SCIENTIFIC ARTICLE
The Hand Surgery Fellowship Application Process: Expectations, Logistics, and Costs Clifton Meals, MD, Meredith Osterman, MD
Purpose To investigate expectations, logistics, and costs relevant to the hand surgery fellowship application process. We sought to discover (1) what both applicants and program directors are seeking, (2) what both parties have to offer, (3) how both parties collect information about each other, and (4) the costs incurred in arranging each match. Methods We conducted on-line surveys of hand surgery fellowship applicants for appointment in 2015 and of current fellowship program directors. Results Sixty-two applicants and 41 program directors completed the survey. Results revealed applicants’ demographic characteristics, qualifications, method of ranking hand fellowship programs, costs incurred (both monetary and opportunity) during the application process, ultimate match status, and suggestions for change. Results also revealed program directors’ program demographics, rationale for offering interviews and favorably ranking applicants, application-related logistical details, costs incurred (both monetary and opportunity) during the application process, and suggestions for change. Conclusions Applicants for hand surgery fellowship training are primarily interested in a potential program’s academic reputation, emphasis on orthopedic surgery, and location. The typical, successfully matched applicant was a 30-year-old male orthopedic resident with 3 publications to his credit. Applicants rely on peers and Web sites for information about fellowships. Fellowship directors are primarily seeking applicants recommended by other experienced surgeons and with positive personality traits. The typical fellowship director offers a single year of orthopedic-based fellowship training to 2 fellows per year and relies on a common application and in-person interviews to collect information about applicants. Applicants appear to be more concerned than directors about the current state of the match process. Applicants and directors alike incur heavy costs, in both dollars and opportunity, to arrange each match. A nuanced understanding of the match process suggests specific changes and may help reduce these costs. (J Hand Surg Am. 2015;(-):-e-. Copyright Ó 2015 by the American Society for Surgery of the Hand. All rights reserved.) Type of study/level of evidence Economic and decision analysis V. Key words Application, fellowship, hand, match, surgery.
From the Department of Orthopedics, George Washington University School of Medicine and Health Sciences, Washington, DC; and the Philadelphia Hand Center, Philadelphia, PA. Received for publication September 15, 2014; accepted in revised form December 23, 2014. No benefits in any form have been received or will be received related directly or indirectly to the subject of this article.
Corresponding author: Clifton Meals, MD, Department of Orthopedics, George Washington University School of Medicine and Health Sciences, 2150 Pennsylvania Ave., NW, Washington, DC 20037; e-mail: [email protected]. 0363-5023/15/---0001$36.00/0 http://dx.doi.org/10.1016/j.jhsa.2014.12.041
Ó 2015 ASSH
r
Published by Elsevier, Inc. All rights reserved.
r
1
2
T
HAND SURGERY FELLOWSHIP APPLICATION PROCESS
match may be understood as an economic market in which applicants offer skill, experience, enthusiasm, and other factors in exchange for advanced training. The National Residency Matching Program (NRMP) facilitates matches and publishes basic information related to each year’s application process. However, precisely what applicants and directors want, what both parties have to offer, and how both parties collect information remain obscure. Two types of costs, monetary and opportunity, are, anecdotally, a source of frustration for applicants and program directors; however, these costs are poorly understood. The monetary costs of the match have not been enumerated. Opportunity cost is the cost of forgoing a next-best option (eg, the cost of putting off clinic) to interview applicants.1,2 These costs are less concrete but no less important and no less obscure. Monetary and opportunity costs might be reduced if applicants and program directors knew how their energies were best directed. If applicants knew how their credentials compared with those of their peers, they might selfselect (ie, chose programs more likely to accept them) early in the application process. If directors knew the interviewing behavior of other programs, they might optimize the number of interviews they offer and/or coordinate interviews with other sites. Data to inform such decision making are lacking. We therefore sought to collect information related to match expectations, logistics, and costs through surveys of applicants for hand fellowship appointment in 2015 as well as current hand surgery fellowship directors.
Between January and May 2014, via in-person and e-mailed inquiries, we collected the e-mail addresses of 118 applicants for 2015 appointment. We invited them to take an on-line survey in June, sent 2 reminder invitations at weekly intervals, and closed the survey after a month. To incentivize participation, we informed potential respondents that their participation in the survey would be acknowledged but that their individual responses would remain anonymous. We identified 84 program directors and codirectors, corresponding to the 81 Accreditation Council for Graduate Medical Education accredited hand surgery fellowships. We invited them to take the on-line survey in August 2014, sent 2 reminder invitations at weekly intervals, and closed the survey after a month. We informed program director respondents that their participation in the survey would be anonymous unless they asked to be acknowledged. Both surveys were facilitated by a commercial Web site (www.surveymonkey.com; SurveyMonkey Inc., Palo Alto, CA). Data analysis involved traditional statistical methods and was aided by the same commercial Web site that facilitated the surveys.
