Patient Perspectives on Racial and Ethnic Implicit Bias in Clinical Encounters: Implications for Curriculum Development

Abstract

Objective

Patients describe feelings of bias and prejudice in clinical encounters; however, their perspectives on restoring the encounter once bias is perceived are not known. Implicit bias has emerged as a target for curricular interventions. In order to inform the design of novel patient-centered curricular interventions, this study explores patients’ perceptions of bias, and suggestions for restoring relationships if bias is perceived.

Methods

The authors conducted bilingual focus groups with purposive sampling of self-identified Black and Latino community members in the US. Data were analyzed using grounded theory.

Results

Ten focus groups (in English (6) and Spanish (4)) with N= 74 participants occurred. Data analysis revealed multiple influences patients’ perception of bias in their physician encounters. The theory emerging from the analysis suggests if bias is perceived, the outcome of the encounter can still be positive. A positive or negative outcome depends on whether the physician acknowledges this perceived bias or not, and his or her subsequent actions.

Conclusions

Participant lived experience and physician behaviors influence perceptions of bias, however clinical relationships can be restored following perceived bias.

Practice Implications

Providers might benefit from skill development in the recognition and acknowledgement of perceived bias in order to restore patient-provider relationships.

1. Introduction

In North America, after several decades of focus on cultural competency instruction in medical education [1–3], health disparities persist [4, 5] and racial and ethnic minority respondents are still more likely to perceive bias when seeking medical treatment than Whites [6]. Implicit bias refers to the unconscious and unintentional assumptions people make about each other. Evidence demonstrates this bias negatively impacts patient’s perceptions of the clinical encounter [7–9] treatment recommendations [10], and trust [11–14]. Although studies from various countries have explored patients’ perceptions of race and/or ethnicity and bias in medicine [6, 12, 14–25], patient perspectives on and suggestions for restoring the clinical and/or therapeutic relationship once bias is perceived are not known. Understanding these perspectives could inform the development of innovations in medical student education addressing implicit bias in clinical encounters.

Implicit bias contributes to health disparities through medical decision-making and interpersonal communication [26]: Evidence demonstrates the influence of physician implicit bias on patient perspectives of encounters [7–9], physician communication patterns [9, 27], clinical outcomes [28], and clinical decision-making [29–33]. Implicit bias regarding race has been demonstrated in medical students [34, 35]. In an effort to decrease physician contributions to health disparities, curricula have been developed to teach about implicit bias across the continuum from undergraduate to graduate and continuing medical education [36–43]. These curricular models have not explicitly provided instruction on detecting the perceptions of bias by patients during the patient-physician encounter, nor in skills to restore the encounter once bias is perceived. Our study addresses this gap in knowledge by exploring patients’ perceptions of physician bias and their suggestions for restoring the relationship and the encounter once bias is perceived.

This study is part of a comprehensive needs assessment [44] to inform curriculum development for medical students in implicit bias recognition and management. Patients represent the intended beneficiaries of any successful future curricular interventions, therefore it is critical to maintain a patient-centered perspective [45] in the development of novel curricular interventions. To inform the design of future patient-centered curricular interventions in implicit bias recognition and management, the purpose of this study is to explore patients’ perceptions of physician bias and their preferences and suggestions for restoring the relationship if they perceive bias.

2. Methods

Given that little is known about patient perspectives on implicit bias we conducted an exploratory focus group interview study using grounded theory, a systematic qualitative methodology involving the discovery of theory through the analysis of data [46]. Recruitment, data collection, and analysis were conducted iteratively to fully capture and explore variation in patients’ perspectives. All aspects of the study were approved by the Institutional Review Board of the Albert Einstein College of Medicine.

