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.
10 DESIGNrNG A NATIONAL HEALTHCARE DISPARITIES REPORT COMMITTEE REPORT It is well established that race, ethnicity, socioeconomic status, and geographic location are among the factors that influence health care independent of patient need (IOM, 2002b; National Center for Health Statistics, 2001; National Quality Forum, 2002; Nerenz et al., 2002~. Growing concern over racial, ethnic, geographic, and other disparities in health care prompted Congress in 1999 to require the Agency for Healthcare Research and Quality (AH:RQ) to produce a new annual report beginning in fiscal year 2003 (October I, 2002 to September 30, 2003~. The National Healthcare Disparities Report (NHDR) will take its place alongside another new annual report to Congress to be called the National Healthcare Quality Report (NHQR). Together, they will provide policy makers, consumers, and others with a more complete picture of the health care that Americans receive and of the areas that need attention. To help it address a number of technical issues related to the NHDR, AHRQ commissioned the Institute of Medicine (IOM) to examine issues related to racial, ethnic, geographic, and socioeconomic access to--and use of--health care services, as well as to the quality of care provided. In addition, the TOM was asked to take into account explanatory factors such as spoken language, literacy, culture, community influences, and attitudes toward health. Also, the ~ As a federal agency AHRQ must use the racial categories specified by the federal Office of Management and Budget (OMB) in OMB Directive 15: American Indian or Alaska Native; Asian; Black or African American; Native Hawaiian or other Pacific Islander; and White. OMB Directive 15 defines ethnicity separately from race, and it is limited to Hispanic or Latino or Not Hispanic or Latino. Currently, there is not a standardized treatment of racial and ethnic subpopulations. See National Quality Forum (2002) for a discussion of the lack of subpopulation definitions. 10
1: COMMITTEE REPORT 11 IOM was asked to examine measures and data sources that could be used in the report.2 AHRQ requested that the TOM consider issues related to the NHDR within the context of the framework developed by the IOM's Committee on the National Quality Report on Health Care Delivery in its study, Envisioning the National Health Care Quality Report (IOM, 2001 c). The framework consists of a matrix of components of health care quality and consumer perspectives on health care needs. The four components of health care quality are based on those presented in Crossing the Quality Chasm (TOM, 2001b): safety, effectiveness, patient centeredness, and timeliness. There are four consumer perspectives on health care needs: staying healthy, getting better, living with illness or disability, and coping with the end of life.3 In the framework, equity is a component that applies to both populations and individuals. It is defined in terms of"providing care that does not vary in quality because of personal characteristics such as gender, ethnicity, geographic location, and socioeconomic status" (TOM, 200 Ib, p. 6~. For populations, equity means reducing disparities in the use of health care services that are related to personal characteristics such as race, ethnicity, socioeconomic background, and geographic location. Research documents that insurance coverage is particularly important to achieving this goal (IOM, 2001a; TOM, 2001b). For individuals, it refers to the receipt of safe and effective services based on need. As Figure 1-1 shows, the framework treats equity as a cross-cutting factor, applicable to each cell of the matrix. AHRQ will use the same framework for the NHDR. This reflects the agency's plan to make health care quality a major focus of this report, which is appropriate since disparities often represent an 2 At the same time that the IOM Committee for Guidance in Designing a National Health Care Disparities Report was meeting, the National Quality Forum (NQF) was producing a report (National Quality Forum, 2002) that addressed the issue of quality measures best suited to capturing health care disparities. To avoid duplicating the work of the NQF, the committee focused on matters relating to service utilization and access. 3 See chapter 2 of IOM (2001 c) for definitions of these terms and an elaboration on the framework as a whole.
12 GUIDANCE FOR THE NATIONAL HEALTH CARE DISPARITIES REPORT "inequality in quality" (FiscelIa et al., 2000, p. 2579~. Within the framework, the NHDR will highlight health care issues related to equity and the extent to which health care disparities undermine its achievement. FIGURE 1-1 Framework for the National Healthcare Quality Report and the National Healthcare Disparities Report Components of Health Care Quality Consumer ~ ~ _ Perspectives on => Heals Care Needs ~ ~ ~ ~ ~ Staying Healthy ~ _ Getting Better Living with Illness or _ Disability _ Coping with the End _ cow cow an .= - ·E~ 1 . ~ 3 a To carry out this work, the IOM established the Committee for Guidance in Designing a National Health Care Disparities Report. The committee met twice. At its initial meeting in January 2002, it planned its work and discussed its charge. it also was briefed on disparities- related issues by experts from AHRQ, the U.S. Bureau of the Census, the National Center for Health Statistics (NCHS), the NQF, and the TOM. The committee commissioned five consultants to address key research topics. At the March meeting, they delivered presentations on the following: Marian E. Gornick (independent consultant): measurement of socioeconomic status in disparities research; Thomas A. LaVeist (Johns Hopkins University): measurement of disparities in health care services and quality; Nicole Lurie (RAND): measurement of disparities in health care access; Thomas C. Ricketts, TII (University of North Carolina-Chape} Hill): measurement of geographic units in disparities research; and Ross Arnett (independent consultant): subnational datasets for use in the NHDR. At the March meeting, the committee also heard testimony from a number of other invited
1: COMMITTEE REPORT 13 experts on health care disparities. These experts participated in one of the following panels: Disparities in Public Health; Disparities in Health Care Purchasing and Providing; Disparities in Health Care Delivery; or General Comments on Disparities.4 1-1. MEASUREMENT OF SOCIOECONOMIC STATUS IN DISPARITIES RESEARCH Socioeconomic status is a complex concept that combines dimensions of social and economic resources as well as societal ranking or prestige. As such, it is related to social stratification, or "a system of social relationships that determines who gets what, why" (Kerbo, 1996, p. 11~; social class, or "social groups arising from interdependent economic relationships among people" (Krieger et al., 1997, pp. 344-5~; and other concepts identified with social inequality. Socioeconomic status influences health care in a number of ways. For example, an individuaT's or family's material circumstances affect health care access, services, and quality since they are directly related to adequate insurance coverage (TOM, 200 1a; Lurie, 2002~. Social status affects health care by influencing the ways in which individuals are perceived. For example, health care professionals are more likely to take seriously those who appear to have higher status (Magnus and Mick, 2000~. In addition, education, well-connected social networks, experience in dealing with professionals, poise, and other aspects of higher social position can help patients effectively navigate a complex health care environment that features health care insurers and individual and institutional providers, among others. The ability to navigate this system can in turn influence the access, services, and quality of care that patients receive (Gornick, 2002; TOM, 2002b; Magnus and Mick, 2000~. The NHDR should contain analyses of racial and ethnic health care disparities that reflect the influence of socioeconomic status. There are two main reasons for paying particular attention to socioeconomic status. First, it would help to clarify the extent to which health care disparities result from socioeconomic factors or from racial 4 See Appendix I for the Workshop Agenda and Appendix II for a summary of the public testimony.