HE HAND SURGERY FELLOWSHIP
RESULTS Applicant survey Of the 118 applicants for 2015 appointment we contacted, 61 (52%) completed the survey, and 3 partially completed the survey. Eighty-one percent of respondents were men. The median age among respondents was 30 years. All respondents were MDs, and 14% had other advanced degrees including 1 PhD. Thirty-five percent of respondents were members of the Alpha Omega Alpha medical honor society, 51% were not, and 14% indicated that their medical schools did not have Alpha Omega Alpha medical honor society chapters. Seventy-eight percent had career or research experience other than publications, presentations, awards, or grants. Eighty percent were enrolled in or had completed orthopedic surgery residency training, and 20% were enrolled in or had completed plastic surgery residency training. One respondent had already completed another fellowship (microsurgery and pediatric plastic surgery). Five percent of respondents had previously applied for hand surgery fellowship training. Respondents’ median number of publications was 3 and their median number of national podium or poster presentations was 2. All but 2 respondents had completed a dedicated hand surgery residency rotation before applying for
MATERIALS AND METHODS We conducted on-line surveys of both hand surgery fellowship applicants for appointment in 2015 and current hand surgery fellowship directors. The full surveys are presented in Appendix A (available on the Journal’s Web site at www.jhandsurg.org). The applicant survey consisted of 37 questions designed to reveal each respondent’s demographics, qualifications, method of ranking hand fellowship programs, expenses and opportunity costs incurred traveling to interviews, ultimate match status, and suggestions for change. The fellowship director survey consisted of 15 questions designed to reveal each respondent’s program demographics; rationale for offering interviews and favorably ranking applicants; application-related logistic details; costs incurred, both monetary and opportunity, during the application process; and suggestions for change. J Hand Surg Am.
r
Vol. -, - 2015
HAND SURGERY FELLOWSHIP APPLICATION PROCESS
Of 61 respondents who answered the question, 100% matched. The greatest share (36%) matched with their first choice. Fifteen percent matched with their second choice. In response to the question “If you did not match, what are you plans for the future?” 1 respondent planned to apply for a different subspecialty fellowship and another planned to apply for 2016 hand surgery fellowship appointment. The final survey question was “How might the match process be improved?” and a free-text response was required. These responses were placed into categories and are represented in Table 1. The raw data obtained from the applicant survey are presented in Appendix A (available on the Journal’s Web site at www.jhandsurg.org).
FIGURE 1: How applicants valued different sources of information in selecting fellowship programs (0, not at all; 1, somewhat; 2, very much).
fellowship training. How applicants valued different sources of information in selecting fellowship programs is represented in Figure 1. How respondents ranked various factors in their decision to pursue a specific fellowship program is represented in Figure 2. Applicants for 2015 appointment applied to a median of 30 fellowships and indicated that, on average, 5 programs asked for material in addition to the Universal Hand Application. Respondents were invited to a median of 20 interviews. Of these, the largest share accepted 12 interview invitations. Seventy-four percent of respondents accepted and then cancelled at least 1 interview. The most common reason cited for cancelling an interview was a conflicting interview. Less common reasons included expense and lack of interest in equal proportion. Sixty-nine percent of respondents had their first interview in January and their last interview in March 2014. Sixty-four percent of respondents traveled more than 2,400 km (1,500 miles) to their most distant interview. The median days spent away from home while interviewing was 20. Thirtyfour percent of respondents spent unplanned and extra days owing to missed or cancelled travel arrangements, with inclement weather being an important cause. The largest share of respondents indicated that they spent between $2,500 and $5,000 on interview-related travel expenses. The majority of respondents paid for their expenses principally with savings, and 36% borrowed money on credit cards and/ or were assisted by family. Seventy-four percent of respondents reported that a faculty member at their home institution made personal contact with 1 or more hand surgery fellowship directors on the applicant’s behalf. Seventy-nine percent of respondents personally contacted at least 1 fellowship director to express particular interest in a program. When asked if visiting fellowship institutions helped respondents to order their rank list, the responses were 48% very much, 33% somewhat, and 20% not at all. Respondents ranked a median of 10 programs. J Hand Surg Am.
3
Program director survey Of the 85 potential respondents, 41 (48%) completed the survey, and 7 partially completed the survey. The proportions of directors offering orthopedic, combined, or plastic programs were 69%, 25%, and 6%, respectively. Thirty-nine (80%) respondents were the sole director of their program and 9 were codirectors. The proportions of directors offering 1, 2, 3, 4, or more fellowship positions are represented in Table 2. Four directors offered optional 2-year fellowship positions. All other available positions were for 1 year. How directors valued different applicant qualifications in deciding to grant an interview is represented in Figure 3. How directors valued different applicant attributes in ordering a rank list is represented in Figure 4. The distribution of directors’ received number of applications is represented in Table 3. The distributions of directors’ number of offered interviews, conducted interviews, and number of applicants ranked are represented in Table 4. The majority (88%) of respondents did not contact applicants after the interview process to express interest, 4 directors contacted only select applicants, and 1 contacted all interviewees. Respondents were asked to estimate the opportunity cost incurred by their department during the interview process with the following formula: Cost ðin hoursÞ ¼ ½# involved faculty ½average hours devoted by each faculty member
The average opportunity cost reported was 65 hours. The average monetary cost incurred by each respondent’s department during the interview process was $4,572. r
Vol. -, - 2015
4
HAND SURGERY FELLOWSHIP APPLICATION PROCESS
FIGURE 2: How applicants valued different factors in their decision to pursue specific fellowship programs (0, not at all; 1, somewhat; 2, very much). Acad., academic; avail., availability; non-acad., non-academic; ortho., orthopedic; surg., surgery.
TABLE 1.
Suggestions for Change
Category
Applicant Survey (61 Respondents, 104 Suggestions)
Director Survey (41 Respondents, 44 Suggestions)
Proportion of Suggestions
Proportion of Suggestions
25%
16%
Centralized or coordinated interviews, or both Universal or electronic application, or both
21%
9%
Change in timing of deadlines, interviews, or match, or all
17%
2%
Other
13%
2%
Higher quality or better availability of program information, or both
12%
5%
Limits on the number of interviews offered or accepted, or both
8%
NA
No suggestion
4%
45%
Reduce actions not permitted by the NRMP
NA
14%
Alternatives to in-person interviews
NA
7%
TABLE 2.
The raw data obtained from the director survey are presented in Appendix A (available on the Journal’s Web site at www.jhandsurg.org).
