Objective: The purpose of this qualitative study was to characterize the adequacy, effectiveness, and barriers related to research mentorship among junior pediatric hospitalists and general pediatricians at a large academic institution.
Methods: Junior faculty and staff physicians in hospital medicine and general pediatrics at a large academic institution were invited to participate in this qualitative study. In-depth interviews were conducted. Experienced mentors were invited to be interviewed for theoretical sampling. Interviews were conducted and analyzed by using grounded theory methodology.
Results: Twenty-six (75%) of the eligible physicians, pediatric hospitalists representing 65% of this sample, agreed to be interviewed about their mentoring experiences. Satisfied and dissatisfied participants expressed similar mentoring themes: acquisition of research skills, academic productivity, and career development. Four experienced mentors were interviewed and provided rationale for mentoring clinicians in research. Both groups of participants agreed that institutional support is vital for promoting mentorship.
Conclusions: Junior pediatric hospitalists and general pediatricians indicated considerable interest in being mentored to learn to do clinical research. Developing faculty and staff physicians to their utmost potential is critical for advancement in academic medicine. Mentoring clinical physicians seeking to add research skills and academic productivity to their practice merits study as an innovative path to develop clinical investigators. Hospital medicine, as a rapidly developing pediatric specialty, is well-positioned to implement the necessary infrastructure to mentor junior faculty in their academic pursuits, thereby optimizing the potential impact for individuals, families, learners, and institutions.
Although scholarship is critical for academic physicians, clinicians seeking to incorporate research into their practice often struggle to find a clearly defined path toward that goal. Research challenges for pediatric hospitalists have been identified as “insufficient research training, dedicated research time, and academic mentorship and role models.”1
For physician scientists, the impact and benefits of mentorship are abundant and include enhancing development as academic physicians2–5; influencing personal development, career choice, and productivity; providing career guidance specifically related to research publications and grant funding; and offering an essential source of stress reduction, feedback, and practical advice.6–8 Academic physicians who choose research careers determine this academic interest during residency with the clear expectation of structured mentorship.9
Physician educators and clinicians’ careers include patient care, teaching medical students and residents, and less often research. Previous investigations have primarily focused on mentorship related to research needs and mentorship for medical trainees (medical students, residents, and fellows)2,10; studies exploring mentorship experiences of academic pediatric hospitalists and general pediatricians are limited. The purpose of this qualitative study was to understand the adequacy, effectiveness, and barriers related to mentorship among junior pediatric hospitalists and general pediatricians at a large academic institution. This institution has a strong commitment to mentorship evidenced by the support of senior physician leadership and the support for the creation of a mentorship program recently designed by the Office of Faculty Development.
Study Participants and Setting
We invited 35 physicians including junior faculty at the instructor and assistant professor levels (23 physicians; 66%) and staff physicians (12 physicians; 34%) to participate. At the time this study commenced, General and Community Pediatrics and Hospital Medicine were a single division. Although staff physicians in this institution are defined as physicians whose role is primarily clinical, they do have research productivity requirements but have not received any research training. The designation is determined by both departmental needs and individual physician interests, and staff physicians are not considered faculty. Junior faculty signifies level of faculty appointment, and we defined it as instructors or assistant professors; promotion depends on scholarship achievement, not years in rank. Faculty at this medical center may be from several tracks; the majority of participants in this study were from the Clinical-Educator Track: “… appropriate for clinical faculty, who focus the bulk of their professional effort in the area of clinical service augmented by program development, clinical research, administrative service and/or educational activities.”11 Four of the participants were from the Research Track: “… expected to be involved in well-designed research programs which are, or have the potential to be, externally funded.”11 Four effective mentors were identified during the initial interviews, so based on the grounded theory concept of theoretical sampling,12 these mentors were invited to be interviewed as a small second group of participants. The mentors were from 3 institutions; 2 were from the host institution and 2 were from 2 medical centers in different parts of the country; these medical centers are unrelated to the host institution or to each other. Participants suggested these mentors from outside institutions based on previous experiences. These mentors were identified as effective by their mentees who voiced gratitude for them in the context of the interviews; no other measure was taken to confirm their level of expertise.
