Delving Below the Surface

Understanding How Race and Ethnicity Influence Relationships in Health Care


  • Lisa A. Cooper MD, MPH,

    1. Welch Center for Prevention, Epidemiology, & Clinical Research, Johns Hopkins University, Baltimore, MD, USA
    2. Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, USA
    3. Department of Health Policy & Management, Johns Hopkins University Bloomberg School of Public Health, Baltimore, MD, USA
    4. Department of Epidemiology, Johns Hopkins University Bloomberg School of Public Health, Baltimore, MD, USA
    Search for more papers by this author
  • Mary Catherine Beach MD, MPH,

    1. Welch Center for Prevention, Epidemiology, & Clinical Research, Johns Hopkins University, Baltimore, MD, USA
    2. Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, USA
    3. Department of Health Policy & Management, Johns Hopkins University Bloomberg School of Public Health, Baltimore, MD, USA
    Search for more papers by this author
  • Rachel L. Johnson MD, PhD,

    1. Department of Pediatrics, Johns Hopkins University School of Medicine, Baltimore, MD, USA
    Search for more papers by this author
  • Thomas S. Inui MD, ScM

    1. Regenstrief Institute, Indiana University Medical Center, Indianapolis, IN, USA.
    Search for more papers by this author

  • The authors have no conflicts of interest to report. This work was presented in part at the Ninth Regenstrief Conference, “Re-Forming Relationships in Healthcare, September 29 to October 1, 2004, Turkey Run State Park, Indiana.

Address correspondence and requests for reprints to Dr. Cooper: Welch Center for Prevention, Epidemiology, & Clinical Research, 2024 East Monument Street, Suite 2-500, Baltimore, MD 21205-2223 (e-mail:


There is increasing evidence that racial and ethnic minority patients receive lower quality interpersonal care than white patients. Therapeutic relationships constitute the interpersonal milieu in which patients are diagnosed, given treatment recommendations, and referred for tests, procedures, or care by consultants in the health care system. This paper provides a review and perspective on the literature that explores the role of relationships and social interactions across racial and ethnic differences in health care. First, we examine the social and historical context for examining differences in interpersonal treatment in health care along racial and ethnic lines. Second, we discuss selected studies that examine how race and ethnicity influence clinician-patient relationships. While less is known about how race and ethnicity influence clinician-community, clinician-clinician, and clinician-self relationships, we briefly examine the potential roles of these relationships in overcoming disparities in health care. Finally, we suggest directions for future research on racial and ethnic health care disparities that uses a relationship-centered paradigm.

Webster's Dictionary defines “relationship” as “the state of being … connected through mutual interests or involvement.”1 Therapeutic relationships are the central interpersonal milieu in which patients are diagnosed, given treatment recommendations, and referred for appropriate tests, procedures, or care by consultants in the health care system. Yet, in health services research, relatively few studies focus on the role of interpersonal relations and social interactions in explaining racial and ethnic disparities in health.2 Relationship-centered care, an important new framework for conceptualizing health care, is founded upon 4 principles: (1) that relationships in health care ought to include the personhood of all participants; (2) that emotions and their expression are important components of these relationships; (3) that all relationships occur in the context of reciprocal influence; and (4) that the formation and maintenance of genuine relationships in health care is morally valuable.3 While relationship-centered care shares many of the same principles as patient-centered care (care that is closely congruent with and responsive to patients' values, needs, and preferences),4 it is a broader conceptualization of health care, in that it considers the unique experiences, values, and perspectives of patients, clinicians, and all other participants in the health care process. It also focuses on the relationships between and among these participants at several levels: clinician-patient, clinician-community, clinician-clinician, and clinician-self (awareness of the clinician's own attitudes and experiences and their impact on interactions with others in the context of health care).

