Shaping the Future of Academic Health Centers: The Potential Contributions of Departments of Family Medicine

Ambulatory and Primary Care

The last 20 years have seen a dramatic change in the role of the hospital in health care. Between 1980 and 1999, the number of days of hospitalization per 1,000 population annually dropped by one half: from 819 to 317 among younger adults, and from 4,098 to 2,257 among the elderly.¹² This drop was the consequence of several interrelated trends—improvements in care, changes in medical practices, and the drive to reduce the cost of care. Paralleling and strengthening these trends have been dramatic shifts in methods of reimbursement, with the imposition of Diagnosis-Related Groups in the mid-1980s and the growth of capitation in the 1990s. The net result of all of these trends has been a massive transfer of care from the hospital to the office or outpatient setting.

The implications of this shift are large. Physicians of all specialties have revised their approaches to provide in the outpatient setting what was formerly done on an inpatient basis, hospital systems have progressively geared themselves to shorter stays, and the rise of the hospitalist movement has focused further attention on the cost-effectiveness of inpatient care. Of course, teaching hospitals have operated outpatient dispensaries for almost 2 centuries, but these pressures have forced hospitals to provide as much care as possible in an outpatient venue. By 2003, about 60% of the average hospital’s operating margin thus depended on outpatient revenues.¹³ The dramatic growth of ambulatory operations has required new organizational structures in AHCs, including designated ambulatory leadership, new financing models, and different relationships between hospitals and physicians.

During this time, many AHCs acquired or developed primary care networks to capture “covered lives” and ensure referrals into the AHC system.^2,3 This proved to be a high-risk strategy, and the benefits of vertical integration have often been elusive. Many academic institutions had no tradition of employing primary care physicians and have faced new structural and cultural challenges in managing primary care networks.^1–3 As capitation receded in many markets, primary care networks often began to lose substantial amounts of money.^14–16 The case study offered by Magill and colleagues¹⁷ describing events at the University of Utah demonstrates how departments of family medicine may be uniquely well suited for leadership in this situation. Family medicine faculty often have substantial clinical and managerial experience in outpatient care, and bring a focus and expertise of critical value as AHCs struggle to make primary care networks fiscally viable. The accompanying commentary by Krugman¹⁸ contrasts the University of Utah experience with the University of Colorado experience and underscores the important role of governance and related academic missions.

Fiscal viability, however, is not the whole picture. The relatively new prominence of ambulatory and primary care within AHCs offers substantial opportunity to address questions of quality and effectiveness in primary care with greater academic vigor. The largest platform of health care delivery in the United States is the small to medium-sized physician practice—not the AHC. Family physicians provide a much larger number of face-to-face patient encounters in these practices than do physicians of any other specialty¹⁰ and are far more likely than subspecialists or other generalists to be the usual source of care for patients with chronic diseases, including coronary vascular disease, asthma, heart failure, chronic obstructive pulmonary disease, depression, diabetes, and hypertension.¹⁰ Our current ambulatory care systems, moreover, are largely based on a model of acute care that came to maturity in the 1950s, when trauma and minor infectious disease were major reasons for seeing patients, and before the dramatic growth of chronic disease.¹⁹ Evidence of a quality chasm between achievable care and actual care in the outpatient setting, as well as health disparities and unequal access issues, have mounted in recent years,^20–22 creating an imperative for substantial change in ambulatory care. Primary care practices of the future will require dramatic redesign,^23–26 with multidisciplinary approaches to patient-centered care, infrastructure reform with advanced scheduling and billing systems, and clinical systems that facilitate measurement of quality and active management of chronic disease.

Academic centers have an opportunity to provide leadership in primary care. Although AHCs have traditionally led development of innovation in new technologies in subspecialty and hospital care, they have failed to emerge as leaders in the testing and diffusion of innovative health services delivery at the level of primary or population care. To the contrary, as asserted by a recent report funded by the Robert Wood Johnson Foundation, quality improvement initiatives appear to be “orders of magnitude more difficult” to implement in academic centers than in community settings²⁷ and will likely remain so until evidence-based system redesign becomes a priority. Visits in hospital outpatient departments are characterized by greater costs and service intensity, and poorer continuity of primary care than visits in community health centers and physicians’ offices.²⁸ Ownership of primary care practices creates new opportunities for AHCs to lead change in a setting in which the potential for impact on the health of the population is greatest.

