Reducing Racial and Ethnic Disparities in Health Care

A sample of recent headlines reveals that disparities in U.S. health care persist:

Health Improves but Racial Disparities Remain (Associated Press in the New York Times, January 25, 2002a)

For Tribes, Traditions May Be Key to a Healthier Future; In Indian Country, the Battle Against Diabetes Draws on Native Traditions—and Emerging Ideas About “Culturally Appropriate” Public Health (Pember in the Washington Post April 9, 2002)

Racial Gap in Cancer Survival Is Not Biological (Reuters in the New York Times, April 24, 2002)

Racial Disparity Is Found in AIDS Clinical Studies (Stohlberg in the New York Times, May 2, 2002)

Hispanic Children Have Health Risks at Greater Rates (Associated Press in the Wall Street Journal, July 3, 2002b)

As these headlines demonstrate, issues of racial and ethnic disparities in health and health care are common and of major concern in the United States. The Institute of Medicine (IOM) report, Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care (Smedley, Stith, and Nelson 2002), has focused additional attention on the topic by providing a comprehensive examination of disparities and inequities in our health care system. The report reviews multiple studies from the past decade and concludes that racial and ethnic minorities receive inferior quality of care compared with whites, even after controlling for insurance status and socioeconomic status. The report recommends a comprehensive strategy to eliminate these widespread disparities, including steps to improve access, increase the practice of evidence-based medicine, and improve patient–provider communication, as well as efforts to improve the collection, reporting, and monitoring of data.

The Agency for Healthcare Research and Quality (AHRQ) has supported efforts to identify and ameliorate racial and ethnic disparities since its inception in 1989. Identification of differences in care associated with patient race or ethnicity was a natural outgrowth of efforts to identify and understand practice variations as a prerequisite to improving care (Stryer et al. 2000). The AHRQ's current initiatives are now an essential component of a larger federal effort to close the unacceptable gaps in health and health care among racial and ethnic groups in the United States (Satcher 1999).

Two recently enacted laws guide AHRQ's current programs: the Agency's reauthorization (Healthcare Research and Quality Act of 1999; U.S. Public Law 106-129) and the Minority Health and Health Disparities Research and Education Act of 2000 (U.S. Public Law 106-525). Both laws, informed by prior, largely descriptive health services research, now challenge the research community to move from documenting disparities to understanding why they occur and evaluating potential solutions. This article provides a selected overview of current AHRQ initiatives, including research, training, and data development for researchers.

New Directions

The AHRQ's reauthorization directed the Agency to establish the Office of Priority Populations Research, focused on racial and ethnic minorities, as well as low-income populations, women, children, elderly, and individuals with special health care needs, including individuals with disabilities and individuals who need chronic or end-of-life care. Both laws direct AHRQ to publish the National Healthcare Disparities Report (NHDR) annually, starting in 2003, to describe prevailing disparities in health care delivery associated with the aforementioned population groups. The NHDR will be developed as a companion to the National Healthcare Quality Report (NHQR) for which equity has been identified by the Institute of Medicine as a key dimension (Hurtado, Swift, and Corrigan 2001).

New Knowledge

Work on disparities fits squarely with the Agency's strategic goals: to support improvements in health outcomes, to strengthen quality measurement and improvement, and to identify strategies that improve access, foster appropriate use, and reduce unnecessary expenditures. While much of the Agency's disparities portfolio examines inequities for discrete subsets of the American population, the implications of the research are quite broad; the entire American health care system will benefit from these efforts to address critical issues of equity and fairness. Indeed, the increased diversity of the U.S. population demands that efforts to improve quality focus on those at highest risk of receiving poor care.

Many studies funded by AHRQ have helped to identify racial and ethnic inequities and to improve our understanding of the magnitude of the challenge the nation faces. For example, a study by John Canto and others reveals that black Medicare beneficiaries are significantly less likely than white Americans to receive reperfusion therapy for acute myocardial infarction (Canto et al. 2000). This potentially life-saving therapy is used in eligible black women and men in 44 percent and 50 percent of the cases respectively as opposed to 56 percent and 59 percent in eligible white women and men. Numerous disparities have also been identified in procedures that affect the quality of life. One study of total hip replacement demonstrates that Hispanic Americans are significantly less likely to undergo the procedure. The difference persists for Medicare patients, suggesting that the inequity could not be attributed to differences in insurance status (Escalante et al. 2002). Other work has revealed that:

• Women of Asian, African, and Hispanic descent wait more than twice as long as white women between having an abnormal screening mammogram and receiving follow-up testing to diagnose breast cancer (Chang et al. 1996).

• African American women who have fibroids are more likely than white women to have them surgically removed with a procedure that preserves the uterus (Agency for Healthcare Research and Quality 2001a).

• Minority racial and ethnic groups are at increased risk of being under assessed and under treated for cancer pain (Agency for Healthcare Research and Quality 2001b).

• African American and Hispanic American HIV patients are only about half as likely as non-Hispanic whites to participate in clinical trials of new medicines designed to slow the progress of HIV (Gifford et al. 2002).

