Health Affairs, 24, no. 2 (2005): 376-387 doi: 10.1377/hlthaff.24.2.376 © 2005 by Project HOPE	New Online   Pay Cuts For Medicare Docs   Access To Care Woes   Public Coverage More Efficient   Empowering Consumers

This Article Abstract Figures Only Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (12) Citing Articles via Google Scholar Google Scholar Articles by Moy, E. Articles by Clancy, C. M. Search for Related Content PubMed PubMed Citation Articles by Moy, E. Articles by Clancy, C. M. Related Collections Access To Care Consumer Issues Determinants Of Health Minority Health Public Health Politics

Public Response

Compiling The Evidence: The National Healthcare Disparities Reports

Abstract

 
Disparities in health care have been described extensively in the literature. The next step in resolving this national problem is to develop the necessary infrastructure for monitoring and tracking disparities. The congressionally mandated National Healthcare Disparities Report begins to build this infrastructure. The 2003 report addressed many of the methodological challenges inherent in measuring disparities. The recently released 2004 report continues the process by summarizing the status of U.S. health care disparities and beginning to track changes over time. Both reports emphasize the need to integrate activities to reduce disparities and to improve the quality of health care.

To gather information about disparities in health and health care, Congress directed the Agency for Healthcare Research and Quality (AHRQ) to produce an annual report to track "prevailing disparities in health care delivery as they relate to racial factors and socioeconomic factors."¹ The first National Healthcare Disparities Report (NHDR), released in 2003, is a comprehensive national overview of disparities in access to and quality of health care among racial, ethnic, and socioeconomic groups.² The recently released 2004 report adds another critical goal: tracking the nation’s progress toward elimination of disparities.³

The NHDR and its companion report, the National Healthcare Quality Report (NHQR), are designed to provide policymakers with snapshots of disparities and quality in the United States and assessments of change over time. Providers and payers also may apply some of the tools used to create these reports to their own data, to assess their performance relative to national benchmarks.

This paper reviews evidence about disparities in health care covered in the 2004 NHDR. Our goals are to describe challenges inherent in understanding which factors are most important for explaining disparities, present selected findings from the report, and discuss the advantages of addressing disparities in health care as an integral component of health care quality improvement.

Challenges To Understanding Disparities

Top Challenges To Understanding... Findings From The 2004... The Quality And Disparities... NOTES

 
Defining "disparities." To develop the NHDR, we had to address methodological limitations related to measurement and data. A key issue was defining "disparities," since existing definitions vary widely. Healthy People 2010 offers this explanation: "In pursuit of the overarching goal of eliminating health disparities, all differences among populations in measures of health and healthcare are considered evidence of disparities."⁴ In the Institute of Medicine (IOM) report Unequal Treatment, "disparities" are differences that remain after accounting for patients’ needs and preferences and the availability of health care.⁵ Still others associate health care disparities with adverse health outcomes, personal responsibility, or provider prejudice.

Because existing federal data needed to monitor and track disparities at the national level typically do not capture patients’ needs or preferences for care, we used a broad definition of "disparities" in the NHDR. Consistent with Healthy People 2010, any differences among populations are considered disparities.

Measurement issues. Based on IOM guidance, identical quality measures are used in the NHDR and NHQR. Most of these measures are relatively nondiscretionary. Three-quarters of the NHDR measures relate to effectiveness, typically with well-defined target populations based on need and clearly preferred treatment options based on rigorous scientific evidence. For example, recommendations for childhood vaccination apply to almost all children and are based on the deliberations of the Advisory Committee on Immunization Practices of the U.S. Centers for Disease Control and Prevention (CDC).⁶ Hence, lack of explicit information about patient need should have little impact on NHDR estimates.

In contrast, lack of information about patients’ preferences may affect estimates if different racial, ethnic, and socioeconomic groups tend to have different preferences. However, as noted in Unequal Treatment, until the perfect study controlling for all potential confounding factors is performed, it is reasonable to examine the consistency of disparities across existing studies.⁷ At a minimum, sizable differences associated with one or more patient characteristics raise important questions about subgroups at highest risk of poor-quality care.

