National Resident Matching Program

from Wikipedia, the free encyclopedia

The National Resident Matching Program ( NRMP ), also known as The Match , is a United States-based, private, non-profit , non-governmental organization . This was founded in 1952 to assign medical students to a specialist training program at American university hospitals. In addition to the annual Main Residency Match , which includes more than 43,000 applicants and 31,000 positions, the NRMP conducts grant assignments for more than 60 sub-specializations through its Specialties Matching Service (SMS).

The NRMP is funded by a board of directors consisting of deans of medical faculties, heads of university clinics, heads of medical training programs and a member of the public.

NRMP International, a subsidiary of the National Resident Matching Program, was founded in 2010 to provide medical matching services outside of the United States and Canada.

history

Shortly after the first specialist training programs were formally introduced in the 1920s, the recruitment process was "characterized by intense competition between hospitals for (insufficient supply of) specialists". In general, hospitals benefited from filling their positions as early as possible, and applicants benefited from delaying accepting positions. The combination of these factors resulted in vacancies being made up to two years before the start of postgraduate training.

In 1945, medical schools decided not to publish testimonials or allow letters of recommendation to be written until a certain date. In this way, they managed to move the date of the choice of place for specialist training back to the fourth year of study. However, the competition for specialists only took a different form. The programs began to send out their offers with a time limit in which to send a response. This limit was gradually reduced from 10 days in 1945 to less than 12 hours in 1950. The students got “exploding” offers that demanded a decision about their education before opening.

In the early 1950s, the National Interassociation Committee on Internships (NICI) examined existing matching plans and selected the Boston Pool Plan, which was being used by programs in the Boston area at the time, as a model for a test run of a new centralized system. In October 1951, student representatives from 79 medical schools formed the National Student Internship Committee (NSIC) to discuss the results of the NICI test-match process and to consider a NICI proposal to replicate the Boston Pool Plan nationally. The NSCI requested a modification of the algorithm so that the applicants are presented more evenly. The modified algorithm was introduced and used in the first match on April 15, 1952. This match was a success, so the NICI recommended that an organization be set up to manage and monitor the matching process. This organization, known as the National Intern Matching Program (NIMP), was founded on January 7, 1953.

In 1951, due to concerns that the matching process favored hospitals over students, students proposed a modification of the algorithm. The students felt that the algorithm gave them the incentive to mask their true preferences. A publication by David Gale and Lloyd Shapley in 1962 stated that there is always a stable solution when universities are matched with students, but it is possible to prefer universities as a group over applicants as a group (and vice versa). That is, Gale and Shapley found that there is a stable match that is optimal for universities and one that is optimal for students. Lloyd Shapley and Alvin Roth won the Nobel Prize in Economics in 2012 for their work on stable allocations.

A debate arose as to whether the matching program was vulnerable to manipulation or inappropriately fair to programs. Indeed, in simple cases (ie, those excluding pairs, second year programs, and special cases for dealing with unfilled spaces) that had multiple "stable" matchings, the algorithm was shown to return a solution that matched programs' preferences the applicant preferred. Correspondence in the New England Journal of Medicine in 1981 found that the algorithm used was program-optimal for individual applicants.

Later researchers such as Marilda Sotomayor 1983, Alvin Roth 1984 and Klaus et al. 2007, noted that if couples are allowed to be matched together, there may not be stable matching.

The NRMP algorithm experienced only minor and incremental changes after its introduction in 1952. In the fall of 1995, however, the NRMP Board of Directors commissioned a preliminary research study evaluating the current algorithm, as well as a study comparing a new algorithm with the existing one. Recommendations for changes to be considered in its functioning and description should also be explored. In May 1997, a new candidate selection algorithm was used and has been in use since it was first used in March 1998. Even though the study showed that the net effect of the change on actual matches was minimal.

Matching algorithm

Matching applicants with programs is a generalization of the stable marriage problem ; therefore the solutions are very similar. A simplified version of the algorithm used to perform the matching process is described below and on the NRMP website. However, this description does not include dealing with couples (couples of applicants who participate in a match together, perhaps to stay in the same geographic location), second-year positions or special dealing with specialist positions that are not filled. The full algorithm is described in Alvin Roth, Elliott Peranson (September 1999).

