Edwards, Janine C. PhD; Elam, Carol L. EdD; Wagoner, Norma E. PhD
Medical school admission is a perennial topic of interest, not only to the administrators and faculty of the 125 allopathic medical schools in the United States but also to the thousands of applicants, their families, and the undergraduate faculty who have taught those applicants and are their advocates. Any number of issues in admission surface from time to time that need attention. Some are of a long-standing nature and serve as grist for the admission committee mills. However, others emerge and gain national attention and require considerably more effort to find some measure of resolution.
In the latter instance, a recent focus on the greater use of qualitative variables for selecting candidates—such as compassion, altruism, respect, and integrity—serves the important goal of emphasizing character qualities that ensure greater professionalism among future physicians. The second and more contentious issue has been the system used to admit white and minority applicants, decried by some as a twotier admission system. Nearly everyone recognizes the educational and societal benefits of a racially and ethnically diverse student body, and in that context, minority student admission has become a compelling question that has reached the highest courts in several states. Using race as a factor in admission, a practice that has been in place since the famous Bakke case1, has come under attack in recent years. Many believe it is a matter of time before this issue will resurface in the U.S. Supreme Court to be debated again.
In this article, we first explore the factors that have led to greater emphasis on qualitative variables in selecting medical students and assessing the conduct of physicians. We then outline an admission model that can be useful in making this or other shifts in emphasis in a school's admission process, and that also can help deal with the difficult issue of diversity in admissions. Last, we give an extended hypothetical example of the use of the model to implement a specific change in emphasis from primary care to research at one medical school.
GREATER USE OF QUALITATIVE VARIABLES
What factors have led the medical education community to call for greater use of qualitative variables in selecting and assessing the conduct of physicians? In the recent book Time to Heal,2 Ludmerer spoke about the need for academic health centers to restore the social contract. He stated that in the 1990s a second revolutionary period in American medical education began. Managed care, with its emphasis on seeing patients quickly, “was making it increasingly difficult for doctors to practice in concordance with many traditional professional teachings and values.”2, p. 387 Further, the inability to use clinical revenues to cross-subsidize education, research, and charity care was destroying the learning environment. As institutions have become more financially burdened, dilemmas faced by American medicine in reestablishing long-held precepts about the importance of the patient, and of educating future physicians to uphold important and long-standing professional values, have begged for solutions. Dr. Ludmerer noted that “over time society tends to reward groups that aspire to improving the human condition,” and that “American society in the twenty-first century is likely to reward the medical profession if it succeeds at placing the interests of patients and the public first.”2, p. 397
Interest in humanistic factors and other attributes of the profession have not always been given high priority in the selection of medical school candidates. This is not out of lack of desire on the part of committees. Difficulties in agreeing upon important variables and how to obtain valid and reliable measures by which to support the use of these variables have proven much more challenging. An important step was taken in 1996, when the Association of American Medical Colleges (AAMC) established the Medical School Objectives Project (MSOP).3 The goal of the first phase of this project was to develop a consensus within the medical education community on the attributes that medical students should possess at the time of graduation. Attributes that included altruism, integrity, respect, empathy, and compassion were identified. In addition, the Liaison Committee on Medical Education (LCME), the accrediting body for U.S. medical schools, is now reviewing the programmatic efforts by medical schools to teach professionalism and to demonstrate, through appropriate assessments, the efficacy of those efforts.
Another sign of the new emphasis on humanistic factors occurred when the National Board of Medical Examiners (NBME) announced its intention to require that examinees seeking licensure to practice in this country pass an examination that assesses professionalism, communication, and interpersonal skills. In a similar vein, the Accreditation Council on Graduate Medical Education (ACGME), in conjunction with the American Board of Medical Specialties (ABMS), recently finalized the definitions describing their specialty-specific professional competencies.4 Among the six defined competencies are two relating to important qualitative aspects that have relevance for the admission process. The first, communication and interpersonal skills, states that residents must demonstrate interpersonal and communication skills that enable them to establish and maintain professional relationships with patients, families, and other members of the health care team. The second is professionalism, where residents must demonstrate behaviors that reflect a commitment to ethical practice, an understanding and sensitivity to diversity, and a responsible attitude toward patients and the profession.
