Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
The Civil Rights Dimension of Racial and Ethnic Disparities in Health Status Thomas E. Perez, J.D., M.P.P.1 Clinic Director and Assistant Professor of Law University of Maryland School of Law It is hard to talk about race. Discussions about race in general and racial discrimination in particular are potentially unnerving, which ex- plains in large measure why such conversations are so few and far be- tween. In the health care context, discussions about race and racial dis- crimination are particularly rare. But this is slowly beginning to change, and two primary forces have triggered the change. First, the Initiative to Eliminate Racial and Ethnic Disparities in Health has focused considerable attention on improving the health status of people of color in the United States. The disparities initia- tive was launched in 1998 by former President Clinton and Secretary of Health and Human Services Donna Shalala. Under this bipartisan initia- tive, President Clinton and Secretary Shalala committed the nation to the ambitious yet attainable goal of eliminating racial and ethnic disparities in six areas of health status while continuing the progress that has been made in improving the overall health of people in America. The six focus areas of this initiative are: (1) infant mortality; (2) cancer screening and management; (3) cardiovascular disease; (4) diabetes; (5) HIV infection/ AIDS, and (6) immunizations. This Initiative enjoys the support of current HHS Secretary Tommy Thompson. 1 Assistant Professor of Law and Director, Clinical Law Programs, University of Maryland School of Law. Former Director, Office for Civil Rights, U.S. Department of Health and Human Services. The author wishes to thank Kathryn A. Ellis, former principal deputy director of the Office for Civil Rights, and Juliet Choi, research assistant and law student, for their input and assistance. 626
627 CIVIL RIGHTS DIMENSION The second factor that has triggered a discussion about the role of discrimination in health care is the dramatic increase in immigrant popu- lations in urban and rural settings across the United States. These major demographic shifts have forced health care providers, government offi- cials, and communities to address the unique challenges confronting im- migrants. The growing dialogue surrounding the need for a culturally competent health profession reflects the recognition of the changing face of America. The disparities initiative and the rapid expansion of immigrant popu- lations across America have focused the attention of health professionals, politicians, and policy makers on the critical question of why it is that communities of color and immigrant communities are lagging behind the rest of America in so many critical measures of health status. Racial disparities in many areas of health status are well-documented, disturbing, and preventable.2 In order to eliminate health disparities, it is important first to understand the root causes. Until recently, the role of discrimination was largely ignored. Instead, disparities have been de- fined in other terms: economicâpoor people are more likely to have diffi- culty accessing quality health care; geographicâthose who live on the wrong side of the tracks and near the toxic waste dump are more likely to encounter health problems; and sometimes geneticâcertain races or ethnicities may have a genetic predisposition to certain illnesses. Finally, disparities frequently have also been defined in terms of education or be- havioral issuesâif we could simply teach better habits. All of these explanations are undoubtedly true depending on the con- text, with some factors perhaps playing a larger role than others. How- ever, another factor is rarely discussedâdiscrimination. This paper ad- dresses the role that discrimination plays in explaining health disparities, and outlines a host of civil rights interventions that can be put into place to address these disparities. This paper is divided into four parts. Part one addresses what specifically is meant by âdiscriminationâ in the legal sense. Discrimination is an often misunderstood concept, and this section seeks to demystify and destigmatize discrimination by ex- plaining what it means legally in the health care context. This section 2 Dr. Jack Geiger has performed perhaps the most comprehensive study to date tracking down the literature on racial and ethnic disparities in health care. Geiger, HJ. 2001. Racial and Ethnic Disparities in Diagnosis and Treatment: A Review of the Evidence and Consider- ation of Causes, Institute of Medicine, Committee on Understanding and Eliminating Racial and Ethnic Disparities in Health Care. Examples of studies documenting disparities include: Bach PB et al. 1999. Racial Differences in the Treatment of Early Stage Lung Cancer. New England Journal of Medicine, 341:1198-205; Ayanian, JZ et al. 1999. The Effect of Patientsâ Preferences on Racial Differences in Access to Renal Transplantation. New England Journal of Medicine, 341:1661-69.
628 UNEQUAL TREATMENT introduces Title VI of the Civil Rights Act of 1964, one of the principal tools used in addressing discrimination in the health care context. Parts two and three explore whether discrimination is actually alive and well in the health care setting. Part two outlines research regarding the potential role of discrimination in explaining health disparities. This section is not intended as an exhaustive review of the literature, but it does outline noteworthy studies that have triggered further discussion about the role of discrimination in explaining disparities. Part three discusses actual discrimination cases in the health care con- text. There are some respected experts who believe that discrimination is not a significant problem in health care.3 Regrettably, the evidence belies these assertions, and there is a substantial body of cases involving dis- crimination in a variety of health care settings, and this section discusses these cases. A generation ago, discrimination in health care was quite similar to discrimination in other settings. That is, hospitals, nursing homes, and other health care facilities, like schools, were segregated and needed to be integrated. Thus, early discrimination cases in the health care context focused on issues such as equal access to medical facilities. Todayâs civil rights challenges in health care generally are more subtle, but no less compelling. The six categories of cases discussed in section three are (1) intentional discrimination cases; (2) access to health care for people with limited English skills; (3) medical redlining in a variety of settings, including managed care and home health care; (4) other man- aged care issues, including physician participation; (5) discrimination in access to treatment; and (6) other unique challenges confronting immi- grant population in addition to language access. Parts two and three lay out the case that discrimination is a root cause of health disparities, and underscore that a comprehensive strategy to eliminate disparities must incorporate a strong civil rights component. Part four contains a series of recommendations for eliminating disparities that focus on the civil rights dimension of the disparities challenge. These recommendations include (1) promoting the collection of data related to race, ethnicity and primary language by federal, state and local govern- ments and health care facilities; (2) enhancing the federal infrastructure for enforcing civil rights laws in the health care context; (3) expanding the capacity of private organizations to enforce civil rights laws in the health settings; (4) implementing a comprehensive language access agenda; (5) amending federal law to re-establish that private parties can enforce all provisions of the Title VI regulations; and (6) reviewing federal law and 3 See, for example, Havighurst, C. and OâNeill, J. September 6, 2001. Invited Presentation at panel discussion, âRace, the Medical Marketplace, and Health Care Disparities,â National Academies of Sciences.
629 CIVIL RIGHTS DIMENSION removing barriers to participation for legal immigrants in critical pro- grams, such as Medicaid and the State Childrenâs Health Insurance Pro- gram (SCHIP). Effective data collection is a lynchpin of any comprehensive strategy to eliminate health disparities. In order to fully appreciate the value of collecting data on race, ethnicity and language of preference in health care, it is useful to draw upon civil rights experiences in non-health care set- tings. Part four explains how advocates and government officials have used federal laws pertaining to data collection in the home mortgage and the policing context to address the issues of mortgage lending discrimina- tion and racial profiling, respectively. Policymakers concerned about data collection in health care can learn a number of valuable lessons from these experiences. Implementing all of the recommendations outlined in part four will assist in reducing, but will not eliminate racial and ethnic disparities in health. Indeed, eliminating racial and ethnic disparities in health will re- quire a comprehensive strategy that reflects a keen understanding of the multi-faceted nature of the challenge. The principal aim of this paper is to give policymakers, providers, advocates and other stakeholders a better understanding of the civil rights dimension of the challenge, as well as a set of suggestions on how to address these civil rights concerns. PART ONE: DEFINING DISCRIMINATION Discrimination can be a scary term, as it frequently conjures up im- ages of nefarious actors engaging in conscious acts of bigotry designed to deny people of color an important benefit or opportunity. While isolated incidents of such discrimination regrettably exist, discrimination in todayâs health care marketplace is much more subtle, but no less serious in its effect. Understanding what discrimination means under federal civil rights laws is critical to understanding precisely how civil rights laws can be used to combat racial and ethnic disparities in health. Title VI of the Civil Rights Act of 1964 (Title VI) and its accompanying regulations are the primary tools used to attack discrimination in health care. Title VI prohibits discrimination based on race, color, or national origin, in any program or activity that receives federal financial assis- tance.4 Virtually every health care provider receives some form of federal 4 Title VI provides that â[n]o person in the United States shall, on the ground of race, color, or national origin, be excluded from participation in, be denied the benefits of, or be sub- jected to discrimination under any program or activity receiving Federal financial assis- tanceâ 42 U.S.C. Â§ 2000d.
630 UNEQUAL TREATMENT financial assistance, so it is not difficult as a practical matter to fall within the ambit of Title VI. The language of Title VI itself addresses intentional discrimination. However, the regulations issued pursuant to Title VI make clear that Title VI itself reaches far beyond intentional discrimination. The regulations prohibit health care providers from using âcriteria or methods of admin- istration which have the effect of subjecting individuals to discrimination on the basis of their race, color, or national origin, or have the effect of defeating or substantially impairing accomplishment of the objectives of the program [with] respect [to] individuals of a particular race, color or national origin.â5 Thus, there are two types of discrimination under Title VI: (1) intentional discrimination; and (2) policies or practices that may be neutral on their face but have the effect of discriminating on the basis of race, color, or national origin (the âdisparate impactâ theory or âeffectsâ test). Title VI applies to a range of discrimination concerns, including ac- cess to health services, redlining, and physician staff privileges and par- ticipation in managed care. Until earlier this year, private litigants were permitted to sue to enforce Title VI regulations prohibiting acts with dis- criminatory effects. However, the United States Supreme Court in Alex- ander v. Sandoval ruled that there is no private right of action under the Title VI regulations (121 S.Ct. 1511 ). The case itself involved a class of non-English speaking residents of Alabama, who alleged that the stateâs policy of offering the driverâs license exam only in English amounted to national origin discrimination under the aforementioned âeffectsâ provi- sion of the Title VI regulation. The Court of Appeals for the Eleventh Circuit agreed, but the Supreme Court ruled that private parties lacked the authority to file a lawsuit to enforce the effects provision of the Title VI regulation. Sandoval represents a significant setback for civil rights advocates, who had pursued a steady diet of civil rights cases in health care using the disparate impact theory under the Title VI regulation. Private plaintiffs must now establish that the conduct in question amounts to intentional discrimination under Title VI. It is important to note, however, that Sandoval only applies to private parties, and has no effect on the federal governmentâs ability to pursue civil rights cases using the effects test un- der the Title VI regulation. The authority of agencies such as the Office for Civil Rights (OCR) at the Department of Health and Human Services (HHS) remains unchanged. There are two ways to invoke the protections of Title VI: an aggrieved person can (1) file a written complaint with OCR and/or (2) file a lawsuit 5 45 C.F.R. 80.3(b)(2).
