eTable 1. Percentage of Participants Reporting Ever Experiencing Biased Patient Behavior by Race/Ethnicity and Gender
eTable 2. Percentage of Participants Ever Experiencing or Witnessing Biased Patient Behavior Targeted at Specific Social Characteristics in the Last Year
eTable 3. Confidence in Responding to Biased Patient Behavior
eTable 4. Beliefs Around Necessity of Policies and Training on Addressing Biased Patient Behavior
eTable 5. Beliefs on Adequacy of Prior Training on Biased Patient Behavior in Terms of Content and Time
eAppendix. Resident Experiences With Biased Patient Behavior Survey
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
de Bourmont SS, Burra A, Nouri SS, et al. Resident Physician Experiences With and Responses to Biased Patients. JAMA Netw Open. 2020;3(11):e2021769. doi:10.1001/jamanetworkopen.2020.21769
How often do resident physicians encounter incidents of biased patient behavior, and how do they respond?
In a survey study of 232 internal medicine residents from 3 institutions, biased patient behavior ranging from belittling comments to refusal of care was experienced or witnessed by nearly all residents. Forty-five percent of Black/Latinx residents experienced epithets or refusal of care, and most women (87%) experienced sexual harassment; however, most residents (84%) did not report these encounters to their institutional leadership.
Given the high prevalence of biased patient behavior, residency programs that aim to foster inclusive training environments should implement resident and faculty training and create patient reporting mechanisms.
Biased patient behavior negatively impacts resident well-being. Data on the prevalence and frequency of these encounters are lacking and are needed to guide the creation of institutional trainings and policies to support trainees.
To evaluate the frequency of resident experiences with and responses to a range of biased patient behaviors.
Design, Setting, and Participants
A retrospective survey was sent via email to 331 second- and third-year internal medicine residents from 3 academic medical centers in California and North Carolina. First-year residents were excluded owing to their limited interactions with patients at the time of participant recruitment. Data were collected from August 21 to November 25, 2019.
Main Outcomes and Measures
Descriptive statistics were used to report the frequency of experience of various types of biased patient behavior, residents’ responses, the factors impeding residents’ responses, and residents’ experiences and beliefs regarding training and policies.
Overall, 232 of 331 residents (70%) participated; 116 (50%) were women; 116 of 247 (47%) were White (participants had the option of selecting >1 race/ethnicity); and 23 (10%) identified as lesbian, gay, bisexual, transgender, or queer. The frequency of resident-reported experience of types of biased patient behaviors varied. The most common behaviors—belittling comments and assumption of nonphysician status—were reported to be experienced 1 or more times per week by 14% of residents (32 of 231) and 17% of residents (38 of 230), respectively. Women, Black or Latinx, and Asian residents reported experiencing biased behavior more frequently. Forty-five percent of Black or Latinx residents (17 of 38) encountered instances of explicit epithets or rejection of care. All 70 Asian residents reported experiencing inquiries into their ethnic origins. Most women residents (110 of 115 [96%]) experienced role questioning behaviors, and 87% (100 of 115) experienced sexual harassment. The need to prioritize clinical care and a sense of futility in responding were the most common factors (cited by 34% of residents [76 of 227] and 25% of residents [56 of 227], respectively) significantly impeding responses to biased behaviors. Eighty-five percent of residents (191 of 226) never reported incidents to their institution. Eighty-nine percent of residents (206 of 232) identified training and policies as necessary or very necessary.
Conclusions and Relevance
This survey study suggests that biased patient behavior is experienced frequently by internal medicine residents. Non-White and women residents reported experiencing a disproportionate burden of these incidents. Residents’ responses rarely included institutional involvement. Residency programs and health care systems should prioritize training and policies to address biased patient behavior and support affected residents.