Data Collection and Procedure
Grounded Theory13 was chosen as the method of data collection and qualitative analysis because the constant comparative process of in-depth interviews and layers of interpretive coding yield a rich description of participants’ mentoring experience and needs. Following Grounded Theory methodology, the research question was not based on hypotheses developed from the literature but rather sought to understand experience and build theory from the data. In this study, we employed in-depth interviews. Initial interview questions for both groups of participants were created by the primary investigator with input from 3 other members of the research team (a physician and 2 psychologists, all experienced in research). The first author, trained in qualitative research methods, conducted each interview. The third author was one of the junior faculty participants interviewed for the study. She was not allowed to participate in data collection to avoid biasing the results. An independent company transcribed the interviews. Either the first author or her research assistant listened to each interview and corrected the transcript for accuracy; all identifying information was removed before coding.
Coding was done in phases: initial, focused, and axial coding.12 The coding team consisted of the first and second authors and 2 other researchers. Coders identified multiple initial coding by using the participants’ own language. Repeated ideas led to focused codes, which in turn were gathered into axial codes, which are codes related to a major concept. Coders discussed which codes emerged as core to our study. Key themes emerge from the large axial codes. We used a constant comparative practice so that interviews and coding happened concurrently. Initial interview questions (Table 1) were used with each participant, and additional interview questions were added based on emerging concepts. We did not provide a definition of the word “mentor” but allowed the participants to describe their experience of mentoring defining the term as they understood it.
In the process of interviewing and coding, the primary investigator made note of important ideas either stated in or stimulated by the interviews under review. She recorded these ideas to track the emerging concepts that led to a proposed model or theory of the process being studied. Interviews were concluded when the study reached saturation, which means no new categories were identified in interviews. When the study team identified clear themes, the final part of the analysis was a confidential member check with participants. This entailed asking the participants to verify accuracy of the de-identified themes and to identify any missing elements.
Twenty-six of the 35 (74%) physicians agreed to be interviewed; 65% were pediatric hospitalists and 35% general pediatricians. This group comprised 12 men and 14 women; 24 white participants, 2 Asian Americans; and included 6 staff physicians, 5 instructors, and 15 assistant professors. All junior faculty and staff physicians were from 1 pediatric academic medical center and provided clinical care to pediatric patients. The participants’ average age was 41. All 4 experienced mentors invited agreed to be interviewed. This group included 2 men and 2 women ranging in age from mid-40s to mid-60s; 3 white participants and 1 Asian American; 2 were from the institution hosting the study and 2 from 2 different medical centers; but all were members of the department of pediatrics at their medical center.
Mentorship and Development of Research Capabilities
Definitions for the word mentor varied in participant responses but included such metaphors as coaching, parenting, and marriage. These metaphors referenced close working relationships; they did not in any way imply poor professional boundaries. Effective mentoring was claimed by 5 participants (19%) who expressed high satisfaction with the quality of mentorship they received. Nine (35%) participants expressed dissatisfaction due to significant barriers to mentorship. The remaining participants were spread out along a continuum based on the adequacy of the mentorship they received. The continuum ranged from highly satisfied to dissatisfied. Three of the highly satisfied were hospitalists; 2 were general pediatricians. Four of the dissatisfied were hospitalists; 5 were general pediatricians. All agreed mentoring had or could make significant contributions to their professional development. The contributions included helping participants identify and write grant applications; supporting their research development by keeping them on focus when faced with multiple opportunities; helping them write institutional review board protocols; providing feedback on manuscripts; and helping them network with other colleagues. The greatest dissatisfaction themes came from physicians who wanted mentoring to develop research capabilities after working as clinicians in the academic environment. Barriers to effective mentoring leading to dissatisfaction included having no mentors either identified or available for research or academic productivity; lack of helpful response from those approached to be mentors; untimely and unhelpful responses to requests for feedback; and well-intentioned but intrusive advice-giving. Mentoring was named as the primary reason participants in this study had experienced success in learning to do research. Participants defined lack of mentoring as the primary reason they were not making progress in learning to do research.
An equal number of hospitalist and general pediatrician staff physicians sought and received faculty appointments. They reported an expectation that securing a faculty appointment would result in investment in their professional development by providing mentorship in research and academic productivity. Although some staff physicians, including 1 hospitalist and 1 general pediatrician, who transitioned to faculty reported satisfaction with receiving mentoring for research, this was not uniform. Some participants, including 1 hospitalist and 1 general pediatrician, who had transitioned to faculty reported receiving no or poor mentorship. The ability to transition from staff physician to faculty results from a decision by the division leadership in conjunction with the Chair of the Department of Pediatrics.