The policy of the American Board of Internal Medicine (ABIM)5 states that internal medicine physicians applying for certification must be evaluated for demonstrated humanistic qualities that include integrity, respect, and compassion in their relationships with patients and their families in order to become board certified. The ABIM has recognized the importance of high moral and ethical standards, such as outstanding professional behavior, and has sought to lead by example in its certification requirements. Other specialties are now beginning to follow suit through revision of their own processes and requirements.
Based on the changes in direction of the groups mentioned above, there is increasing pressure for medical schools to ensure that entering students bring the requisite attributes to succeed in the additional humanistic areas that will be required of them for licensure and certification. It will not be long before medical schools themselves are required by some organization or accrediting entity to demonstrate how they are making judgments about whether candidates entering their medical schools possess these attributes. Moving toward the inclusion of greater use of qualitative variables will require a paradigm shift in how admission committees go about their business.
There are recognized barriers to making this paradigm shift away from the highly focused quantitative assessment that now drives the selection process. In June 1998, the Arnold P. Gold Foundation held an invitational conference entitled “Challenging the Barriers to Sustaining Humanism in Medicine: Selecting Humanistic Candidates for Medical School,”6 in which barriers to selecting such candidates were identified and a range of possible solutions promulgated. Publication of rankings of U.S. medical schools based on quantitative variables as the primary source data for the rankings is an obstacle. Detailed information of that type about medical schools can be found in the Association of American Medical Colleges' medical school admission requirement handbook (MSAR).7 For example, this publication enables schools to provide relevant data to applicants, including the mean grade-point average (GPA) of each school's applicants and their Medical College Admission Test (MCAT) scores. Rarely does the school's information include documentation about the important personal attributes and character qualities that are relevant to their selection process. Although it is true that producing details of cognitive-based scores for candidates helps the prospective candidate focus on his or her chances for obtaining entrance at a particular school, such publications also give the message that these variables are what count most in the selection process.
One important reason that expanding schools' admission focus to include humanistic qualities is difficult is that the admission process is very expensive in terms of time, money, and effort for all involved. To ask overburdened applicants to do more, or in any way lengthen the process through more extensive screening or by increasing the number of interviews granted, is a difficult concept to sell to faculty. Faculty members are not rewarded through any tangible institutional means for the time they invest in the admission process. Another reason that introducing a humanistic focus is not easy is that considerable faculty time and effort are sometimes spent on issues of diversity; the solutions reached are sometimes not ideal. For example, when attempting to meet an institutional goal of having a diverse class, coming to grips with the fear of lawsuits often causes committees to “play it safe” by resorting to numbers. In addition, helping interviewers understand issues of diversity, including one's own biases, is a significant hurdle. Yet, as is explained in the next section, the introduction of a humanistic focus can help admission committees achieve greater diversity in the school's student body.
Solutions to some of the challenging problems suggested at the “barriers conference” will require institutional buy-in as to the importance in making this paradigm shift possible. The medical school representatives at the conference felt strongly that they needed to enlist the help of many groups involved in this process. One such group that was identified was the health-professions advisors. Enlisting their assistance in identifying important humanistic traits in candidates could serve as a very important first step. The medical schools would need to hold workshops for advisors and establish collaborative arrangements to determine what important qualitative information they believe would be possible to gather and to include in letters of evaluation.
At the local level, each school would need to identify and define criteria important for selecting humanistic candidates and promulgate this information throughout the institution. The admission committee would need to develop guidelines for assessing criteria such as altruism, service orientation, respect for others, empathy, judgment, and honesty. Interviewers in the medical school would need to be trained, and an institutional measurement instrument would need to be constructed to ascertain how well the admission process succeeded in selecting candidates with the desired attributes. Each institution would also need to strategize about how to market humanism as a value within and outside the institution, through publicity, awards, and the admission materials themselves. Alumni support for the importance of humanism would need to become an institutional goal. Ultimately, institutions would need to connect their efforts to the requirements of the LCME, the NBME examination, the ACGME, and the certification and recertification processes of the various specialities.
