Health disparities in the United States are among the highest in the developed world, and reducing them is a major public health priority.1–3 Any strategy to reduce disparities must address the mismatch between the greater burden of morbidity in disadvantaged populations and these populations’ limited access to care compared with people from middle and upper socioeconomic groups. Because the supply, specialty mix, and geographic distribution of physicians are important determinants of a population’s access to care, successful strategies to reduce disparities must address the physician workforce needs of underserved populations.
Our aim in this article is not to perform a broad review of the literature about the challenges of supplying physicians to underserved urban and rural areas of the United States. Instead, we will highlight several important principles and incorporate them into a framework for building a physician workforce that would be responsive to the needs of America’s disadvantaged populations. Our framework is based on the following principles:
- Health disparities will not be completely eliminated until underserved people, communities, and populations—rural and urban—have physicians available locally to care for them.
- U.S. medical schools have a responsibility to educate physicians who will care for the entire population, while meeting societal goals for effectiveness, efficiency, and equity.
- Achieving the objectives of decreased health disparities and physician supply for all population subgroups requires training students interested in delivering primary care to underserved people.4
- The medical students we currently train are not choosing to enter primary care or serve underserved populations in adequate numbers and, therefore, we are selecting the wrong students.
- The epidemiology of medical student career choice is sufficiently understood to permit medical schools to identify and admit students likely to work in underserved communities.
- There are successful model programs that improve physician supply to underserved areas, and these programs should be expanded dramatically.
In this article we emphasize how medical schools, individually and in aggregate, can focus their efforts on meeting the needs of the disadvantaged for access to care. Much of the debate about physician workforce policy has centered around the need to expand physician supply,5 but expansion without a change in the prevailing determinants of physician distribution is unlikely to meet the needs of disadvantaged Americans. Other important aspects of a comprehensive plan for physician recruitment and retention in underserved areas, including strategies targeting K–12 education and physician reimbursement incentives, are outside the scope of this paper.
Principle 1: Health Disparities Will Not Be Completely Eliminated Until Underserved People, Communities, and Populations—Rural and Urban—Have Physicians Available Locally to Care for Them
This principle is a basic and core assumption for anyone (presumably including those in the medical profession) who believes medical care is good for health. Although medical care may not be the primary determinant of population health, it is an important one,6 and the greatest benefit of medical care is realized when populations move from “none” to “any” on the spectrum of available care.7
Principle 2: U.S. Medical Schools Have a Responsibility to Educate Physicians Who Will Care for the Entire Population, While Meeting Societal Goals for Effectiveness, Efficiency, and Equity
Whether medical schools have a social responsibility for population health is an old question.8–10 We believe no credible argument to the contrary has ever been made. The first academic unit for medicine in the English-speaking world, at King’s College, Aberdeen, Scotland, was created for the “pursuit of health in the service of society.”11 Additional compelling arguments for the role of medical education in promoting the health of populations track from Virchow,12 to recent work by the Association of American Medical Colleges (AAMC) Advisory Panel on the Mission and Organization of Medical Schools (APMOM).9 Modern perspectives, summarized by ethicist and health researcher Tom Inui, suggest that by spending public funds in support of medical education and medical care, society acknowledges that medical care is an important societal resource. Public support of medical schools—and the schools’ acceptance of that support—enters them into an implicit contract to serve the public interest and to promote equity of health benefits across all populations, advantaged and disadvantaged.
Principle 3: Achieving the Objectives of Decreased Health Disparities and Physician Supply for all Population Subgroups Requires Training Students Interested in Delivering Primary Care to Underserved People
The problem of recruiting students to work in rural and inner-city America must be understood in the broader context of declining student interest in primary care careers, because it is family physicians and general internists who provide the bulk of care to residents of underserved areas.13–16 Fewer students in the primary care pipeline will almost certainly mean fewer physicians serving disadvantaged patients.17 Thus, promoting primary care as a desirable career option is a general strategy that will benefit the underserved. Evidence that this is true has been compiled by Starfield and her collaborators6,18–24 during the past two decades: there is an important link between an effective primary care workforce, better population-health outcomes, and lower health care costs.