Indigent Care

Seventy-five percent of the nation’s 126 AHCs are seated within communities of underserved populations,²⁹ and care of the medically indigent and vulnerable populations is a key component of the mission of most AHCs. Historically, the provision of indigent care has always been linked to medical student and house staff education, both in the inpatient setting and in the dispensary or clinic setting. With the rise in numbers of uninsured and underinsured in recent years, however, AHCs are seeing substantial increases in the cost of indigent care. Hospitals have reported dramatic increases in charity and bad debt over the last 20 years,³⁰ reflected in both the absolute costs of care and the relative impact on net revenues. Academic centers care for a large minority of all indigent patients, but indigent care represents a disproportionately large segment of their inpatient, and increasingly outpatient, care. Teaching hospitals represent 6% of the nation’s hospitals but provided 40% of the nation’s hospital charity care in 2000.³¹

The financial impact of indigent care on AHCs has been compounded by the challenges of the larger health care economic environment. Modern medicine and especially academic medicine is a capital-intensive industry, requiring a large operating margin to continue to invest in physical facilities and technology. Prospective payment, utilization constraints, decreases in clinician reimbursement by Medicare and other insurers, and slowdowns in other sources of federal support have limited the ability of AHCs to subsidize their missions of research, education, and indigent care. The increasing cost of indigent care thus threatens the core missions of the AHC.

How have AHCs responded? Not surprisingly, many have responded with attempts to limit access to care in a variety of ways, such as increasing wait times for appointments or demanding advance payment for services. Some, however, have responded more proactively, identifying ways to provide more effective and efficient care rather than less care. In the absence of a usual source of care, uninsured patients may overly rely in emergency departments as a point of care, thus greatly inflating the costs of care. With the recognition that the uninsured represent one of the last “unmanaged” populations remaining in a highly managed medical marketplace, health systems have begun to look for cost saving through improved coordination of services for this population.

Efforts to manage the care of the uninsured range greatly in degree. At a minimal level are attempts to reduce demand through health education or outreach clinics. At the other end are more ambitious efforts to transform the regional health care landscape, identifying and developing partnerships with private primary care clinicians, community health centers, and state and local agencies to develop organized networks of care.^32–34 Community partners may be better equipped to provide access to culturally competent care, and a robust primary care system is the best antidote to avoidable hospitalizations and emergency department visits. Case management programs for high users can target support for patients at highest risk, and disease management programs and coordinated mental health services may also prove cost-effective. Community-academic partnerships have the potential to improve the health of the community, further the service missions of schools, and often provide educational opportunities for both learners and community clinicians.^35–39

The case study by Kaufman and colleagues⁴⁰ at the University of New Mexico conveys the enormous potential of effective academic-community partnerships for improving the care of the uninsured while cutting avoidable costs. Like many AHCs, the University of New Mexico has been the major source of care for a large uninsured population spread over a large geographic area. In a series of initiatives spanning a decade, however, the institution has put in place the elements of a comprehensive system of care for the uninsured. These elements include a primary care home, affordable copayments for visits and medications, and integration of public health, medical, behavioral, and social services into common sites at the community level. The Department of Family Medicine provided key intellectual and organizational leadership in bringing together various units of the academic center and building bridges with community partners. The commentary by Roth⁴¹ puts the University of New Mexico’s strategy in national context, contrasts investment in tertiary and quaternary care and in prevention and primary care, and underscores the political context of care for the uninsured.

Education

Powerful forces of change in health care delivery and renewed emphasis on population-based thinking in health care will require equally substantial reform in the educational mission of AHCs. We must consider how the clinical setting of education should prepare graduates to practice in modern health care settings. As described originally by White and others⁴² and replicated with current data by Green and others,⁴³ the vast majority of care takes place in community-based outpatient practices, substantially distinct from AHCs. Nearly 1 billion ambulatory visits are made annually in the United States, compared with 32 million hospitalizations,¹² but medical school curricula often place a major emphasis on the delivery of tertiary services in the inpatient setting, and reform without change is common.^44,45 With trends toward shorter hospital stays and much diagnostic evaluation being done before hospital admission, learners who remain in hospitals may lose some of the intellectual opportunity to approach undifferentiated patient complaints, as well as the opportunity to establish relationships with patients that offer insight into the myriad individual determinants of health and well-being.^46–48 Reflecting this shift, the recent AAMC policy statement underscored the need for considerable change in clinical curricula to reflect contemporary clinical practice as well as to support teachers and teaching.⁴⁹

