Disparities in health outcomes and health care among different racial and ethnic groups have been well documented in the literature. Infant mortality rates, for example, are higher among African Americans than they are among white Americans (14.1 per 1,000 births for African Americans versus 5.7 per 1,000 for whites; see National Center for Health Statistics, 2002) and the rate of death caused by cardiovascular disease was 146.9 per 100,000 for whites and 230.5 per 100,000 for blacks (U.S. Department of Health and Human Services [U.S. DHHS], 1990). There are substantial differences across racial groups in the percentage of women over the age of 40 who received a mammogram within the past 2 years: according to data from the 2000 National Health Interview Survey, among women at least 40 years old, 71.4 percent of white women received a mammogram compared with 67.8 percent of African American women, 47.3 percent of American Indian or Alaska Native women, and 53.3 percent of Asian women (National Center for Health Statistics, 2003). Disparities among racial and ethnic groups also show up in many chronic and acute disease conditions (e.g., diabetes, cardiovascular disease, obesity, hypertension, and asthma), as well as in the risk factors associated with them. Disparities exist within broad racial and ethnic categories as well. For example, among Asian Americans, 34 percent of Koreans have no health insurance, compared with 27 percent of Southeast Asians (Vietnamese, Cambodians, Laotians), 22 percent of South Asians (Indians, Pakistanis, and Bangladeshis), 20 percent of Filipinos, 19 percent of Chinese, and 13 percent of Japanese (Henry J. Kaiser Family Foundation, 2000).
The reasons for these disparities are not fully understood. Differences in access to health care (that is, health insurance coverage and/or the ability to pay for health care) might explain part of these differences. Among persons under 65 years of age in 2001, 19.3 percent of African Americans and 34.8 percent of Hispanics did not have health insurance coverage, whereas 14.7 percent of whites and 17.1 percent of Asians did not (National Center for Health Statistics, 2003). Some racial and ethnic groups tend to have lower income and wealth levels than others and thus may be less able to afford health care. Significant and persistent differences in wealth across racial groups have also been documented (Barsky et al., 2002; Oliver and Shapiro, 1995).
But differences in access are only part of the story. Differences in health care treatment even among the insured and beyond differences in access to health care also contribute to disparities. For example, among Medicare beneficiaries aged 65 and older, white women were more likely to get mammograms and to receive angioplasties than black women (Gornick et al., 1996).
Disparities in health and health care illuminate weaknesses in the health care and public health systems. Interest in better understanding the causes of these differences and formulating strategies to ensure the highest quality of care for everyone has generated significant attention to disparities across racial and ethnic groups. In addition to studies being conducted across many disciplines in a variety of academic, clinical, governmental, and other settings, Congress has initiated several important projects to provide better information on health disparities. In 1999, Congress asked the Institute of Medicine (IOM) to assess the extent of racial and ethnic differences in health care beyond those that are attributable to access to care, to evaluate potential sources of racial and ethnic disparities in health care, and to recommend interventions to eliminate the biases. This study resulted in the 2003 IOM report Unequal Treatment: Confronting Racial and Ethnic Disparities in Healthcare (IOM, 2003a). A key finding in this report was that racial and ethnic disparities exist beyond what can be attributed to differences in access to care. The panel also found that these disparities contribute to worse outcomes in many cases (IOM, 2003a). This study also examined the factors that contribute to disparities and offered guidance on interventions to reduce and eliminate disparities.
In 1999, Congress required the Agency for Healthcare Research and Quality to produce an annual National Healthcare Disparities Report (NHDR) that would monitor disparities in health care by race, ethnicity, socioeconomic status, and geography. Congress also asked the IOM to provide guidance on the development of the NHDR, a study that culminated in the 2002 production of Guidance for the National Healthcare Disparities Report (IOM, 2002). The first of the NHDR annual reports was released in December 2003 (U.S.
DHHS, 2003a).1 The purpose of this series of reports is to “track prevailing disparities in health care delivery as they relate to racial factors and socioeconomic factors in priority populations” (P.L. 106-129).
DHHS has implemented initiatives aimed at better understanding and addressing disparities. Eliminating health disparities is one of two primary goals of the Healthy People 2010 program. Some agencies within DHHS have implemented initiatives of their own to correspond to Healthy People 2010. The department has also initiated an educational campaign called Closing the Health Gap. A key element of this campaign is Take a Loved One to the Doctor Day, which is an effort to “encourage individuals to take charge of their health by visiting a health professional (a doctor, a nurse, a nurse practitioner, a physician assistant, or another health provider), making an appointment for a visit, attending a health event in the community, or helping a friend, neighbor, or family member do the same” (http://www.healthgap.omhrc.gov/index.htm).
