The Challenges of Understanding and Eliminating Racial and Ethnic Disparities in Health

Former Surgeon General Dr. David Satcher alerted the nation to the gross disparities in health that exist in this country and pushed for elimination of disparities as a high-profile objective for the Clinton administration.¹ His use of the bully pulpit to alert this nation, viewed by some as having the best health care in the world, that disparities in health care access, treatment, quality of care, and health outcomes exist between Americans from racial and ethnic minority groups and white Americans will influence research and policy in this nation for years to come. Current efforts to address disparities by the Department of Health and Human Services often focus on individual behaviors such as diet, exercise, and tobacco use. However, the evidence about why health disparities exist suggests complicated interactions between national and local health policies, institutional policies and procedures, individual provider behavior, and personal health–related behaviors in the larger context of economic and educational inequalities and opportunities and other social determinants of health. In all likelihood disparities in health between African Americans, Latinos, American Indians, Alaska Natives, and other Americans from racial and ethnic minority groups and white Americans will continue to exist unless we as a nation address the root causes of disparities from every dimension.

In this special issue of the Journal of General Internal Medicine on health disparities, our goal was to publish a series of manuscripts to increase knowledge about why disparities exist and strategies for eliminating disparities. More than 100 manuscripts were submitted for this special issue. We appreciate the interest from across the country. The large number of submissions made our task extremely difficult. We believe that this issue provides some insight into why disparities exist and how we can focus attention on eliminating disparities through improved clinical care, education of physicians and other providers, further research, and changes in policy.

The emergence of cultural competence as a major concern in the health care setting tracks with the promise of insurance for all Americans and the rise of managed competition in the 1990s. Viewed by many as a marketing tool to attract immigrant populations and those with limited English proficiency to managed care settings, cultural competence has often been limited to awareness of the health beliefs of specific immigrant populations and effectively communicating with patients who do not speak English. Training in cultural competence has been touted as a means to decrease health disparities, though there are no data to support this relationship. That native-born black Americans exhibit the worst health outcomes and quality of care suggests that other factors contribute to racial disparities besides lack of understanding of “foreign” health beliefs and language barriers. Medical educators have recently provided an expanded view of the major components of cultural competence,² but from a research perspective there is limited information on how one can measure this construct or how it should differ by racial and ethnic group.

In this issue, Johnson et al. describe patients’ perceptions of bias and cultural competence based on a structured, national telephone interview about health care quality.³ In their analysis of survey data, cultural competence is defined as respondents’ answers to questions about treatment by individual doctors (dignity and respect, understanding background and values, being looked down upon) and to questions about their experiences in health care systems rather than direct responses to questions about cultural competence of clinicians and systems of care. In a study from a public hospital clinic, Fernandez et al. describe the association of physicians’ self-report of cultural competence and language skills to their patients’ assessment of interpersonal processes of care.⁴ Clinicians who rate their Spanish language skills highly are more likely to be viewed as providing appropriate care by their patients. Communication, demographics, and source of care were also important in explaining differences in assessment of bias and cultural competence between racial and ethnic groups but did not explain all the difference. Accepted measures of cultural competence, that account for several domains of cultural competence, would provide insight into what factors account for patients’ perceptions of cultural competence and the more precise attitudes, skills, and knowledge required by clinicians.

Karliner et al. also provide important insight into how language ability impacts perceptions of care.⁵ Clinicians who have had some training in working with interpreters were more satisfied with interpreted mediated visits. However, physicians cite many barriers to effective communication related to diagnosis and treatment even when they use interpreters. Others have demonstrated that the use of interpreters can improve access to and quality of care.⁶ Physician satisfaction and sense of effectiveness of encounters with patients using interpreters may be affected by other factors such as time pressure, lack of knowledge about cultural beliefs, and other challenges posed by patients with limited English proficiency.

Three papers provide insight into disparities and screening and treatment for cancer. Peek and Han remind us that populations are heterogeneous and that, while mammography rates appear to have equalized between black and white women, lower income, uninsured, and older black women still experience lower rates of mammography screening, as do women from other ethnic groups and rural women.⁷ This is important to consider as researchers explore disparities in breast cancer mortality between white women and black women. Walsh et al. explore colorectal cancer screening among understudied populations and report surprising results in terms of attitudes and knowledge of the Latino and Vietnamese-American populations.⁸ Shavers et al. raise more questions about treatment for prostate cancer and describe disparities in cancer treatment among white, black, and Hispanic men with prostate cancer.⁹ Comorbid disease, expected life expectancy, stage, and grade did not explain higher rates of watchful waiting among black and Hispanic men compared with those of white men. Their study emphasizes the limits of administrative datasets to further our knowledge of why disparities exist. Education, sources of information about treatment choices, and other nonclinical factors are likely to influence the choice to “wait and see.”¹⁰

Improving clinical care for racial and ethnic minority groups will require comprehensive interventions that address the skills, knowledge, and attitudes of clinicians providing care. Furthermore, we need valid measures that work across diverse groups for assessing care to determine whether in fact it is culturally competent. These measures should be linked to concrete skills, standards for care, and improved quality of care. Educating physicians about cultural competence requires moving beyond theories about cultural interpretations of illness and health and how these beliefs affect personal health behaviors. Training in fundamental communication skills including explicit ways to overcome perceptions of bias in treatment are necessary to improve racial and ethnic minorities’ experiences in health care settings. As Ngo-Metzger et al. illustrate in their analysis of national survey data, Asians were less likely to be satisfied and express trust and more likely to change doctors than whites.¹¹ However, after adjusting for health care experiences, such as having the doctor take adequate time to listen to what they say and to understand their background and values, the association of race with satisfaction and trust is less impressive. The lack of adequate time, lack of listening, and lack of understanding are arguably mechanisms for dissatisfaction and not independent confounders. How data are analyzed and interpreted affect the way we come to understand racial and ethnic disparities in health, highlighting the importance of peer review and involvement of communities experiencing disparities in the design of research and the interpretation of the results. In addition, the questions raised about causal factors points to the need for a more unified approach to researching causal factors and the pathways that lead to disparities. More methodological and empiric work is needed to provide an understanding of the interactive effects of various factors.

Over the past 20 years, health disparities by race and ethnicity have been extensively documented.^12–13 Now the mandate is to reduce and eliminate these disparities. However, a better understanding of the mechanisms and contributing factors to observed disparities is essential in order to develop and implement effective interventions. Thus, we need more research on how to understand these relationships and how to address them in health care settings when the physician and the patient are different. The critical interactions of social class, race and ethnicity, language proficiency, cultural background, environment, and health are at the core of understanding the mechanisms for health care disparities. Some of the conclusions drawn in the articles in this issue are not surprising. We need appropriate resources to care for the diverse population of people in our health care system and to reach those who are not engaged. Lack of insurance, lack of outreach workers and health educators, and lack of interpreters all contribute to health disparities. The societal consensus to address the lack of diversity in the health work force has been weakened and adds another barrier for addressing disparities. As a nation, it is important to track how well we are doing to eliminate disparities and we hope that the Health Disparities Report Card, with appropriate data and descriptions of multiple interactions¹⁴ creates opportunities to develop appropriate health policies that will ensure equal treatment and equal quality of care for all Americans.—JudyAnn Bigby, MD,Brigham and Women"s Hospital, Boston, Mass, and Eliseo J. Pérez-Stable, MD,University of California, San Francisco, CA?