Education and Training for Primary Care
If primary care is to move in the direction advocated by this committee, many aspects of education and training of primary care clinicians must be restructured. The committee has already drawn attention to the wide range of responsibilities that primary care clinicians might have, the equally broad array of settings in which they might practice, and the need for a team approach to the delivery of primary care. Various other issues, more widely examined in the arena of health professions education, also impinge on primary care and have implications for the recommendations this committee is making.
Considerable attention has been focused on these important issues. Christakis (1995) reviewed reform proposals for undergraduate medical education in 19 major reports issued from 1910. He found consistent themes in these reports, including the need to increase generalist training and exposure of students to ambulatory care. In recent years, many statements regarding the content and financing of graduate medical education and primary care education of other health professionals have been issued. Moreover, targeted grants from The Robert Wood Johnson Foundation, The Pew Charitable Trusts, and The W.K. Kellogg Foundation have addressed the changes in academic infrastructure, curricula, and financing that must be implemented to respond successfully to a mandate to increase the availability of well-trained primary care clinicians. Most recently, The Robert Wood Johnson Foundation has funded Generalist Initiative grants to medical schools with a goal of promoting primary care and interesting medical students in generalist training.
To this rich mix the present IOM committee adds its particular perspective, which relates more explicitly to primary care. Specifically, this chapter addresses essential changes that need to be made in undergraduate and graduate health
professional training and the need for clinical training to include multidisciplinary team practice; attention is directed to the three types of primary care clinicians—physicians, nurse practitioners, and physician assistants—focused on in Chapter 6. The need to identify common core competencies across these professions is an important ramification of the discussion. The chapter also explores retraining of physicians for primary care. Finally, it offers nine recommendations by which the committee's vision of primary care might be brought closer to reality through appropriate changes in education and training of health care personnel.
Appropriate Training In Primary Care
The scope of primary health care services is broad and often complex. Both the content and the challenges of primary care demand a considerable period of education. The committee believes that all newly trained primary care clinicians must have adequate and discipline-appropriate training—that is, specific training in primary care appropriate to their expected roles. For physicians (many of whom will ultimately provide the gamut of primary care services), this means a residency with emphasis on primary care followed by certification by an appropriate specialty board. For the nurse practitioner, it means graduate education and national credentialing. For the physician assistant, it means graduation from an accredited physician assistant program and certification by the National Commission on Certification of Physician Assistants.
The Education Of Physicians
In considering the education of a physician, this committee concluded that attention ought to be directed at both undergraduate and graduate training, because it believes that new efforts to produce a primary care doctor will be far less productive if instituted only at the graduate level. Thus, this section examines issues for both medical students and residents, noting in particular that models of practice to which physicians-to-be and newly graduated physicians are exposed play a critical role in long-term career directions (Stimmel, 1992; GAO, 1994; Martini et al., 1994; Kassebaum and Haynes, 1992).
Undergraduate Medical Education
Experience in Primary Care Settings
The challenges of revamping the undergraduate medical curriculum should not be underestimated, and this committee was not empaneled to explore such issues in depth. One aspect of primary care is especially important in this context, however, and the committee spent considerable time debating it. Specifically, a true appreciation of a patient's family and community context—a tenet of
this committee's definition of primary care—requires that students gain experience in practices and sites that are primary care based.
This does not now happen to nearly the extent the committee sees as desirable. The reasons are varied. Financing issues have been a major impediment to undergraduate education in ambulatory settings. Training costs are increased, and the logistics can be complex; finding ways to offset such costs has been difficult. Other objections to ambulatory training have been raised as well (Petersdorf and Turner, 1995). Some faculty, for example, believe that inpatient education with its intense exposure to acute disease provides better education and can be transferred to the ambulatory setting more readily than vice versa. Others are concerned that, during office visits, patients may not be willing to devote the extra time that might be required to accommodate undergraduate teaching and that, similarly, community-based physicians may be unwilling to have their patient schedules disrupted by student involvement.
The committee did not find these arguments about the problems of conducting some undergraduate medical education in outpatient or primary care settings persuasive. Calls for greater emphasis on out-of-hospital primary care training in both undergraduate and graduate medical training are not new; they have been raised with increasing frequency in the last several decades (Alpert and Charney, 1973; IOM, 1983). As discussed below, therefore, the committee concluded that the benefits of such training can and do outweigh the drawbacks and that concrete steps therefore need to be taken to provide all future medical students with such exposure. For this reason, it recommends the following:
Recommendation 7.1 Training in Primary Care Sites
All medical schools should require their undergraduate medical students to experience training in settings that deliver primary care as defined by this committee.
The committee concluded that useful, indeed crucial, educational experiences can take place in doctors' offices, community health centers, and other out-of-hospital community sites. It also judged that such exposure to primary care settings and practices should be relatively intense; that is, an occasional short rotation in several sites is unlikely to provide an adequate experience.
References in this chapter to ambulatory in regard to student and resident training should be understood as ambulatory care in primary care settings. The committee strongly cautions against the view that a "rotation in an ambulatory setting" is equivalent to experience with primary care. Substituting ambulatory for inpatient service at either the undergraduate or graduate level will not necessarily yield primary care experience to trainees, because much of ambulatory care is not primary care. For example, many procedures that were once performed in an inpatient setting are now done in offices or ambulatory surgery
facilities—including subspecialty procedures in ophthalmology, gastroenterology, neurology, and others.
In the committee's view, undergraduate medical education in sites like those in which doctors are expected to practice in the future has several benefits. First, it will expand their knowledge of the goals and processes of primary care, improve the skills required in primary care, and raise students' sensitivity toward core elements of primary care, such as prevention. Second, it may affect the choices that students make about their careers, especially if they encounter, in those sites, role models who are competent and enthusiastic about their work (Osborn, 1993; Martini et al., 1994). Third, past resistance of residents in graduate medical training to off-campus or out-of-hospital clinical rotations is understandable, to some extent, given the absence of any earlier undergraduate experience in community-based, ambulatory settings. Providing such training at the undergraduate level might go far toward reducing such resistance.
Curricular and Other Structural Reforms
Curricula and clerkships. Medical schools of course have a certain degree of latitude to determine what their students must know and be able to do when they graduate, and the committee was heartened by information demonstrating that many schools are responding to the challenge of devising innovative undergraduate programs. In 1992, the Association of American Medical Colleges (AAMC) appointed a Generalist Physician Task Force to develop a policy statement for the association and to recommend actions to help reverse the trend away from generalism. The task force report recommended that, as an overall national goal, a majority of graduating medical students be committed to generalist careers and that appropriate efforts be made by all schools to reach this goal quickly (AAMC, 1992).
The AAMC task force found that medical schools are adding courses with a primary care focus during the first two (preclinical) years and are offering or requiring clerkships in one of the generalist disciplines during the third or fourth years, including clerkships that emphasize experience in primary care settings. At some medical schools, even first-year medical students can apply for primary care clerkships, where they can observe generalist physicians in hospital clinics and doctors' offices. At other medical schools, first-year medical students take required longitudinal primary care clinical care experiences during which they observe generalist physicians in their own office practices. Several schools teach beginning physical diagnosis to their first-year students and supervise patient care interactions such as interviewing and simple clinical examinations.
Many schools now include primary care or ambulatory experiences as part of their basic clerkships.1 Gradually more of the core clerkships in family practice,
internal medicine, and pediatrics are being conducted in physicians' offices, community health centers, and group practices. Specialty societies such as the American Academy of Family Practice and the American Society of Internal Medicine (ASIM) actively support such activities with advice, curricula, and evaluation tools. Efforts to encourage states to fund placement of students with practicing preceptors are also under way, with Texas already having passed legislation to fund such programs.
According to a later AAMC report (1994), responses to the 1993 Medical School Graduation Questionnaire found that 36 percent of third-year students and 49 percent of fourth-year students had a primary care clerkship, and 57 percent of these third- and fourth-year respondents had taken the clerkship as a required course.
The AAMC task force also found that curricula are being modified to emphasize the evaluative sciences that are associated with primary care, such as epidemiology and evidence-based medicine. This point is especially relevant with respect to the research agenda issues discussed more fully in Chapter 8. Furthermore, schools are developing programs to provide experience in a number of other fields thought important for a fully rounded primary care education. For example, Dartmouth Medical School requires its students to teach preventive medicine in nearby public schools. Medical students are also matched with needy families whom they advise on health care and social services (New York Times, 1992).
These are illustrative examples only, and a broader set of examples of office-based clerkships is provided in a ''mentorship kit" developed by the ASIM (ASIM, 1995). This kit encourages local efforts (in part because ASIM is dubious about whether federal funding for such programs will be forthcoming), and it offers practical advice for implementing and evaluating community-based internal medicine teaching for students. Collectively, these examples demonstrate that medical schools across the country can act on, and indeed already are acting on, the above recommendation (Recommendation 7.1) in creative and productive ways. In so doing, schools can also lay the groundwork for acceptance of greater out-of-hospital training during residency years, as discussed more fully below.
Competencies and clerkships. Medical schools and various health policy groups have also begun to consider the competencies that should be required of all graduating medical students. As a case in point, The Pew Health Professions Commission (1994) identified seven capabilities that it believes will be essential for all future practitioners, clearly including primary care:
- Care for the community's health.
- Provide contemporary clinical care.
- Participate in the emerging system (including new health care settings and interdisciplinary team arrangements) and accommodate expanded accountability.
- Ensure cost-effective and appropriate care.
- Practice prevention and promote healthy lifestyles.
- Involve patients and families in the decisionmaking process.
- Manage information and continue to learn.