2.1. Sample

Participants were selected through purposive sampling, a useful way to select study subjects that will ensure “information-rich cases for study in-depth [47].” To be eligible, community members spoke English or Spanish, were aged 18–90, had sought medical care for themselves or their child(ren) in the previous year, and lived or sought medical care in New York City, NY, USA. Two investigators recruited participants from community board meetings in Bronx, NY, USA, a borough of New York City. There are eleven community boards in the Bronx representing various neighborhoods. All residents represented by a given community board and who met our study inclusion criteria were eligible to attend. Investigators also pursued referrals from colleagues. Participants were selected to span the socioeconomic spectrum within the United States (US). We sampled for participants who self-identified as Black (African American, Caribbean, and African), and Latino (US-born and immigrants), or as both Black (race) and Latino (ethnicity).

2.2. Interview guide development

We developed the interview guide (Appendix) based on review of the literature related to racial discordance, trust, and discrimination in clinical encounters [6–25]. Questions focused on racial and ethnic bias, our construct of interest. It was supplemented by our clinical experiences (e.g. patient anecdotes of mistreatment perceived related to race/ethnicity). We revised items through discussion among the investigative team until all investigators agreed to the final questions.

In the US there have been historical differences in societal acceptance and social status between White and racial and ethnic minority populations. Therefore, our open-ended questions explored patient perspectives on how they were treated and/or judged by both individual providers and within the healthcare system, the consequences of those experiences, and their suggestions for actions physicians can take to restore the relationship if patients perceive bias. The questions served as a starting point for the discussion, and facilitators were able to probe unanticipated lines of discussion that occurred in the focus groups.

2.3. Data Collection

Focus groups were conducted in English or Spanish with participants in community settings. The bilingual PI (CMG) followed a semi-structured interview guide and a bilingual research assistant (MLD) took field notes of nonverbal behaviors. Focus groups were digitally recorded and professionally transcribed. Spanish focus groups were professionally translated and transcribed. Investigators cross-referenced the transcripts to the audio to check for accuracy. Focus groups continued until data analysis demonstrated we had reached thematic saturation, i.e., no new concepts in subsequent focus groups emerged [48]. Participants received a meal and a $25 gift card. Written, informed consent was obtained.

2.4. Analysis

We conducted the data analysis in three phases [49]. The first phase was to develop the codebook. Three investigators (CMG, MLD, EK) independently read three transcripts each line-by-line to identify phrases that related to patient perspectives on implicit bias. These phrases were discussed and consensus reached on a list of codes and their definitions to create the preliminary codebook. This codebook was applied to three more transcripts and further refined after discussion. During the second phase, the codebook was used to code the remaining transcripts, which were coded independently by two investigators each. Using inductive reasoning, the investigators began with low inference codes, discussed their meaning, and developed conceptual themes. Finally, they met to discuss the relationships between themes and reach consensus on representative quotes. Once the final themes and their representative quotes were identified, these data were presented to select participants to ensure accurate representation of their perspectives, for member checking [50].

3. Results

We conducted ten focus groups, six in English and four in Spanish, with N=74 participants. Demographic data demonstrated successful sampling across the socioeconomic spectrum of the US (Table 1). Our analysis identified four themes relating patient experiences with discrimination to perceptions of bias in their physician encounters, the outcomes of perceived bias, and suggestions for physician actions to restore the relationships within such encounters when bias is perceived.

Table 1.

Demographic data of participants in focus group study exploring perceptions implicit bias in clinical encounters.

Demographic Data	N=74
Self-identified gender
Female	44 (59%)
Male	28 (38%)
Transgender M to F	2 (3%)
Age
Mean	41.8 years
Range	18–81 years
Self-identified race/ethnicity
Latino	39 (51%)
US born	12/39
Non-Hispanic Black	27 (35%)
African American	24/27
Non-Hispanic White	1 (1.3%)
Preferred language
English	48 (65%)
Spanish	26 (35%)
Annual household income
Mean	$57,356
Range	$5000–$300,000
Educational attainment*
Mean	Some college
Range	Less than high school to doctoral degree
Residency
Bronx	57 (77%)
Number of medical problems
Mean	1.04
Range	0–4

^*Educational attainment was scored as 1 = less than high school, 2 = high school, 3= some college, 4 = bachelor’s degree, 5= master’s degree, 6 = doctoral degree. Participants had an average educational attainment of 3.04, equivalent to some college.