Another measurement issue was how to handle the confounding effects of race, ethnicity, and socioeconomic status (SES).⁸ People included in racial and ethnic minority subgroups tend to have lower incomes and less education than whites; hence, racial and ethnic disparities often are highly correlated with SES disparities. To distinguish between the two types of disparities, the NHDR includes stratified and multivariate analyses when possible.

Multivariate analyses adjust for age, sex, race, ethnicity, household income, education, insurance, and residence location as appropriate. Stratified analyses identify groups by race, ethnicity, and SES that often experience the largest quality problems. For example, many racial and ethnic disparities in initiation of prenatal care in the first trimester persist even after stratification by education (Exhibit 1). Only college-educated white women achieved the Healthy People 2010 goal of 90 percent. Black, Native Hawaiian and other Pacific Islander (NHOPI), and American Indian and Alaska Native (AI/AN) women who did not graduate from high school had the lowest rates of early prenatal care and could gain the most from targeted interventions.

View larger version (30K): [in this window] [in a new window]   EXHIBIT 1 Mothers With Prenatal Care In The First Trimester, By Race And Ethnicity, Stratified By Education, 2001

Data issues. We also encountered many data issues that had to be addressed in producing the NHDR. Specific racial, ethnic, and socioeconomic groups often are not identifiable in national surveys. Federal agencies had until 2003 to be compliant with new federal guidelines for the collection of race and ethnicity data that created the "NHOPI" category and allowed respondents to choose more than one race.⁹ Because NHDR data were collected before 2003, estimates for multiple-race and NHOPI people were not possible for most measures.

Even when data are collected on specific groups, the data are not always comparable. Statewide data organizations often report race and ethnicity as a single field in their hospitalization data. People are identified as white or Hispanic but not both, which differs from federal standards.¹⁰ In addition, a patient’s race and ethnicity may be assigned by a hospital worker instead of being collected through self-identification. In the absence of federal standards for collection of socioeconomic data, different data sources often collect different income and education variables, again complicating comparisons.

When collected in a comparable fashion, data for specific racial and ethnic groups often are insufficient to generate reliable estimates because of small samples. These effects are magnified in attempts to assess disparities within groups, such as women and children. Across the 150 measures of quality tracked, reliable estimates for blacks and Hispanics were possible for more than 90 percent of measures. Reliable estimates for Asians and Pacific Islanders were possible for about three-quarters of measures; for AI/ANs, for about half of measures. However, reliable estimates were possible for less than one-fifth of measures for NHOPI and multiple-race people. Even fewer data are collected on racial and ethnic subgroups (for example, by national origin).

Noncontinuous data collection and data gaps make it difficult to track disparities over time. For most measures, the NHDR is able to analyze developments for only two or three years, a period too brief to reveal many changes. However, some changes are evident. Receipt of all recommended vaccinations by children ages 19–35 months is tracked from 2000 to 2002 (Exhibits 2 and 3). In all three years, the proportion of children receiving all recommended vaccines was lower among blacks than whites; Hispanics than non-Hispanic whites; and poor, near-poor, and middle-income than high-income children. Vaccination improved among Asian, Hispanic, and high-income children between 2000 and 2002.

View larger version (32K): [in this window] [in a new window]   EXHIBIT 2 Children Ages 19–35 Months Who Received All Recommended Vaccines, By Race And Ethnicity, 2000–2002

View larger version (30K): [in this window] [in a new window]   EXHIBIT 3 Children Ages 19–35 Months Who Received All Recommended Vaccines, By Family Income Level, 2000–2002

Findings From The 2004 Disparities Report

Top Challenges To Understanding... Findings From The 2004... The Quality And Disparities... NOTES

 
Addressing the challenges. To address challenges posed by inconsistent definitions, measurement problems, and variations in data standards, the general methodological approach of the NHDR is to examine many measures from a variety of national data sources, standardize data and comparisons whenever possible, and apply uniform and rigorous thresholds for identifying differences. Based on the consistency of disparities across measures, the 2003 NHDR concluded that inequalities persist at the national level.

