|Home | About | Journals | Submit | Contact Us | Français|
We conducted a review to examine current literature on the effects of interpersonal and institutional racism and discrimination occurring within health care settings on the health care received by racial/ethnic minority patients.
We searched the PsychNet, PubMed, and Scopus databases for articles on US populations published between January 1, 2008 and November 1, 2011. We used various combinations of the following search terms: discrimination, perceived discrimination, race, ethnicity, racism, institutional racism, stereotype, prejudice or bias, and health or health care. Fifty-eight articles were reviewed.
Patient perception of discriminatory treatment and implicit provider biases were the most frequently examined topics in health care settings. Few studies examined the overall prevalence of racial/ethnic discrimination and none examined temporal trends. In general, measures used were insufficient for examining the impact of interpersonal discrimination or institutional racism within health care settings on racial/ethnic disparities in health care.
Better instrumentation, innovative methodology, and strategies are needed for identifying and tracking racial/ethnic discrimination in health care settings.
Racial/ethnic minorities suffer disproportionate morbidity and mortality from chronic diseases, including cancer, heart disease, diabetes, and stroke. US racial/ethnic health disparities are a consequence of several factors including the disproportionate prevalence of less healthy lifestyles, low socioeconomic status, resource-poor neighborhood environments, and poorer access to care. Another factor is the poorer care received by minority patients after they enter the health care system. The 2005 National Healthcare Disparities Report indicated that White patients receive better quality of care than 53% of Hispanic, 43% of African American, 38% of American Indian/Alaska Native, and 22% of Asian and Pacific Islander patients.1 An updated report in 2010 showed no changes in disparities in 30 of 41 quality core measures for Hispanics, 40 of 47 measures for African Americans, 13 of 19 measures for Asians, and 15 of 22 measures for American Indian or Alaska Natives compared with Non-Hispanic Whites.2 Efforts to eliminate these disparities are hampered by the lack of a full understanding of all proximal causes including any role that racial/ethnic discrimination within the health care system might play.
Racial discrimination is defined as “(1) differential treatment on the basis of race that disadvantages a racial group and, (2) treatment on the basis of inadequately justified factors that disadvantage a racial group,”3(p39) and has been linked to racial/ethnic disparities in health outcomes. Not all discrimination occurs at the individual level or is intended. Seemingly benign policies, practices, structures, and regulations also have the potential to be discriminatory and are collectively referred to as institutional racism. Individual level discrimination and institutional racism may compound the negative effects of other health determinants,4 thereby placing racial/ethnic minorities in double jeopardy.
Results of a recent nationwide poll of the United States showed that 74% of African Americans, 69% of other non-Whites, and 30% of Whites report personally experiencing general race-based discrimination.5 Research studies have shown that general experiences with racial/ethnic discrimination are associated with a variety of adverse health outcomes including higher mortality6; lower use of cancer screening7; elevated blood pressure8,9; higher levels of C-reactive protein10; substance use11,12; mental and physical health13,14 including mood, anxiety, and psychiatric disorders12; increased depressive symptoms15; weight gain16; high body mass index17; and smoking.18 Not all studies, however, have found a significant association between general race/ethnicity-based discrimination and health.19,20 Although racial/ethnic discrimination within health care settings and health systems has also been implicated in health disparities,21 little is known about the empirical evidence supporting its prevalence or the association with poor health outcomes.
We provide a review of the scientific literature on the prevalence, perception of and effect of racial/ethnic discrimination and institutional racism within health care settings. Our specific objectives were to examine the extent to which recent literature addressed the following research questions:
This review summarizes results of recent research, identifies currently used instrumentation and methodology, and identifies areas where additional research is needed and is a resource for researchers with interest in working in this topic area.
We searched the PsychInfo,22 PubMed,23 and SciVerse Scopus (Scopus) databases24 for articles that focused on US populations published between January 1, 2008 and November 1, 2011. The beginning of the timeframe (2008) was chosen because of the 2009 publication by Williams and Mohammed25 that reviewed the literature on discrimination and health from 2005 to 2007 from a methodological perspective. Kressin et al.26 reviewed the psychometric properties of instruments used to examine discrimination. Our review differs from these in that we specifically examined current literature with a focus on racial/ethnic discrimination by health professionals or that occur within health care settings as opposed to discrimination occurring in the general community. Our cutoff date of November 1, 2011 reflects the publication date of the most current literature available at the time of our review.
