Breast Surgical Oncology Fellowship Applicant Selection and Ranking: a Survey of Society of Surgical Oncology Programs

Abstract

Background and Objectives

In 2003, the Society of Surgical Oncology (SSO) initiated a breast surgical oncology fellowship, which has now grown to 60 SSO accredited programs as of 2021. Limited knowledge exists on the traits of successful applicants and the factors influencing the rank list.

Methods

A web-based, anonymous survey was sent to all SSO Breast Surgical Oncology Fellowship program directors. The survey consisted of 26 questions. Descriptive statistics were used to analyze survey responses and evaluate impact on applicant interview and rank list.

Results

Thirty-four programs (57% response rate) completed the survey. Programs received an average of 70 applications and granted 24 interviews. Most programs reported a minimum ABSITE cut-off score (n= 28, 82%) and a defined publication requirement (n= 22, 65%), including a first-author requirement (n= 18, 53%) to extend an invitation to interview. For post interview rank, applicant interpersonal skills were highly valued. The interview was the most important aspect for the rank list.

Conclusions

Many programs have ABSITE and publication thresholds before offering an interview. Upon receiving interview invitation, the applicant’s interview performance, interpersonal skills, and letters of recommendation were the most important aspect in rank list decision making.

INTRODUCTION

Fellowship applications have steadily risen over the past decades with recent estimates of 85% of graduating general surgery residents are interested in fellowship training(1–3). The Society of Surgical Oncology (SSO) approved Breast Surgical Oncology Fellowship training in 2003, which now encompasses 60 programs across the US(4). Despite rapid growth and interest among trainees, limited knowledge exists on the traits of successful applicants to SSO Breast Surgical Oncology Fellowship training. Our group and others have previously conducted descriptive studies that have evaluated the qualities of successful complex general surgical oncology, pediatric surgery, colorectal surgery, and thoracic fellowship applicants, but no such data exists on the qualities that make for a successful Breast Surgical Oncology Fellowship applicant(5–8).

Given the growing interest in post-graduate fellowship training and the myriad of academic, professional, and research training considerations that are now available for general surgical residents, it is important to understand the specific factors that are associated with successful matriculation to the Breast Surgical Oncology Fellowship. Understanding the attributes of successful fellowship applicants enables residents, residency and research programs, mentors of general surgery trainees, as well as fellowship program directors to assist applicants to focus their clinical, academic, and research time to be most successful in their professional development.

In the present study we aimed to evaluate the characteristics associated with an invitation to interview and inclusion on the rank-list by querying all SSO accredited Breast Oncology Fellowship directors with a web-based survey. Programs were questioned about candidate background, most important factors for extending an interview offer, and the attributes they were looking for during the interview process. Programs were further queried regarding factors that were most important for inclusion on their rank list. Our goal was to establish the candidate academic and research benchmarks for obtaining an interview and, assuming that this threshold was met, what specific applicant traits were most valued in offering interviews as well as ranking a candidate highly.

METHODS

Data Acquisition

An anonymous, web-based survey was sent to all SSO Breast Oncology Fellowship program directors and program coordinators. Initial notification letters were sent to all program directors in June 2021. The survey was available for a total of two months from June 2021 to September 2021 via a link embedded in the request for participation emails. As of March 2021 there were 60 SSO approved Breast Oncology Fellowship programs(4). Only one response per program was accepted. No incentives were offered for participation. Gender and racial demographic data were not collected as this was considered deidentified data and given the small number of programs inquired this information could potentially lead to identification of the survey participant. Due to the anonymized nature of the survey, this study was reviewed and considered to be exempt from institutional review board (IRB) review at both the National Cancer Institute and Moffitt Cancer Center.

The web-based survey was built using the SurveyMonkey (SurveyMonkey Inc. Palo Alto, CA; www.surveymonkey.com). The survey was developed using the modified Delphi method and consisted of 26 questions with a mixture of dichotomous yes/no, multiple choice, five-point Likert scale, rank, and free response questions(6). The complete survey is available as Supplemental Document 1. The Likert scale varied from “Not at All Important” to “Critical to Further Consideration”. A second Likert scale evaluated the likelihood of offering an interview to applicants based on particular characteristics ranging from “Much Less Likely” to “Much More Likely”. Program directors were queried regarding factors that affect selection of applicants both pre and post interview. Free response questions allowed respondents to identify the number of applications they receive on average, the average number of interviews they offer, and the number of fellowship positions offered. They also allowed for more granular details on prior questions. The final free response question asked about ways the application process could be improved, ranked questions evaluated the importance of ABSITE score, letters of recommendation, interview performance, publications/research experience, and personal phone calls supporting the candidate compared to one another.