Building upon the 2003 report, efforts to summarize disparities and track changes over time are refined in the 2004 report. To begin to quantify disparities systematically, a subset of measures for which comparable data are available for 2000 and 2001 are highlighted. This subset consists of thirty-eight measures of health care effectiveness (Exhibit 4). For each measure, racial, ethnic, and socioeconomic groups are compared with an appropriate reference group.¹¹

View larger version (39K): [in this window] [in a new window]   EXHIBIT 5 Percentage Of Measures Of Quality Of Care For Which The Group Performed Better Than, About The Same As, Or Worse Than The Comparison Group, 2001

View larger version (30K): [in this window] [in a new window]   EXHIBIT 6 Percentage Of Measures Of Quality Of Care For Which The Group Performed Worse Than The Comparison Group, 2000 And 2001

However, national findings may mask large variations in disparities nationwide. Disparities that are significant nationally may not exist in some communities but may be substantial in others. Apparent inertia in reducing disparities at the national level likely reflects sizable improvement in some communities and worsening disparities in others. Further, while some interventions may be effective nationally, others should be tailored to specific populations in specific communities.

Change occurs locally, and population subgroups vary geographically. Hence, community analyses are important for developing efficient strategies for reducing disparities. For example, because most black patients are seen by a relatively small fraction of providers, interventions focused on these providers could have a particularly strong impact on disparities for blacks.¹³ Communities and providers can use the NHDR measures and methods to identify their most serious disparities, target interventions, and track their progress against national standards.

The Quality And Disparities Chasms

Top Challenges To Understanding... Findings From The 2004... The Quality And Disparities... NOTES

 
One key insight gained from planning and producing the NHDR and NHQR is that quality and disparities are not isolated phenomena. Instead, they are integrally related concepts that benefit from coordinated interventions. In its report Crossing the Quality Chasm, the IOM recognized six dimensions of high-quality health care: effectiveness, safety, timeliness, patient-centeredness, efficiency, and equity.¹⁴ Although most of these dimensions have been the subject of study and intervention, equity has received little attention by the quality improvement (QI) community. QI efforts often do not explicitly identify elimination of disparities as an essential component. This apparent oversight misses an important opportunity because many potential synergies exist between activities to reduce disparities and improve quality.¹⁵

Focusing on disparities to improve quality. The disparities perspective may provide useful information to hone QI activities. Examining disparities may help identify the largest quality deficiencies and foster development of focused interventions for these problems. Although the gap between high-quality and actual care has been described as a chasm, the size of this chasm varies between groups. For the general population, there is a gap between ideal care and care received, but there is also a gap between advantaged and disadvantaged populations. These gaps compound forming the chasm between the actual care received by disadvantaged groups, including minorities and people of lower SES, and ideal high-quality health care. Targeting interventions for these disadvantaged groups may be a more efficient strategy to improve quality.

Analyses of disparities also may be useful in identifying achievable quality goals. Although different facilities may vary in their ability to maximize quality of care, it should be possible for all groups within a facility to do as well as the best-performing group. Additionally, facilities or communities with few disparities can provide valuable insight into factors that reduce disparities and interventions that have proved effective and warrant expansion.

Further, it may be necessary to take disparities into consideration to interpret public reporting fairly. Public reports of quality may be misinterpreted if they lack information about the demographic profile of the populations they serve. The increasing unreliability of further stratifying public reports does not mitigate this challenge, but it should prompt debate about what level of aggregation (for example, facility, county, state, or region) is most meaningful.