Combinations of the following terms were used to search the 3 databases: discrimination, perceived discrimination, race, ethnicity, racism, institutional racism, stereotype, prejudice, and bias combined with health or health care in the text, title, or abstract (e.g., racial discrimination and health, racial discrimination and health care, racism and health and racism and health care; see Appendix A [available as a supplement to the online version of this article at http://www.ajph.org] for all search terms). A search of all 3 databases was also conducted with the names of commonly used instruments used to assess discrimination (e.g., Everyday Discrimination, Experiences of Discrimination, Perceptions of Racism Scale, Schedule of Racist Events, Implicit Association Test, Racism in Health Care Index, Perceived Prejudice in Health Care, Multiple Discrimination Scale, and General Experiences of Ethnic Discrimination) as search terms. These instruments were identified through a PubMed search with the search terms “measurement and racial discrimination” and through personal knowledge of the lead author. Other than time of publication, there were no other limits applied for the search. The last date the search was performed was November 18, 2011 for PubMed, November 23, 2011 for PsychNet, and December 13, 2011 for Scopus; however, only articles published between January 1, 2008 and November 1, 2011 were used in the review.
To be eligible for inclusion in this review articles had to be published in English; focused on US health care providers, patients, or US health care settings; and original research articles that reported quantitative or qualitative results of racial/ethnic discrimination, patient or provider perceptions of race/ethnicity-based discrimination within US health care settings, or discriminatory attitudes and beliefs of US patients or health care providers. We initially identified 5024 articles, of which 1185 were found to be duplicates of articles found in other searches of the same database or of one of the other databases. After these were deleted, 3839 unique articles remained from the combined searches of the 3 databases using the search terms in Appendix A (available as a supplement to the online version of this article at http://www.ajph.org). Titles of the remaining articles were first examined to determine their general relevance to the current review. Abstracts of articles that appeared to focus on racial/ethnic discrimination were then reviewed for inclusion (n = 686). Abstracts from dissertations; commentaries, letters to editors, editorials or that did not report original research; did not report results of research on US populations, or that did not provide information on the health care setting were deleted (n = 509). One hundred seventy-seven abstracts were further reviewed for inclusion, of which 94 were later found not to meet the inclusion criteria, leaving 83 articles of which 58 met the inclusion criteria after further review (Figure 1). These 58 articles are summarized in the current study.
The results of the literature search are presented by topic area. Each section addresses 1 of the 5 research questions.
The accuracy of research findings is only as good as the tools used to measure them. Optimal measures of racial/ethnic discrimination in health care settings assess the actual occurrence of or potential for discriminatory events, the impact of discriminatory events among individuals who experienced them, and the effect if any, on the patients’ interactions with their health care provider.26 Three published reviews examined measures used to assess discrimination in health research.25-27 In general, measures examined in 2 of the previous reviews26,25 primarily assessed general experiences of discrimination and thus only provided limited information on measures to assess racial/ethnic-discrimination in health care settings. As a component of a review that focused on personally mediated racism, Kressin et al.26 examined 16 measures that contained at least 1 item on perceived discrimination in health care settings. The majority of the measures reviewed by Kressin focused on attitudes and behaviors of health care workers perceived to be discriminatory (e.g., poorer service, less respect, and unfair treatment), and only 2 contained items that assessed discriminatory receipt of health care.26 Table 1 lists the questions used by investigators to assess racial/ethnic discrimination in the health care setting in the articles included in the current review (for a more complete list, see Table A, available as a supplement to the online version of this article at http://www.ajph.org).
