Objective
We sought to determine the attributes of successful and unsuccessful fellowship applicants of the American Board of Obstetrics and Gynecology Inc (ABOG)-approved fellowship programs and to identify salient differences between subspecialty applicants.
Study Design
Anonymous questionnaires were completed by obstetrics and gynecology fellowship applicants using a web-based survey after match day of 2012. Fellowship applicant practices were evaluated and included importance of prematch preparations, interview process, networking practices, and postmatch reflections.
Results
A total of 327 fellowship applicants applying to programs accredited by the ABOG were surveyed, and 200 completed the survey (61% response rate). A comparison between prematch educational preparations pursued by applicants showed that matched applicants were more likely to come from allopathic medical schools (94%), attain membership in Alpha Omega Alpha and/or Phi Beta Kappa (27%), and receive a letter of recommendation from a nationally known subspecialist (77%) than unmatched applicants ( P = .03, .005, and .007, respectively). Applicants to reproductive endocrinology and infertility were more likely than female pelvic medicine and reconstructive surgery to be members of academic honor societies ( P = .008). Research publication was common among matched subspecialist applicants, with over half publishing 1-3 peer-reviewed manuscripts prior to matching. Applicants to gynecologic oncology did more visiting electives than any other specialty applicants ( P < .001).
Conclusion
Successful obstetrics and gynecology fellowship applicants have superior prematch preparations, strong letters of recommendation from leaders in their field of interest, and multiple research publications. These data will guide applicants to a critical self-analysis before deciding to apply.
See related editorial, page 500
The obstetrics and gynecology subspecialties of female pelvic medicine and reconstructive surgery (FPMRS), gynecologic oncology, maternal-fetal medicine (MFM), and reproductive endocrinology and infertility (REI) proved competitive for the 2013 appointment year, with 94% of programs filling offered positions, and 31% of applicants going unmatched. A total of 196 active obstetrics and gynecology fellowship programs had 240 positions available. Of a total of 327 active applicants, 226 filled these positions, leaving a total of 14 positions unfilled. By subspecialty, all 46 available positions in gynecologic oncology filled, with 46 (61%) of 75 applicants matching. Eighty-nine of 98 positions filled in MFM, with 88 (72%) of 123 applicants matching. Regarding REI, 43 of 45 positions filled, with 43 (67%) of 64 applicants matching. Lastly, 48 of 51 positions filled for FPMRS, with 48 (74%) of 65 applicants matching. The competitive nature of the obstetrics and gynecology subspecialty match over the last few years compels us to consider the attributes that would qualify a particular applicant to obtain an interview and, ultimately, match into these programs. While previous studies on fellowships for subspecialties such as pediatric general surgery have recognized the considerable preparation required for applicants, the specific prerequisites for obtaining an interview in a competitive subspecialty are unknown. Some progress has been made in the field of obstetrics and gynecology with regard to desirable attributes of applicants; however, the study was undertaken from the perspective of fellowship directors, not the applicants themselves. In addition, with the exception of some online resources available to prospective applicants, information on the fellowship match remains scant, and several areas remain unexplored, for instance, applicant demographics as well as the number of publications and interviews needed for a successful match.
The purpose of our investigation was to determine the attributes of successful applicants to the American Board of Obstetrics and Gynecology Inc (ABOG)-approved fellowship programs and to identify differences between subspecialty applicants.
Materials and Methods
The residents applying to all subspecialties available at the authors’ institutions were contacted. We also queried social media sites like Facebook to contact a group of fellowship applicants that discuss interview plans and programs on these listservs. These listservs were used to further encourage participation since the survey invitation was coming from known colleagues. We were unable to get a list of applicant contact numbers or e-mail addresses from the National Resident Matching Program (NRMP) due to privacy policies. Using a web-based survey tool ( www.surveymonkey.com ), we sent a questionnaire to participants. Anonymous surveys were sent to fellowship applicants to programs in FPMRS, gynecologic oncology, MFM, and REI. Applicants with urology residency training applying to FPMRS were excluded. The survey was sent the day after the match was completed. A brief statement describing the intended use of data informed respondents that their participation constituted their voluntary consent for the study.