A USEFUL ADMISSION MODEL
Gaining momentum for this paradigm shift remains one of the great challenges faced by all medical schools. Convincing faculty members that qualitative variables should have a more equal weight in selecting candidates often leads to accusations of making the process too soft, too diffuse, and too difficult to defend. However, we believe the time is right to take the steps toward change. To make a paradigm shift in the admission process, we must think about all the elements of that process and their interrelationships. An admission model can foster greater understanding of the admission process and can serve as a heuristic guide.8 Use of a graphic or pictorial model, such as that in Figure 1, is helpful in imaging the components of the model and suggesting how those components, or dimensions, relate to each other.9 Referring to the dimensions of the admission model can assist in thinking through admission issues, thus making the whole planning process more systematic, since the model synthesizes those factors essential for admission.
The model for admission embodies dimensions of admission that are generally recognized and accepted: (1) the applicant pool; (2) criteria for selection; (3) the admission committee; (4) selection processes and policies; and (5) outcomes—each of these is briefly discussed in the following paragraphs. Criteria for admission serve as a major focal point among the dimensions. Expanding such criteria has already helped establish the new humanistic paradigm in the admission process in the difficult area of ensuring diversity among medical students. In many instances, in states where anti—affirmative-action issues have been particularly prominent, the medical schools have already taken the lead to expand their admission criteria to include greater qualitative aspects to ensure that they can continue to attract diversified classes.10 We feel strongly that the contribution of a model such as the one proposed here is not to create new admission criteria or other content, such as a research study would create, but to synthesize elements to provide a framework for thinking.
The Applicant Pool
This dimension of the model can be thought of by group and also by individuals. That is, the applicant pool includes the number of applicants each year as well as all the variables that define each individual applicant. Everyone involved in the admission process scrambles each year to keep abreast of the latest statistics of this sort. The Association of American Medical Colleges regularly publishes statistics related to number of applicants, MCAT scores, and demographic trends of applicants throughout the cycle of admission. Each medical school compiles and disseminates its set of descriptive statistics related to its applicant pool. At annual national, state, and regional medical education meetings these statistics are presented, compared, and discussed. Studies of this dimension can help determine the extent of the applicant pool fluctuations over time, which could be useful in interpreting changes in the applicant pool and perhaps in predicting future applicant pool size.11
Variability among individual applicants is another aspect of this dimension. The character traits of individual applicants are of special concern in our society because medical educators recognize that fostering attributes of the profession must become central to medical education. Therefore, there are important relationships between the applicant dimension and the dimension of criteria for admission, discussed next.
Criteria for Admission
Academic, or cognitive, criteria are the mainstay of most medical schools' selection processes. Of the cognitive criteria, undergraduate science GPA and MCAT scores hold sway as the most important. These quantitative criteria have yielded some evidence of validity and reliability,12 and they are efficient to process. In general, it is fair to say that underrepresented minority applicants as a group do not have quantitative qualifications equal to those of majority applicants; yet there is evidence that most underrepresented minority students do succeed in medical school and in obtaining residency positions.13 The differential achievement of underrepresented minority applicants on the MCAT has been documented and studied.14 However, given the current well-organized and well-financed attack on affirmative action programs in higher education and professional education, medical school administrators must document the objectivity of their selection processes. As a result, they are reluctant to use subjective information related to personal characteristics, since these show little evidence of validity and reliability. This difficulty does not decrease the need for a greater focus on qualitative variables, for the reasons discussed earlier in this article.
The Admission Committee
The composition of the admission committee can account, in large part, for admission policies in each medical school. In many schools, the admission committee is directly responsible for defining the criteria for admission in concert with the institutional mission. Recommending faculty members, students, community members, and others to serve on the committee is an important task each year for the chief admission officer. Negotiating the appointments for the committee with the dean and department chairs requires experience and skill to achieve a committee that operates effectively. The methods by which the committee assimilates new committee members and inculcates them with the committee's values is critical. Studies of decision-making styles of the committee and voting patterns among members can shed light on these relationships.15 Determining ways to recognize and reward faculty members for service on the committee is also a challenge.