With increasing interest in the third-year clerkship in primary care, the latest addition to efforts to define appropriate curricula for medical students has been developed by Goroll and Morrison with support from BHP/HRSA (Bureau of the Health Professions of the Health Resources and Services Administration) and approved by the Society of General Internal Medicine (SGIM) and the Clerkship Directors in Internal Medicine (CDIM) (SGIM/CDIM, 1995). This model curriculum for the third-year medicine clerkship is based on a national survey of internal medicine faculty. It emphasizes the importance of training students in basic generalist competencies and shifting a greater portion of their educational experiences from the inpatient to the primary care setting.
As described in their materials, the model curriculum divides the competencies into three categories2 that should be taught to third-year students:
- Category one competencies (taught in all cases when appropriate): diagnostic decisionmaking; case presentation; history and physical examination; communication and relationships with patients and colleagues; test interpretation; therapeutic decisionmaking; bioethics of care; self-directed learning; and prevention.
- Category two competencies (taught in some but not all cases): coordination of care and teamwork; basic procedures; geriatric care; community health care; and nutrition.
- Category three competencies (taught occasionally): advanced procedures; occupational and environmental health care.
For each competency, a set of corresponding learning objectives, divided into knowledge, skills, and attitudes, has been devised to help guide the learning agenda.
Faculty. Other changes proposed by the AAMC have included raising the prominence of generalist physicians in teaching and medical school administrative positions. Some medical schools have responded by appointing faculty from the generalist disciplines to serve on important administrative committees. For example,
in 1990, one medical school had an associate dean for primary care; five years later, eight schools had created such a position, and many more had added special advisers to the dean on primary care (Fein, 1995).
Examinations. The National Board of Medical Examiners (NBME) administers the United States Medical Licensing Examination (USMLE), which was first administered in 1992. Taken by medical students at the end of their undergraduate years, it has also begun to move in a direction that supports greater emphasis on education and training for primary care. In testimony submitted to the committee, the NBME acknowledged that several areas of primary care practice had been underemphasized in its licensure examination—namely, ambulatory care, chronic care, care of the elderly, and preventive care. Acting on its belief that these areas are critically important, it has revamped the examination and placed a priority on generalist knowledge and skills (NBME testimony to the IOM Committee on the Future of Primary Care, 1994).
Remaining issues. Despite these encouraging examples, the dominant model continues to be education in the inpatient services of teaching hospitals, and such training can be expected to have a lasting influence. When medical students begin their third- and fourth-year clinical rotations in the hospital, the role models tend overwhelmingly to be those in the increasingly acute, inpatient setting with high-technology interventions (GAO, 1994). Thus, the committee believes that Recommendation 7.1, above, must be acted on more forcefully at the medical school level as a counter to these long-standing traditional dynamics.
The committee has discussed the system of undergraduate medical education as a whole, perhaps leaving the impression that medical schools are essentially the same institutions across the nation. This is clearly not the case, however. Different medical schools have quite different missions: Some focus more on research and the production of specialists, others focus more on education and the production of primary care clinicians. Moreover, the effect of the structure of universities within which medical schools function and of the history within each institution of its departmental affiliations can be substantial (a point noted in another recent IOM report [IOM, 1995] on dental education). The committee was not ignorant of these factors, but it judged that exploring them would exceed both its charge and its resources. The basic conclusion is that efforts to overcome some of the problems of changing the mission and the curriculum of medical schools will need to take issues of the larger university organization and aim thoroughly into account.
Graduate Medical Education
Graduate medical education (GME) provides the opportunity to train physicians for a field of practice and to prepare them for independent practice and
certification. The medical school graduate is an undifferentiated physician who is not capable of independent practice and who must take at least one year of residency training to be eligible for licensure. For practical purposes a physician will require residency training leading to certification to establish his or her place as an appropriately and completely trained physician. Thus, GME becomes as essential for the production of a physician as medical school and is the time when differentiation occurs.
Unlike medical schools, which have relatively broad discretion about teaching curricula, graduate programs in primary care (i.e., residencies) are much more closely defined by the residency review committees (RRCs) of each primary care discipline and by the Accreditation Committee on Graduate Medical Education (ACGME). RRCs approve residency programs, which must comply with their requirements. The specialty boards that examine graduates of residency programs for board certification also influence the curricula by determining what is included—and emphasized—on examinations. In short, regardless of the impact of the above-mentioned changes in medical school curricula, how residency programs are structured will remain a dominant factor in creating a cadre of primary care physicians with the characteristics thought to be significant by this committee.
Residency Programs in Family Practice, Internal Medicine and Pediatrics
Primary care has begun to attract more residents (Fein, 1995). Part of this trend is attributable to external forces, both the growth of managed care (and its greater demand for primary care clinicians) and trends in public policy. For example, several state legislatures have mandated or attempted to mandate that a given proportion, such as 50 percent, of medical school graduates go into primary care residency programs (M. Garg, University of Illinois, Chicago, personal communication, October, 1995). Nevertheless, the main physician specialty areas of primary care—family practice, internal medicine, and pediatrics—have some distance to go in creating training experiences that match the committee's vision of the capabilities that will be needed by primary care clinicians of the future, especially a future dominated by managed care organizations.
Managed care organizations made clear to the committee that the current products of family practice, internal medicine, and pediatric residencies lack key competencies required to function maximally in their systems. Based on its public hearing and site visits, the committee shares with many medical educators and the medical directors of integrated health care delivery systems concerns about traditional GME, especially about the extent to which such training is preparing tomorrow's doctors for the new ways and settings in which they will be expected to function. Graduates of residency programs often lack knowledge of population-based health promotion and disease prevention, evidence-based clinical decisionmaking, and patient interviewing skills (particularly communication
and consultation skills). Many are not taught how to function as a member of a team and have little knowledge of information systems or time and resource management.
Internal medicine and pediatrics merit special attention, in the committee's view, because tertiary care and specialty care still constitute too much of the training in their programs; internal medicine residents may lack experience in ambulatory clinical specialty areas such as dermatology, ophthalmology, office gynecology, behavioral health care, behavioral medicine, and preventive medicine (Kantor and Griner, 1981; Kern et al., 1985; Linn et al., 1986; McPhee et al., 1987). Other commonly cited deficiencies are training in clinical nutrition, occupational medicine, working with other primary care clinicians (e.g., nurse practitioners, physician assistants), use of community services, resource management, and setting up an office practice (Barker, 1990).
Primary Care Tracks
Family practice residency programs are unambiguously committed to preparation for primary care practice, whereas internal medicine and pediatric residencies have competing interests in training for referral practice. In the late 1970s, however, residency programs in primary care internal medicine and general pediatrics were established to train more general internists and pediatricians.
Primary care tracks provide more office-based training in gynecology, dermatology, orthopedics, otolaryngology, ophthalmology, psychiatry, and preventive and occupational medicine than traditional programs, and they offer much greater continuity experience. Residents in internal medicine primary care tracks spend considerable time in ambulatory settings, serving as the principal physician for their patients. Less emphasis is placed on hospital-based and subspecialty training; more attention is directed to ambulatory specialties, medical interviewing, and clinical epidemiology (Lipkin et al., 1990). In general these curricula are closer to what the committee is advocating, but they are still small in number and remain the exception rather than the rule.
Other Content Issues In Training For Primary Care
Academic health centers educate and train all types of primary care clinicians (physicians, physician assistants, and nurses practitioners) as well as many other health professionals. Their role is evolving, however, as health care restructuring moves rapidly ahead, and their responsibilities with respect to creating innovative education and training programs will likely be more complex in the future than today. One particular challenge will be to identify, in concert with professional and other groups, common core competencies for primary care, so that tomorrow's training efforts will reflect the committee's vision of primary care and primary care teams.
Future Steps for Academic Health Centers
The above-mentioned trends toward reform of undergraduate curricula, changes in graduate training, and more physicians opting for primary care training are encouraging, but they do not tell the entire story. Traditional curricula, training sites, and distinguished role models can all have a powerful reinforcing influence once residents begin their training. Unless primary care faculties are in prestigious administrative and departmental positions (e.g., deans and department chairs), and unless medical students and residents encounter enthusiastic role models, mentors, and teaching methods that support prerequisite skills described in this report, market-driven changes are likely to be short-lived and may eventually give rise to dissatisfied and demoralized physicians who resent not being able to practice medicine as they choose or were trained.
The required changes are complex. Academic health centers must undertake fundamental alterations in their missions, administrative structures, practice environments, and curricula. The logistical difficulties are formidable; for example, emphasizing nonhospital settings is costly under current reimbursement policies. Moreover, they come at a time when academic health centers are struggling to change quickly enough to survive in competitive markets, and these pressures do not foster long-term planning strategies.
The committee believes that the survival of academic health centers depends on their adoption of primary care teaching and service as a central mission, while continuing and maintaining their roles in providing extraordinarily complex patient care and pursuing biomedical research that has justly earned an international reputation. Further, society needs to support these changes by providing funds for primary care just as it has supported the traditional teaching and research missions of the academic health center. In short, academic health centers will have to change to reflect the practice environment in which its graduates will practice; but society, if it is to enjoy the health care system and practitioners it evidently wants, will need to provide the policy and financial support without which academic health centers will not be able to move forward.3
Common Core Competencies
Defining core competencies is a requisite for every field in health care. Credentialing of health practitioners—whether by hospitals or managed care organizations—depends on defined competencies. For primary care to prosper, these competencies must be sufficiently well defined for patients, residents, faculty,
managed care organizations, other health practitioners, and physicians seeking retraining to understand clearly what is expected of the professional who provides primary care.