DATA TO SUPPORT HEALTH DISPARITIES INTERVENTIONS AND RESEARCH
The availability of high-quality data on race, ethnicity, and other characteristics of individuals receiving health care is critical to documenting disparities in health and health care. But there are many weaknesses in the data sources currently available.
National-level surveys sponsored or conducted by the federal government are rich in information on health and health care outcomes as well as other characteristics of individuals; but while most have large enough sample sizes to obtain reliable information about broad racial and ethnic groups (i.e., blacks and whites), sample sizes are often not large enough for analyses of smaller racial groups (e.g., American Indian and Alaska Native) or for analyses within some of the racial and ethnic groups (e.g., to analyze differences between individuals of Mexican descent and individuals from other Hispanic backgrounds).
Data from Medicare claims and enrollment files have been widely used for analysis of racial and ethnic disparities, but such data are not available for all enrollees or potential enrollees, nor do they include information to permit analysis for more refined categorizations of race and ethnicity.
We note that two versions of this report were released; the executive summaries of these versions were the only differences in the two releases. The first version was released in December 2003; a second version was released in February 2004. The second version contained what was the original executive summary of the report, which had been changed in the process of departmental review (see Pear, 2004). The key findings of these two releases of the report are different (see http://www.house.gov/reform/min/politicsandscience/example_disparities.htm for both versions of the executive summary).
While state-based sources of data—vital records, administrative data from Medicaid, the State Children’s Health Insurance Program (SCHIP), and disease registry systems, for example—are potentially valuable for the analysis of health disparities, they often do not contain standardized data on race and ethnicity and contain very little information on other characteristics of individuals that would be used in an analysis. As a result, data from these sources often need to be matched with data from other sources.
Finally, data on health and health care are obtained from private records (those of health insurance plans, hospitals, and medical groups). But the racial and ethnic data on hospital records are unstandardized and include little information on the economic and social standing of patients. Health plans collect very little data on race and ethnicity, although there is some movement among plans to begin to collect such data.
Concerns about the data infrastructure for analyzing, understanding, and eliminating racial and ethnic disparities motivated Congress, in the Minority Health and Health Disparities Research and Education Act of 2000, to ask the National Academies to conduct a comprehensive study of the adequacy of Department of Health and Human Services (DHHS) data collection systems for measuring racial, ethnic, and socioeconomic disparities in health. In response to this request, the Office of the Assistant Secretary for Planning and Evaluation (ASPE) asked the Committee on National Statistics (CNSTAT) of the National Academies to convene a panel of experts to review DHHS data systems. The panel was charged to review the collection of data on race and ethnicity in data collection or reporting systems of the programs or activities of DHHS. These include other federal data collection systems (such as that of the Social Security Administration) with which the department interacts to collect data on race and ethnicity, as well as such systems in states and in the private health care sector (U.S. DHHS, 2003b). The panel was asked to:
identify the data needed both to support efforts to evaluate the effects of race, ethnicity, and socioeconomic position on access to health care and on disparities in health and to enforce existing protections for equal access to heath care;
assess the effectiveness of the data systems and collection practices of DHHS and the effectiveness of selected systems and practices of other federal, state, and local agencies and the private sector in collecting and analyzing such data; and
identify critical gaps in the data and suggest ways in which they could be filled, including the possible establishment of new systems.
ASPE and CNSTAT developed the charge for the study based on this legislation and on the department’s own needs for review of its data systems, giving the panel the flexibility to review related data needs as they arose.
We note some specific distinctions the panel made in interpreting its charge. First, the panel reviewed a very broad set of data collection systems both within and outside DHHS. These systems include health surveys, administrative records from programs operated by DHHS and the states (e.g., Medicare, Medicaid, and SCHIP), and records from private data systems such as health insurance records, hospital discharge abstracts, and physician and medical group records. The research purposes and uses of these data collection systems are quite varied—some are used to understand broad determinants of health (e.g., the effect of income on mortality) while others are used to understand very specific outcomes of health care treatment (e.g., the effects of ethnicity and race on medical outcomes of patients with hypertension or diabetes). The panel focused on the collection of data on race, ethnicity, and socioeconomic position, as called for in the originating legislation and charge to the panel. The panel also considered the collection of data on acculturation and language use because it believed these to be important covariates in understanding racial, ethnic, and socioeconomic aspects of health and health care. To comprehensively study health and health care, much more data are needed on individuals (e.g., genetic, behavioral, environmental, and cultural), and their health care treatments (e.g., what treatment they received, the cultural competency of the health care professionals administering the treatment, etc.). Collection of these data is affected by issues of race, ethnicity, and socioeconomic position because concepts of health and health care may differ or may be communicated in different ways in different cultural and socioeconomic groups in the United States. For example, the Western medical concepts of mental and physical health may not be consonant with all survey respondents, or specific concepts such as “family planning” or “gestational age” may be interpreted quite differently by Hispanic and Anglo respondents. A review of these issues is, however, beyond the scope of this panel, whose charge was to review the collection of basic racial, ethnic, and socioeconomic characteristics rather than to explore fully their implications for health research.