Of the staff physicians who transitioned to faculty, some were members of the group very satisfied with mentoring. They described development of research skills as a result of excellent mentoring. One participant described the mentorship he had received: “Everybody who’s involved in my professional development is constantly kind of talking and figuring out what’s best for me.” His experience mirrored that of 2 of the research-trained participants. Another participant who had made the transition to faculty described the role mentorship played in her development as a researcher:
“I love taking care of patients; I love working with learners…and teaching them and that one-on-one personal relationship that you develop when you work with someone over time. That’s really where I saw myself, and then I was pushed through a variety of mentors to say, ‘Well how do you know what you’re doing is effective?’ And you can’t just say, ‘Well, it feels good,’ but how do you prove it? If the push was ‘operate on mice and figure out stuff like that,’ I would have no interest and I would never go there. But these things were important outcomes that I would like to show, but were not in my realm of thinking. So (mentoring) opened me up to new thinking, ideas, and questions.”
Themes from participants expressing satisfaction with mentoring also included attention to career development; help choosing projects within their areas of interest and corresponding guidance setting limits; and attention to work/life balance. Themes of concern they raised included difficulty obtaining a mentor and the delivery of feedback: “I think it’s hard for people to give difficult feedback, so people don’t do it often and a lot of people don’t do it well.”
Dissatisfaction with mentoring themes emerged including learning to do research, academic productivity, and career development. This staff physician who transitioned to faculty articulated a level of frustration shared by clinical physicians dissatisfied with mentoring:
“Once we get in the faculty track, we have no clear pathway of what we’re supposed to do or how to advance. We know that at least part of that advancement is going to require research, but none of us know how to do a research project or publish. So I feel like our division loses a lot because we don’t have the support to teach us how to do these things. And I think personally we lose a lot because so much good experience falls through our fingers because we have no guidance.”
Clinicians dissatisfied with mentoring proposed potential solutions to some of the barriers to clinicians learning the necessary research skills: “The opportunities the institution offers—how to write a grant, how to write an IRB—the times offered are geared very much toward people who don’t have clinical responsibilities.” Two participants wondered whether leaders imagined clinicians are receiving peer mentoring by consulting with researcher colleagues:
“I think the higher ups may think there is some collaboration, maybe there is some mentorship that goes on, and so I feel like maybe that’s why it’s not a huge priority because they think, ‘Oh we have a research section in our division, I’m sure they’re all collaborating together,’ but the reality is there’s no crossover collaboration.”
Although not wanting to become primarily researchers, these clinicians said they would like to add a research component to their work by spending a limited amount of time conducting clinical research to contribute information pertaining to their patient population:
“It’s been a detriment not to have a mentor...I think we have a patient population that’s worth studying; I think a lot could be done. But I don’t have any formal education in doing research, publication, or grant writing. I don’t really know where to start...I don’t do any real research and I think that’s a great ‘underdeveloped’ (opportunity for my professional development). I like working with the patients and this is what I want to do…I think it would be worth sharing with people. It’s the idea of being able to share that knowledge with other people and helping other people do kind of what we do in this (clinical area) that I think would be very rewarding.”
Perspectives of Experienced Mentors Regarding Clinicians and Research
Experienced mentors who were interviewed provided triangulation to the concept of clinical physicians learning to do research. The experienced mentors along with several of the physician participants suggested that someone should have a defined role to provide mentorship to clinicians pursuing research. This would address the barrier of limited number of effective senior research mentors. One experienced mentor stated:
“I think the clinical faculty should...say, ‘We need a mentor to advance our careers…to help us do some clinical research because we’re at the front lines and we see a lot of clinical questions that need to be answered, but we don’t know how to answer them. So, we need someone with a research background to help us design studies to answer the questions.’”
Another of the experienced mentors saw difficulties with this concept: “There’s no mechanism to pay for the time to support the clinicians to learn research…so it’s an infrastructure challenge.” Additionally, mentoring is time consuming and mentor time needs support:
“It has to be part of someone’s job description…supported either through internal or external funding… Someone has to do some pretty intense mentoring...So, unless you say, ‘We want you, person X, to help achieve this mission. We want (to) make our clinical environments more academically productive. I’m going to recognize that you’re spending 10% of your time doing that. Here’s this internal fund that’s going to help you and here’s the deliverables that you as a mentor need to help these people provide.’ But nothing like that exists.”