Other AHRQ-funded work has also helped to clarify some of the underlying causes of disparities. For example, Hispanic Americans with diabetes often face economic barriers to treatment and are reluctant to place their own medical needs over the needs of family members. Other common barriers include a distrust of insulin therapy, a preference for more familiar traditional remedies, and a fatalistic acceptance of the course of the disease (Lipton et al. 1998; Noel et al. 1998).

In an effort to further improve our understanding of why disparities exist and to identify effective strategies to eliminate them, AHRQ supports EXCEED (Excellence Centers to Eliminate Ethnic/Racial Disparities). Under EXCEED, a five-year, $45-million initiative, the Agency is supporting nine centers across the country, each of which is performing a number of projects that are related by a common theme. The themes include communication, cultural competence, Native American elders, and the interaction of the patient, the provider, and the health care system. For example, the center at the University of California, San Francisco, led by A. Eugene Washington, M.D., M.Sc., is addressing effective communication and decision making for diverse populations. Projects include examining patient and clinician factors influencing variations in management of coronary artery disease, communication of risk related to common cancers, and the relationship between race/ethnicity and trust in doctor–patient relationships.

Additional ongoing projects related to disparities include:

• Development and testing of quality-of-care measures for hypertension in a population of Hmong refugees (Wong 1999), and of patient-centered quality-of-care measures for Asian Americans of Chinese and Vietnamese descent (Phillips 1999).

• Exploring the extent to which citizenship plays a significant role in access to Pap smears for Latina women (Katzburg 2001).

• Studying psychosocial influences on regular mammography screening among African American and white women (Calvocoressi 2001).

• Determining the effect of formally trained Navajo medical interpreters on diabetes outcomes (McCabe 2000).

New Tools and Talent

Databases

In addition to funding discrete research projects, AHRQ supports the development of databases for research. Three AHRQ-sponsored databases and tools provide the basis for a wide variety of research on racial and ethnic disparities in care: the Healthcare Cost and Utilization Project (HCUP), the Medical Expenditure Panel Survey (MEPS), and the Consumer Assessment of Health Plans (CAHPS®). All are available as public use files for interested researchers.

The HCUP currently provides encounter-level administrative data for inpatient hospital stays in 29 states and ambulatory surgery encounters in 15 states. Research using HCUP data finds that Hispanic Americans have higher rates of inpatient noninvasive diagnostic testing for cerebrovascular disease while having lower rates of invasive testing and therapeutic procedures (Elixhauser et al. 2002). The HCUP data also demonstrate that there were no substantial changes in racial/ethnic disparities in heart, liver, and kidney transplants between 1988 and 1997 (Xiao, Campbell, and Song 2002).

The MEPS is a nationally representative, longitudinal survey that collects detailed information on the specific health services that Americans use, how frequently they use them, the cost of these services, and how they are paid for. The MEPS-based studies show that while African American children are less likely than white children to use any prescription drugs, similar proportions of Asian and Hispanic American children use prescription drugs as compared to white children (Chen and Chang 2002). Similarly, while Hispanic Americans aged 65 and older are about as likely as white Americans to receive influenza vaccinations, African Americans are considerably less likely to be currently vaccinated (Marin, Johanson, and Salas-Lopez 2002).

The CAHPS is a set of survey and report tools that provides reliable and valid information to help consumers and purchasers assess and choose among health plans. The National CAHPS Benchmarking Database (NCBD) facilitates comparisons of CAHPS results among different sponsors, including Medicaid agencies, public and private employers, and individual health plans. Research using the NCBD shows that Hispanic Americans and Native Americans generally report similar experiences with health care as white Americans, while Asian Americans/Pacific Islanders report worse access to care, promptness of care, and communication with providers (Morales et al. 2001). Other studies describe that, compared with white Americans, Asian and Hispanic American parents who speak a language other than English at home report worse timeliness of care, staff helpfulness, and plan service for their children, while those who speak English at home report care similar to that of white American parents (Weech-Maldonado et al. 2001).

Training

The AHRQ supports the training and development of minority researchers and those interested in minority health through a number of mechanisms. Both pre- and postdoctoral training are available through the National Research Service Award (NRSA) program. These programs support trainees for one or more years of full-time supervised experience in health services research. Dissertation grants for those enrolled in doctoral programs are also available and minority researchers are strongly encouraged to apply (Agency for Healthcare Research and Quality 1998). Other opportunities for individuals include minority supplements to ongoing studies. Under this support, grants and cooperative agreements can be expanded to involve a minority researcher or to investigate additional questions that relate to minority health issues.

The Minority Research Infrastructure Support Program (M-RISP) provides institutional research development support in order to enhance the infrastructure and capacity for health services research of minority institutions. It also provides funding for an individual investigator's involvement in projects and for developing a new investigator's capacity to conduct focused research, with the expectation that such experiences may lead to applications for larger-scale health services research grants. Last year, AHRQ funded institutions in Texas, Hawaii, and Tennessee under this program.

An additional program, Building Research Infrastructure and Capacity (BRIC), is intended to enhance the competitiveness of institutions in areas that have received little funding in the past. While not specifically focused on minority health, this program has also expanded opportunities to address disparities. For example, a partnership under this program between Mississippi State University and the American Academy of Pediatrics is addressing the health needs of minority, low-income, and underserved children in the Mississippi Delta (Agency for Healthcare Research and Quality 2001c).