Confusion arises over what it means to be a primary care clinician when members of diverse disciplines and specialties (within medicine as well as outside it) declare that they are practicing primary care. Not everyone who declares that he or she is practicing primary care is, in fact, doing so. Despite efforts to define competencies within each discipline and specialty (as illustrated above), no common, cross-discipline competencies have yet been defined and agreed on, either within medicine or across all primary care clinical fields. The remainder of this section reviews efforts by medicine or other health care professions to articulate sets of capabilities or proficiencies for generalist practice.
Defining Core Competencies in Medicine
Medical training programs have remained separate for historical and understandable reasons. Those reasons and the values they represent—clear and justifiable as they may be to those within the medical establishment—are murky to those outside it. The idea of core competencies, however, is reminiscent of the first-year rotating internship that, at one time and in some states, was required for licensure. The committee does not think that GME ought to return to those days. It holds, rather, that in the long term GME programs in primary care would do better to be based on a core set of competencies for all primary care residents and that such core training ought to be augmented by a series of specialty modules (e.g., in the care of the elderly, of children, or of persons in rural areas).
At its most general, training in primary care should equip the clinician to practice competently in a number of areas; for example, for physicians the following competencies would be important:
- periodic assessment of the asymptomatic person,
- screening for early disease detection,
- evaluation and management of acute illness,
- assessment and either management or referral of patients with more complex problems needing the diagnostic and therapeutic tools of the medical specialist and other professionals,
- ongoing management of patients with established chronic diseases,
- coordination of care among specialists, and
- provision of acute hospital care and long-term care.
What specific competencies would enable primary care physicians to fulfill these roles? For half a century or more, the various primary care disciplines have been engaged in defining core competencies within their own fields. For example, in internal medicine, the Federated Council of Internal Medicine Curriculum
Task Force (FCIMCTF) has developed a list of learning experiences that would lead to needed competencies in general internal medicine (FCIMCTF, 1996, forthcoming). Another case in point is the American College of Obstetricians and Gynecologists, which has developed program requirements for training residents in obstetrics-gynecology (OB-GYN) (ACOG, 1995). These requirements, which have been approved by the ACGME, include experiences in some areas that reflect a primary care orientation: patient education and counseling, screening appropriate to patients of various ages, management of the health care of patients in a continuous manner, appropriate use of community resources, awareness and knowledge of the behavioral and societal factors that influence health among women, and behavioral medicine and psychosocial problems.
The American Academy of Family Physicians (AAFP) put forward a comprehensive competency-based curriculum for family practice training (Family Health Foundation of America, 1983). The curriculum includes three sets of skills: general skills, systems, and skills needed for care of special problems and populations. General skills include: interaction and involvement with patients and families; the family; health promotion and disease prevention; nutrition; community involvement and public health; patient education; research skills; practice management; medico-legal problems; personal and professional issues; ethical decisions; general laboratory knowledge and medical imaging; and anesthesia. System skills are organized by body system—cardiovascular, musculoskeletal, and so forth. The third set of skills is titled "Special Problems and Special Populations." This set includes pregnancy, childbirth, and the puerperium; the developing child; the elderly; environmental and occupational problems; accidents, poisonings, violence, and emergencies; behavioral and psychological patterns; and recreational and athletic health care. The curriculum was intended to be open-ended and flexible to accommodate changing knowledge and regional differences. Currently a task force of the Society of Teachers of Family Medicine (STFM) is in the process of updating this curriculum (Roger Sherwood, STFM, personal communication, November 1995).
Various joint residencies and activities by specialty boards also reflect concerns about common core competencies, typically involving internal medicine with either pediatrics or family practice (JAMA, 1994). A joint statement of the American Board of Internal Medicine and the American Board of Family Practice identified the following essential features of generalist physicians (Kimball and Young, 1994, p. 315):
Generalist physicians must be highly skilled in using appropriate medical consultation and referral to other specialists and community resources when necessary … and must aggressively encourage health promotion and disease prevention and be knowledgeable about the efficient use of resources, behavioral medicine, the information sciences, and the principles of population medicine.
One broad effort reflected a review of residency curricula for family practice, general internal medicine, pediatrics, emergency medicine, and OB-GYN (Rivo et al., 1994). The authors identified 7 categories and 60 key components that primary care clinicians should have. The seven categories were (1) care of the population; (2) care of patients in multiple settings; (3) comprehensive preventive care; (4) treatment of common acute illnesses; (5) ongoing treatment of common chronic conditions; (6) ongoing treatment of common behavioral problems; and (7) other special topics for generalist practice. The authors urged that residency programs require use of these categories and components as the framework for determining resident training.
Barker (1990) offered six "proficiencies" and suggested a residency time-table for achieving the tasks related to each proficiency. Similarly, Lipkin et al. (1984), noting the clinical importance of patient-physician interaction, described a core curriculum for teaching medical interviewing.
Though little collective progress has been made regarding formal approval of a common core set of competencies for a generalist curriculum, one thorough analysis of the educational content of curricula developed for pediatrics, family medicine, and general internal medicine residencies identified 15 educational components shared by the three disciplines (Noble et al., 1994):
- community, creating networks of health care workers and services in the community, teaching prevention, and identifying health problems of the community;
- Prevention: prevention of illness, accidents, and health problems;
- Ethics: sensitivity to issues such as those surrounding birth, abortion, emancipated minors, confidentiality and disclosure of information, conflicts of interest, and the obligations of the physician to society;
- Continuous learning: ability to update medical knowledge throughout one's professional life, to appraise literature critically, and to use evidence-based medicine;
- Medical informatics: ability to use computers and information systems and understanding of biostatistics, epidemiology, and health care policy;
- Consultation: the skill set necessary to recognize professional limitations and obtain appropriate consultative assistance, including the rational choice and timing of referrals and effective interaction with colleagues;
- Advocacy: efforts to seek access to care and other needed resources for segments of the population that cannot obtain them; and
- Practice management: some knowledge of the business of practice, financial and legal management, time management, and similar topics.
Although the most comprehensive effort of its sort to date, and although it had the participation of three professional societies, the primary care disciplines have not officially adopted these competencies. These core areas were arrived at despite the differences among the three disciplines (Lipkin et al., 1990). Specifically, pediatrics is distinguished by the young age of patients and by an emphasis on prevention and developmental stages. Family practitioners care for a fuller range of ages of patients and tend to emphasize the family (and sometimes the community) as a unit (as compared to primary care internists), and they may provide obstetrical services; by contrast, internists have more in-depth training in the pathophysiology, diagnosis, and management of complex medical illnesses. Family practitioners tend to see a higher volume of patients in the office setting compared to internists, who focus more on complicated problems and older adult patients in both office and hospital settings. For both internal medicine and family practice, training in geriatrics is becoming essential.
Defining Core Competencies in Nursing
The nursing profession has also recognized the need for core competencies and the desirability of instilling these during training. Nurse training programs for advanced practice, for example, include a primary care track that has separate branches for older adults and for young people.
The National Organization of Nurse Practitioner Faculties (NONPF, 1995) has identified competencies for nurse practitioners, many of which are related to primary care. The competencies are organized into six domains: (1) managing
client health or illness status; (2) maintaining the nurse-client relationship; (3) carrying out the teaching-coaching function; (4) developing the professional role; (5) managing and negotiating health care delivery systems; and (6) monitoring and ensuring the quality of health care practice.
Defining Core Competencies Across Primary Care Clinical Fields
Reaching a mutually agreed-upon set of core competencies across all primary care clinical fields—that is, physicians, nurse practitioners, and physician assistants—poses formidable obstacles. The committee supports and encourages the efforts of health professional societies, residency review committees, academic medical centers, and specialty boards to define a set of common core competency requirements for primary care.
Recommendation 7.2 Common Core Competencies
The committee recommends that common core competencies for primary care clinicians, regardless of their disciplinary base, be defined by a coalition of appropriate educational and professional organizations and accrediting bodies.
This committee urges the formation of a coalition of appropriate professional organizations, certifying boards, and other groups that provide perspectives about desirable competencies in primary care. Tracking the commonalities of topics and content and mapping them to the definition of primary care is an important task for such a coalition. This is probably a task first for medicine, including schools of medicine; medical residency program directors in family practice, general internal medicine, and general pediatrics; and practicing physicians.
Ideally, however, this effort should eventually include all primary care clinicians, from essentially the same constituencies as for physicians. In addition, important viewpoints will come from representatives with expertise in public health, managed care, the social sciences, and bioethics. The aims are to assist with revamping curricula, promote greater coherence of purpose, and advance understanding and collaboration among primary care clinicians.
The several efforts cited above, such as the 15 components cited by Noble and his coauthors (1994), might form the basis of the work of the coalition proposed in Recommendation 7.2.
Implementing Common Core Competencies
Defining common core competencies will not, in the end, be sufficient. Professional societies and associations, especially those involved with training
primary care clinicians and certifying their capabilities at the end of training, have a major role to play as well in implementing the vision of this committee.
Recommendation 7.3 Emphasis on Common Core Competencies by Accrediting and Certifying Bodies
The committee recommends that organizations that accredit primary care training programs and certify individual trainees support curricular reforms that teach the common core competencies and essential elements of primary care.
Apart from the efforts at defining core competencies already mentioned, the committee notes other specific steps being taken by various physician groups. According to the American Board of Internal Medicine, for instance, internal medicine training is in transition to a broader, evidence- and competency-based curriculum; it will place added emphasis on specific ambulatory skills, training in geriatric and behavioral medicine, clinical epidemiology, and medical informatics. It supports generalist training with other primary care (nonphysician) professionals (ABIM testimony to the IOM Committee on the Future of Primary Care, 1994).