The experienced mentors and many of the participant physicians agreed that mentorship needs to be clearly incentivized by institutional leadership.
Junior faculty and staff physicians in pediatric hospital medicine and general pediatrics identified the ability to successfully perform research and become academically productive as the root cause for both mentoring effectiveness and barriers to mentoring. This is especially important in building the appropriate infrastructure in pediatric hospital medicine as the specialty continues to mature and develop members with clinical, educational, and research expertise.
Goldhamer et al14 named difficulties faced by physician clinical investigators: “debt acquired during medical training, long training periods required for research careers…difficulty obtaining grant funding, lack of protected research time, and more lucrative clinical opportunities.” If physician clinicians can perform clinical research and become academically productive with the help of mentoring, this approach could be a beneficial path for developing a cadre of clinical physician researchers.
Our study’s finding that clinical physicians have a difficult time receiving mentorship for academic pursuits is consistent with Reid et al’s15 previous finding. Only 42% of academic hospitalists report having a mentor, even though mentorship has been associated with producing a peer reviewed first author publication, publishing a nonpeer reviewed article, and leading a teaching session at a national meeting.15 The perception that clinicians are “workhorses” for the institution who don’t warrant attention to their professional development is costly in the sense of lost potential and perhaps ultimately decreased job satisfaction and retention.16–18
Physicians in our study perceived that the lack of research mentoring was hampering their professional development. Although physicians did not explicitly express a need for protected time, Reid et al15 found that academic hospitalists with both 20% protected time and an understanding of promotion requirements were more likely to be academically productive. Our participants’ perception that mentoring will lead to increased research capabilities is consistent with previous studies that demonstrated effectively mentored junior faculty noted an increase in “research skills and preparation.”15 In a prospective study in academic medicine, promotion and scholarly productivity were closely related.19 These findings support the participants’ concerns that lack of mentoring about research design, development, and dissemination could impede their professional advancement.
For junior faculty and staff physicians who have the opportunity to work with mentors, the “managing up” model described by Zerzan et al20 provides an approach that equips the mentee to get the most helpful feedback from the mentor by taking ownership of the process. Also, the shortage of available senior mentors in hospital medicine suggests that alternative models of mentorship (ie, peer, group, and distance) may need to be explored as potential solutions to enhancing mentorship.21
Our study has several limitations. First, all staff physician and junior faculty respondents were from 1 division of an academic medical center that in the process of the study became 2 divisions. Because the Hospitalist Division was just becoming separate from the General and Community Pediatrics Division, the mentoring experiences between general pediatricians and hospitalists were not discernibly different. Although qualitative research does not aim for generalizability, our findings may nonetheless be transferable to other pediatric departments or departments in other disciplines to guide development or improvement of mentoring programs. Second, although our sample of junior faculty and staff physicians was robust for a qualitative study, our experienced mentor group was small. Third, we did not attempt to document outcomes or assess any particular program, but key themes identified may be used to inform future mentoring initiatives.
Developing academic physicians to their utmost potential is critical for advancement in academic medicine. Hospital medicine, as a rapidly developing pediatric specialty, is well-positioned to implement mentoring standards to ensure clinical physicians are mentored in academic pursuits, thus optimizing the potential beneficial impact for individuals, families, learners, and institutions.
Special thanks to Janet Schultz, PhD, Director of Graduate Psychology at Xavier University, Cincinnati, Ohio, and to Brian Matevia, MDiv, Chaplain Research Assistant at Cincinnati Children’s Hospital Medical Center for their help coding the transcripts. Special thanks to Constance Baldwin, PhD, Professor of Pediatrics, University of Rochester Medical Center, and Raymond Baker, MD, MEd, Professor Emeritus of Pediatrics, Cincinnati Children’s Hospital Medical Center, for their editorial feedback. Very special thanks to the junior faculty, staff physicians, and experienced mentors who were willing to be interviewed for this study.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
- Wilson KM,
- Shah SS,
- Simon TD,
- Srivastava R,
- Tieder JS
- Kroenke K
- 11.↵Pediatric faculty appointments, reappointments, promotions, and tenure: Academic criteria. 2009. Available at: www.sagepub.com/books/Book217586. Accessed October 2, 2013.
- Charmaz K
- Glaser BG,
- Strauss AL
- Leavitt LL
- Kay FM,
- Wallace JE
- Copyright © 2014 by the American Academy of Pediatrics