3.1. Racism/discrimination is exhausting

Participants discussed their experiences with racism within society and the healthcare system and voiced frustration with the ubiquitous nature of bias. Previous experiences with both explicit racism and subtle slights were common to many of our participants.

One participant described an example of such a slight:

“In my profession it happens all the time because in New York City most attorneys aren’t of color. So when you come into the court…they usually think that you’re a litigant, not an attorney. That happens often.” [Latino Man]

Patients may be experiencing subtle slights in the form of bias and discrimination in their day to day lives, potentially affecting their perceptions of bias when interacting with the healthcare system. Being referred to other physicians and not being able to run back in and see the doctor if they forgot something during their appointment are examples of factors that affect all patients. These routine behaviors have more severe consequences when related to being a member of a racial or ethnic minority rather than White.

“Then when I told the nurse please get me a doctor—I couldn’t walk, I was bent over—this doctor that came to my bed, ‘Oh Miss [NAME] who is the president’. I said, ‘I know who the president is. I know what you’re doing. I didn’t call for a head doctor. I called for a doctor.’” [Black Woman]

Frequent experiences of discrimination may lead to patient perceptions of bias in otherwise routine clinical practice behaviors.

Discrimination based on limited English proficiency was a commonly voiced concern among the Spanish-speaking participants.

“The first thing they ask me, ‘Do you speak English?’ That’s where the discrimination steps in, that’s where we begin. Where I tell you that this woman [the doctor] wants to hit me because I don’t know English.” [Spanish Speaker]

Members of minority groups who perceive discrimination in the course of their routine day may be more primed to notice non-verbal behaviors and identify them as biased behaviors [51].

Participants described instances in which they felt they were treated differently based on their race, ethnicity, or the language they spoke. Frequently described behaviors included varying levels of courtesy, respect, how gently patients were treated if they were in pain, and how efficiently they were seen by the physician.

“When the doctor came in [after a surgery] she proceeded to show me how I had to get up because I’m being released that day “whether I like it or not”…She yanked the first snap on the left leg…So I’m thinking, ‘I’m human!’ And she was courteous to the White lady [in the next bed], and I’ve got just as much age as her. I qualify on the level and scale of human being as her, but I didn’t feel that from the doctor.” [Black Woman]

The suggestion of disparate treatment affecting the core of humanity is profound and adds an additional challenge for patients in already difficult situations.

In response to these insults to their humanity some participants utilize compensatory strategies.

“You need a doctor. That’s why you tolerate it. I go [to the encounter] in a very nice way in order to get the same response from them [as compared to a White patient]. And I practically make a face, like a sad face, to get them to feel pity for me and to get treated well and to avoid conflict. Because if not, I will lose.” [Spanish Speaker]

Patients may avoid conflict with their physician because they perceive a power imbalance and prioritize their need for medical care [52]. Our participants suggest they are even willing to bear an additional burden and overcompensate in order to avoid potential conflicts in anticipation of perceived bias and racism in their medical encounters.

3.2. Meaning of privilege in society

Participants identified effects of privilege based on race or ethnicity, socioeconomic status, language, and documentation status for immigrants. They felt that in clinical situations, from checking in at the front desk to the outcome of the visit with the physician that they would have been treated differently if they were White.

“I was in the ER, and a guy came in…He was Caucasian and I do not know what his medical problem was, but I know that he was in and out of there within an hour. Triage, doctors, everything. We are just sitting here, all the Black people, Puerto Rican people, we are just looking at each other, ‘Did you see that?’ It was unbelievable... Almost started a riot in there…This is Manhattan. Wow, that is all it takes huh?...I have got to bleach my skin then.” [Latino Man]

Other participants described perceived differential treatment prior to seeing the physician in the office setting.

