Mistreatment of medical trainees is a pervasive issue in academic medical centers.1,2 The topic of mistreatment was first brought to the attention of the medical community in 1982 when Silver3 described the transformation of medical students’ attitudes from “enthusiastic” and “excited” to “cynical,” “depressed” and “filled with frustration” as they progressed through their training. Mistreatment, he suggested, was a significant factor in this change. Although the subject received little attention in the following years, discrimination and other forms of mistreatment have resurfaced as critical issues for medical schools to address. All undergraduate programs are now required to document instances of mistreatment towards trainees by physicians, peers, and members of the medical staff. In 2013, the Accreditation Council for Graduate Medical Education was also charged to address this problem in residencies.4–6
Numerous studies published in the last five years have found that trainee mistreatment and discrimination is a widespread phenomenon with between 17% and 95% of trainees reporting its occurrence.5,7,8 Variable forms of abuse have been reported. One meta-analysis found that verbal harassment was the most common form, with discrimination based on gender and race most prevalent, ranging from 4% to 19%, respectively.8 Although definitions of mistreatment vary, we use the definition offered by Mavis et al,5 which defines mistreatment as any behavior that shows disrespect for the dignity of others including discrimination based on race, gender, and religion.
Despite the prevalence of mistreatment overall, there remains a paucity of data regarding mistreatment in graduate medical education, and even less is known about the nature of mistreatment from patients and their families. One study of first-year residents found that of nearly 400 respondents, 93% had experienced “disruptive behavior” including abusive language, gender and racial bias, berating, and exclusion from decision making.9 To our knowledge, only one study has examined the prevalence of discrimination and other forms of mistreatment towards trainees by patients. This study found that patients accounted for nearly 40% of this behavior.10 In 2015, we conducted a survey of all pediatric residents at our institution and found a disproportionately high rate of patient and family mistreatment towards trainees. Fifteen percent of residents had personally experienced or witnessed mistreatment; of those instances, 67% involved mistreatment, including discrimination by patients’ families. Further, 50% of respondents did not know how to respond to these instances, and 25% believed no action would be taken if they alerted hospital leadership.
Discrimination towards trainees by patients and families does not appear to be uncommon, yet to our knowledge, the literature offers no recommendations for how to respond. To address this gap, we sought to qualitatively explore the perspectives of experienced faculty educators to (1) identify approaches trainees and physicians can use to respond to discrimination from patients and their families, and (2) describe specific educational strategies trainees and faculty can employ to address this issue.
We employed a constructivist grounded theory approach, a qualitative methodology well suited to exploring complex social phenomena not explained by preexisting theory.11 From April to June 2014, we purposefully selected members of the pediatric residency Program Evaluation Committee (PEC) at Stanford University to participate. We chose these participants because we wanted to ensure representation of individuals who held both clinical and teaching responsibilities, who occupied leadership positions within the residency program (e.g., rotation directors, associate program directors), and who had extensive experience teaching and mentoring trainees. The 110 members of the PEC were separated by leadership role to capture a range of positions and were recruited via e-mail to participate. Members were not included in the invitation pool if they were nonfaculty (fellows) or their e-mail was not found, leaving us with 96 eligible participants.
Following grounded theory methodology, recruitment, data collection, and analysis proceeded iteratively, allowing us to seek new participants and insert new prompts into the interview guide to probe emerging concepts and themes.11 Recruitment stopped after 13 interviews once thematic saturation was achieved (i.e., no new concepts emerged, and the concepts were well developed).
Interview guide development
The semistructured interview guide consisted of three clinical scenarios of trainee discrimination (Table 1) and a series of open-ended questions to probe reflection and responses to these scenarios. The scenarios were developed following an extensive literature review by the authors and discussion about real experiences at our institution. One scenario (racial discrimination) was adapted from the literature. Two scenarios (religious and gender discrimination) were adapted from real-life experiences. Prior to study initiation, two authors [E.W., A.B.] pilot-tested the semistructured interview guide with three faculty members to assess content and clarity. Questions that were unclear were revised until all authors approved the final guide.
One nonphysician author who is a trained qualitative interviewer [A.B.] conducted one-on-one 75-minute interviews. Interviews were conducted in private offices to maximize privacy, and all participants completed an informed consent. To prime participants to consider the trainee perspective, participants were asked to first describe how they would respond to each scenario as the trainee and then to describe their response as the supervising physician. Participants were also asked to describe strategies that could be employed to teach trainees and faculty how to respond effectively in these situations. After the interview, participants completed a brief demographic survey.
Interviews were audio-recorded, transcribed verbatim by a professional transcription service, and reviewed for accuracy by two authors [E.W., A.B.]. All identifying information was removed prior to analysis. All data were electronically recorded and maintained on a secure password-protected file.