Principle 4: The Medical Students We Currently Train Are Not Choosing to Enter Primary Care or Serve Underserved Populations in Adequate Numbers and, Therefore, We Are Selecting the Wrong Students
It is clear that our long-standing approaches to creating a physician workforce that is responsive to disadvantaged communities are not working, and, indeed, that we are losing ground. Thirty years after the establishment of the National Health Service Corps to help provide physicians for rural and urban underserved areas, we are no closer to meeting this need. Dwindling numbers of medical students are choosing rural or inner-city practice. Similarly, the downturn in U.S. medical school seniors choosing family medicine, the most common specialty of rural physicians,16 has been dramatic.25,26 General internists, another important source of care for rural and inner-city residents, are also in decline: from 1998 to 2003, the percentage of third-year residents in internal medicine planning to pursue careers in general medicine dropped from 54% to 27%, and only 19% of first-year residents in 2003 were planning such careers.27 Primary care physicians practicing in poverty zip codes have continued to decline from the year 1990 to the present.28
Ethnic and racial minority groups constitute an increasing proportion of the U.S. population, and medical students from underrepresented minority groups and low-income backgrounds are more likely to care for underserved populations.29 However, the gap between the increasingly diverse population of our communities and the much less diverse group enrolled in our medical schools is widening. In 2000, African Americans constituted 12.3% and Latinos 12.5% of the U.S. population, yet they accounted for only 7.4% and 6.7% of entering medical students, and the proportion of matriculants from these minority groups is falling.30 In its “Diversity in the Physician Workforce: Facts and Figures 2006,” the AAMC notes that “Blacks, Hispanics/Latinos, and Native Americans comprise only 6.4% of all physicians graduating from U.S. allopathic medical schools.”31
Diversity in social class, or socioeconomic origins, is another important measure against which medical education is failing. We continue to produce physicians who come from upper-middle and upper-income families. Sixty percent of medical students come from families in the top 20% of income, whereas only 20% come from the bottom 60%, and this trend is even more dramatic than the trends in race/ethnicity discussed above30; from 1997 to 2004 there has been a marked redistribution of entering students from lower to higher income brackets. Low-income minority students are thus doubly affected: in 2004, only 137 African Americans, 52 Latinos, and 231 non-Latino whites whose parents’ annual incomes were below $20,000 gained admission to medical school. This income group, which comprises 22% of the population,32 is thus represented by less than 3% of entering medical students.30
Why have such inequities persisted, and even increased? A recent paper from the United Kingdom provides an illustration of how medical educators may directly contribute to admission inequalities for those from underserved or low-income backgrounds.33 In describing the difficulties of diversifying the medical student body, the authors note that social exclusion is sometimes raised as a concern in medical school admissions, but they dismiss this concern, saying “It has been suggested that a pool of talented individuals capable of becoming good doctors is excluded by current admission methods. Even if that were so (and we know of no evidence to support it) there is no basis for believing that intellectual aptitude tests are capable of identifying them.” The authors thus demonstrate that, if one takes a sufficiently narrow perspective of medical school “success” (i.e., grades), it is hard even to see the problem with aptitude-based selection criteria for admissions and advancement in medical education. That is, selecting students with high grades is the best predictor of students who will achieve high grades, but it is not the best predictor of who will enter the specialties and practice in the areas where they are most needed. Only when we step back to a societal-level perspective and understand the service needs of disadvantaged populations can we observe a problem that relates to the preadmission applicant pipeline and eventual practice patterns in graduates.
To do this well requires us to follow the lead of medical practice, business, and program development and examine outcome rather than process measures. That is, we need to know whether we are getting the results we want, rather than simply knowing whether we are doing what we wanted to do. If the outcome we desire is to produce physicians who will provide care to our entire population, to the most as well as to the least needy, we are failing, despite the degree to which we “succeed” as measured by interval process variables such as grades, U.S. Medical Licensing Examination scores, or residency match rates. By defining a “qualified” applicant as one who meets certain cognitive criteria that predict little more than success in a preclinical biomedical curriculum, we select a population largely uninterested in generalist specialties and care for the underserved, and unlikely to become the kinds of doctors who enter these practices. The principles we have outlined above provide a basis for the selection of students who are not only likely to successfully complete the process of medical education, but who are also likely to make a positive difference to the health of our society—including the underserved—once they enter practice.