This paper provides a review and perspective on the literature that explores the role of relationships and social interactions across racial and ethnic differences in health care. Our discussion is mainly focused on the clinician-patient relationship as it has been studied the most. However, we acknowledge that race and ethnicity also impact clinician-community, clinician-clinician, and clinician-self relationships and that these relationships in turn influence disparities in health care quality (Fig. 1). We provide a brief discussion of the potential role of relationship-centered care in overcoming racial and ethnic disparities in health care and conclude with directions for future research on racial and ethnic disparities in health care using a relationship-centered paradigm.

Figure 1.

 Dimensions of relationship-centered care with a potential link to racial and ethnic health care disparities.

The Social and Historical Context for Disparities in Health Care

Racial and ethnic disparities in health care quality have been extensively documented.5 Much of this literature has focused on technical aspects of health care, such as whether or not patients receive appropriate tests, procedures, or medications. However, there is also increasing evidence that racial and ethnic minority patients receive lower quality interpersonal care than white patients.6 In 2002, the Institute of Medicine report Unequal Treatment7 confirmed that racial and ethnic disparities in health care are not entirely explained by differences in access, clinical appropriateness, or patient preferences, and suggested that disparities in health care exist in the broader historical and contemporary context of social and economic inequality, prejudice, and systematic bias.

Perhaps because of their lengthy presence in American society, health and health care disparities are most documented for African Americans relative to whites.8 However, the literature that documents disparities in health and health care for other ethnic minority groups has also grown. To the extent that the term ethnicity describes national origin and language in addition to culture and social status, its role in producing disparities may be distinct from that of race. The latter term has been used throughout U.S. history to de-humanize African Americans in particular.9–11 For example, while there is evidence that language proficiency, socioeconomic status, and acculturation can explain disparities in health and health care for many Hispanic populations,12 racial disparities for African Americans persist despite adjusting for factors such as socioeconomic position.12 Nevertheless, the consistency of patterns of disparity in different aspects of society supports the argument that a common underlying set of mechanisms exists through which race and ethnicity affect inequalities in health care and health status.13–15

The social environment is one such set of mechanisms. It includes socioeconomic factors (e.g., employment and education), physical surroundings (e.g., neighborhood and work conditions), social relations within one's community and/or workplace, and power arrangements (e.g., political empowerment, individual and community control, and influence).14 A detailed examination of the myriad of ways in which the social environment might impact race relationships in health care is beyond the scope of this paper. However, there is substantial evidence to suggest that the social meaning attributable to race and ethnicity within a given societal context is at the root of its largest effects on health and health care.16–18 We submit that race and ethnicity are both socially constructed and have unique societal and individual meanings, and that the health implications of race and ethnicity result primarily from their effects on social interactions rather than on biology.

Race/Ethnicity and the Clinician-Patient Relationship

To guide our selection of the aspects of interpersonal care that might be most relevant to clinician-patient relationships across racial and ethic differences, we reviewed 2 conceptual articles. First, we reviewed van Ryn's model of hypothesized mechanisms through which clinician factors might influence racial and ethnic disparities in patient care. This model incorporates clinicians' beliefs about patients as antecedents to clinicians' interpersonal behavior; it also incorporates patient cognitive and affective factors as antecedents of patient behavior and outcomes of clinicians' interpersonal behaviors in encounters.19 Second, we reviewed the conceptual framework proposed by Stewart et al.20 for understanding interpersonal processes of care in diverse populations from the patient's perspective. This framework distinguishes 3 dimensions, each with multiple domains: communication, decision making, and interpersonal style. With these dimensions in mind, we focused on studies of the impact of race or ethnicity on the following attitudes and behaviors reflected in interpersonal care: communication, partnership, respect, knowing, affiliation or liking, and trust. Additionally, as concordance (whether defined as a state of agreement or as shared identities between persons) is often a core aspect of successful relationships, we also briefly reviewed findings from studies of racial or ethnic concordance in clinician-patient relationships.

