Factors Influencing Graduate Program Choice Among Factors Influencing Graduate Program Choice Among Undergraduate Women Undergraduate Women

Context: Despite equal enrollment proportions in MD and PhD programs, there are fewer women than men in MD-PhD programs and academic medicine. Factors important in degree program selection, including the perception of gender disparities, among undergraduate students were characterized. Methods: In 2017, women pre-health students at four public North Carolina universities were invited to participate in an online survey regarding career plans, decision factors, and perceptions of gender disparities in MD, PhD and MD-PhD pathways. This study characterizes factors important to program selection, and evaluates the association of intended graduate program with perceived gender disparities using Fisher’s exact tests. Results: Among the n=186 female survey participants, most were white (54 %) and intended MD, PhD, and/or MD-PhD programs (52 %). Sixty percent had heard of MD-PhD programs, over half had no research experience, and half were considering but uncertain about pursuing a research career. The most common factors influencing degree program choice were perceived competitiveness as an applicant, desired future work environment, and desire for patient interaction. Twenty-five percent of students considering MD, PhD, and MD-PhD programs stated that perceived gender disparities during training for those degrees will influence their choice of program, however intended degree was not statistically associated with perceived gender disparities.


INTRODUCTION
Women are underrepresented in academic medicine, particularly in high-ranking leadership positions, 1 as physician-scientists, 2 and in full-time clinical positions. 3 The latest Physician -Scientist Workforce Report prepared by the National Institutes of Health (NIH) revealed that women represent just 2 2 % of MD-PhD research project grant awardees. 4 Factors that differentially affect the career trajectories of women and men include, but are not limited to, a stated interest in a research-based career, 5 participation in domestic and parenting responsibilities, 5 available opportunities for mentorship, 7 financial compensation, 8 grant funding, 9 and career attrition. 10 Gender disparities, defined here as differences in the treatment of students who identify as female and students who identify as male, are the focus of ongoing dialogues among faculty in academic medicine but are also relevant to students entering the training pipeline at the university and graduate school level s. 11 Despite approximately equivalent representation of men and women in MD-only and PhD-only training programs, only 39% of students enrolled in combined MD-PhD programs in the United States in the 2017-2018 academic year are women. 12−14 Among those interested or enrolled in MD-PhD programs, gender has not been associated with differences in academic preparation, interest or experience in research, 15 or in program attrition rates. 16 The proportion of women compared to men among applicants and among matriculants are similar and both below gender parity of 50%, 17 suggesting that programs have fewer women than men primarily because fewer women choose to apply to MD-PhD programs.
Faculty with both MD and PhD training represent 45% of the NIH-funded physician-scientist workforce even though dual degree students are a small minority of students pursuing MD degrees. 18 Equalizing the gender balance in MD-PhD training programs and ultimately, in the physician-scientist workforce is a priority for the future 11,19 and necessitates an examination of why women are currently underrepresented in MD-PhD programs. Goals for this study were to identify factors central to the selection of a graduate school program (MD, PhD, or MD-PhD) by women, to examine the association between perceived gender disparities and intended graduate school program, and to pinpoint opportunities for intervention that may encourage equivalent representation between men and women in MD-PhD programs. It is conceivable that the volume of discussion and media attention surrounding gender disparities in STEM fields may deter qualified candidates from applying to advanced degree programs that they would otherwise consider. Here, the authors consider the example of MD-PhD programs and hypothesize that female undergraduates' perception of gender disparities in graduate program training is negatively associated with their stated intention to matriculate to an MD-PhD program.