Input

The application process for residency training begins before the main residency match opens in September. Applications are usually routed to the programs through the Electronic Residency Application Service (ERAS), a service of the Association of American Medical Colleges. After applicants apply for programs, the programs review applications and invite selected applicants for interviews. The interviews take place between October and February. At the end of the interview period, programs and applicants each create “ranking lists” which they submit to the NRMP. The programs list all applicants they wish to train, in order from most to least preferred. The applicants categorize the programs in which they would like to be trained accordingly. For applicants who would like to be matched as a pair, the ranking lists contain pairs of selected programs that are simultaneously taken into account by the matching algorithm. Applicant rankings may include a combination of categorical programs (training that lasts 3-5 years and begins in the first postgraduate year); preparatory programs (training that lasts one year and begins in the first postgraduate year); or advanced programs (training that lasts 3-4 years and begins after one or more years of preparatory training). For advanced programs on the leaderboard, applicants can add a supplementary list of preparatory programs to be assigned to a full training course.

Simple case

The matching process begins with an attempt to match an applicant to the program he or she most prefers according to his ranking. If the applicant cannot be assigned to the program of their first choice, attempts will be made to place the applicant in the program of their second choice and so on until the applicant is provisionally assigned to a program which has an open position and prefers that applicant or all Choices from the applicant's ranking list have been exhausted. This process is carried out for all applicants until each applicant has either been provisionally assigned to his or her most preferred possible choice or all of the applicants' choices have been exhausted. Subsequently, preliminary matches become final.

To understand how the current NRMP algorithm works, it is helpful to first consider the simpler case where there are no pairs or secondary programs.

As with the stable marriage problem, the basic goal is to match applicants with programs so that the results are "stable". "Stability" in this case means that there is no applicant A and no program P, so that both of the following statements are true:

  • A is not assigned to any program or would prefer to be in program P than the program A was matched with
  • P has a vacancy or would prefer A to another applicant assigned to the program.

It is possible to show that for every example of the problem there is at least one valid solution. Under the old (pre-1995) NRMP algorithm, which favored the programs' preferences over applicants, in certain cases programs could benefit from lying about their preferences. This no longer applies under the current algorithm. Neither applicants nor programs can benefit from being untruthful about their preferences, even if they are perfectly informed about everyone's preferences.

Also, under the current system, it is not possible for an applicant to fare worse by adding more specialist training programs to the bottom of a list when those programs are in fact preferred to non-allocation.

Couples

The ranking lists of pairs are processed simultaneously by the matching algorithm, which complicates the problem. In some cases there is no stable solution (stable as defined in the simple case). Indeed, the problem of whether there is a stable solution and the determination of its existence has been shown to be NP-complete . Even if there is no randomization in the NRMP algorithm, it always delivers the same output if it receives exactly the same input. Different results can be generated by changing trivial characteristics of the data, such as the order in which applicants and programs are to be processed. However, when the algorithm was first tested with specialist matching data from 5 years and a large number of different initial conditions, the current NRMP algorithm always quickly ended in a stable solution. Tests also showed that "none of the [trivial] sequencing decisions had a large or systematic effect on matching" - the maximum number of applicants ever affected in a single run was 12 out of 22,938.

Once the programs are ranked, there is usually no opportunity for an applicant to be placed in a better position if they choose to be matched as part of a couple. For example, if a very strong applicant and a very weak applicant can be matched as a pair, there is no mechanism in the algorithm that enables the stronger applicant to improve the attractiveness of the weaker applicant in any way. Of course, if the programs know before processing the matching algorithm that the stronger and weaker applicants are participating in the match as a couple, they are free to change their lists accordingly. This, in turn, could affect the bottom line.

Failure to match

It is possible that an applicant will not be matched with any program. Until the Main Residency Match in 2010, applicants who had not been given a job went through a process called scramble. At 12:00 noon on the Monday of Match Week, the NRMP notified applicants whether they had been assigned to a program (but without disclosing the name of the program) and published a list of the unoccupied programs. Applicants then applied en masse to programs with open positions, often having to change their preferred specialization in the process. The scramble was widely seen as chaotic, disorganized, and not very transparent. The scramble ended on Match Day, the Thursday of Match Week. Most positions were filled within the first few hours and almost all within the first 48 hours. Scrambling was extremely competitive: In 2008 around 10,600 applicants, many of whom were trained abroad, competed for only 1,392 specialist positions.

After the Main Residency Match in 2010, the Scramble was replaced by the Supplemental Offer and Acceptance Program, or SOAP. In the SOAP, unassigned applicants are offered positions in unoccupied programs through a series of rounds, creating a systematic opportunity for applicants to find training positions without the chaos of scramble. In SOAP, all appointments are made via the NRMP (in contrast to scramble, direct matching is not possible). Data on unmatched applicants and unfilled programs will be released on the Monday of Match Week at the same time. Match Day is now the Friday of Match Week.