Each medical school designs its own selection process, although many aspect of selection processes are similar among the various schools. Many medical schools continue time-honored traditions in their selection processes year after year. Screening initial applications, deciding which applicants should receive secondary applications, reading letters of recommendation and complete applications, weighting various criteria, deciding which applicants to interview, decision making, extending offers of acceptance, and attending to the national “traffic rules” regarding due dates are all aspects of the selection process. These aspects interrelate with one another and with the other dimensions of the model. The introduction of AMCAS 2002, the electronic application system, may change screening procedures and have an impact on decisions.16
The interplay between the criteria and selection processes is particularly important. For example, if the criteria include attributes of the profession such as compassion and respect for persons, do the selection processes include appropriate methods of evaluating those attributes, such as interviewing and evaluation of compassion and respect in the interview? If there is a tradition to value “legacy” in the medical school, is that value explicitly acknowledged, or are decisions to admit applicants from alumni families made “behind the scenes”? Is value added by having committee members deliberate on the overall competitiveness of each candidate for admission, or should weighting formulas be used to determine the admission decision? If both academic criteria and attributes are included in decision formulas, do the weighting formulas give equal weight, or do these formulas favor one type of criterion over the other type? These examples illustrate the complex interplay between criteria and selection processes.
What are the outcomes of admission to medical school? Which admission criteria are the best predictors of academic performance in medical school? Which admission criteria are most helpful in predicting the academic success of underrepresented minority students in medical school? Which admission criteria are the best predictors of faculty or peer ratings of professionalism and other qualitative behaviors? Short-term outcomes are the matriculating student's progress through the medical school curriculum marked by course grades and licensure exam scores, selection of a specialty, and progress into graduate medical education. The ultimate outcome may be the skillfulness and humaneness of care for individual patients. In addition, population care is very much a current issue and may be perceived as a more important outcome in the future.
APPLYING THE MODEL
Consider this example: Assume that your medical school is an institution established by a state legislature to produce more primary care doctors than the traditional specialty-oriented medical schools. Your admission criteria and process have been tailored to fit that mission of your medical school for the past 20 years. Applicants have been recruited and selected on the basis of their interest in practicing primary care medicine.
Now your dean wants to step away from that primary care mission and build the research enterprise in the medical school. He or she has conducted a strategic planning process in which research emerged as a new priority for the school. The dean has advised you, the director of admission, that the new institutional mission is to develop research and that this mission must become a priority in admitting applicants.
How do you proceed? Using the model of admission discussed here, you can begin thinking systematically about each dimension in order to change your admission process's orientation. The applicant pool is an obvious starting point. In your current applicant pool, about how many candidates are interested in research? Can you expand that number, through both in-state and out-of-state recruitment? Do you need to have the policy about the number of in-state students in the class changed? What recruiting programs are in place or need to be developed to increase potential students' interest in research? What marketing materials need to be prepared for your applicants about research opportunities? What information does your Web site contain about research at your institution? How can you revise your Web site? How would you go about enlisting the aid of the premed advisors to interest more applicants in research?
The criteria for admission are another dimension of the admission model to consider in relation to research. How will the school criteria have to be changed to admit more students who want to combine research with clinical patient care? What courses should undergraduates take to prepare themselves to participate in research in a medical center? Would you anticipate that the minimum acceptable science GPA and MCAT scores would be different for students interested in research compared with the scores of those interested in primary care? What personal qualities are important for future biomedical researchers? Who can provide information to help you define those necessary personal qualities? How would you gather information about those qualities during the application process? Should you ask for different information from the premedical advisors in the letters of recommendation that they forward to the medical school? Obviously, there will have to be a criterion about research experience. What are the nature and the extent of prior research experiences that one should expect from candidates for admission? Is an MD—PhD program necessary? How are the selection criteria for that program interrelated with the selection criteria for the MD program? It seems likely that all of these factors will require policy changes.
Policy changes lead to another dimension of the model—the admission committee. How should this new research priority be presented to the committee? Is the dean mandating this change in priority, or does he or she want the committee to develop a consensus to make this change? How many of the committee members will be receptive to this change? If the majority are not receptive, how can they be persuaded to change? Will a new committee have to be constituted? How long will it take for the committee to change the policies? How will the new policies be implemented? How can the committee be educated about the opportunities for research within the medical school so that they can discuss this issue intelligently with the applicants?
What changes need to be made in the selection process? On the secondary application, should new questions be asked about research interests? How can interviewers be trained to explore an applicant's qualifications and level of interest in research? Will the committee decision process have to be changed? Discussion of that question may depend upon how many of the committee members are receptive to the new research priority. If the majority will accept the new priority, then perhaps the decision process can remain the same. But if a critical mass of the committee members are skeptical or negative, the decision process may have to be changed so that they will not block implementation of the research priority.