Survey instrument
The 40-item survey was developed to assess the demographics, post-undergraduate training plans, and factors influencing choice of graduate program among undergraduate students (Appendix 1). Respondents who indicated interest in applying to MD, PhD, and/ or MD-PhD programs were asked questions about who discouraged or encouraged them to apply to the indicated program s and about factors impacting selection of programs and institutions. Some questions were adapted from the Summer Undergraduate Research Program at Icahn School of Medicine at Mount Sinai survey (2015).
All respondents who indicated interest in graduate training answered questions about their perception of gender disparities in MD, PhD, and MD-PhD training. "Gender disparities" were defined in the survey as "differences in the treatment of students who identify as female and students who identify as male." More specifically, specific aspects of training that may be differentially experienced by students who identify as female (e.g., differential treatment by faculty or peers, acceptance of having children, impact of length of training on family planning) were queried. Respondents were asked to rate their agreement with the statement "Gender disparities during training will play NO role in my decision about the type of degree-granting program(s) to which I will apply" with response options of true, false, or I don't know.
All respondents were asked demographic information, including whether they identified as an underrepresented minority, which institution they attend(ed ), current year of study, and familiarity with MD-PhD programs. "Underrepresented minority" was defined using the American Association of Medical Colleges definition: "racial and ethnic populations that are underrepresented in the medical profession relative to their numbers in the general population." Participants who indicated interest in MD, PhD, and/or MD-PhD programs were asked which program they would select if they had to pick one today; this single choice is referred to as the "intended" training program.

Survey distribution
The survey was hosted electronically via Qualtrics at UNC-Chapel Hill. Survey invitations were emailed to undergraduate listservs for students interested in health-related careers at UNC-Chapel Hill (UNC-CH; PhD and MD-PhDs were originally on a 5-point Likert scale ranging from "strongly disagree" to "strongly agree", with an additional option "I don't know." Response patterns for "neutral" and "I don't know" were examined via Fisher's exact tests and were not meaningfully different, so they were grouped into one option ("neutral/I don't know") in analyses. "I don't know" comprised approximately half of this group