International Medicine Graduates

To participate in the NRMP, an international medical graduate must meet the requirements for ECFMG certification by the Rank Order List Certification Deadline in February of the matching year (not required for medical graduates from Canadian medical schools in the USA are not considered as international medical graduates).

To obtain ECFMG certification, international medicine graduates must:

  • achieve the required number of points in USMLE Step 1, USMLE Step 2 Clinical Knowledge and USMLE Step 2 Clinical Skills.
  • Receive a medical diploma from a medical education from an institution registered in the International Medical Education Directory (IMED).

In comparison, undergraduate and postgraduate medical school graduates in the United States are bound by college graduation requirements. Applicants educated in both the U.S. and overseas can participate in the NRMP while completing their final year of medical school before earning their medical degree.

Legal proceedings

In 2002, 16 law firms filed a lawsuit on behalf of 3 medical specialists to represent all medical specialists matched by the NRMP. Plaintiffs brought a case in which they sought to demonstrate that the NRMP was working with other national medical organizations, medical training organizations and institutions promoting specialist training to lower specialist doctors' wages in breach of antitrust law United States violates.

In 2004, after lobbying by the American Medical Colleges and the American Hospital Association, an add-on clause granting the NRMP specific immunity was added to a pension bill signed by President George W. Bush. The ordinance was supported by Senators Edward M. Kennedy, Democrat of Massachusetts, and Judd Gregg, Republican of New Hampshire. A federal district court then dismissed the case.

Implementation in software packages

  • R : The Roth-Peranson algorithm used in the NRMP is implemented as part of the package matchingMarkets.
  • API : The MatchingTools API provides the Roth-Peranson algorithm for the matching problem with pairs used in the NRMP via a free programming interface.