The joint statement of the ABIM and the ABFP already cited (Kimball and Young, 1994) praised efforts at designing interdisciplinary generalist clerkships and endorsed the reduction of institutional and interdepartmental barriers to training in coordinated care. Goals for this model include revising curricula and teaching methods and sharing educational resources as a means of conserving educational resources and improving the quality of ambulatory GME programs. These are all worthy steps that other accrediting and certifying bodies, for physicians as well as nurse practitioners and physician assistants, could adopt or adapt.
Special Areas of Curricular Emphasis
Two areas of competency are of particular interest to this committee: communication skills and cultural sensitivity.
Recommendation 7.4 Special Areas of Emphasis in Primary Care Training
The committee recommends that the curricula of all primary care education and training programs emphasize communication skills and cultural sensitivity.
The committee assumes that primary care trainees should and will learn excellent prevention, diagnostic, and management skills and the other types of
core competencies described above. It wishes to emphasize, however, the two particular skills mentioned above, communication and cultural sensitivity—one more generally applicable to all patients and the other accommodating the needs of some patients.
Good communication skills are essential for primary care clinicians. These involve interviewing, communicating risks and information, answering questions, addressing the concerns of patients and their families, and helping patients make difficult decisions based on ambiguous or conflicting scientific evidence. Skills in facilitating communication—whether for patients who have hearing impairments, are illiterate, or have language or other barriers to communication—can and should be taught to primary care clinicians. Novack et al. (1992) have described a course for medical students that effectively teaches interview skills using a variety of instructional methods including simulated patients and role-playing.
Apart from straightforward communication skills are issues posed by patients with cultural backgrounds and languages that are different from those of primary care clinicians (or trainees). The ability to accommodate these patients' styles of coping with illness and their values, belief systems, and language is critical. Training could include teaching about the health beliefs, practices, and mores of specific ethnic and cultural groups that are in the patient populations to which trainees or future clinicians are likely to be accountable.
Many examples could be given: African-Americans tend to use eye contact differently from white Americans (Shabazz and Carter, 1992). Asian men may refuse to be examined by a female doctor, and their wives may expect their husbands to be present throughout an examination. Latino patients may speak of susto, an illness arising from fright (Allshouse, 1993). Southeast Asians may believe that touching the head is taboo because the head holds the essence of life; consequently, disturbing the head will cause loss of the soul (Sherer, 1993). Other aspects of cross-cultural competence include creating a comfortable atmosphere, encouraging the possibility of disclosure of sexual orientation by using neutral terms, and conveying appropriate trainee and staff behavior toward patients regarding forms of address and rules of propriety (Rigoglioso, 1995).
Emerging links between health professional schools and approximately 600 federally funded health centers are beneficial to both students and health centers because in culturally diverse areas primary care clinicians are expected to be familiar with the cultural context and environmental conditions that affect their patients' health. In many areas of the country, primary care settings are uniquely positioned to fulfill the dual purposes of education—providing students with a very broad set of clinical conditions and offering cultural diversity that helps them gain appropriate cultural competence. Further, given the complexity of presenting problems, especially in underserved communities, students in these settings can learn firsthand about the interdependency of members of a health care team and observe their respect for the complementary skills of individual
team members. The W.K. Kellogg Foundation has funded the Community-Based Public Health Initiative to improve the practice and teaching of primary care through collaborative efforts between academic health centers, health professions institutions, and communities. The project involves interdisciplinary education of graduate nurses and medical residents in community clinics.
The shift away from inpatient training permits early access to preventive and primary care. It also reinforces the change that many communities wish to make, namely, away from the prevailing attitude that patients must find their own way to their clinicians, regardless of barriers presented by language, geography, or culture. On their part, communities with significant unmet health problems have begun to welcome involvement with nonhospital-based training programs.
Financial Support For Graduate Training In Primary Care
In addition to issues of the content of graduate training in primary care, the committee devoted considerable attention to the question of how such training might be supported in the future. Two topics were paramount: where funding will come from (i.e., what parties in this country ought to be responsible for underwriting graduate training) and how support for primary care training in nontraditional (e.g., nonhospital) settings can best be achieved.
Current Sources of Graduate Medical Education Funding
A considerable array of sources provide GME funding: the Medicare and Medicaid programs, the Department of Defense (DOD) and Department of Veterans Affairs (VA), universities and practice plans, state and local governments, and other third-party payers. The largest single funding source, however, is the federal government, primarily through Medicare. In 1994, Medicare payments for GME totaled $5.8 billion and have been estimated to cost $70,000 per physician resident (COGME, 1994).4 This program is described in more detail below. The VA and DOD provide 16 percent of total national support of residents' salaries.
Federal funding of a different type comes from the U.S. Public Health Service (Title VII), also described below. This funding is very sparse, however; together with all other sources of support from professional fees, medical school funds, foundation grants, and gifts for GME, it amounts to only 5 percent of total national support (Eisenberg, 1989). Finally, state and local governments provide
an additional 10 percent, and some states also support physician assistant and nursing education.
Historically, Medicare funds supporting GME have been divided into two categories: direct (DME) and indirect (IME) payments. DME payments include reimbursements for salaries and fringe benefits of the teaching hospitals' residents, the portion of faculty salaries devoted to teaching, and the overhead allocated by the hospital for teaching. IME payments support teaching hospitals to compensate for higher expenses associated with their teaching mission as well as their patients' greater severity of illness. The payments are based on a set of complex formulas that are intended to recognize the urban location of most teaching institutions, their more complex case mix, the higher costs attributable to inefficiencies as part of the training mission of the teaching hospital (e.g., more testing by residents as part of the teaching process, longer operating room time), and unreimbursed costs of clinical research.
Some readers might wonder why if ambulatory training is so essential for primary care training it has not supplanted hospital-based training. The answer lies to a large extent in how GME is financed. Medicare's system of GME funding makes training in ambulatory care exceedingly difficult to finance. When the Health Care Financing Administration (HCFA) began using the prospective payment system to reimburse hospitals for services to Medicare patients in 1983, it included residents trained in ambulatory settings in its calculations of indirect payments. In 1985, however, a HCFA regulation mandated that training in outpatient settings be excluded from the determination of indirect GME payments. Congress responded by passing the Consolidated Omnibus Budget Reconciliation Act of 1985, which reversed the HCFA regulations and required that Medicare IME payments include training in ambulatory settings. This step did not, however, fully solve the problem.
Before 1986, the time that residents spent in ambulatory settings and the cost of administering outpatient education were recognized by Medicare only if the setting was part of the hospital. In the Omnibus Budget Reconciliation Act of 1986, Congress required that Medicare acknowledge the time that residents spend in ambulatory settings if the hospital incurs ''all or substantially all" of the costs of the training. Although this legislation was important in establishing that ambulatory centers do not have to be located in or owned by the hospital, interpreting how much of the cost of education constitutes "substantially all" has made implementation of this law difficult (Eisenberg, 1989).
Furthermore, the rules by which payment is determined for faculty teaching time have also complicated GME financing. Teachers who are not hospital employees cannot be paid through Medicare unless the hospital pays them directly or by written agreement. Even if such an arrangement were made, the
hospital would be limited to the costs it showed in 1984, the base year from which Medicare payments are calculated. This has created a financial disincentive to physicians to teach in hospital outpatient training programs and in programs that are separate from the hospital, despite their very real educational advantages (Eisenberg, 1989). Moreover, physical additions to hospitals after 1984 are not recognized in the payment formula, which means that hospitals find it difficult to build new facilities such as outpatient centers to support primary care residency programs.
In addition, resident time spent in outpatient settings other than hospital clinics is not included in the full-time-equivalent calculations for the payment of indirect costs. If a resident's training moves from a hospital-run clinic to a faculty-run clinic—even at the same location—the resident's time no longer counts toward the indirect adjustment (National Governors' Association, 1994). In sum, hospitals have learned that if they want to maximize GME payments for services provided to patients, they should keep trainees as house staff (and thus their site of training) in the hospital.
Another hindrance to the training of primary care physicians is the fact that Medicare GME payments are made to any certified residency program, whether or not such programs further national health care workforce goals and need. In the face of many calls for decreases in the training of specialists and increases in the production of primary care physicians, this aspect of Medicare GME funding in effect encourages the training of more specialists.
Title VII Funds for Primary Care Training
Federal targeted support for residencies in primary care—including general medicine, general pediatrics, and family medicine—was authorized in the 1976 health professions legislation, specifically Title VII, Section 784, of the Public Health Services Act. Title VII also provides support for physician assistant programs and general dentistry. Currently $59.8 million in funds support approximately 405 grants awarded for medical residency training programs, faculty development, and predoctoral training in General Internal Medicine/Pediatric and Family Medicine Programs (Bureau of Health Professions, personal communication, November 1995).
Grant support for physicans assistant (PA) educational programs promotes educational preparation of physicans assistants for roles in primary care settings and utilization in medically under-served areas. Since 1972 these grants have encouraged curricula to focus on primary care and deploying physicans assistants in areas of need. Like the medical training grants, these grants, which totaled $5 million in fiscal year 1993, are administered through the Division of Medicine, Bureau of Health Professions, Health Resources and Services Administration (HRSA) (HRSA, no date). Title VIII provides support for nursing education and is discussed below.
Title VII was not the first federal effort to support primary care graduate medical education. Health professions legislation in the 1960s increased medical school enrollment through capitation and encouraged the establishment of new medical schools. Within a decade medical school enrollments doubled. It was widely believed that by graduating more physicians, the need to produce more generalists would be addressed. Instead, an increasing proportion of the new graduate students pursued subspecialty training.
