Understanding the Sex-Based Professional Development Priorities of Medical Students: Results from a novel self-assessment survey tool

Introduction: A novel survey tool was created in order to compare the professional development priorities of medical students by sex and year in training and to medical school faculty. Method: An electronic survey was developed and emailed to all first and second year medical students and medical school faculty across two campuses. Students ranked seven professional development needs and eight career development events in order of priority. Faculty ranked items based on what they thought would be priorities for medical students. Kruskal-Wallis non-parametric tests were used for all comparisons. Results: 65/200 (33%) students and 78/224 (35%) faculty responded. All students ranked future career planning as highest priority and financial planning as lowest priority. Students were most likely to attend career development events and least likely to attend social events. Males ranked networking (p<0.01) higher than females. Male and female faculty rankings and faculty vs. student rankings were not entirely consistent. Discussion: A novel survey tool was created which is easy to implement and analyze. Results imply sex-based differences in professional development priorities in early medical training. This survey tool can be used to address disparities in female student professional development and provide a more tailored experience for all medical students.


Introduction
In 2018, more women than men matriculated into medical school for the first time in history (Travis, 2018). Larger numbers of women entering medicine, however, has not translated to equal number of women and men in leadership positions (Bickel et al., 2002;Travis, 2018). For example, in 2018 at academic medical institutions only 22.8% of women physicians were full professor, 15% were permanent chair, and 16% of dean positions were held by women (Association of American Medical Colleges, 2017;Lautenberger et al., 2019;Lewis et al., 2018).
The Association of American Medical Colleges (AAMC) 2002 committee report on women leadership in academic medicine proposed that the scarcity of women leaders could be attributed to "constraints of traditional gender roles, manifestations of sexism in the medical environment, and lack of effective mentors" (Association of American Medical Colleges, 2017). To address the scarcity, the committee report recommended a multifactorial approach including evaluating development of female faculty, attracting more female candidates, financially supporting Women in Medicine programs in medical schools, and targeting the professional development needs of women (Bickel et al., 2002). Some medical schools and societies are already attempting to address women's leadership development through such strategies, including the American Medical Women's Association, the Association of Women Surgeons, the AAMC Group on Women in Medicine and Science (GWIMS), and medical school campusorganized women in medicine support groups (Carr et al., 2017;Faucett et al., 2017).
However, for institutions wishing to implement a peer mentorship program there is little research on best practices or standardized guidelines. There is even less research on which aspects of peer mentorship groups students prioritize and how needs may differ by sex and year of training (Meschitti and Smith, 2017). Our goal was to quantitatively assess the professional development priorities of medical students through the development and administration of an online survey. We hypothesize men and women will have different professional development priorities in medical school, which may also differ by year in school. Additionally, medical school faculty may have differing views of what professional needs are most important to medical students. These results can inform peer mentorship programs to provide a more tailored experience based on year and sex.

Survey Development
A literature review with the assistance of a medical librarian did not reveal any published questionnaires which assessed the professional development needs of medical students. However, two previously published surveys were identified which focused on professional development and mentorship of faculty and residents (Colón-Emeric, Bowlby and Svetkey, 2012;Gaetke-Udager et al., 2018). The first survey was published by Gaetke-Udager et al. in the American Journal of Radiology in 2018. The purpose of their survey was to evaluate trainee satisfaction with a Women in Radiology mentorship group and the events hosted by this group. The variety of events hosted by this group was used to inform the creation of our 'Professional Development Events' list (Gaetke-Udager et al., 2018). The second survey, published in 2012 in Medical Teacher by Colon-Emeric et al., asked 66 faculty from different backgrounds to prioritize lists of unmet professional needs and peer-mentoring activities (Colón-Emeric, Bowlby and Svetkey, 2012). The most highly ranked ten professional development needs and seven peer-mentoring group activities reported in their analysis were used as the basis for the needs and events categories in our survey. Some of the language for the needs and themes was adjusted to be more relevant to medical students (Supplementary File 1). A team of medical students and medical school faculty with different levels of teaching experience reviewed all survey items for language and relevance. Talbott J, Foley B, Kling J, Kraus M, Misra L, Kosiorek H, Lim E, Patel B MedEdPublish https://doi.org/10.15694/mep.2020.000236.1 Page | 3