A 4-part (educational data, prematch preparation, interview process, and role of the match) electronic survey was developed. The category for educational data included items related to the quality of medical education and research experience of the applicants. Prematch preparation focused on previous publications and networking at national meetings. The last 2 categories, interview process and role of the match, evaluated the experience of meeting with fellowship programs. The survey included a section for respondents to make free-text comments.
Identical 38-item questionnaires were e-mailed to fellowship applicants following their match day. In an attempt to increase response rate, 3 more personalized e-mails were sent over the next 3 weeks. A $5 gift card to an Internet retailer was provided as an inducement for completion of the survey. A final e-mail was sent to all nonresponders with a second $5 gift card.
To compare the rankings of fellowship characteristics between subspecialty applicants, a series of Kruskal-Wallis tests were performed for each of the characteristics. Post hoc pairwise comparisons were performed to identify the specific differences between specialties. To control for multiple comparisons, we used Bonferroni adjustment to the type I error rate. To further elucidate whether or not any of the key variables were sufficient predictors of whether or not residents matched to a fellowship program, we conducted a multiple logistic regression analysis to control for multiple factors predicting match. We used a backward procedure for removing predictors with P values > .10 to reduce the model to the simplest model that included only statistically significant predictors of match. Responses were analyzed using statistical software (SPSS Statistics 20.0; IBM Corp, Armonk, NY). Some respondents chose not to answer all questions, accounting for the varying denominators. An exemption from St. Luke’s Hospital of Kansas City Institutional Review Board approval was obtained for this study.
Results
We received 200 completed surveys from a total of 327 fellowship applicants in FPMRS, gynecologic oncology, MFM, and REI who were invited to participate, with an overall response rate of 61% (200/327). Respondents were mostly female (71%), Caucasian (72%), and were most likely to have applied for MFM fellowship (30%). The response rate was similar among the 4 subspecialties ( Tables 1 and 2 ).
| Characteristic | Overall | FPMRS | GO | MFM | REI | P value |
|---|---|---|---|---|---|---|
| Overall | 200 | 41 (21%) | 58 (29%) | 60 (30%) | 41 (21%) | |
| Age, y (n = 198) | ||||||
| <30 | 72 (36) | 17 (42) | 23 (40) | 15 (25) | 17 (44) | – |
| 30-32 | 77 (39) | 10 (24) | 24 (41) | 28 (47) | 15 (39) | |
| 33-35 | 29 (15) | 8 (20) | 11 (19) | 5 (8) | 5 (13) | |
| 36-38 | 15 (8) | 3 (7) | 0 (0) | 11 (18) | 1 (3) | |
| ≥39 | 5 (3) | 3 (7) | 0 (0) | 1 (2) | 1 (3) | |
| Sex (n = 196) | ||||||
| Male | 57 (29) | 12 (29) | 19 (33) | 13 (22) | 13 (33) | .52 |
| Female | 139 (71) | 29 (71) | 38 (67) | 46 (78) | 26 (67) | |
| Race (n = 190) | ||||||
| White | 137 (72) | 25 (63) | 42 (75) | 39 (71) | 31 (80) | – |
| Black | 12 (6) | 3 (8) | 3 (5) | 4 (7) | 2 (5) | |
| Asian | 29 (15) | 9 (23) | 8 (14) | 8 (15) | 4 (10) | |
| Multiple/other | 12 (6) | 3 (8) | 3 (5) | 4 (7) | 2 (5) | |
| Type of medical school (n = 194) | ||||||
| Allopathic medical school | 177 (91) | 34 (83) | 56 (98) | 51 (90) | 36 (92) | – |
| Osteopathic medical school | 17 (9) | 7 (17) | 1 (2) | 6 (11) | 3 (8) | |
| Did you match? (n = 198) | ||||||
| Yes | 148 (75) | 29 (71) | 41 (71) | 48 (80) | 30 (77) | .61 |
| No | 50 (25) | 12 (29) | 17 (29) | 12 (20) | 9 (23) | |
| Rank at matched program (n = 126) | ||||||
| First | 66 (52) | 11 (42) | 17 (49) | 24 (62) | 14 (54) | – |
| Second | 21 (17) | 9 (35) | 6 (17) | 2 (5) | 4 (15) | |
| Third | 13 (10) | 1 (4) | 4 (11) | 5 (13) | 3 (12) | |
| Fourth | 13 (10) | 3 (12) | 3 (9) | 5 (13) | 2 (8) | |
| ≥Fifth | 13 (10) | 2 (8) | 5 (14) | 3 (8) | 3 (12) | |
| Member of Alpha Omega Alpha or Phi Beta Kappa (n = 183) | ||||||
| Yes | 40 (22) | 4 (10) | 14 (25) | 9 (17) | 13 (36) | .04 |
| No | 143 (78) | 35 (90) | 42 (75) | 43 (83) | 23 (64) | |
| CREOG score >230 (n = 184) | ||||||
| Yes | 57 (31) | 10 (26) | 21 (38) | 15 (29) | 11 (31) | .74 |
| No | 91 (50) | 20 (51) | 26 (46) | 29 (56) | 16 (44) | |
| Do not know | 35 (19) | 9 (23) | 9 (16) | 8 (15) | 9 (25) | |
| Did visiting elective prior to match (n = 181) | ||||||
| Yes | 59 (33) | 8 (21) | 33 (60) | 9 (17) | 9 (26) | < .001 |
| No | 122 (67) | 31 (80) | 22 (40) | 43 (83) | 26 (74) | |
| Letter of recommendation from nationally known subspecialist (n = 181) | ||||||
| Yes | 130 (72) | 28 (74) | 46 (82) | 33 (64) | 23 (66) | .14 |
| No | 51 (28) | 10 (26) | 10 (18) | 19 (37) | 12 (34) |
| Characteristic | Overall | Matched | Unmatched | P value |
|---|---|---|---|---|
| Overall | 198 | 148 (75%) | 50 (25%) | |
| Age, y (n = 198) | ||||
| <30 | 72 (36) | 53 (36) | 19 (38) | – |
| 30-32 | 77 (39) | 61 (41) | 16 (32) | |
| 33-35 | 29 (15) | 21 (14) | 8 (16) | |
| 36-38 | 15 (8) | 11 (7) | 4 (8) | |
| ≥39 | 5 (3) | 2 (1) | 3 (6) | |
| Sex (n = 196) | ||||
| Female | 139 (71) | 103 (71) | 36 (72) | .85 |
| Race (n = 190) | ||||
| White | 137 (72) | 107 (76) | 30 (61) | – |
| Black | 12 (6) | 8 (6) | 4 (8) | |
| Asian | 29 (15) | 17 (12) | 12 (25) | |
| Multiple/other | 12 (6) | 9 (6) | 3 (6) | |
| Type of medical school (n = 194) | ||||
| Allopathic medical school | 177 (91) | 136 (94) | 41 (84) | .03 |
| Osteopathic medical school | 17 (9) | 9 (6) | 8 (16) | |
| Rank at matched program (n = 126) | ||||
| First | 66 (52) | |||
| Second | 21 (17) | |||
| Third | 13 (10) | |||
| Fourth | 13 (10) | |||
| ≥Fifth | 13 (10) | |||
| Member of Alpha Omega Alpha or Phi Beta Kappa (n = 183) | ||||
| Yes | 40 (22) | 36 (27) | 4 (8) | .005 |
| No | 143 (78) | 97 (73) | 46 (92) | |
| CREOG score >230 (n = 183) | ||||
| Yes | 57 (31) | 46 (35) | 11 (22) | .06 |
| No | 91 (50) | 59 (44) | 32 (64) | |
| Do not know | 35 (19) | 28 (21) | 7 (14) | |
| Did visiting elective prior to match (n = 181) | ||||
| Yes | 59 (33) | 41 (31) | 18 (37) | .47 |