In 1974, the Bureau of Health Professions in the Health Resources and Services Administration (BHP/HRSA) awarded six contracts to support the development of residencies in general pediatrics and general internal medicine. Shortly thereafter, The Robert Wood Johnson Foundation provided support for some of the original six residency programs as well as others and added a major evaluative component. The documented success of these programs supported continuation of the Title VII effort.
One objective of the 1976 Title VII program was that 50 percent of medical school graduates would choose primary care careers. Consideration was given to requiring 50 percent of graduating students to enter primary care in order for medical schools to receive federal support for GME; however, the legislation did not include such a requirement. Shortly after the legislation was passed, the AAMC reported that 50 percent of GME first-year positions were already in the fields of internal medicine, pediatrics, and family medicine. Of greater concern than the number of the entry-level positions, however, was the number of graduates at the completion of residency training who would be generalists, because many residents who enter general residencies go on to subspecialty training.
The legislation, a product of efforts of the American Academy of Family Practitioners (AAFP) and a small group of academic pediatric and internal medicine generalists, not only supported primary care graduate education but also undergraduate departments of family medicine. Family practice was unquestionably a primary care discipline, and eligibility for grant funding was determined by having an approved residency. The same model did not work for internal medicine and pediatrics because these programs trained large numbers of subspecialists, and no mechanisms were available by which primary care training could be distinguished from the more typical training available to these two specialties.
To help address this problem, BHP/HRSA—with consultation from appropriate medical groups including the AAMC, the AAFP, the American Academy of Pediatrics, and the American College of Physicians—developed eligibility criteria for application to general internal medicine and pediatrics. These criteria included 25 percent "continuity" experience,5 a psychosocial curriculum, and
sizeable ambulatory experiences; the last point was especially important for internal medicine, which at that time was 90 percent or more inpatient training. These criteria, although modified in the intervening 20 years, have remained in principle the distinguishing features of primary care training in general internal medicine and pediatrics.
Despite the success of the Title VII program, the period 1980 through 1992 saw funding remain flat (in fact, funding actually decreased because of the failure to keep up with inflation). Funding was especially problematic because various administrations during the period were opposed to reauthorization of the program as a whole. Again, as of this writing, the present climate of budget cutting creates some doubt about whether the Title VII program will be reauthorized.
Funding of Nurse Practitioner Education
Since its original enactment 30 years ago, Title VIII of the Public Health Service Act (P.L. 104-12) has played a significant role in helping to improve health care delivery in our nation by providing federal support to nursing education and students in nursing programs. Specifically, Title VIII programs fund the development of innovative programs to reach underserved areas, the development of educational programs for advanced practice nurses, and the special programs for nursing education for individuals from disadvantaged backgrounds.
Professional Nurse Traineeships for nurse practitioner and other advanced nurse education at the master's and doctoral level are also provided under this authority (Janet Heinrich, American Academy of Nursing, personal communication, January 1996; Bureau of Health Professions, Division of Nursing, 1994). Title VIII support totals roughly $60 million a year and goes directly to educational programs. Of the $60 million, $16.14 is earmarked for nurse practitioner and certified nurse midwife (CNM) programs with $10.9 million for nurse practicition programs and $4.8 million for CNM training. Medicare funds also support nursing, but the $248 million in 1994 Medicare funds support primarily preprofessional (diploma) nursing education (Aiken and Gwyther, 1995).
Graduate Medical Education as a Public Good
A supply of well-trained clinicians is a national resource for all Americans. This benefit, plus the very high cost of graduate training for physicians, justifies the use of public funds to help support such education (Schroeder et al., 1989). Such a resource can be understood, in the classic economic sense, as a public good. "Public goods" are those consumed collectively or those from which everyone can benefit, and where one person's use does not, in theory, prevent any other person from using or benefiting from the goods in question; roads, national defense, and information are cases in point. The contrast is made with "private goods," where consumption or use is exclusive and benefits are internalized; if
left to private markets, enough of these goods or services will not be produced to meet public need.
In this context, training (and the costs thereof) should be regarded as a public good; the private market, left to itself, will underproduce GME (or, more specifically, fully trained physicians), whether for primary care or specialty care. Health plans (or health institutions) in the private marketplace will not invest in training clinicians (at least not to the extent necessary). The reason is that the eventual benefits would accrue to all health plans because any physician, having completed his or her training, can work for any plan or institution, but the costs of his or her training are borne by one plan, and those incurred costs might make that plan less competitive.
Furthermore, the costs of training are too great for many medical trainees to pay entirely without incurring very large debts. Indeed, the debt burden incurred by students—which might be repaid sooner if the students enter a highly paid specialty—is often cited as a deterrent to their entering a primary care discipline, where incomes have traditionally been much lower.
To spread GME costs among all sources of payment for medical care, the committee has concluded that the societal benefit of well-trained primary care clinicians is so valuable that it should be supported by all health care payers, including self-insured employers, managed care organizations, and private insurers, as well as federal payers.
Given the importance to our society of a well-trained primary care workforce, this committee recommends that a portion of all health care spending go to supporting primary care training. Because managed care organizations have a clear stake in training primary care clinicians to meet their needs, it is logical that they should play an important role in their education and in the financing of graduate medical education. Medicaid contracts with private sector health plans should, for example, acknowledge the positive role of those organizations that are involved in primary care training. Various legislative, regulatory, professional, or marketplace alternatives might be explored to implement this recommendation (described by Petersdorf, 1985). However such support is structured, the future of primary care depends on explicit support for primary care physicians. Although the committee endorses the support of all primary care clinicians, it emphasizes medical training because that training is long and expensive in comparison to the shorter and less expensive training of nurse practitioners and physician assistants.
Recommendation 7.5 All-Payer Support for Primary Care Training
The committee recommends the development of an all-payer system to support health professions education and training. A portion of this pool of funds should be reserved for education and training in primary care.
In making this recommendation, the committee endorses those of several other groups and commissions, including: COGME (1992, 1994, 1995); PPRC (1993, 1994); the Pew Health Professions Commission (1994, 1995); and previous congressional legislative proposals in both the House of Representatives and the Senate. BHP/HRSA estimates that an allocation of 1 percent of all third-party payments including Medicare would generate approximately $5.5 billion; an allocation of 1.2 percent would provide $6.5 billion (COGME, 1994). Several ways of collecting these funds can be considered; these include a tax on health insurance premiums (or gross revenues) or a tax based on the number of covered lives. The committee is not recommending an increase in funding for GME; rather, it believes that funding should come from all sources, not just Medicare or the much smaller sources cited earlier.
Many health insurance plans now refuse to contract with other health plans, delivery systems, or institutions that have higher costs attributable to teaching. Alternatively, they negotiate rates without regard to these costs and thus avoid paying a share of the cost of education; this is essentially the free-rider problem. If all plans contribute to the financing of GME, however, this problem can be circumvented, and competition among health plans can occur without penalizing a few plans that support primary care training.
Support for Advanced Training in Primary Care Sites
Rather than relying overwhelmingly on public payers such as Medicare, the committee has recommended just above that all payers support graduate medical education (indeed, support the education and training of all health professions). It now takes this position one step further and encourages the federal government to implement policies to designate a portion of those funds to support primary care training. This might be done in the context of proposed modifications to federal DME and IME policies now in place in the Medicare program.
Bills before the current Congress have proposed modification of DME and IME Medicare payments, and the financing of GME may be substantially restructured. Whatever the outcome of the current legislation, one thing is certain: if primary care is to achieve the fundamental role in health care that this committee believes it should, continued federal support for training, or at least graduate training, through direct and indirect payments will be necessary. Furthermore, such support will have to be structured to encourage primary care training in ambulatory settings. Finally, financial support should follow the trainee to his or her site of training—whether ambulatory or hospital-based—to produce high-quality primary care clinicians.
Restructuring Medicare GME financing needs to pay specific attention to advanced training for primary care in ambulatory sites. The committee is not alone in this view. For example, in a policy paper for The W.K. Kellogg Foundation, Garg (1995) describes three options for reforming graduate medical educational
financing at the federal level. The first option, which is also endorsed by this committee, was that Medicare should extend its direct and indirect reimbursements to ambulatory settings. Thus, to have funds flow to the settings where primary care training for physicians takes place, the committee makes the following recommendation:
Recommendation 7.6 Support for Graduate Medical Education in Primary Care Sites
The committee recommends that a portion of the funds for graduate medical education be reallocated to provide explicit support for the direct and overhead costs of primary care training in nonhospital sites such as health maintenance organizations, community clinics, physician offices, and extended care facilities.
The committee emphasizes that, regardless of the level of training, support from Medicare trust funds should be reallocated to ensure that a portion will be used for training in primary care settings. How to ensure through regulation that funds going to plans are used for training at these sites is of concern to the committee, but beyond the scope of its charge.
Regarding support for nurse practitioner (and other advanced nursing) education, many observers have advocated shifting in Medicare monies now spent on diploma education to advanced practice nursing education (Aiken and Gwyther, 1995; Pew Health Professions Commission, 1995).