Patients who demean physicians on the basis of physicians’ social characteristics pose multiple clinical and ethical challenges.1,2 Biased patient behaviors arise when patients encounter physicians whose social identity (eg, race, ethnicity, or gender identity) is not compatible with their notion of a competent or appropriate health care professional. Biased patient behaviors can range from offensive quips to outright refusal of care3 and can exact a heavy psychological toll on physicians.2-9 It is possible that physicians’ encounters with biased patients have become more common in recent years: the health care workforce is increasingly diverse along multiple social dimensions, and the number of hate groups and hate crimes in the US has grown during the past 5 years, suggesting increased social polarization.10 Characterizing the nature and frequency of these occurrences may be particularly important for academic medical centers, as these incidents undermine learning for trainees and may result in delayed or suboptimal patient care.3,4
Data on the frequency of these occurrences among trainees—and even more so on physicians’ responses to them—are limited, as much of the literature consists of first-person physician accounts of interactions with racist patients or online surveys of practicing physicians.5-7,11-18 A recent national survey of general surgery residents reported that 43% of residents experienced discrimination on the basis of gender identity and 47% experienced racial/ethnic discrimination from patients or patients’ families.17 The same study also found that residents who experienced demeaning behavior were more likely to experience burnout and to have suicidal thoughts. In a qualitative study, internal medicine residents who encountered biased patients reported experiencing painful emotions including fear, self-doubt, exhaustion, and cynicism; residents who did not have these encounters reported moral distress and uncertainty about how to respond and support their colleagues when these situations arise.3
To create institutional policies to support and teach residents and faculty how to manage these challenging patient encounters, academic medical centers and medical educators need data on the frequency of specific types of biased patient behaviors as well as a better understanding of how resident experience may vary according to social characteristics, such as gender identity and race/ethnicity. In addition, it is important to know how residents commonly respond to these behaviors and what individual, patient-related, or institutional elements factor into their decisions on how to respond. We conducted a survey of physician residents at 3 large, academic internal medicine residency programs to evaluate the frequency of resident experiences with and responses to a range of biased patient behaviors.
We administered an electronic survey to second- and third-year internal medicine residents at 3 academic medical centers in California and North Carolina (University of California, San Francisco; University of California, Los Angeles; and Duke University). We collected data from August 21 to November 25, 2019. We excluded first-year residents owing to their limited interactions with patients at the time of participant recruitment. The institutional review boards at each site approved the study, and all participants provided written informed consent. This study followed the American Association for Public Opinion Research (AAPOR) reporting guideline.19
We developed a survey with both multiple choice and open-ended questions based on prior qualitative research exploring the range of demeaning behaviors experienced by physicians and trainees, types of responses to such behavior, and barriers to responding.3 The survey was iteratively revised by the study team. We conducted cognitive testing with residents and faculty from the University of California, San Francisco Department of Medicine to assess the overall coherence, balance, and clarity of the survey. The full survey is available in the eAppendix in the Supplement.
We asked participants to indicate how frequently they had directly experienced the following demeaning behaviors from patients: (1) belittling or demeaning stereotypes (belittling comments, inquiries into racial/ethnic origins, generalizations about social identity, confusing physician with team member of the same race/ethnicity, and nonverbal disrespect), (2) role questioning (credential questioning, assumption of nonphysician status, and addressing less experienced physician or student rather than the participant), (3) explicit epithets or rejection of care (epithets, refusal of care, and request to change physician), and (4) sexual harassment (verbal or nonverbal sexual advances or sexual harassment). Participants were provided the option to describe other demeaning behaviors not defined above. These behaviors are hereafter referred to as “biased patient behaviors.”
Participants were asked to indicate how frequently patients targeted their own social identities (race/ethnicity or national identity, gender identity or expression, Muslim faith, non-Muslim faith, sexual orientation, and disability status) and how frequently they witnessed biased patient behavior targeting other physicians’ social identities. Response options were classified as never, sometimes (once to a few times per year), often (once to multiple times per month), or very often (at least once per week).
We asked how residents respond to these biased patient behaviors (including types and frequencies of responses, barriers to responding, and confidence in responding), residents’ prior experience with training on how to respond to these behaviors, and the degree to which residents believe that training and institutional policies for guiding responses are necessary. We asked about specific responses (1-on-1 limit setting, debriefing with friends or family, debriefing with a team member, creating a team response plan, reporting to attending physician or chief resident, reporting to the institution, switching the patient to another team member, and not addressing the incident). In addition, we asked about the negative impact (from “no impact” to “significant impact”) of specific barriers on responding (prioritizing clinical care; feeling unsupported by the team, senior physicians, or institution; lack of knowledge or skills; perceived ineffectiveness or sense of futility in responding; and feeling emotionally overwhelmed). Participants could also describe other responses and barriers not defined above.
Participants were asked to report their clinical postgraduate year; gender identity; race/ethnicity; identification as lesbian, gay, bisexual, transgender, or queer (LGBTQ) or as another sexual or gender minority; and whether they were an immigrant to the US. Options were defined by the research team with an open field option for participants to describe demographics outside those listed. These demographic characteristics were collected to understand how residents’ experiences varied by social characteristics.
