点击显示 收起
医院医学学会2005年会
Society of Hospital Medicine (SHM) 2005 Annual Meeting
2005年4月28－30日
美国伊利诺斯州芝加哥
April 28 - 30, 2005, Chicago, Illinois The State of Hospital Medicine
Joseph Li, MD
Introduction
The Society of Hospital Medicine (SHM) Annual Meeting in Chicago, Illinois, marked the 8th annual meeting of a society that has grown to nearly 5000 members. There are now approximately 12,000 hospitalists in the United States. Within the next 5 years, hospital medicine may be the largest medical specialty in the United States. The term "hospitalist" was coined by Dr. Robert Wachter in 1998, making this trend nothing short of remarkable. We are witnessing a paradigm shift in who is caring for America's hospitalized patients.
Despite the growing numbers, most Americans and many healthcare providers remain ignorant of the hospitalist's role. Two years ago, the SHM Education Committee, chaired by Dr. Alpesh Amin, University of California at Irvine, established an agenda to define the role of a hospitalist. In January 2006, SHM will publish the first "Core Curriculum for Hospital Medicine." Dr. Amin and his colleagues described the core curriculum at the annual meeting. It will contain 3 sections: 1) clinical conditions, 2) systems, and 3) procedures, and each section will address 3 distinct competency domains: A) cognitive (knowledge), B) affective (skills), and C) psychomotor (attitudes). With this core curriculum, SHM will define the hospitalists' core competencies.
SHM will publish this curriculum as a supplement to the inaugural issue of The Journal of Hospital Medicine (JHM). Since SHM's inception, The Hospitalist has been its sole continuous publication. At the annual meeting, SHM announced plans to transform The Hospitalist, an amalgam of newsletter, practice profiles, and clinical updates, into a society newspaper. While The Hospitalist will remain the voice of hospital medicine, JHM will become the journal for original research related to hospital medicine. Like other medical journals, JHM will be peer-reviewed. Mark Williams, MD, Director of the Hospital Medicine Unit and Associate Professor of Medicine at Emory University, will be the Senior Editor and is currently seeking submissions for the inaugural edition of the journal, scheduled for January 2006. Meanwhile, James Pile, MD, a hospitalist at Cleveland Clinic Foundation, is resigning as Editor of The Hospitalist. SHM has announced that it plans to restructure the editorial leadership of this publication. In addition to a Staff Editor, SHM is currently seeking a new Physician Editor to replace Dr. Pile.
Despite the many changes at SHM, some things stay the same. Dr. Robert Wachter (Figure 1), Associate Chair, Department of Medicine and Professor of Medicine, University of California at San Francisco, gave another provocative annual address on the state of hospital medicine. Dr. Wachter titled his talk, "Hospital Medicine 9 Years Later...Still Crazy After All These Years."
Dr. Wachter contrasted the external forces driving growth in hospital medicine in 1998 to those affecting its growth today. He postulated what factors will be the driving force in the field in 2010.
In 1998, cost pressures were pervasive in the healthcare marketplace, and there was a demand for physicians with expertise in the care of hospitalized patients. He described cost as the main driver behind the development of hospitalist programs at that time. [Made change because it reads as if cost pressure was the only factor then, right?] Quality improvement and patient safety were only beginning to be recognized as major issues.
Today, hospitals are filled with patients, and administrators are looking for ways to improve "through-put." Primary care physicians are retreating from the hospital setting, leaving the care of hospitalized patients to hospitalists. At the same time, resident physician work-hour restrictions have forced teaching hospitals to consider hospitalists as an alternative staffing solution. Last, but not least, the mandate to improve quality and safety is now front and center in American healthcare thought and dialogue. Dr. Wachter believes these are the main reasons for the continued growth in hospital medicine today.
Dr. Wachter listed the 3 reasons why hospital administrators and CEOs looking for healthcare value will continue to rely on hospitalists: nurse retention/shortage, hospital efficiency, and resident physician work-hour limits.
Nurse retention/shortage. Wachter quoted statistics that cite a shortage of 126,000 nurses in the American healthcare system today. With one third of nurses older than 50 years of age, this shortage may get worse before it gets better. With poor doctor-nurse relationships as a major cause of discontent among nurses, he believes hospitalists will serve an important role in improving nurse retention rates in hospitals.