Principle 5: The Epidemiology of Medical Student Career Choice is Sufficiently Understood to Permit Medical Schools to Identify Students Likely to Work in Underserved Communities
Extensive work has identified a cluster of characteristics in medical school applicants that predict an increased probability of eventual rural or urban underserved practice. These include gender, older age at matriculation, rural or urban underserved community of influence, and relatively lower family income.29,34–38 It is important to note that the demographic predictors of practicing in rural or urban underserved locations differ. Women and African Americans more often choose to practice in urban underserved areas, whereas men and underrepresented minorities other than African Americans more often choose rural practice. In the case of rural practice, more than 20 years of experience in the Physician Shortage Area Program at Thomas Jefferson University have shown that two characteristics identifiable at enrollment—growing up in a rural area, and a plan to enter family practice on entering the first year of medical school—were strong independent predictors of rural recruitment and retention.39 A key lesson from this experience supports our emphasis on selection: although curricular elements are important in supporting students planning rural careers, curriculum is less important than selection, because curriculum is unlikely to change the career plans of students who are not predisposed towards rural careers.
We have relatively little data about how a host of other beliefs and personality traits (political orientation, altruism, moral maturity) affect specialty choice and practice location. Because not all minority, rural, older, or low-income students will practice in underserved areas, and because there are likely to be some high-scoring students from high-income, suburban, nonunderrepresented backgrounds who will, the study and analysis of these predictors is important. Changing demographics suggest that society will have a greater and greater need to identify applicants whose histories demonstrate that they have the values and commitment to work in underserved communities. Selection will benefit more from tracking the prior activities of a specific applicant (e.g., volunteer work, Peace Corps/Americorps, work in underserved or rural communities) than from simply searching for empathic and caring statements on an essay. That is to say, until we find a way to measure attitudes that is not susceptible to prospective medical students gilding their applications, we need to rely on validated predictors that cannot be falsified: demographics and past behavior.
Principle 6: There Are Successful Model Programs That Improve Physician Supply to Underserved Areas, and These Programs Should Be Expanded Dramatically
A number of medical school programs have been developed and implemented to promote graduates’ eventual practice among the underserved, and many of these programs have continued for several decades.40–43 In addition, many schools have postbaccalaureate programs designed to help minority and disadvantaged college students gain admission to medical school, and these have been demonstrated to be successful.44 It is useful to consider the potential benefit of expanding such programs across the nation’s landscape of physician workforce training.
Four reports in Academic Medicine’s August 2005 issue40,45–47 describe excellent programs to increase the number of students entering and staying in rural practice; they are worthy of emulation, and in some respects they are not unlike programs at many medical schools across the country. From a purely methodological perspective, it can be argued that the evaluation of their success is “complicated” by selection bias; that is, the same characteristics that drive students to opt for participation in these programs also are predictive of eventual practice among the underserved, and this correlation, rather than the features of the special programs themselves, may account for the programs’ effects. From another perspective, however, this “bias” has identified many of the applicant characteristics (identified in Principle 5) that medical schools can use to indirectly address the tenacity of our nation’s health disparities by using the characteristics in preadmission screening; that is, creating a “selection bias” of medical school applicants who possess these characteristics. Framing the issue in this way allows us to understand better the weaknesses of our existing admissions process by highlighting the current lack of consideration of qualities that may indicate an applicant’s likelihood to care for the underserved after graduation. It also allows us to begin evaluating the relative costs of building new programs versus altering selection criteria for applicants of current programs in pursuit of eventual equitable distribution of practicing physicians. Selecting the “right” students into medical school in the first place gives us more time to evaluate what is “right” and “wrong” with the medical school curricula and its influences on the students’ intentions to practice among the underserved. The 2005 studies clearly indicate the importance of selecting the “right” students for medical school, indeed consciously redefining the idea of “qualified.” As Whitcomb notes, “ …the favorable outcomes of the programs are simply the result of preselection bias. And for some students this may be the case. But so what¡”48 We should be less interested in defining a “qualified” applicant as one who will be capable of navigating through medical school, and more interested in defining a “qualified” applicant as one who is likely to help meet society’s needs.
How Can We Meet the Needs of the Underserved?
In the above discussion, we attempted to demonstrate that:
- there is a major problem with health disparities in the United States;
- one requirement for addressing these disparities is to have more doctors, especially primary care doctors, practicing in areas that are currently underserved;
- fewer, rather than more, medical school graduates are choosing the specialties and practice locations that will provide the medical care necessary (if not sufficient) to help reduce health disparities;
- although there are many factors, including money (generally higher reimbursement in non–primary care specialties) and lifestyle (time off from work, preference for the opportunities available in urban areas) that affect specialty and practice location choice, medical schools should do what they can, through the admissions process and curriculum, to affect this choice; and
- what schools are doing now is not working.
