Health Affairs, 24, no. 2 (2005): 398-408 doi: 10.1377/hlthaff.24.2.398 © 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 (15) Citing Articles via Google Scholar Google Scholar Articles by Lillie-Blanton, M. Articles by Hoffman, C. Search for Related Content PubMed PubMed Citation Articles by Lillie-Blanton, M. Articles by Hoffman, C. Related Collections Access To Care Business Of Health Consumer Issues Determinants Of Health Insurance Coverage Minority Health Politics

Insurers' Response

The Role Of Health Insurance Coverage In Reducing Racial/Ethnic Disparities In Health Care

Abstract

 
Research showing racial/ethnic disparities in medical care obtained by people with comparable insurance has raised questions about the extent to which health insurance improves opportunities for care. To assess whether insurance expansions could be expected to reduce racial/ethnic disparities in access to care, this paper reviews evidence from studies specifically designed to quantify the contribution of health insurance to racial/ethnic disparities in access. The studies provide evidence that a sizable share of the differences in whether a person has a regular source of care could be reduced if Hispanics and African Americans were insured at levels comparable to those of whites.

View larger version (16K): [in this window] [in a new window]   EXHIBIT 1 Uninsurance Rates Among Nonelderly Americans, By Race/Ethnicity, 2003

View larger version (17K): [in this window] [in a new window]   EXHIBIT 2 Profile Of Nonelderly Uninsured Americans, By Race/Ethnicity, 2003

Differences among the uninsured. Several studies provide evidence that uninsured African Americans and Hispanics fare worse than uninsured whites in obtaining access to care.⁴ For example, a nationally representative survey of adults in 2001 found that African Americans and Hispanics were more likely than whites to lack a regular provider, regardless of insurance status; however, the racial/ethnic gap between the uninsured was twice the size of the gap between those with insurance.⁵ Since these findings were not adjusted for other factors that might explain the disparity, they are useful in determining that an access gap existed but not whether factors other than insurance status might explain it. Nonetheless, they suggest that people of color, when uninsured, are doubly disadvantaged in the health system by their race and insurance status and that racial/ethnic disparities in health care could become worse if current levels of coverage among minority Americans decline.

Jobs with insurance offerings. Racial/ethnic minority Americans have lower rates of health insurance than whites largely because they are less likely to have employer-sponsored coverage (Exhibit 3). Although 70 percent of whites are insured through an employer-sponsored health plan, less than half of African Americans and Hispanics, the two largest U.S. racial/ethnic minority groups, had such coverage in 2003. This lower rate is explained in part by minorities’ greater employment in low-wage jobs, in which insurance coverage is either unavailable or unaffordable. Also, restrictions on legal and undocumented immigrants’ access to public programs such as Medicaid contribute to their lower rates of insurance.⁶

Citizenship. Uninsured Hispanics are less likely than whites or blacks to be U.S. citizens. About half (54 percent) of uninsured Hispanics (compared with less than 10 percent of uninsured blacks and whites) were not U.S. citizens in 2003.

Research showing that health care disparities persist, regardless of income and insurance, has led to an understandable emphasis on improving quality of care for those with public or private sources of coverage. However, the lack of health insurance is a barrier to medical care that is amenable to change and perhaps could have the single greatest impact on reducing racial/ethnic disparities in care. Therefore, we review the available evidence to assess whether, as well as the degree to which, expansions in coverage could be expected to reduce race- and ethnicity-associated barriers to health care.

Study Approach

Top Study Approach Study Results Discussion Policy Implications NOTES

 
Research review. To better understand the role of insurance coverage in reducing racial/ethnic disparities in access to care, we began by reviewing cited evidence from seminal reports on the issue—including the two IOM reports mentioned earlier and HHS’s National Healthcare Disparities Report.⁷ We also reviewed articles cited in background papers for the IOM reports that specifically summarized the literature addressing racial/ethnic health care disparities.⁸ Finally, we conducted a MEDLINE search, spanning 1995 to the present, to identify articles that answered the research question: How much of racial/ethnic disparities in access to care can be explained by differences in health insurance status?⁹ Or, stated another way, How much improvement in access to care might be expected if racial/ethnic differences in health insurance were eliminated?

