AMA Journal of Ethics®

Illuminating the art of medicine

Journal of Ethics Header

AMA Journal of Ethics®

Illuminating the art of medicine

Virtual Mentor. May 2002, Volume 4, Number 5.

Medical Professionalism

  • Print

Justice in Residency Placement

A medical school professor argues against the idea of a random lottery for residency placement, as proposed in a recent journal article.

Timothy F. Murphy, PhD

Residency training is an essential component of medical education and is required in most jurisdictions for licensure as an independent medical practitioner. In the United States, the match system assigns approximately 23,000 applicants to residency training programs in the areas of pediatrics, obstetrics and gynecology, internal medicine, and the rest. The system has been in place since 1952 and is overseen by the National Residency Match Program, a non-profit organization. Rank order lists are at the heart of the match. An applicant picks a number of residency programs and ranks them according to preference. The residency program prepares a similar list, ranking the candidates it most wants in its program. A computer program compares the rankings and makes assignments according to a certain algorithm. These assignments are then announced to all parties on specific days. This system is effective in a number of ways. First of all, it standardizes the timetable for decisions, and applicants are in no position to tie up offers while waiting to hear from another institution. Institutions are not held captive either in making assignments while waiting to hear from particular parties.

A bioethicist at Mercer University, D. Micah Hester, has recently argued that the match system is incompatible with the core values of medicine [1]. Hester's chief criticism is that current match program embodies a competitiveness that corrupts core values in medicine. According to Hester the competition involved in the match encourages values that are antithetical to the medical profession. He says that "so long as competitive practices run rampant in institutionalized activities such as residency matching, medicine simply will never fully meet the concerns of the people who need its help and a society that needs its comfort" [2]. In short, medicine is hyper-competitive, the match is part of this syndrome, and we all suffer as a result.

Hester proposes an alternative to the match as it currently exists: random assignment. Under his proposal, all candidates would stand an equal chance of being selected for each residency opening in a designated discipline. For example, each candidate interested in a pediatrics residency would be assigned at random to one of the available slots in pediatrics residencies around the country. There would be no rank order lists and no communication between candidates and institutions. This approach would—as Hester does not fail to note—eliminate personal choice altogether. In addition to curbing the competition involved in jockeying for candidates and positions, Hester says this approach would free up new resources and energies: "Eliminating the competitive match system would provide residency programs and candidates with the resources to work on other more pressing issues. More time, energy, and money could go to support such concerns and activities as better salaries and hours for residents, outreach programs, deeper professionalism, and ethics and humanities education—concerns and activities that go to the heart of moral medical care" [3].

This is a drastic proposal and, I think, unwarranted both in terms of damaging effects on residency programs and the undermining of personal choice.

Medical residencies are not equal in terms of what they prepare their residents to do and how well they achieve their goals. There is a social division of labor in terms of what residencies are training their physicians to do, and they are not interchangeable replicas of one another. Some residencies are much more likely than others to encourage their trainees to engage in clinical research, to assume academic posts, and to go on to leadership roles in the profession. Others are much more likely to channel their trainees into certain kinds of practice, for example, working in institutions that provide large amounts of charitable care. It does matter which students are tracked into residencies because these programs train particular people whose knowledge and skills are fundamental to the design of the health care system, produce trainees who are expert in the management of certain kinds of patients, and develop the skills of particular people who will fill specific roles in the delivery of health care. It is reasonable to believe that random assignment of residents would undercut this division of labor and compromise the ability of residencies to achieve their important social goals.

When it comes to the fate of residents themselves, there are important reasons to avoid complete randomness in residency assignment. For example, many residents are married and have children. Some residents have primary responsibility as caregivers for aging or sickly parents. It would be a fundamental hardship to say that these residents should have no say whatsoever in where they train. A decision to pursue a particular residency is not only about where one continues medical education; it also reflects choices about one's familial and financial interests. For some residents it would be a hardship in the extreme to move their families from Florida to Alaska or to relocate them to rural programs far from their families. In another instance, it would be an undue economic hardship to ask some residents to shoulder the unwanted costs of residency in Manhattan when they actually prefer less costly living in a smaller city in the South. These kinds of complications could be multiplied without much difficulty. That some residencies last 6 and 7 years makes it all the more important to recognize that random assignment in residency could create and magnify all kinds of problems for trainees.