Hospital efficiency. It is common for a hospital staff to strive for a 90% occupancy rate. As Wachter states, "in the full house, the surgeon is king," meaning that hospital administrators prefer to care for patients with surgical rather than medical needs because the margin of profit is higher in the former group of patients. Without evolving into a specialty facility to eliminate medical patients, the only way hospital administrators can make beds available for surgical patients is to provide efficient care and minimize the medical patients' length of stay. Wachter notes that no one is better suited for this role than the hospitalist. However, he adds that hospitalist relationships with surgeons are not risk-free for the hospitalists, citing concerns of liability surrounding co-management of surgical patients as well as "dumping." He advised hospitalists to "proceed cautiously but to welcome the surgeons' interest."
Resident physician work-hour limits. The Accreditation Council for Graduate Medical Education (ACGME) work-hour regulation is achieving 1 of its goals: limiting the number of hours resident physicians are present in teaching hospitals. As a result, Wachter notes, many hospitals have turned to hospitalists to lead nonteaching services in traditional teaching hospitals.
Dr. Tom Baudendistal, a hospitalist at California-Pacific Medical Center, and Dr. Jeff Wiese, Chief of Medicine at the Medical Center of Louisiana at New Orleans and Charity Hospital, led a session about this topic at the annual meeting. They explained that hospital staffs are turning to hospitalists for 2 main reasons:
Hospitalists champion efficient care.
No other group of physicians, including "ambulists" or medical subspecialists, are as ideally suited to this role.
Wachter believes that the trend of relying on hospitalists, rather than resident physicians, for inpatient care in hospitals will continue. The ACGME work-hour regulation has also had unintended consequences -- for example, causing fragmentation of care. Wachter states that medical housestaff average 15 handoffs during a 5-day hospitalization. He notes that even in the traditional teaching service, the hospitalist may be the only physician who has a longitudinal perspective of the patient's hospital stay.
The Executive Board of the SHM shares this view. In May, they published a supplement to The Hospitalist, titled "How Hospitalists Add Value." Articles in this supplement include: "Providing Extraordinary Availability," "Maximizing Throughput and Improving Patient Flow," and "Improving Patient Safety and Quality of Care." Dr. Larry Wellikson, CEO of the SHM, clearly states that the supplement is targeted not only to hospitalists but also to the healthcare community. He said, "The success of hospitalist programs is just as dependent on the development of an external support network as it is on the existence of a strong internal infrastructure."
With increasing demand for hospitalists despite a relatively fixed supply of physicians, hospitalists often find themselves working many more hours than expected. An increasing number are describing themselves as "burned out." Dr. Michael Anthony Williams, President of Inpatient Services, PC, in Denver, CO, led a session titled, "Real Life, Real Answers." He believes that the nature of the job itself, which involves the following factors, leads to stressors that place hospitalists at risk for burnout:
He stressed that although change is inevitable, burnout is not. He described the following as major factors in "burnout":
Values. Individual values often differ among members of a hospitalist group. Dr. Williams emphasized that it is important to clearly understand individual and group values. To minimize burnout, he advises groups to (1) define values, (2) align goals with values, (3) stay consistent with the goals and values, and (4) recruit like-minded people.
Autonomy and control. He urged individuals and hospital groups to learn who makes the decisions. Is it the group leader? The hospital? The payer? To minimize burnout, he advised decision leaders to maximize individual autonomy while balancing individual interests with those of the group.
Communications and expectations. Dr. Williams stressed that it is important to define expectations and maintain clear lines of communication. Effective teamwork depends on constant communication.
Ownership. He advised groups to value individuals and allow them to "own" projects and plans. He emphasized that it is important to look for areas of ownership, as they come in many forms.
Stability and predictability. He described the daily practice of hospital medicine as "inherently unpredictable." But setting and adhering to schedules can provide the group and individuals with control.
Support systems. He advised groups to establish common goals, emphasize teamwork, create backup plans, hold regular meetings, and encourage feedback.
Imagine that the CEO of a Fortune 500 company has selected you to help design a strategic vision that will both improve the quality of care administered to its employees and reign in its spiraling cost. This was the challenge put forth by Dr. Arnold Milstein, the medical director of the Pacific Business Group on Health and cofounder of the Leap Frog Group for Patient Safety, at the Society of Hospital Medicine (SHM) 2005 annual meeting.
In his talk, "Outswimming the Shark: How Purchasers Will Make Provider Excellence in Cost Efficiency and Quality a Matter of Business Survival," Dr. Milstein stated, "The first step in tackling any problem is to define its scope." In recent years, employee healthcare expenditure was the number 1 problem facing Fortune 500 CEOs. Studies by the Rand Corporation, the Institute of Healthcare Improvement, and Harvard and Dartmouth Medical Schools suggest that up to 40% of healthcare dollars are wasted. If best practice and resource efficiency models were applied nationwide, it would be possible to cut 30% of all healthcare expenditures without a change in health care quality, patient health status, or patient perception about the service and quality they receive.