631 CIVIL RIGHTS DIMENSION under Title VI.6 It is not necessary to first file an administrative complaint in order to file a lawsuit. OCR is the federal agency responsible for enforc- ing Title VI in the health and human service context, and has been doing so for over 35 years. OCR will investigate that complaint by, among other means, reviewing the pertinent practices and policies of the hospital or provider that is the subject of the complaint, the circumstances under which the possible noncompliance occurred, and other factors relevant to a determination as to whether the hospital or other recipient has failed to comply with Title VI.7 OCR may also, on its own, review the practices of a recipient of federal funds to determine whether they are complying with Title VI. A complaint is not necessary.8 If OCR finds noncompliance, it will first seek voluntary compliance by the recipient. OCRâs ultimate sanction is to terminate federal funding, either in an administrative proceeding or by referring the case to the De- partment of Justice for litigation. The advantages of filing an OCR com- plaint are that: (1) it does not require a lawyer; (2) a finding of discrimina- tion by a federal agency can be very powerful; (3) the involvement of OCR frequently results in a resolution of the case, sparing the expense and un- certainty of protracted litigation; and (4) OCR can investigate both allega- tions of intentional discrimination under Title VI and disparate impact under the Title VI regulations, whereas private litigants can only sue un- der an intentional discrimination theory. Title VI is not the only civil rights tool that is employed to address discrimination in health care. For instance, the Hill-Burton Act proscribes discrimination in health care settings, and provides a statutory hook to attack racial discrimination in health care. However, relatively few health care providers fall within the jurisdiction of the Hill-Burton Act, whereas virtually every actor in the health care system receives federal financial assistance, and is therefore covered by Title VI. Title VII of the Civil Rights Act of 1964, and 42 U.S.C. 1981 are anti-discrimination provisions that can be used by employees or other providers who feel that they have been victims of discrimination.9 For instance, physicians of color who have been 6 45 C.F.R. 80.7(b). 7 45 C.F.R. 80.7(c). 8 45 C.F.R. 80.7(a). 9 Title VII provides, in part, that it is unlawful for an employer to âfail or refuse to hire or to discharge any individual, or otherwise to discriminate against any individual with re- spect to his compensation, terms, conditions or privileges of employment, because of such individualâs race, color, religion, sex, or national originâ 42 U.S.C. 2000e-2(a)(1). Like Title VI, Title VII prohibits intentional discrimination as well as disparate impact discrimination. Section 1981 prohibits racial discrimination in the making and enforcement of contracts, and applies to public and private contracting.
632 UNEQUAL TREATMENT terminated from managed care networks have sought redress under Title VII and/or section 1981. These statutes provide legal framework for addressing the role of dis- crimination in health care. PART TWO: IS DISCRIMINATION ONE OF THE ROOT CAUSES OF DISPARITIES: WHAT DOES THE RESEARCH SAY? Research Regarding the Potential Role of Bias in Physician Decision-Making There is a growing body of research addressing whether discrimina- tion is one of the root causes of racial and ethnic disparities in health. The 1999 Schulman study generated the most attention.10 In this study, re- searchers addressed whether the patientâs race or gender affected the physicianâs decision to recommend cardiac catheterization. The study involved simulated patients (white males, white females, African-Ameri- can males, African-American females) who had identical backgrounds in all respects except race and gender, identical risk factors and symptoms. The researchers controlled for other factors, and concluded that the patientâs race and gender indeed affected the physicianâs decision to rec- ommend cardiac catheterization. Specifically, the study found that Afri- can Americans and women with chest pain had relative odds of referral for cardiac catheterization that were 60 percent of the odds for whites and men. African-American women faced the greatest disparity, as they had relative odds that were 40 percent of those for white men. The real implication of the study was actually quite simple: doctors are human. Like lawyers, businesspeople, and other professionals, doc- tors are fallible and may discriminate, consciously or subconsciously. In other civil rights contexts, it has been shown that racial bias can infect the corporate boardrooms, the schoolrooms, and the police precinct rooms. This study simply concluded that racial bias can affect who gets to the operating room. Yet, the Schulman study provoked a firestorm of criticism, as if it were the only study to document the potential role of discrimination in explain- ing racial and ethnic disparities. In a âSounding Boardâ article in the New England Journal of Medicine a few months later, three physicians described the results of the Schulman study as âoverstated,â although they also noted that their purpose in writing was ânot to deny the occurrence of 10 Schulman K. et al. 1999. The Effect of Race and Sex on Physiciansâ Recommendations for Cardiac Catheterization. New England Journal of Medicine, 340:618-626.
633 CIVIL RIGHTS DIMENSION racial or sex bias.â11 These commentators and others questioned the un- derlying methodology, and suggested that the results are not as stark as the study suggested. Most recently, a study by a team of researchers led by Dr. Jersey Chen concluded that minority and non-minority physicians alike were more likely to recommend a white patient for cardiac catheterization than a minority patient.12 The Schulman study did not control for the race of the physician. The Chen study prompted one commentator to opine at a panel discussion sponsored by the Institute of Medicine that the Schulman study has been âlargely discredited.â13 Such statements seeking to dismiss the Schulman study miss the mark, and ignore the wide body of research, both before and after the Schulman study, suggesting that race continues to matter in health care, and racial bias may contribute to racial and ethnic disparities in health status.14 Such statements also ignore serious methodological flaws in the Chen study that a number of experts have noted.15 Most notably, the African-American physicians in the Chen study tended to be internists, not cardiologists, when compared with the white physicians. There are so few African-American cardiologists that it would likely be difficult, if not impossible, to devise a study that addresses the question presented in the Chen study using a sufficient sample of both African-American and white cardiologists. The reality is that discrimination may be hard to quantify, but is diffi- cult to deny. As Drs. Arnold Epstein of the Harvard School of Public Health and John Ayanian of the Harvard Medical School pointed out in response to the Chen study, 11 Schwartz, L. et al. 1999. Misunderstandings About the Effects of Race and Sex on Phy- siciansâ Referrals for Cardiac Catheterization, New England Journal of Medicine, 341:279-282. 12 Chen, J. et al. 2001. Racial Differences in the Use of Cardiac Catheterization After Acute Myocardial Infarction, New England Journal of Medicine, 344:1443-1449. 13 Symposium of September 6, 2001, at the National Academies of Sciences on The Role of Discrimination in Explaining Racial and Ethnic Disparities in Health Status, testimony of June OâNeill. 14 See e.g., Gornick, M.E. 2000. Vulnerable Populations and Medicare Services. New York: Century Foundation. van Ryn, M. Burke, J. 2000. The Effect of Patient Race and Socioeco- nomic Status on Physiciansâ Perceptions of Patients, Social Science Medicine, 50:813-828; U.S. Civil Rights Commission. 1999. Smith, D.B. 1999. Health Care Divided: Race and Healing a Nation (University of Michigan Press). The Health Care Challenge: Acknowledging the Disparity, Confronting Discrimination, and Ensuring Equality. Washington, DC: U.S. Civil Rights Com- mission. Mayberry, R.M. 1997. Racial and Ethnic Differences in Access to Medical Diagnosis and Treatment. Atlanta, GA: Morehouse School of Medicine. 15 See e.g., 2001. Report of Institute of Medicine Committee on Understanding and Elimi- nating Racial and Ethnic Disparities in Health Care.
634 UNEQUAL TREATMENT âRacial bias is difficult to detect in surveys of physicians or medical records. Physicians may have various biases, but they often do not perceive them, and would not report them as such. Thus, it is no surprise that previous studies have not directly documented bias. Instead, racial bias has always remained a pos- sible explanation for residual racial differences in treatment, after other explana- tory factors have been accounted for.â16 Dr. Neil Calman, a professor of family medicine at Albert Einstein School of Medicine and President and co-founder of the Institute for Ur- ban Family Health, has written and spoken extensively about his own efforts to come to grips with his own racial bias.17 A white physician practicing in the Bronx with a predominantly minority patient mix, Dr. Calman has written about how this experience has forced him to confront his own racial stereotypes that have stood in the way of quality treatment of his patients. Based on his experience, he has concluded that âthe shadow of racial prejudice looms over us all.â According to Dr. Calman, until and unless physicians develop the capacity to confront their own stereotypes, it will be difficult to prevent conscious or subconscious bias from interfering with the physician-patient relationship. Dr. Calmanâs thesis is consistent with those put forward by Dr. Vanessa Gamble, former vice president of the Association of American Medical Colleges, and a frequent contributor on issues of race and medicine. In a 1997 article, Dr. Gamble wrote about race and medicine, and discussed a newspaper story reporting on an African-American woman who went to the emergency room of a county hospital in Los Angeles for treatment of a suspected broken arm. 18 According to the article, interns, who were white, asked her to posi- tion her arm âlike she would having a beer on a Saturday night.â The patient responded: âDo you think Iâm a person on welfare?â âWell arenât you?â was the response. The patient was actually an administrator at the University of Southern California Medical School.19 Dr. Gamble used this experience to discuss the role of discrimination in health care, and exam- ine the forces at work that would enable a physician to basically look at a person and conclude they are on welfare. 16 Epstein, A., Ayanian, J. 2001. Racial Disparities in Medical Care. New England Journal of Medicine, 344:1471-73. 17 Calman, N. 2000. Out of the Shadow, A White Inner City Doctor Wrestles With Racial Prejudice, Health Affairs, 19:170-74. 18 Gamble, V. 1997. Under the Shadow of Tuskegee: African Americans and Health Care, American Journal of Public Health, 87:1773-78. 19 Moustafa, K. April 1, 1995. U.S. Remains Worlds From Top Notch Health Care. The Fresno Bee.