Communication. All relationships are made possible through communication between and among participants, and communication is the behavioral action through which the other features of relationships are observed. Few studies have used direct observation of physician communication behaviors to examine how patient race and ethnicity influence physician interpersonal behaviors.21–23 In their interactions with African-American patients, physicians have been shown to exhibit less nonverbal attention, empathy, courtesy, and information giving,21 to adopt a more “narrowly biomedical” communication style,22 to spend a lower proportion of time intervals providing health education, chatting and answering questions,23 and to be more verbally dominant and exhibit more negative emotional tone24 than with white patients. These studies are consistent with social cognition studies that document negative interpersonal behaviors in interracial interactions.25,26 Because reciprocal influences exist in communication, it is not surprising that African-American patients have been shown to ask physicians fewer questions, provide less information when asked questions, seek less clarification of information provided by physicians,27 and exhibit less positive emotional tone in their visits with physicians.24 With a few exceptions,28–30 data regarding communication between physicians and Hispanic and Asian Americans are limited to patient and physician reports of language barriers and abilities or physicians' inattentiveness.31–33 A recent national survey of household respondents showed that a higher percentage of African Americans, Asians, and Hispanics than whites reported at least 1 communication problem when seeing a physician.34

Partnership. A partnership between patient and clinician is a union that recognizes and values the unique perspective, knowledge, and opinion of each participant. Strong evidence links partnership (participatory or shared decision making) to positive patient outcomes such as adherence, satisfaction, and improvements in health status.35 Racial/ethnic minority patients are less actively engaged in partnerships with their physicians. In 1 small study, Latina mothers of developmentally disabled children expressed that physicians ignored their own expertise about their children and treated them as if they did not know anything.36 Similarly, large studies of primary care patients have found that ethnic minority patients rate their physicians as significantly less participatory than do white patients.34,37,38 Participatory decision making is strongly and significantly related to satisfaction across all racial and ethnic groups, suggesting that all patients of all racial and ethnic groups would like physicians to allow them to participate in medical decision making.38

Respect. Respect for persons, defined as the recognition of each person's inherent value, might be considered the cornerstone of all human interactions. Disrespect, devaluing, and biased treatment have historically occurred and continue to be directed at persons of racial/ethnic minority backgrounds. Given this broader societal context, it is not surprising to find that racial/ethnic minority patients perceive themselves as being treated with disrespect in health care settings. In a small study, Latina mothers felt that professionals pathologized their children and did not recognize the children's unique strengths and abilities (something the mothers considered essential for establishing a good relationship). Mothers also complained that professionals were rude, rushed them through meetings, and treated them “like dirt” and as if they were wasting their time.36 The Commonwealth Fund's 2001 Healthcare Quality Survey found similarly that African Americans and Hispanics were the most likely to feel that they had been treated with disrespect and Asian Americans were the most likely to feel they had been looked down on by health professionals.34 Even after controlling for potential confounders, Hispanics, Asians, and African Americans were all more likely than whites to believe that they would have gotten better care if they belonged to a different race/ethnic group and that medical staff judged them or treated them unfairly based on their race.39

Knowing. To know another person is to be familiar with them and their unique life story. To be in relationship with someone, one must know them to some degree. Knowing in the patient-physician relationship is important; in a large study of outpatients, patients' perception that their physician “knows them as a person” was more highly correlated with patient adherence to medical treatment than any other dimension of care measured.40 Stereotyping, in which group characteristics are ascribed to individuals, is a cognitive shortcut that one takes precisely when one does not know another person as an individual. Stereotyping is more likely to be used as a cognitive shortcut in busy clinical settings when time is short. Given the concern about stereotyping of racial/ethnic minorities, the clinician's “knowing” of a patient may be a particularly important aspect of relationships in health care for racial/ethnic minority patients. In 1 study, physicians rated African-American patients more negatively with regard to intelligence and educational level, and likelihood to comply with medical advice and to abuse alcohol or other drugs, even after controlling for patients' self-reports of many of these same variables.41 In the study of Latina mothers described earlier, mothers had the pervasive belief that professionals were often lacking in intimate personal knowledge of the child and family and could not really be helpful.36 In the Commonwealth Fund's 2001 Healthcare Quality survey, Asian Americans, but not African Americans or Hispanics, were less likely than whites to feel that their doctor understood their background and values.34,42