We conducted the analysis in three phases using constant comparative analysis to develop themes.11 First, E.W. and A.B. independently read the first two transcripts line by line to identify and descriptively label content areas with codes. Codes were applied to phrases, sentences, and paragraphs. The authors met to discuss their lists of codes and create an initial coding scheme. This coding scheme was applied to new data as interviews were conducted and was revised over time through discussion and reconciliation. Incoming data informed our decision to merge or remove codes until a final code list was created. In the second stage of analysis, the authors reread all transcripts to manually code using the code list and to cluster codes around emerging themes. In the third stage, the authors met to discuss their list of themes and the relationships between them and to identify exemplary quotations. Differences of opinion about the meaning of specific passages were discussed until consensus was reached. Themes were refined until the authors developed a fully conceptualized list. Data were evaluated for trustworthiness using participant review, a process where participants are asked to review the themes for accuracy.12
All study procedures were approved by Stanford University’s IRB prior to initiation.
Thirteen faculty participated in this study. There were 96 education leaders eligible; 24 education leaders were sent e-mails, 15 of whom responded; 13 were interviewed (54%). The majority were female (n = 8), non-Hispanic white (n = 10), and Christian (n = 6). Four interviewees were division chiefs, 3 were rotation directors, 3 were clinical coaches, 2 were other academic advisors, and 1 was a scholarly concentration leader. Demographics are in Table 2.
Four themes emerged in how participants would respond to discrimination in the scenarios: (1) assess illness acuity, (2) cultivate a therapeutic alliance, (3) depersonalize the event, and (4) ensure a safe learning environment for trainees.
Assess illness acuity
Acuity of illness emerged as an important influence over how participants would respond to discriminatory remarks and parents’ requests for alternative providers. Participants suggested that as a first response, providers should assess the medical needs of patients and determine whether immediate medical intervention was necessary. In emergencies, participants suggested that providers ignore or avoid responding to the discriminatory remark. Participants also believed that parents’ requests for alternative providers should not be granted under these circumstances.
It’s an urgent versus routine thing. If we’re just doing a routine checkup, it’s one thing. If you’re here to deliver a baby or the baby urgently needs attention and I’m the only person to provide the care, then there’s not a lot of choice in the matter.
Cultivate a therapeutic alliance
The majority of participants endorsed cultivation of a therapeutic alliance with parents and patients as a constructive response to discrimination or other forms of mistreatment when medical issues are not acute. Participants believed identifying, naming, and validating the emotional experience underlying the discriminatory remark was an important step in establishing trust with families. They believed discrimination was often motivated by parents’ fears and suggested empathizing with this anxiety to draw parents’ attention away from their personal prejudices and toward a shared goal of addressing the child’s immediate health needs.
Trying to get at what the real fears are can be helpful in building a trusting relationship. If you’re willing to listen to them, explore “What are the things that you’re really concerned about? Tell me what you’re afraid of. Tell me what your desires are.…” It begins to build a trusting relationship, which is really critical.
Participants advocated building rapport and trust with families by deemphasizing the target of the discriminatory remark (e.g., the trainee’s religion) and emphasizing the child’s medical care. This goal could be achieved by highlighting the qualifications of the team as a whole.
I would say to parents, “At this juncture, we don’t really have the luxury to worry about the religious beliefs or origins of the people that are available. All members of the medical staff are highly qualified. Your child needs attention from people with the right expertise. Let’s see if we can get past [the prejudice] so we can be sure we’re doing the right thing for your child. I wouldn’t want to compromise the care for your baby because of concerns about who is doing the providing.”
Notably, four participants believed that accommodating families’ requests for alternative providers gave credence to their prejudicial beliefs and wondered if it reinforced discriminatory behaviors. These participants were less likely to cultivate an alliance with families and explore the emotions underlying the behavior. Rather, they adhered to a pragmatic approach of explaining which providers were available and offering families the choice of receiving care elsewhere.
If I think there’s some mistaken thought that is contributing to this prejudice or to not wanting this provider to take care of the child, then I’m willing to go there. But if it seems to be a situation of just prejudice then I’m not going to get into that conversation. We’re just going to focus on “These are our providers. This is what we do. Let’s focus on getting what you came here for.”
Depersonalize the event
Participants recommended depersonalizing the event as another strategy to effectively manage discriminatory situations. They believed it was important that parents’ comments not be taken personally and to recognize that discriminatory remarks are often more reflective of the parents than they are of the provider. Although participants acknowledged that this strategy may not lessen the discomfort providers feel, they believed it could minimize heedless responses and negative emotional reactions that could interfere with patient care. One participant described how he would talk through this strategy with a trainee.