1 4 GUIDANCE FOR THE NATIONAL HEALTH CARE DISPARITIES REPORT and ethnic factors. Socioeconomic status is associated with race and ethnicity: racial and ethnic minorities are more likely to have Tower socioeconomic status as measured in a variety of ways, including income, wealth, and education (National Research Council, 2001~. Better understanding the relative effects that socioeconomic factors and racial and ethnic factors have on disparities is critical to identifying ways to eliminate them. Secondly, socioeconomic status deserves attention in its own right because it has a pervasive influence on both health status and health care. It is linked to health status in a number of ways. For example, those with lower socioeconomic status are more likely to lead socially isolated lives and to be unemployed or to work at jobs that are unfulfi~ling and stressful (Marmot, 2002~. They more frequently live in places where it is difficult to buy fresh food and to exercise. Their environments are more apt to be polluted by such elements as hazardous wastes, unclean air, and lead paint (IOM, 1999~. Socioeconomic status is also linked to differences in health care. Those with Tower socioeconomic status are more commonly uninsured and have more limited access to preventive, primary, and specialized care. They are more likely to suffer adverse health outcomes and poorer health status (IOM, 200 1a; TOM, 2002a). 1. The National Healthcare Disparities Report should present analyses of racial and ethnic disparities in health care in ways that take into account the effects of socioeconomic status. For the NHDR to adequately take into account the effects of socioeconomic status, it should use socioeconomic status in two different ways: first, as a stratification variable in collecting sample data on racial and ethnic disparities in health care access, service utilization, and quality; and secondly, as an independent vanable that serves as a control in analysis. Stratification would ensure adequate sample sizes of racial and ethnic populations with varying levels of socioeconomic status. Further, controlling for socioeconomic status would help to identify the extent to which disparities result from factors associated with race and ethnicity and the extent to which they result from factors associated with socioeconomic status. Using socioeconomic status as both a stratif~er and control would be more likely to yield reliable findings of the role that socioeconomic status
1: COMMITTEE REPORT 15 plays in racial and ethnic health care disparities. Using socioeconomic status only as a control variable assumes that a single model fits all subgroups in the population. This assumption, however, may not be valid. It may be, for example, that there is a significant positive relationship between socioeconomic status and a particular dependent variable among Blacks, but no significant relationship among Whites. There might even be a significant negative relationship between the variables among Asian Americans or American Indians. The NHDR, therefore, should include analyses that stratify by race and ethnicity to explicitly test whether the relationships among variables are the same when each subgroup is considered in turn. To understand the independent impact of socioeconomic status on disparities, it is important to identify which of its dimensions have greater effects on health care, facilitating the development of more targeted and effective interventions. For example, are material resources more strongly related to receipt of quality care than educational level or occupation? If material resources are critical, then efforts can be focused on formulating and implementing policies and interventions to enhance economic well-being or to reduce the cost of medical care. If occupation is strongly associated with receipt of quality care, interventions can be focused on those in affected jobs. It is also important to determine the extent to which disparities are rooted in racial, ethnic, or socioeconomic issues. For example, is poor access to specialized care more strongly related to race or to income? Clarifying the impact of socioeconomic status on disparities would enhance the accuracy of the NHDR as well as add to the mix of issues that interventions need to address. The complexity of socioeconomic status is more exactly expressed in the many different ways that it has been operationalized (Liberatos et al., 1988~. There does not appear to be a single right way to operationaTize it. Different measures either taken by themselves or together capture important aspects of socioeconomic status and how it affects health care (Gornick, 2002; Liberatos et al., 1988~. There are several major approaches to measuring socioeconomic status. They include income, wealth, education, poverty level, occupation, and deprivation indices, which are composite measures consisting of such variables as employment status and access to a car. Some of the major
1 6 GUIDANCE FOR THE NATIONAL HEALTH CARE DISPARITIES REPORT approaches, namely income and education, are commonly used in the research literature. Wealth, including property holdings and income from investments, is less often used. Occupation is more frequently applied in British studies of the relationship between socioeconomic status and health status. Deprivation indices have not been extensively used in studies of health care. Each approach has advantages and disadvantages (Gornick, 2002~. At present, education and income are the measures of socioeconomic status for which enough data are available for use in the NH:DR. Education is a stable measure for adults, with nearly complete reporting in surveys. Income is included in most publicly reported datasets. Each has significant advantages, which have been noted by Gornick (Gornick, 2002) and others (Krieger et al., 19971. Survey respondents readily report information on their educational backgrounds, and education is commonly regarded as a meaningful and valid measure of socioeconomic status. Data on income are relatively accessible to researchers. It should also be noted that each has important disadvantages. Education may have different social meanings across generations and races and ethnicities. For example, a high school degree for the postwar generation was associated with more economic opportunity than a high school degree for younger generations. In addition, minorities often attend schools with fewer resources and less prestige, which can make their educational achievements less valued. Similarly, an immigrant with a college degree from another country may not receive the same economic returns as a person who graduated from an American college. Income data are missing for a significant proportion of people in most health care surveys. Survey respondents also tend to underreport income. In addition, income more accurately captures the financial resources available to minorities than to Whites, who are more likely to own real estate and have other investments and assets (Oliver and Shapiro, 2001; Smith, 2001~. Is education or income the better measure of socioeconomic status? Education and income are related variables: higher educational levels are associated with higher income levels. But this does not mean that they are interchangeable. For example, Gornick (Gornick,
1: COMMITTEE REPORT 17 2002) examined the proportion of White and Black women over 65 who received mammograms, flu shots, and Pap smears in 1998 by two levels of income (below and above $25,000) and two levels of education (less than high school and high school or more). The data reveal disparities by race whether socioeconomic status is measured in terms of income or education. Nonetheless, stratification by income and education yielded appreciably different numbers for analysis. Almost nine times the number of Black women (985,000) had a lower level of education than the number (111,000) who had a lower level of income. More than two times the number of White women had a Tower level of income (11,434,000) than education (5,472,000~. As this also suggests, analyses of the relationship between socioeconomic status and health care can vary depending on the quality indicator used and the population or subpopulation examined.s Which measure or measures of socioeconomic status should AHRQ use in the NH:DR? This is an important question. To adequately answer it will require a clearer understanding of the relationship between socioeconomic status and health care. There is a striking dearth of studies of the relationship of socioeconomic status to health care. Income does appear to be a critical variable, and it is related to insurance coverage. Those on the higher end often receive private health insurance as a work-related benefit. While some on the Tower end of the income distribution may qualify for Medicaid depending on federal and state eligibility requirements, they are most likely to lack it. Many of those in the middle also lack health insurance (IOM, 2001a). However, research does not show whether income mainly accounts for the relationship of socioeconomic status to health care. Nor does it indicate whether other variables with which income is associated actually account for more of the relationship. For example, research reveals that those at different income levels tend to use the health care system in different ways. Those with higher incomes have a greater tendency to use preventive services while those with lower incomes have a greater tendency to use acute care sentences due in part s It should also be noted that analyses can vary depending on the cutpoints used for education, income, and other measures of socioeconomic status (Liberatos et al., 19 8 8~.