Several basic methodological approaches can be used to obtain information about occurrences of or effects of discrimination including official counts; matched, residual, observational, and laboratory studies; in-depth interviews; and surveys.28 The strengths and limitations of each are provided in Table 2. In the current review, patient survey was the most widely used approach for assessing perception of health care discrimination (Table B describes the methods and results of the reviewed studies and is available as a supplement to the online version of this article at http://www.ajph.org).12,29-63
Survey data were used to examine the proportion of respondents who ever experienced a discriminatory event attributed to race/ethnicity in the health care setting or when getting medical care,31,32,37,40,41,43-45,50,56,58,60-65 recent discriminatory experiences (e.g., within last month—5 years),12,30,33-35,46,48,49,52-54,59 frequency of discriminatory events,31,33 perceptions of specific provider behaviors (e.g., treats me with respect and dignity, looks down on me)39,43,45,48-50,53 length of time since last discriminatory event,33 while receiving treatment of a specific condition,29,36,63 general beliefs regarding discrimination in health care settings,38,51,66 and methods of coping with discriminatory treatment.12
Assessing the prevalence of perceived discrimination among patients was the sole focus of 11 articles,35-37, 43,48,67-71 whereas the association of perceived race/ethnicity-based discrimination with specific health outcomes was the focus of 20 articles.12,29,30,32-34,39,40,44-46,49-54,57,59,63,72 Outcomes examined included treatment adherence,29 health care utilization,30,41,46 depression,68 cancer screening,33 health status,34,40,46,54 functional limitations,34 patient—provider communication,39,53 comfort with providing information on race/ethnicity,44 blood pressure control,45,50 quality of care,49,50,53,57 psychiatric disorders,12 quality of life,51 locus of control,54 and chronic health conditions.59
In-depth interviews and focus groups were the second most frequently used approach and were used to elicit information on patient perspectives of discrimination in cancer care,69 provider explanations for racial disparities in medical treatment,68 patient perceptions of discriminatory behavior by providers and its effect on compliance with treatment recommendations,67 description of health provider behaviors perceived to be discriminatory,73,74 perspectives regarding race and race-related conversations within the health care work-place,75 patient perspectives of barriers to care,71 and perception of the role of racial discrimination in health care disparities.76
The Implicit Association Test (IAT),77 a variation of the matched study approach was used in 7 articles to examine implicit racial/ethnic biases.76,78,79,80-84 Another previously popular method the case vignette, another form of the matched pair study; was used in only 2 of the reviewed studies to assess racial differences in treatment recommendations.82,84 None of the reviewed studies used other methodological approaches listed in Table 2. Because residual studies would not specifically focus on measurement of discrimination, articles using this approach were not included in this review.
Twenty-six articles measured respondent perception of health care discrimination (Table B, available as a supplement to the online version of this article at http://www.ajph.org). Established scales or their adaptations were used to measure perceived discrimination in 14 articles,12,30-32,38,39,43,45,48-51,53,54 and investigators developed their own measures in 8 articles.29,36,67,68,71,75,76 The original or an adapted version of the Experiences of Discrimination Scale was the most frequently used measure of perceived discrimination.12,30,31,43,48,49,79
Survey data specific to health care settings were obtained from a single question which assessed the occurrence of any race/ethnicity–based discriminatory health care event12,29,30,36,42,44,48,49,52,56,57,59 with either a question on attribution35 or length of time since the experience33,35 (Table 1; see also Tables A and B, available as supplements to the online version of this article at http://www.ajph.org). Only 1 survey conducted longitudinal assessments (6-month intervals) of discriminatory health care experiences.29