“What are some of the skills that we can give you so [the situation] won’t put up a barrier, but somehow you find a place where you feel okay about that?” Not that these things are ever okay, but again to say it’s their problem. We have to figure out how to move around that rather than, “Oh, my God. There was some truth in that, and it hurt,” or “I can’t believe someone would ever think that about me.”
Another participant suggested depersonalizing the event and reducing reactivity in the moment by encouraging reflection on the values of the medical profession and one’s commitment to patient care.
There’s always a question of how much do you want to take on in a professional setting when you’re trying to be professional but someone is challenging your beliefs or your feeling of what is right and wrong.… The emotional heaviness of this can be alleviated if you rest on your professional values.… I’m here to provide medical care for the child. The child is my patient, not the parent. Do no harm.
Ensure a safe learning environment for trainees
All participants described a commitment to trainees and a desire to protect them from these situations. They acknowledged that instances of discrimination and other forms of mistreatment could not be prevented and recognized that creation and maintenance of a safe learning environment are essential. Participants believed that trainees are particularly vulnerable and felt that the medical team and learning environment as a whole needed to provide sufficient support to counteract the potential harmful effects of discrimination.
There’s need for follow-up when anything in the room comes up that was unexpected, let alone something so personal as this could be. Sitting down and having a one-on-one or even having the whole team discuss it is important because I think all residents will be discriminated against no matter what at some point in their career and need to learn skills to handle that … at hospitals these days a very big concern is patient satisfaction … that’s of high importance to me as well. But I also think as educators, it’s our responsibility to be protective of our learners and create as optimal an environment as we can, knowing they are going to have to deal with these things long-term.
Participants hoped that trainees would feel empowered to remove themselves from a discriminatory encounter if it compromised patient care, but acknowledged that this was not always possible or easy to do. All participants believed that trainees should make their own decisions to remove themselves; faculty making this decision for them would further undermine trainees’ autonomy.
I certainly would respect [the trainee] saying, “Maybe this isn’t something that I should be put through, and I don’t want to have to deal with this family” or “I know I can’t give this patient the care he or she deserves now.” I would respect that if that was the choice they made. Or, if they felt like they wanted to continue to provide care, I would absolutely support them and be there to back them up … it can be hard for trainees to make these decisions though.
Participants described several specific responses both trainees and faculty can take immediately following a discriminatory event that fall within these themes (Table 3). Finally, participants generated several recommendations of educational strategies to teach trainees and faculty how to respond to these scenarios (Chart 1).
This study extends the medical education community’s understanding of mistreatment by describing concrete approaches for addressing acts of discrimination, one form of mistreatment,5 by patients and families. Studies suggest that these instances are not rare, and although many authors discuss ways to respond to the “difficult patient,” few specifically describe how to respond to the difficult discriminatory patient.8,13,14
Participants described a range of acceptable responses to the discriminatory patient or family, and four specific approaches emerged that can be used as a first response: (1) assess illness acuity, (2) cultivate a therapeutic alliance, (3) depersonalize the event, and (4) ensure a safe learning environment for trainees. We believe these approaches begin to form a model that can be taught to trainees and faculty alike to ensure adequate preparation for these events.
Although participants generally agreed on these approaches, the respondents who did not strive to cultivate a therapeutic alliance offered an intriguing perspective. These respondents described an approach to patient care that was focused on immediate medical needs and did not allow room to explore the motivations behind the discriminatory behavior. They did not tolerate acts of discrimination by patients or families and believed a pragmatic approach offered the best solution. Participants endorsing this view had more clinical and teaching experience than other participants in our study and may have had personal experiences that shaped these opinions.
All participants agreed that trainees should be prepared to address discriminatory patient encounters because prevention is impossible. Case-based discussions were unanimously endorsed as one way to practice working through these scenarios. These discussions can help physicians recognize what behaviors they are willing to accept and how to respond constructively. We recommend discussion of this type of mistreatment early in training so that trainees feel equipped to respond and feel permission to remove themselves from care when necessary.15
Participants also agreed that maintaining a safe learning environment is crucial in the event of discrimination. They recommended that the attending acknowledge the event and initiate a debriefing with the team as soon as possible to emphasize that the trainee’s safety is a priority, that such behavior is not tolerated, and to identify and reinforce opportunities for professional growth and learning. Individual trainees targeted by these acts should be given the opportunity to reflect in person with the attending or with another trusted source for support and guidance. This approach is consistent with other studies that have encouraged the use of critical reflection as a way of processing emotionally difficult clinical encounters.16
The option of separating oneself from a discriminatory encounter also arose as an important element of ensuring a safe learning environment. Nevertheless, we believe that terminating a patient encounter when it is uncomfortable should not be a medical trainee’s first response, as potential learning opportunities could be missed. Part of the debriefing practice that faculty members can initiate with their trainees should include discussions of alternative ways to engage with patients and families.