It is clear that the clinical settings of the future will look very different from those of the past. The aging of the population, the rise in chronic illness, and technologic advances all contribute to an increasing shift in medical care to an office and community focus. Twenty-first century challenges require training beyond the traditional biomedical focus, to incorporate social, cultural, behavioral, and environmental influences on population health and prevention. These challenges also require better background on resource allocation and should emphasize chronic disease in addition to acute disease management.^50–52 Physicians must be adept at information management and critical appraisal of the literature, and prepared to work effectively within interdisciplinary teams and with community partners to deliver patient-centered care. AHCs are called on, then, to cultivate physicians who are not just healers, researchers, and educators, but also leaders, managers, and public health partners.⁵³

Financing mechanisms for clinical education that are oriented toward hospital inpatient care (via direct and indirect payments by Medicare for graduate medical education) are thus increasingly out of synchrony with the public good.^54,55 Unfortunately, capacity for sustaining educational activities in AHCs is simultaneously threatened by narrower operating margins in the current health care marketplace, associated with trends toward reduced federal subsidies for physician education.⁵⁶ In addition, shifting medical education to ambulatory settings presents considerable challenges beyond the disincentives inherent in current financing streams. Although more than one half of surveyed young physicians believe that they had too little training in physician offices, organized care settings, or long-term care facilities,⁵⁷ students generally rate the quality of instruction in ambulatory settings lower than that in inpatient settings.⁷ The need for accreditation and oversight of educational programs in ambulatory settings, as well the need for faculty development and organization, present further obstacles to the development of strong ambulatory education. The evidence base for determinants of quality and effectiveness in clinical education remains weak, leaving little to guide AHCs as they implement changes that will better prepare students for their future in clinical practice.⁶

Training physicians to be aware of—and act on—their social responsibility should also be an important part of the curriculum. It is notable that the social status of physicians, as reflected in opinion polls, has declined substantially during the last 5 years.⁵⁸ To be sure, medicine’s successes are chronicled extensively, and physicians continue to have high prestige, but over the last 5 years, there has been a marked drop in the prestige attributed to this profession. This relative decline in status is greater than the drop of status in the clergy associated with widespread publicity about sexual abuse scandals.⁵⁸ The last 10 years has also seen a relative decline in applications to medical schools,⁵⁹ despite the increase in number of women coming into medicine. These shifts may reflect that, to some extent, the profession of medicine is under siege by a combination of adverse trends: the steady negative publicity of liability suits; ethical lapses at major professional societies, major journals, and the NIH; and the juxtaposition of astronomical salaries made by some physicians alongside escalating health care costs increasingly borne out of pocket by consumers. In this context, an emphasis on service for medical students in underserved communities, abroad, or both strikes a particularly positive chord for students, and development of curricula addressing “meaning in medicine” have also proved extraordinarily attractive.^60,61

Community-based and international education is an area of traditional emphasis and strength for family medicine. In many institutions, and facilitated by Title VII funding, departments of family medicine have taken a leadership role in curricular innovation. The case study presented by Smith and Weaver⁶² provides one example of such leadership—an international elective that successfully brings vitality to the medical curriculum, promotes understanding of the social and environmental determinants of health, and nurtures the idealism and compassion of students. The commentary by Eckhert⁶³ gives a national perspective on the increasingly recognized need to prepare students to work in the global community and discusses the measures by which we might determine success.

Workforce Planning

Training physicians was the original and unquestioned purpose of medical schools. In the 1960s and 1970s, the United States doubled the number of medical graduates, recognizing that it was important to provide physicians to meet the needs of the population. It was this rationale that justified the development of the new discipline of family medicine as well as parallel initiatives to build more hospitals and develop new disciplines of clinicians, such as physician’s assistants and nurse-practitioners. The landmark 1981 report on physician workforce from the Graduate Medical Education National Advisory Committee (GMENAC)¹⁶ codified this “planning approach” to physician workforce planning based on estimates of population growth. The authors predicted that by the year 2000, there would be an excess supply of specialist physicians. The Bureau of Health Professions, the Council on Graduate Medical Education (COGME),⁶⁴ and others long espoused such needs-based methods, endorsing policies that aimed to restrict specialist training and increase the ratio of generalists to specialists.

In recent years, however, a different approach to estimating workforce need has gained popularity. Emphasizing the influence of economic growth on demand for physicians, Cooper and others^65,66 have argued that we are now facing a shortage of specialists and perhaps an overabundance of primary care physicians. Assuming marked growth in the economy and allowing for marketplace allocation of physicians, Cooper’s supply-demand model argues that economic growth will increase demand for specialist care. His work has influenced COGME to recommend increases in physician training, without any special attention to primary care.⁶⁴ Similarly, the AAMC has recently called for removal of Medicare restrictions on funded positions to allow AHCs to respond to physician work-force market dynamics.

The differences in workforce projection between planning-based and demand-based models are substantial. Workforce models are often imbued with professional self-interest, and many methodologic issues remain controversial. Neither of these approaches considers the impact of new models of care, although we know that the enormous growth in the numbers of allied health professionals and substantial changes in care delivery are transforming practice. In a recent comparison of 3 methods for estimating family physician workforce needs—a planning model, a supply and demand model, and an estimate of need based on new models of care—analysts noted that the estimated number of family physicians necessary varies by more than 50,000 physicians!⁶⁷ Importantly, the characteristically uneven distribution of physicians in America means that an increase in per-capita physician supply alone is not sufficient to resolve problems of access to care in many populations of patients.

Empirical evidence increasingly shows, however, that the proportion of primary care clinicians and their integration into the health care system are strongly correlated with both quality and cost-effectiveness of health care of the population.^68–71 In contrast, greater specialist supply is not associated with better population health and is more likely to inflate health care costs.^72–74 Value-neutral approaches such as that espoused in market demand–based models fail to address this situation. If quality and cost-effectiveness of care for the population are important, the balance of generalists to subspecialists is a key leverage point for governmental policy.

How should AHCs respond? For the public, training physicians remains a major mission of the AHC. That mission requires that AHCs take a position on the workforce debate. The need for taking a stance has become increasingly urgent as interest in family medicine and primary care has dropped precipitously since the late 1990s, with worrisome implications for safety net clinicians.⁷⁵ Even more pronounced has been the stunning drop in interest in general internal medicine among internal medicine residents,⁷⁶ and strong shifts of physician’s assistants, nurse-practitioners, and osteopathic physicians toward subspecialization.⁶⁷ The case study presented in this supplement by Berg and Norris⁷⁷ describes how the University of Washington, a well-regarded institution with both substantial NIH-funded research and an explicit mission to support primary care, decided to use workforce data to drive their educational policy, including admissions and curriculum design.

Social accountability of AHCs remains an important issue. Over the last decade, the World Health Organization has promoted an explicit call for the integration of social accountability into the education, research, and service activities of academic centers. Workforce development is a key element of that social accountability.⁷⁸ In practice, this means “a systematic approach to addressing priority health needs, including issues of access to services, determining and educating the appropriate number and mix of physicians, and facilitating the geographic distribution necessary to meet the needs of the community.”⁷⁹ The commentary by Rourke⁸⁰ provides a perspective on the role of AHCs from Canada, a nation whose medical schools have espoused the principle of social accountability and grappled with putting it into practice.⁸¹ He comments on the University of Washington approach and addresses both the specifics of workforce planning and the practical aspects of integrating responsibility for the health of a province and the management of a medical school.

Research

How can AHCs best contribute to improving the health of the population through research? For 2 generations, medical science has emphasized the discovery of basic mechanisms of disease and, at a clinical level, the development of increasingly sophisticated methods of diagnosis and treatment for individual patients. Over that time span, the NIH, along with other federal, philanthropic, and commercial funders of research, has reshaped the AHC, adding research as a major mission of many schools and making possible a dramatic increase in the number of faculty and specialty training programs. This advance has been especially impressive over the last decade, as another explosion of work is leading to the development of genetically targeted testing and treatment strategies. The potential benefit of these advances for individual patients is tremendous.

Increasingly, however, as reflected in the IOM report,⁶ leaders recognize that capacity for continued basic science discovery by itself is not sufficient. The full benefit of these discoveries is unlikely to be realized without greater focus on clinical, health services, and prevention research.⁶ The formal distinction is between efficacy—what ought to work in ideal circumstances—and effectiveness—what actually happens with usual patients in typical situations. More than 30 years ago, the classic study of Brook and colleagues⁸² of the effectiveness of nonurgent care in the emergency department documented that, even in our most prestigious institutions, what actually happens to patients differs greatly from what we think should happen—and outcomes are much worse than we want to admit. With important exceptions, our academic culture has not prioritized the exploration of effectiveness, promoting technical triumph for individuals over benefits in practice for populations.