Affiliation/Liking. Affiliation is the degree to which a person feels a shared identity with another person. Liking, a related but distinct concept, is often based on the degree to which a person produces pleasure or enjoyment and is dependent on the subjective whim of the person who likes. Few studies have examined liking and affiliation in the context of the patient-physician relationship.43,44 We are only aware of 1 study that examined physician liking and affiliation (using a measure that seems to combine the 2 concepts) between patients and clinicians of different races and found, after controlling for other patient characteristics, that physicians were less likely to rate black compared with white patients as someone with whom they could see themselves as friends.41 This is concerning for the quality of relationships and for health care disparities, as patients who are not liked by their physicians have received differential treatment decisions45 and poorer medical care.46,47

Trust. Interpersonal trust, a fundamental component of all relationships, depends on the degree to which people see one another as competent, responsible, caring, tactful, and ethical.48 It also includes other dimensions such as fiduciary responsibility, confidentiality, and trustworthiness.49,50 Trust in one's physician has been linked to satisfaction, adherence to recommended treatment, use of preventive services, continuity of care, and self-rated health.40,51–55 Many studies show that racial and ethnic minorities have less trust in physicians, researchers, and in the health care system.34,55–59 In 1 small qualitative study, African-American patients with positive cardiac stress test results expressed a preference for building a relationship with physicians (trust) before agreeing to an invasive cardiac procedure, and consistently complained that trust was lacking.60 Although some studies fail to show such an association,34,42,61 the fact that other important elements of the patient-physician relationship are impacted negatively for racial and ethnic minority patients makes it possible that trust is also impaired.

Racial and Ethnic Concordance. Ethnic minority patients are frequently treated by professionals who differ from them in racial or ethnic background in the so-called “race-discordant” relationships. The term “concordance” has been used to indicate shared identities between patients and clinicians.62,63 There are visible demographic characteristics (age, gender, social class, ethnicity, race, language) and relatively invisible characteristics below the surface or tip of the “cultural iceberg” (beliefs, attitudes, values, preferences, and role orientations) across which patients and clinicians may have concordance (Fig. 2). Researchers have developed a new line of inquiry that seeks to understand how patients and physicians relate with each other across these similarities and differences.

Figure 2.

 Iceberg concept of culture applied to race relations in health care.

Several studies suggest that racial/ethnic concordance between patients and physicians is positively related to partnership, respect, and communication. A telephone survey of over 1,800 adult managed care enrollees attending primary care practices in a large urban area found that patients in race-concordant relationships with their physicians rated their physicians as significantly more participatory than patients in race-discordant relationships.38 A nationally representative survey study found that black respondents with black physicians were more likely than those with non-black physicians to rate their physicians as excellent in treating them with respect, explaining problems, listening, and being accessible to them.64 Another study showed that patient-provider racial concordance accounted for the gaps in ratings of respect and satisfaction between whites and African Americans.65 A recent study that used measures of actual communication behaviors of physicians and patients found that race-concordant visits were longer and had higher ratings of patient positive affect by independent observers than race-discordant visits.66 In this study, patients in race-concordant visits were also more satisfied and rated their physicians as more participatory.66 Similarly, in a European study, researchers showed that patient-provider ethnic discordance was associated with less social talk and less positive physician affect, lower patient ratings of mutual understanding with physicians, satisfaction, and self-reported compliance.67 Most promising for the future is a study that showed communication problems attributed to race discordance to be diminished over time when there was continuity in the patient-provider relationship.68 However, to our knowledge, there are no data to address whether an association exists between concordance and feelings of affiliation, liking, knowing, and trust.