Following what we learned from this study, we recommend that institutions develop procedures for addressing these situations at the trainee, faculty, staff, and hospital administration levels. We recommend the creation of task forces that create institutional policies addressing patients who discriminate against providers. Processes must be in place to ensure protection of trainees and the faculty who train them, and to ensure continuity of care for patients needing transfer when they refuse to be treated by the team. Our political and social climate is such that discrimination is at the forefront of many human interactions. As medical professionals, our duty is to prepare our trainees, faculty, staff, administrators, and ourselves to appropriately navigate these situations. To achieve this goal, trainee and faculty development needs to be championed and routinely included in our clinical practice.
Despite these important findings, this study does have several limitations. First, this study was conducted at a single institution. Second, our sample was small, and there may be self-selection bias in those who chose to participate. Although our sample size was in the range recommended by qualitative methodologists, and data collection continued until saturation of themes was reached,17 our findings may not represent the full range of approaches that can be taken in these instances. Future research will need to determine the transferability of our findings to other settings with more diverse faculty. We deliberately chose a nonphysician interviewer because of the anticipated benefit of encouraging physicians to speak freely about a sensitive topic. Although we recognize the potential for her to introduce her own perspectives into the interviews and analysis, our constructivist grounded theory approach affords strategies for accounting for this including using more than one analyzer and researcher reflexivity, which involves the creation of memos and field notes that allow the researcher to engage in continuous self-awareness of her role in analysis.11
This study describes a range of acceptable responses that trainees and faculty can take with discriminatory patients and families, and provides strategies for teaching our learners how to best respond in these situations. Although physicians may vary in the degree to which they accommodate discriminatory preferences, their dedication to the care of their patients and to the protection of their trainees is constant. Advance preparation through the strategies described can equip trainees and faculty to respond constructively in ways that ensure the safety and well-being of patients and trainees.
Acknowledgments: The authors wish to thank Laura Bachrach, MD, and Elizabeth Stuart, MD, MSEd, for their contributions to the conception of this project and Lavjay Butani, MD, and Jennifer Plant, MD, MEd, for their expertise on critical reflection.
1. Brooks KC. A silent curriculum. JAMA. 2015;313(19):19091910.
2. Reynolds KL, Cowden JD, Brosco JP, Lantos JD. When a family requests a white doctor. Pediatrics. 2015;136:381386.
3. Silver HK. Medical students and medical school. JAMA. 1982;247:309310.
4. Li SF, Grant K, Bhoj T, et al. Resident experience of abuse and harassment in emergency medicine: Ten years later. J Emerg Med. 2010;38:248252.
5. Mavis B, Sousa A, Lipscomb W, Rappley MD. Learning about medical student mistreatment from responses to the medical school graduation questionnaire. Acad Med. 2014;89:705711.
6. Schenning KJ, Tseng J. Toward a humanistic learning environment: Addressing resident mistreatment. J Grad Med Educ. 2013;5:344.
7. Baldwin DC Jr, Daugherty SR. Do residents also feel “abused”? Perceived mistreatment during internship. Acad Med. 1997;72(10 suppl 1):S51S53.
8. Fnais N, Soobiah C, Chen MH, et al. Harassment and discrimination in medical training: A systematic review and meta-analysis. Acad Med. 2014;89:817827.
9. Mullan CP, Shapiro J, McMahon GT. Interns’ experiences of disruptive behavior in an academic medical center. J Grad Med Educ. 2013;5:2530.
10. Crutcher RA, Szafran O, Woloschuk W, Chatur F, Hansen C. Family medicine graduates’ perceptions of intimidation, harassment, and discrimination during residency training. BMC Med Educ. 2011;11:88.
11. Charmaz K. Constructing Grounded Theory: A Practical Guide Through Qualitative Analysis. 2014.2nd ed. Thousand Oaks, CA: Sage Publications.
12. Corbin J, Strauss A. Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory. 2015.4th ed. Thousand Oaks, CA: Sage Publications.
13. Breuner CC, Moreno MA. Approaches to the difficult patient/parent encounter. Pediatrics. 2011;127:163169.
14. Miedema B, Hamilton R, Lambert-Lanning A, et al. Prevalence of abusive encounters in the workplace of family physicians: A minor, major, or severe problem? Can Fam Physician. 2010;56:e101e108.
15. Irby DM. Three exemplary models of case-based teaching. Acad Med. 1994;69:947953.
16. Wald HS, Davis SW, Reis SP, Monroe AD, Borkan JM. Reflecting on reflections: Enhancement of medical education curriculum with structured field notes and guided feedback. Acad Med. 2009;84:830837.
17. Watling CJ, Lingard L. Grounded theory in medical education research: AMEE guide no. 70. Med Teach. 2012;34:850861.