Data collection
The final survey was generated on a privacy-compliant REDcap server and sent in September of 2018 to the emails of all first and second year students on the Mayo Clinic Alix School of Medicine campuses in both Arizona and Minnesota (with a total class size of 200), as well as to all faculty involved in medical education at the Arizona and Minnesota campuses. Both the admissions process and curriculum are the same across schools and both campuses are accredited under the same Liaison Committee on Medical Education credential. Lists of faculty including deans, professors, interest group leaders, and designated academic advisors were provided by the Office of Academic Affairs on both campuses. An email reminder was automatically generated for all participants with an incomplete questionnaire and sent weekly for a total of three weeks. All surveys and scripts were approved by the Mayo Clinic Institutional Review Board. Consent for each participant was gathered at the beginning page of each survey. No personal identifiers were collected in the survey, and all emails were deidentified from survey results.
The student survey consisted of a list of eight professional development needs and seven professional development events. Demographics were also collected including age, sex, race, ethnicity, year in school, marital status, and number of children. Students were asked the following two questions: 1) "What professional development needs are not being fully met, or would you like to see enhanced? (ranked in order of importance or need)" and 2) "Please rank your interest in attending the following professional development events from most to least important." All professional needs rankings had values ranging from 1 to 7. Professional events had values ranging from 1 to 8.
The faculty survey asked respondents to prioritize what they believed were the most important needs for medical students to consider: 1) Thinking about current medical students' education, what professional development needs are not being fully met, or should be enhanced? and 2) Please rank how interested you believe medical students would be in attending the following events.

Analysis plan
Demographic variables were summarized by student and faculty responses. Continuous variables were summarized by median (range) and categorical variables by frequency distribution. We compared rankings between students and faculty by use of Kruskal-Wallis non-parametric tests. Similarly, rankings were compared by sex and academic year for students. We analyzed data with SAS version 9.4 (SAS Institute Inc. Cary, NC). A p-value <0.05 was considered statistically significant.

Demographics of respondents
The student response rate was 33% (65/200), and they were predominantly Caucasian, aged 20-25, and unmarried. Fifty percent of student respondents were women. One respondent had children. Faculty response rate was 34.8% (78/224), and they were predominantly Caucasian, married, and with children. Forty-three percent of faculty respondents were women. A summary of demographics is included in Table 1. Faculty samples were combined across Minnesota and Arizona campuses. When stratified by sex and professional rank statistically significant differences were seen between professional development priorities and events. Statistically significant differences in responses were not seen between academic years of students. Results for students are summarized in Table 2 and 3. Caption: Table 2 reports the mean and median ranking scores for each of the seven professional needs categories in the survey. Results are compared between male and female students and all students and faculty. Samples from Arizona and Minnesota sites were combined. Bolded values indicate Kruskall-Wallis p-value < 0.05.

Unmet professional needs (Figure 1)
All students regardless of sex or year ranked future career planning as their highest priority need and financial planning as their lowest priority need. All students ranked research skills and future career planning higher than that of faculty. Additionally, all faculty irrespective of sex ranked academic/studying skills higher than what students had ranked. When faculty were compared by sex, more female faculty identified work-life balance as a perceived priority for medical students than their male faculty counterparts.

Professional development events (Figure 2)
All students, regardless of sex, ranked career development events as most likely to attend and social gathering events as the least likely to attend. Additionally, all students were more likely to attend resource sharing events than what faculty predicted. All faculty, on the other hand, predicted informal peer discussions would be more popular than what was observed with students. When students were stratified by sex, male students were more likely to attend networking events than female students whereas female students were more likely to attend professional skills training events than male students. Faculty did not demonstrate any significant differences in opinions of events by sex.