635 CIVIL RIGHTS DIMENSION The Calman and Gamble articles are also consistent with a more re- cent study led by Dr. Michelle van Ryn examining whether physiciansâ perceptions of patients are affected by the patientâs race or socioeconomic status.20 Researchers concluded that physicians tended to perceive Afri- can Americans and members of low and middle SES groups more nega- tively on a number of measures than they did whites and upper SES pa- tients. As a result of this study, it is difficult to dismiss the aforementioned encounter in Los Angeles as an unrepresentative anecdote. It is also interesting to examine the van Ryn research in the context of studies on race concordance in the physician-patient relationship. A re- cent study of white and African-American patients by a research team led by Dr. Lisa Cooper-Patrick found that patients seeing physicians of their own race rated their physiciansâ decision-making styles as more partici- patory.21 Another study found that African American and Latino patients were more likely to rate care as excellent and very good from physicians of the concordant race.22 These studies on race concordance and the physician-patient relation- ship do not stand for the proposition that racial bias on the part of physi- cians is the reason why African-American and Latino patients tend to prefer to be treated by African-American and Latino physicians, respec- tively. However, these studies are frequently cited in discussions on cul- tural competency, and suggest that Dr. Calman and Dr. Gamble are cor- rect in stating that it is important for physicians to develop an enhanced capacity to step into the shoes of their patients. These studies, and Dr. Calmanâs insights, reinforce the conclusion of the Schulman study, as well as other studies, that it is important to focus on the potential role of racial bias as a possible factor explaining racial and ethnic disparities. More recently, Dr. Kevin Schulman was the co-author of another study similar in nature to his aforementioned 1999 study that provoked such widespread debate. In this study, the research team again used simu- lated patients in an effort to understand the effect of race and gender on provider decision-making, except that medical students, as opposed to physicians, were the providers in question.23 Once again, the researchers concluded that the race and gender of the patient had a significant effect 20 van Ryn, M. et al. 2000. The Effect of Patient Race and Socioeconomic Status on Physi- ciansâ Perceptions of Patients, Social Science and Medicine, 50 (2000):813-828. 21 Cooper-Patrick L. et al. 1999. Race, Gender, and Partnership in the Patient-Physician Relationship, JAMA, 282:583-589. 22 Saha, S. et al. Patient-Physician Racial Concordance and the Perceived Quality and Use of Health Care. Archives of Internal Medicine, 159:997-1004. 23 Rathore, S., Schulman, K. et al. May 2000. The Effects of Patient Sex and Race on Medical Studentâs Ratings of Quality of Life, American Journal of Medicine, 108:561-566.
636 UNEQUAL TREATMENT on the medical studentâs assessment. In other words, race matters for both medical students and physicians alike, according to the two Schul- man studies. Some medical students appear to be entering medical school with certain stereotypical notions of people of other races, and these erro- neous notions can have an ill effect on their ability to treat people of dif- ferent races effectively and fairly. The discussion thus far has focused on physician perceptions and physician behavior in providing patient care. In the debate about the potential role of racial bias in physician decision-making, what is fre- quently overlooked is the perceptions of the patients themselves. As mentioned earlier, Drs. Epstein and Ayanian, in responding to the re- cent Chen article, noted that many physicians may not perceive bias. It is important to understand the starkly contrary perceptions of many patients of color. In a 1999 survey by The Henry J. Kaiser Family Foundation, African Americans were more than twice as likely as whites to state that discrimi- nation in health care is a major problem, almost three times as likely to believe that African Americans receive lower quality health care than whites, and fourteen times as likely to report that they were treated unfairly because of race when seeking medical care in the recent past.24 According to the Public Health Special Report of Seattle and King County, nearly one-third of African-American respondents living in central and southeast Seattle reported experiencing racial discrimination when seek- ing health care, compared to 13% of all respondents.25 Surveys of perceptions undeniably have certain limitations; yet, they are critically important because perceptions so frequently drive behavior, and perceptions are often a function of historical, negative experiences with the health care system. The aforementioned data suggest that there is a strong perception within communities of color that discrimination, whether conscious or subconscious, indeed plays an important role in ex- plaining racial and ethnic disparities in health. Overall, while there are some who believe that the research does not support the conclusion that discrimination plays an important role in ex- plaining racial and ethnic disparities, the weight of the evidence from the research is to the contrary. It would be imprudent to fixate on the use of odds ratios in one study and ignore the strong body of research suggest- ing that physicians are indeed human, and are susceptible to conscious 24 The Henry J. Kaiser Family Foundation. 1999. Race, Ethnicity, and Medical Care: A Survey of Public Perceptions and Experiences. 25 Seattle and King County Public Health Department. 2001. Racial and Ethnic Disparities in Health Care Settings, Public Health Special Report.
637 CIVIL RIGHTS DIMENSION and subconscious racial bias. It would be equally imprudent to ignore the apparently strongly held beliefs of communities of color that discrimina- tion is indeed alive and well in health care. The Need for Broader Research on the Potential Role of Discrimination in Explaining Racial and Ethnic Disparities in Health The Schulman studies and others have triggered a much-needed dia- logue about the potential role of discrimination as one factor explaining the persistence of racial and ethnic disparities in health status. However, it is vitally important not to limit the scope of the research regarding the potential role of discrimination as a factor in racial and ethnic disparities to physician and provider behavior. As noted earlier, there are two types of discrimination under Title VI: intentional discrimination and disparate impact discrimination. Intentional discrimination addresses whether an individual acted intentionally or with a discriminatory purpose. Dispar- ate impact discrimination does not require proof of discriminatory intent, and cases alleging disparate impact frequently address systemic policies or practices that have a disproportionate adverse impact on the basis of race, color or national origin. The overwhelming percentage of actual discrimination cases in health care involves disparate impact. The current research focus on potential discrimination in physician behavior is important, and should continue, but it is not enough. In addition, it is important for researchers to exam- ine the remainder of the health care system, broadly defined, to determine whether there are policies and practices in place that have the effect of discriminating against communities of color. For instance, as discussed in Part three, it is important to examine whether there are policies or prac- tices in place in the process of applying for Medicaid and SCHIP that have the effect of discriminating against communities of color. Another poten- tial area of research may focus on how health care providers market them- selves in their communities. Are there marketing practices in place that have the effect of discriminating on the basis of race, color or national origin? Do referral rules and policies regarding privileges have the effect of discriminating on the basis of race, color or national origin? There are a wide range of systems issues that should be but are not currently, the subject of a sustained research agenda. Overall, it is important to continue to support a robust, broadly de- fined research agenda that focuses on the potential role of discrimination at a number of points in the health care system, and is not limited to the area of physician behavior.
638 UNEQUAL TREATMENT PART THREE: IS DISCRIMINATION ONE OF THE ROOT CAUSES OF DISPARITIES: WHAT DOES THE TITLE VI ENFORCEMENT HISTORY SUGGEST? Introduction There are at least two ways to determine whether discrimination plays a role in explaining racial and ethnic disparities. The first method, discussed in the preceding section, is to look at the research. The second way is to look at actual cases, and this section discusses discrimination cases in the health care setting, and flags emerging civil rights issues in health care. It is not intended as an exhaustive history of discrimination in health care. David Barton Smith has written an exhaustive chronicle of discrimination in health care, from 1920 to the present. In Health Care Divided, Race and Healing a Nation, Smith documents the role of race in shaping our system of medical care, and concludes that discrimination indeed is a force that explains in part the persistence of racial and ethnic disparities.26 This section focuses on the current and recent landscape of Title VI cases, and demonstrates that discrimination regrettably is not a thing of the past. As noted earlier, Title VI and the Title VI regulations are quite broad in their reach. Anyone who receives federal financial assistance of any kind is subject to these anti-discrimination provisions. The regula- tions prohibit not only intentional discrimination but also âfacially neu- tralâ policies or practices that have the effect of discriminating on the ba- sis of race, color, or national origin. Given its broad reach, Title VI has been used in many contexts. For instance, Title VI has been used to challenge the closure of a health care facility in a minority community, the relocation of a health care facility away from a minority community, or the denial of admitting privileges to a physician. This section focuses on six areas that are of particular rel- evance to the disparities debate: (1) intentional discrimination cases; (2) managed care; (3) medical redlining; (4) access to treatment; (5) access to health care for people with limited English skills; and (6) barriers (in addi- tion to language) for immigrant populations. The bulk of the cases dis- cussed herein emanate from OCR. 26 For additional documentation of discrimination in health care, see Watson, S. 1990. Reinvigorating Title VI: Defending Health Care DiscriminationâIt Shouldnât Be So Easy, Fordham Law Review, 58:939-978.
639 CIVIL RIGHTS DIMENSION Types of Discrimination Intentional Discrimination A generation ago, discrimination in health care, like discrimination elsewhere, was quite overt. Hospitals, nursing homes, and other health facilities were segregated, and the challenge was to integrate them. To- day, discrimination generally is more subtle, but no less serious. Yet, from time to time, Title VI has been used recently to attack discrimination that one would think was a thing of the past. A sample of recent cases settled by OCR provides sobering evidence that intentional discrimination, while no longer as prevalent as a generation ago, still rears its ugly head.27 In recent years, OCR has been involved in the following cases: â¢ OCR reached a settlement with a national pharmacy chain that had a franchise that repeatedly refused to fill the prescription of an African- American Medicaid recipient. As part of the settlement, the chain apolo- gized to the victim, posted notices of the companyâs non-discrimination policy in stores nationwide, and posted the same notice in an advertising circular that was distributed nationwide to 25 million people. â¢ OCR settled a case involving a hospital in New York City that had de facto segregated maternity wards; the segregation appeared to be based on source of payment, although source of payment was a proxy for race. â¢ OCR reached a settlement with a hospital located near the U.S.- Mexico border in McAllen, Texas, that ordered its security personnel to dress up in uniforms that closely resembled the U.S. Border Patrol. This policy had the effect of deterring Latinos in the area from using the facility. â¢ OCR reached a settlement with a national home health agency that had a franchise that refused to provide assistance to an African-American family after the family moved from one part of town into a predominantly minority low-income housing development in a different part of town. The investigation revealed that the agency served a predominantly non- minority elderly housing complex that was located a very short distance from the predominantly minority housing development that it refused to serve. â¢ OCR reached an agreement with a hospital in South Carolina that had a policy of prohibiting women with limited English skills from re- ceiving an epidural during labor and delivery. These cases are not meant to illustrate that there is widespread pat- tern of intentional discrimination in health care. Rather, these cases are 27 Additional information about the cases discussed herein can be obtained by contacting the Office for Civil Rights.