Individual evidence

  1. ↑ NRMP homepage.Retrieved January 31, 2018
  2. ^ A b Dan Gusfield: 1.1.1 . In: The Stable Marriage Problem: Structure and Algorithms . The MIT Press, 1989, ISBN 0-262-07118-5 , pp. 3-4.
    Description of market based on AE Roth: The evolution of the labor market for medical interns and residents: a case study in game theory . In: Journal of Political Economy . 92, 1984, pp. 991-1016. doi : 10.1086 / 261272 .
  3. ^ Alvin E. Roth: The Theory and Practice of Market Design . Nobel Media AB. December 8, 2012.
  4. ↑ Medical seniors hit internal plan. New York Times 1951; 22 Oct: 25 (col. 1)
  5. ↑ JG Shiller. An alum recalls the first matching plan. P&S: The Journal of the College of Physicians & Surgeons of Columbia University 1985; 5:29
  6. ↑ KJ Williams. A reexamination of the NRMP matching algorithm. Acad Med 1995; 70: 470-476
  7. ↑ E. Peranson, R. Randlett. Comments on Williams' "A reexamination of the NRMP matching algorithm". Acad Med. 1995; 70: 490-494 (footnote at the bottom of page 492)
  8. ↑ a b D. Gale, LS Shapley. College admissions and the stability of marriage. American Mathematics Monthly. 1962; 69: 9-15
  9. ↑ Roth "Redesign" 748
  10. ^ Sara Robinson: Are Medical Students Meeting Their (Best Possible) Match? Archived from the original on November 18, 2016. Info: The archive link was inserted automatically and has not yet been checked. Please check the original and archive link according to the instructions and then remove this notice. In: SIAM News . No. 3, April 2003, p. 36. Retrieved January 31, 2018. @1@ 2Template: Webachiv / IABot / www.siam.org
  11. ↑ Gusfield “Stable Marriage” 64 references AE Roth: The evolution of the labor market for medical interns and residents: a case study in game theory . In: Journal of Political Economy . 92, 1984, pp. 991-1016. doi : 10.1086 / 261272 . as proving that the pre-1995 algorithm is essentially the hospital-optimal algorithm described in Gusfield 39. Gusfield 41 demonstrates that the hospital-optimal algorithm is also applicant-pessimal.
  12. ↑ KJ Williams, VP Werth, JA Wolff. An analysis of the resident match. N Engl J Med. 1981; 304: 1165-1166; correspondence in N Engl J Med. 1981; 305: 526
  13. ^ Roth AE. The evolution of the labor market for medical interns and residents: a case study in game theory. Journal of Political Economy 1984; 92: 991-1016
  14. ^ B. Klaus, F. Klijn, J. Massó. Some things couples always wanted to know about stable matchings (but were afraid to ask). Review of Economic Design 2007; 11: 175-184
  15. ↑ a b c Roth "Redesign" 749
  16. ^ Evaluation of changes to be considered in the NRMP algorithm , by Alvin E. Roth. October 24, 1995
  17. ↑ Roth "Redesign" 752, 760
  18. ↑ NRMP site
  19. ↑ Alvin Roth, Elliott Peranson: The redesign of the Matching Market for American Physicians: Some Engineering Aspects of Economic Design . In: The American Economic Review . 89, No. 4, September 1999, pp. 756-757. doi : 10.1257 / aer.89.4.748 . Accessed January 31, 2018.
  20. ↑ Gusfield "Stable Marriage" 38
  21. ↑ Gusfield "Stable Marriage" 41
  22. ^ Gusfield "Stable Marriage" 59
  23. ^ Analysis of the National Residency Matching Program . Tedlab.mit.edu. Accessed January 31, 2018.
  24. ↑ Gusfield "Stable Marriage" 54 gives an example of a situation with no stable solution and states that proof of NP completeness comes from Eytan Ronn: NP-complete stable matching problems . In: Journal of Algorithms . 11, No. 2, June 1990, ISSN  0196-6774 , pp. 285-304. doi : 10.1016 / 0196-6774 (90) 90007-2 .
  25. ↑ Roth "Redesign" 759
  26. ↑ Roth "Redesign" 757
  27. ↑ Roth "Redesign" 758
  28. ↑ Alvin E. Roth, Elliott Peranson: The Redesign of the Matching Market for American Physicians: Some Engineering Aspects of Economic Design Archived from the original on June 11, 2011. Information: The archive link was automatically inserted and not yet checked. Please check the original and archive link according to the instructions and then remove this notice. In: The American Economic Review . 89, No. 4, September 1999, pp. 756-757. doi : 10.1257 / aer.89.4.748 . Accessed January 31, 2018. @1@ 2Template: Webachiv / IABot / www.bepress.com
  29. ↑ a b c [1] https://web.archive.org/web/20111226032103/http://www.nrmp.org/soap.pdf ( Memento from December 26, 2011 in the web archive archive.today )
  30. ↑ a b Using "SOAP" to Clean Up the Scramble . Studentdoctor.net. October 24, 2010. Accessed January 31, 2018.
  31. ^ The Residency Scramble: How It Works and How It Can Be Improved . Studentdoctor.net. March 12, 2009. Archived from the original on August 18, 2013. Info: The archive link was inserted automatically and has not yet been checked. Please check the original and archive link according to the instructions and then remove this notice. Accessed January 31, 2018. @1@ 2Template: Webachiv / IABot / studentdoctor.net
  32. ↑ [ https://web.archive.org/web/20111202180322/http://www.nrmp.org/res_match/faq/us_seniors_faq.html Frequently Asked Questions US Seniors] . Archived from the original on December 2, 2011. Retrieved January 31, 2018.
  33. ↑ nrmp.org> 2012 Main Match Schedule ( Memento of the original from July 16, 2011 in the Internet Archive ) Info: The archive link was automatically inserted and not yet checked. Please check the original and archive link according to the instructions and then remove this notice. Updated May 19, 2011 @1@ 2Template: Webachiv / IABot / www.nrmp.org
  34. ↑ nrmp.org> Independent Applicants ( Memento of the original from July 21, 2011 in the Internet Archive ) Info: The archive link was automatically inserted and not yet checked. Please check the original and archive link according to the instructions and then remove this notice. Updated September 8, 2010 @1@ 2Template: Webachiv / IABot / www.nrmp.org
  35. ↑ nrmp.org> US Seniors> Registering with the NRMP ( Memento of the original of July 21, 2011 in the Internet Archive ) Info: The archive link was automatically inserted and not yet checked. Please check the original and archive link according to the instructions and then remove this notice. Updated August 2010. @1@ 2Template: Webachiv / IABot / www.nrmp.org
  36. ^ Antitrust Against the Resident Match. http://isites.harvard.edu/fs/docs/icb.topic926634.files/7.1%20Antitrust%20Lawsuit%20Against%20the%20Resident%20Match.pdf
  37. ^ S. Robinson. Antitrust Lawsuit Over Medical Residency System Is Dismissed. Aug. 2004. https://www.nytimes.com/2004/08/14/politics/14match.html
  38. ^ T. Klein: Analysis of Stable Matchings in R: Package matchingMarkets . In: Vignette to R Package matchingMarkets . 2015.
  39. ^ MatchingMarkets: Analysis of Stable Matchings . In: R Project .
  40. ↑ MatchingTools API .

Web links