The IAT is a timed computer-based measure of unconscious bias assessed by measuring the time it takes to match members of social groups to particular attributes (e.g., good, bad, cooperative, and stubborn).77 Unconscious biases in health care settings were measured with adaptations of the IAT in all identified articles that measured implicit bias.76,78-84 Seven of the reviewed studies used the IAT to assess implicit race/ethnicity-based bias among current health care providers or clinical trainees.76,78,80-84 Studies using the IAT investigated implicit racial attitudes and cultural competency,76 the role of physician’s implicit biases in shaping physicians’ and patients’ perceptions in racially discordant medical interactions,78 unconscious bias and its association with clinical assessment among medical students,79 and an educational intervention to reduce implicit bias among medical students.80 Sabin et al.81 used the race attitude, compliant patient, and quality of care IATs to assess racial bias and quality of care provided by pediatricians and White-Means et al. examined objective and subjective cognitive processes among allied health and medical students.82
Two studies examined the correlation between measures of implicit and explicit racial attitudes.83,84 In a somewhat novel application among US-born Blacks, Krieger et al. examined the correlation between implicit and explicit measures of being a target of racial/ethnic discrimination personally and for the respondents’ racial/ethnic group.83 Sabin et al. examined the correlation between implicit and explicit racial attitudes in a physician subgroup.84
Surveys with questions on discrimination identified in the reviewed articles included the Behavioral Risk Factor Surveillance System (BRFSS; http://www.cdc.gov/BRFSS), Coronary Artery Risk Development in Young Adults (CARDIA; http://www.cardia.dopm.uab.edu/o_ucd.htm), the California Health Interview Survey (CHIS; http://www.chis.ucla.edu), and the Commonwealth Fund Health Care Quality Surveys (http://www.commonwealthfund.org). Data from the California Health Interview Surveys and the Commonwealth Fund Health Quality Survey were the most frequently used secondary data sources and were used in 14 studies.33-35,44-46,50,53,56-59,63,70
Because of the populations targeted, sampling design, and response rates, data from these surveys could not be used to provide reliable national estimates of the prevalence of discrimination encountered in health care settings. For example, although the BRFSS incorporated the Reactions to Race module as an optional module in 2002, only 9 states fielded the module in 2004 and only 2 in 2009; furthermore, overall BRFSS survey response rates are low in general.85 The Commonwealth Fund last fielded the Health Care Quality Survey in 2006. Data from cohort studies such as CARDIA could be used to assess trends in the prevalence of perceived discrimination within defined populations, but CARDIA is not a national probability sample. CHIS, a biennial survey, included 1 question on discrimination encountered in health care settings and an attribution question on the 2001, 2003, and 2005 surveys and tested a discrimination module in 2007 and 2009; however, the discrimination module is not publicly available, and the sample includes California residents only.
The reported prevalence of race/ethnicity based discrimination from local, regional, or state cross-sectional or cohort studies was between 6.9% and 52.0% for African Americans, 4.2% and 13.4% for Latinos and 0.4% and 6.0% for Non-Hispanic Whites.30,31,34,40,43,52,60,62,73 A small number of studies also reported perceived discrimination in health care of between 8.0% and 8.9% for American Indian patients,30,34 12.3% for Puerto Ricans in the United States,66 7.5% for Southeast Asians,30 and between 3.0% and 9.1% for Asian/Pacific Islanders,34,35,47,48,57,59 which varied by nativity and time since immigration in 1 study.59 Perceptions of racial/ethnic discrimination was also found in Veterans Affairs facilities, generally considered to be equally accessible to all veterans, retirees, and their family members.31,39,42,43 Patients did not report that they were subjected to race/ethnicity-based health care discrimination in all studies69 nor among all minority groups.36 The percentage of people who reported race/ethnicity-based discrimination in health care settings was higher in studies that covered longer reporting periods compared with studies with shorter periods (2 months—5 years).
