INTRODUCTION

Currently, 11.2% of physicians (vs. 33.4% of the US population) identify as a member of a historically underrepresented ethnic or racial group (Black or African American, American Indian or Alaska Native, Native Hawaiian or Other Pacific Islander, or Hispanic or Latinx), and 17.1% of physicians (vs. 5.9% of the US population) identify as Asian (which includes multiple subgroups of cultures and languages).1, 2 International medical graduates (IMGs) comprise 25% of the physician workforce—the majority from countries without predominantly European heritage.3 In this paper, we refer to “physicians of color” (POC) as either US medical school graduates who are members of ethnic or racial minority groups historically underrepresented in the USA4 or who have other non-European heritage (e.g., Middle Eastern or Asian), or IMGs who come from countries without predominantly European heritage.3 POC are more likely to encounter workplace discrimination than their White counterparts and report negative impacts of such discrimination on career advancement, work satisfaction, and health which are accompanied by higher rates of job turnover, changing specialties, and leaving medicine altogether.5,6,7,8,9 The cumulative burden of discrimination toward POC has been termed a “minority tax10” and described as “racial fatigue.11”

Research has identified colleagues, leaders, and evaluation practices as sources of workplace discrimination for POC, largely overlooking the patient as a potential source.5 In a systematic review,5 we identified only one survey study of POC that inquired about discrimination from patients with a single question about patients refusing care.11 Given the paucity of research on this important issue, we explored experiences with and observations of discrimination from patients toward POC.

METHODS

Study Design

In this qualitative study, we conducted and recorded in-depth, semi-structured, one-on-one telephone interviews of practicing general internists at academic medical centers about their experiences with and observations of patients exhibiting discriminatory behavior toward POC. Interviews occurred between May and September 2019. We used thematic analysis to inductively generate themes from transcribed interviews. We followed the Consolidated Criteria for Reporting Qualitative Research (COREQ).12 The University of Wisconsin-Madison institutional review board approved the study.

Participants

We recruited a national, purposive sample of academic general internists.13, 14 To obtain rich, qualitative data from a wide range of participants who have experience with the phenomenon of interest, we recruited participants by randomly selecting both private and public institutions accredited by the Association of American Medical Colleges (AAMC) across the four US geographic regions.15, 16 Historically Black Colleges or Universities (HBCUs) were grouped separately (n=3).

We collected physicians’ names, email addresses, and demographics online from institutional websites17 and randomly selected 5 physicians from each of 4 demographic groups (White-male, White-female, POC-male, POC-female) from each randomly selected institution. Once all POC from an institution received invitations, we randomly selected a new institution from that group (e.g., private, Northeastern institution) to continue our recruitment of physicians from the four demographic groups. We invited participants via email and verified demographics with a pre-interview questionnaire. We expanded recruitment with snowball sampling.18 Recruitment and data collection continued until no new ideas emerged (thematic saturation).19 In total, we sent 955 email invitations including 529 POC over 8 rounds of recruitment, with rounds 7 and 8 only recruiting POC (48 and 56 individuals, respectively). At some institutions, we invited all POC on their list of internists. Of the 44 who responded to our email invitation, 16 participated. We invited 21 additional physicians via snowball referrals; 8 participated. In our final sample of 24 participants, 14 participants self-identified as White and 10 as a physician of color, which included Asian, Black, Native American, and self-identified other race. The physician who identified as Native American also identified as Hispanic/Latinx. No physician from an HBCU volunteered to participate in our study. One physician indicated that they were an IMG (Table 1).

Table 1 Characteristics of General Internists (n=24) Participating in a Qualitative Study on Physicians’ Experiences With Discriminatory Patients
Full size table

Data Collection

A.F., L.A., or A.R. audio-recorded one-on-one phone interviews from a private office with participants in their preferred location. A.F. is a researcher with a PhD, L.A. is a nurse researcher with a PhD, and A.R. was a medical student with a BA. A.F. and L.A. identify as White, A.R. identifies as African American, and all three identify as female. All had training in qualitative methods. Interviewers introduced themselves at the start of the call and explained their role in the study. Some participants knew M.C. through her research.

We developed a semi-structured interview script of questions to ask all participants, pilot tested at a single site (Appendix, Table 3). Interviewers took field notes to facilitate the interview process; these notes were not included in analyses. Based on analysis of the first few interviews, we added two questions relevant to the phenomenon of interest to the interview script. We did not conduct repeat interviews.