1 8 GUIDANCE FOR THE NATIONAL HEALTH CARE DISPARITIES REPORT to greater morbidity. This pattern holds true even when cost is not a factor, as is the case with the use of influenza vaccinations by Medicare beneficiaries, a benefit that is completely covered by Medicare (Gornick, 2000~. Is it income per se that accounts for these different patterns? Or is it factors with which income is associated, such as access to transportation and proximity to health care providers? 2. AHRQ should pursue a research initiative to more accurately and meaningfully measure socioeconomic status as it relates to health care access, service utilization, and quality. AHRQ should sponsor research on the relationship of socioeconomic status to health care. Exploration of the relationship between socioeconomic status, health care, and other factors such as race and ethnicity would help to identify the most appropriate measures of socioeconomic status to use in studies on health care disparities. Appropriate measures in turn would help to indicate which of the many dimensions of socioeconomic status are most likely to affect a particular aspect of health care and therefore to be associated with outcomes. The following are examples of important areas where more research is needed: · In general, which dimensions of socioeconomic status most affect health care and why? How strongly are different measures of socioeconomic status such as income, education, and occupational prestige associated with health care? · Should socioeconomic status be studied at the individual, household, or community level? Research indicates that different races, ethnicities, and nativities call for different levels of analysis. For example, community-level measures may better capture the social and economic status and environments of immigrants and some races and ethnicities while individual- and household-level measures may be more appropriate for others (Krieger et al., 1997~.
1: COMMITTEE REPORT 19 The status dimension of socioeconomic status is particularly understudied, both in terms of how it does and does not empower patients and how it affects the relationship that patients have to providers and other aspects of the health care system. How might status be operationalized? Is income sufficient? Or are other measures of status, such as car and home ownership, needed? · Is one measure of socioeconomic status adequate? Or do composite measures, such as deprivation indices, perform better? Are different dimensions of socioeconomic status implicated in preventive, acute, or other aspects of health care? For example, is it the case that attitudes toward health, health literacy, cost and availability of transportation, or work schedule flexibility influence a patient's use of preventive services, while income more strongly accounts for use of acute care? · The NHDR will be focused on issues of access to and within the health care system, health care service utilization, and health care quality. How are patterns in each related to a patient's socioeconomic status? Are the results of quality measurement affected by the socioeconomic status of the population more likely to use particular services? 1-2. MEASUREMENT OF DISPARITIES IN ACCESS TO AND WITHIN THE HEALTH CARE SYSTEM Access can be defined as entry to the system of care as well as entry within the system of care. Access is a central aspect of quality, and the NHDR should give it prominent attention. The framework of the NHQR treats access as an important aspect of all four components of health care quality: safety, effectiveness, patient centeredness, and timeliness. However, access deserves greater prominence in its own right because it is a critical starting point for quality care. As such, it is a fundamental aspect of quality, especially for racial and ethnic minorities, those with fewer socioeconomic resources, and those in
20 GUIDANCE FOR THE NATIONAL HEALTH CARE DISPARITIES REPORT disadvantaged geographic areas (TOM, 200 1a; TOM, 2002a; TOM, 2002b; National Center for Health Statistics, 2001~. 3. Access is a central aspect of health care quality. As such, the National Healthcare Disparities Report should give it prominent attention. Access should be reconceptualized as a dimension of health care rather than as a dimension of medical care alone. Access to physicians and other medical care professionals is essential: it provides the diagnoses, medical interventions, and monitoring that can be critical to preventing and treating illness. However, primary care physicians typically coordinate care received from other health care specialists including nutritionists, dentists, and occupational, physical, and mental health therapists. Disparities in access depend in part on the social and human capital of the patient and the community. Some factors play important roles, including adequate and secure material resources; community norms that favor healthy lifestyles; social support networks supplied by families, mends, and religious, professional, social, and civic organizations; the availability of safe and convenient places to exercise and shop for fresh food; well-developed transportation systems; high literacy rates; and Tow crime rates (Aday, 2001; Fiscelia, 2002; TOM, 2002b; Lurie, 2002; Ricketts, 2002~. As such, a wide range of measures of access influence entry to the system of care as well as entry within the system of care. They include factors as diverse as the extent of insurance coverage, language access services, and other aspects of culturally competent care. For example, cultural competency is critical in the diagnosis and treatment of mental illness: behavior that is interpreted as mental illness in one culture may be an appropriate way of displaying emotion in another culture (DHHS, 1999~. The following are some of the new measures of access suggested by Lurie for development and use in the NHDR: Adequacy of insurance coverage. Based on work by Bashshur et al. (Bashshur et al., 1993), Lurie defines underinsurance as "a
1: COMMITTEE REPORT 21 situation in which the consequences of having less than full coverage are so burdensome that they inhibit realized access to needed care likely to result in the best possible outcome" (Lurie, 2002, p.123~. Lurie recommends presenting data on benefits, co- payments, and deductibles for different racial and ethnic groups by income. She maintains that assessing the adequacy of insurance coverage in this way would promote a more accurate analysis of the adequacy of insurance coverage than relying on a uniform standard of uninsurance that did not take into account the ability to afford out-of-pocket expenses. . Language access. Millions of Americans have difficulty communicating in English. It is important that patients be able to clearly communicate their medical and social histories, symptoms, and concerns to their providers, and that providers be able to clearly communicate their questions, diagnoses, and recommendations for treatment (Lavizzo-Mourey and Mackenzie, 1996; Morales et al., 1999; Perez-Stable et al., 1995~. New measures should be developed that address whether language access is available through either hi- or multi-lingual providers or trained interpreters and whether it is of adequate quality (California Pan-Ethnic Health Network (CPEHN), 2001~. Frequently, language interpretation including American Sign Language is unavailable, leaving the task to volunteers, Fiends, and family. This can compromise privacy and free exchange of information (IOM, 2002b; Woloshin et al., 19951. . . Regular site of care and regular provider. Access to a regular source of care whether to a particular site such as a local clinic or a particular provider such as a primary care physician makes it more likely that individuals will receive a greater number of appropriate health care services. This in turn makes it more likely that individuals will experience better health outcomes (IOM, 2002a; Lurie, 2002~. Access to mental health care. Racial and ethnic minorities are especially likely to have limited access to mental health care (DHHS, 2001~. The dimensions of limited access that measures should capture include service availability and insurance coverage.