Participants were emailed a link to an anonymous survey via Qualtrics. Follow-up reminder emails were sent to nonrespondents as needed by study authors as well as chief residents at each institution. Participants who completed the survey were emailed a gift card for $10.00 USD to Amazon.
We conducted descriptive analyses of (1) the prevalence of residents’ experiences with biased patient behaviors overall that were stratified by resident race/ethnicity and gender identity, (2) the frequency of behaviors targeting specific resident identities, (3) the frequency of witnessed biased patient behaviors targeting other residents’ identities, and (4) the types and frequencies of residents’ responses to biased patient behaviors. Because of small sample sizes, we combined participants who identified as Black and/or Latinx into 1 category (Black + Latinx) for analysis. In reporting the frequency of different categories of behaviors stratified by race/ethnicity and gender identity, we present means of the component behaviors of those categories.
Two of us (A.B. and S.S.N.) independently reviewed all responses to open-ended questions about experiences with and responses to biased patient behavior and coded responses using grounded theory.20 The reviewers then met to reconcile discrepancies and define themes, which allowed identification of new categories of patient behavior and resident responses not defined in the survey.
Of 331 residents, 232 (70%) responded to the survey. The response rate (using AAPOR Response Rate Calculation 2) was 70% at each of the 3 participating programs.19 Residents were diverse in gender identity and race/ethnicity (Table 1). Residents identifying as LGBTQ or another gender or sexual minority made up 10% of participants (23 of 232); 16% of participants (36 of 232) identified as immigrants to the US. Respondents were representative of the racial/ethnic and gender identity demographics of their residency programs as a whole.
Nearly all residents (228 of 232 [98%]) reported experiencing or witnessing biased behavior at least once in the past year. The frequency of specific biased patient behaviors varied (Table 2). A total of 14% of residents (32 of 231) experienced belittling comments at least once a week, 11% (25 of 230) experienced questioning of credentials or abilities, and 17% (38 of 230) experienced assumption of nonphysician status occurring at least once a week. Behaviors reported by one-third of participants as occurring at least once per month included belittling comments (87 of 230 [38%]), assertive inquiries into racial/ethnic origins (75 of 231 [33%]), generalizations about social identity (70 of 231 [30%]), and credential or ability questioning (77 of 230 [34%]). In contrast, epithets, refusal of care, and requests to change physicians were less common yet were experienced at least 1 to 3 times per year by 40% of residents (91 of 230), 30% of residents (69 of 230), and 27% of residents (61 of 229), respectively. Sexual harassment was also common and experienced at least 1 time per year by 60% of participants (138 of 230).
Experiences with biased patient behavior were more common for residents who identified as women, Black or Latinx, and Asian (Figure). Most residents identifying as women (100 of 115 [87%]), Latinx or Black (29 of 38 [76%]), and Asian (43 of 70 [61%]) reported experiences of sexual harassment within the last year compared with 32% of those identifying as men (36 of 113) and 54% of those identifying as White (59 of 109) (Figure). In addition, 96% of women residents (110 of 115) reported encountering role-questioning behaviors at least once within the past year compared with 42% of male residents (47 of 113) (Figure). All 70 residents identifying as Asian reported experiencing inquiries into ethnic origins, and 99% (69 of 70) reported being confused with team members of the same race/ethnicity at least once within the past year (eTable 1 in the Supplement). All 115 residents identifying as women reported experiencing assumptions of nonphysician status at least once within the past year (eTable 1 in the Supplement). Experiences of refusal of care and requests to change physicians were more common among residents identifying as Black or Latinx—45% (17 of 38) reported encountering these behaviors within the last year compared with 28% of White-identifying residents (31 of 109) (Figure).
In general, physicians reported witnessing all types of biased behavior more frequently than they reported directly experiencing it (eTable 2 in the Supplement). Biased patient behavior directed toward a physician’s race/ethnicity or national identity and gender identity or expression were witnessed very commonly, with 89% of residents (205 of 230) and 74% of residents (169 of 230), respectively, reporting having witnessed at least one instance within the past year. Behavior directed toward a physician’s sexual orientation and Islamic faith were also common, as 42% of residents (95 of 227) and 40% of residents (91 of 228), respectively, reported witnessing that behavior at least once within the past year (eTable 2 in the Supplement). Finally, content analysis of written responses did not reveal any additional types of biased behavior.