The “quality chasm” described by the IOM⁸³ is thus not new, just newly listened to. Three basic streams of evidence document widespread lack of effectiveness in American medicine. The first relates to quality of care, with mounting reports documenting a quality gap in both inpatient and outpatient settings.^20,21,83 Care patterns vary considerably between geographic regions, without strong relation to quality or outcomes; almost one half of adults in a national sample do not receive recommended care for prevention, acute episodes, or chronic conditions.²⁰ Similarly, more than one half of patients with diabetes,⁸⁴ hypertension,⁸⁵ high cholesterol levels,⁸⁶ congestive heart failure,⁸⁷ chronic atrial fibrillation,⁸⁸ asthma,⁸⁹ depression,⁹⁰ and tobacco addiction⁹¹ do not receive proven treatment.²¹ As the IOM summarized, “there is a large gap between what we know and what we do.”⁸³

A second issue is equity. Despite a rhetoric of equity in American health care, disparities in health care access, quality, and outcomes related to race/ethnicity and socioeconomic status are pervasive and have been well documented across many clinical conditions and many care settings.⁹² Health disparities represent an important challenge to deeply held American values of equality. They demand correction.

Finally, the effectiveness of US health care is called into question by the health status of the population as a whole. Despite enormous investments in health care, the United States lags well behind other first world countries—and, indeed, increasing numbers of formerly third world countries—in both population health indicators and patient satisfaction.^68,71

These are huge issues for American society as well as American medicine, and AHCs, by nature and by design, are not yet well suited to addressing them.⁹³ These issues are, however, the agenda for the future of care,⁹⁴ particularly as the basic science achievements of the last decade are translated to the frontlines of care. The new NIH Roadmap initiative,⁹⁵ along with related changes in the organization of our research agenda,⁹⁶ represents a nascent recognition that our research infrastructure must be broadened to include more emphasis on the effectiveness of medical care and the translation of innovation from the bench not only to the bedside but also to the community. Increasingly, translational research will include a component of bidirectionality, recognizing the important role that communities and other new research partners can play in informing both the questions and methods of medical research.⁹⁷

Achieving a goal of evidence-based practice will require practice-based evidence. Departments of family medicine are well positioned to link traditional research with the living laboratory required for this new kind of science. The case study by Schwenk and Green⁹⁸ from the University of Michigan describes the development of the research infrastructure necessary to work in the community in the context of a successful submission for an NIH Roadmap proposal. The commentary by Roper and Newton⁹⁹ puts the experience in Michigan in national context, from the perspective of Dr Roper’s leadership roles at the Health Care Financing Administration, the Centers for Disease Control and Prevention, and a commercial health insurer, and in public health, and his current role as dean and chief executive officer of the University of North Carolina Health Care System.

AHCs as a Public Good

Almost 40 years ago, Hardin¹⁰⁰ described the tragedy of the commons:

“Picture a pasture open to all. It is to be expected that each herdsman will try to keep as many cattle as possible on the commons. Such an arrangement may work reasonably satisfactorily for centuries because tribal wars, poaching and disease keep the numbers of both man and beast well below the carrying capacity of the land.

“Finally, however, comes the day of reckoning … as a rational being, each herdsman seeks to maximize his personal gain. Explicitly or implicitly, he asks, ‘what is the utility of me adding one more animal to my herd?’ … The rational herdsman concludes that the only sensible course for him to pursue is to add another animal to his herd. And another; and another. But this is the conclusion reached by each and every rational herdsman sharing a commons. Therein is the tragedy. Each man is locked into a system that compels him to increase his herd without limit—in a world that is limited.”

AHCs represent a kind of commons—a public trust for the future of medicine and society. Public investment through the states, Medicare, Medicaid, NIH, and other branches of government has played a major role in the development of AHCs—and, over the last 2 generations, AHCs have played a major leadership role in the clinical and social success of American medicine. American medicine is graven in the image of the AHCs.

Now, however, medicine and society face new challenges. Our patients are changing, what medicine can do is expanding rapidly, and public accountability is high: the stakes are greater than ever before. The herdsman’s incentive, however, is to continue to develop ever more elaborate technologies that may help individuals and individual AHCs, but may distract from or even worsen the health of the population.

To respond to this changed environment—and to continue to lead—AHCs must change substantially. The IOM report on the future of AHCs⁶ is an urgent call to attend to the commons, to respond to the changing needs of our society—and to move with urgency and impact. By pushing for change, by reinvigorating the relationship between AHCs and the communities they serve, and by emphasizing fundamental innovation across all missions, departments of family medicine can help lead the renewal of AHCs.