| Letter of recommendation from nationally known subspecialist (n = 181) | ||||
| Yes | 130 (72) | 102 (77) | 28 (57) | .007 |
Prematch preparations
A comparison between prematch educational preparations pursued by applicants showed that 77% of matched applicants had a letter of recommendation from a “nationally known” member of their subspecialty, while only 57% among unmatched applicants did ( P < .01). Matched applicants (94%) were more likely to come from allopathic medical schools, attain membership in Alpha Omega Alpha and/or Phi Beta Kappa (27%), and receive a letter of recommendation from a nationally known subspecialist (77%) than unmatched applicants ( P = .03, .005, and .007, respectively). Comparing differences between applicants according to subspecialty, we found that membership in alpha omega alpha or phi beta kappa was significantly higher among applicants to REI than it was for applicants to FPMRS (36% vs 10%, P = .008). All other differences for these academic honor society memberships between subspecialties were not statistically significant ( P > .008 for each). Another measure of academic achievement is in service training scores. Almost a third (31%) of all applicants had Council on Resident Education in Obstetrics and Gynecology (CREOG) in service training examination scores >230. Lastly, research publication was common among matched subspecialist applicants, with over half publishing 1-3 peer-reviewed manuscripts prior to matching ( Table 3 ). Eighteen applicants (18/182, 10%) took time off to complete basic or clinical research.
| Variable | Overall | FPMRS | GO | MFM | REI |
|---|---|---|---|---|---|
| Poster presentations (n = 131) | |||||
| None | 20 (15) | 1 (4) | 9 (23) | 8 (21) | 2 (7) |
| 1-3 | 70 (53) | 15 (58) | 16 (41) | 21 (54) | 18 (67) |
| 4-6 | 27 (21) | 6 (23) | 10 (26) | 7 (18) | 4 (15) |
| ≥7 | 14 (11) | 4 (15) | 4 (10) | 3 (8) | 3 (11) |
| Published manuscripts (n = 130) | |||||
| None | 23 (18) | 5 (19) | 7 (18) | 8 (20) | 3 (12) |
| 1-3 | 80 (62) | 16 (62) | 25 (66) | 26 (65) | 13 (50) |
| 4-6 | 16 (12) | 2 (8) | 4 (11) | 4 (10) | 6 (23) |
| ≥7 | 4 (15) | 3 (12) | 2 (5) | 2 (5) | 4 (15) |
| Oral presentations (n = 130) | |||||
| None | 66 (51) | 13 (50) | 13 (34) | 24 (62) | 16 (59) |
| 1-3 | 52 (40) | 10 (39) | 21 (55) | 12 (31) | 9 (33) |
| 4-6 | 7 (5) | 0 (0) | 4 (11) | 2 (5) | 1 (4) |
| ≥7 | 5 (4) | 3 (12) | 0 (0) | 1 (3) | 1 (4) |
Prematch elective
Fifty-nine applicants (59/181, 33%) completed a visiting elective in their subspecialty of choice prior to the match; there was no difference between applicants who matched (31%) and those who did not match (37%) in terms of likelihood of completing a visiting elective prior to the match ( P = .47). Of applicants to gynecologic oncology fellowships, 60% had completed a visiting elective prior to the match, a greater percentage as compared to FPMRS applicants (21%, P < .001), MFM applicants (17%, P < .001), and REI applicants (25%, P = .001). There were no significant differences among FPMRS, MFM, and REI applicants with regard to an elective ( P > .008 for each).