Results of these studies show that African Americans and Latinos more frequently report race/ethnicity-based discrimination during their health care encounters compared with Non-Hispanic Whites,31,34,35,40,43,47,48,51,52,57 as do minority women compared with minority men33,41 and US-born Latinos compared with foreign-born Latinos.53 Perceived racial/ethnic discrimination within health care settings was also more frequently reported by patients who were younger than 65 years, of low socioeconomic status and uninsured or publically insured or with no usual source of care.34
Behaviors perceived to be discriminatory were described by Mexican Americans patients and families or African American women in 2 studies.72,73 Behaviors perceived by Mexican American patients of terminally ill children to be discriminatory included preferential treatment to White patients such as moving the Mexican American child out of a patient room into hall when a White child was admitted, allowing entire White families to visit after hours while limiting visits to 1 person at a time for Mexican American patients, and attending to the needs of White patients and family members while ignoring Mexican American family members and patients. In another study behaviors perceived by African American women to be racial profiling, biased or discriminatory included patients who arrived after they did being treated ahead of them, White patients being permitted to see the doctor without an appointment, the condescending tone of White providers/staff such as references to “you people.”73
Two studies examined geographic variation in reports of discrimination in health care encounters.34,52 Of the 114 people (4.8%) who reported health care discrimination in the 2004 through 2008 Behavioral Risk Factor Surveillance System (BRFSS) 13.1% resided in the North, 37.1% in the South, 48.6% in the East and 1.2% in the West.52 In a study that examined urban and rural differences in discrimination by site of care D’Anna et al., found that urban residents more frequently than rural residents reported race/ethnic-based discrimination when receiving health care in a private doctor’s office or within an HMO plan but not at other health care sites.34
Experiences with racial/ethnic discrimination in health care settings were not limited to patients. Three studies examined the bias, prejudice, stereotyping, and discriminatory attitudes or behaviors of patients toward providers.86-88 In a national cross-sectional survey of 529 physicians, October 2006 through February 2007, 60% of African American, 33% of Asian, 17% of Hispanic or Latino, 30% of Non-Hispanic White, and 42% of other race physicians reported a belief that patients refused care from them because of their race/ethnicity.86 After 9/11, 41.2% of Arab American nurses in another survey reported more intimidation, 32.4% reported more suspicious treatment, and 15.2% reported more frequent patient refusals of treatment by them. In a study of direct care workers 32% to 54% reported hearing racial/ethnic comments perceived to be intentionally hurtful from residents or clients, other staff members, and patient family members.88
Eighteen studies addressed the impact of patient perception of racial/ethnic discrimination within health care settings (Table B, available as a supplement to the online version of this article at http://www.ajph.org).12,29-33,35, 37,41,44-46,49,52,53,57,77 Perceived racial/ethnic discrimination during health care encounters was associated with poorer self-reported health status,40,46,53 lower perceived quality of care,57 greater bodily pain,31 more psychiatric disorders,12,32 lower colorectal screening among women but not men,33 worse diabetes care and more diabetic complications,52 poor adherence to antiretroviral therapy,29 under-utilization of health services,46 physical and emotional functional limitations,35 delays in seeking care,30,84 failure to adhere to recommendations,30,45,84 societal distrust,37 unmet needs for health care utilization,47 lower levels of comfort in providing information about race/ethnicity in health care settings,44 and interruptions in care, mistrust of providers, and avoidance of health care systems.84 Not all studies however, found a negative impact of perceived discrimination.41,49,51 Perceived racial discrimination when seeking health care was not significantly associated with lower utilization of the flu shot41 or less medication intensification among diabetes patients.49 In the Myaskovsky et al. study, African American patients with spinal cord injuries who reported more discrimination also reported better occupational functioning.51
None of the reviewed studies examined trends or temporal patterns of perceived racial/ethnic discrimination in health care settings.
Eight articles examined racial/ethnic attitudes, biases, stereotypes, and behaviors among current or health care providers in training.68,71,76,77,80-82,84 Findings from the majority of the reviewed articles provide evidence of the prevalence of provider explicit and implicit biases, attitudes and beliefs that could negatively affect the health care delivered to racial/ethnic minority patients. Included were less patient involvement in decision-making,77 disbelief of the existence of health care disparities,68 belief of a lack of role of racial discrimination in health care disparities,76 and implicit preferences for White race80-82 and light skin color.81 By contrast, Sabin et al. found that medical faculty and residents had more explicit positive beliefs regarding patient compliance for African Americans compared with European Americans.80
The causes of discriminatory care are not limited to the personal biases and prejudices of providers or patients. Relatively absent from the literature were studies that examined institutional racism or racial/ethnic variation in the impact of specific federal and institutional regulations, policies, and practices on the receipt of health care. In a notable exception, an audit study of 273 specialty practices in Cook County, Illinois, showed that 66% of persons posing as parents of a sick or injured child requiring urgent care who mentioned Medicaid-CHIP (Children Health Insurance Program) were denied an appointment compared with 11% of patients with private insurance.89 Among clinics that scheduled appointments for both types of patients, those with Medicaid-CHIP waited 22 days longer for an appointment than did privately insured patients. The potential discriminatory impact of this practice becomes apparent when considering the substantial numbers of minority children insured through state CHIP programs.90 Although low Medicaid reimbursement rates were thought to contribute to this practice, it provides a good example of how policies, regulations, and practices can result in discriminatory behavior.