“You ask [the front desk staff], ‘Where do I sign in?’ You get dismissed. They’re like, ‘I’m on the phone,’ or whatever. Then you turn around for a second and you have a Caucasian that comes in and they are like, ‘Hello, how can we help you?’” [Black Man]

Patients may feel slighted within healthcare settings prior to actually seeing the physician. Such experiences may have a negative synergistic effect with discrimination in society and prime patients to perceive bias in their physician encounters.

Participants who identified themselves as having privilege based on socioeconomic status and education lamented their perceived lower status when interacting with the healthcare system.

“They put me sort of in the corner [in the ER] and I can’t talk very well because I can’t breathe so well. The nurse comes over to me and actually says, ‘Tu tiene tu Medicaid?’ I whispered out, ‘I’m a doctor…and I have insurance.’ I said it in perfect English. Literally, the color on her face went completely white, like whiter than it already was…Within two minutes there was an orthopedic team around me…I kept wondering about what if I hadn’t been a doctor, you know? Pretty eye opening and very sad.” [Latino Man]

Different types of privilege may not always confer the same benefits. The protective effect of socioeconomic privilege may be nullified when minority patients are being judged solely on their minority status.

3.3. Perceived bias influences the outcome of present and future clinical encounters

Many behaviors that were perceived as bias or discrimination were based on things physicians and staff should be doing for all patients, such as showing respect, having common courtesy, avoiding jargon, and being patient centered. Physician and staff behaviors, both verbal and nonverbal, contributed to the participant’s perception of the presence or absence of bias in a clinical encounter.

“Sometimes the doctor will see you walking in and as soon as they see who you are, their head goes down. Like you have to say ‘Excuse me…’” [Black Woman]

Patients can be sensitive to verbal and nonverbal cues to bias resulting in feeling dismissed or belittled by physicians, nurses, and support staff. Reactions for many of our participants included, “Well what’s wrong with me?;” “I was feeling like he was trying to belittle me and my intellect;” “So, I’m thinking like I’m human;” and “Well, to me he embarrassed me.” These perceptions and their effects demonstrate the centrality of the effects of the dehumanizing nature of bias and discrimination.

When participants did perceive bias in a clinical encounter, trust in their physician suffered, they delayed seeking medical care, or they subsequently avoided medical care altogether, as one participant describes in her encounter with a surgeon.

“You know what I did? I got up, put my coat on, went to the reception desk, and told them, ‘Take my name off, I’m never coming back here.’ I had a kidney stone. I told my primary doctor, she got the paperwork done for me and I went to another doctor in a week. Yeah, that kidney stone was kicking my butt.” [Black Woman]

Perceived bias can have significant consequences. Patients may even endure painful conditions in order to preserve their dignity.

3.4. Restoring the relationship when bias is perceived

Participants fondly remembered behaviors including respectful communication, rapport building, advocacy behaviors, and appropriate utilization of interpreter services. When bias was not perceived in an encounter, participants felt more trust in their physicians and the medical system in general. Participants made suggestions for preserving the patient-physician relationship when they did perceive bias in an encounter. Although many participants described events that affected them deeply, most just wanted the incident to be acknowledged and were satisfied with an apology from the physician. Many participants felt that the acknowledgement of an incident of biased behavior (or the perception of biased behavior) followed by an apology was the most important thing. Ignoring or dismissing it could lead to the negative outcomes described above.

“I’m sorry. I apologize if I made you feel any kind of way. Will you accept my apology? Just be joking with it because most of the time we as patients we want the lightness because we’re here in a heavy moment and everything is so heavy. We appreciate that little jokey thing.” [Black Woman]

Suggestions for improving the clinical encounter frequently centered around open communication, courtesy, and respect. Participants also made suggestions addressing nonverbal behaviors that can be perceived as bias.