Data Analysis

We sent verbatim transcripts of interviews to participants for verification. Of the 17 who responded, one participant sent back grammatical edits; others confirmed the accuracy of the text. Using thematic analysis,20 four authors (A.F., L.A., A.R., K.P.) independently coded 10 transcribed interviews line-by-line, meeting to discuss codes, and build our codebook. A.F. and L.A. then coded all the interviews using the established codebook. Facilitated by QSR NVivo 12 software,21 A.F. and L.A. grouped codes into potential themes and subthemes, refining them through an iterative process. We sent descriptions of our themes to all study participants for verification (member checking).22 Six participants responded and confirmed that our findings reflected their experiences. The labels applied to the themes were adjusted until the authors were satisfied that they accurately captured the theme. Triangulation of all authors’ perspectives and member checking helped ensure that the themes reflected the full range and depth of the data.23 From the finalized themes and subthemes, we created a thematic map.

RESULTS

The 24 interviews lasted 10–52 min (median=37 min); in the interview that lasted 10 min, the participant chose to skip over questions. Overall, White participants reflected on what they have observed while POC focused on both their lived experiences and observations. Although our focus was on discrimination directed at POC, physicians with other identities described experiencing discrimination from patients. We included these experiences in our analyses.

Four major themes emerged (Fig. 1): (1) assumption that a legitimate doctor is White, male, and able-bodied; (2) legacy of the Black experience; (3) working through the struggle of discrimination; and (4) ethical dilemma of providing care to discriminatory patients. These major themes and their subthemes are described in the following sections. Table 2 provides representative quotes with participants’ self-described demographics and assigned number.

Figure 1
figure 1

Thematic map of practicing general internists’ observations of or interactions with patients exhibiting discriminatory behavior toward their physician illustrating the relationship between themes (rectangles) and subthemes (ovals) identified in the data.

Full size image
Table 2 Illustrative quotations from participants
Full size table

Themes and Subthemes

Assumption That a Legitimate Doctor Is White, Male, and Able-bodied

Male and female POC, White female physicians, and a physician with a disability reported how patients challenge their role as a physician from outright refusal of care to unintentional microaggressions. These experiences sent the message to our participants that patients assume doctors are White, male, and able-bodied, often prompting the need to assert their physician status.

Refusal of Care

Some physicians described patients who exhibited explicit prejudice by refusing care from their physician. For example, a participant witnessed a patient refusing care from an Asian physician colleague because he was “not even an American” (American Indian, female, P274). Another physician had a patient that “just started yelling saying ‘I am not going to see her, she’s a terrorist’” to a colleague (White, female, P272). This refusal of care was driven by both gender and race: “the patient specifically wanted to see a White, male doctor” (White, female, P263).

Microaggressions

As described by a White, male physician (P213), microaggressions may be “meant in a positive way”; for example, “a patient, in an effort to be kind, said to an African American resident, ‘it’s good to see more of you people in medicine.’” Other patients assumed a POC was in a lower status role without being overtly discriminatory: “predominantly Black or Latino colleagues…have been asked by patients, ‘when are you gonna come by and clean our room,’ thinking they were housekeeping staff” (Asian, female, P240). Female physicians reported being mistaken for a nurse which was usually gender-based but intersected with race: “people are still calling me nurse after I introduce myself as a doctor” (Black, female, P292). Having a disability also led patients to question a physician’s role: “as a person with a disability…patients have a hard time believing that I was the supervising physician on the team” (White, male, P211).

Indicating Physician Status

Participants described both verbal and non-verbal ways to establish their physician status.

Because discrimination “adds to that insecurity of ‘Oh…I’m not really this authority figure that I’m trying to be right now’” (White, female, P229) some “women physicians are much more likely to wear white coats with their names on them” (White, male, P288). One physician described it as, “‘doctoring up’…to be able to give…that external symbol…of authority” (Asian, female, P245). Female physicians and POC reported that they introduce themselves using their title and “repeat [that they are the] ‘doctor’ a few times” (Asian, female, P247).