Dr. Milstein asked how we could achieve this goal. Cutting provider reimbursements would be unpopular and require annual battles. Trying to raise consumer awareness about costs by increasing deductibles would also be problematic -- the public already feels overwhelmed by healthcare premiums, and shifting further costs to them would exacerbate this problem.
Dr. Milstein suggested an alternate strategy. First, he proposed that physicians look for ways to reduce the intensity of services, especially for "flat-of-the-curve" spending. This term refers to the practice of spending increasing amounts of money to realize only incremental improvements in prognosis or quality of life -- a phenomenon that may account for as much 30% of healthcare costs. Second, perhaps we can improve quality and cost by trying to understand the relationship between physician activities, their costs, and their ultimate impact on patient outcomes.
Dr. Milstein suggested looking at all providers in the healthcare network and developing measures to see how well they track best practice and evidence-based medicine. How well they utilize clinical resources should also be measured. Ideally, a physician would do well with both measures, but as we know from aggregate data, this is not always the case.
Once we have a better sense of the problem, we need a way to measure it. A useful model might be to define healthcare delivery in terms of a "care unit." This is a measure of an individual doctor's quality index (outcomes or adherence to evidence-based medicine) and longitudinal cost-efficiency index (total cost per case-mix adjusted per treatment episode). In this way, best practice would be more clearly quantified as medical care that finds the optimal balance between quality and longitudinal cost efficiency, which, in turn, provides the framework for continuous fine-tuning, application of information technology, and pay-for-performance incentives.
If this model sounds like an attempt to industrially engineer healthcare, in many ways it is, Dr. Milstein suggested. However, doctors have traditionally resisted such a notion -- many will ask, isn't medicine an art? How can you apply these types of industrial practices to healthcare, which is really about the doctor-patient relationship? Indeed, if executed effectively, this process may actually be the best way to improve quality and safety and avail resources to make care more accessible. Moreover, Dr. Milstein points out that this model is actually identical to a vision put forth by the Institute of Medicine in their 2001 health care quality initiative report titled Crossing the Quality Chasm,[1] the most widely quoted position statement on the future of healthcare in America.
In closing, Dr. Milstein charged hospitalists with a unique responsibility in this transformation, both as physicians in hospitals and administrators with expertise in hospital systems. Future directions include reengineering of hospitalist services by means of information technology and a broader scope of hospitalist services to include effective clinical resource utilization as well as posthospital management of high-risk (for clinical and/or financial outcomes) patients.
President George Bush has called this the decade of healthcare information technology, and he has appealed for widespread adoption of electronic health records to improve quality and contain costs. Many people are unaware of how far behind the curve healthcare is when it comes to integrating information technology into daily practice. It is difficult to imagine any other business sector of such size, expense, and importance that would rely on technology that has not changed in decades. Fewer than 5% of hospitals have implemented a computerized provider order entry system (CPOE), which allows providers to order medications using a computer, thereby reducing transcription errors and delays. Even fewer hospitals have electronic medical records, which allow instantaneous access to critical patient data.
Few would dispute the fact that physicians care deeply about their mission and their commitment to their patients. Yet despite their physicians' best efforts, patients continue to suffer the effects of medical errors and lack of adherence to best practice at the cost of life and limb. Perhaps "trying harder doesn't work," suggested Peter Lindenauer, MD, MSc, a hospitalist at Baystate Medical Center in Massachusetts, and Assistant Professor of Medicine at Tufts Medical School. In his talk, "The Role of Information Technology in Quality Improvement and Patient Safety," he quoted the revered management consultant W. Edward Demings, who famously said, "Put a good person in a bad system, and the system wins, no contest." This is where we are in 2005 when it comes to quality and patient safety.
The practice of medicine is time-sensitive and highly technical. It has experienced an explosion in the complexity and volume of information that must be mastered and easily retrieved at the point of care. The traditional models of practice, which rely on education, memory, and vigilance, may be good for roundsmanship but do little to foster optimal care and patient safety. This was the core message of Dr. Peter Lindenauer and his co-presenters, Andrew Karson, MD, MPH, from the Massachusetts General Hospital in Massachusetts and Joshua Lee, MD, from the University of California at San Diego, in their presentation on the emerging role of information technology in quality improvement and patient safety.
The promise of healthcare information technology is manifold and includes the following:
Preventing medication errors through accurate dosing on the basis of weight, age, and kidney function.
Allergy cross-checks and identifying drug-to-drug interactions.
Embedding decision support with best practice guidelines.