Response Analysis

Responses were deidentified, so responses were not able to be connected to respondent or program. Data was entered into Statistical Package for the Social Sciences (SPSS) (IBM, Armonk, NY) for statistical analysis. Responses were reported as a percentage of respondents unless otherwise noted. Likert scale questions were evaluated by weighted analysis with a response of “Not at all Important” being assigned a value of one and a response of “Critical to Further Consideration” receiving a value of five with graded responses in between. Similarly, for responses evaluating applicant characteristics “Much Less Likely” was assigned a value of one and “Much More Likely” received a value of five. For Likert scale questions a response of “Not Applicable” was not included in analysis, so the denominator for percentages reflects the number of responses for a given question not based upon the total number of respondents. Mean and mode were calculated to evaluate the overall importance of each characteristic. Ranked responses were calculated by assigning the highest ranked variable as one and lowest ranked variable as five to reflect ordinal conventions. Free response questions pertaining to virtual interviews were tallied with percentages based upon the number of participants entering a free response.

RESULTS

Survey Respondent Information

Thirty-four out of the sixty Breast Surgical Oncology Program contacted responded to the survey (57% response rate) (Table I). These programs reported a median of 1 fellowship position per program (range 1 – 4). For these positions, the programs reported receiving a mean of 70.4 applications (range 20 – 100). The mean number of interviews granted per program was 23.7 (range 8 – 40). This equated to a mean interview rate of 35% (interviews granted per applications received) for all the programs taken together, with individual program rates ranging between 16% to 100%. Of the respondents, 38.2% had applicants selected by the program director alone while 61.8% involved multiple faculty members in the selection process. For programs with multiple people involved in decision making, the median number of faculty members involved in decision making was 3 (range 2–6 individuals). In response to the application process in general, 24% of respondents commented on the negative effects of the virtual interview format during COVID, and no respondents commented favorably about the virtual interview format in its current state.

Table I.

Characteristics of the fellowship programs responding to the survey

Fellowship Program Characteristics
Breast Surgical Oncology Fellowship Program Response Rate	34/60 (57%)
Mean number of positions per program (Median, Range)	1.4 (1, range 1 – 4)
Mean number of applications received per program (Median, Range)	70.4 (70, range 20 – 100)
Mean number of interviews granted (Median, Range)	23.7 (22, range 8 – 40)
Mean interview rate (interviews granted/applications received)	35% (range 16 – 100%)

Factors that influence fellowship interview invitation

All thirty-four programs responded to the survey questions regarding the importance of ABSITE scores, publication record, and applicant characteristics on the likelihood of receiving an interview. Of the respondents, 82% of programs reported having a minimum pre-requisite ABSITE score prior to granting interviews; of these 60.7% reported this minimum to be between 25 – 50^th percentile and 21.4% reported a minimum of 10 – 25^th percentile (Figure 1A). 65% of programs reported a publication requirement prior to granting interview, with 22.7% of these programs requiring at least 2 manuscripts, 9.1% requiring 3, 4.5% requiring 4, and 18.2% of programs requiring 5+ manuscripts for interview (Figure 1B). Finally, 53% of programs reported requiring a first author manuscript for consideration for interview, with most of these programs requiring between 1 and 3 first author manuscripts (66.7%) (Figure 1C).

Figure 1.

Importance of ABSITE score (A), Publication Number (B), and First Author Publications (C) on securing interviews for Breast Surgical Oncology Fellowship Programs

In addition to reporting on minimum ABSITE and publication milestones, applicants spending time at the fellowship institution for either clinical rotations (mean score 4.18 +/− 0.76) or research activity (4.06 +/− 0.69) were deemed to be more likely to influence programs to offer an interview (Figure 2). Secondary degrees such as MPH (mean 3.56 +/− 0.75), PhD (3.53 +/− 0.86), or MBA (3.29 +/− 0.72) had a negligible effect on likelihood of offering interview. Most programs (n=23, 67%) rated dedicated research time to be only slightly or not at all important in their decision to extend an interview, however12% (n=4) of programs reported dedicated research time to be very important. Lastly, applicants who had previously applied to but were not accepted to a Breast Surgical Oncology Fellowship were deemed to be the less likely to receive an interview (mean score 2 +/− 0.67).

Figure 2.

Importance of Applicant Characteristics on securing interviews for Breast Surgical Oncology Fellowship Programs. Mean scores on Liked Scale are depicted by blue horizontal bars, with higher scores representing applicant characteristic more likely to positively affect receipt of interview.

Factors that influence inclusion on the rank list

The scoring of factors pertinent to applicant final, post-interview rank is summarized in Figure 3. Applicant interpersonal skills such as ability to work in a team, good communication skills, ability to interact well with other applicants/non-physician staff during interviews, and ethical decision-making skills were valued the most in the rank determination. In contrast, overall publication quantity, interest in becoming a surgeon-scientist, or laboratory-based publications were deemed only slightly to moderately important in establishing the final rank list. In directly comparing the interview, letters of recommendation, personal phone calls recommending applicants, publication and research, and ABSITE scores, 75% of programs noted the interview to be the most important factor influencing rank list position. Letters of recommendation were the second most important factor (noted as such by 50% of programs), while interestingly, the importance of publication and research, and ABSITE scores, were noted by most programs to be the 4^th or 5^th important factors for the final rank list Table II.

Figure 3.