What would work? In this article we identify programs that do work (Principle 6), and we contend that the reason that they work is, in significant part, because of whom they select to participate (Principle 5). Thus, the lesson for medical schools is that they must take a different mix of students to meet the health care needs of society. Many, if not most, medical schools have programs for identifying students who meet either the demographic (coming from rural and underserved areas or from underrepresented ethnic and socioeconomic groups) or personal (demonstrated commitment to the underserved through prior work) characteristics identified in Principle 5. These programs, however, have not been sufficiently successful, or of adequate size, to significantly change the service profile to our most underserved population segments. We would argue that this is because such characteristics are given secondary priority, after high academic performance characteristics. We first select a group likely to get high grades in medical school, and then look for those characteristics likely to meet societal needs for medical practice. This would not be a problem if these two criteria were not in conflict, but very often they are. Cognitive-score criteria select overwhelmingly for upper- and upper middle-class suburbanites who have gone to academically strong public and private schools. Schools in rural and inner-city neighborhoods, without comparable resources, cannot produce the same proportion of high-scoring graduates as do suburban schools.49 Even when the schools impart a solid cognitive base, disadvantaged students may not have the socialization necessary to move successfully through application, interview, and matriculation to join a largely upper-class cohort of undergraduates.50
An alternative approach, one much more likely to be successful in changing the outcomes of medical schools by increasing the choice of primary care specialties and underserved practice locations, would be to change the order that the medical school admissions algorithms use to consider applicant characteristics. We suggest that the first consideration of “qualified” students be based on those characteristics that suggest humanism (age and prior life experiences, volunteerism and commitment to community, socioeconomic and ethnic origin, etc.), orientation toward service, or ability to promote diversity among the student body. Only then should we consider previous academic performance, selecting those in the first group whose grades and MCAT scores suggest their ability to succeed in the medical school curriculum. Indeed, as a general rule, we should set threshold values for cognitive indicators on the basis of levels previously demonstrated to indicate probability of success in the curriculum, rather than “ranking” students on the basis of grades. Of course, just as we now take some students whose background characteristics and demonstrated commitment may be seen to overcome weaker performance in their academic cognitive performance, in the future we should take some students whose stellar academic accomplishments make us overlook their deficiencies in the core personal and background characteristics that we know correlate with eventual practice in the target areas we have identified.
Is this likely to be a hard sell? No doubt. Deans’ offices are seldom evaluated on the extent to which their graduates fill niches in underserved areas. Our current admissions committees are overwhelmingly made up of the people from the same backgrounds as our current students—only 14% of medical school admissions committee members are underrepresented minorities, and in half of all medical schools the actual number is zero or one.51 If, however, our educational goal is to meet an outcomes need based on objective criteria of access to care—that is, more doctors working in underserved communities (which may include subspecialists working in rural areas and small cities)—this is where we need to go. Our deans should be evaluated on the degree to which their graduates begin to meet the health needs of all U.S. population subgroups, and if the current members of our admissions committees find such proposals unacceptable, then we believe membership on those committees needs to change.
Identifying Our Mission and Achieving Our Goal
The fundamental question that medical school admissions committees must confront is this: Is the mission of medical schools to feed market demand for a commodity, operating under the tacit assumption that the “invisible hand” will manage population health, or is the mission to address population health more actively with explicit social objectives, such as increasing health care access and equity? Although some economists believe that health care is a “common” or possibly a “luxury good” used in proportion to income,52 this is at odds with the ethical traditions of medicine, and medical schools need not acquiesce to this view. If the responsibility for population health is explicitly embraced by schools, then the following facts should serve as the basis for action. First, the distribution of U.S. physicians is changing in a way that is likely to increase rather than decrease disparities in health. Second, without changes in medical school admissions policies, the goal of a physician workforce responsive to the needs of America’s underserved is unlikely to be achieved. Third, the epidemiology of medical students’ specialty and geographic choices is now understood and could be applied to meeting policy objectives to reduce health care disparities. Reconsidering medical school admissions policies on the basis of this understanding has the potential to improve America’s health. The health care marketplace has failed to meet the needs of America’s disadvantaged residents. A coordinated effort among our medical schools can help rectify that failure, but only if academic leaders recognize the shortcomings of present approaches and create a system that will support equity and fairness in the delivery of medical care.
1 Agency for Healthcare Research and Quality. National Healthcare Disparities Report: Department of Health and Human Services. Washington, DC: Agency for Healthcare Research and Quality. 2004;05–0014.