Race/Ethnicity and the Clinician-Community Relationship

The nature of health care providers' relationships with the communities that they serve remains largely unexplored. However, a significant body of literature examines the individual characteristics of primary care clinicians who provide care to underserved communities. These studies have shown that clinician race/ethnicity, family background, training, and economic factors are related to care of ethnic minority communities. Ethnic minority physicians are more likely to care for patients of their own race or ethnic group, to practice in health care manpower shortage or underserved areas, to care for poor patients, patients with Medicaid insurance or no health insurance, or to care for patients who report poor health status and use more acute medical services such as emergency rooms and hospital care.69–75 Additionally, physicians who expressed a strong interest in practicing in an underserved area prior to medical school and who grew up in an underserved (inner city or rural) area themselves are more likely to provide care to underserved populations.76 In some studies, international medical graduates are also more likely to provide primary care in urban poor and rural underserved areas.77–80 This literature suggests that a systematic commitment from educational programs and health care organizations is needed to produce activated clinicians, teachers, and researchers who can work collaboratively with communities for a more just and egalitarian medical system.

Race/Ethnicity and the Clinician-Clinician Relationship

America's health professional workforce is dominated by white and upper- or middle-class individuals. Until the mid-1960s, health professional schools and organizations remained segregated. Even today, substantial literature documents the disproportionate underrepresentation of African Americans, Hispanics, and American Indians among physicians, nurses, dentists, and other health care providers.81 Relatively less research explores how race/ethnicity affects interactions among members of the health care team such as clinician-clinician relationships. However, there is historical and current evidence that racial and ethnically based biases affect opportunities and practice conditions of health professionals who are members of ethnic or racial minorities.18,82 In 1 study, more than 99% of African-American physicians reported some degree of racial discrimination in the practice of medicine including peer review, obtaining practice privileges at hospitals, hospital staff promotions, Medicaid and Medicare reimbursements, malpractice suits, private insurance oversight and reimbursements, and referral practices of white colleagues.83 In another study, physicians treating black patients were more likely to report that they were unable to provide high-quality care and that they faced greater difficulties in obtaining access to high-quality subspecialty care and services and elective hospital admission for their patients.84 These studies suggest that in addition to structural barriers of inequality, physicians caring for ethnic minority patient populations may also experience interpersonal biases that impact upon their ability to provide patient care.

Race/Ethnicity and the Clinician-Self Relationship

The clinician-self relationship may be characterized as the degree to which an individual clinician is aware of his or her own background, attitudes, and values, and their impact upon behaviors and interactions with others in the context of health care. Perhaps the least is known about this level of relationships in health care. We found 1 small study, in which physicians were trained to identify previously unrecognized, negative attitudes that interfered with learning patient-centered interviewing skills.85 This study showed that increasing self-awareness led to improvement of physicians' interviewing skills. The impact of race on the clinician-self relationship has not been extensively studied. In cultural competence training, the sensitivity/awareness approach uses educational exercises and techniques that promote self-reflection to help clinicians increase their awareness of their own beliefs, values, experiences, and behaviors (including subconscious biases).86 These programs are designed to improve clinicians' interactions with patients and colleagues across cultural differences and are promising strategies for overcoming racial and ethnic disparities in health care; at the same time, however, evidence for the effectiveness of cultural competence training programs is limited, largely because of the lack of overall methodological rigor in these studies and the paucity of valid and reliable measures of clinician's attitudes and their impact on intercultural interactions or patient outcomes.87,88