In response to biased patient behavior, residents used strategies such as limit setting (67 of 227 [30%]), debriefing with friends or family (80 of 227 [35%]), and debriefing with team members (77 of 225 [34%]) most of the time or always (Table 3). Reporting biased patient behavior to the institution or switching the patient to another team member were uncommon: most residents reported never using those options (85% [191 of 226] and 78% [176 of 226], respectively). The factors with the greatest negative association with a resident’s ability to respond to biased behavior included the need to prioritize clinical care and a sense of futility in responding, with 34% of residents (76 of 227) and 25% of residents (56 of 227), respectively, identifying these factors as having significant impact (Table 4). Content analysis of written comments also revealed a few additional responses; most commonly, a witness in the room addressed biased behavior. Rarer responses included documenting biased behavior in the electronic medical record and withdrawing from the interaction with the offending patient as much as possible.
Only 32% of residents (73 of 231) were confident or very confident in responding to biased patients (eTable 3 in the Supplement). Most residents (206 of 232 [89%]) identified training for medical students, residents, and faculty on dealing with biased patients as necessary or very necessary; similar numbers believed institutional policies to be necessary or very necessary for addressing biased patient behavior (eTable 4 in the Supplement). With regard to training, 81% of residents (187 of 232) reported receiving 2 hours or less of training on caring for biased patients during residency (Table 1). Although 72% of residents (165 of 230) had prior training on patient bias in medical school, 74% (150 of 203) rated the content of their prior training as below adequate (eTable 5 in the Supplement).
In this survey study of second-year and third-year residents at 3 large, academic internal medicine residency programs, we found that nearly all residents reported witnessing or experiencing biased patient behavior within the past year. The frequency and extent of specific biased patient behaviors varied substantially, with belittling comments experienced on a monthly or weekly basis by more than half the residents and bias-based requests to change physicians experienced a few times a year by one-third of residents. Although ethnicity/race or national origin and gender identity or expression were the most commonly targeted sociodemographic characteristics, sexual harassment was particularly prevalent. Most residents do not report these encounters to institutional leadership, preferring instead to respond to incidences of patient bias on their own or debrief with friends, family, and team members.
To address the issue of biased patient behavior, interventions are needed at the institutional and interpersonal levels. The high prevalence of biased patient incidents may be largely unknown by residency program directors and academic medical center leaders as evidenced by the 80% of residents who stated that they never reporting biased patient behavior “up” within their institution. At the institutional level, low use of a formal structure to address and report incidents may indicate either a lack of formal reporting processes, a lack of knowledge of existing processes, or a reluctance to engage in these processes for fear of negative repercussions. This finding should be explored further and clear reporting mechanisms instituted, while recognizing that reporting systems succeed only when institutional culture encourages reporting without fear of retaliation. Specific policies on approaching patients who explicitly reject or refuse care based on sociodemographic bias requires a specialized approach to navigate the ethical and legal implications of such behaviors.1 On an interpersonal level, training on dealing with biased patients should be incorporated into resident and faculty development curricula. Most residents described prior training on biased patient behavior as below adequate, and only one-third felt confident in responding to biased patient behavior. Developing a deeper understanding of residents’ sense of futility in responding to bias will be necessary for effective development and implementation of trainings and policies. Well-facilitated team debriefing should be continued through allocated time and space where residents’ experiences can be acknowledged, validated, and addressed.21 Bystander trainings on supporting targeted colleagues should also be encouraged given how frequently residents reported witnessing biased behavior.