2 Kelley E, Moy E, Stryer D, Burstin H, Clancy C. The national healthcare quality and disparities reports: an overview. Med Care. 2005;43(3 suppl):I3–I8.
3 Siegel S, Moy E, Burstin H. Assessing the nation’s progress toward elimination of disparities in health care. J Gen Intern Med. 2004;19:195–200.
4 Politzer RM, Hardwick KS, Cultice JM, Bazell C. Eliminating primary care health professional shortage areas: the impact of Title VII generalist physician education. J Rural Health. 1999;15:11–20.
5 Salsberg E, Grover A. Physician workforce shortages: implications and issues for academic health centers and policymakers. Acad Med. 2006;81:782–787.
6 Shi L, Starfield B, Politzer R, Regan J. Primary care, self-rated health, and reductions in social disparities in health. Health Serv Res. 2002;37:529–550.
7 Donabedian A. Explorations in Quality Assessment and Monitoring Vol. 1. The Definition of Quality and Approaches to Its Assessment. Ann Arbor, Mich: Health Administration Press; 1980.
8 Inui TS. Reform in medical education: a health of the public perspective. Acad Med. 1996;71(1 suppl):S119–S121.
9 McCurdy L, Goode LD, Inui TS, et al. Fulfilling the social contract between medical schools and the public. Acad Med. 1997;72:1063–1070.
10 Inui T. The social contract and the medical school’s responsibilities. In: White KL, Connelly JE, eds. The Medical School’s Mission and the Population’s Health: Medical Education in Canada, the United Kingdom, the United States, and Australia. New York: Springer Verlag; 1992:23–59.
11 Lewkonia RM. The missions of medical schools: the pursuit of health in the service of society. BMC Med Educ. 2001;1:4.
12 Virchow RLK, Rather LJ. Collected Essays on Public Health and Epidemiology. Canton, Mass: Science History Publications, USA; 1985.
13 Green LA, Phillips RL Jr. The family physician workforce: quality, not quantity. Am Fam Physician. 2005;71:2248–2253.
14 Pugno PA, Epperly TD. Residency Review Committee for Family Medicine: an analysis of program citations. Fam Med. 2005;37:174–177.
15 Schwartz MD, Basco WT Jr, Grey MR, Elmore JG, Rubenstein A. Rekindling student interest in generalist careers. Ann Intern Med. 2005;142:715–724.
16 American College of Physicians. Rural primary care. Ann Intern Med. 1995;122:380–390.
17 Phillips RL Jr, Dodoo MS, Green LA. Adding more specialists is not likely to improve population health: is anybody listening? Health Aff (Millwood). 2005;(suppl):W5-111–W115-114.
18 Starfield B, Shi L, Macinko J. Contribution of primary care to health systems and health. Milbank Q. 2005;83:457–502.
19 Shi L, Macinko J, Starfield B, Wulu J, Regan J, Politzer R. The relationship between primary care, income inequality, and mortality in US States, 1980–1995. J Am Board Fam Pract. 2003;16:412–422.
20 Shi L, Macinko J, Starfield B, et al. Primary care, infant mortality, and low birth weight in the states of the USA. J Epidemiol Community Health. 2004;58:374–380.
21 Starfield B. Is primary care essential? Lancet. 1994;344:1129–1133.
22 Starfield B. Public health and primary care: a framework for proposed linkages. Am J Public Health. 1996;86:1365–1369.
23 Starfield B, Shi L. Policy relevant determinants of health: an international perspective. Health Policy. 2002;60:201–218.
24 Starfield B. Primary Care: Balancing Health Needs, Services, and Technology. Rev. ed. New York: Oxford University Press; 1998.
25 Pugno PA, Schmittling GT, Kahn NB Jr. Entry of US medical school graduates into family medicine residencies: 2004–2005 and 3-year summary. Fam Med. 2005;37:546–554.
26 Pugno PA, Schmittling GT, Fetter GT Jr, Kahn NB Jr. Results of the 2005 national resident matching program: family medicine. Fam Med. 2005;37:555–564.
27 Garibaldi RA, Popkave C, Bylsma W. Career plans for trainees in internal medicine residency programs. Acad Med. 2005;80:507–512.
28 Luo W, Wang F, Douglass C. Temporal changes of access to primary health care in Illinois (1990–2000) and policy implications.J Med Syst. 2004;28:287–299.
29 Komaromy M, Grumbach K, Drake M, et al. The role of black and Hispanic physicians in providing health care for underserved populations. N Engl J Med. 1996;334:1305–1310.
