In 1952, the National Resident Matching Program (NRMP or “match”) was established to improve uniformity in residency recruitment.1 Policies require programs that participate in the match to accept current, allopathic, US, fourth-year medical students only through the match. However, independent applicants (ie, graduates of US medical schools, students and graduates of US osteopathic medical schools, Canadian medical schools, Fifth Pathway programs, or international medical schools2) may be accepted outside the match, which creates a system that some feel is inequitable.
Before the NRMP, program directors offered positions to applicants increasingly earlier, which reduced the time applicants could consider options: Applicants were often forced to choose immediately between a guaranteed job versus a potentially “better” later option. In addition, some applicants accepted multiple early offers while waiting for the best offer, which thereby eliminated choices for other applicants and left some programs ultimately with open positions. By using the match, program directors and applicants accepted some limitations in exchange for overall greater security and fairness.
Currently, outside match offers (“prematch”) are common in many specialties. A 2010 review reported 23% (1490 positions) in internal medicine (IM), 23% (385 positions) in preliminary surgery, 21% (704 positions) in family medicine, 21% (277 positions) in transitional years, 10% (273 positions) in pediatrics, and 7% (81 positions) in obstetrics-gynecology.3
NRMP Policy Changes
The NRMP has periodically considered plans to implement a policy whereby either individual residency programs or entire institutions would be required to recruit all or none of their residency positions through the match. Concern has been raised regarding such “all-in” policies, particularly because of recruiting practice variation, perceived fairness of the recruitment process, enforceability, and concerns that international medical graduates (IMGs) participating in the match would not be able to obtain visas in time to begin their first year.4
In May 2011, the NRMP unanimously voted to implement a program-based all-in rule (ie, individual residency programs, but not entire institutions, will be required to place all positions in the match or withdraw from the match) beginning in 2013.4 The NRMP solicited feedback from the Association of Program Directors in Internal Medicine (APDIM) and other NRMP constituents regarding the all-in rule changes. This article will discuss program director responses to the 2 surveys from the APDIM and the NRMP.
In 2007, and again in 2011, the APDIM Survey Committee surveyed all member programs of APDIM to explore the characteristics of programs and opinions of program directors regarding an all-in match policy. The survey instruments and summary files are available on the APDIM website (http://www.im.org/toolbox/surveys/APDIMSurveyData). Program directors were asked whether they support or oppose the all-in policy, and their reasons for their opinions in free-text responses.
Responses to both surveys were tabulated, with logistic regression models fit to examine associations between program and program director covariates and disagreement with the all-in proposal. This study was approved by the Mayo Clinic Institutional Review Board.
APDIM Survey Results
A total of 236 program directors (61.9%) responded in 2007, with 122 (51.7%) indicating their program filled some categorical positions outside the match. A total of 109 (46.2%) disagreed with the proposed all-in rule. Summaries of covariates and associations with all-in rule disagreement are displayed in table 1. After controlling for all covariates simultaneously through a multiple logistic regression model, only having more than 10% IMGs remained significantly associated with increased odds of disagreement (odds ratio [OR] = 4.76; P = .005).
In 2011, 223 program directors (62.5%) responded, with 129 (57.8%) indicating their program filled positions outside the match. Seventy-nine (35.4%) disagreed with the proposed all-in rule. Summaries of covariates and associations with all-in rule disagreement are displayed in table 2. No significant associations were found.
In 2011, 70 program directors responded to the question of potential consequences of the all-in policy and cited 85 expected consequences (table 3). The most commonly cited concern was that smaller, nonuniversity programs and those in geographically less-desirable areas would suffer in recruitment. Many program directors reported the all-in policy would improve fairness of recruitment for IMGs, but many also reported the cost and effort of recruitment and interviewing would increase for both programs and applicants.
From these 2 surveys, discussions at APDIM meetings, and NRMP surveys targeted at program directors from multiple specialties,4 it is clear that program directors have varying opinions regarding the all-in policy. Although regional and demographic variation still exists in outside match positions, no usual demographic factors are associated with program directors' agreement with such a policy.
Proponents and opponents of all-in agree that, although some applicants would benefit (eg, academically stronger IMGs and osteopathic students), other applicants may have more difficulty (eg, less-competitive US medical graduates and IMGs). There is general agreement that some programs' recruitment might be unsuccessful, and recruitment costs and effort would increase for programs and applicants who previously had outside match options. Program directors also express concerns that IMGs may not be able to secure timely visas.
Supporters of the all-in policy primarily cite enhanced support for fairness and professionalism. Without an all-in policy, applicants may be pressured to make career decisions prematurely, to accept offers before having adequate opportunity to interview at all programs, and be subject to an unregulated hiring process. The US, allopathic, fourth-year medical students may sense unfairness at not having the same outside match option that independent applicants have. There are also potential risks for programs offering positions outside the match: Without regulation, programs have been left with unfilled positions after an outside match candidate withdraws acceptance of an offer.