Proportions of Program Sizes Proportion of Directors Offering Different Numbers of Fellowship Positions (48 Respondents)
1
33%
2
46%
3
13%
4
6%
>4
2%
DISCUSSION The hand surgery fellowship application process was initially informal. Applicants contacted fellowship programs by phone or mail and provided their credentials before being selected to interview. Positions were offered and accepted variably by a centralized service, agreement between fellowship directors, and/ or by arrangement between fellowship directors and applicants (A. L. Osterman and R. Meals, personal communications, February, 2015). As interest in hand surgery has expanded, the fellowship matching process has increased in scale and sophistication. Since appointment year 1993, matches have been facilitated by a third-party matching service. The centralized hand surgery fellowship match is a model of success. Several orthopedic subspecialties do not offer a third-party matching program. Others
Sixty-three percent of fellowship directors believed it was important for an applicant to visit their institution in order to effectively rank the applicant. The remainder of directors did not think this was important. The final survey question was “How might the match process be improved?” and a free-text response was required. These responses were placed into categories and are represented in Table 1. J Hand Surg Am.
r
Vol. -, - 2015
5
HAND SURGERY FELLOWSHIP APPLICATION PROCESS
TABLE 3. Proportions of Number of Received Applications Proportion of Directors’ Received Number of Applications (42 Respondents)
FIGURE 3: How directors valued different applicant qualifications in deciding to grant an interview (0, not at all; 1, somewhat; 2, very much). AOA, American Orthopedic Association; educ., education; ortho, orthopedic surgery; plastics., plastic surgery; pregrad, pregraduate; reput., reputation; USMLE, United States Medical Licensing Examination.
14% 22%
51e75
26%
76e100
21%
101e125
12%
126e150
5%
ask applicant survey respondents, whose participation would be acknowledged in aggregate, whether they had been asked to arrange for a match outside of the NRMP, because we assumed responses to this question would be unreliable. Anecdotally, neither of us encountered any obvious attempts to subvert the match process. Although we were not specifically asked to provide postinterview assurances, we recall that applicants on interview days were sometimes asked to express interest in a program should applicants wish to match there. This may skirt the NRMP mandate that “program directors shall not solicit or require post-interview communication from applicants.”4 Solicited or not, we discovered that thank you letters were sent from interviewees to directors and vice versa. This heeds the NRMP’s directive that “both applicants and programs may express their interest in each other; however, they shall not solicit verbal or written statements implying a commitment.”4 Only one applicant respondent referred to illicit deal-making. Although 17% of director respondents suggested that directors sometimes cheat, it seemed some of these respondents viewed expressing postinterview interest as prohibited, when in fact the NRMP permits this. The NRMP governs only a few of the application process’s many variables, and whereas the match may be fair, the process likely places undue burden on both program directors and applicants. Applicants, in particular, may bear a disproportionate share of this burden as recent trends suggest a sellers’—that is, directors’—market. The number of hand fellowship applicants increased from 155 for appointment in 2010 to 205 for appointment in 2014. The number of available fellowship positions has increased slightly, although applicants have outnumbered available spots. Beginning with the 2011 appointment year, roughly one-quarter of applicants for hand fellowship training have not matched.5
FIGURE 4: How directors valued different applicant attributes in ordering a rank list (0, not at all, 1; somewhat; 2, very much). Acad., academic; geo., geography; med., medicine.
have seen their centralized matching scheme corrupted by arrangement of appointments outside the match, ignored by a portion of eligible programs, or abandoned.3 The match sets up an economic “Prisoner’s Dilemma.” The Prisoner’s Dilemma is illustrated by a scenario in which 2 accomplices have been arrested. Their fate varies according to whether they both keep mum, one betrays the other, or if they inform on one another. They will benefit from cooperation but both are motivated to cheat and are likely to do so.1 As such, both applicants and program directors benefit by universal adherence to the NRMP but are incentivized to cheat (ie, arrange matches outside of the centralized matching service). It appears to us, however, that little cheating actually occurs. The hand surgery fellowship match has met the NRMP’s requirements for fairness since hand surgery began using this service in 1992. American hand surgery fellowship programs have universally adhered to the NRMP service for the same period. We declined to J Hand Surg Am.
1e25 26e50
r
Vol. -, - 2015
6
HAND SURGERY FELLOWSHIP APPLICATION PROCESS
TABLE 4.