30 Association of American Medical Colleges. Association of American Medical Colleges Matriculating Student Questionnaire. Medical Education Division. Washington, DC: Association of American Medical Colleges; 2005.
31 Association of American Medical Colleges. Diversity in the Physician Workforce: Facts and Figures 2006. Washington, DC: Association of American Medical Colleges; 2006.
32 DeNavis-Wait CPB, Lee CH. Income, Poverty, and Health Insurance Coverage in the United States: 2004. Washington, DC: U.S. Census Bureau; 2004.
33 McManus IC, Powis DA, Wakeford R, Ferguson E, James D, Richards P. Intellectual aptitude tests and A levels for selecting UK school leaver entrants for medical school. BMJ. 2005;331:555–559.
34 Petersdorf RG, Turner KS, Nickens HW, Ready T. Minorities in medicine: past, present, and future. Acad Med. 1990;65:663–670.
35 Xu G, Veloski J, Hojat M, Politzer RM, Rabinowitz HK, Rattner SL. Factors influencing primary care physicians’ choice to practice in medically underserved areas. Acad Med. 1997;72(10 suppl 1):S109–S111.
36 Kassebaum DG, Szenas PL, Caldwell K. Educational debt, specialty choices, and practice intentions of underrepresented-minority medical school graduates. Acad Med. 1993;68:506–511.
37 Moy E, Bartman BA. Physician race and care of minority and medically indigent patients. JAMA. 1995;273:1515–1520.
38 Ko M, Edelstein RA, Heslin KC, et al. Impact of the University of California, Los Angeles/Charles R. Drew University Medical Education Program on medical students’ intentions to practice in underserved areas. Acad Med. 2005;80:803–808.
39 Rabinowitz HK, Diamond JJ, Markham FW, Paynter NP. Critical factors for designing programs to increase the supply and retention of rural primary care physicians. JAMA. 2001;286:1041–1048.
40 Rabinowitz HK, Diamond JJ, Markham FW, Rabinowitz C. Long-term retention of graduates from a program to increase the supply of rural family physicians. Acad Med. 2005;80:728–732.
41 Sempowski IP. Effectiveness of financial incentives in exchange for rural and underserviced area return-of-service commitments: systematic review of the literature. Can J Rural Med. 2004;9:82–88.
42 Schwarz MR. The WAMI Program: 25 years later. Med Teach. 2004;26:211–214.
43 Pathman DE, Konrad TR, King TS, Taylor DH Jr, Koch GG. Outcomes of states’ scholarship, loan repayment, and related programs for physicians. Med Care. 2004;42:560–568.
44 Grumbach K, Chen E. Effectiveness of University of California postbaccalaureate premedical programs in increasing medicalschool matriculation for minority and disadvantaged students. JAMA. 2006;296:1079–1085.
45 Lang F, Ferguson KP, Bennard B, Zahorik P, Sliger C. The Appalachian Preceptorship: over two decades of an integrated clinical-classroom experience of rural medicine and Appalachian culture. Acad Med. 2005;80:717–723.
46 Smucny J, Beatty P, Grant W, Dennison T, Wolff LT. An evaluation of the Rural Medical Education Program of the State University of New York Upstate Medical University, 1990–2003. Acad Med. 2005;80:733–738.
47 Pacheco M, Weiss D, Vaillant K, et al. The impact on rural New Mexico of a family medicine residency. Acad Med. 2005;80:739–744.
48 Whitcomb ME. Who’s going to take care of the folks? Acad Med. 2005;80:789–790.
49 Nickens HW, Smedley BD, Institute of Medicine (U.S.). The Right Thing to Do, the Smart Thing to Do: Enhancing Diversity in the Health Professions: Summary of the Symposium on Diversity in Health Professions in Honor of Herbert W. Nickens, MD. Washington, DC: National Academy Press, Institute of Medicine; 2001.
50 Greenhalgh T, Seyan K, Boynton P. “Not a university type”: focus group study of social class, ethnic, and sex differences in school pupils’ perceptions about medical school. BMJ. 2004;328:1541.
51 Kondo DG, Judd VE. Demographic characteristics of US medical school admission committees. JAMA. 2000;284:1111–1113.
52 Elofsson S, Unden AL, Krakau I. Patient charges—a hindrance to financially and psychosocially disadvantage groups seeking care. Soc Sci Med. 1998;46:1375–1380.