Legacy of the Black Experience

Black physicians shared some experiences with other POC, but given encounters with racism outside of medicine, their experiences were also unique. As described by an Asian, female physician: “speaking on behalf of my colleagues who are Black…I think they get it way more than I do” (P240). One Black, female physician (P218) shared that, “a patient…referred to me as the ‘n’ word.” Even in situations that were not explicitly racist, one noted “it’s always in the back of your mind as a Black physician that somebody’s hatred or hostility towards you is partially due to [race]” (Black, female, P292). One White female physician (P272) noted that patients may make racist statements like “wow, it is so nice to see a smart, Black doctor.” Some Black physicians described spending more of their “emotional and cognitive energy battling” discrimination every day (Black, female, P278), and perceived their experiences with discrimination as having a negative impact on their health: “I think I’m gonna have a stroke just having to work around racism and its impact on health…I’m increasing my cardiovascular risk” (Black, female, P278).

In addition to overt instances of discrimination, the frequency of microaggressions encountered by Black physicians led one physician to take the following position:

I’ve taken the attitude…to dismiss a lot of the microaggressions because they’re my everyday life so I wish that there was more known about how best to combat this from an internal way and a social way. (Black, female, P292)

Black physicians sometimes described the need to be “resilient…let it roll off us without it affecting us too deeply” (Black, female, P278). One described how being trained at a highly regarded predominantly White medical school served as a form of protection from microaggressions:

Flash a [medical school] badge and…doors would open… Those credentials of goodness that can sort of cover you like White skin when you don’t have the White skin…so it makes it extra challenging when you have Black skin and a Black school (Black, female, P278).

Working Through the Struggle of Discrimination

Participants described variable support from colleagues, the long-term impact of encounters with discriminatory patients, and commonly used responses.

Lack of Support

The lack of support from colleagues when faced with a discriminatory patient was exemplified by one physician: “afterwards people were pretty quiet about it except for a pharmacist”; as a result, she has “never mentioned it to my boss or my chair” (Black, female, P218). Some wanted institutions to recognize discrimination from patients as an issue and provide support: “the institution could do a better job of (A) naming and talking about these issues and (B) educating providers about what kind of practical strategies could be used” (White, male, P210). A White, female physician (P297) felt the institution needed to be more explicit to discriminatory patients about the consequences of their behavior: “it should be clear to patients we are changing a provider because of the provider’s mental health, not because you are a racist.”

Impact of Discriminatory Patient

Interacting with a discriminatory patient made some physicians feel isolated, excluded, and alone. An Asian, male physician (P202) reflected on how being frequent targets of discrimination from patients may compel physicians to “quit medicine or make a change or leave the practice.” Such interactions could contribute to the attrition of POC “because patients are being racist and they [POC] can’t be a doctor because they’re not being seen as a doctor” (Asian, female, P258). This wore POC “down; it makes [them] tired…and some days [they] don’t even feel like coming to work” (Asian, female, P240). As perceived by a White, female physician (P272), interacting with discriminatory patients can have “a huge impact on wellness and confidence and…overall mental health.”

Supporting One’s Self and One’s Colleagues

Some participants chose to confront a discriminatory patient: “if it’s something more overt…taking a firm stance that this behavior is not going to be accepted” (Asian, male, P202), while others used “humor a little bit” (White, female, P287). Sometimes the target removed themselves: “I just left the room and I took the team with me” (White, female, P297). A White, male physician (P237) described how “we would need to support the [physician] in not being discriminated against.”

Some physicians described means of coping with the aftermath of discriminatory patients. One chose to “journal and write things down to sort of process it” (Asian, female, P260). Another recommended that physicians “reach out to others in their social support network,” including mentors, colleagues, friends, and family: “I still need other people to take care of me” (Black, female, P278).

Communication Toolbox

Participants described having a “communication toolbox” (Asian, female, P240) for responding to a discriminatory patient. One White, male physician (P237) will say, “‘it’s interesting that you feel comfortable saying that,’… as a way of noting the language without…an accusation of its motivation or underlying bias.” Physicians without a “toolbox” were curious “what other people have had ready in their pocket” and noted their institutions need to “develop policies to address [discriminatory patients] in a way that protects healthcare providers” (White, male, P211).

Ethical Dilemma of Providing Care to Discriminatory Patients

When they encountered a discriminatory patient, physicians in our study were often torn between their aversion to the patient and their professional obligation to provide care, pondering the impact on the patient’s quality of care.

Internal Conflict

Encountering a discriminatory patient affected the clinical interaction and physician-patient relationship. Physicians described an “ethical dilemma” (White, male, P210) between confronting the patient and an obligation to provide care. A White, male physician (P213) described that in working with a neo-Nazi patient with swastika tattoos, the physician team still wanted to stay “true to the professional code,” and provide quality care.