22 GUIDANCE FOR THE NATIONAL HEALTH CARE DISPARITIES REPORT Mental health therapy is especially dependent on communication and understanding between patients and providers (DHHS, 2001; Takeuchi et al., 1995~. Therefore, measures of access should also indicate obstacles posed by racial, language, and cultural differences (CPEHN, 2001~. 1-3. MEASUREMENT OF DISPARITIES IN HEALTH CARE SERVICES AND QUALITY Health care disparities, including those in service utilization and quality, can be measured in absolute or relative terms. Measuring disparities in absolute terms means comparing the care received to evidence-based standards of quality. Measuring disparities in relative terms usually involves comparing the care that racial and ethnic minorities receive to the care that majorities receive. However, this is problematic if the care that majorities receive is not consistent with evidence-based standards of quality. For example, greater receipt of a particular kind of procedure may reflect health care overuse, which results in greater exposure to medical errors and other risks of needless treatment. AHRQ should explore the possibility of using an absolute standard of evidence-based quality care. This would imply that the NHDR would report on the proportions of those of different races and ethnicities that receive evidence-based quality care. At the same time, many areas of health care lack absolute standards of evidence-based quality care, which will limit the ability of the NHDR to rely on them. While disparities usually describe care in which minorities receives fewer services or procedures, it is important for the NHDR to educate policy makers, consumers, and other audiences of the report about another kind of health care disparity in which minorities receive more services or procedures. However, this kind of disparity does not indicate that minorities are in these cases more likely to receive better health care. Instead, it reveals that minorities often received poorer prior care, necessitating more extreme and less desirable interventions. 4. The National Healthcare Disparities Report should include measures of high utilization of certain health care services that indicate poor
1: COMMITTEE REPORT health care quality. It should also include measures of low utilization of certain health care services, which are more commonly used to indicate poor health care quality. 23 Research has revealed a number of health care services in which high minority utilization holds true (IOM, 2002b). For example, compared to Whites, African Americans with asthma are more likely to be treated in emergency departments and are more likely to be hospitalized, indicating poorer preventive and primary care (Zoratti et al., 1998~. Compared to Whites, African American diabetics are more likely to receive amputations, again indicating poor prior care or lack of access to primary care services (Andersen et al., 1983~. In mental health care, minorities in some cases receive more antipsychotic medications administered in emergency departments (Segal et al., 1996~. This kind of disparity stands in contrast to the more familiar kind that should also be featured in the NHDR. These occur when minorities receive fewer health care procedures than other races and ethnicities (TOM, 2002b; Mayberry et al., 2000~. For example, it is the case that Whites were more likely than African Americans, Hispanics, and Asian Americans to receive coronary angiography, angioplasty, and coronary artery bypass graft (CABG) surgery. In addition, compared to African Americans, Whites were more apt to receive an effective range of diagnostic procedures for colon cancer as well as post-treatment surveillance care. Whites were also more likely to receive adequate pain medication for cancer (TOM, 2002b). 1-4. MEASUREMENT OF GEOGRAPHIC DISPARITIES IN HEALTH CARE The NHDR should present national-level information on health care disparities. However, geographic data on health care is also available on subnational units that include states, counties, regions, ZIP code areas, primary care service areas, hospital service areas, rural and metropolitan areas, and census units such as tracts and block groups (Ricketts, 2002~. While the number of geographic units is
24 GUIDANCE FOR THE NATIONAL HEALTH CARE DISPARITIES REPORT large, at least in the short term the NHDR should focus on states and rural and urban areas. However, in the future, policy trends, data availability, and other issues could suggest that the NHDR should also include analyses at smaller subnational levels. 5. The National Healthcare Disparities Report should present data on disparities at the state level. It should also present data on disparities along a rural-urban continuum. The NHDR should focus on states for two major reasons. First, states play important roles in health care and, by extension, must play important roles in any future efforts to eliminate health care disparities. States, for example, help to fund and primarily administer such programs as Medicaid and the State Children's Health Insurance Program (SCHIP). They are also charged with key public health activities such as data collection and reporting, prevention programs, and health care delivery. In addition, they carry out key regulatory and oversight functions that affect patients, providers, provider organizations, facilities, and insurers, among others (Ricketts, 2002~. Secondly, the NHDR should focus on states because they would interest the report's primary audiences, which consist of Congress, other policy makers, and consumers. Congress has indicated that it is a primary audience by mandating that AHRQ annually report to it on health care disparities, including geographic disparities. The TOM Committee on the National Quality Report on Health Care Delivery also identified other policy makers and consumers as the primary audiences for the NH:QR (IOM, 200Ic). It is likely that other policy makers and consumers will be primary audiences for the NHDR as well. Policy makers are increasingly interested in disparities issues and initiatives to eliminate them (Murray-Garcia, 2002; OMH [DHHS], 2002~. At a time in which the nation is becoming more diverse, growing minority populations have registered increasing concern about disparate treatment. The NHDR should also present information on rural and urban areas. Rural areas are strongly represented in Congress. In particular, in the Senate, rural states such as North Dakota and Wyoming have the same number of Senators as highly populous states such as California.
1: COMMITTEE REPORT 25 Urban areas also draw substantial interest in Congress, where they are more strongly represented in the population-based House. Both rural and urban health care disparities have long concerned Congress. It has supported a number of programs aimed at addressing issues such as physician supply and hospital and clinic support, and monitors the impact of programs such as Medicare and Medicaid on rural and urban health care delivery (Me6PAC, 2001~. While the NHDR should feature information on geographic disparities featuring states and rural and urban areas, it should be noted that states and rural and urban areas are heterogeneous units. State comparisons can produce misleading results. For example, a large state with a racially and ethnically diverse population such as California faces very different challenges in health care delivery than a smaller state such as Vermont, which has a more racially and ethnically homogenous population. Similarly, comparing all rural areas with all urban areas can mask considerable differences within both. For example, rural Wyoming and rural Pennsylvania significantly vary in how far their residents are from major medical centers while urban areas significantly vary in the degree of access their residents have to health care. It can also be misleading to draw a dichotomy between rural and urban areas. Instead, there is a continuum of geographical areas that is defined by gradations in population density, distance from population centers and services, and other characteristics. In addition to states and rural and urban areas, the NHDR could make use of such population-based measures as residential segregation, community impoverishment, and income inequality. Residential segregation has most often been measured using the dissimilarity index, which indicates the proportion of one group that would need to relocate from one community to the other to achieve an even distribution in both (Sakoda, 1981~. Residential racial segregation has been linked to greater risk for infectious disease (Acevedo-Garcia, 2000), greater hospital segregation (Smith, 1998), higher Black infant mortality (Polednak, 1996), and higher all-cause mortality for Blacks and Whites (Acevedo-Garcia, 2000; Cooper et al., 2001; Jackson et al., 2000~. Segregated Black communities have higher concentrations of toxic air (Lopez, 2002), fewer food stores (Moriand et al., 2002), more liquor stores (LaVeist and Wallace,
26 GUIDANCE FOR THE NATIONAL HEALTH CARE DISPARITIES REPORT 2000), and fewer medical providers (Gaston et al., 1998; Williams and Collins, 2001~. Similarly, community impoverishment has been shown to affect coronary artery disease and cause mortality independent of individual-level risk factors (Anderson et al., 1997; Diez Roux et al., 2001; Waitzman and Smith, ~ 998~. Income inequality, whether measured at the state or community level, has been linked to higher rates of violent crime (Kennedy et al., 19981; increased teen births (Gold et al., 2001~; obesity (Kahn et al., 19981; and poor health (Fiscella and Franks, 2000; Lochner et al., 2001~. Effects of income inequality may be improved to some extent by better access to primary care (Shi and Starfield, 2000; Shi and Starfield, 2001~. Many of these indicators have been used with periodic national surveys such as the National Health and Nutrition Examination Survey (NHANES), the National Health Interview Survey (NHTS), and the Behavioral Risk Factor Surveillance System (BRFSS) and so could lend themselves to monitoring over time in the NHDR. 1-5. SUBNATIONAL DATASETS A variety of datasets could help to supply the needs of the NHDR.6 These include surveys produced by single states that have sizable racial and ethnic populations and subpopulations. In addition are surveys of all or multiple states, localities, or other entities in multiple states sponsored by public or private sources. Still others are nationally representative surveys, databases, or data-oriented reports that could also be useful for the NH:DR. Table 1-1 contains the name of the datasets along with information on where to obtain further information on them. 6 The long-term data needs of the NHDR and other disparities-related programs and activities of the Department of Health and Human Services (DHHS) will be addressed in another study underway at the National Academy of Sciences. This study, which is being conducted by the National Research Council's Committee on DHHS Collection of Race and Ethnicity Data, will examine data requirements and limitations and ways to improve data sources. It expects to issue a final report in late 2003.