1e10
Proportions of Number of Applicants Offered Interviews, Interviewed, and Ranked Proportion of Directors’ Number of Offered Interviews (42 Respondents)
Proportion of Directors’ Number of Conducted Interviews (42 Respondents)
Proportion of Directors’ Number of Applicants Ranked (42 Respondents)
12%
17%
24%
11e20
22%
29%
36%
21e30
41%
45%
31%
31e40
21%
5%
5%
41e50
2%
2%
2%
51e60
2%
2%
2%
Additional evidence of a sellers’ market includes our finding that, in response to a prompt to suggest change to the current system, 45% of director replies could be categorized as “no change,” whereas only 4% of applicant responses to the same prompt could be similarly described. In this setting of increased competition and asymmetrical concern regarding the application process, no third party regulates what fellowship programs may ask of applicants, how many interviews a program may offer, how many interviews an applicant may accept, or when and where interviews will take place, and so on (P. Stern, A. L. Osterman, P. Benhaim, and R. Meals, personal communications, October, 2014). It has been unclear what makes an applicant competitive or programs desirable. As such, both applicants and program directors have had little to inform the best (ie, efficient) use of resources. We make several suggestions based on our investigation. There is likely a role for increased third-party oversight of those variables outside the jurisdiction of the NRMP. Hand surgery fellowship directors meet at the annual American Society for Surgery of the Hand meeting. This group’s responsibilities, or the responsibilities of a new, independent organization, potentially including both applicants and directors, might be expanded to include:
could take place at national meetings or regional sites, as occurs with other subspecialties.6 Video or phone-based interviews deserve consideration. 6. Encouraging fellowship programs to offer comprehensive, up-to-date, on-line information. 7. Moving the application process to either the fall or the spring to facilitate safe and efficient travel. Earlier interviews might be favored by applicants eager to determine their career plans. Later interviews might allow applicants to make more informed decisions, complete a dedicated hand surgery rotation, and/or better develop relationships with mentors ultimately influencing the match. Our investigation was based on 1 anonymous survey and 1 in which the respondents knew their identities would be published only in aggregate and is, therefore, not entirely reliable. For instance, we are both recent hand surgery fellowship applicants and recall receiving approximately 10 thank you letters and other “expressions of interest” from fellowship programs. This is anecdotal evidence at odds with the finding that only 5 program directors reported contacting applicants after their interview. A second weakness is related to the proportion of potential respondents who completed the survey: roughly onethird of the predicted number of applicants for 2015 appointment and roughly one-half of current fellowship program directors. In addition, selection bias likely resulted in a disproportionate number of successfully matched applicants responding to the survey. Nonetheless, our data reveal several trends. Applicants for hand surgery fellowship training are primarily interested in a potential program’s academic reputation, emphasis on orthopedic surgery, and location. The typical, successfully matched applicant is a 30year-old male orthopedic resident with 3 academic publications to his credit. Applicants rely on peers and Web sites for information about fellowships. Fellowship directors are primarily seeking applicants
1. Limiting the number of interviews offered by fellowship programs and/or accepted by applicants. 2. Encouraging disclosure from program directors regarding the qualifications of their fellows and of their ideal applicants. 3. Instituting an electronic centralized application system, similar to the Electronic Residency Application Service. 4. Coordinating interview dates between programs in the same region. 5. Establishing alternatives to traditional in-person and/ or on-site interviews. Final or screening interviews J Hand Surg Am.
r
Vol. -, - 2015
HAND SURGERY FELLOWSHIP APPLICATION PROCESS
recommended by other experienced surgeons and with positive personality traits as listed in the director survey: maturity, candor, poise, and confidence. The typical fellowship director offers a single year of orthopedic-based fellowship training to 2 fellows per year and relies on a common application and in-person interviews to collect information about applicants. Applicants, in general, appear disproportionately concerned about the match process compared with program directors. Perhaps most importantly, the high costs of the current process, in both money and opportunity are apparent. Monetary cost is familiar. Opportunity expenses (eg, what a resident loses in missing 20 days of training) deserve special attention because they may be important but difficult to quantify. Careful consideration of both monetary and nonmonetary inefficiencies will aid directors and applicants in eliminating them (eg, data presented here may allow applicants and directors to make better decisions early in the application process, minimize travel demands, and decrease time away from work). Thirdparty oversight, potentially involving both applicants and directors, of those variables outside the jurisdiction of the NRMP may be warranted, and our data will inform any future top-down decision making.
Michelotti, B. Taylor, B. Drolet, B. Ting, C. L. Dwyer, C. Meals, C. Lehrman, D. Cassidy, D. Meister, D. Perretta, E. Paryavi, E. Tsai, F. Gerold, G. Faucher, G. Malhorta, G. Giusti, I. Smithson, J. Solomon, J. Brubacher, J. Coppage, J. Kutsikovich, J. Miles, J. Foley, J. Pirolo, J. Zumsteg, J. Janowski, K. Carlberg, K. Lim, K. Faust, K. Vogeli, L. Al-Shihabi, L. Daniali, L. Robinson, M. Brown, C. Maugans, M. Arief, M. Vosbikian, M. Wigton, S. Moen, J. Morgan, M. Gust, N. Jarrett, N. Oak, O. Lamikanra, O. I. Franko, P. Letourneau, R. Wiltfong, R. Odgers, S. M. Fuller, J. J. Schreiber, S. Rogers, S. A. Glait, T. Theman, J. Munns, T. Gaddie, J. V. Unadkat, V. Cuellar, Z. Sarmast. REFERENCES 1. Gregory P, Ruffin R. Economics. New York: HarperCollins College Publishers; 1994. 2. Walsh K. Economic analysis in medical education: definition of essential terms. Med Teach. 2014;36(10):890e893. 3. Ranawat A, Nunley R, Genuario J, Sharan A, Mehta S, Washington Health Policy Fellows. Current state of the fellowship hiring process: are we in 1957 or 2007? AAOS Now. 2007;1(8). 4. Specialties Matching Service match participation agreement for all matches opening after June 1, 2014. 2014. Available at: www.nrmp. org/wp-content/uploads/2014/08/2015-MPA-SMS-Final-for-WWW. pdf. Accessed Sept 6, 2014. 5. National Resident Matching Program. NRMP results and data specialities matching service, 2014 Appointment Year. Available at: www. nrmp.org/wp-content/uploads/2013/08/National-Resident-MatchingProgram-NRMP-Results-and-Data-SMS-2014-Final.pdf. Accessed June 16, 2014. 6. Griffin SM, Stoneback JW. Navigating the orthopaedic trauma fellowship match from a candidate’s perspective. J Orthop Trauma. 2011;25(Suppl 3):S101eS103.
ACKNOWLEDGMENT The authors acknowledge the participation of the Hand Match 2015 Study Group: A. Micev, A. Kupperman, A. Ivy, A. Schannen, A. Argenta, A. Sull, B.
J Hand Surg Am.