We examined the availability of data on the prevalence, trends, mechanisms, and institutional policies and practices associated with racial/ethnic discrimination in health care settings. Although there were a number of studies that described race/ethnicity based discriminatory behaviors, attitudes, biases and preferences that could potentially contribute to discriminatory health care we found no studies that specifically addressed the US prevalence or trends. Also, relatively absent were studies that addressed how institutional racism impacts the health care received by racial/ethnic minority patients.
None of the measures used in the reviewed studies captured information on all 3 aspects of effective measures as described by Kressin et al. (i.e., assessed the actual occurrence of or potential for discriminatory events, impact of discriminatory events among individuals who experienced them, and the effect if any, on the patients’ interactions with their health care provider).26 There was also wide variation in the length of time for which discrimination was assessed (i.e., lifetime, varying time intervals), which would be expected to greatly influence the reported prevalence of perceived discrimination and adds to the difficulty in comparing rates across studies. Furthermore, many of the studies that examined perceived racial discrimination in health care settings provided little, if any, information about the specific actions perceived to be discriminatory, specific context in which the discriminatory act occurred (e.g., emergency room, doctors office, or health clinic), or the specific perpetrator (e.g., nurse, office staff, or physician), which would be helpful in developing targeted interventions.
Unconscious biases and stereotyping that can underlie decision-making contribute to the difficulty of assessing the actual impact of race/ethnic discrimination in health care settings on access to and receipt of optimal care. Whereas patients are a good source of information about perceived discriminatory provider behaviors, system characteristics perceived to be discriminatory, and the personal consequences of perceived discrimination, they may be unknowledgeable about the standard of care for their disease or condition. Provider and other staff surveys that examine their observations of practices within their institutions therefore may be a better source of information on the prevalence of racial/ethnic discrimination in health care settings and its association with the receipt of health care.
Findings from the reviewed studies suggest that racial/ethnic discrimination may be prevalent in health care settings and potentially influence the health care received by minority patients. However, little is known about the national prevalence or trends in race/ethnicity-based discrimination in US health care systems. This situation results in part from the fact that Federal statistics on racial/ethnic discrimination in the United States are primarily limited to findings from audit programs for housing, hate crimes, and other complaints filed with US agencies such as the Equal Employment Opportunity Commission (EEOC) and the Fair Housing Commission.28 In the absence of federally mandated surveillance, much of what we know about the receipt of discriminatory health care comes from small research studies that focus on racial/ethnic disparities in receipt of treatment and outcomes, and that are not generally designed to provide information on what, if any, portion of observed disparities is a result of racial/ethnic discrimination. Accurate measurement and tracking of the incidence and prevalence of prejudice, bias, and other discriminatory attitudes in health care settings, therefore, is not only important to increasing the visibility of discrimination as a health risk but will help clarify its relationship to racial/ethnic disparities in health outcomes.91
There is a continuing need for innovative methodology, better instrumentation, and strategies for identifying racial/ethnic and other types of discrimination in health care settings, particularly because of the somewhat subjective manner in which health care is delivered. For example, Shields et al. were successful in using actors (standardized patients) portraying patients with stage IV lung cancer to evaluate patient-centered communication.92 Using standardized patients to monitor receipt of discriminatory care in a manner similar to housing and employment audits might be a feasible method for directly assessing racial discrimination in health care receipt. There is also a need to create data resources that facilitate tracking of reports of the receipt of discriminatory care and to establish a system of accountability that facilitates positive change.