Building on quality improvement to reduce disparities. The existing QI infrastructure is well developed and can be adapted to address disparities: (1) Numerous organizations exist to promote and advance quality improvement. (2) Priority conditions for quality improvement and measures to track quality have been identified through consensus.¹⁶ (3) Quality report cards and other tools for public reporting have been produced and disseminated to consumers, providers, and purchasers.¹⁷ (4) A strategic plan for a national quality measurement and reporting system has been developed.¹⁸ Most of these QI activities are applicable or adaptable for addressing disparities. For example, most priority conditions for quality improvement, quality-of-care measures, and tools for public reporting are relevant to multiple groups.¹⁹

Many organizations have appreciated the synergies between disparities and quality. For example, the Health Resources and Services Administration (HRSA) created Health Disparity Collaboratives to eliminate disparities in health centers. The success of this program is evident in the absence of significant disparities and the sustained improvements in quality of care delivered by participating centers.²⁰

The Centers for Medicare and Medicaid Services (CMS) began the End-Stage Renal Disease Clinical Performance Measures Project to track and improve the quality of dialysis services provided under Medicare. This led to both improvements in dialysis care and reductions in disparities related to race and sex.²¹ Starting in 2005, the CMS will require Quality Improvement Organizations (QIOs) to begin work on reducing disparities in chronic illness care in physicians’ offices.

Extending pioneering work done by Aetna on the collection of data on race and ethnicity, ten of the leading U.S. health plans have joined with AHRQ and the Robert Wood Johnson Foundation (RWJF) to form the National Health Plan Learning Collaborative to Reduce Disparities and Improve Quality. Similarly, the Health Research and Educational Trust (HRET) partnered with six premier hospitals and the Commonwealth Fund to examine disparities and design and implement effective interventions through the new project, Eliminating Disparities through Community and Hospital Efforts. In another effort, the Center for Health Care Strategies formed the Best Clinical and Administrative Practices Group, comprising twelve Medicaid managed care teams working with consumers, providers, and state purchasers to assess and address disparities.

Future work. Despite these successes at integrating quality improvement and reduction of disparities, much work remains. Existing quality measures need to be validated in different populations, and new measures are needed to address the unique health care needs of specific groups. Methods are needed for summarizing disparities across measures and groups. As recommended by the National Research Council, data on racial, ethnic, and SES disparities need improvement.²² Public reporting on quality should account for differences in patient populations. Information for different groups should be produced when feasible, although reliable estimates for many groups will not be possible at the provider or facility level.

Research by AHRQ’s Excellence Centers to Eliminate Ethnic/Racial Disparities (EXCEED) and the National Center on Minority Health and Health Disparities (NCMHD) can be used as a foundation. EXCEED programs analyze underlying causes for disparities in health care and implement strategies to reduce disparities. For example, the EXCEED program at the Medical University of South Carolina demonstrated that voluntary primary care audit and feedback can improve control of cardiovascular risk factors, but people with multiple cardiovascular risks, particularly women and blacks, may require specifically tailored treatment guidelines to achieve desired levels of control.²³

The NCMHD leads the National Institutes of Health (NIH) research effort to reduce health disparities. Activities include the Endowment Program to strengthen the capacity of health disparities research, the Centers of Excellence Program to develop an infrastructure for disparities research at academic institutions, the Research Infrastructure in Minority Institutions Program to assist non–doctoral degree institutions in developing a research infrastructure, and the Loan Repayment Program to increase the number of health professionals conducting health disparities research.

Additionally, the NCMHD partners with other NIH institutes to identify and address risk factors for disparate health outcomes. For example, the NCMHD and National Eye Institute cofunded the Los Angeles Latino Eye Study. This study found that half of Latinos with diabetes age forty and older had diabetic retinopathy and that many were undiagnosed.²⁴

More work is required to understand the root causes of disparities and how these relate to the root causes of quality problems.²⁵ Causes common to both disparities and quality problems would likely be ideal targets for intervention. In addition, objective evaluations of interventions are needed to answer several fundamental questions: (1) How can synergies between disparities and quality improvement best be exploited? (2) When should QI interventions be applied broadly with the expectation that all groups will improve (a rising tide raises all boats), and when will tailored interventions be needed? (3) Recognizing that health care is sometimes segregated, should quality interventions target specific patient groups or specific provider groups with the most room for improvement? (4) Should quality interventions target affluent patients and providers who may be better able to sustain costs of higher-quality care with the expectation that improvements will diffuse to disadvantaged populations? (5) Is a combination of approaches more effective than a single strategy?