7
r
Vol. -, - 2015
The Hand Surgery Fellowship Application Process: Expectations, Logistics, and Costs Clifton Meals, MD, Meredith Osterman, MD
Purpose To investigate expectations, logistics, and costs relevant to the hand surgery fellowship application process. We sought to discover (1) what both applicants and program directors are seeking, (2) what both parties have to offer, (3) how both parties collect information about each other, and (4) the costs incurred in arranging each match. Methods We conducted on-line surveys of hand surgery fellowship applicants for appointment in 2015 and of current fellowship program directors. Results Sixty-two applicants and 41 program directors completed the survey. Results revealed applicants’ demographic characteristics, qualifications, method of ranking hand fellowship programs, costs incurred (both monetary and opportunity) during the application process, ultimate match status, and suggestions for change. Results also revealed program directors’ program demographics, rationale for offering interviews and favorably ranking applicants, application-related logistical details, costs incurred (both monetary and opportunity) during the application process, and suggestions for change. Conclusions Applicants for hand surgery fellowship training are primarily interested in a potential program’s academic reputation, emphasis on orthopedic surgery, and location. The typical, successfully matched applicant was a 30-year-old male orthopedic resident with 3 publications to his credit. Applicants rely on peers and Web sites for information about fellowships. Fellowship directors are primarily seeking applicants recommended by other experienced surgeons and with positive personality traits. The typical fellowship director offers a single year of orthopedic-based fellowship training to 2 fellows per year and relies on a common application and in-person interviews to collect information about applicants. Applicants appear to be more concerned than directors about the current state of the match process. Applicants and directors alike incur heavy costs, in both dollars and opportunity, to arrange each match. A nuanced understanding of the match process suggests specific changes and may help reduce these costs. (J Hand Surg Am. 2015;(-):-e-. Copyright Ó 2015 by the American Society for Surgery of the Hand. All rights reserved.) Type of study/level of evidence Economic and decision analysis V. Key words Application, fellowship, hand, match, surgery.
From the Department of Orthopedics, George Washington University School of Medicine and Health Sciences, Washington, DC; and the Philadelphia Hand Center, Philadelphia, PA. Received for publication September 15, 2014; accepted in revised form December 23, 2014. No benefits in any form have been received or will be received related directly or indirectly to the subject of this article.
Corresponding author: Clifton Meals, MD, Department of Orthopedics, George Washington University School of Medicine and Health Sciences, 2150 Pennsylvania Ave., NW, Washington, DC 20037; e-mail: [email protected]. 0363-5023/15/---0001$36.00/0 http://dx.doi.org/10.1016/j.jhsa.2014.12.041
Ó 2015 ASSH
r
Published by Elsevier, Inc. All rights reserved.
r
1
2
T
HAND SURGERY FELLOWSHIP APPLICATION PROCESS
match may be understood as an economic market in which applicants offer skill, experience, enthusiasm, and other factors in exchange for advanced training. The National Residency Matching Program (NRMP) facilitates matches and publishes basic information related to each year’s application process. However, precisely what applicants and directors want, what both parties have to offer, and how both parties collect information remain obscure. Two types of costs, monetary and opportunity, are, anecdotally, a source of frustration for applicants and program directors; however, these costs are poorly understood. The monetary costs of the match have not been enumerated. Opportunity cost is the cost of forgoing a next-best option (eg, the cost of putting off clinic) to interview applicants.1,2 These costs are less concrete but no less important and no less obscure. Monetary and opportunity costs might be reduced if applicants and program directors knew how their energies were best directed. If applicants knew how their credentials compared with those of their peers, they might selfselect (ie, chose programs more likely to accept them) early in the application process. If directors knew the interviewing behavior of other programs, they might optimize the number of interviews they offer and/or coordinate interviews with other sites. Data to inform such decision making are lacking. We therefore sought to collect information related to match expectations, logistics, and costs through surveys of applicants for hand fellowship appointment in 2015 as well as current hand surgery fellowship directors.
Between January and May 2014, via in-person and e-mailed inquiries, we collected the e-mail addresses of 118 applicants for 2015 appointment. We invited them to take an on-line survey in June, sent 2 reminder invitations at weekly intervals, and closed the survey after a month. To incentivize participation, we informed potential respondents that their participation in the survey would be acknowledged but that their individual responses would remain anonymous. We identified 84 program directors and codirectors, corresponding to the 81 Accreditation Council for Graduate Medical Education accredited hand surgery fellowships. We invited them to take the on-line survey in August 2014, sent 2 reminder invitations at weekly intervals, and closed the survey after a month. We informed program director respondents that their participation in the survey would be anonymous unless they asked to be acknowledged. Both surveys were facilitated by a commercial Web site (www.surveymonkey.com; SurveyMonkey Inc., Palo Alto, CA). Data analysis involved traditional statistical methods and was aided by the same commercial Web site that facilitated the surveys.