1: COMMITTEE REPORT TABLE 1-1 Possible Subnational Datasets to Support the National Healthcare Disparities Report (NHDR) 27 Dataset Responsible Website Address Organization Behavioral Centers for www.cdc.gov/brfss Risk Factor Disease Control Surveillance and Prevention System (CDC) (BRFSS) California UCLA Center for www.healthpolicy.ucla.edu/chis Health Health Policy Interview Research, CA Survey (CHIS) Dept. of Health Services Consumer Agency for www.ahrq.gov/qual/cahps Assessment of Healthcare Health Plans Research and (CAMPS) Quality (AHRQ) Commonwealth Commonwealth www. cmwf. org Fund Health Fund Care Quality Survey Community Health Resources www.communityhealth.hrsa.gov Health Status and Services Report Administration (HRSA) Community Center for www.hschange.org Tracking Studying Health Survey (CTS) System Change Hawaii Health Hawaii Dept. of www.hawaii.gov/health/stats/ Survey (HHS) Health surveys/hhs.htm Healthcare AHRQ www/ahrq. gov/data/hcup Cost and Utilization Project (HCUP) HMO Research AHRQ ht~://hmoresearchnetwork.org Network Medical Care Maryland Health www.mhcc.state.md.us Data Base Care Commission (MHCC)
28 GUIDANCE FOR THE NATIONAL HEALTH CARE DISPARITIES REPORT Dataset Responsible Website Address Organization- Medicare Centers for www.hcLa.gov/stats Administrative Medicare and Data Medicaid Services (CMS) National Urban Institute www.urban.org/Content/Research/ Survey of NewFederalism/NSAF/Overview/ America's NSAFOverview.htm Families (NSAF) Peer Review CMS www.hcfa.gov/stats Organization Data Proprietary MED STAT www. medstat. com Health Care Group Databases State Health Kaiser Family www.statehealthfacts.kff.org Facts Online Foundation (KFF) State Surveys State Health www.shadac.org on Health Access Data Insurance Assistance Center Coverage (SHADAC) There are advantages and disadvantages to the use of subnational datasets in the NHDR, on which AHRQ plans to rely at least in the short term (Burstin, 2002~. Their primary advantage lies in the samples of subpopulations that some contain and that currently are inadequately represented in national datasets. For example, the new California Health Interview Survey (CHIS) provides estimates on such aspects of care as insurance coverage, access, and barriers to health services for the major racial and ethnic populations in the state. It also provides estimates for smaller subpopulations, including Asian American and Pacific Islander ethnic groups, American Indians, and Alaska Natives. AHRQ should take into account the following methodological considerations in using subnational datasets:
1: COMMITTEE REPORT 29 · Sampling designs. While racial and ethnic minority populations have grown significantly over the past two decades, they remain a relatively small proportion of the total population in many communities. In studies of small populations, certain sampling decisions must be made to decrease cost and increase survey efficiency. However, these decisions can limit the extent to which conclusions can be drawn about the target population. For example, a study of Chinese Americans in Los Angeles County lowered screening costs by restricting its sampling frame to geographic areas (census tracts) where Chinese Americans comprised at least six percent of the population (Takeuchi et al., 1998~. The six percent criterion provided coverage of approximately sixty percent of the Chinese American population in Los Angeles. At the same time, it limited the extent to which conclusions can be drawn about all Chinese Americans living in Los Angeles since it excluded Chinese Americans in low density Chinese American geographic areas, who tend to be native born and more prosperous. . · Translations. Immigrants comprise a substantial proportion of Asian Americans and Latinos. Since many immigrants may not speak English, it is important to know whether respondents were offered translated versions of survey questionnaires. If a region has a sizable number of immigrants and translated versions of a questionnaire were not provided, this limits the extent to which results can be generalized to all members of a group. Response rates. Survey response rates have declined, especially in major urban areas. It is critical to know the response rate for subnational data sources and, if possible, for different racial and ethnic minority groups. Lower response rates can limit the generalizability of survey findings. AHRQ and other governmental entities have productively worked with private entities in the past to develop and disseminate health care measures. Public and private utilization of the Consumer Assessment of Health Plans (CAMPS) is a major example of this kind of partnership. AHRQ could work with public and private organizations that sponsor key subnational data sources to help make
30 GUIDANCE FOR THE NATIONAL HEALTH CARE DISPARITIES REPORT these data sources more useful for the NHDR. These data sources could shed light on how race, ethnicity, socioeconomic status, and geography affect health care. For example, very little is known about the health care of Native Hawaiians and Pacific Islanders. Since a major proportion of Native Hawaiians and Pacific Islanders lives in California and Hawaii, it may be advantageous for AHRQ to work with sponsors of subnational data sources in these states to address issues of race, ethnicity, and health care. Public and private organizations would also realize advantages from this partnership. It is likely that datasets compatible with the NHDR would also be more useful as separate data sources or as data sources used in conjunction with others. 6. In the future, if A~Q continues to rely on suhnational data sources for the National Healthcare Disparities Report, it should work with public and private organizations that sponsor key suhnational data sources to identify core elements in these surveys that can be standardized. More specifically, AHRQ and other organizations must address the fact that subnational datasets measure core elements such as race, ethnicity, and socioeconomic status in ways that differ from each other and differ from many national datasets. For example, race and ethnicity are measured in several ways. Federal policy for national surveys calls for respondents to have the option of selecting multiple races or, in the case of ethnicity, Hispanic or Latino or Not Hispanic or Latino. However, some subnational surveys require respondents to choose only one racial category. The question that is used in a survey will affect the size of the denominator and therefore estimates of the prevalence of what is being measured. It will also influence the extent to which generalizations can be made about racial and ethnic communities. In addition, subnational surveys and other data sources differ on whether racial and ethnic identities are reported by respondents or by observers.7 Standardized measurement of socioeconomic status is also lacking. For example, the way that respondents are asked to estimate 7 Studies indicate that self reporting is more accurate (Williams, 2001).