640 UNEQUAL TREATMENT set forth as a reminder that while the civil rights landscape in health care is indeed evolving, pockets of straightforward discrimination persist. These cases also illustrate the wide-ranging contexts in which forces of discrimination can inhibit or prevent communities of color from accessing critical services. As noted earlier, focusing exclusively on whether racial bias affects the physician-patient relationship is important, but it is equally important to look at whether discrimination is at work in other entry points within the health care system. Limited English Proficiency Under Title VI and its implementing regulations, health providers who receive federal financial assistance have a legal obligation to ensure that people with limited English skills can meaningfully access health care. Language access cases under Title VI typically proceed under a disparate impact theory pursuant to the Title VI regulations. The failure to provide effective language assistance services amounts to national origin discrimi- nation. There is no need to prove intentional discrimination. In August 2000, OCR published policy guidance in the Federal Register outlining the responsibilities of health and human service providers un- der Title VI to ensure that people with limited English skills can meaning- fully access health and human services.28 It also provides a roadmap to assist providers in meeting their obligations. Although Title VI has been on the books since 1964, the publication of the guidance reflected the first time that OCR had formally published guidance in the Federal Register outlining the requirements of providers under Title VI to serve people with limited English skills, OCRâs process for investigating such com- plaints and initiating compliance reviews, and the steps that providers can take to ensure compliance. The failure of health providers to ensure meaningful access for people with limited English skills is one of the most prevalent civil rights prob- lems in health care. Language access cases are easily OCRâs most fre- quently encountered type of Title VI case, and have been the subject of a steady diet of private litigation. Language access cases are also the most frequent subjects of OCR Title VI compliance reviews and technical assis- tance. According to many immigrant advocates, the large number of OCR complaints actually understates the extent of the problem, because many victims are immigrants who are reluctant to come forward. 28 HHS (Office for Civil Rights). 2000. Title VI Prohibition Against National Origin Dis- crimination As It Affects Persons With Limited English Proficiency: Policy Guidance. Federal Register 65(169):52763-774.
641 CIVIL RIGHTS DIMENSION The language access challenge arises in virtually every health care setting. OCR has investigated and resolved language access complaints in hospital settings, managed care settings, fee for service settings, and human service settings. The most frequently encountered problems are providers who (1) require the patients to provide their own interpreter service, through family members or friends; (2) fail to provide interpreter service, or provide untrained personnel; and (3) subject people with lim- ited English skills to lengthy delays as a result of the lack of readily avail- able interpreter services. The failure to ensure meaningful access for people with limited En- glish skills can have serious, even life or death, consequences. A brief sampling of cases illustrates this point in vivid detail: â¢ OCR settled a case with a hospital in the mid-Atlantic region whose absence of an effective language assistance policy almost had fatal conse- quences for a pregnant woman with limited English skills who came to the hospital with severe bleeding. â¢ A Laotian woman in California who was diagnosed with tubercu- losis was jailed for months because local law enforcement and public health authorities, as a result of communication gaps resulting from the failure to make use of a Laotian interpreter, mistakenly concluded that she would refuse to follow her treatment regimen unless she was jailed. â¢ A young boy in Los Angeles interpreted a consent form for his fa- ther that pertained to his ailing mother. The son thought the form meant that a nurse would make daily visits to care for his mother, and the father signed the form. Instead, the mother was sent to a nursing home. Until recently, language access cases were perhaps the most frequent cases brought by private plaintiffs. However, as noted earlier, the Supreme Court in Sandoval recently made it much more difficult for private litigants to bring language access cases under Title VI. As a result of Sandoval, private litigants can no longer attack defective language access policies using the disparate impact theory under the Title VI regulation. Instead, private liti- gants will have to prove that the failure to provide effective language assis- tance services amounts to intentional discrimination under Title VI itself.29 29 Private plaintiffs may still be able to meet the higher intent standard in language access cases. By virtue of the OCR guidance issued last year, and subsequent outreach efforts, health providers have been put on notice of their obligation under Title VI to ensure mean- ingful access for people with limited English skills. As a result, a private plaintiff can argue that the failure to comply with a civil rights obligation that has been clearly communicated amounts to intentional discrimination under Title VI. In addition, a number of states, such as California, have laws and/or regulations requiring the provision of language assistance services. Private plaintiffs can continue to avail themselves of these provisions.
642 UNEQUAL TREATMENT Given the difficulty that private plaintiffs may encounter as a result of Sandoval, OCRâs role in promoting language access has become even more important. The need for health care providers to ensure meaningful access is growing, as the population becomes more diverse, and immigrant com- munities migrate to both urban and rural settings. According to recent census data, the nation gained more immigrants in the 1990s than in any previous decade. One in five people in America do not speak English at home, and more than 10.5 million said they speak little or no English, up from 6.5 million in 1990. Over 3 million, or approximately one-third of the people who speak little or no English, reside in California.30 Overall, OCRâs enforcement experience, and the recent census data outlined above, demonstrates that language access challenges abound in health care, and the stakes are quite high, and getting higher. For immi- grant populations, the lack of effective language assistance services is one of the most important factors impeding access to health care. Addressing the language access challenge would go a long way in reducing racial and ethnic disparities among many immigrant populations across America. Redlining Redlining is most frequently associated with housing and mortgage lending, and historically related in those settings to the practice of liter- ally drawing a red line around certain poor, predominantly minority sec- tions of cities and refusing to provide a mortgage or sell homeownerâs insurance. If a person could not get a mortgage or homeownerâs insur- ance, it was impossible to own a home. The Department of Justice, and many private civil rights organizations, have successfully pursued a host of redlining cases in the mortgage lending and insurance setting. In the health care setting, there is insufficient evidence to determine the extent of the redlining problem. Unlike the housing context, the prob- lem has not been studied in any detail in health care. Anecdotal evidence shows that redlining has taken a number of forms. In recent years, OCR has handled cases involving allegations of redlining in the home health industry. The cases involved claims that home health agencies refused to provide service to people who resided in certain areas of town. OCR reached settlements in separate cases involv- ing two national home health agency chains. As noted earlier, in one case involving a national home health chain, the investigation revealed that the home health agency, while refusing to serve an African-American cli- 30 Cohn, D., Cohen, S. August 6, 2001. Census Sees Vast Change in Language, Employ- ment, Washington Post, p.A-1, 5.
643 CIVIL RIGHTS DIMENSION ent who lived in a predominantly minority public housing development, regularly served a resident of a predominantly non-minority elderly hous- ing development located a short distance away. Redlining is a concern in the pharmacy setting. A study that sur- veyed 347 pharmacies in New York City found that in nonwhite neigh- borhoods, only 25 percent of the pharmacies had sufficient supplies of pain medication, while 72 percent of white neighborhoods had adequate supplies. Experts disagree on the root cause of this disparity. Some cited potential racial bias, while others opined that low demand in minority neighborhoods, or fear of robbery, were the explanations.31 The National Medical Association (NMA) has raised concerns about redlining in managed care. Specifically, the concern is that managed care organizations are bypassing predominantly minority zip codes in their marketing. To date, the concerns are anecdotal. However, under man- aged care, there are market pressures to ensure that the patient pool of a managed care organization is as healthy as possible. As a result, concerns have been raised that a managed care organization may conclude that a minority community is poorer, sicker, and therefore not an economically viable area in which to conduct business. Managed Care The advent of managed care has raised a variety of civil rights chal- lenges, in terms of both patient care and physician participation. The aforementioned allegations of redlining in managed care are closely re- lated to concerns expressed by a large number of physicians of color that they are terminated from managed care networks, or prevented from joining these networks, because of race. For instance, a 1994 sur- vey of African American physicians found that 92% believed that man- aged care organizations terminated the contracts of African-American physicians more often than those of white physicians.32 At the 2001 an- nual conference of the National Medical Association, an entire afternoon was devoted to discussion of discrimination against physicians of color. The most frequent topic of discussion was discrimination in managed care. The concerns of the NMA and physicians of color may stem at least in part from a number of studies showing that African-American and Latino physicians are more likely to care for African-American and Latino pa- 31 Grady, D. April 6, 2000. Little Access to Pain Drugs in Some Areas, New York Times. 32 Lavizzo-Mourey, R. et al. 1996. The Perceptions of African-American Physicians Con- cerning Their Treatment By Managed Care Organizations, Journal of the National Medical Association, 86:191-199.
