and modestly funded in some cases, Gainor noted. But they rarely address the highest priority topic in any given state. Moreover, the various federal activi- ties are not always in sync with each other. As a result, the federal partners provide a somewhat tattered patchwork to the states. Gainor said the first priority should be the safety of those providing care or transportation. The safety of the patient is a close second. “I think that we have largely failed, especially in the prehospital sector, to put the safety of EMS personnel first and foremost,” she remarked. 

QUALITY AND PATIENT SAFETY 29 States have a significant and established role in the direct regulation of prehospital and, often, interfacility transport, Gainor continued, going back nearly 40 years in some cases. However, we have not moved ourselves away from quality checks that involve mere checklists of equipment and a head count of licensed personnel. EMS needs to adopt methods to ensure that quality is inherent in every practice that is undertaken by every individual EMS provider and agency, she said. Emergency Care as a Health System Barometer Charlotte Yeh, chief medical officer of AARP Services Inc. and former regional administrator for CMS, believes that emergency care is a barometer for the overall effectiveness of the health care system. “We are the window into whether the health care system is working or not,” Yeh said. She asserted that whether one is focused on pharmacological adherence, care coordination, care management, caregiver support, trauma, fall prevention, or any other topic in health care, the emergency department has the data within it to indicate how well the health care system is performing. Conse- quently, the ED should be seen not just as the safety net for the health care system, but its vital signs. Yeh emphasized that seniors make up a sizable proportion of the patients seen in emergency departments. For this reason, Yeh emphasized, they are impacted more than any other population group by boarding and overcrowding. Nationally, 46 percent of all emergency department admis- sions are covered by Medicare admissions. In Massachusetts, the figure is 56 percent. In Massachusetts, Yeh said, if you look at the top 25 diagnosis-related groups in terms of margin, they are all elective admissions that involve proce- dures such as knee replacements, hip replacements, and cardiac interventions. They do not enter the hospital through the ED. This is why, she believes, hos- pitals have no financial incentive to resolve ED overcrowding because direct admissions compete for beds and pay a higher profit margin (see Chapter 6). She cited a study that showed that when hospitals go on diversion, their net revenue increases. Yeh reported that, since January, Massachusetts has completely banned ambulance diversion statewide. The consensus is that the policy is working. Hospitals have stood up to the challenge and realized ED crowding is a problem for the entire hospital, not just the ED. Yeh’s presentation also highlighted the impact that mental health patients have on hospital emergency departments. In addition to seniors, mental health patients are “the other silent story,” she said. In her experience with Massachusetts hospitals, although only 5 percent of ED patients have primarily mental health problems, they represent 30 percent of occupancy in the emergency department. The biggest reason is that few hospitals have 

30 NATIONAL EMERGENCY CARE ENTERPRISE sufficient in-patient beds for mental health patients and the few regional centers are full. So if a patient is too risky to discharge, but no bed is avail- able, he or she may be held in the ED for days at a time. As a result, mental health patients’ length of stay in the ED is about twice the mean of any other diagnoses. This is especially relevant for hospitals now that mental health parity is being implemented. What is Quality? Dr. Ramon Johnson, an emergency physician practicing in Southern California, said that after 25 years of practicing emergency medicine, he’s still not exactly sure what quality is. But, he said, “You know it when you see it.” He likes to measure quality by what he calls the Aunt Bessy Test: What would be best for Aunt Bessy? For example, it’s probably better not to make her sit in the ED for hours on end, waiting for an in-patient bed. Johnson said that, when measuring quality, we tend to focus on mea- sures such as whether antibiotics were prescribed within 4 hours to patients with pneumonia. That focus, he said, drives emergency care providers to order chest X-rays and start antibiotics on everybody who comes to the triage desk. This has driven up the cost of care for patients—even for those who didn’t need it, he said. In some cases we will find the diagnosis of pneumonia in patients who never had pneumonia, just because we are trying to meet a quality benchmark. Enacting a statewide ban on boarding more than 4 hours in the ED would be the best way to improve quality, Johnson said. Finding Quality by Connecting the Links of the Chain The AMA’s Nedza described a cultural issue surrounding quality mea- surement that must be overcome: Medical providers often do not want to be measured on what happens to a patient after the patient has left their control. Efforts to design quality measures for emergency medicine have at times been hampered by well-founded concerns about not being the one who is responsible for what happens to patients after they leave the ED, she said. However, emergency medicine is practiced by teams, and measure- ment needs to be team- or community-based. “The connectivity for quality has to be outside of our department,” she said. Taking an example from her own career, she noted, “I thought quality care for an AAA [abdominal aortic aneurysm] was to get into the operating room.” But later she found out that the majority of patients who had an AAA rupture outside the r ­ etroperitoneum did not survive to discharge or did not return to health or independence. This raised the difficult issue of how to define quality across 