Our results are consistent with those of prior studies of harassment and discrimination of physicians that have found a disproportionate burden on those identifying as women and/or racial/ethnic minorities.17,22-28 We found that Latinx and Black residents experienced all types of biased behaviors more frequently than their White counterparts. More than 40% of Latinx and Black residents experienced the more extreme forms of patient bias—racial epithets, bias-based refusals of care, and bias-based requests to change physician—at least once in the last year. Academic medical centers need to communicate clear policies on physician refusal based on race/ethnicity to their patients and key medical educators, including attending physicians, chief residents, and other front-line supervisors.1
Asian residents reported the greatest prevalence of belittling and demeaning behaviors. All Asian respondents surveyed reported bias-based inquiries into their ethnic origins. Asian residents also experienced more role questioning than White residents. Sociologists have highlighted the “perpetual foreigner” status of Asian Americans, wherein Asian immigrants or their descendants are considered to be not fully US citizens or to be more closely tied to their country of origin than to the United States.29 Given the high prevalence of biased patient behaviors experienced by Asian residents—and the common misconception that Asians are a homogeneous, overrepresented group in medicine (in actuality, many Asian subgroups, such as Cambodian Americans, are underrepresented)30—our data highlight the need to consider the Asian American trainee experience when designing programs to address patient bias. Moreover, such support is particularly urgent given the increase in biased behavior and hate crimes toward Asian Americans in the time of coronavirus disease 2019 (COVID-19).31
Consistent with other reports,32,33 we noted a high prevalence of sexual harassment. Eighty-seven percent of female respondents reported experiencing verbal or nonverbal sexual harassment, meaning that this behavior may well be regarded as a norm within clinical practice. We also found that a common barrier to responding to biased patient behavior is a sense of futility, reflecting a culture that rarely holds abusers accountable. Harassment constitutes one of many gender identity–based inequities in medicine alongside gaps in salary, career advancement, and leadership that will require a concerted effort to address.34
Prior qualitative work has shown that individuals reporting multiple marginalized group identities often encounter uniquely complex, compounded discriminatory behavior.3 This concept of intersectionality is critical to understanding, validating, and addressing the multidimensional experiences faced by individuals who are members of multiple marginalized groups (eg, ability status, race/ethnicity, immigration status, nationality, religion, sex, gender identity or expression, and sexual orientation).35,36 Our ability to examine the frequency of biased patient behavior toward these individuals in our study was limited by small sample size. Larger studies are needed to measure the frequency and nature of bias that trainees belonging to multiple marginalized groups experience in the workplace.
Our study has several limitations. First, survey responses are inherently subject to recall bias. Second, as noted, our sample size was too small to analyze the impact of the intersection of identities or evaluate the experience of respondents with less common social identities, such as those who identified as gender nonbinary. Third, while we surveyed residents in 3 urban areas in 2 geographically distinct regions of the US, our findings may not reflect the experiences of residents in other parts of the country. Despite these limitations, our study is, to our knowledge, the first to evaluate the frequency of specific biased patient behaviors and physician responses to these behaviors. Prior studies have shown that asking about specific behaviors greatly increases reporting of harassment,32 suggesting that prior studies without this survey’s level of granularity may have underreported these behaviors.
In this survey study, resident physicians’ reported that experiences with biased patient behavior were common, varied both in the type of behavior and the intended target, and were disproportionately prevalent for women, Latinx, Black, and Asian residents. Despite the diversification of the physician workforce and the steady work to decrease bias and discrimination within the workplace, there has been a lack of training and policies to address the social, emotional, and clinical challenges that biased patient behavior produces. Addressing biased patient behavior will require medical institutions to bring forth training, policies, and cultural change that support their trainees and workforce and promote an environment free from harassment and bias for physicians and other health care workers.
Accepted for Publication: September 7, 2020.
Published: November 23, 2020. doi:10.1001/jamanetworkopen.2020.21769
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 de Bourmont SS et al. JAMA Network Open.
Corresponding Author: Alicia Fernandez, MD, Department of Medicine, University of California, San Francisco, PO Box 1364, DGIM Zuckerberg, San Francisco General Hospital, San Francisco, CA 94143 (email@example.com).
Author Contributions: Ms de Bourmont and Mr Burra had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Ms de Bourmont and Mr Burra contributed equally as co–first authors.
Concept and design: All authors.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: de Bourmont, Burra, El-Farra, Mohottige, Fernandez.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Nouri, Friedman.
Obtained funding: Fernandez.
Administrative, technical, or material support: de Bourmont, Burra, Mohottige, Schaeffer, Friedman.
Supervision: de Bourmont, Nouri, Schaeffer, Fernandez.
Conflict of Interest Disclosures: Dr Mohottige serves as a member of a cohort of Duke Department of Medicine faculty and trainees who have been trained to offer informal support in response to incidents of bias, harassment, or other unprofessional behaviors and has lectured in grand rounds format regarding the above program (which addresses discrimination, bias, and micro-aggressions and macro-aggressions at Duke University, University of North Carolina–Chapel Hill, and the University of Washington. No other disclosures were reported.
Funding/Support: The Greenwall Foundation funded this study and supported efforts of Ms de Bourmont and Dr Fernandez. Mr Burra was funded by the PROF-PATH Program, University of California, San Francisco Latinx Center of Excellence (HRSA D34HP31878). Dr Nouri was funded in part by a National Research Service Award fellowship training grant (T32HP19025). Dr Mohottige was funded by training grant T32-DK007731-22 from the National Institutes of Health, National Institute of Diabetes and Digestive Kidney Diseases. Dr Fernandez was funded by grant K24DK102057 from the National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases.
Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Create a personal account or sign in to: