Although internal medicine residency programs participate in the National Resident Matching Program (NRMP) under an All In Policy, most medical subspecialty fellowships have not adopted this policy as part of participation in the NRMP's Medical Specialties Matching Program (MSMP). To participate in the MSMP, adult endocrinology and other fellowships commit to at least 75% of programs participating in, and at least 75% of positions allocated through, the Match, but some programs fill some or all positions out of the Match. Growing stakeholder interest in an All In Policy led the Association of Program Directors in Endocrinology, Diabetes, and Metabolism (APDEM) leadership to consider adopting such a policy. In essence, an All In Policy requires that programs participating in the Match must attempt to fill all offered positions via the Match (unless an exception is granted).1 

In 2015, APDEM surveyed endocrinology program directors and fellows regarding a potential All In Policy. Although participation rates were less than 50%, approximately two-thirds of program director and fellow respondents indicated a preference for an All In Policy (provided as online supplemental material). As a result, APDEM assembled an All In Match Working Group (the authors of this article) to explore the potential advantages and disadvantages of such a policy (table 1). By intention, this group included members known to be skeptical of an All In Policy. We benefitted from a review of several resources from other medical subspecialities,24  but we also considered challenges that had not been fully addressed previously. This article provides a summary of our deliberations in addition to APDEM's formal decision-making process.

table 1

Advantages and Disadvantages of an All In Match Policy

Advantages and Disadvantages of an All In Match Policy
Advantages and Disadvantages of an All In Match Policy

As a working group, we valued candidates' autonomy to pursue any opportunities they found desirable. We acknowledged that, when out-of-Match offers are extended, some applicants might feel unwanted pressure to make commitments before they are ready to do so (eg, prior to a full exploration of programs). This potential for undue pressure is especially relevant when accompanied by a time limit for acceptance; candidates may accept an offer to obviate a failure to Match, even if the position is otherwise less desirable than other options. We noted a power asymmetry in this regard, with programs disproportionately possessing the ability to offer and revoke out-of-Match offers. Anecdotal data (eg, APDEM surveys) confirmed that out-of-Match offers can exert unwanted pressure on candidates, but the scope of this problem was unclear.

In some cases, out-of-Match agreements can be highly desirable for the program and candidate alike. Our primary dilemma concerned candidates with an exclusive preference for a specific program, especially when highly desired by that same program. In such situations, both the candidate and the program would presumably place each other at the top of their rank order lists, ensuring the desired match. However, the MSMP matching process is associated with some degree of pre-Match uncertainty. While candidates can voluntarily communicate their ranking intentions, the NRMP prohibits a program from asking applicants to disclose such intentions.57  Similarly, while a program can communicate its ranking intentions, program directors may be dissuaded from doing so in light of NRMP's policy regarding potentially misleading communication with applicants.7  Regardless, declarations of exclusive or unique interest are difficult, if not impossible, for either party to substantiate formally. Thus, without the security of a formal pre-Match agreement, candidates with an exclusive preference for a specific program may feel compelled to interview at other institutions as a safeguard against not matching. This can be costly in terms of time, effort, and money. Similarly, without a pre-Match agreement, a program may feel a need to interview additional candidates as a safeguard against not filling.

We acknowledged that out-of-Match offers could be employed as a competitive advantage; if a program can secure an early commitment from a desirable candidate, the candidate is effectively removed from the applicant pool. This can represent a lost opportunity for other programs; it may undercut other programs' ability to compete for the candidate's interest, and interview cancellations may negatively affect other programs' ability to achieve the desired number of interviews. Our group endorsed the notion that recruitment efforts should focus on how program characteristics align with candidates' needs, and we agreed that introducing other considerations (eg, early position security) could inappropriately disrupt the competitive landscape. We were also cognizant that the endocrinology candidate-to-position ratio had decreased from 1.6 to 1.0 in 2012 to 1.2 to 1.0 in 2016.8  If endocrinology were to become undersubscribed—as has recently occurred with nephrology2,3—competitive considerations could prompt programs to more aggressively pursue out-of-Match arrangements, perhaps even offering positions before sufficiently evaluating all available applicants.4  Such an environment could undermine trainee confidence in the prevailing system of fellowship position allocations in addition to eroding the morale of training program leaders.

Both the working group and APDEM Council noted a number of tensions inherent to this policy issue. A desire to allow out-of-Match agreements that would be highly beneficial to candidates was in tension with a desire to prevent out-of-Match offers that would not be in candidates' best interest. Similarly, a desire to maximize candidates' autonomy to pursue any opportunity they find desirable was in tension with a desire to mitigate the power asymmetry inherent to most out-of-Match offers. Additionally, APDEM's desire to maximize program autonomy was in tension with its desire to ensure a level playing field for programs.

The APDEM Council carefully considered the working group's deliberations, results from the 2015 program director and fellow surveys (provided as online supplemental material), results from a follow-up fellow survey in 2016 (provided as online supplemental material), and informal feedback from program directors. In late 2016, the APDEM Council endorsed transition to an All In Policy, citing a belief that, compared to available alternatives, it would (1) maximize applicant autonomy by protecting each applicant's ability to evaluate programs without undue pressure; (2) enhance the overall success of the Match, with more applicants achieving better Match outcomes vis-à-vis their individual preferences; (3) best safeguard the integrity of endocrinology fellowship position allocation; and (4) maximize procedural fairness among programs competing for a common pool of applicants. However, the APDEM Council resolved to allow program directors to make the decision via a formal registration of preferences. The council stipulated that it would implement an All In Policy if both (1) the participation rate met or exceeded 75%, and (2) at least two-thirds of registered preferences aligned with the policy. All endocrinology program directors were provided with a summary of working group deliberations, a proposed All In Policy, and other materials for careful consideration (available at http://www.apdem.org). Importantly, the proposed policy included planned exceptions in selected circumstances (table 2). The formal survey was administered from April through June 2017; the participation rate was 92% (130 of 141 program directors), and 80% of respondents indicated a preference for an All In Policy. Eleven program directors did not participate, making 74% the lowest possible favorable rate among program directors.

table 2

Potential Exceptions to Association of Program Directors in Endocrinology, Diabetes, and Metabolism's (APDEM's) All In Match Policya

Potential Exceptions to Association of Program Directors in Endocrinology, Diabetes, and Metabolism's (APDEM's) All In Match Policya
Potential Exceptions to Association of Program Directors in Endocrinology, Diabetes, and Metabolism's (APDEM's) All In Match Policya

Based on the foregoing, APDEM voluntarily adopted an All In Policy beginning with the 2018 Match season. This decision reflected careful longitudinal deliberation by many individuals (provided as online supplemental material) in addition to the endorsement of a majority of program directors. Although we conducted 2 fellow surveys to assess the values and concerns of resident candidates for endocrinology fellowship programs (provided as online supplemental material), we did not include endocrinology-bound residents or endocrinology fellows in the working group; doing so would have likely strengthened our deliberation process. The APDEM also recognizes that it may need to refine the All In Policy in response to ongoing stakeholder feedback, and it recognizes the importance of promoting the highest professional standards with regard to fellowship recruitment.6,7  The APDEM's overarching goal is to achieve a Match policy that will be as beneficial as possible for candidates and programs.

In light of these considerations, we call on all medical subspecialties to carefully consider their current Match policies—regardless of perceived urgency—and to prioritize candidate needs in such deliberations. We suggest that the unique facets of our All In Policy (eg, planned exceptions as described in table 2) can be considered by the minority of medical subspecialties that have already adopted similar policies. However, we believe that our deliberations may be even more relevant to the majority of medical subspecialties that have not adopted All In policies. We suspect that the values and tensions of other subspecialty candidates are not substantially different from our subspecialty's candidates, and we suggest that some perceived advantages of an All In Policy may be even more compelling for other subspecialties. For example, when a subspecialty is oversubscribed, with substantially more candidates than positions, candidates may be at even higher risk for undue influence exerted by out-of-Match offers (eg, such candidates may be more willing to accept a relatively undesirable position as a hedge against not matching). While each subspecialty must make its own decisions in this regard, we challenge such subspecialty communities to consider carefully whether there are compelling reasons not to adopt an All In Match policy.

1
National Resident Matching Program
.
All In policy: fellowship matches
. ,
2019
.
2
Hsu
CY,
Parker
MG,
Ross
MJ,
Schmidt
RJ,
Harris
RC;
ASN Nephrology Match Task Force. Improving the nephrology match: the path forward
.
J Am Soc Nephrol
.
2015
;
26
(
11
):
2634
2639
. .
3
Alliance for Academic Internal Medicine
.
View summit report: the benefits and obstacles for subspecialty fellowship applicants and programs if the NRMP/SMS “All-In” policy is adopted
. ,
2019
.
4
Proctor
DD,
Decross
AJ,
Willis
CE,
Jones
TN,
Pardi
DS.
The Match: five years later
.
Gastroenterology
.
2011
;
140
(
1
):
15
18
. .
5
National Resident Matching Program
.
Specialties matching service Match participation agreement for all matches opening after June 30, 2018
. ,
2019
.
6
National Resident Matching Program
.
Professionalism and the Match
. ,
2019
.
7
National Resident Matching Program
.
Match communication code of conduct
. ,
2019
.
8
National Resident Matching Program
.
Results and data: specialties matching service 2016 appointment year
. ,
2019
.

Author notes

Editor's Note: The online version of this article contains the 2015 Association of Program Directors in Endocrinology, Diabetes, and Metabolism (APDEM) Program Director Survey results, the 2015 and 2016 APDEM Fellow Survey results, and a table of participants in APDEM deliberations.

The authors would like to thank the Association of Program Directors in Endocrinology, Diabetes, and Metabolism staff members for their support and guidance throughout the decision-making processes described in this manuscript: Ailene Cantelmi, Nancy Chill, Amanda Perl, Barbara Small, Alexa Stout, and Stacey Taylor.

Supplementary data