PREVIOUS REVIEWS OF DATA NEEDS
The panel’s work comes on the heels of previous DHHS efforts to examine collection of racial and ethnic data.2 In 1999, the department
implemented the DHHS Policy for Improving Race and Ethnicity Data, or the “Inclusion Policy,” which requires the inclusion of information on race and ethnicity in all DHHS-funded and -sponsored data collection systems. Also in 1999, the DHHS Data Council and the Data Work Group of the DHHS Initiative to Eliminate Racial and Ethnic Disparities in Health produced a report called Improving the Collection and Use of Racial and Ethnic Data in Health and Human Services (U.S. DHHS, 1999). This report contained recommendations for DHHS to improve the collection, analysis and interpretation, dissemination and use, research and maintenance of racial and ethnic data. Previous reports produced by DHHS or their contractors have also addressed the topic of data collection for measuring racial and ethnic disparities (Waksberg, Levine, and Marker, 2000; U.S. Public Health Service, 1992, 1993; U.S. DHHS, 1985). The National Committee on Vital and Health Statistics, an advisory committee to DHHS, has devoted serious attention to the collection of data on race and ethnicity within DHHS by holding hearings around the nation, writing reports, and making recommendations that encourage the collection of such data.
The panel considered these DHHS reports during its information-gathering phase and also invited experts on various data-related topics—including the Health Insurance Portability and Accountability Act (HIPAA) and the National Disparities Report Card—as well as members of the National Committee on Vital and Health Statistics Subgroup on Populations to brief the panel in open meetings. The panel also sponsored a 2-day Workshop on Improving Racial and Ethnic Data in Health, which focused on state-based and private sector-based data collections and resulted in a summary report (NRC, 2003). The agenda for this workshop is included in Appendix B. The panel commissioned a paper for this report to review the collection of socioeconomic position measures in health and health care databases. This paper, Recommendations on the Use of Socioeconomic Position Indicators to Better Understand Racial Inequalities in Health (O'Campo and Burke), is included in Appendix C. Four papers also were commissioned for the workshop: The Role of Racial and Ethnic Data Collection in Eliminating Disparities in Health Care (Fremont and Lurie); State Collection of Racial and Ethnic Data (Geppert, Singer, Buechner, Ranbom, Suarez, and Xu); Collection of Data on Race and Ethnicity by Private-Sector Organizations: Hospitals, Health Plans, and Medical Groups (Nerenz and Currier); and Racial and Ethnic Data Collection by Health Plans (Bocchino). These papers are included in Appendixes D-G of this report.
The rest of this report contains the panel’s assessment of the strengths and weaknesses of racial and ethnic data collection efforts in the health and
health care fields. The panel believes very strongly that social and economic factors as well as language and acculturation factors play an important role in understanding racial and ethnic disparities in health and health care, and therefore considers the collection of measures of these concepts just as important as measures of race and ethnicity. Chapter 2 sets the stage by briefly reviewing some of the literature on such disparities, citing examples of disparities, and making the case for the importance of collecting data to monitor and understand them and to develop interventions to eliminate them. Chapter 3 reviews measures of race and ethnicity and recent regulations governing the collection of these data. The chapter also briefly reviews measures of socioeconomic position, language, and acculturation that are used to understand the relationship between health and these constructs.
The remaining chapters of the report review current practices in the collection of data on race, ethnicity, socioeconomic position (SEP), and language and acculturation in the health care field, highlight weaknesses, and recommend improvements. The discussion is organized by the source of the data—federal, state, and private-sector sources. There is occasional overlap in this partition, as some data generated by private-sector entities are collected and organized by states, sometimes with federal funding support. Recommendations for improved data collection are given in each of the last three chapters. In making these recommendations, the panel did not consider specific assessments of the cost of improved data collection but did broadly consider the costs of data collection among different types of data collection systems.
The primary focus of the panel was on data collection systems within DHHS, over which the department has the most control. Chapter 4 reviews data collected by DHHS. The panel also reviewed data collected by states and by private sector entities. Both state health departments and private sector entities have roles to play in the collection of data for research on disparities. States are responsible for the collection and maintenance of a number of data systems in health and health care—many of which are then forwarded to DHHS for use on a national level (e.g., vital records and hospital discharge abstracts). Chapter 5 reviews data collected by state governments. Private entities also collect data relevant for researching disparities. These records of diagnoses, treatments, and insurance enrollments and claims are especially important sources of information on health care. Some of these records are the basis of federal and state-based data collection systems. Although outside of DHHS and state health agencies, these privately based records are an important component of data systems needed for the federal government and for state governments to monitor the health of their respective populations. Chapter 6 reviews data collected by private sector organizations.