As knowledge about disparities in health care and the commitment to reducing them continue to grow, our ability to monitor and track improvements in disparities will become critical. Work begun in the 2003 NHDR to build the information infrastructure needed to track the nation’s progress toward eliminating disparities in health care is continued and expanded in the 2004 NHDR. Future editions will continue to build on advances in quality improvement and leverage the natural synergies that exist between efforts to improve quality and efforts to reduce disparities. Working together, using the NHDR as a guide, U.S. patients, providers, purchasers, and policymakers can make full access to high-quality health care a reality for all.

Editor's Notes

 
Ernest Moy (emoy{at}ahrq.gov) is director of the National Healthcare Disparities Report project, Center for Quality Improvement and Patient Safety, at the Agency for Healthcare Research and Quality (AHRQ) in Rockville, Maryland. Elizabeth Dayton is a junior service fellow in that center. Carolyn Clancy is director of AHRQ.

The views expressed in this paper are those of the authors and do not necessarily reflect the official position of the Agency for Healthcare Research and Quality or the U.S. Department of Health and Human Services.

NOTES

Top Challenges To Understanding... Findings From The 2004... The Quality And Disparities... NOTES

 

1. Healthcare Research and Quality Act of 1999 (P.L. 106–129), 113 Stat. 1653, 6 December 1999, www.ahcpr.gov/hrqa99a.htm (26 January 2005).
2. Agency for Healthcare Research and Quality, National Healthcare Disparities Report (2003), Pub. no. 04-0035, July 2003, www.qualitytools.ahrq.gov/disparitiesreport/download_report.aspx (1 November 2004).
3. AHRQ, National Healthcare Disparities Report (2004), Pub. no. 05-0014 (Rockville, Md.: AHRQ, 2004).
4. U.S. Department of Health and Human Services, Healthy People 2010: Understanding and Improving Health, 2d ed. (Washington: U.S. Government Printing Office, 2000).
5. B.D. Smedley, A.Y. Stith, and A.R. Nelson, eds., Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care (Washington: National Academies Press, 2002).
6. U.S. Centers for Disease Control and Prevention, "Recommended Childhood and Adolescent Immunization Schedule—United States, 2004," Morbidity and Mortality Weekly Report 53, no. 16 (2004): Q1–Q4.
7. Smedley et al., eds., Unequal Treatment.
8. E.K. Swift, ed., Guidance for the National Healthcare Disparities Report (Washington: National Academies Press, 2002).
9. Office of Management and Budget, "Revisions to the Standards for the Classification of Federal Data on Race and Ethnicity," Federal Register 62, no. 210 (1997): 58782–58790.
10. T. Hasnain-Wynia, D. Pierce, and M.A. Pittman, Who, When, and How: The Current State of Race, Ethnicity, and Primary Language Data Collection in Hospitals (New York: Commonwealth Fund, 2004).
11. The following data sources were used for the NHDR summary measure: the Surveillance, Epidemiology, and End Results (SEER) program; the U.S. Renal Data System (USRDS); the Medical Expenditure Panel Survey (MEPS); the CDC AIDS Surveillance System; the National Vital Statistics System–Natality (NVSS-N); the National Immunization Survey (NIS); the National Health Interview Survey (NHIS); and the National Hospital Discharge Survey (NHDS). Each racial, ethnic, or income group could receive care that is poorer than, about the same as, or better than the comparison group. Data on all measures were not available for all groups. Because these databases vary greatly in size and scope, two criteria were used to assess importance: statistical significance and the relative magnitude of differences. Only relative differences of at least 10 percent that are statistically significant at the 5 percent level are discussed. Without the magnitude criterion, many small differences from large databases would be labeled as significant.