Interdisciplinary Education Of Primary Care Clinicians
Some physicians continue to organize their practices in traditional forms such as single or small, physician-only practices, but multidisciplinary team practice will be an increasingly common mode of practice in the future. Rapid changes of these types are taking place in large managed care organizations. The committee visited many sites that provided team delivery of care and held a three-day conference that explored the roles of health professionals as they practice in teams. In several sites, medical students and other health professional students were incorporated into multidisciplinary teams, and the committee observed the benefits of cross-disciplinary decisionmaking and management that would be useful experiences for students and medical residents during their primary care training. During its site visits, the committee found a remarkable array of organizational uses of health professionals, and it expects that experimentation and evolution will continue. A case study of two mature staff-model health maintenance organizations (HMOs) conducted for the committee illustrated clear diversity and ongoing changes in clinical staffing patterns (Scheffler, Appendix E).
Education must, therefore, equip trainees not only with specific skills but also with the ability to adapt to and create new clinical roles as members of a team.
Surprisingly, however, many students in the health professions are not currently taught in multidisciplinary settings, and they are not exposed at all to working models of team delivery of care in medical and nursing schools or physician assistant programs (MacPherson and Sachs, 1982). The committee urges that this be changed. It is crucial that educational programs recognize the need to prepare their trainees for effective team practice.
Recommendation 7.7 Interdisciplinary Training
The committee recommends that (a) the training of primary care clinicians include experience with the delivery of health care by interdisciplinary teams; and (b) academic health centers work with health maintenance organizations, group practices, community health centers, and other health care delivery organizations using interdisciplinary teams to develop clinical rotations for students and residents.
Educational experiences in interdisciplinary models of practice help trainees to learn the strengths, capabilities, and orientation of other disciplines so that in practice they can more easily appreciate overlapping and complementary skills. The Pew Health Professions Commission recently developed a model curriculum to assist educators in the creation of individualized courses on interdisciplinary collaboration in primary care, and The W.K. Kellogg Foundation has awarded 12 grants for a planning phase to develop interdisciplinary graduate nursing and medical education in community sites. These are important steps.
Such teams need to be truly integrated in how they approach their work and to be organized to provide the kinds of coordinated, comprehensive, and continuous care that primary care trainees are expected to learn. It is difficult to expect such training to take place simply by aggregating medical, nursing, pharmacy, and dental students in the same environment, because of different curricula, scheduling problems, varying levels of preparation, and similar problems.
Rather, students should be incorporated during their training into already functioning teams of practitioners. In this model, students from more than one discipline are assigned to a team that itself reflects an array of health professionals. The committee strongly urges academic health centers to move toward team delivery in their own clinics and inpatient settings and that they structure their primary care clinical practices into teams that can be models for teaching students.
Several academic medical centers have taken this approach. For example, since the early 1970s, the George Washington University School of Medicine and Health Sciences has included primary care residents, medical students, nurse practitioner students, and physician assistant students on multidisciplinary teams
in its university-affiliated HMO and its geriatric practice. Similarly, in 1994, with support from The Josiah Macy, Jr. Foundation, the Harvard Medical School and then Harvard Community Health Plan began to implement an educational model of this sort that incorporates practicing and learning in a managed care setting (Moore et al., 1994).
The committee is acutely aware of the logistic difficulties of accomplishing this goal in institutions that are organized and funded by program (e.g., physician assistant, nursing, medicine) and by department (e.g., family practice, medicine, pediatrics). Further, these programs and departments may have differing, but deeply held, values that make merging curricula and faculty problematic. It also believes, however, that the commonalities of primary care curriculum content and the realities of the practicing environment make multidisciplinary training both desirable and necessary.
Health professionals must develop a common understanding of each other's roles and feel comfortable working with other health professions; they must have confidence about which clinical areas can be appropriately delegated or referred and to whom, and about whose skills augment their own, especially for the complicated medical and social problems that some patients present. Thus, a considerable amount of innovation, experimentation, study, and evaluation of new approaches is called for, and attention should be directed at the best ways to accomplish such teaching.
Recommendation 7.8 Experimentation and Evaluation
The committee recommends that private foundations, health plans, and government agencies support ongoing experimentation and evaluation of interdisciplinary teaching of collaborative primary care to determine how such teaching might best be done.
Although there is no one way that teams should be configured, active exploration of different models can improve our understanding of what works best for patient care and, by extension, what works best for teaching primary care. Three questions about teams have been of particular interest: (1) Who should be on the team? (2) How should work be distributed? (3) Who should provide leadership to the primary care team?
Preparing clinicians to practice in a team is a considerable challenge to health professions educators. Long-held distrust between professions, as well as issues of the autonomy of different disciplines, such as nursing and medicine, underlie systems of education. Furthermore, given the differences in length of training and the costs of that training, facilitating the experience of learning together is understandably difficult.
Despite such financial and political realities, it is nevertheless essential that interdisciplinary education be pursued if there are to be effective primary care
teams. Otherwise, it is unrealistic, despite a common commitment to patient care, to expect different health professions magically to come together after the completion of their programs and work effectively and efficiently to provide primary care services.
Students need to be placed on teams that provide good models of primary care in order to appreciate each clinician's role. Cross-professional preceptorships—such as nurse practitioners working with medical or physician assistant students—convey to all concerned the message that all health disciplines have valuable knowledge and skills. Trainees will also learn to manage the conflicts that are bound to arise as the result of different disciplinary approaches, overlapping roles, and competing demands for team resources and time (Doyle et al., 1993).
Integrated Delivery Systems And Primary Care Training
Cooperation between academic health centers and integrated delivery systems is currently not occurring to any meaningful degree. Barriers include competition for patients, inflexibility and resistance to change on both sides, and failure of leadership to grasp the long-term potential for community benefit. In the committee's view, however, this should change.
Integrated delivery systems (IDSs) can derive benefits from academic centers, and the converse is also true. Shortages of primary care clinicians can be alleviated by creating or participating in primary care residency programs with IDSs providing training sites. To address an oversupply of specialists, academic health centers and IDSs may cooperate in implementing retraining programs in primary care (discussed below under Physician Retraining).
Other health professional students—in particular physician assistants and nurse practitioners—can and should be included in IDS sites as well. For example, IDSs may develop training programs for physician assistants and nurse practitioners and then employ these clinicians to increase the efficiency of care in their system. They may also develop programs to expand the skill of nurses who are no longer needed in hospitals to enable these nurses to take on roles in homes, skilled nursing facilities, and with medical groups that need personnel for telephone triage and care management.
Such teaching practices can thus be models of multidisciplinary training. If the health plan or system has a primary care residency program, the teaching faculty may be given clinical appointments, and these practices can be the center of physician graduate education. Patients will accept that some clinicians are faculty and that residents and other health professional trainees will participate in their care in these sites. Such partnerships or affiliations between academic health centers and IDSs can include clinical rounds and other linkages with the academic health center as a way to ensure that patients have the benefit of up-to-date
clinical knowledge. Teaching practices can also serve as test sites for new models of care and new technologies such as computer-based patient records.
Academic health centers would gain primary care facilities to expand teaching resources in the community. Costs of education and related research could be spread over a broader base. IDSs could provide support for medical, PA education, or advanced practice nursing education in exchange for services and graduates that meet their particular personnel needs. For example, IDS practices that include residents might be able to provide preventive services and continuity of care to a population that otherwise uses an emergency department for its care.
Funding will be a critical issue in considering the role for IDSs in primary care education and training. If funding for teaching in these systems is absent or inadequate, IDSs will refuse to participate or will invest only enough to meet their immediate needs. This may result in short-lived programs and programs of questionable educational quality.
If, however, IDSs are supported by general revenues or other monies for their medical education activities, as recommended above, they are more likely to be longer-term participants. Because IDSs can bring a defined population—even a community—to medical education, they should be understood as indispensable resources for education and training in primary care. Thus, funding to support cooperation between academic medical centers and integrated systems in primary care education is in the public interest and should be encouraged.
Continuing Medical Education
The knowledge base of medicine continues to grow, and clinicians change their practices over time. Attention needs to be paid to how primary care clinicians maintain and improve their skills. Traditional forms of continuing medical education (CME) such as conferences and journals may be augmented increasingly by computer-based methods such as CD-ROM learning materials, telemedicine conferences (both presentations and case conferences), and simulated clinical situations that provide learning experiences tailored to an individual clinician's need and interests. Increasingly powerful search methods are available for locating reference materials, experts, and clinical guidelines through the Internet, and these could be especially useful to those in rural and underserved areas where participation in CME is more difficult. Other promising methods give clinicians feedback about test ordering, prescribing, reminders about needed preventive care, and the like (Davis et al., 1995).
The development of large IDSs may provide especially appropriate settings for relevant CME that can take place in the practice setting itself and bring within reach rural practitioners and those whose care settings are more isolated. This should include training of primary care nurse practitioners, certified nurse midwives, and physician assistants in addition to the training of primary care physicians.
One area of concern to the committee is physicians who are currently practicing in non-primary-care fields and who have not had primary care training. Some of these physicians are now interested in practicing primary care, and some assert that they are already doing so. The basic question is to what extent such physicians, never having had any grounding in primary care, ought now to be regarded as primary care clinicians.
The American Board of Family Practice (ABFP) has taken the position that a full residency is required to qualify one for primary care practice. Nevertheless, the ABFP position may be unrealistic for most subspecialists, because few physicians are able to return to a training program that reduces their incomes by substantial margins for a year or two, and public or private funding is not likely to be available for substantial retraining, especially not for physicians who may already be earning considerable incomes. In any case, it will be necessary (and more practical) to evaluate the results of current shorter programs before concluding that full primary care residencies are needed for retraining purposes.
On the one hand, the committee takes issue with the notion that one can ''self-declare" as a primary care physician if one has never received the relevant training or that a weekend or so of continuing medical education will suffice. The committee strongly affirms that primary care requires special training, but it also believes that requiring currently practicing physicians to undertake a full residency equivalent to those of a newly graduated medical student in order to practice primary care is neither desirable nor feasible. "Retraining" is a middle-ground solution.