644 UNEQUAL TREATMENT tients, respectively, and are more likely to care for poor patients.33 As a result, the practice patterns of many physicians of color may be on a colli- sion course with the market pressures that managed care organizations face to engage in economic credentialing, which relates to the use of eco- nomic criteria, unrelated to quality of care or professional competence, in making contracting decisions.34 A recent federal lawsuit illustrates the concerns of physicians of color. Eight primary care physicians of color were part of a network of physi- cians who had contracts with Humana and its local affiliate in Florida. In their federal complaint, the physicians allege, among other things, that their contracts were not renewed because of their race, and that all 17 white physicians on the Humana panel remained with the managed care organization. Rather than file a claim under Title VI, the plaintiffs relied on 42 United States Code, Section 1981, a civil rights statute that prohibits racial discrimination in contracting and contains broader damages provi- sions than Title VI. Humana contends that race had nothing to do with the decision not to renew the contracts, and that the terminations were for business reasons, which is acceptable under Florida law. In all likelihood, this case will boil down to one question: Were Humanaâs and its local affiliateâs actions motivated by race, or by legitimate business consider- ations.35 This case reflects an emerging civil rights issue in managed care: Does a qualified physician with a sizeable minority patient population have an actionable civil rights claim under Title VI if a managed care organization terminates him or her from a provider network, or refuses to permit him or her to join the provider network? It is impossible to address this ques- tion without gathering additional information. Termination cases are fre- quently complicated by the fact that there are often provisions in the man- aged care contracts permitting the managed care organization to terminate the contract âwithout cause.â Such provision means that even if a physi- cian is delivering quality care to his or her patients, the managed care organization can still terminate the contract. The managed care organiza- tion cannot, however, terminate the contract for a discriminatory reason. As a result, the analysis will frequently return to whether a decision was motivated by business reasons, or discrimination. 33 Kington, R. et al. 2001. Increasing Racial and Ethnic Diversity Among Physicians: An Intervention to Address Health Disparities? The Right Thing To Do, The Smart Thing to Do, Enhancing Diversity in Health Professions, pp. 57-90 (Institute of Medicine). 34 Dallon, C. 2000. Understanding Judicial Review of Hospitalsâ Physician Credentialing and Peer Review Decisions, Temple Law Review, 73:597. 35 Jackson, C. January 8, 2001. Florida Doctors Charge Racial Discrimination By Humana After Being Dropped From Panel, American Medical News.
645 CIVIL RIGHTS DIMENSION âBusiness reasonsâ in this context often does not mean that the physician delivered substandard care. Instead, it often means that the physician spent too much money caring for a patient, so that it becomes economically unfea- sible for the managed care organization to maintain this relationship. As noted above, this places many physicians of color in a difficult conundrum, because they tend to care for a larger percentage of patients of color who are frequently poorer and sicker. It is not difficult to understand why so many physicians of color are concerned about discrimination in managed care. Such discrimination, if present, affects not simply physicians of color, but the dis- proportionately minority patient population that they serve. Proving discrimination is a different matter. Given the prevailing dynamic, it will not be surprising for allegations of discrimination in man- aged care to emerge as one of the primary areas of civil rights litigation in the health care setting.36 Other civil rights in managed care include language access. OCR con- tinues to investigate a steady diet of cases involving managed care orga- nizations. In addition, OCR has performed testing in an effort to measure the extent to which provider networks can meet the needs of people with limited English proficiency. Working in a community with a sizeable Spanish and Vietnamese population, testers contacted providers to deter- mine whether and how they would ensure access to a Spanish- or Viet- namese-speaking patient. A substantial percentage of providers indicated that they would not provide such language assistance. In resolving managed care cases, especially involving Medicaid man- aged care, it is important to work not only with the managed care organiza- tions themselves, but also with the state agencies that oversee managed care programs. As recipients of federal financial assistance, both the state and the managed care organizations are responsible for Title VI compli- ance. Consequently, OCR has worked both with states and with managed care organizations themselves to ensure that individuals with limited En- glish skills enrolled in Medicaid managed care can meaningfully access the programs. The emergence of Medicaid managed care, with its high per- centage of enrollees who do not speak English, heightens the importance of aggressive enforcement of Title VI in the managed care setting. Access to Medical Treatment A number of recent reports have raised questions about the ability of people of color to access certain types of critical health care. For instance, 36 For further discussion of civil rights issues in managed care, see Rosenbaum, S. et al. 1999. U.S. Civil Rights Policy and Access to Health Care by Minority Americans: Implica- tions for a Changing Health Care System, Race, Diversity and Health Care, Improving Access in a Diverse Society (The Henry J. Kaiser Family Foundation).
646 UNEQUAL TREATMENT a lengthy report in Newsday in 1999 documented racial disparities in a wide range of health areas in Queens, Nassau County, and Suffolk County, New York.37 For instance, according to the report, â¢ Even though they die of coronary heart disease at lower rates than African Americans, whites had 138 cardiac bypass procedures per 100,000 residents, as opposed to 31 per 100,000 among African Americans. â¢ One cardiac surgeon from a nationally recognized facility, who drew patients from a broad geographic area, performed 267 bypass op- erations during the year in question; one of the patients was African- American, two were on Medicaid, and three had no insurance; â¢ Another well-respected cardiologist performed 284 inpatient angioplasties during a given year; there were no Medicaid patients; one African American, and two without insurance. â¢ Whites received hip replacements at a rate twice as high as African Americans, knee replacements at a rate 23 percent higher, and gall blad- der operations at a rate 53 percent higher. It is extremely difficult to understand the root causes of these dispari- ties. However, it is impossible to rule out race discrimination. As a result, in 1999, OCR, led by Michael Carter in New York, initiated a series of investigations designed to determine the root causes of disparities in a number of discrete areas. The initial OCR investigation focused on a num- ber of facilities in the New York City area, in part because there was a significant amount of data available from New York State. Using the ser- vices of outside experts, OCR began the difficult process of ascertaining the root cause of the apparent disparities. These analyses seek to address a host of questions, including: â¢ Why is it that of the scores of cardiac catheterizations that a given provider performed over a certain period, only a handful were performed on people of color? â¢ Why is it that a particular facility with a sizeable minority popula- tion within its service area has extremely low utilization rates among this population? There are no easy answers to these questions, but this is the first time that these questions are being addressed through the civil rights prism of Title VI. It may turn out that discrimination is not a factor in explaining these facts. Again, it is important to reiterate that there are two types of discrimination, intentional discrimination and disparate impact, and these 37 Fessenden, Ford. November 29-December 6, 1998. The Health Divide, Newsday.
647 CIVIL RIGHTS DIMENSION investigations have focused on both theories. That is, is intentional dis- crimination somehow at work? Alternatively, and perhaps more likely, is there perhaps something in the hospitalâs marketing plan, referral pat- terns, or rules governing admitting privileges that has the effect of dis- couraging communities of color from using the facility? A recent study on the experience of minority physicians in obtaining referrals for specialists and hospital admissions concluded that African- American physicians were more likely to report problems obtaining hos- pital admissions, and Hispanic physicians were more likely to report prob- lems obtaining referrals to specialists.38 The authors were unable to pinpoint the precise cause of the problems, although they noted that dis- crimination was a potential explanation that could not be ignored. These inquiries are not capable of rapid resolution, but they are criti- cally important to the disparities debate. In addition, it is virtually impos- sible for private litigants to undertake this ambitious task because it is extremely costly, is often more difficult for private litigants to obtain criti- cal information, and the likelihood of recouping the costs of pursuing such an inquiry is speculative. As a result, OCR and the Department of Justice must take the lead in pursuing these questions. Regardless of whether discrimination is the explanation for these ap- parent disparities, there is cause for concern, and time is of the essence. As a result, OCR, in addition to conducting the investigation under Title VI, has embarked upon an ambitious program of outreach and technical assistance designed to bring stakeholders together to discuss the dispari- ties challenge, and identify and implement solutions. For instance, OCR developed a civil rights self-assessment tool designed to assist providers in asking a series of important questions that will allow them to assess current policies and practices, and make the necessary modifications to ensure that the provider is serving communities of color effectively. Over- all, investigation, prevention, outreach, and enforcement are all equal parts of the overall OCR strategy. Discrimination Concerns Unique to Immigrant Populations The immigration explosion of the 1990s has forced communities across America, urban, suburban, and rural, to identify solutions to meet the unique needs of immigrant populations. The language barrier is perhaps the most frequently encountered challenge that adversely af- fects the ability of immigrants to access health care. It is by no means the only challenge. 38 Hargraves, JL. et al. August 9, 2001. Minority Physiciansâ Experiences Obtaining Refer- rals to Specialists and Hospital Admissions, Med Gen Med.
648 UNEQUAL TREATMENT In addition, fear plays an important role in explaining the reluctance of immigrant populations to seek health care. Many citizens and legal immigrants are very reluctant to seek medical treatment, or apply for pub- lic benefits, such as Medicaid or State Childrenâs Health Insurance Pro- gram (SCHIP) benefits, to which they are entitled. They are fearful for at least two reasons: (1) legal immigrants fear that accessing public benefits or uncompensated care will jeopardize their pending application for citi- zenship; and (2) there is a fear that accessing such services or benefits may force them to disclose information about the immigration status of a household member who is undocumented. In reality, these fears are sometimes unfounded, and are based on a misunderstanding of federal policies. For instance, a legal immigrant can access Medicaid and SCHIP benefits without jeopardizing his or her ap- plication for citizenship in any way. Immigration officials do not take this into account in processing the application. In these cases, the solution is aggressive public education to ensure that immigrant populations have an accurate understanding of their rights. Some fears are triggered by policies or practices by governments and health care providers that are neutral on their face, but have the effect of deterring many immigrants from seeking critical benefits to which they are entitled. These are the policies and practices that must be changed in order to maximize participation in critical health care programs. OCR and private advocacy organizations spend considerable time and energy identifying and eliminating unnecessary barriers to participa- tion in critical health programs for immigrant populations. For instance, OCR initiated an investigation of the state of Georgia upon learning that Georgiaâs application for Medicaid benefits required all applicants to cer- tify under penalty of perjury that all members of the household were legal residents of the United States. The only relevant immigration question for Medicaid purposes was the immigration status of the applicant him- self or herself. The effect of including such a question was to deter eligible applicants from applying, because they were fearful that family members who were not documented would be reported to the Immigration and Naturalization Service. This practice not only deterred eligible applicants from applying; it also violated the Title VI regulations because it constituted a policy or practice that had a disproportionate adverse impact on the basis of na- tional origin. As a result, OCR reached a settlement with Georgia, and the application form has been redesigned. OCR then reviewed the application forms for benefits of all states, and found that many states asked a number of irrelevant questions that had the effect of deterring eligible immigrants and citizens from seeking critical benefits. OCR led an HHS effort in conjunction with the Depart-
649 CIVIL RIGHTS DIMENSION ment of Agriculture that resulted in the dissemination of policy guidance to states outlining how states could fine-tune their application forms to maximize participation in critical benefits programs, and avoid potential liability under Title VI.39 Maximizing participation of eligible immigrants in programs such as Medicaid and SCHIP is a critical measure that will assist in reducing disparities and enhancing health status among immi- grant populations. The Social Security Administrationâs âEnumeration at Birthâ program is another example of a laudable initiative that deterred immigrants from accessing critical benefits as a result of one irrelevant question in the ap- plication process. This important program is designed to obtain social security numbers for babies at birth. Obtaining social security numbers at birth enables the baby to become eligible immediately for benefits, such as Medicaid. An implementation problem occurred when the application form required the parents to provide their social security numbers in or- der for the baby to receive a social security number. However, many of the immigrant parents of children born in the United States do not have Social Security numbers. Even though the baby was clearly eligible to receive a social security number, and access benefits, this irrelevant ques- tion had the effect of deterring parents from applying for a social security card. This in turn prevented or delayed the citizen child from accessing critical health benefits that are necessary to a healthy start. Upon learning of the problem in 2000, the Social Security Administration prompted is- sued guidance that corrected the matter, and eliminated this barrier to access for citizen children of immigrant parents. Identifying barriers that prevent or inhibit immigrants from seeking health care is critical to reducing health disparities among this growing population. Upon close analysis, a number of the barriers for immigrants have a civil rights dimension, so that civil rights laws can be employed to eliminate them, and promote access to critical health benefits. PART FOUR: USING CIVIL RIGHTS STRATEGIES TO ASSIST IN ELIMINATING RACIAL AND ETHNIC DISPARITIES The research and the Title VI enforcement experience demonstrate that eliminating racial and ethnic disparities in health is both a civil rights and a public health challenge. As shown above, a number of barriers confronting communities of color have a clear civil rights dimension. 39 Department of Health and Human Services and Department of Agriculture. Septem- ber 21, 2000. Policy Guidance Regarding Inquiries Into Citizenship, Immigration Status and Social Security Numbers in State Applications for Medicaid, SCHIP, Temporary Assistance for Needy Families and Food Stamp Benefits (on file with OCR).