1: COMMITTEE REPORT 31 their incomes can differ. Most ask respondents to place themselves within specified earning ranges that differ in size. However, they use the general concept of"earnings" without distinguishing wage and salary earnings from investment earnings so that there is uncertainty about what is being reported. There are ways to address at least some of these differences. For example, AHRQ could provide leadership on the measures of socioeconomic status that are most useful for explaining disparities in health care. It could also work with public and private organizations to develop measures of socioeconomic status that can be supported by data that are feasible to collect for a range of surveys. The agency could also support longitudinal efforts, which allow health status to be measured before health care is utilized rather than at the same time. This makes it possible to draw stronger inferences about the causes behind racial, ethnic, socioeconomic, and geographic disparities in health care. Many subnational datasets are derived from cross- sectional surveys that provide detailed information about specific racial or ethnic groups or about specific geographic areas. AHRQ could explore ways to encourage or support conversion of some of these cross-sectional surveys into panel studies that would follow respondents over time. In any relationship AHRQ develops with other sponsors of subnational datasets, it should be remembered that agency policies are to an extent determined by federal policy. For example, AHRQ does not have the authority to formulate racial and ethnic categories that differ from those in OMB Directive 15. In addition, there are technical issues associated with subnational datasets that will be difficult to resolve for financial, administrative, and other reasons. For example, different datasets rely on different methods to determine the size of the subpopulation in question, which can produce varying estimates of prevalence and which also can affect estimates of the success of particular interventions. 1-6. GENERAL ISSUES Several general issues apply to the NHDR as a whole or to some or all of its topic areas in measurement and data source development, including the criteria that should guide the development
32 GUIDANCE FOR THE NATIONAL HEALTH CARE DISPARITIES REPORT of new measures for the NHDR. They include the importance of providing measures that policy makers, consumers, and others would find readily understandable. Measures should indicate areas in which policy makers can target legislative or other action to eliminate disparities. AHRQ should rely on the criteria for measurement development identified by the Committee on the National Quality Report on Health Care Delivery (IOM, 200Ic, pp. 82-7~. More specifically, measures should be: important, as indicated by their relationship to health status and relevance to policy makers and consumers; · Scientifically sound, as reflected by their validity, reliability, and substantiation; and Feasible, as indicated by current use or prototype development, data availability (including data for subpopulation analyses), and expense and difficulty of gathering the data. Applying these criteria, for example, could mean that the NHDR should rely on measures for conditions such as hypertension that are highly prevalent and/or that pose a substantial health burden in minority populations, as do cancer and end stage renal disease. It could also mean that the NHDR should focus on populations who have conditions with clinical needs that are well defined. This would include diabetics who need eye examinations and patients with histories of myocardial infarction who need beta blockers. In addition, it could mean that disparities highlighted in the NHDR should be related to national goals for health care specified in Healthy People 2010 such as access to clinical preventive care, emergency services, and Tong term care and rehabilitative services (DHHS, 2002~. As noted above, AHRQ will need to develop new measures in the areas of health care access, service utilization, and quality if the NHDR is to fulfill its potential of becoming a new and important government report on the nation's health care disparities and a critical too! to help guide action on their elimination. In addition, the agency
1: COMMITTEE REPORT 33 will need to develop new measures for both the NHQR and the NHDR that can cover components of their common framework, such as safety, for which few measures currently exist (TOM, 2001c; Lurie, 2002~. Both new and current measures of health care disparities will in turn require data on race, ethnicity, socioeconomic status, and geography that can support them. 7. AHRQ should receive adequate resources to develop datasets and measures needed for the National B:ealthcare Disparities Report. AHRQ has the responsibility of ensuring that the measures and datasets needed for the NHDR are developed. The exact nature and amount of resources that the agency wit! require are beyond the scope of this report. However, AHRQ will need expertise and substantial means to develop measures on access, service utilization, and quality that can meet the criteria of importance, scientific soundness, and feasibility. To do so, the agency will have to formulate and evaluate measures based on their scientific soundness and feasibility; solicit input from providers, patient groups, and others on measurement development and evaluation; and reexamine the measure set and its ability to reflect changes in the health care system and the nature of health care disparities (IOM, 200Ic). AHRQ also faces a number of challenges in obtaining the data necessary to support the report. While there are a number of datasets available, each has real limitations that the agency must develop strategies to address. The NHDR should not devote attention to the impact of behavior such as attitudes toward health on health care disparities. The contribution that personal attitudes make can be difficult to interpret, and many areas in which disparities occur do not involve patients' personal attitudes. Certainly, attitudes can make a difference in those areas of health care that depend on individual initiative or discretion, such as seeking the preventive care that insurance covers and complying with providers' instructions. However, in many of these areas, individual initiative or discretion can be more constrained than might first appear to be the case. For example, seeking preventive care that insurance covers can be more difficult when time off from work is limited and access to transportation is uncertain (Heckrnan et
34 GUIDANCE FOR THE NATIONAL HEALTH CARE DISPARITIES REPORT al., ~ 998; Heymann, 2000; Lannon et al., 1995; Perioff et al., 1997~. Complying with providers' instructions can be more difficult when literacy is a problem or when health insurance is lacking or does not fully cover the recommended treatment. Should the NHDR only devote attention to matters under the direct influence of the health care system? As defined by the Committee on the National Quality Report on Health Care Delivery, these matters refer to "care that can be influenced by the health care system as it exists or as it is envisioned" (TOM, 200Ic, p. 84~. To elaborates those in rural areas are more likelier to die from car accidents (Ricketts, 2002~. Although the health care system can seek to improve trauma care, many policy responses such as speed limits, road design, and car design fall outside of its purview. The NHDR should mainly, but not exclusively, address issues that the health care system could directly affect. It is appropriate for AHRQ to make health care quality and the quaTity-related performance of the health care system an important focus of the NHDR as well. However, disparities are inextricably related to issues that fall outside of the primary influence of the system such as the availability of public and private transportation (IOM, 2001a; TOM, 2002a; TOM, 2002b; Lurie, 2002~. Therefore the NHDR should address these disparities too. For example, the report could include data on disparities in reliance on public transportation, which can make timely treatment difficult for those in rural and inner city communities. The NHDR will change and improve over time. For practical reasons, it is likely that in the short term AHRQ would make use of current measures and data sources on racial, ethnic, socioeconomic, and geographic disparities in the early editions of the report. However, with time and adequate resources, AHRQ will have the opportunity to introduce more specific measures for use in later editions that will more accurately detect the magnitude of health care disparities. Measures could be drawn from those used in the NHQR that are particularly relevant to racial, ethnic, socioeconomic, and geographic disparities. These include measures of areas with unusually large disparities; those likely to result in death or serious illness; or those susceptible to improvement. Because of space limitations, AHRQ will monitor a larger set of measures than it can include in the NHQR.
1: COMMITTEE REPORT 35 Measures in this larger set could be used in the NHDR if they are more pertinent to disparities. Also, measures of disparities in health care service utilization and quality described by LaVeist (LaVeist, 2002) and measures of disparities in access described by Lurie (Lurie, 2002) could be used in the NHDR. Lastly, the design and dissemination of the NHDR will be critical to the report's success. A design that does not appeal to the report's audiences wall discourage them from reading it, using it as a reference source, and recommending it to others. Poor dissemination will mean that fewer people will leam about it, with the result that its annual updates on health care disparities will not have the impact that they should. The design and dissemination strategies for the NHQR developed by the Committee on the National Quality Report on Health Care Delivery (IOM, 2001 c) are sound approaches that could in general be applied to the NHDR as well. 1-7. CONCLUSION It is important to accurately identify the disparities that are primarily racial and ethnic and those that are primarily socioeconomic. Therefore, the NHDR should present findings on racial and ethnic health care disparities that reflect the impact of socioeconomic status. Also, there is currently an inadequate understanding of the relationship of socioeconomic status to health care. AHRQ should initiate research on the relationship with the goal of producing more useful, accurate, and meaningful measures of socioeconomic status. The NHDR should include measures of high utilization of certain health care services, such as greater minority use of emergency department care, that may indicate poor access to care or quality of care. To increase interest in the report by policy makers, consumers, and other key audiences, the NHDR should present data on disparities by state and by urban and by rural areas. Also, if the NHDR draws from subnational data sources in the Tong term, AHRQ should collaborate with data source sponsors to identify core elements in these surveys that can be standardized.