Experience with retraining of acute-care-based clinical nurse specialists as nurse practitioners has shown that assumptions about the skills that trainees bring to a program based on their educational background are often unwarranted and that more is required than might have been expected. Given discipline-specific demands, nursing should similarly consider that retraining may require significant education.
Although commonly used, the term "retraining" in this context is something of a misnomer. Many physicians in medical practice have never been trained in primary care, so retraining in reality refers more to the need to augment the training of clinicians who have been engaged solely or predominantly in subspecialty practice (e.g., a subspecialty of medicine or pediatrics, dermatology, ophthalmology, or anesthesiology) or in specialties that generally involve little or no patient contact (e.g., radiology or pathology). The term does not include training for management positions. A host of issues might be raised about training experienced clinicians to provide primary care.
Reasons for Retraining
Several reasons can be given for retraining. First, from a practical standpoint, some managed care plans now require that physicians be classified as primary care physicians either on the basis of specialty training or by self-declaration; in the latter case, the plans may require evidence of some primary care training. Thus, the most recent impetus for retraining is to enable physicians who are already in practice to participate in managed care plans and to continue to see their patients.
Second, retraining would avoid a waste of human resources and clinical experience in situations where specialists, because of an excess supply in some areas, are unable to practice. Although some might argue that it would be more efficient for specialists to reduce their practice or retire early, 50 percent of all physicians in practice today are 40 years of age or younger with many productive years ahead of them, so early retirement is not an option for many.
Third, there are issues of quality of patient care. On the one hand, if physicians self-identify as primary care clinicians without appropriate training, they may provide poor quality care to their patients. Appropriate training can provide the requisite knowledge and skills. By contrast, newly retrained subspecialists in internal medicine, psychiatry, dermatology, OB-GYN, or other fields could bring needed expertise to a primary care team and thus expand its internal resources.
Fourth, the nation needs some additional primary care clinicians now. Retraining could be an efficient way to produce a well-qualified primary care workforce. By implication, training specialists to practice primary care could help to reduce the specialist-generalist imbalance described in Chapter 6. Because subspecialists may not be needed in many rural areas that would welcome a primary care clinician, it might also assist in recruiting and retaining primary care physicians in rural and urban underserved areas.
Kinds of Retraining
In November 1994 the Pew Health Professions Commission identified 25 different retraining efforts in 13 states (Pew Health Professions Commission, 1994). Of these 25 programs, 10 were in existence, another 6 were under development, 6 task forces or committees were examining research initiatives, and 3 groups were addressing retraining issues. Some of these programs are designed specifically for OB-GYNs; others are directed to internal medicine subspecialists or physicians who have been out of the workforce for a time. One program at the Medical College of Pennsylvania has been in existence for about 20 years, but most are very recent. Length and intensity range widely—from a fully accredited residency program at the University of Tennessee to much briefer programs that might last half a day per week for 6 weeks or more and that are usually described as dependent on the needs of the individual.
A single curriculum is not likely to be either adequate or necessary for all clinicians. Different needs by specialty and type of practice expected (e.g., elderly, large group, urban, rural) are likely to be substantial. Lundberg and Lamm (1993) have made the reasonable suggestion that methods be developed to assess the extent to which practicing specialists possess primary care competencies as a means of determining their retraining needs.
The process of adding competencies will almost surely be different for those who, for example, have had three years of training in internal medicine than for those who were trained in a surgical subspecialty. A different curriculum is required to retrain an internist subspecialist who has had exposure to primary care as a resident and has provided some primary care to his or her patients than to retrain an anesthesiologist who has had no primary care training since medical school and has delivered no primary care as a practitioner. The core set of competencies, when developed (see Recommendations 7.2 and 7.3), could form the basis for a retraining curriculum. The length and intensity of the program needed by an individual would be individually determined, and additional modules could be added as necessary and appropriate. In addition, programs might augment ongoing specialty practice with gradually increasing responsibilities in primary care until "retrainees" can demonstrate adequate capabilities in primary care.
Certification After Retraining
The appropriate certification that should be awarded after retraining is an unresolved question. Many trainees would want a certification that would be more widely transferable than one given by the organization in which they practice or even by a specific state. Without an accreditation policy for retraining or certification examination for individuals based on defined competencies, however, it will not be possible to compare or judge the competence of graduates of widely varying programs. In internal medicine, one section of the recertifying boards is on general internal medicine, and this might be one avenue considered for certification.
Recommendation 7.9 Retraining
The committee recommends that (a) curricula of retraining programs in primary care include instruction in the core competencies proposed for development in Recommendations 7.2 and 7.3 and (b) certifying bodies in the primary care disciplines develop mechanisms for testing and certifying clinicians who have undergone retraining for primary care.
A major oversupply of specialists is perhaps a time-limited problem. In the
short term, retraining of specialists may represent an important opportunity to expand the primary physician workforce, but retraining is basically a coping mechanism, not a preferred route to becoming a primary care physician. The committee believes that the specialist oversupply problem may be largely self-correcting in the longer term, as the proportion of newly trained primary care clinicians increases and the supply of specialists decreases. As a start, a study using focus groups to explore issues of specialist retraining has been funded by The Josiah Macy, Jr. Foundation. The committee suggests that foundations and federal agencies such as HRSA and HCFA conduct or support studies on retraining. These studies should include examination of needed competencies and the feasibility and outcomes of various approaches to retraining for various kinds of clinicians. Questions that might be studied include the following:
- What is the level of interest in retraining and who are the interested clinicians?
- Which critical primary care competencies are already known and which need to be taught? Can this be viewed as expanding an impressive set of skills rather than starting over?
- What types of physicians are successfully retrained and enter primary care practice?
- What sort of retraining is most appropriate and for what kinds of programs? What elements of GME and CME work best for retraining of the sort contemplated here? Are short CME courses, part-time study, tailored mini-residencies, full residencies, or on-the-job training adequate?
- What are the characteristics of appropriate mentors or preceptors for experienced colleagues, and are these characteristics different for new residents? What are the most appropriate learning methods for mid-career physicians?
- Who should do the training? Medical schools? Professional associations? HMOs?
- Who should pay for retraining—the trainee, the organizations that will or have hired them, or state or federal government?
- Does the incorporation of retrained specialist and subspecialty physicians into a primary care team augment that team's resources, add a new dimension to the team's capabilities, and allow it to function more effectively?
- What sorts of standards are needed for retraining? Currently the specialty boards such as those for family practice, internal medicine, and pediatrics have taken different positions. What types of standards could be used, or imposed, by the managed care industry?
If primary care is to move in the directions advocated by this committee, then many aspects of health professions education and training will need to be restructured.
This chapter explored the changes likely to be required in undergraduate and graduate training, argued that clinical training ought to involve exposure to multidisciplinary team practice, and examined issues of retraining physicians for primary care.
To reach these goals, the committee put forward several recommendations. With respect to undergraduate medical education, the committee was concerned that students gain experience in primary care settings; with respect to graduate training, the committee explored issues of residency programs in family practice, internal medicine, and pediatrics and the value of primary care tracks. Education in ambulatory sites, community health clinics, and managed care organizations is essential to create a primary care workforce that will serve the needs of men and women, children and adults, rich and poor, individuals in rural and urban locations, and persons of all ethnic backgrounds.
More broadly, the committee examined questions of advanced training for all primary care clinicians and called attention to the need for the development of a set of common core competencies for all primary care clinicians. In addition, the committee highlighted its concerns about two special areas of emphasis—communication skills and cultural sensitivity.
A major consideration for the committee was financial support for primary care training. Consistent with earlier recommendations about universal coverage for health care, the committee called for an all-payer system to support health professions education and training, with some of this support reserved for primary care and directed to training in nonhospital sites such as offices, clinics, and extended care facilities. Adopting the recommendations in this chapter will require a realignment of funding and power to create incentives for different institutional behaviors (for example, in academic health centers and in integrated delivery systems) to focus on primary care and on training in ambulatory as well as hospital-based settings. Similarly, funding mechanisms for graduate medical education will need to be revamped to support training sites other than the traditional hospital base. Because the graduates of these programs will increasingly be needed by integrated delivery systems and the managed care industry generally, the committee believed that all payers should share the burden of establishing and maintaining the required educational infrastructure.
Finally, the committee examined other elements of education and training and called for the development of more innovative and interdisciplinary training programs. It also advocated that better mechanisms be created by which nonprimary-care physicians can be formally and adequately retrained for primary care practice.
AAMC (Association of American Medical Colleges). Task Force on the Generalist Physician, October 8, 1992.
AAMC. Academic Medicine and Health Care Reform: Roles for Medical Education in Health Care Reform. Washington, D.C.: AAMC, 1994.
ACOG (American College of Obstetricians and Gynecologists). Primary Care Provisions: Program Requirements for Residency Training in Obstetrics and Gynecology. Accreditation Council for Graduate Medical Education, February 1995 (effective January 1996).
Aiken, L.H. and Gwyther, M.E. Medicare Funding of Nurse Education. The Case for Policy Change. Journal of the American Medical Association 273:1528–1532, 1995.
Allshouse, K.D. Treating Patients and Individuals. Pp. 19–27 in Through the Patient's Eyes: Understanding and Promoting Patient-Centered Care. M. Gerteis, S. Edgman-Levitan, J. Daley, and T.L. Delbanco, eds. San Francisco: Jossey-Bass, 1993.