Importance of Applicant Characteristics on final rank list position for Breast Surgical Oncology Fellowship Programs. Median scores on Likert Scale are depicted by blue horizontal bars, with yellow up arrowhead depicting 75^th percentile score and down arrowhead depicting the 25^th percentile response. Higher scores represent applicant characteristic that are more important for favorable rank list position.

Table II:

Rank of Importance of Listed Factors Considered in Decision Process

	1st	2nd	3rd	4th	5th
Interview	24 (75%)	3 (9.4%)	3 (9.4%)	2 (6.2%)	0 (0%)
Letters of Recommendation	2 (6%)	16 (50%)	8 (25%)	4 (13%)	2 (6%)
Personal Phone Calls Recommending Applicant	1 (3%)	4 (13%)	9 (29%)	6 (19%)	11(36%)
Publications and Research	4 (13%)	6 (19%)	9 (28%)	11 (34%)	2 (6%)
ABSITE Scores	1 (3.1%)	3 (9.4%)	3 (9.4%)	8 (25%)	17 (53.1%)

Responding programs ranked each factor in comparison to the other stated factors when considering a candidate. Bold indicates the aggregate most common ranking for each of the categories. Numbers may not add up to 38 as not all programs responded to each question.

DISCUSSION

Fellowship applications for all general surgery residents continue to increase with over 70% of residents applying to post residency fellowship training(9, 10). Breast Oncology is a popular fellowship with more applicants than available positions. Data regarding applicant characteristics associated with successful matriculation into Breast Surgical Oncology Fellowship Training Programs match process and successful is limited despite equivalent information available for other general surgery subspecialty fellowship applications. The study aims to was to identify the establish academic and research benchmarks, and applicant-specific qualities that are of greatest importance, as identified by the program directors, for interview and high ranking for breast oncology fellowship. We identified that over 50% of programs have a minimum ABSITE score of 25^th percentile and a minimum number of publications (most commonly 2 publications) required for interview offer. Additionally, programs valued interpersonal skills of teamwork, communication, and ethical decision making, which they assessed during the interview.

While we were able to identify modifiable factors in the match process for applicants to optimize, program directors most valued applicants’ interpersonal skills. The interview was weighted most heavily in determining the program’s rank list followed by letters of recommendation, phone calls of recommendation, publications, and finally ABSITE scores. During the interview program directors valued an applicant’s perceived teamwork ability, interpersonal/communication skills, and ethical decision making. While these results are similar to the CGSO fellowship survey in the values of program directors in their selection process, we did identify subtle differences. For instance, our survey indicates that Breast Oncology Fellowship directors placed greater value in applicants’ specific interest in their fellowship program than their CGSO counterparts(6). Another difference from CGSO that was identified in our survey was in the publication requirement for interview. The CGSO survey found that 83% of programs had a publication requirement for interview offer compared to 65% for Breast Oncology and a mean minimum publication requirement for CGSO of 4.4 publications compared to 2 publications most commonly reported for Breast Oncology(6). Laboratory publications were considered slightly/moderately important as compared to clinical publications which were found to be moderately important. However, overall publication quantity was not highly ranked in the applicant selection process, nor was the necessity of basic science research Altogether, our results suggest residents should pursue research that most aligns with their academic interests when considering dedicated research time.

This study provides insight into the Breast Oncology Fellowship match, which had not previously been explored, but it does have several limitations. Given then we were unable to assess raw applicant data we were reliant upon a survey study of program directors. This method was inherently biased as questions were subjective and respondents could vary in their interpretation of the survey questions. We tried to mitigate this bias by using the modified Delphi method in constructing this study. A similar survey was used for CGSO fellowship which demonstrated potential limitations of the study in prior iterations(6). Since the survey relied on ranking of subjective variables, there was limited ability for statistical analysis of the data. This study was also subject to the Hawthorne effect whereby subjects subconsciously or consciously alter their answers because they know they are being observed(11). We attempted to mitigate this by making the survey anonymous and using a web-based survey platform that does not allow for respondent identification. We excluded any question that could provide identifiable data about a respondent or their affiliated program. Although there are many limitations to a survey study, this method provides the best perspective to date on the breast oncology fellowship application process and enables applicants and mentors to understand the characteristics that make an applicant successful.

CONCLUSION

This study helps to establish the characteristics that Breast Surgical Oncology Fellowship program directors evaluate in applicants. Most programs have an academic threshold as measured by ABSITE scores and publication numbers that are associated with receiving an interview invite. After these baseline criteria are achieved, the interview was the most important factor in deciding the rank list. Our study suggests that program directors prioritize applicants with strong interpersonal skills. Applicants who are most successful in the fellowship match are the candidates who demonstrate their ability to work as a team, have strong communication skills, display strong ethical decision making, and are able to convey their accomplishments effectively during the interview process.

Synopsis:

ABSITE scores and publication numbers are used by more than half of breast surgical oncology fellowship programs to select applicant for interviews. The applicant’s performance during interview, perception of interpersonal skills, and letters of recommendations were the important factors affecting position on the rank list.