RESULTS
In November 2017, a total of 186 women completed the survey. Responses were comparable between R1 (highest research activity) and R2/R3(higher or moderate research activity) universities (1.4% and 1.8% of listserv subscribers, respectively). Most participants were white (5 4 %), 41% identified as an underrepresented minority (Table 1). Most participants (92%) were currently completing their undergraduate degrees. Overall, 52% of participants were considering MD, PhD, and/or MD-PhD training.
When asked to pick one program, most intended to pursue MD degrees (77%), followed by MD-PhD (15%) or PhD (8%) degrees (data not shown).  (8) Sixty percent of women indicated familiarity with MD-PhD training ( Figure 1). However, familiarity varied significantly by institution: students attending universities classified as R2 or R3 (ECU and UNC-Charlotte) were less familiar with MD-PhD training than students at R1 institutions (UNC-Chapel Hill and NCSU). About half of respondents (n=98, 53%) stated they were unsure of their interest in a career that involves research, and over half had no prior research experience (n=101, 54%). Overall, interest in a research career varied by experience with research (p=0.01), with lack of research experience most prevalent among those unsure of their interest in a research career (Figure 2).   Table 2). Some women (15%) ranked gender barriers to success in career as a factor that influences their selection of a graduate program. When selecting an institution for training, the five most important considerations listed were location (58%), program funding (50%), prestige of institution (48%), perceived competitiveness as an applicant (33%), feeling of "fit" with current students (33%). There were no statistically significant differences in the selection of intended degree program across any gender disparity questions (Table 3). Collectively between all those intending an MD, PhD, or MD-PhD program, about half of participants agreed that there were gender barriers in training (range 43-60%) and to career success (range 42-53%) as an MD, PhD, or MD-PhD. While not significant, slightly more MD intending participants agreed that gender disparities exist in MD training (64%) and career success (57%) than in training and success as a PhD (53% and 43%, respectively) or as an MD-PhD (42% and 39%, respectively). Among participants intending to pursue an MD or a PhD program, the majority of women (64% and 63%, respectively) agreed that there are gender disparities in their intended programs' training process, whereas a lower proportion of students intending an MD-PhD program endorsed "agree" (43%) that there are disparities in MD-PhD training, however,43% of those intending an MD-PhD program answered "neutral/I don't know ". Among students intending an MD-PhD program, about half agreed that gender disparities exist in training for an MD-PhD(43%), PhD(50%), and MD (50%). Students intending an MD-PhD agreed more often to the presence of gender disparities in the success of an MD-PhD career (5 0%) compared to in a career as a PhD (36%) or an MD (4 3 %).
When explicitly asked, 57% of women said disparities would play no role in their choice of program, while 25% said disparities would play a role, and 18% were not sure (Table 3). Those intending MD training more frequently indicated that gender disparities would play no role in their choice of program compared to those intending a PhD or an MD-PhD (MD: 61%, PhD: 38%, MD-PhD: 50%) ( Table 3).
Uncertainty regarding whether gender disparities would influence their degree program decision was highest among those intending a PhD (50%), compared with only 14% of those intending an MD and 21% of those intending an MD-PhD. However, none of the differences reached statistical significance.
(Page Break for Table 3 on the following page.)  The working hypothesis was that women undergraduates' perception of gender disparities in graduate program training is negatively associated with stated intention to matriculate to an MD-PhD program.
Encouragingly, none of the gender disparity responses were significantly associated with intended degree program. However, multiple data points in this study's results suggest gender disparities are a concern for some aspiring biomedical professionals. When asked explicitly whether gender disparities would play a role in the participant's choice of degree program, 25% of those intending to pursue an MD, PhD or MD- Over three-quarters of female participants considering MD, PhD, and/or MD-PhD programs stated that their perceived competitiveness as an applicant was a determining factor in their choice of graduate degree program. Women consistently underrate their competence, even at an early age. 20,21 While not unique to MD-PhD training, women who consider MD-PhD training as the best way to achieve their career goals may inappropriately self-select out of the applicant pool because they perceive that they are underqualified to apply. Because attitudes and perceptions are internalized at an early age, it would be beneficial for all womenand especially those attracted to hyper-competitive careers and environments like academic medicineto be able to accurately assess their competencies and apply with confidence to programs that match those qualifications. In addition to perceived competitiveness, other highly ranked factors concerned the nature of the work itself (work environment and patient interaction) and burden of training (financial support, stress, and duration of training). Encouragingly, these factors were endorsed more often than other reasons commonly thought to account for the gender disparity, including the desire for work-life balance, perceived gender barriers to success, and role models. 11 It is likely that experience s that allow women to assess their competencies and qualifications (i.e. undergraduate research initiatives, faculty mentorship) will also help them learn about the day-to-day mechanics and training pathways of different research careers.
Familiarity with MD-PhD training programs was lowest among students at R2 or R3 universities without affiliated MD-PhD programs. It is likely that awareness of MD-PhD programs and the physician-scientist career path is even lower in the general student population compared to this study's sample of pre-health listserv subscribers, 60% of whom indicated knowledge of the joint degree pathway. Additionally, about half of the "neutral/I don't know" responses in the questions about gender disparities in training and success for PhD and MD-PhD careers were originally coded as "I don't know" and were from students who intended to pursue an MD, which indicates that many aspiring physicians are not sure about the presence or effect of gender disparities in more research-oriented training programs and careers. The data from the current study indicate an opportunity for an intervention to disseminate information about MD-PhD programs to both the student population as well as undergraduate research and pre-health advisers. Students at universities without MD-PhD programs may benefit from stronger collaborations between the MD-PhD program and undergraduates, or on-campus information sessions co-hosted by MD-PhD trainees at neighboring institutions. s Furthermore, mentoring support from faculty members has been shown to strengthen the scientific identities and motivation of early undergraduate women, and a limitation of this study is that the authors do not know to which program students will ultimately apply, and their responses reflect their perceptions, rather than empirical experiences, with graduate programs.
Participants' hypothetical choice one degree program at the time of the survey approximates their future path, but their decision may change over time. Strengths of this study include the participation of students from multiple universities as well as those who identified as underrepresented minorities.
In summary, undergraduate women's perceptions of gender disparities in MD, PhD, and MD-PhD training programs; factors that influence their choice of degree program; and sources of encouragement or discouragement to pursue specific paths are presented. In the current study, one in four students intending to pursue an MD, PhD, or MD-PhD program stated that gender disparities during training will influence their choice of program. The most commonly cited factor influencing choice of degree program was perceived competitiveness as an applicanta factor that women consistently underrate. Finally, self-rated understanding of MD-PhD programs varied significantly by institutional research intensity, and very few students were encouraged by pre-health advisers to pursue MD-PhD training. Altogether, there is room for increased dissemination of information about MD-PhD programs to both undergraduate students and pre-health advisers. Future research should evaluate whether such initiatives generate earlier interest in physician-scientist training and encourage women to apply for and remain in physicianscientist careers.