QUALITY AND PATIENT SAFETY 31 an episode of care and how to hold everyone responsible for outcomes. She concluded that we need better ways to link up what happens to patients after the emergent care has been provided in order to know what constitutes quality care. Nedza quoted Arthur Kellermann from the previous session, who said (jokingly), one thing a physician never wants to be asked is, “Do you remember that patient you saw yesterday?” She argued that we have to own the care we provide in the emergency department and link it to what happens the next day. But this is not typically what happens. “I taught residents for years,” she said. “I can tell you, very few times did I ever take them by the hand and say, let’s go upstairs and find out what happened to that patient we took care of yesterday.” Roger Lewis of Harbor–UCLA Medical Center observed that “many of the quality measures that we have been exposed to, or been beaten over the head with, have been based on single diseases.” However, these types of measures invite gaming of the system. For example, “in some triage areas, it has been suggested to hand everybody a tab of an antibiotic, and then if they turn out to have pneumonia, you look good.” The only way to avoid this problem, Lewis said, is to come up with quality measures that apply across broad classes of patients and disease settings. We need to look for multivariate, broad measures of what Aunt Bessy experiences, regardless of her disease, he said. Legislating Quality Michael Rapp said that Johnson’s suggestion of going to Capitol Hill and lobbying legislators to outlaw boarding is not the only alternative. The executive branch also has the means to promote this objective. For example, CMS could establish a performance measure to track whether or not hospitals board, or the extent to which boarding exceeds 4 hours. CMS already has a pay-for-reporting program for hospitals. If they do not report the quality measures CMS requires, their annual payment update is reduced by 2 percent. As a result, virtually all hospitals provide these reports. If a boarding measure were constructed and endorsed by health care stakeholders through the National Quality Forum, CMS could imple- ment the benchmarks. AARP’s Yeh agreed that outright bans are not the only regulatory solu- tion. Others include aligning financial incentives and public reporting. In addition, she said, it is important to recognize that implementing strict rules will not succeed unless technical assistance is provided to facilities. One reason the ambulance diversion ban has worked in Massachusetts, she said, is because of all the work that accompanied it, including trial runs, discus- sion groups, and dissemination of best practices. In Massachusetts, Yeh said, 

32 NATIONAL EMERGENCY CARE ENTERPRISE there has been a constant feedback loop that has allowed people to share stories and capture best practices. The hospital CEOs have been at the table. This has transformed how they look at their EDs. When the diversion ban went into effect, she said, “people were ready, they were prepared.” Bundling Payments to Improve Provider Linkages The panel also discussed the potential for CMS to provide bundled pay- ments in order to align financial incentives and promote better care. Some have proposed that Medicare provide bundle payments to hospitals, which could then manage the payments to care providers. Some have argued that payments could be structured in a way that aligns incentives toward better patient outcomes. The objective would be to promote a less fragmented team approach among hospitals, physicians, and postacute providers and improve problem areas, such as hospital throughput. Anderson agreed that to improve patient outcomes, “We have to all be aligned to the same incentives.” Currently, financial outcomes for emergency medicine physicians and trauma surgeons are not matched up with those of the hospital, because they do not receive Medicaid disproportionate share payments or upper payment limit reimbursements. He acknowledged that the hospitals need to take better care of their doctors financially. However, a bundled payment arrangement can be problematic for a number of ­reasons. For example, it runs up against the current payment system, in which greater utilization means greater payment. Also, he said, it can be difficult for the hospital to align interests with physicians if the latter have purchased an equity position in a competitor hospital across the street, or own their own ambulatory surgery center. “That is a little bit difficult to bundle,” he said. In addition, bundled payment systems require big investments in informa- tion systems. Yeh cautioned that payouts from the payment bundle should be risk adjusted, not just by the patient’s clinical conditions and comorbidities, but also by socioeconomic factors, literacy, and language. Homeless patients require a higher level of resources than other patients. If we create a uniform bundling based on simple measures such as clinical condition, Yen pointed out, it may worsen disparities of care. Patchwork Regulatory Systems David Thompson, an emergency physician from Syracuse, New York, observed that many states seem to have “system creation” as a part of their regulatory function. He said we seem to be coming up with a patchwork of systems where the states have their own regulatory quality initiatives, the fed- eral government has its initiatives, and in the end, there is no system at all. 