Alpert, J.J., and Charney, E. The Education of Physicians for Primary Care. DHEW Publ. No. (HRA)74–3113. Rockville, Md.: Bureau of Health Services Research, Health Resources and Services Administration, Department of Health, Education, and Welfare, 1973.
ASIM (American Society of Internal Medicine). What's So Special About Being an Internist? A Resource Kit for Internists on Internal Medicine Preceptorship Programs. Washington, D.C.: ASIM, 1995.
Barker, L.R. What and How to Teach. Curriculum for Ambulatory Care Training in Medical Residency: Rationale, Attitudes, and Generic Proficiencies. Journal of General Internal Medicine 5 (Jan./Feb. Suppl.):S3–S14, 1990.
Bureau of Health Professions, Division of Nursing. Fact sheets on advanced nurse education; nurse practitioner and nurse midwifery grants; professional nurse traineeships; and nurse anesthetist traineeships. Rockville, Md.: Division of Nursing, Health Resources and Services Administration, Public Health Service, 1994.
Christakis, N.A. The Similarity and Frequency of Proposals to Reform U.S. Medical Education. Journal of the American Medical Association 274:706–711, 1995.
COGME (Council on Graduate Medical Education). Third Report. Improving Access to Health Care Through Physician Workforce Reform: Directions for the 21st Century. Rockville, Md.: Health Resources and Services Administration,Department of Health and Human Services, 1992.
COGME. Recommendations to Improve Access to Health Care Through Physician Workforce Reform. Rockville, Md.: Health Resources and Services Administration, Department of Health and Human Services, 1994.
COGME. Council on Graduate Medical Education Seventh Report—COGME 1995 Physician Workforce Funding Recommendation for Department of Health and Human Services' Programs. Rockville, Md.: Health Resources and Services Administration, Department of Health and Human Services, 1995.
Davis, D.A., Thomson, M.A., Oxman, A.D. and Haynes, R.B. Changing Physician Performance. A Systematic Review of the Effect of Continuing Medical Education Strategies. Journal of the American Medical Association 274:700–705, 1995.
Doyle, D., Hanks, W.C., and MacDonald, N. The Interdisciplinary Team. Pp. 21–28 in Oxford Textbook of Palliative Medicine. New York: Oxford University Press, 1993.
Eisenberg, J.M. Special Article. How Can We Pay for Graduate Medical Education in Ambulatory Care? New England Journal of Medicine 320:1525–1531, 1989.
Family Health Foundation of America. MERIT Project. A Compendium of Topics for Curricular Development in Family Practice. Vols. I and II. Kansas City, Mo.: American Academy of Family Physicians Foundation, 1983.
FCIMCTF (Federated Council of Internal Medicine Curriculum Task Force). November 1995 draft, Curriculum report developed by the FCIM Task Force on Curriculum, 1996, forthcoming.
Fein, E.B. More Young Doctors Forsake Specialty for General Practice. New York Times, October 16, 1995.
GAO (General Accounting Office). Medical Education: Curriculum and Financing Strategies Need to Encourage Primary Care Training. Washington, D.C.: General Accounting Office, 1994.
Garg, M.L. Primary-Care Physicians: The Case for Reforming Public Financing of Medical Education. January, 1995, working paper.
HRSA. Physician Assistants in the Health Workforce 1994. Rockville, Md: U.S. Public Health Service, no date.
IOM (Institute of Medicine). Medical Education and Societal Needs: A Planning Report for the Health Professions. Washington, D.C.: National Academy Press, 1983.
IOM. Dental Education at the Crossroads: Challenge and Change. M.J. Field, ed. Washington, D.C.: National Academy Press, 1995.
IOM. The Nation's Physician Workforce: Options for Balancing Supply and Requirements. K.N. Lohr, N.A. Vanselow, and D.E. Detmer, eds. Washington, D.C.: National Academy Press, 1996.
JAMA (Journal of the American Medical Association). Appendix II. Graduate Medical Education. Table 1. Resident Physicians on Duty in ACGME-Accredited and in Combined Specialty Graduate Medical Education (GME) Programs in 1993. Journal of the American Medical Association 272:725–726, 1994.
Kantor, S.M., and Griner, P.F. Educational Needs in General Internal Medicine as Perceived by Prior Residents. Journal of Medical Education 56:748–756, 1981.
Kassebaum, D.G., and Haynes, R.A. Relationship Between Third-Year Clerkships in Family Medicine and Graduating Students' Choices of Family Practice Careers. Academic Medicine 67:217–219, 1992.
Kern, D.C., Parrino, T.A., and Korst, D.R. The Lasting Value of Clinical Skills. Journal of the American Medical Association 254:70–76, 1985.
Kimball, H., and Young, P. Commentary. A Statement on the Generalist Physician From the American Boards of Family Practice and Internal Medicine. Journal of the American Medical Association 271:315–316, 1994.
Linn, L.S., Brook, R.H., Clark, V.A., et al. Evaluation of Ambulatory Care Training by Graduates of Internal Medicine Residencies. Journal of Medical Education 61:293–302, 1986.
Lipkin, M., Quill, T.E., and Napodano, R.J. The Medical Interview: A Core Curriculum for Residencies in Internal Medicine. Annals of Internal Medicine 100:277–284, 1984.
Lipkin, M., Levinson W., Barker, R., et al. Primary Care Internal Medicine: A Challenging Career Choice for the 1990s. Annals of Internal Medicine 112: 371–378, 1990.
Lundberg, G.D., and Lamm, R.D. Editorial. Solving Our Primary Care Crisis by Retraining Specialists to Gain Specific Primary Care Competencies. Journal of the American Medical Association 270:380–381, 1993.
MacPherson C., and Sachs, L.A. Health Care Team Training in U.S. and Canadian Medical Schools. Journal of Medical Education 57:282–287, 1982.
Martini, C.J.M., Veloski, J.J., Barzansky, B., et al. Medical School and Student Characteristics that Influence Choosing a Generalist Career. Journal of the American Medical Association 272:661–668, 1994.
McPhee, S.J., Mitchell, T.F., Schroeder, S.A., et al. Training in a Primary Care Internal Medicine Residency Program: The First Ten Years. Journal of the American Medical Association 258:1491–1495, 1987.
Moore, G.T., Inui, T.S., Ludden, J.M., and Schoenbaum, S.C. The "Teaching HMO:" A New Academic Partner. Academic Medicine 69:595–600, 1994.
National Governors' Association, Curley, T., Orloff, T., et al. Health Professions Education Linkages: Community-Based Primary Care Training . Washington, D.C.: National Governors' Association, 1994.
New York Times. Dartmouth Redesigns Medical Training to Give Future Doctors a Human Touch. New York Times, (Education, part 1–3): Sept. 2, 1992.
Noble, J., Bithoney, W., MacDonald, P., et al. The Core Content of a Generalist Curriculum for General Internal Medicine, Family Practice and Pediatrics. Journal of General Internal Medicine 9(Supplement 1):S31–S42, 1994.
NONPF (National Organization of Nurse Practitioner Faculties). Curriculum Guidelines & Program Standards for Nurse Practitioner Education. Washington, D.C.: NONPF, 1995.
Novack, D.H., Dube, C., and Goldstein, M.G. Teaching Medical Interviewing. A Basic Course in Interviewing and the Physician-Patient Relationship. Archives of Internal Medicine 152:1814–1820, 1992.
Osborn, E.H.S. Factors Influencing Students' Choices of Primary Care or Other Specialties. Academic Medicine 68:572–574, 1993.
Petersdorf, R.G. A Proposal for Financing Graduate Medical Education. New England Journal of Medicine 312:1322–1324, 1985.
Petersdorf, R.G., and Turner, K.S. Medical Education in the 1990s—and Beyond: A View from the United States. Academic Medicine 70(Suppl.):S41–47, 1995.
Pew Health Professions Commission. Primary Care Workforce 2000. Federal Policy Paper. San Francisco: Pew Health Professions Commission, 1994.
Pew Health Professions Commission. Shifting the Supply of Our Health Care Workforce. A Guide to Redirecting Federal Subsidy of Medical Education. San Francisco: Pew Health Professions Commission, 1995.
PPRC (Physician Payment Review Commission). Annual Report to Congress, 1993. Washington, D.C.: PPRC, 1993.
PPRC. Annual Report to Congress, 1994. Washington, D.C.: PPRC, 1994.
Rigoglioso, R.L. Multiculturalism in Practice. The Picker Report 3(2):1,12,13, 1995.
Rivo, M.L., Saultz, J.W., Wartman, S.A., et al. Defining the Generalist Physician's Training. Journal of the American Medical Association 271:1499–1504, 1994.
Schroeder, S.A., Zones, J.S., and Showstack, J.A. Academic Medicine as a Public Trust. Journal of the American Medical Association 262:803–812, 1989.
SGIM/CDIM (Society of General Internal Medicine and the Clerkship Directors in Internal Medicine). Core Medicine Clerkship Curriculum Guide: A Manual for Faculty. A.H. Goroll, and G. Morrison. (project investigators). Rockville, Md.: Division of Medicine, Bureau of Health Professions, Health Resources and Services Administration, 1995.
Shabazz, C.D., and Carter, J.H. Multicultural Diversity in Medicine (editorial). Journal of the National Medical Association 84:313–314, 1992.
Sherer, J. New Waves: Hospitals Struggle to Meet the Challenge of Multiculturalism Now—And in the Future. Hospitals May 20, pp. 29–31, 1993.
Stimmel, B. The Crises in Primary Care and the Role of Medical Schools: Defining the Issues. Journal of the American Medical Association 268:2060–2065, 1992.