HE HAND SURGERY FELLOWSHIP
RESULTS Applicant survey Of the 118 applicants for 2015 appointment we contacted, 61 (52%) completed the survey, and 3 partially completed the survey. Eighty-one percent of respondents were men. The median age among respondents was 30 years. All respondents were MDs, and 14% had other advanced degrees including 1 PhD. Thirty-five percent of respondents were members of the Alpha Omega Alpha medical honor society, 51% were not, and 14% indicated that their medical schools did not have Alpha Omega Alpha medical honor society chapters. Seventy-eight percent had career or research experience other than publications, presentations, awards, or grants. Eighty percent were enrolled in or had completed orthopedic surgery residency training, and 20% were enrolled in or had completed plastic surgery residency training. One respondent had already completed another fellowship (microsurgery and pediatric plastic surgery). Five percent of respondents had previously applied for hand surgery fellowship training. Respondents’ median number of publications was 3 and their median number of national podium or poster presentations was 2. All but 2 respondents had completed a dedicated hand surgery residency rotation before applying for
MATERIALS AND METHODS We conducted on-line surveys of both hand surgery fellowship applicants for appointment in 2015 and current hand surgery fellowship directors. The full surveys are presented in Appendix A (available on the Journal’s Web site at www.jhandsurg.org). The applicant survey consisted of 37 questions designed to reveal each respondent’s demographics, qualifications, method of ranking hand fellowship programs, expenses and opportunity costs incurred traveling to interviews, ultimate match status, and suggestions for change. The fellowship director survey consisted of 15 questions designed to reveal each respondent’s program demographics; rationale for offering interviews and favorably ranking applicants; application-related logistic details; costs incurred, both monetary and opportunity, during the application process; and suggestions for change. J Hand Surg Am.
r
Vol. -, - 2015
HAND SURGERY FELLOWSHIP APPLICATION PROCESS
Of 61 respondents who answered the question, 100% matched. The greatest share (36%) matched with their first choice. Fifteen percent matched with their second choice. In response to the question “If you did not match, what are you plans for the future?” 1 respondent planned to apply for a different subspecialty fellowship and another planned to apply for 2016 hand surgery fellowship appointment. The final survey question was “How might the match process be improved?” and a free-text response was required. These responses were placed into categories and are represented in Table 1. The raw data obtained from the applicant survey are presented in Appendix A (available on the Journal’s Web site at www.jhandsurg.org).
FIGURE 1: How applicants valued different sources of information in selecting fellowship programs (0, not at all; 1, somewhat; 2, very much).
fellowship training. How applicants valued different sources of information in selecting fellowship programs is represented in Figure 1. How respondents ranked various factors in their decision to pursue a specific fellowship program is represented in Figure 2. Applicants for 2015 appointment applied to a median of 30 fellowships and indicated that, on average, 5 programs asked for material in addition to the Universal Hand Application. Respondents were invited to a median of 20 interviews. Of these, the largest share accepted 12 interview invitations. Seventy-four percent of respondents accepted and then cancelled at least 1 interview. The most common reason cited for cancelling an interview was a conflicting interview. Less common reasons included expense and lack of interest in equal proportion. Sixty-nine percent of respondents had their first interview in January and their last interview in March 2014. Sixty-four percent of respondents traveled more than 2,400 km (1,500 miles) to their most distant interview. The median days spent away from home while interviewing was 20. Thirtyfour percent of respondents spent unplanned and extra days owing to missed or cancelled travel arrangements, with inclement weather being an important cause. The largest share of respondents indicated that they spent between $2,500 and $5,000 on interview-related travel expenses. The majority of respondents paid for their expenses principally with savings, and 36% borrowed money on credit cards and/ or were assisted by family. Seventy-four percent of respondents reported that a faculty member at their home institution made personal contact with 1 or more hand surgery fellowship directors on the applicant’s behalf. Seventy-nine percent of respondents personally contacted at least 1 fellowship director to express particular interest in a program. When asked if visiting fellowship institutions helped respondents to order their rank list, the responses were 48% very much, 33% somewhat, and 20% not at all. Respondents ranked a median of 10 programs. J Hand Surg Am.
3
Program director survey Of the 85 potential respondents, 41 (48%) completed the survey, and 7 partially completed the survey. The proportions of directors offering orthopedic, combined, or plastic programs were 69%, 25%, and 6%, respectively. Thirty-nine (80%) respondents were the sole director of their program and 9 were codirectors. The proportions of directors offering 1, 2, 3, 4, or more fellowship positions are represented in Table 2. Four directors offered optional 2-year fellowship positions. All other available positions were for 1 year. How directors valued different applicant qualifications in deciding to grant an interview is represented in Figure 3. How directors valued different applicant attributes in ordering a rank list is represented in Figure 4. The distribution of directors’ received number of applications is represented in Table 3. The distributions of directors’ number of offered interviews, conducted interviews, and number of applicants ranked are represented in Table 4. The majority (88%) of respondents did not contact applicants after the interview process to express interest, 4 directors contacted only select applicants, and 1 contacted all interviewees. Respondents were asked to estimate the opportunity cost incurred by their department during the interview process with the following formula: Cost ðin hoursÞ ¼ ½# involved faculty ½average hours devoted by each faculty member
The average opportunity cost reported was 65 hours. The average monetary cost incurred by each respondent’s department during the interview process was $4,572. r
Vol. -, - 2015
4
HAND SURGERY FELLOWSHIP APPLICATION PROCESS
FIGURE 2: How applicants valued different factors in their decision to pursue specific fellowship programs (0, not at all; 1, somewhat; 2, very much). Acad., academic; avail., availability; non-acad., non-academic; ortho., orthopedic; surg., surgery.
TABLE 1.
Suggestions for Change
Category
Applicant Survey (61 Respondents, 104 Suggestions)
Director Survey (41 Respondents, 44 Suggestions)
Proportion of Suggestions
Proportion of Suggestions
25%
16%
Centralized or coordinated interviews, or both Universal or electronic application, or both
21%
9%
Change in timing of deadlines, interviews, or match, or all
17%
2%
Other
13%
2%
Higher quality or better availability of program information, or both
12%
5%
Limits on the number of interviews offered or accepted, or both
8%
NA
No suggestion
4%
45%
Reduce actions not permitted by the NRMP
NA
14%
Alternatives to in-person interviews
NA
7%
TABLE 2.
The raw data obtained from the director survey are presented in Appendix A (available on the Journal’s Web site at www.jhandsurg.org).