12. The NHDR defines "poor" as those with household incomes below 100 percent of the federal poverty level and "high income" as those with household incomes 400 percent of poverty or higher. The federal poverty level for a family of four in 2003 was $18,400.
13. P.B. Bach et al., "Primary Care Physicians Who Treat Blacks and Whites," New England Journal of Medicine 351, no. 6 (2004): 575–584.[Abstract/Free Full Text]
14. IOM, Crossing the Quality Chasm: A New Health System for the Twenty-first Century (Washington: National Academies Press, 2001).
15. K.F. Aaron and C.M. Clancy, "Improving Quality and Reducing Disparities: Toward a Common Pathway," Journal of the American Medical Association 289, no. 8 (2003): 1033–1034.[Free Full Text]
16. IOM, Priority Areas for National Action: Transforming Health Care Quality (Washington: National Academies Press, 2003).
17. M.N. Marshall et al., "Public Reporting on Quality in the United States and the United Kingdom," Health Affairs 22, no. 3 (2003): 134–148[Abstract/Free Full Text]; J.A. Rainwater, P.S. Romano, and D.M. Antonius, "The California Hospital Outcomes Project: How Useful Is California’s Report Card for Quality Improvement?" Joint Commission Journal on Quality Improvement 24, no. 1 (1998): 31–39; and RAND, The First National Report Card on Quality of Healthcare in America: Research Highlights (Santa Monica, Calif.: RAND, 2004).
18. E.A. McGlynn, "An Evidence-based National Quality Measurement and Reporting System," Medical Care 41, no. 1 Supp. (2003): I8–I15.[CrossRef][ISI][Medline]
19. National Quality Forum, Workgroup on Healthcare Quality Measurement and Reporting for Minority Populations, Improving Healthcare Quality for Minority Patients (Washington: NQF, 2002).
20. R.M. Politzer et al., "Inequality in America: The Contribution of Health Centers in Reducing and Eliminating Disparities in Access to Care," Medical Care Research and Review 58, no. 2 (2001): 234–248.[Abstract/Free Full Text]
21. A.R. Sehgal, "Impact of Quality Improvement Efforts on Race and Sex Disparities in Hemodialysis," Journal of the American Medical Association 289, no. 8 (2003): 996–1000.[Abstract/Free Full Text]
22. National Research Council, Panel on HHS Collection of Race and Ethnic Data, Eliminating Health Disparities: Measurement and Data Needs (Washington: National Academies Press, 2004).
23. K.H. Hendrix, D.T. Lackland, and B.M. Egan, "Cardiovascular Risk Factor Control and Treatment Patterns in Primary Care," Managed Care Interface 16, no. 11 (2003): 21–26.[Medline]
24. R. Varma et al., "Prevalence of Diabetic Retinopathy in Adult Latinos: The Los Angeles Latino Eye Study," Ophthalmology 111, no. 7 (2004): 1298–1306.[CrossRef][ISI][Medline]
25. K.F. Aaron and F.D. Chesley Jr., "Beyond Rhetoric: What We Need to Know to Eliminate Disparities," Ethnicity and Disease 13, no. 3, Supp. 3 (2003): 9–11.

                      What's this?

This article has been cited by other articles:

				C. Clancy Improving Care Quality and Reducing Disparities: Physicians' Roles Arch Intern Med, June 9, 2008; 168(11): 1135 - 1136. [Full Text] [PDF]

				L. T. Bilheimer and J. E. Sisk Collecting Adequate Data On Racial And Ethnic Disparities In Health: The Challenges Continue Health Aff., March 1, 2008; 27(2): 383 - 391. [Abstract] [Full Text] [PDF]

				A. N. Trivedi, A. M. Zaslavsky, E. C. Schneider, and J. Z. Ayanian Relationship between quality of care and racial disparities in medicare health plans. JAMA, October 25, 2006; 296(16): 1998 - 2004. [Abstract] [Full Text] [PDF]

				R. F. Redberg Gender, Race, and Cardiac Care: Why the Differences? J. Am. Coll. Cardiol., November 15, 2005; 46(10): 1852 - 1854. [Full Text] [PDF]