New Research Models

An important complement to support for research projects, data development, and training has been the development of two new research models to facilitate the rapid and efficient study of racial and ethnic disparities. Integrated Delivery System Research Networks (IDSRNs) were initiated in 2000 and link researchers with large health care systems. Two examples of disparities-related projects utilizing the IDSRN mechanism are:

• Examining the capacity of health plans to study racial/ethnic disparities using data from managed care organizations and identifying strategies to improve such capacity;

• Exploring the role that differential diffusion of technology plays in contributing to racial, ethnic, and socioeconomic disparities in health care (Agency for Healthcare Research and Quality 2002a).

Practice Based Research Networks (PBRNs), each of which is composed of a group of primary care providers or clinics, work collaboratively to investigate topics that arise in the course of routine clinical practice. An underlying premise is that because of the relevance of the questions and their involvement in the research process, participants will be very motivated to assimilate findings into practice. Since 2000, AHRQ has funded numerous PBRNs that have enhanced our capacity to study ethnically and socioeconomically diverse populations. For example, the Southeast Regional Clinicians' Network is a partnership of health professionals from 142 federally funded community health centers in eight southern states and the Morehouse School of Medicine (Agency for Healthcare Research and Quality 2000a).

Finally, AHRQ is in the process of increasing its activities in community-based participatory research (CBPR), a method of conducting research that actively involves communities in all stages of the research process. Substantial involvement of community members and representatives of community-based organizations offers researchers insights into research priorities, assistance in the development of valid hypotheses, recommendations regarding methods that are sensitive to the needs and values of research subjects, and perspectives to interpret data. Also, involvement of communities increases the likelihood that findings will be incorporated into practice rapidly (Agency for Healthcare Research and Quality 2002b). While the use of CBPR is not limited to research on minorities, it has been shown to be an effective process with which to address disparities (Israel et al. 1998).

Translating Research into Practice (TRIP)

A crucial component of AHRQ's work in the area of racial and ethnic disparities is aimed at reducing the gap between what is known and what is done. One component of the Agency's TRIP activities seeks to improve our understanding of what interventions are effective for changing practices. The Agency has made a major investment in TRIP trials, a subset of which is focused on testing interventions in environments with large numbers of minority patients. For example, a study led by Perla Vargas of the Arkansas Children's Hospital is examining the effectiveness of an evidence-based asthma care management program focused on minority children enrolled in Head Start programs. Other studies include evaluations of culturally competent interventions to improve pain management in nursing homes, and multimedia interactive computer programs aimed at improving diabetes and obesity-related knowledge, self-efficacy, and compliance among African American and Hispanic patients (Agency for Healthcare Research and Quality 2001d).

A second component of the Agency's implementation activities is aimed at state and local policymakers through the User Liaison Program. This program presents workshops, teleconferences, and other learning opportunities on priority issues in a format that is easily understandable and highly usable. Over the past few years, workshops directly related to inequities have included Strategies to Reduce Health Disparities, Addressing Critical Concerns of Health Care Systems Serving American Indians/Alaska Natives, and Providing Care to Diverse Populations: State Strategies for Promoting Cultural Competency in Health Systems (Agency for Healthcare Research and Quality 2001e).

Several efforts are aimed at reaching minority patients directly. The Agency has prepared some of its patient education material in Spanish (Agency for Healthcare Research and Quality 2002c). For example, tips are available to assist patients in finding high-quality providers, health plans, and hospitals as well as in learning how to avoid medical errors. A public service announcement for radio is also available (Agency for Healthcare Research and Quality 2001f). Another guide assists patients interested in finding reliable information about their disease and the tradeoffs of various treatment options (Agency for Healthcare Research and Quality 2000b).

Future Directions

The breadth of the current portfolio promises to inform much-needed solutions to the challenges presented in the IOM report, Unequal Treatment. However, the extent and persistence of disparities in health care delivery are a sobering reminder that many important questions remain to be addressed. While we know that socioeconomic position, race, and ethnicity exert independent effects on health care, we have much to learn about the pathways through which these factors operate. Effective collaborations between researchers who examine social determinants of health and those who focus on disparities in health care will help clarify the proportion of disparities in health that can be mediated through improved health care. Efficient strategies for routinely collecting requisite data need development, and our ability to interpret the “effect size” of efforts to reduce disparities needs enhancement. Finally, as with other areas of health care research, identifying incentives for sustainable improvements in care is necessary to effect substantial change.

The AHRQ has recently developed a new policy on the inclusion of priority populations in research that expands upon the previously established policies for inclusion of women and minorities. We will evaluate the impact of this policy periodically, and continue to assess the impact of prior research investments on improving health care for minority populations. The NHDR will provide important benchmarks for the success of prior and current efforts to address this critical challenge and stimulate much-needed discussion and action by health care organizations. We look forward to working with the research community and health care leaders to assure that our success in defining problems in health care delivery will be exceeded by our capacity to apply analytical methods to address those problems. We welcome your participation and feedback.