An interesting area for future investigation is the role of stereotype threat defined as “being at risk of confirming, as self-characteristic, a negative stereotype about one’s group.”93(p797) A number of racial/ethnic stereotypes are prevalent in health care settings.77,79,84,94-96 Stereotype threat in the clinical setting is posited to be more likely to occur when features of the setting make prevailing stereotypes of minority patients salient.94 However, only 1 research study within the review period examined the impact of stereotype threat on the receipt or utilization of health care. In a review of studies on stereotype threat, Burgess et al.94 concluded that stereotype threat in clinical settings contributed to treatment nonadherence and influenced patient outcomes. Stereotype threat also resulted in impaired communication between patients and providers, with patients discounting feedback and disengaging by avoiding going to the doctor and exhibiting the stereotyped behavior which, in turn, reinforced provider beliefs and their clinical decision-making.94
Of note, our review only included studies published in English on US populations from 2008 to November 1, 2011 that focused on racial/ethnic discrimination occurring in health care settings. Therefore, other studies are likely to exist that may provide insight into available measures, prevalence, trends, and systematic factors that were not examined in this review. In addition, because we were interested in data sources and measures currently used by researchers as evidenced by the published literature, we only evaluated surveys used in the reviewed studies.
The existing literature suggests that racial/ethnic discrimination may be prevalent in US health care settings but more research including national studies are needed. Several gaps in the research literature were also identified and should be considered for future research studies. As Gee97 suggests, there is a need to address discrimination at both the interpersonal and structural levels. Doing so will help us to understand how discrimination operates within health care settings while identifying targets for intervention. Studies should include systematic examinations of patient-physician interactions, particularly as they relate to communication styles and nonverbal behaviors that have the potential to elicit the perception of discrimination among diverse patients. Although it is possible that some providers purposely engage in discriminatory care, unconscious bias among well-meaning providers is the more likely culprit. Additional research is needed that explores whether and under what conditions the implicit attitudes of providers affect the quality of the medical care delivered to racial/ethnic minority patients. Physicians who are trained to be aware of implicit biases can be sensitized to their potential for bias,77,79,94,96 which may encourage self-regulation77 and facilitate decision-making based on the specific needs and resources available to individual patients.
There is also need to assess how racial/ethnic discrimination faced by racial/ethnic minority health providers within their work-places (i.e., hospitals and clinics) influence the availability of minority health care providers, and as a consequence, minority patient perception of the accessibility of appropriate care.
The introduction of health care reform, which has provisions that affect access to and the composition of health insurance plans, hospital availability, and other federal policies; provides a unique opportunity for research on the effect of health system policies on the receipt of discriminatory care. In recognition of the need for additional research, The National Cancer Institute (NCI) has recently reissued the program announcement “The Effect of Racial/Ethnic Discrimination on Healthcare Delivery,”98 for investigators interested in pursuing research in this topic area.
The authors would like to acknowledge Cecilia Casale, Office of Extramural Research, Education and Priority Populations. Agency for Health Care Quality and Research and Wendy J. Nilsen, PhD, NIH Office of Behavioral and Social Sciences Research for their contributions during the conceptual phase of this project. Thank you also to Martin Brown, PhD, Rachel Ballard-Barbash, MD, MPH, and Robert Croyle, PhD, for their review of this article for NCI publication clearance and Lawrence Fine, PhD, for his review for NHLBI publication clearance.
ContributorsAll authors contributed to the conceptualization of the project. V. L. Shavers, P. Fagan, and D. Jones reviewed the literature, participated in writing, reviewing and editing the article. W. M. P. Klein reviewed and edited the draft article. J. Boyington and C. Moten provided comments on the draft article and E. Rorie performed the initial literature search.
Human Participant Protection
Human participant protection was not required because no human participants were involved.
Vickie L. Shavers, Behavioral Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute (NCI)
Pebbles Fagan, University of Hawaii Cancer Center.
Dionne Jones, Services Research Branch Division of Epidemiology, Serves and Prevention Research (DESPR) National Institute on Drug Abuse.
William M. P. Klein, Behavioral Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute (NCI)
Josephine Boyington, National Heart Lung and Blood Institute (NHLBI), National Institutes of Health, Bethesda, MD.
Carmen Moten, Center to Reduce Cancer Health Disparities National Cancer Institute Rockville, MD.
Edward Rorie, Nova Research Company Bethesda, MD.