The final dimension to be considered comprises the outcomes. Would the school expect all medical students to perform research? How would the school evaluate their performances? Would the curriculum have to be changed to allow more flexibility in the third and fourth years for students to participate in research? Will licensure exam scores increase or decrease? Would it be expected that the school will graduate more clinical researchers and fewer practitioners? What admission criteria will best predict the outcome of producing clinical investigators? How will the advising program have to change to support that new outcome? Will a greater percentage of the students enter academic medicine?
This is a scenario that, in a variety of forms, is being played out in a number of medical schools established 20 years ago. During the past ten years, many medical school admission committees have had to grapple with the opposite scenario—that is, they have had to change their priorities to produce more primary care doctors. Thinking about admission in the dimensions of the proposed model can make planning more systematic. The model synthesizes those factors essential for admission.
In this article, we have tried to show that formulating a model for medical school admission can be helpful, especially given the societal forces for change that can affect admission policies. Emphasis is being placed on teaching and licensure testing of the attributes of the profession—by greater use of qualitative variables—during the four-year curriculum. It is logical, therefore, to include those attributes in the admission criteria and selection processes. Furthermore, as indicated earlier, the selection of underrepresented minority applicants who can succeed in medical school may be facilitated by the use of criteria concerning attributes of the profession in addition to academic criteria. A model of medical school admission, such as the one we have proposed, can represent the complexities of admission and help decision makers organize and interpret facts, generalize without attempting to standardize, and stimulate hypotheses for further study.
A major benefit of building a model is that it can stimulate and guide future research. A thorough review of research studies about admission to medical school during the past decade is needed. The last review of literature about admission was published in a thematic issue of Academic Medicine in 1990.17 The model can provide a framework for reviewing existing studies and also point out which dimensions require further study. Through this and other uses of the model, the forces that challenge us are more likely to lead to creative and productive thinking for the future.
1. Regents of the University of California v. Bakke, 438 U.S. 265, 1978.
2. Ludmerer KM. Time to Heal: American Medical Education from the Turn of the Century to the Era of Managed Care. New York: Oxford University Press, 1999.
3. The Medical School Objectives Writing Group. Learning objectives for medical student education—guidelines for medical schools: Report I of the Medical School Objectives Project. Acad Med. 1999;74:13–8.
4. Leach DC. Evaluation of competency: an ACGME perspective. Am J Phys Med Rehabil. 2000; 79:487–9.
5. Blank LL. Humanistic Qualities: The Art of Medicine and a Policy of ABIM. Philadelphia, PA: American Board of Internal Medicine, 1999.
6. Arnold P. Gold Foundation. Challenging the Barriers to Sustaining Humanism in Medicine: Focus on Humanism in the Medical School Selection Process. Alexandria, VA, June 1998.
7. Coleman CL (ed). Medical School Admission Requirements United States and Canada 2001–2002. Washington, DC: Association of American Medical Colleges, 2000.
8. Tzeng OC, Jackson JW. Common methodological framework for theory construction and evaluation in the social and behavioral sciences. Genet Soc Gen Psychol Monogr. 1991;117:49–76.
9. Green M. Theories of Human Development: A Comparative Approach. Englewood Cliffs, NJ: Prentice—Hall, 1989.
10. Edwards JC, Maldonado FG, Englegau GR. Beyond affirmative action: one school's experiences with a race-neutral admission process. Acad Med. 2000;75:806–15.
11. Kassebaum DG, Szenas PL. The decline and rise of the medical school applicant pool. Acad Med. 1995;70:334–40.
12. Mitchell KJ. Traditional predictors of performance in medical school. Acad Med. 1990;65:149–58.
13. Davidson RC, Lewis EL. Affirmative action and other special consideration admissions at the University of California, Davis, School of Medicine. JAMA. 1997;278:1153–8.
14. Koenig JA. Evaluating the predictive validity of MCAT scores across diverse applicant groups. Acad Med. 1998;73:1095–106.
15. Elam CL, Johnson MM. An analysis of admission committee voting patterns. Acad Med. 1997;72(10 suppl):S72–S75.
16. Association of American Medical Colleges, AMCAS Admissions Workstation Users Manual. Washington, DC: AAMC, 2000.
17. Edwards JC (ed). Medical school admission. Theme issue of Academic Medicine. 1990;65:133–90.