650 UNEQUAL TREATMENT Whether discrimination plays a role in explaining other barriers is less certain, and requires additional inquiry. This section outlines specific rec- ommendations for eliminating racial and ethnic disparities in health that focus on the civil rights dimension of the disparities challenge. 1. Mandate the Collection of Data on Race, Ethnicity, and Language of Preference It is impossible to address racial and ethnic disparities in health status without adequate data. Data provide knowledge, and knowledge pro- vides power to construct effective interventions. Yet, there are currently widespread data gaps that prevent stakeholders from obtaining the nec- essary information that permits the development of a comprehensive plan to eliminate health disparities in a particular area. The need for adequate data collection on race, ethnicity, and language of preference is first and foremost a quality of care issue. For instance, health plans seeking to improve care for minority populations are often hindered because they do not collect data on the race and ethnicity of their members. As a result, it is impossible to study disparities in care, and measure the success of efforts to eliminate disparities in a particular area. Data collection is also useful as a civil rights compliance tool. In other civil rights settings, data on race and ethnicity is routinely collected. For instance, the Department of Education collects data on race and ethnicity in the public school setting. The Equal Employment Opportunity Com- mission (EEOC) collects data on race, gender, and ethnicity of employers with greater than 15 employees. These data are useful to both employers and employees alike. In some cases, data show a pattern or practice of discrimination. On the other hand, the vast majority of employment dis- crimination claims filed with the EEOC result in a finding in favor of the employer, and the statistical evidence can be very helpful in providing the big picture that places an individual case in a useful context. Regrettably, data collection is not required in the health care setting. The Department of Health and Human Services has the authority under the Title VI regulation to require providers and states to collect data on race, ethnicity and language of preference.40 However, HHS does not have the legal obligation under the regulations to require the collection of data on race, ethnicity and language of preference.41 Thus, the decision to require the collection of data on race, ethnicity and language of preference rests within the discretion of HHS, and HHS 40 U.S. Department of Health and Human Services Title VI Regulation, 45 CFR 80.6. 41 Madison Hughes v. Shalala, 80 F.3d 1121 (6th Cir. 1996).
651 CIVIL RIGHTS DIMENSION has exercised this discretion in a piecemeal fashion. For instance, OCR has required data collection as part of a resolution of a particular discrimi- nation case. In addition, the OCR Title VI policy guidance on language access notes the importance of collecting data on language of preference. The recent SCHIP regulation, published June 25, 2001, in the Federal Regis- ter, requires states to collect data on race and ethnicity. The original ver- sion of the SCHIP regulation also required collection of data on language of preference. The final regulation omitted this critical requirement, which is quite regrettable. Nonetheless, the collection of data on race and ethnicity will enable state officials to measure more effectively their level of success in enroll- ing children of color in SCHIP. Many states are already collecting data on race, ethnicity and language of preference. Consequently, there is useful data available in a number of states. It is important to note that collecting data on race, ethnicity and lan- guage of preference does not violate federal law. A number of providers have raised concerns that it is illegal under federal law to collect data on race and ethnicity. To the contrary, as noted earlier, Title VI regulations explicitly empower HHS to require the collection of such data. HHS has taken a number of steps to educate providers on this issue. For instance, in January 2001, OCR and the Surgeon General sent letters to over 30 prominent health care organizations underscoring the importance of data collection, and providing assurances of its legality under federal law. HHS has also funded numerous projects on data collection. Overall, the current data collection system is patchwork at best. The issue of whether the federal government can require providers to collect racial and ethnic data is not a question of statutory or regulatory author- ity. HHS has the authority to require data collection. Mandating data collection boils down to a question of political will. It is useful to look at the evolution of data collection efforts in other civil rights settings in order to fully appreciate the efficacy of data collection. Data Collection in Mortgage Lendingâ The Home Mortgage Disclosure Act The historical debate surrounding the role of discrimination in mortgage lending is quite similar to the current debate in health care. Communities of color had difficulty obtaining a mortgage, and many groups were concerned about redlining. To put it slightly differently, widespread concerns existed about whether there were racial dispari- ties in the rate of denial of applications, and whether these disparities, if any, were a function of discrimination, or some other non-discrimi- natory factor.
652 UNEQUAL TREATMENT In 1968, Congress had passed the Fair Housing Act, which contained an explicit anti-redlining provision. However, problems of discrimina- tion in housing persisted, and Congress in 1975 passed the Home Mort- gage Disclosure Act (HMDA).42 HMDA illustrates how federal regula- tions mandating data collection can serve as a powerful tool for social and policy change to improve the lives of marginalized and disenfranchised minority populations. The stated purposes of HMDA are: (i) To help determine whether financial institutions are serving the housing needs of their communities; (ii) To assist public officials in distributing public sector investments so as to attract private investment to areas where it is needed; and (iii) To assist in identifying possible discriminatory lending patterns and enforcing anti-discrimination statutes. HMDA is a pure data collection statute. There is no private right of action under HMDA, meaning that the enforcement of HMDA is entirely dependent upon federal regulators. HMDA data are insufficient to state a civil rights claim under the Fair Housing Act or any other federal provi- sion. However, HMDA data frequently give a broad snapshot of trends that allows investigators to determine whether additional investigation is warranted. A number of federal agencies are involved in the data collection ef- fort, including the Federal Reserve, the Federal Deposit Insurance Corpo- ration, the National Credit Union Administration, the Comptroller of the Currency, and the Office of Thrift Supervision. Collectively, these agen- cies comprise an interagency entity called the Federal Financial Institu- tions Examination Council (FFIEC). Financial institutions must comply with HMDA, and the term âfinancial institutionâ is broadly defined to include banks, savings associations, credit union and mortgage lending institutions, as well as their subsidiaries. They are required to report a variety of information on their mortgage lending practices to the FFIEC, including but not limited to information relating to the race and ethnicity of applicants for mortgages. From 1975 to 1989, HMDA was in place but had little effect on lend- ing practices. According to one expert, it had little effect because the data were collected but were not widely publicized or collected in a publicly accessible format.43 In 1989, HMDA was amended to make the data pub- 42 12 United States Code 2801. 43 Interview with Kenneth Zimmerman, executive director, New Jersey Institute for Social Justice, Newark, New Jersey.