36 CUID~CE fOR THE NAT~AL HEALTH CAM DISPARITIES UPON ARM has He oppo~i~ to maw Tic ~DR ~ valu~lc Ed c~cOvc tool far cbminshng racist chick socioccono~c, Ed gco~bic disp~bcs in Tic nabon~s basalt chic system. lo tsar adv~tapc of ~k opposing ~RQ should reccivc adequ~c Rang Ed rcso~ccs to develop Tic dstascts Ed mess~cs ~~ Fig be needed far Tic repod.
1: COMMITTEE REPORT 37 Reference List Acevedo-Garcia, D. 2000. Residential segregation and the epidemiology of infectious diseases. Social Science & Medicine 51 (8~: 1143-61. Aday, L. A. 2001. At Risk in America. 2nd edition. San Francisco: Jossey- Bass Publishers. Andersen, R. M., A. McCutcheon, L. A. Aday, G. Y. Chin, and R. Bell. 1983. Exploring dimensions of access to medical care. Health Serv Res 18 (1~:49-74. Anderson, R. T., P. Sortie, E. Backlund, N. Johnson, and G. A. Kaplan. 1997. Mortality effects of community socioeconomic status. Epidemiology 8 (1~:42-47. Bashshur, R., D. G. Smith, and R. A. Stiles. 1993. Defining underinsurance: a conceptual framework for policy and empirical analysis. Med Care Rev 50 (2~:199-218. Burstin, H. 2002. The Importance of Health Care Disparities and the National Healthcare Disparities Report. Presentation to the Second Meeting of the IOM Committee for Guidance in Designing a National Healthcare Disparities Report (March 19~. California Pan-Ethnic Health Network (CPEHN). 2001. "Diverse patients, disparate experience: the use of standardized patient satisfaction surveys in assessing the cultural competence of health care organizations." Online. Available at http://www.cpehn.org/pdfs/diverse~atients.pdf [accessed June 13, 20023. Cooper, R. S., J. F. Kennelly, R. Durazo-Arvizu, H. J. Oh, G. Kaplan, and J. Lynch. 2001. Relationship between premature mortality and socioeconomic factors in black and white populations of US metropolitan areas. Public Health Rep 1 1 6 (5~:464-73. DHHS. 1999. "Mental Health: A Report of the Surgeon General." Online. Available at http :llwww. surgeongeneral. gov/library/mentalhealth/home.html "accessed July 18, 20023.
38 GUIDANCE FOR THE NATIONAL HEALTH CARE DISPARITIES REPORT . 2001. "Mental Health: Culture, Race and Ethnicity. A Supplement to Mental Health: A Report of the Surgeon General." Online. Available at http://www.surgeongeneral.gov/library/mentalhealth/cre/ [accessed Aug. 6, 20023. 2002. Healthy People 2010. Washington DC: DHHS. Diez Roux, A. V., S. S. Merkin, D. Arnett, L. Chambless, M. Massing, F. J. Nieto, P. Sortie, M. Szklo, H. A. Tyroler, and R. L. Watson. 2001. Neighborhood of residence and incidence of coronary heart disease. N EnglJMed 345 (2~:99-106. Fiscella, K. 2002. Using existing measures to monitor minority healthcare quality. In Improving Healthcare Quality for Minority Patients. National Quality Forum. Washington DC: National Forum for Health Care Quality Measurement and Reporting. Fiscella, K. and P. Franks. 2000. Individual income, income inequality, health, and mortality: what are the relationships? Health Serv Res 35 (1 Pt 2):307-18. Fiscella, K., P. Franks, M. Gold, and C. Clancy. 2000. Inequality in health: addressing socioeconomic, racial and ethnic disparities in health care. JAMA 283 (19):2579-82. Gaston, M. H., S. E. Barrett, T. L. Johnson, and L. G. Epstein. 1998. Health care needs of medically underserved women of color: the role of the Bureau of Primary Health Care. Health & Social Work 23 (2):86-95. Gold, R., I. Kawachi, B. P. Kennedy, J. W. Lynch, and F. A. Connell.2001. Ecological analysis of teen birth rates: association with community income and income inequality. Maternal & Child Health Journal 5 (3~: 161-67. Gornick, M. 2000. Vulnerable Populations and Medicare Services; Why Do Disparities Exist. New York: The Century Foundation Press. . 2002. Measuring socioeconomic status in health care. In Guidance for the National Healthcare Disparities Report. E. Swift, ed. Washington DC: National Academy Press.
1: COMMITTEE REPORT 39 Heckman, T. G., A. M. Somlai, J. Peters, J. Walker, L. Otto-Salaj, and C. A. Galdabini. 1998. Barriers to care among persons living with HIV/AIDS in urban and rural areas. AIDS Care 10 (3~:365-75. Heymann, J. 2000. The Widening Gap: Why America 's Working Families Are in Jeopardy and What Can Be Done About It. New York: Basic Books. IOM. 1999. Toward Environmental Justice: Research, Education and Health Policy Needs. Washington DC: National Academy Press. 2001a. Coverage Matters: Insurance and Health Care. Washington DC: National Academy Press. . 2001b. Crossing the Quality Chasm. Washington DC: National Academy Press. 2001 c. Envisioning the National Health Care Quality Report. M. Hurtado, E. Swift, and J. Corrigan, eds. Washington DC: National Academy Press. . 2002a. Care Without Coverage: Too Little, Too Late. Washington DC: National Academy Press. 2002b. Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. B. Smedley, A. Stith, and A. Nelson, eds. Washington DC: National Academy Press. Jackson, S. A., R. T. Anderson, N. J. Johnson, and P. D. Sortie. 2000. The relation of residential segregation to all-cause mortality: a study in black and white. Am J Public Health 90 (4~:615-17. Kahn, H. S., L. M. Tatham, E. R. Pamuk, and C. W. Heath, Jr. 1998. Are geographic regions with high income inequality associated with risk of abdominal weight gain? Social Science & Medicine 47 (1~:1-6. Kennedy, B. P., I. Kawachi, D. Prothrow-Stith, K. Lochner, and V. Gupta. 1998. Social capital, income inequality, and firearm violent crime. Social Science & Medicine 47 (1~:7-17.
