AMA Journal of Ethics^®

Lloyd Shapley and Alvin Roth were recently awarded the Nobel Prize in economics for their theoretical and applied work in matching theory. This concept was adapted and implemented by the National Residency Matching Program (NRMP) and has been successfully matching applicants with programs by prioritizing the applicant’s choices rather than relying exclusively on programs’ input. Pre-matches, traditionally used by programs to provide an incentive for desirable applicants for early matching, remain unpredictable and unreliable and lack standardization. They are seen by the NRMP to compromise the system by creating “inequities in how residency programs recruit U.S. allopathic senior students and other applicants and increase the risk of undue persuasion when residency programs offer positions outside the Match” [1].

The NRMP is now implementing an “all-in” policy, which will require every residency program to fill every first-year position either exclusively through the NRMP match or outside of it. Programs that continue to offer pre-matches will do so for all residency positions, outside the match. The impact of this new change on both programs and applicants is not yet known. Arguably those most affected by the policy include foreign medical graduates (FMGs), seniors in United States osteopathic schools, and the programs that rely on pre-matches to fill their residency positions. This paper will discuss some potential effects of all-in and the associated ethical dilemmas.

Upon acceptance of a pre-match, FMGs begin seeking their J-1 or H1-B visas. The early appointment allowed supplementary time to obtain visas prior to the internship start date. However, even with the pre-match, some FMGs still started late. With all-in matching, it is likely that the number of applicants requiring a delayed start date will increase because they must wait until they match to apply for visas. The impact of these delays is expected to be minimal in comparison to a resident’s entire training experience. However, residency programs are already confronted with challenges to adequate resident training and all-in could make this more difficult. This raises several questions. First, is it okay to accept a matching system that will delay training for some residents? Second, if this is acceptable, where is the threshold between an acceptable number of affected residents and an unacceptable number? Lastly, is it okay to accept a system if it disproportionately affects a small number of residency programs?

The policies and rules set forth by the NRMP are justified by the fact that participation in the NRMP match is voluntary, and the program is designed to help protect applicants and residency programs. The decision by Saint Barnabas and New York Methodist to send applicants notice that they will no longer be participating in the NRMP match highlights the voluntary aspect of the program [2, 3]. It is unlikely that many programs, however, would be able to fulfill their own recruitment needs outside of the match. For those programs, it is not clear that the match is truly voluntary. Consequently, it could be argued that it is not truly voluntary for those who want to apply to those programs. One concern in this system is that residency programs and applicants who are pressured to participate in the NRMP match may also be pressured to violate rules in the future.

Survey data show that 37 percent of surveyed program directors with more than 10 FMGs residents oppose all-in, compared to only 25 percent of program directors with fewer than 10 FMGs [4]. The same survey noted that program directors were most concerned about the potential adverse effects of the policy on smaller, less competitive, nonuniversity-based programs. For smaller programs with greater dependency on FMGs and graduates of osteopathic schools, securing applicant commitments outside the match is often less costly than securing positions within it. Due to greater uncertainty about filling residency positions in the match, programs will either need to interview more applicants or try to obtain early commitments from those they select. The latter option is a match rule violation. That leaves smaller, less competitive programs, which often focus on primary care training, at a distinct disadvantage—;at a time when we need more high-quality primary care physicians.

Concerns regarding the ability of small, community- and primary care-based residencies to fill positions may be further augmented by the impact of all-in on osteopathic student recruitment. While some osteopathic students apply to both the American Osteopathic Association match and the NRMP match, many have historically applied to AOA programs while interviewing out of the match at NRMP-participating programs. It is unclear if students in osteopathic programs will choose to participate in both the AOA and the NRMP matches. If they do not participate in the NRMP match, once again, smaller programs will be the ones that face the most significant challenges.

While the all-in policy strives to make the match program more equal for all participants, it is not clear that it makes the recruitment process more just for all programs involved. The effort to create widespread equity has the potential to lead to stronger residency programs and more satisfied residency applicants, but it does raise ethical concerns, and the policy will likely have unforeseen consequences. It will be important to monitor the ability of the programs most affected by this policy to manage the change.