Impact on Quality of Care

A discriminatory patient can jeopardize the quality of care they receive. A White, male physician (P210) questioned “how do you not let that affect your obligation to this person…if you find their beliefs morally reprehensible?” These situations could be draining and required physicians “to turn on some active energy to maintain…objectivity with the patient” (White, male, P213). These types of interactions may affect quality of care: “you might not be as willing to go as out of your way…may not go the extra mile for them” (Asian, male, P202). One physician expressed concern about a discriminatory patient’s refusal to share relevant clinical information with a POC, “which makes the doctor visit take longer and makes the interaction get off to a bad start” (White, female, P258).

DISCUSSION

From microaggressions to refusal of care, physicians in our study described discrimination from patients based on the physician’s race, gender, or disability status. Our physicians’ responses to such discrimination ranged from ignoring what was said to direct confrontation, with the interaction leaving some feeling isolated and alone and others enduring some adverse effects. Some physicians felt torn between providing quality care and confronting the discriminatory patient. Both targets of and witnesses to discrimination from patients expressed a need for support from their colleagues and clearer institutional guidance.

The issue of encountering discriminatory patients has emerged in social media,24,25,26 the popular press,27, 28 and personal essays.29,30,31,32 However, our systematic review of discrimination experienced by POC found little research on patients as a source of discrimination.5 In the interim, a focus group study from Wheeler et al. explored this issue using a convenience sample of hospitalist physicians, residents, and medical students.33 Despite differences from our study design and sampling, participants in both studies encountered similar types of discrimination from patients and identified lack of institutional support as a barrier for change.

Even in the absence of overt prejudice, our physicians encountered microaggressions.34, 35 This was sometimes in the form of “status leveling” where a person is assumed to belong to a group of lower status in a social hierarchy (e.g., a Black physician assumed to be housekeeping staff or a female physician assumed to be a nurse).36,37,38 Black physicians in particular experienced the well-described cumulative psycho-emotional impact of frequent microaggressions.34 The current COVID-19 pandemic underscores the challenges and vulnerabilities of Black physicians practicing in the USA.39

When faced with a discriminatory patient, the American Medical Association’s Code of Ethics recommends transferring their care to another physician.40 Mayo Clinic has developed policies to facilitate transfer of patient care in such situations taking into account the condition and mental state of the patient, the effect of transfer on the patient’s health, and the provider’s safety.41 While well-intended, it is not yet known whether having an institutional policy helps resolve the conflict our physicians voiced between their aversion to a discriminatory patient and their responsibility to care for that patient.

Arming physicians with stock phrases for use with discriminatory patients would appear to provide an area for immediate action. Analogous to the “communication toolbox,” Mayo Clinic provides physician targets and bystanders with scripts to use with discriminatory patients.41 Mello and Jagsi provide scripts to address patients who exhibit sexual harassment or gender bias42 and Goldenberg et al. found that behavioral rehearsal of specific scripts in a workshop increased skills and confidence of physician faculty to intervene with patients exhibiting prejudice toward trainees.43, 44 In addition to training physicians to competently deliver appropriate scripts when experiencing or witnessing discrimination from patients,41,42,43,44,45 health care organizations can implement bias literacy training so members of the organization have a common language46 to more easily recognize, label, discuss, and address episodes of discrimination.47, 48 Organizations can develop and visibly display diversity and inclusion messages that emphasize a growth mindset for overcoming the influence of stereotypes49 and look to social marketing research to construct messages directed at patients to reduce prejudice toward their physician.50, 51

Our use of a national sample, authentication of transcripts and themes with participants, collection of data until thematic saturation was achieved, and convergence of findings with existing research strengthen the trustworthiness of our data and transferability of our findings; however, our study has limitations.22 Our findings are limited to general internists in academic medical centers at predominantly White institutions, and although thematically consistent in their perceptions and experiences, our sample contained only three Black physicians and one Hispanic/Latinx physician. We have no information on reasons physicians did not participate which could have been numerous; e.g., the invitation went to “junk mail,” no experience with or observations of patients discriminating against POC, no interest in participating because of the topic or in a time consuming interview, or racial fatigue of POC at having to constantly “tell their story.”30

In conclusion, our findings contribute to a mounting body of evidence that physicians who do not match patients’ mental model of a physician as White, male, and able-bodied can be targets of demeaning and belittling behaviors from patients. While achieving a respectful work environment for all physicians will require a cultural change, there are actions health care organizations can take immediately to work toward this goal.52, 53