653 CIVIL RIGHTS DIMENSION lic for the first time. Once the data were publicly disseminated, commu- nity-based organizations and other advocacy groups were empowered because they could use the HMDA data to compare data on area lenders and put pressure to bear on lenders that appeared to have problematic data. It was not until the Sunshine Amendment of 1989 that an industry of social scientists emerged who spent considerable effort breaking down HMDA and other data to pinpoint the potential role of various explana- tory factors, including but not limited to discrimination. As a result of these analyses, investigators were able to answer for the first time the critical question: did discrimination infect the process, or were the dis- parities a function of other, non-discriminatory factors? In a number of cases, discrimination was pinpointed as a root cause, and the Department of Justice settled a number of mortgage lending cases during the 1990s, including: â¢ United States v. Albank: The bank agreed to provide $55 million in loans at below market rates to settle a suit alleging that the bank refused to take mortgage loan applications from significant minority populations. â¢ United States v. Decatur Federal Savings and Loan: The bank agreed to pay $1 million to compensate 48 victims of discrimination after extensive review of bank records revealed that the bank applied stricter underwriting standards to African-American applicants than it did to white applicants. The Lessons of HMDA There are a host of lessons that can be gleaned from the HMDA expe- rience. First, HMDA has given meaning to the adage that âknowledge is powerâ because it has assisted in answering critical questions about the role of discrimination. Knowledge can also be powerful in the health care context. Second, HMDA illustrates that it is not simply important to collect data; rather, it is important to collect the right data. In the health context, stakeholders can and must develop consensus on standard data collec- tions methods, as well as the types of data that should be collected. Oth- erwise, it becomes difficult to compare and analyze data. Perhaps data collection will differ between geographic areas. However, it is important to develop a commonly accepted language of data collection. Third, it is also important to reevaluate the data being collected on a periodic basis. In the HMDA context, for instance, there currently is no data being collected by race and ethnicity on the interest rate that is being charged. With the emergence of predatory lending as a formidable civil rights challenge, the absence of these data is problematic. Similarly, in the
654 UNEQUAL TREATMENT health care settings, it is important to reexamine data collection protocols regularly and adjust to meet emerging concerns. Finally, and perhaps most importantly, it is not simply important to collect data. It is also important to report data, and ensure that data are accessible to the public at large. The 1989 Sunshine Amendment marked a critical turning point under HMDA. Similarly, allowing public access to data will enable policymakers and the public at large to have a better handle on critical questions and potential trends. In so doing, it is impor- tant to be mindful of considerations of medical records privacy, as well as the recently enacted federal medical records privacy regulation. How- ever, it is possible to obtain basic necessary information without running afoul of the privacy regulation. Racial Profiling: The Importance of Data Collection Racial profiling by law enforcement is one of the most frequently dis- cussed civil rights issues. Once again, the debate mirrors the racial dis- parities discussion. Critics contend that police officers target African Americans and Latinos on account of their race, and stop them improp- erly and illegally. Police officials counter that a personâs race is not the basis for a traffic stop. Rather, a police officer takes a host of factors into account. Racial profiling was the subject of debate during the past presi- dential election race, with both candidates decrying the practice. There is considerable disagreement about how to go about answering the question of whether police officers are engaging in racial profiling. The Special Litigation Section of the Civil Rights Division of the Depart- ment of Justice has the authority to investigate racial profiling cases. It has concluded that an effective means of resolving this issue is to collect data on traffic stops. As a result, a number of recent consent decrees in police misconduct cases have included provisions mandating the collection of racial and ethnic data on traffic stops. Agencies that have agreed to col- lect data include the Pittsburgh Police Department and the New Jersey State Police.44 Data collection is not limited to consent decrees. According to a Feb- ruary 2000 fact sheet issued by the Bureau of Justice Statistics of the United States Department of Justice, a total of 37 police agencies collect racial demographic information on traffic-related arrests with close to a dozen states mandating such efforts. This number is growing. Congress is de- bating a bill that would mandate the collection of data on race and ethnicity in traffic stops. If passed, this bill, among other things, would 44 To review these consent decrees, see www.usdoj.gov/crt/split.
655 CIVIL RIGHTS DIMENSION require the development of a uniform system for collecting the data, so that the appropriate data are collected. Critics are concerned that the data will somehow be misused. The federal legislation addresses this concern with its provision for the devel- opment of a uniform data collection system. Opponents of data collection continue to express concerns about potential misuse of data, and a belief that collecting the data will not answer the question. Regarding data collection, two things appear clear in both the racial pro- filing and the health care context. First, simply collecting the data will not answer the question of whether discrimination is at work. The data will have to be analyzed and, as shown in the HMDA setting, such inquiry can shed light on critical questions. Second, failing or refusing to collect data guaran- tees that the critical questions about the root causes of disparities will never be answered definitively. As one high-ranking official of the San Diego County Sheriffâs Department stated, after the Department agreed to collect racial and ethnic data on traffic stops, â[the community] deserve[s] an an- swer on whatâs really going on here.â45 The public also deserves an answer on what is really going in the disparities context, and the collection of racial and ethnic data holds a key to answering so many critical questions. Conclusions Regarding Data Collection Effective data collection is the lynchpin of any comprehensive strat- egy to eliminate racial and ethnic disparities in health. Collecting data on race, ethnicity, and language of preference is a quality of care as well as a civil rights issue. Health providers should not wait for HHS to mandate the collection of data relating to race, ethnicity and language of prefer- ence. Fortunately, many health care providers have already begun to col- lect these data, and many states also require the collection of data. But these patchwork efforts are not nearly sufficient. HHS must exert leader- ship and mandate the collection of data, and assist in the development of uniform data collection systems. The expertise certainly exists within HHS to perform this critical task. 2. Strengthen the Federal, State, and Private Health Care- Related Civil Rights Infrastructure Discrimination in health care is the forgotten frontier of civil rights. It is seldom discussed. There are relatively few private advocacy organizations 45 February 26, 1999. San Diego To Study Racial ProfilingâOfficers Will Collect Race Data on Every Traffic Stop, www.apbnews.com/cjprofessionals/behindthebadge/1999/02/26/ sandiego0226_01.html.
656 UNEQUAL TREATMENT involved in a steady diet of health care related civil rights advocacy. The National Health Law Program is perhaps the most effective national advo- cacy organization that focuses on the intersection of health care and civil rights. Due to resource constraints, other private civil rights organizations have been forced to scale back their health care practices in recent years. The Department of Justiceâs Civil Rights Division has 10 different sec- tions that address critical areas such as education, employment, housing, disability rights, and the like. There is no section whose exclusive or even primary mission is to address the intersection of health care and civil rights. There are sections that address health-related civil rights issues, and these sections make an important contribution. However, given the importance of health care, and the history of discrimination in health care, it would be useful if health care were on equal footing with education, housing, employment, and other critical building blocks of self sufficiency. The Office for Civil Rights at HHS is the lead federal agency combat- ing discrimination in the health care setting. The first budget of HHS OCR (fiscal year 1980) was $22 million, which supported approximately 550 employees. The budget remained stagnant for the ensuing two de- cades, and the budget for fiscal year 2000 was also $22 million, which supported only 215 employees. As a result, it has been difficult for OCR to carry out its critically important mission in a fully effective manner. Nonetheless, OCR has established a body of cases that document continu- ing instances of discriminatory activity, in violation of Title VI. A 1999 report from the U.S. Civil Rights Commission was quite criti- cal of some of OCRâs work in the Title VI context, and cited the chronic underfunding. Among other things, the Report recommended a substan- tial boost in funding for OCR. The budget for fiscal year 2001 increased substantially to $28 million. However, this substantial increase is only a fraction of OCRâs true need. OCRâs staff has decreased over 50 percent during the past 20 years, even though its enforcement responsibilities have increased dramatically with the passage of new civil rights laws, such as the Americans with Disabilities Act. As a result of Sandoval, OCRâs role in enforcing Title VI has become even more important, because it is more difficult for private plaintiffs to bring many Title VI claims. In order to address the disparities challenge effectively, OCR must have substantial increases in its budget. The substantial hike in 2001 was important, but should be viewed as just a start. HHS as a whole is spend- ing hundreds of millions of dollars addressing racial and ethnic dispari- ties. OCR should have a much greater share of these dollars. Among other things, increased resources would enable OCR to develop the insti- tutional capacity to perform the sophisticated analyses necessary to an- swer difficult questions discussed earlier, such as why so few people of color are accessing certain medical procedures in a particular facility.
657 CIVIL RIGHTS DIMENSION State governments must also expand their health care-related civil rights enforcement infrastructure. California, for instance, has more people with limited English skills than any other state. Yet, the state office responsible for ensuring that counties comply with their civil rights obligations in the language access context has only a handful of people. It is simply impossible to carry out effective enforcement and compliance monitoring with a skeleton crew. The foundation community has been supportive of private advocacy organizations that are involved in a wide-ranging set of initiatives de- signed to address racial and ethnic disparities in health. The California Endowment, for instance, has made a major investment in the area of language access. However, additional foundations must step to the plate and expand the capacity of private organizations to address these dispari- ties. In some circumstances, such as combating discrimination in man- aged care, this may mean supporting litigation activities, which a number of foundations have been reluctant to do. Strengthening the governmental and private civil rights infrastruc- ture will go a long way toward ensuring that the civil rights concerns are addressed. 3. Develop a Comprehensive Language Access Agenda Ensuring language access for people with limited English skills is ar- guably the most important measure that could be taken to reduce dispari- ties among the rapidly expanding immigrant populations. Ensuring lan- guage access is also a quality of care issue. The solutions are not difficult to envision. In fact, many providers across the country, large and small, urban and rural, have put into place model programs that are both cost- effective and are improving the quality of life for people with limited En- glish skills. Yet, the solutions are all too elusive in too many parts of the coun- try, and the demand for these services is increasing. Under the leader- ship of Secretary Donna Shalala, HHS focused substantial time and en- ergy on language access issues. The OCR Title VI policy guidance was issued in August 2000, on the heels of an Executive Order in which former President Clinton directed leaders across the federal government to address language access challenges. Following the policy guidance, HHS developed an agency-wide Strategic Plan on language access, de- signed to ensure that HHS has the capacity in its own programs to en- sure meaningful access to people with limited English skills. In short, HHS has attempted to set the tone on language access, and these efforts are continuing.