Hester does acknowledge that some residents would be resentful about assignments given to them by chance, but he thinks that this resentment would be offset by the value of being exposed to trainees from all across the country. He says that residency programs with their supply of trainees chosen at random "would benefit from having fully supplied medical staffs and residents from an array of educational backgrounds, and the diverse residents could learn from each other while providing care for otherwise underserved patients. On the flip side of the equation, if it is in fact the case that some residency programs are better than others, these so-called "top" programs would have the opportunity to work with a variety of residents from different schools and backgrounds, residents who might not otherwise have had the opportunity to learn from the "best." In response, it must be said that it is not clear how random assignment would necessarily improve care for underserved patients. A lottery might help distribute talent more broadly across residencies, but by itself this would not mean that underserved patients would necessarily receive better care. If "the best" medical graduates do not like their placement, lingering resentment could work to sabotage quality of patient care they deliver as residents.

Hester goes on to compare his proposal for random assignment to the kind of drafting that occurs in professional sports. Many professional athletes are assigned to teams without their having a say in the matter—and their families and living preferences are not taken into consideration. Hester wonders why this same attitude—it is enough for small town heroes to play in the National Football League no matter where they end up—she should not also prevail in medicine. In other words, the rewards of being in medicine should override any specific concerns about where one wants to live and train. It is not clear, however, that random assignment would promote selfless values in physicians any more than it does in professional football recruits. Random assignment would undoubtedly disrupt important interests for more than a few residents—which disruption could easily undermine selfless attitudes. Moreover, it is certainly not clear that random assignment would make trainees better diagnosticians, better therapists, or even help them exhibit more humane behaviors toward patients. Simmering resentment could corrode humane values and foster poor clinical habits just as badly—if not more so—than the competitive aspects of the match. Even professional football players—the best of them anyway—try to control where they play, especially those concerned with the rewards of league victories, championship rings, and commercial endorsements.

Over and above the effect random assignment would have on residents, a lottery system could also be expected to undercut motive and effort among medical students. Certain medical school graduates are better than others with regard to their capacities in diagnostics, in therapeutic judgments, and interpersonal skills. It is to be wondered what incentive there would be for medical students to strive toward superior achievement if residency assignment turned a blind eye toward all accomplishment and occurred only by chance. It might well be true that some students would go the extra mile in medical school, those who do so for personal satisfaction or some other intrinsic reward. However, it is hard to believe that performance would not suffer if extra efforts could not, could never help in securing a preferred residency. And, to continue the theme of performance for a moment, it is hard to see that one would be doing the poorest performing students any favor by placing them in the most demanding residencies in the nation. By extension, it would be doing little favor to burden highly functional residencies with applicants who have done little more than stumble and limp their way to a diploma. After all, it is not only talent that would be randomly distribute, a lottery would distribute the opposite of talent—whatever one wants to call that—as well.

Hester does acknowledge that random assignment would mean the loss of personal choice, but he believes this loss is acceptable because of the way in which residency competition would be undercut. It is simply untrue, of course, that once some resources are freed up that these would flow automatically to more noble causes. For example, physicians involved in overseeing residency recruitment—interviews, answering questions, preparing promotional materials, ranking candidates—might just as easily turn their attention to clinical revenue as to improving humanistic education of residents.

Lastly, there is also an important difficulty in Hester's claim that the match operates in a way that is inconsistent with the core values of medicine. The preservation of choice is one core value in medicine, one that undergirds patient-physician encounters. The American Medical Association Principles of Medical Ethics asserts that "A physician shall, in the provision of appropriate patient care, except in emergencies, be free to choose whom to serve, with whom to associate, and the environment in which to provide medical care" [4]. In other words, choice is a core value of medicine because it is important to both patient and physician alike to enter into mutually satisfactory relationships. To put it another way, except for emergency or court-ordered treatment, health care relationships should not be random or involuntary. It is hard to understand why this principle—so important to health care relationships—should not also extend to educational relationships.

In one study of residency applicants, only 4 percent of the respondents believed that the match should be completely overhauled [5]. Whatever the problems of the match system are, it is not clear that residency assignments made at random will solve them without also causing broad, systemic problems on a large scale. It is one thing to dream of medicine that is shorn of the worst effects of competition and that is fully committed to ethics, professionalism, and humanities. There is no reason, though, to think that a lottery system would help achieve these goals in any meaningful way. There is no obvious reason to think that revolution rather than reform should be the appropriate response to problems in the match.


  1. Hester DM. Rethinking the residency matching process and questioning the value of competition in medicine. Acad Med. 2001;76:345-347.
  2. Hester, 345.
  3. Hester, 346.
  4. American Medical Association. Principles of medical ethics. Accessed April 10, 2002.
  5. Anderson KD, Jacobs DM, Blue AV. Is match ethics an oxymoron? Am J Surg. 1999;177:237-239.

Timothy Murphy, PhD, is professor of philosophy in the biomedical sciences in the Department of Medical Education at the University of Illinois College of Medicine in Chicago.

The viewpoints expressed on this site are those of the authors and do not necessarily reflect the views and policies of the AMA.