Proportions of Program Sizes Proportion of Directors Offering Different Numbers of Fellowship Positions (48 Respondents)
1
33%
2
46%
3
13%
4
6%
>4
2%
DISCUSSION The hand surgery fellowship application process was initially informal. Applicants contacted fellowship programs by phone or mail and provided their credentials before being selected to interview. Positions were offered and accepted variably by a centralized service, agreement between fellowship directors, and/ or by arrangement between fellowship directors and applicants (A. L. Osterman and R. Meals, personal communications, February, 2015). As interest in hand surgery has expanded, the fellowship matching process has increased in scale and sophistication. Since appointment year 1993, matches have been facilitated by a third-party matching service. The centralized hand surgery fellowship match is a model of success. Several orthopedic subspecialties do not offer a third-party matching program. Others
Sixty-three percent of fellowship directors believed it was important for an applicant to visit their institution in order to effectively rank the applicant. The remainder of directors did not think this was important. The final survey question was “How might the match process be improved?” and a free-text response was required. These responses were placed into categories and are represented in Table 1. J Hand Surg Am.
r
Vol. -, - 2015
5
HAND SURGERY FELLOWSHIP APPLICATION PROCESS
TABLE 3. Proportions of Number of Received Applications Proportion of Directors’ Received Number of Applications (42 Respondents)
FIGURE 3: How directors valued different applicant qualifications in deciding to grant an interview (0, not at all; 1, somewhat; 2, very much). AOA, American Orthopedic Association; educ., education; ortho, orthopedic surgery; plastics., plastic surgery; pregrad, pregraduate; reput., reputation; USMLE, United States Medical Licensing Examination.
14% 22%
51e75
26%
76e100
21%
101e125
12%
126e150
5%
ask applicant survey respondents, whose participation would be acknowledged in aggregate, whether they had been asked to arrange for a match outside of the NRMP, because we assumed responses to this question would be unreliable. Anecdotally, neither of us encountered any obvious attempts to subvert the match process. Although we were not specifically asked to provide postinterview assurances, we recall that applicants on interview days were sometimes asked to express interest in a program should applicants wish to match there. This may skirt the NRMP mandate that “program directors shall not solicit or require post-interview communication from applicants.”4 Solicited or not, we discovered that thank you letters were sent from interviewees to directors and vice versa. This heeds the NRMP’s directive that “both applicants and programs may express their interest in each other; however, they shall not solicit verbal or written statements implying a commitment.”4 Only one applicant respondent referred to illicit deal-making. Although 17% of director respondents suggested that directors sometimes cheat, it seemed some of these respondents viewed expressing postinterview interest as prohibited, when in fact the NRMP permits this. The NRMP governs only a few of the application process’s many variables, and whereas the match may be fair, the process likely places undue burden on both program directors and applicants. Applicants, in particular, may bear a disproportionate share of this burden as recent trends suggest a sellers’—that is, directors’—market. The number of hand fellowship applicants increased from 155 for appointment in 2010 to 205 for appointment in 2014. The number of available fellowship positions has increased slightly, although applicants have outnumbered available spots. Beginning with the 2011 appointment year, roughly one-quarter of applicants for hand fellowship training have not matched.5
FIGURE 4: How directors valued different applicant attributes in ordering a rank list (0, not at all, 1; somewhat; 2, very much). Acad., academic; geo., geography; med., medicine.
have seen their centralized matching scheme corrupted by arrangement of appointments outside the match, ignored by a portion of eligible programs, or abandoned.3 The match sets up an economic “Prisoner’s Dilemma.” The Prisoner’s Dilemma is illustrated by a scenario in which 2 accomplices have been arrested. Their fate varies according to whether they both keep mum, one betrays the other, or if they inform on one another. They will benefit from cooperation but both are motivated to cheat and are likely to do so.1 As such, both applicants and program directors benefit by universal adherence to the NRMP but are incentivized to cheat (ie, arrange matches outside of the centralized matching service). It appears to us, however, that little cheating actually occurs. The hand surgery fellowship match has met the NRMP’s requirements for fairness since hand surgery began using this service in 1992. American hand surgery fellowship programs have universally adhered to the NRMP service for the same period. We declined to J Hand Surg Am.
1e25 26e50
r
Vol. -, - 2015
6
HAND SURGERY FELLOWSHIP APPLICATION PROCESS
TABLE 4.