658 UNEQUAL TREATMENT A number of additional steps can be taken on the language access front. Potential steps include the following: A. More Foundations Must Get Involved Foundations must continue to provide support to language access ef- forts. For instance, there are many best practices across America, but sur- prisingly little sharing of information. Foundations can promote sharing of best practices, as can HHS. Foundations must also continue to support efforts to implement innovations. The United Hospital Fund, for instance, has been instrumental in bringing together providers and community- based organizations to develop a promising system of medical interpreta- tion that uses wireless remote technology and would enable providers to meet a wide range of need. More foundations need to make a substantial commitment to language access. B. Educate and Train Providers on Their Obligation to Provide Lan- guage Assistance Services A massive training and education campaign must be undertaken to ensure that providers understand their obligations to ensure meaningful access to people with limited English skills, and also appreciate that it is possible to implement solutions that work and are not prohibitively ex- pensive. OCR spends as much time on language access issues, including training and outreach, as any issue in its portfolio of responsibilities. De- spite these efforts, and despite the recent publication of the OCR guid- ance, all too many providers are unaware of their responsibilities under Title VI to ensure meaningful access to people with limited English skills. Foundations and HHS should support the establishment of technical assistance centers that would assist providers in developing language as- sistance programs, educate communities on their rights under Title VI, and conduct research on a number of critical issues in the language access context, such as whether the provision of effective language assistance services is actually cost-effective. Providers frequently express a desire for technical assistance in developing effective programs. However, some providers are reluctant to reach out to OCR for assistance, because they fear that OCR will conduct a review and find them in violation of Title VI. They are sometimes fearful of reaching out to community-based organi- zations or advocacy groups, because they perceive that these groups may sue them. Technical assistance centers would be the neutral entity with expertise to assist providers in a non-threatening manner, disseminate best practices, and perform research. C. Address the Critical Financing Challenge Any plan to enhance language access must address the financing chal- lenge. For providers, the financing challenge is the critical issue in lan- guage access. Many providers contend that they are willing to comply
659 CIVIL RIGHTS DIMENSION with Title VI, but simply cannot afford the cost of an interpreter. For instance, following the issuance of the OCR policy guidance on language access, the American Medical Association expressed concerns to then Sec- retary Shalala and, subsequently, to Secretary Thompson that the costs of compliance for physicians would be prohibitive. Cost concerns are certainly understandable. It is important to learn from the numerous providers and states that have developed comprehen- sive programs that have not placed the exclusive burden of compliance upon physicians. For instance, Washington State has developed an im- pressive program of cost reimbursement for language assistance that makes substantial use of Medicaid matching funds. As a result, if a phy- sician is seeing a Medicaid patient with limited English skills, he or she simply makes arrangements with the state agency that will provide a qualified interpreter at no charge to the physician or patient. The state also provides translation of critical documents and forms, such as consent forms, into over 60 languages. Washingtonâs experience illustrates that it is possible to meet the fi- nancing challenge. However, meeting the challenge requires leadership, ingenuity and financial commitment from the state. It can be done, but Washington regrettably is an exception rather than the rule. The lesson from the Washington experience is that the financing challenge requires leadership from state government leaders. In the Medicaid and SCHIP context, HHS already provides matching funds for the costs of language assistance services, and there is no upper limit on the amount of matching funds that HHS will provide. States must step forward and commit to tapping into these matching funds by declaring that language assistance services are essential services, just as a physician is essential. More states need to follow Washington Stateâs lead. D. Encourage Innovation There are a number of promising practices in language assistance, and the federal government and private foundations must expand their support for innovation. The Office of Minority Health and the Agency for Health Research and Quality (AHRQ) within HHS, along with a number of private foundations, have funded a number of innovative programs. However, other federal agencies, as well as additional foundations, can and should support these efforts. E. Develop a Research Agenda on Language Access A robust research agenda exists in language access, and should be supported by HHS and private foundations. Questions that merit further research include: â¢ Is the provision of adequate language assistance services cost- effective for providers?
660 UNEQUAL TREATMENT â¢ Are certain types of language assistance more effective in ensuring meaningful communication? â¢ Does fluency in English affect access to critical services, such as immunizations? â¢ What do the most effective language assistance programs have in common? Overall, language access is the low hanging fruit of racial disparities in that many promising interventions have been identified, and are ca- pable of implementation. Yet, the low hanging fruit is still on the tree, and additional leadership and visibility must be given to this issue. 4. Identify and Eliminate Other Barriers for Immigrants Language access is critical for immigrant communities. However, all stakeholders must be vigilant in continuing to identify and eliminate addi- tional barriers that inhibit many legal immigrants from accessing critical health care. OCR has led a nationwide initiative to examine application forms for public benefits to ensure that they do not make irrelevant inquir- ies that have the effect of chilling participation in programs such as Medic- aid and SCHIP. Considerable work lies ahead on this issue, and other bar- riers frequently emerge. Community-based organizations are critical to this effort, as their collective fingers are closest to the pulse of immigrants, and they enjoy the trust of immigrants. As a result, it is important to support their efforts, and important for government to partner with community- based organizations on outreach and education campaigns. 5. Preventing Discrimination Through Education of Providers and Patients In addressing the root causes of disparities, Dr. Jack Geiger noted re- cently that racial bias and lack of cultural competence on the part of health care workers âmay be the most directly remediable problem, if they are honestly recognized and if programs are designed to address them.â46 Many providers and schools have begun to implement programs of study on cultural competence. The Office of Minority Health at HHS issued Standards on Cultural and Linguistic Competence in 2000.47 There is sub- 46 Geiger, J. June 12, 2001. Racial Stereotyping and Medicine: The Need for Cultural Com- petence, Canadian Medical Association Journal, 164(12):1699-1700. 47 Department of Health and Human Services. 2000. Office of Minority Health National Standards on Culturally and Linguistically Appropriate Services (CLAS) in Health Care. Federal Register 65(247).
661 CIVIL RIGHTS DIMENSION stantial financial support from governments and foundations for these overall efforts to promote cultural competence. There are proposals to make such efforts a requirement for the accreditation of health profes- sional schools. But the work is just beginning, and it is necessary to ex- pand these efforts, and to develop tools that will enable policymakers and providers to measure the effectiveness of these training programs. This training must include discussion of civil rights obligations of health care providers. It is important not only to educate health care providers, but also to educate and empower patients themselves, and assist them in navigating the often-treacherous terrain of health care. Harlem Hospital in New York, for instance, has implemented a patient navigator program designed to provide advocates for patients who can assist them in asking the ap- propriate questions, and making the necessary inquiries as they access the health care system. Many immigrants are reluctant to question au- thority figures, such as physicians, and patient navigator programs are helpful in ensuring that the necessary questions are asked. Community- based organizations can play a critical role in educating and empowering patients. It would be useful to implement and study the effectiveness of a number of patient empowerment programs that affirmatively reach out to vulnerable populations and educate them about a wide panoply of is- sues in health care, including but not limited to civil rights. 6. Fix Sandoval Congress should act to restore the status quo that existed prior to Sandoval by passing legislation to reestablish that there is a private right of action for disparate impact discrimination under the Title VI regulation. The failure to restore the private right of action will mean that private advocacy organization will have to fight many health care discrimination battles with one hand tied behind their backs. The failure to restore the private right of action will also place more pressure on an already over- burdened OCR to pursue disparate impact cases. Sandoval is not limited in its impact to language access cases. Private plaintiffs interested in pursuing disparities complaints in other contexts, such as access to treatment in hospitals, will find it more difficult, if not impossible, to make use of Title VI. For instance, if the evidence demon- strates that a hospital policy, such as a referral practice or admission policy, is having a disparate impact on the ability of minorities to use the facility, a private plaintiff would be prohibited under Sandoval from suing the hospital under Title VI. Proving disparate impact discrimination under Title VI is hard enough, given the difficulty of unpacking the complex maze of interactions in the
662 UNEQUAL TREATMENT health system. Sandoval provides an additional disincentive for an already small cadre of lawyers who address civil rights issues in health care. 7. Develop the Capacity and Infrastructure to Address Critical Civil Rights Questions in Managed Care Some of the most complex civil rights issues occur in the managed care context. Racial disparities may be a function of a potential collision of market forces and practice patterns of minority physicians, who tend to have patient pools that are disproportionately minority, poor, and poten- tially less healthy. Given this potential collision, it is not difficult to un- derstand why over 90 percent of African-American doctors believe that managed care organizations discriminate against them in contracting. The critical question is whether the perceptions of discrimination are the real- ity, or whether managed care organizations are engaging in legitimate business practices. Regrettably, there is very little research that has addressed this criti- cal question.48 There are few lawsuits that have addressed this issue; how- ever, as a result of continuing concerns raised by the National Medical Association and its membership, this may change. Grantmakers both within and outside government should support efforts to study this issue in greater depth, and attempt to determine whether the perceptions of many physicians of color are accurate or not. The lack of a sufficient pri- vate infrastructure of organizations that address discrimination in health care is hindering the effort to answer these critical questions. OCRâs ac- tivities in managed care have focused primarily on language access is- sues. Issues of redlining in managed care and potential discrimination in selection and de-selection of physicians for provider networks, have re- ceived less attention, not because they are less important. Supporting a research agenda in this area, and supporting the efforts of private organi- zations to study the critical questions of potential discrimination in man- aged care, either in a litigation or non-litigation context, is critical to ad- dressing the broader concerns once and for all. 8. Perform a Civil Rights Self-Assessment It is exceedingly difficult, time-consuming, and costly to perform the sophisticated regression and other analyses that will enable investigators 48 For a further discussion of civil rights issues in managed care, see Rosenbaum, S. 2001. Racial and Ethnic Disparities in Health Care: Issues in the Design, Structure and Adminis- tration of Federal Health Care Financing Programs Supported Through Direct Public Fund- ing, Institute of Medicine Committee on Understanding and Eliminating Racial and Ethnic Dispari- ties in Health Care.
663 CIVIL RIGHTS DIMENSION and researchers to pinpoint with precision the precise causes of a particu- lar disparity. Many providers are not waiting for the completion of this analysis, and initiated a series of activities, such as cultural competence training and review of language access programs. As part of its outreach and technical assistance, OCR, led by Michael Carter, regional manager in New York, developed a civil rights self-assessment tool. This instrument is aimed at enabling providers to ask a series of important questions that will enable the provider to assess current policies and practices, modify policies and practices that may have a discriminatory effect, and institute innovative measures to ensure that communities of color can have mean- ingful access to the facility. The self-assessment tool is a prevention tool designed to encourage voluntary compliance efforts. Providers should consider using this tool or some other instrument that will enable it to perform a critical civil rights self-assessment. CONCLUSION Eliminating racial and ethnic disparities in health status is a moral imperative for which there is no single magic bullet. As policymakers, politicians, and the health care profession grapple with how to eliminate disparities, it is time to acknowledge certain realities about disparities. It is indeed deadly to be poor in America, and communities of color are disproportionately poor. Yet, the disparities story does not end with eco- nomics, and any story that does is simply incomplete. Similarly, a dis- parities story that ignores the potential role of race and racial discrimina- tion is also incomplete. This paper has documented specific settings where forces of discrimi- nation are at work in the health care system, and has identified other ar- eas where there is insufficient information to make a definitive judgment about the role of discrimination. It is critically important to understand the civil rights dimension of the disparities challenge, and acknowledge the unpleasant reality that race often matters in health care. It is equally important to implement solutions, many of which are described above, to address areas where discrimination is clearly at work, and to support the necessary research and advocacy to resolve the unanswered questions.