Factors driving support for outside match offers are related to program director concerns that they may not fill all available spots with highly qualified candidates and the preference for some independent residency applicants for outside match offers. The APDIM 2007 survey indicated that the outside match option appears to be more attractive to programs that rely heavily on IMGs. Smaller, nonuniversity programs and those in less-desirable geographic locations may have concerns about their ability to secure qualified candidates without the outside match option.
From the independent-applicant perspective, they currently have the ability to prematch, and thus, eliminate concerns of not securing a position. However, these outside match offers have significant potential to promote lapses in ethics and professionalism by both applicants and program directors. Recent studies in multiple specialties have confirmed match agreement violations as well as other ethical and professionalism concerns. In one study,5 94% of family medicine program directors felt they needed to be dishonest with applicants to have the best possible match. Ethical transgressions have been witnessed during residency recruitment in dermatology6 and radiology.7 Radiology applicants reported more than 50% of programs were violating the match policies by misleading applicants and by pressuring them to commit early.7
These violations contribute to mutual skepticism and dishonest communication between applicants and program directors8 and undermine the professionalism that the match process seeks to preserve. Outside match offers may pressure applicants to make commitments before adequate consideration and thus facilitate later retraction of these commitments with concomitant erosion of professional integrity.9
In the APDIM surveys, IM program directors were more likely to support the all-in proposal in 2011 than they were in 2007. There are several possible reasons for that. The NRMP's previous all-in proposal (which was queried by the 2007 survey) would have required an entire institution to enter all of its residency positions in the match, whereas the 2011 policy limits the all-in match to individual residency programs. Program directors may recognize that institutions are likely to allow an all-in policy for individual residencies because individual program recruitment strategies may vary widely within a single institution. In addition, the visa process may have now improved sufficiently that program directors have less concern regarding IMGs beginning residency on time. Some program directors may now agree with the policy out of a sense of inevitability. Finally, increasing competitiveness for IM residency positions between 2006 and 2011 may also have swayed program director opinions (in 2011, the ratio of postgraduate year-1 positions per active US senior was 1.41, the lowest since 1999).10
The issue of prematch offers is not unique to IM; 1 in 5 positions (20%) in all primary care specialties are attained via the prematch.3 Outside the Match offers contributed to the demise of the match in the late 1990s for gastroenterology fellowship positions.11 After initial abandonment, the gastroenterology match was reinstituted and has largely been successful.12
The APDIM represents a diverse group of residency programs whose constituent program directors have expressed varying opinions regarding all-in NRMP match policies. The variation likely depends on many factors, including those intrinsic to programs, institutions, and geographic regions. Although many program directors prefer the consistency and fairness of the match, others have concerns that a “one size fits all” policy would benefit some programs while harming others. The NRMP has requested comments for consideration regarding policy exceptions.4 A desire for fairness for both residencies and applicants was expressed in the 2 APDIM surveys. We suggest the APDIM survey committee, other national specialty and subspecialty program director groups, and medical student advocacy organizations continue to track constituents' opinions regarding this important topic.
Michael Adams, MD, FACP, is Program Director of Medicine at Georgetown University; Thomas B. Morrison, MD, is Fellow in Cardiovascular Medicine at Vanderbilt University; Stephanie Call, MD, is Program Director of Medicine at Virginia Commonwealth University; Andrew J. Halvorsen, MS, is Biostatistician and Project and Data Manager in the Internal Medicine Residency Office of Educational Innovations at Mayo Clinic; Jared Moore, MD, is Chief Medical Resident in Internal Medicine at Ohio State University; Maria Lucarelli, MD, is Associate Program Director in Internal Medicine at Ohio State University; Steven Angus, MD, is Program Director of Medicine at the University of Connecticut Health Center; and Furman S. McDonald, MD, MPH, is Program Director of Internal Medicine Residency Office of Educational Innovations at the Mayo Clinic.
Funding: This study was supported in part by the Mayo Clinic Internal Medicine Residency Office of Educational Innovations as part of the ACGME Educational Innovations Project. The Mayo Clinic Survey Research Center provided assistance with the survey design and data collection.
We are grateful for the support of the Association of Program Directors of Internal Medicine and the members of the Survey Committee and to the residency program directors who completed the Association of Program Directors in Internal Medicine (APDIM) survey.
While we are reporting the results of the APDIM Survey Committee, we are not presuming to speak for the organization and our paper does not constitute an official policy statement of APDIM, the APDIM Council, or any other organzation with which any of the authors may be affiliated.