“Acknowledge me. Don’t just acknowledge the paperwork that the nurse brought in. Don’t just go to the computer. Acknowledge me. ‘How are you Miss [Name]?’ So just common courtesy.” [Black Woman]

Spanish speaking participants shared similar ideas as English speaking participants for restoring the encounter whether the biased behavior was deemed relative to race/ethnicity or language.

“‘Excuse me. I’m sorry,’ you know? We can all make mistakes.” [Spanish speaker]

4. Discussion and Conclusion

4.1. Discussion

Our analysis revealed the multi-faceted nature of the factors leading to patients perceiving bias in their physician encounters. Their perceptions of bias may be due to actual biased behaviors or interpretations of routine proceedings as based on these multiple factors. The theory emerging from our analysis suggests the outcome of the clinical encounter once bias is perceived can still be positive. Whether the outcome is positive or negative depends on whether the physician acknowledges this perceived bias or not, and his or her subsequent actions. The identification of this opportunity to restore the relationship within the clinical encounter suggests a viable target for future curricular innovations (Figure 1).

Figure 1. Conceptual model for patient perceptions of bias in clinical encounters and the potential consequences.

This conceptual model represents the contributions the patient lived experience to the presence or absence of perceptions of bias in the clinical encounter. If bias is perceived, outcomes may change depending on whether the physician acknowledges this perceived bias or not. The role of the curricular intervention would be to enable the recognition of real and/or perceived bias and help learners recognize and acknowledge it, to improve the outcomes of the clinical encounter.

Our participants’ lived experiences often include explicit racism, perceived or real incidents of discrimination, and/or subtle slights. Subtle slights are termed microaggressions. Microaggressions are the “brief and commonplace daily verbal, behavioral, and environmental indignities, whether intentional or unintentional, that communicate hostile, derogatory, or negative racial, gender, sexual-orientation, and religious slights and insults to the target person or group [53].” The cumulative effect of microaggressions can lead to increased perception of discrimination, contributing to decreased trust in physicians, as well as a priming effect for patients to perceive bias [51, 54].

In addition to their lived experience, treatment by support staff prior to meeting the physician can also prime patients to perceive bias from their physician, even before the physician demonstrates any bias. While some physicians may be acting in a racially biased fashion, other generally negative behaviors, including poor communication and interpersonal skills, may be interpreted as racial bias by the patient even if this poor doctoring is offered to all, irrespective of race or ethnicity. Regardless, the perception exists and according to our participants, can lead to negative outcomes of avoiding or delaying medical care, mistrust, and misperception of routine proceedings, all with potentially detrimental effects on health.

Perceptions of experiencing discrimination in society have been associated with delays in seeking medical care and nonadherence to medical recommendations [18]. The importance of nonverbal behaviors and race-based assumptions by the physician have been described by others as contributing to perceived discrimination in clinical encounters, leading to mistrust, negative emotional reactions, changing providers, or avoidance/delay in seeking care [14, 15]. Perceived prejudice by providers along racial, ethnic, and immigration status cause patients to worry that they will receive lower quality of care [17, 19, 20]. Minority patients have described episodes of disrespect, and feeling that they would be treated differently if they were of a different race [6, 21]. Patients with trust in their physician and/or the system are more likely to continue to engage in medical care, continue with the same physician, and adhere to treatment recommendations [55].

The similarities among our participants’ perspectives and those of patients and participants in other studies demonstrates the pervasive nature of bias, discrimination, and racism affecting patients. These experiences may have an effect on the clinical encounter. Our study extends the prior work of patients’ perceptions of bias by eliciting participants’ suggestions to mitigate the effect of bias in clinical encounters and restoring the relationship. Restoring the relationship, from our participants’ perspective can lead to the same outcome as never having demonstrated bias in the first place. Accounting for this perspective may enhance efforts to provide patient-centered care and decrease health disparities. It has implications for the design of novel curricular interventions in implicit bias recognition and management.