Discussion
An original survey tool was created which was easy to implement and analyze. Pilot data of pre-clerkship medical students attending the same medical school at two geographic sites reported similar priorities in certain professional development areas, regardless of year. However, the students' development priorities were not entirely consistent with faculty's perception of the needs and events which they believed to be important for students. This distinction is important, as faculty are often the organizers and leaders of interest groups and mentorship groups on campus. An understanding of these student-identified, high-need areas will help programs better tailor their activities to meet students' interests. Furthermore, facilitating conversations between students and faculty can also allow discussion regarding the reasoning behind the prioritization of professional development areas from each perspective.
For example, results from this survey suggest students do not believe academic skills training are as high need as faculty predicted. Students may feel more equipped to tackle their studying skill responsibilities on their own perhaps due to access to plentiful study resources outside of the medical school. Prior research has suggested many medical students use third-party study resources, such as Anki flashcard decks, SketchyMedical informational videos, and Firecracker practice questions. These resources complement or even substitute their medical school's curriculum (Burk-Rafel, Santen and Purkiss, 2017;Choi-Lundberg et al., 2016;Deng, Gluckstein and Larsen, 2015). By this point in training, pre-clinical students likely have access to appropriate resources and feel confident in their studying skills. If this survey repeated at other institutions yields similar results, school administrators may consider time spent on developing studying strategies may be better used introducing students to basics of medical research and future career development opportunities.
Many of the surveyed students in this sample preferred to attend structured events which facilitate resource sharing as opposed to informal, peer-to-peer social events. It may be that students are meeting their social needs elsewhere outside of school organized events. Programs or social groups may choose to focus their efforts on more formal sharing of resources, funding sources, and research tools. For example, learning communities, which randomly divide students into longitudinal mentorship groups across year, are widely prevalent across medical school campuses. Per a recent Liaison Committee on Medical Education (LCME) survey conducted in 2014, 102 out of 140 responding AAMC schools reported organizing their students into formal "colleges or mentorship groups" (Shochet et al., 2019). Furthermore, these learning communities provide opportunities for advising and mentoring by cultivating long-term student-to-student and student-to-faculty relationships, building social networks, and organizing professional development activities (Shochet et al., 2019). If survey results at other sites suggest resource sharing is in demand, learning communities may provide an appropriate venue.
We observed sex-based differences in professional development events among students, with males preferring networking events and females having a stronger preference towards professional skills training. "Imposter phenomena" may be a possible explanation for why women preferred additional skills training compared to their male counterparts. Imposter syndrome occurs when female professionals feel inadequate, unskilled, or incompetent despite having high success, high skills, or accomplishments (Fitzpatrick and Curran, 2014;Shochet et al., 2019;Villwock et al., 2016). More female physicians and female medical students experience an 'impostor phenomenon' when compared to their male colleagues (Fitzpatrick and Curran, 2014;Shochet et al., 2019;Villwock et al., 2016). The presence of gender bias or microaggressions in the workplace may additionally make females feel inadequate (Files et al., 2017). As a result, female students in this sample may feel more unprepared in terms of their professional skills despite having the same level of training as male students. Therefore, female students may focus on more education and development opportunities as a way to over-compensate for this feeling of inadequacy. We also hypothesize these results will look similar across medical school campuses.
The imposter syndrome, and a lack of confidence in ones networking skills, may be one possible explanation for why female students in this sample were less likely to prioritize networking compared to males. Other research on female physicians and female entrepreneurs suggests women's more limited engagement in networking may be due to a lack of aspiration for organization leadership, lack of time, or a focus on seeking advice from within a smaller social circle (Chae, 2018;Mariam Malende, 2017). Regardless of the specific reasons, promoting same-sex mentorship is important as female students and trainees traditionally lack academic and professional mentors (Association of American Medical Colleges, 2017; Butkus et al., 2018;Colón-Emeric, Bowlby, and Svetkey, 2012;Burk-Rafel, Santen, and Purkiss, 2017;DeFilippis et al., 2016;Faucett et al., 2017;Gaetke-Udager et al., 2018;Lautenberger et al., 2019;Meschitti, Smith, 2017). Multiple studies have shown that same-sex mentors can target professional development disparities for women early in their career and can be especially influential when students are determining their future career choice (Faucett et al., 2017;Gaetke-Udager et al., 2018;Lautenberger et al., 2019;Lewis et al., 2018). Women in medicine mentorship groups therefore may have a consequential role in addressing feelings of imposter syndrome, highlighting the importance of mentoring in building professional networks, and providing opportunities for women-focused networking (Faucett et  survey tool provides an opportunity for such mentorship groups to assess the specific priorities of their students and then tailor events and activities accordingly. We also observed sex-based differences among faculty members, with more female than male physicians prioritizing work-life balance for students. Female physician mothers bear the burden of childrearing and household tasks, even when in dual-physician relationships (Lyu et al., 2019). This additional work and stress load may contribute to higher rates of burnout among women (Bickel et al., 2002;Lyu et al., 2019). Female physicians therefore may advocate that students address work-life balance earlier in their careers. Given the relatively young age and stage of training of the sample population, the students themselves may not yet see this as high priority. This provides another justification for promoting conversations between faculty and students to discuss reasoning behind prioritization of professional development and life skills during medical training. This novel survey tool can provide a means by which to promote these conversations.

Strengths and Limitations
Strengths include the development of a survey tool to evaluate the professional development needs of medical students. This short and easy to implement electronic survey can be used in a variety of settings, such as mentorship groups hoping to improve their programming or medical school administrations targeting professional development gaps within a student class. While certain conclusions were drawn based upon pilot data from this sample, other institutions are encouraged to repeat this survey in order to make decisions using the most relevant data. Another strength of the pilot data is students and faculty from geographically diverse sites were surveyed and contributed to results of the study.
The limitations of our research include the homogeneity of the respondents and small sample size due to a partial response rate and single institution study. This survey included first and second year medical students only, so the priorities presented are only for pre-clinical years. We also collected data at a single time point; therefore, this study cannot provide trends of information or insight into how perception of needs may change over the course of education. In future study we will evaluate a larger, more diverse student population across more academic years in order to accurately assess needs and describe how needs may change over the course of training.

Conclusion
Using an original survey in a group of pre-clinical medical students, sex based differences and faculty to student differences were found for professional development priorities. These results can inform the activities of on-campus peer mentorship groups, interest groups, and learning communities. As academic medicine seeks to address the sexbased career disparities of female students and physicians, it is critical to understand and enhance professional development opportunities for female students in training. This novel survey tool may help programs understand the professional development priorities of medical students in order to best prepare our next generation of successful physicians.

Take Home Messages
Male and female medical students may have different professional development needs. 1.
There are few evidence-based assessments for the professional development of medical students.

2.
This short, electronic survey asks students and faculty to rank seven professional development needs and 3.
eight career development events in order of priority.