The Theoretical Potential of Relationship-Centered Care to Reduce Disparities

Few studies examine whether the quality of relationships explains disparities in health service use or outcomes.66,89 However, we believe that relationship-centered care has the theoretical potential to reduce disparities in care in a variety of ways. Relationship-centered care directly addresses many of the hypothesized mechanisms by which patient race/ethnicity impacts clinician behaviors.19 First, physicians have been shown consciously or unconsciously to exhibit bias toward patients based on race/ethnicity.19 Although stereotyping behavior is often unconscious, relationship-centered care embraces the notion put forth first by proponents of patient-centeredness—that is, the goal that each person should be viewed as “a unique human being.”90 Second, physicians make differential medical decisions based on patient race.19,91 As relationship-centered care has the goal of equalizing power between patients and doctors, relationship-centered care can aim to reduce disparities in clinical decisions by increasing patient involvement. Finally, physicians have differential interpersonal behavior, characterized by more affective distance (less warmth and empathy), when interacting with racial/ethnic minorities.21,24,67 Here, too, relationship-centeredness places emphasis on improving these qualities of the patient-physician relationship.

In addition to improving the interactions between patients and clinicians in a way that could reduce disparities in interpersonal care, relationship-centered care also emphasizes the importance of the clinician's relationships with other clinicians and with the community. A physician who does not know or understand the patients' multiple communities formed by families, neighborhoods, cultures, work groups, and circumstances cannot possibly offer the best medical care. This aspect of relationship-centered care is particularly important when the patient and the physician come from different communities. Finally, through promotion of self-reflection of one's values and biases, relationship-centered care holds promise for reducing clinicians' engagement in stereotyping behavior directed toward ethnic minority patients. Clinician-patient, clinician-community, clinician-clinician, and clinician-self relationships are all targets of cultural competence training.


In summary, the existing literature demonstrates several relational aspects that are influenced by the race and/or ethnicity of participants in health care. A growing number of studies document racial differences in the quality of clinician-patient relationships and differences in communication and partnership between race-concordant and race-discordant clinician-patient pairs. Several studies show that ethnic minority and foreign-born physicians, as well as physicians from underserved areas or with an early interest in caring for underserved populations, are more likely to deliver care in ethnic minority communities. Recent work suggests that minority physicians and physicians who care for predominantly ethnic minority patient populations experience structural barriers and interpersonal biases from other clinicians in delivering care to their patients. Finally, although limited, new evidence from the medical education literature suggests that clinicians' self-awareness may promote behaviors that indicate cultural competence, and that enhancing the clinician-self relationship is a promising strategy for improving the quality of other relationships in health care across racial and ethnic differences.

Directions for Future Research

Future research should use a relationship-centered paradigm to explore how clinicians interact with patients, other clinicians, and the communities they serve, and how they examine and act upon their own beliefs in the context of race/ethnicity. There is a need for more empirical studies of the role of clinician-community, clinician-clinician, and clinician-self relationships in overcoming racial and ethnic disparities in health care. These studies should focus on dimensions of relationship-centeredness associated with race/ethnicity in clinician-patient relationships, including communication, partnership, respect, affiliation/liking, knowing, trust, and concordance. With regard to clinician-patient relationships, more studies are needed to identify the underlying mechanisms by which race concordance or discordance influence health care processes and outcomes. For example, little is known about how concordance or discordance across other dimensions influences race-concordant and race-discordant relationships. Another interesting question relates to the relative importance of concordance with regard to more visible demographic characteristics and characteristics that are below the surface or tip of the “cultural iceberg” (Fig. 2). It may also be important to examine common understanding (a concept related to concordance) across racial and ethnic differences and its contribution to health care disparities. Common understanding may be reflected by mutuality in perceptions of partnership, respect, knowing, liking, and trust, or by perceived similarities between clinicians and their patients, other clinicians, and the communities they serve. Studies of the role of implicit attitudes or biases (those that are “below the surface”) held by clinicians, patients, and the communities they come from are also needed, and will require collaboration among clinicians and behavioral scientists, including social psychologists. Through an enhanced understanding of these important dimensions of relationship-centered care across racial and ethnic lines, this research will provide new knowledge that informs the development of health systems, health professions education, and community-based interventions to reduce racial and ethnic disparities in health care and outcomes.


This work was supported by a research grant from the Fetzer Institute through its Relationship-Centered Care Research Network.