4.2. Practice Implications: Implications for curriculum development

From our data analysis we have developed a conceptual framework (Figure 1) that could inform curriculum development delivering instruction in the knowledge, attitudes, and skills necessary to recognize and manage racial and ethnic implicit bias in clinical encounters. Given the influence of patients’ lived experiences on perceptions of bias as articulated by our participants, it may benefit students to participate in perspective taking exercises [56], and other exercises meant to build empathy. Additionally it may help students to realize that the patients perception of bias is not always a direct reflection of provider (e.g. student, physician, etc.) actions within a clinical encounter. This realization may help mitigate the ego-dystonic reaction [57] that being accused of being biased may engender. For students working hard with good intentions to care for their patients, having implicit bias would be contrary to their perceptions of themselves as good people who would do the right thing. If they are being accused of acting in a biased way, when in fact, they were not, they may become defensive and be unable to effectively care for the patient. If, however, students acquire knowledge of the lived experience’s influence and the potential for their patients to be primed to perceive bias, they may be able to acknowledge the perception of bias, and then mitigate its influence in the clinical encounter. Even if students do act in a biased manner, the relationship can still be restored.

A complete absence of bias or biased behavior is unrealistic, and according to our participants, unnecessary. Participants stressed the importance of physicians acknowledging biased behavior and recognizing and managing implicit bias in clinical encounters. Instruction could be designed that moves students from awareness of their implicit biases to skill building so that they could achieve this. The theory emerging from our analysis suggests that a curriculum providing: 1) instruction focusing on the influence of the patient’s lived experience will enhance students’ ability to partner with their patients to acknowledge perceived bias and 2) skill development in recognizing both perceived and real implicit bias in an encounter will enable students to mitigate its effect on the clinical encounter leading to continued patient engagement in the present and future medical encounters.

4.3. Limitations

Our study has some limitations. We sampled in a large urban center in the United States, purposively for Black and Latino participants; other geographic areas and racial, ethnic, and language minorities may have different perspectives. Bias along with gender, sexual orientation, disability status, among others, may affect the clinical encounter in different ways. Therefore we might have missed the intersectionality of the discrimination as well as the possibility of not uncovering issues related to other races and ethnicities. Given the preponderance of data on physician racial and ethnic bias and its impact on communication patters and clinical decision-making, and the racial and ethnic diveristy of Bronx, NY, USA, we focused on racial and ethnic bias.

4.4. Conclusion

In a patient-physician dyad it is important to consider the dual contributions of the physician’s potential implicit biases, and the patient’s lived experience, which may augment his or her perceptions of bias during the clinical encounter. Given that patients globally perceive negative aspects of some clinical encounters to be attributed to their race or ethnicity, and the potential influence of physician implicit bias in clinical encounters, medical educators worldwide could consider curricular interventions to improve the outcomes of the patient-physician dyad. Developing curricula to address this interplay in clinical encounters may enable students and physicians at any stage of training and practice to mitigate the effect of their implicit bias and contribute to the reduction of healthcare disparities. For example, improved communication patterns may have immediate and downstream effects as patients would then be more likely to trust their physicians and remain engaged in care [58, 59]. To our knowledge, no published interventions exist to help physicians at any stage of training to develop the skills to recognize and manage their implicit bias in clinical encounters. Further research is required to develop such interventions. The patient voice is an important part of this area of investigation. Understanding their perspective and valuing their input can shape curricular interventions in meaningful ways. Our conceptual model highlighting the effect of the patient’s lived experience, the importance of the physician’s ability to acknowledge perceived bias, and the potential for positive outcomes even after the perception of bias may facilitate the design of future interventions with the ultimate goal of providing exceptional, equitable care to all patients.

I confirm all patient/personal identifiers have been removed or disguised so the patient/person(s) described are not identifiable and cannot be identified through the details of the story.