40 GUIDANCE FOR THE NATIONAL HEALTH CARE DISPARITIES REPORT Kerbo, H. R. 1996. Social Stratification and Inequality: Class Conflict in Historical arid Comparative Perspective. 3rd edition. New York, NY: McGraw-Hill. Krieger, N., D. Williams, and N. Moss. 1997. Measuring social class in US public health research. Annu Rev Public Health 18:341-78. Lannon, C., V. Brack, J. Stuart, M. Caplow, A. McNeill, W. C. Bordley, and P. Margolis. 1995. What mothers say about why poor children fall behind on immunizations. A summary of focus groups in North Carolina. Arch Pediatr Adolesc Med 149 (10~: 1070-75. LaVeist, T. 2002. Measuring disparities in health care quality and service utilization. In Guidance for the National Healthcare Disparities Report. E. Swift, ed. Washington DC: National Academy Press. LaVeist, T. A. and J. M. Wallace, Jr. 2000. Health risk and inequitable distribution of liquor stores in African American neighborhood. Social Science & Medicine 51 (4~:613-17. Lavizzo-Mourey, R., and E. R. Mackenzie. 1996. Cultural competence: essential measurements of quality for managed care organizations. Ann Intern Med 124 (10):919-21. Liberatos, P., B. G. Link, and J. L. Kelsey. 1988. The measurement of social class in epidemiology. EpidemiolRev 10:87-121. Lochner, K., E. Pamuk, D. Makuc, B. P. Kennedy, and I. Kawachi. 2001. State-level income inequality and individual mortality risk: a prospective, multilevelstudy. AmJPublicHealth 91~3~:385-91. Lopez, R. 2002. Segregation and black/white differences in exposure to air tonics in 1990. Environ Health Perspect 110 (Suppl 2~:289-95. Lurie, N. 2002. Measuring disparities in access to care. In Guidance for the National Healthcare Disparities Report. E. Swift, ed. Washington DC: National Academy Press. Magnus, S. A., and S. S. Mick. 2000. Medical schools, affumative action, and the neglected role of social class. Am J. Public Health 90 (~8~:1 197- 201.
1: COMMITTEE REPORT Marmot, M. 2002. The influence of income on health: views of an epidemiologist. Health Aff(Millwood) 21 (2~:31-46. 41 Mayberry, R. M., F. Mill, and E. Ofili. 2000. Racial and ethnic differences in access to medical care. Med Care Res Rev 57 (Supp 1~:108-45. MedPAC. 2001. Report to the Congress: Medicare in Rural America. Washington DC: Medicare Payment Advisory Committee. Moore, R. D., D. Stanton, R. Gopalan, and R. E. Chaisson. 1994. Racial differences in the use of drug therapy for HIV disease in an urban community. N EnglJMed 330(11~:763-68. Morales, L. S., W. E. Cunningham, J. A. Brown, H. Liu, and R. D. Hays. 1999. Are Latinos less satisfired with communicatio by health care providers? J Gen Intern Med 14 (7) :409- 1 7. Morland, K., S. Wing, R. A. Diez, and C. Poole. 2002. Neighborhood characteristics associated with the location of food stores and food service places. Am J Prev Med 22 ~ 1~:23-29 Murray-Garcia, J. L. 2002. Multicultural Health 2002; An Annotated Bibliography. Woodland Hills CA: The California Endowment. National Center for Health Statistics. 2001. "NCHS publications and information products. Health, United States, 2001." Online. Available at http://www.cdc.gov/nchs/hus.htm [accessed Apr. 10, 20023. National Quality Forum. 2002. Improving Healthcare Quality for Minority Patients. Washington DC: National Forum for Health Care Quality Measurement and Reporting (National Quality Forum). Nerenz, D. R., M. J. Gunter, M. Garcia, R. R. Green-Weir, K. Wisdom, and C. Joseph. 2002. Developing a Health Plan Report Care on Quality of Care for Minority Populations. New York NY: The Commonwealth Fund. Oliver, M., and T. Shapiro. 2001. Wealth and racial stratification. In America Becoming: Racial Trends and Their Consequences. Vol. 2. N. J. Smelser, W. J. Wilson, and F. Mitchell, eds. Washington DC: National Academy Press.
42 GUIDANCE FOR THE NATIONAL HEALTH CARE DISPARITIES REPORT OMH [DHHS]. 2002. Good Health for All: Eliminating Racial and Ethnic Disparities in Health. The National Leadership Summit on Eliminating Racial and Ethnic Disparities in Health. Rockville MD: Department of Health and Human Services. Perez-Stable, E. J., F. Sabogal, and R. Otero-Sabogal. 1995. Use of cancer- screening tests in the San Francisco Bay area: comparison of Latinos and Anglos. JNatl CancerInstMonogr 18:147-53. Perloff, J. D., P. R. Kletke, J. W. Fossett, and S. Banks. 1997. Medicaid participation among urban primary care physicians. Med Care 35 (2~: 142-57. Polednak, A. P. 1996. Trends in US urban black infant mortality, by degree of residential segregation. Am J. Public Health 86 (5~:723-26. Ricketts, T. 2002. Geography and disparity in health. In Guidance for the National Healthcare Disparities Report. E. Swift, ed. Washington DC: National Academy Press. Sakoda, J. M. 1981. A generalized index of dissimilarity. Demography 18 (2~:245-50. Segal, S. P., J. R. Bola, and M. A. Watson. 1996. Race, quality of care, and antipsychotic prescribing practices in psychiatric emergency services. PsychiatrServ 47~3~:282-86. Shi, L. and B. Starfield. 2000. Primary care, income inequality, and self-rated health in the United States: a mixed-level analysis. Int J. Health Serv 30 (3~:541-55. . 2001. The effect of primary care physician supply and income inequality on mortality among blacks and whites in US metropolitan areas. Am J. Public Health 9 1 (8~: 1 246-50. Smith, D. B. 1998. The racial segregation of hospital care revisited: Medicare discharge patterns and their implications. Am J. Public Health 88 (3~:461-63. Smith, J. P. 2001. Race and ethnicity in the labor market. In America Becoming: Racial Trends and Their Consequences. Vol. 2. N. J. Smelser, W. J. Wilson, and F. Mitchell, eds. Washington DC: National Academy Press.
1: COMMITTEE REPORT 43 Takeuchi, D. R., K. Chung, H. Lin, K. Shen, C. Kurasaki, Chun, and S. Sue. 1998. Lifetime and twelve-month prevalence rates of major depressive episodes and dysthymia among Chinese Americans in Los Angeles. Am JPsychiamy 155 (10):1407-14. Takeuchi, D. T., S. Sue, and M. Yeh. 1995. Return rates and outcomes from ethnicity-specific mental health programs in Los Angeles. Am JPublic Health 85 (5):638-43. Waitzman, N. J. and K. R. Smith. 1998. Separate but lethal: the effects of economic segregation on mortality in metropolitan America. Milbank Q 76 (3):341-73. Williams,D.R. 2001. Racialvariationsin adult health status. In America Becoming: Racial Trends and Their Consequences. Vol. 2. N. J. Smelser, W. J. Wilson, and F. Mitchell, eds. Washington DC: National Academy Press. Williams, D. R. and C. Collins. 2001. Racial residential segregation: a fundamental cause of racial disparities in health. Public Health Rep 1 16 (5):404-1 6. Woloshin, S., N. A. Bickell, L. M. Schwartz, F. Gany, and H. G. Welch. 1995. Language barriers in medicine in the United States. JAMA 273 (9):724-28. Zoratti, E. M., S. Havstad, J. Rodriguez, Y. Robens-Paradise, J. E. Lafata, and B. McCarthy. 1998. Health service use by African Americans and Caucasians with asthma in a managed care setting. Am JRespir Crit Care Med 158 (2):371-77.