Warnings about abnormal lab results, which require immediate action or follow-up.
Tracking errors and adverse events and providing feedback to improve patient safety.
Imagine a physician beginning an admission order set on the computer, ordering strict bed rest for a patient. The computer sees the order and automatically responds by asking the physician to choose from among a list of approved medications that decrease the frequency of venous blood clots -- a major cause of death among immobilized patients. This function is actually built into the CPOE system used at Massachusetts General Hospital and was demonstrated by Dr. Karson during his presentation. We can all imagine how important such a safety check would be to a busy physician who may be distracted by another critically ill patient.
If these systems are so beneficial, why are so few physicians using them? Dr. Lindenauer suggests that the answer is complicated and related to economic, technical, and cultural factors. While it is true that many tertiary care centers have successfully transitioned to CPOE systems, they benefit from economies of scale in ways that other centers do not. In particular, the upfront expenditure of implementing a CPOE system is large, but the cost for each additional user is relative small. Moreover, because many of these systems are homegrown or purchased from proprietary third-party vendors, hospitals and physician practices may face problems with interoperability. Integrating computers into a physician's workflow is also a departure from traditional practice and as such is not universally welcomed. Finally, Dr. Lee cautioned that while these systems show great promise, their deployment is complicated, and may result in unintended consequences. While initial experiences with these systems are mainly positive, results are inconclusive. Dr. Lee emphasized the need for further multicenter studies that focus on vendor deployment of CPOE/electronic health record products.
Nevertheless, all 3 discussants agreed that it is not a matter of if these systems will be adopted, but when, as there is now significant pressure from the government, private sector, and increasingly the public to utilize these technologies. Healthcare is on an inexorable track toward transparency and accountability, and information technology is a tool we cannot afford to squander.
Along with tangible benefits, the hospital medicine movement has brought unwelcome problems. In the traditional model of care, the primary care physician would see the patient in the office, hospital, and rehabilitation center if needed. Now, patients are "handed off" over time and seen by multiple providers. Patients often become frustrated with having many so providers who practice in multiple settings and often do not communicate effectively with each other. Even worse than their frustration, critical information may be lost in translation, leaving patients with inadequate instruction, inadequate follow-up, and a limited understanding of the medications they are supposed to take.
Several research abstracts presented at the SHM 2005 were aimed at identifying and remedying these problems.
In an abstract titled, "Post-hospital Medication Discrepancies: Prevalence, Type, and Contributing Factors," Eric Coleman, MD, and colleagues from the University of Colorado Health Science Center at Denver, comprehensively assessed medications of 375 patients, aged 65 years of age or older, 24 to 72 hours after the patient was discharged from the hospital. They found that 14.1% of patients experienced 1 or more medication discrepancies, 51% of which were attributable to errors at the patient level and 49% of which were attributable to the system. An even greater concern was that patients with medication discrepancies were almost 2.5 times more likely to be readmitted to the hospital within 30 days of discharge.[1]
In an accompanying abstract titled, "The Care Transitions Intervention: Results from a Randomized Controlled Trial," Dr. Coleman and colleagues presented the results of randomized, controlled trial of 750 patients, aged 65 years of age or older, who were admitted with 1 of 9 major diagnoses. Of these patients, 370 were assigned to a novel "care transitions intervention," which included (1) encouragement to take a more active role in their care; (2) tools to promote cross-site communications; (3) continuity across settings; and 4) guidance from a "transition coach." The hospital readmission rates at 30, 90, and 180 days were less for patients in the intervention group than for those in the control group, as were the readmission rates for the same initial diagnosis. Moreover, study participants were significantly more likely to achieve physical function and symptom control, based on their own predefined goals. Such interventions will probably become commonplace in the future as a tool to improve outcomes and minimize acute care expenditures.[2]
In a similar vein, Gail M. Burniske, PharmD, and her colleagues from Boston Medical Center, Boston, Massachusetts, and Lovelace Medical Center, Albuquerque, New Mexico, presented an abstract titled, "Post-Discharge Follow-up Telephone Call by a Pharmacist and Impact on Patient Care." This was a prospective trial intended to answer whether a postdischarge telephone call, aimed at educating patients and answering medication-related questions, would have a measurable impact on 30-day readmission rates. Interim data show that patients assigned to a follow-up pharmacist phone call were 18% less likely to present to the emergency department or to be admitted at 30 days.[3]
Another problem that results from fragmentation of care is that of critical lab results, which are discovered only after a patient has been discharged. In an abstract titled, "Test Results that Return after Hospital Discharge: A Patient Safety Concern During Transition of Care," Christopher Roy, MD, and his colleagues at Brigham and Women's Hospital in Boston presented the results of a prospective review of 2644 discharged patients for whom a total of 2033 test results were reported after the time of discharge. Of these results, 191 (9%) were potentially actionable and yet physicians were unaware of 62% of them; 37% of these test results led to change in plan and 12% were considered urgent. This clearly represents an area that needs further improvement.[4]
The US Senate proclaimed March 2005 as DVT [deep venous thrombosis] Awareness Month. Melanie Bloom, the widow of David Bloom and the national spokesperson for the Coalition to Prevent DVT (http://www.preventdvt.org), was an invited speaker at this year's meeting. David Bloom was a news correspondent for NBC News "embedded" in the 3rd Infantry Division in Iraq. He died in April 2003, not from an insurgent's bullet or a mortar shell, but from pulmonary embolism. Melanie Bloom reminded us that more than 2 million Americans will develop DVT this year and more than 200,000 will die of this disease. This number is more than the number of Americans who will die from breast cancer and AIDS combined.