QUALITY AND PATIENT SAFETY 33 Nedza described her experience with the trauma system in Illinois, where many community service areas cross jurisdictional boundaries. They decided to look at the preexisting referral patterns, rather than dictating them to the various centers. This produced informal models of regionaliza- tion that transcended traditional federal–state barriers. Yeh reminded the audience that health care is local. That is why not everybody operates under one system. Nationalizing the system would not allow local perspectives and local solutions, and she said there must be a balance between national consistency and local flexibility. Where systems exist, she said, we should identify the criteria where local input should be permissible versus areas that require national standardization. Gainor pointed out the fragmentation can exist within a single state. Idaho has a state office that regulates the EMS system, a bureau of facility standards that licenses hospitals, and its state Medicaid program. She was not able to think of any occasion on which all three agencies met to talk about emergency care overall or how to improve the regulatory environment. Nor, she said, have they ever been asked to do so by federal partners. Thompson said the problem is even more complicated because of cross- state interactions. He asked Gainor how Idaho interacts with the analo- gous agencies in Washington or Montana, pointing out that patients move through all of these areas. He asked if the states are going to regulate, how we can integrate all of these functions. He agreed with Yeh that there is a need for some federal mandates, but that local variation and experimenta- tion is also needed. Gainor noted that there are a number of ways to ensure interstate integration and cooperation. For example, EMS systems have used interstate compacts, so that licensure of an ambulance service is recognized at the state line, and there is no additional red tape upon repatriation. Recommendations to Federal Partners In the United Kingdom, Kellermann observed, the National Health Service considers an overcrowded emergency department an institutional failure, and management suffers the consequences. In the United States, an overcrowded ED is considered a failure of the ED staff and patients suffer the consequences. CMS’s Rapp noted that, for every hospital in the United Kingdom, the National Health Service posts on a website the percentage of the time the hospital meets the 4-hour waiting standard. Rapp then asked the panelists to provide their recommendations to the U.S. federal partners on what they should do to improve the quality and patient safety of emergency care in the United States. Nedza said that, on a practical level, we need to standardize and har- monize measures, measurements, and measurement priorities. The federal 

34 NATIONAL EMERGENCY CARE ENTERPRISE government should provide national frameworks, perhaps standards, and also funding where appropriate, but local solutions and sustainable regional models are needed. A key issue, she said, is that measurement should follow the patient. It should not be based on who provides your insurance coverage, or which federal program is receiving the data. Yeh provided three recommendations to the federal government to improve quality and patient safety in emergency care. First, she said that focusing on Medicare patients will provide significant opportunities for cost savings and quality improvement. Second, she recommended that research be conducted to examine whether “carve-outs” for mental health have actually increased the cost of medical care. Third, she recommended that a national ban on ambulance diversion be considered. Instituting this ban, she said, would make everybody work together—prehospital personnel, ED personnel, in-patient staff, and hospital administration. “The most striking thing was we had to work together if we were going to fix this,” Yeh said. It created a collaborative environment and fostered a cross-disciplinary, consensus approach. Gainor advocated a federal platform for interagency planning and implementation. The agency or group that performs this function, such as the Federal Interagency Committee on Emergency Medical Services, must be appropriately resourced to become a true lead agency. The buck should stop there. Gainor noted that a number of models demonstrate federally promoted, state-led, and locally implemented solutions, often involving industry. For example, the U.S. Department of Transportation (and agencies such as the Federal Aviation Administration and the Federal Railroad Administration) has helped to ensure that the stop signs in rural Idaho look exactly the same as those in Boston. EMS needs to establish a nationwide culture of consistency and quality. Fildes said top-down leadership is needed from the federal government to address the terrible disparity among and between cities, counties, and states in their trauma systems. Systems currently range from exceptionally well-developed to absolutely nothing. “When I drive from New York to Los Angeles, 80 percent of the time I am not in a coverage area.” What is needed is a national trauma system to provide federal oversight to ensure that the level across all 50 states is consistent and essentially the same. Parkland’s Anderson said that right now, providing trauma care is a bad business decision. He said we need to take away the financial disincentives for providing it. The federal government can help level the playing field and reward those who actually make emergency and trauma care work. In these systems, readiness costs are substantial, whether or not there are patients who require care. So trauma and emergency care are not like other commodities; they should be treated more like a public health utility model, 

QUALITY AND PATIENT SAFETY 35 Anderson said. He noted that most utilities are planned. People sit down and they plan across county lines, across state lines, and across jurisdic- tions. More planning is necessary to ensure the public’s health. Some states, including Texas, have done a great job in developing their trauma systems, but even in Texas, 80 percent of the land mass doesn’t have trauma system coverage. “We literally call these dead zones,” he said. Instead, there should be funding for regional planning through states. “We also have to realize that some places will never have trauma hospitals,” he said. “We need to plan for transporting people in from rural areas, and ensure the availability of telemedicine to stabilize patients until they can reach definitive care.” Anderson believes we must confront the fact that subspecialists keep migrating away from emergency and trauma care. “Many trauma surgeons are aging out,” he asserted. “And other specialists want to go into sub­ specialty care so they can better control their likelihood of being sued, their likelihood of being paid, and their fee schedule. This is true for orthopedic surgeons and neurosurgeons and others who support the trauma system.” Changes in how physicians are trained may be necessary. Finally, Sandra Schneider, vice president of the American College of Emergency Physicians, noted that, in a survey of a thousand emergency physicians, emergency department crowding ranked as their top patient safety issue. Crowding has been associated with delays in care, increased morbidity, and increased mortality. “There is consensus among emergency physicians that this is their number one issue. Anything that can be done to improve this will be better for our patients,” Schneider said.