1e10
Proportions of Number of Applicants Offered Interviews, Interviewed, and Ranked Proportion of Directors’ Number of Offered Interviews (42 Respondents)
Proportion of Directors’ Number of Conducted Interviews (42 Respondents)
Proportion of Directors’ Number of Applicants Ranked (42 Respondents)
12%
17%
24%
11e20
22%
29%
36%
21e30
41%
45%
31%
31e40
21%
5%
5%
41e50
2%
2%
2%
51e60
2%
2%
2%
Additional evidence of a sellers’ market includes our finding that, in response to a prompt to suggest change to the current system, 45% of director replies could be categorized as “no change,” whereas only 4% of applicant responses to the same prompt could be similarly described. In this setting of increased competition and asymmetrical concern regarding the application process, no third party regulates what fellowship programs may ask of applicants, how many interviews a program may offer, how many interviews an applicant may accept, or when and where interviews will take place, and so on (P. Stern, A. L. Osterman, P. Benhaim, and R. Meals, personal communications, October, 2014). It has been unclear what makes an applicant competitive or programs desirable. As such, both applicants and program directors have had little to inform the best (ie, efficient) use of resources. We make several suggestions based on our investigation. There is likely a role for increased third-party oversight of those variables outside the jurisdiction of the NRMP. Hand surgery fellowship directors meet at the annual American Society for Surgery of the Hand meeting. This group’s responsibilities, or the responsibilities of a new, independent organization, potentially including both applicants and directors, might be expanded to include:
could take place at national meetings or regional sites, as occurs with other subspecialties.6 Video or phone-based interviews deserve consideration. 6. Encouraging fellowship programs to offer comprehensive, up-to-date, on-line information. 7. Moving the application process to either the fall or the spring to facilitate safe and efficient travel. Earlier interviews might be favored by applicants eager to determine their career plans. Later interviews might allow applicants to make more informed decisions, complete a dedicated hand surgery rotation, and/or better develop relationships with mentors ultimately influencing the match. Our investigation was based on 1 anonymous survey and 1 in which the respondents knew their identities would be published only in aggregate and is, therefore, not entirely reliable. For instance, we are both recent hand surgery fellowship applicants and recall receiving approximately 10 thank you letters and other “expressions of interest” from fellowship programs. This is anecdotal evidence at odds with the finding that only 5 program directors reported contacting applicants after their interview. A second weakness is related to the proportion of potential respondents who completed the survey: roughly onethird of the predicted number of applicants for 2015 appointment and roughly one-half of current fellowship program directors. In addition, selection bias likely resulted in a disproportionate number of successfully matched applicants responding to the survey. Nonetheless, our data reveal several trends. Applicants for hand surgery fellowship training are primarily interested in a potential program’s academic reputation, emphasis on orthopedic surgery, and location. The typical, successfully matched applicant is a 30year-old male orthopedic resident with 3 academic publications to his credit. Applicants rely on peers and Web sites for information about fellowships. Fellowship directors are primarily seeking applicants
1. Limiting the number of interviews offered by fellowship programs and/or accepted by applicants. 2. Encouraging disclosure from program directors regarding the qualifications of their fellows and of their ideal applicants. 3. Instituting an electronic centralized application system, similar to the Electronic Residency Application Service. 4. Coordinating interview dates between programs in the same region. 5. Establishing alternatives to traditional in-person and/ or on-site interviews. Final or screening interviews J Hand Surg Am.
r
Vol. -, - 2015
HAND SURGERY FELLOWSHIP APPLICATION PROCESS
recommended by other experienced surgeons and with positive personality traits as listed in the director survey: maturity, candor, poise, and confidence. The typical fellowship director offers a single year of orthopedic-based fellowship training to 2 fellows per year and relies on a common application and in-person interviews to collect information about applicants. Applicants, in general, appear disproportionately concerned about the match process compared with program directors. Perhaps most importantly, the high costs of the current process, in both money and opportunity are apparent. Monetary cost is familiar. Opportunity expenses (eg, what a resident loses in missing 20 days of training) deserve special attention because they may be important but difficult to quantify. Careful consideration of both monetary and nonmonetary inefficiencies will aid directors and applicants in eliminating them (eg, data presented here may allow applicants and directors to make better decisions early in the application process, minimize travel demands, and decrease time away from work). Thirdparty oversight, potentially involving both applicants and directors, of those variables outside the jurisdiction of the NRMP may be warranted, and our data will inform any future top-down decision making.
Michelotti, B. Taylor, B. Drolet, B. Ting, C. L. Dwyer, C. Meals, C. Lehrman, D. Cassidy, D. Meister, D. Perretta, E. Paryavi, E. Tsai, F. Gerold, G. Faucher, G. Malhorta, G. Giusti, I. Smithson, J. Solomon, J. Brubacher, J. Coppage, J. Kutsikovich, J. Miles, J. Foley, J. Pirolo, J. Zumsteg, J. Janowski, K. Carlberg, K. Lim, K. Faust, K. Vogeli, L. Al-Shihabi, L. Daniali, L. Robinson, M. Brown, C. Maugans, M. Arief, M. Vosbikian, M. Wigton, S. Moen, J. Morgan, M. Gust, N. Jarrett, N. Oak, O. Lamikanra, O. I. Franko, P. Letourneau, R. Wiltfong, R. Odgers, S. M. Fuller, J. J. Schreiber, S. Rogers, S. A. Glait, T. Theman, J. Munns, T. Gaddie, J. V. Unadkat, V. Cuellar, Z. Sarmast. REFERENCES 1. Gregory P, Ruffin R. Economics. New York: HarperCollins College Publishers; 1994. 2. Walsh K. Economic analysis in medical education: definition of essential terms. Med Teach. 2014;36(10):890e893. 3. Ranawat A, Nunley R, Genuario J, Sharan A, Mehta S, Washington Health Policy Fellows. Current state of the fellowship hiring process: are we in 1957 or 2007? AAOS Now. 2007;1(8). 4. Specialties Matching Service match participation agreement for all matches opening after June 1, 2014. 2014. Available at: www.nrmp. org/wp-content/uploads/2014/08/2015-MPA-SMS-Final-for-WWW. pdf. Accessed Sept 6, 2014. 5. National Resident Matching Program. NRMP results and data specialities matching service, 2014 Appointment Year. Available at: www. nrmp.org/wp-content/uploads/2013/08/National-Resident-MatchingProgram-NRMP-Results-and-Data-SMS-2014-Final.pdf. Accessed June 16, 2014. 6. Griffin SM, Stoneback JW. Navigating the orthopaedic trauma fellowship match from a candidate’s perspective. J Orthop Trauma. 2011;25(Suppl 3):S101eS103.
ACKNOWLEDGMENT The authors acknowledge the participation of the Hand Match 2015 Study Group: A. Micev, A. Kupperman, A. Ivy, A. Schannen, A. Argenta, A. Sull, B.
J Hand Surg Am.
7
r
Vol. -, - 2015