SHM announced the development of the VTE [venous thromboembolic] Resource Room to educate hospitalists about the symptoms, signs, and treatment of VTE. The VTE Resource Room will be available to everyone at the SHM Web site: www.hospitalmedicine.org.
In the United States, more than 4 million hospitalizations occur annually among people with diabetes mellitus, at a cost of approximately $40 billion.[1] The American Association of Clinical Endocrinologists (AACE) has stated that there needs to be greater emphasis on the management of blood glucose levels during hospital stays.[3] Hyperglycemia is a common comorbid condition in all hospitalized patients. While most patients with hyperglycemia enter the hospital with a diagnosis of diabetes mellitus, many do not. Data suggest that hyperglycemia is an independent marker of in-hospital mortality in patients with previously undiagnosed diabetes.[2]
Dr. Guillermo Umpierrez, Associate Professor of Medicine at Emory University School of Medicine, discussed the issues surrounding the management of blood glucose management in hospitalized patients in his talk, "Diabetes Management in Inpatients: The New Paradigm Shift." He cited a number of common reasons for poor blood glucose control in hospitalized patients: 1) stress response, 2) concomitant steroid therapy, 3) changes in insulin regimen, and 4) use of hyperalimentation feeding solutions.[4]
He also proposed 2 major reasons why hyperglycemia is linked to poor hospital outcomes:
Hyperglycemia leads to immune dysfunction. Data suggest that hyperglycemia hinders the phagocytic activity of neutrophils.[5]
Hyperglycemia leads to increased serum free fatty acids (FFAs.) Dr. Umpierrez suggested that increased serum FFAs result in endothelial dysfunction, increased oxidative stress, and sympathetic overactivity.[6]
Dr. Umpierrez reviewed the clinical evidence surrounding the role of glucose control in various clinical conditions:
Acute myocardial infarction. Insulin and glucose infusion followed by an outpatient multidose insulin regimen reduced long-term mortality in patients with type 2 diabetes who had had acute myocardial infarction.[7]
Acute stroke. Patients with stress hyperglycemia, when compared with those patients without stress hyperglycemia, have a relative increased risk for 30-day mortality after stroke.[8]
Cardiovascular surgery In patients with diabetes mellitus, improved perioperative glycemic control is associated with decreased risk for wound infection after cardiac surgery.[9]
Critically ill patients in intensive care units (ICUs). Intensive insulin therapy in critically ill patients treated in the ICU is associated not only with a shorter length of stay in the ICU but also with improved clinical outcomes.[10]
Dr. Umpierrez reviewed the AACE consensus for upper limit of inpatient glycemic targets:
ICU: 110 mg/dL (6.1 mmol/L)
Non-ICU patients: Preprandial: 110 mg/dL (6.1 mmol/L); maximum: 180 mg/dL (10 mmol/L)
He then proposed the following strategies to improve blood glucose control in hospitalized patients:
Admission orders. He reminded hospitalists not to automatically discontinue insulin or oral agents upon hospital admission.
Glucose targets. In the hospital, Dr. Umpierrez noted, hyperglycemia is too readily tolerated. A blood glucose level of 200 mg/dL is too high and requires immediate attention.
Therapeutic adjustments. He stated that, too often, providers do not review the results of finger-stick glucose testing. He urged hospitalists to use this information to guide pharmacologic treatment of hyperglycemia.
Insulin sliding scales. Dr. Umpierrez believes that regular insulin sliding scales are overutilized. He encouraged hospitalists to schedule insulin therapy, whenever possible, for hospitalized patients.