Fellowship interview and ranking
The majority of all applicants applied to >25 programs and then ranked >10 programs ( Table 4 ). Over half of the candidates (66/126, 52%) were appointed to the first fellowship program on their rank order list while 79% (100/126, 79%) were matched to 1 of their top-3 rank list choices. Of the 129 respondents who provided their residency program and their matched fellowship program, 14 (11%) stayed at their home institution. There was a statistically significant greater percentage of applicants who matched if they applied to a subspecialty that was present at their home institution ( P < .0003).
| Variable | Overall | FPMRS | GO | MFM | REI | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| Matched | Unmatched | Matched | Unmatched | Matched | Unmatched | Matched | Unmatched | Matched | Unmatched | |
| Fellowships applied (n = 180) | ||||||||||
| <5 | 2 (2) | 1 (2) | 1 (4) | 0 (0) | 0 (0) | 0 (0) | 1 (3) | 0 (0) | 0 (0) | 1 (13) |
| 6-15 | 28 (21) | 5 (10) | 8 (30) | 2 (17) | 3 (8) | 1 (6) | 12 (30) | 2 (17) | 5 (19) | 0 (0) |
| 16-25 | 33 (25) | 10 (21) | 8 (30) | 1 (8) | 6 (15) | 3 (19) | 11 (28) | 4 (33) | 8 (31) | 2 (25) |
| ≥26 | 69 (52) | 32 (67) | 10 (37) | 9 (75) | 30 (77) | 12 (75) | 16 (40) | 6 (50) | 13 (50) | 5 (63) |
| Invitations offered (n = 179) | ||||||||||
| <5 | 7 (53) | 22 (46) | 1 (4) | 6 (50) | 2 (5) | 7 (44) | 3 (8) | 4 (33) | 1 (4) | 5 (63) |
| 6-15 | 66 (52) | 20 (42) | 13(50) | 4 (33) | 14 (37) | 7 (44) | 25 (63) | 6 (50) | 14 (52) | 3 (38) |
| 16-25 | 44 (34) | 6 (13) | 8 (27) | 2 (17) | 17 (45) | 2 (13) | 11 (28) | 2 (17) | 9 (33) | 0 (0) |
| ≥26 | 14 (11) | 0 (0) | 5 (19) | 0 (0) | 5 (13) | 0 (0) | 1 (3) | 0 (0) | 3 (11) | 0 (0) |
| Interviews attended (n = 181) | ||||||||||
| <5 | 11 (8) | 25 (52) | 1 (4) | 6 (50) | 4 (10) | 8 (50) | 5 (13) | 5 (42) | 1 (4) | 6 (75) |
| 6-15 | 95 (71) | 21 (44) | 20 (74) | 6 (50) | 25 (64) | 7 (44) | 32 (80) | 6 (50) | 18 (67) | 2 (25) |
| 16-25 | 27 (20) | 2 (4) | 6 (22) | 0 (0) | 10 (26) | 1 (6) | 3 (8) | 1 (8) | 8 (30) | 0 (0) |
| ≥26 | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
| No. of programs ranked on NRMP list (n = 179) | ||||||||||
| 0-1 | 5 (4) | 12 (25) | 1 (4) | 1 (8) | 2 (5) | 3 (18) | 2 (5) | 5 (46) | 0 (0) | 3 (38) |
| 2-4 | 9 (7) | 11 (23) | 3 (11) | 3 (25) | 0 (0) | 4 (24) | 4 (11) | 2 (18) | 2 (7) | 2 (15) |
| 5-7 | 23 (18) | 8 (17) | 3 (11) | 4 (33) | 8 (21) | 2 (12) | 6 (16) | 1 (9) | 6 (22) | 1 (13) |
| 8-10 | 16 (12) | 8 (17) | 5 (19) | 2 (17) | 2 (5) | 3 (18) | 9 (24) | 2 (18) | 2 (7) | 1 (13) |
| 11-13 | 30 (23) | 6 (13) | 7 (26) | 2 (17) | 7 (18) | 3 (18) | 11 (29) | 0 (0) | 7 (26) | 1 (13) |
| 14-16 | 26 (20) | 2 (4) | 3 (11) | 0 (0) | 12 (31) | 1 (6) | 5 (13) | 1 (9) | 2 (7) | 0 (0) |
| ≥17 | 20 (15) | 1 (2) | 1 (4) | 0 (0) | 7 (18) | 1 (6) | 1 (3) | 0 (0) | 8 (30) | 0 (0) |
| Matched outside of NRMP | 2 (2) | 0 (0) | 1 (4) | 0 (0) | 1 (3) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | 0 (0) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