Research criteria. For this review we focused only on original research that met all of the following criteria: (1) compared measures of access to health care between whites and a specific racial/ethnic group; (2) measured the effects of racial/ethnic differences in social, economic, or health system factors that may contribute to disparities in access to health care; and (3) quantified the contribution of racial/ethnic differences in health insurance on disparities in access to care. As research on health care disparities has evolved, researchers have only recently begun to separate and quantify the unique effect of specific factors known to affect access to care. We identified four studies to date that have applied regression-based methods so that multiple factors can be isolated to determine each factor’s contribution to racial/ethnic disparities in access to care.¹⁰ This group of studies provides the most compelling evidence on the potential role of health insurance in narrowing the racial/ethnic health care gap.

Common measure: usual source of care. Although each study examined two or three access measures, the only common measure across all of the studies was whether the person had a usual source of care.¹¹ This is a commonly used measure of a person’s ability to access high-quality care.¹² Numerous studies have found that people with a regular source of care are more likely than those without to obtain preventive, primary, and specialty care services.¹³ There is also evidence that having a regular provider is an equally, if not more, important factor than insurance status in the receipt of health services.¹⁴ Thus, we consider this to be an appropriate measure of health care access for our analysis.

Study Results

Top Study Approach Study Results Discussion Policy Implications NOTES

 
All four studies used national survey data from the late 1990s (1996, 1997, 1998, and pooled 1996–1999 surveys), and all of the researchers used statistical models that allowed them to simultaneously examine the effects of health insurance and other demographic, socioeconomic, and community factors. This method enabled them to weigh, and thereby compare, the contribution of each of the factors in explaining differences in access to care.

Researchers’ models varied, but all of the studies used data sources that could yield national estimates. Two studies used data from the Medical Expenditure Panel Survey (MEPS) (Weinick, Zuvekas, and Cohen, 2000; and Zuvekas and Taliaferro, 2003). Waidmann and Rajan (2000) used the National Survey of America’s Families, providing both national and several state-level estimates, while Hargraves and Hadley (2003) pooled two rounds of data from the Community Tracking Study (CTS) household surveys.

Hispanic-white disparities. Hispanics were significantly less likely than whites to have a usual source of care; for example, Samuel Zuvekas and Gregg Taliaferro found that 30 percent of Hispanics compared with 15 percent of whites did not have a usual source of care.¹⁵ This access gap of fifteen to sixteen percentage points was consistent across all four studies (Exhibit 5).

In no study did income differences between Hispanics and whites explain more of the access gap than health insurance differences. In fact, income explained considerably less of the access difference (for example, 8 percent and 19 percent compared with 33 percent explained by insurance differences between Hispanics and whites). Other socioeconomic factors, such as education, family status, employment, and health status, all explained less than 10 percent of the Hispanic-white disparity in having a usual source of care, with one exception. In the study by Timothy Waidmann and Shruti Rajan, differences between whites and Hispanics in citizenship explained 19 percent of the access disparity (still less than the 33 percent explained by health insurance in their study).¹⁶

Black-white disparities. Each of the four studies also found significant disparities between blacks and whites in having a usual source of care (Exhibit 5). The access gap was smaller, however, than that between Hispanics and whites, with differences ranging from 4.4 to 8.4 percentage points.

In all but one study, differences in health insurance between blacks and whites were again the single largest observable factor explaining the disparity in having a usual source of care. In these three studies, health insurance explained 24–42 percent of the difference, with income explaining 14–20 percent. The only other single factor explaining a sizable share of the difference between blacks’ and whites’ access to a usual source of care was family status (26 percent, Waidmann and Rajan), which was incorporated into just one of the studies’ explanatory models.

Interestingly, two studies included some health system capacity measures in addition to the individual characteristics that were common to most of the studies. These studies found that the indicators of provider capacity in a county (Zuvekas and Taliaferro) and the availability of more "safety-net services" and charity care (Hargraves and Hadley) contributed very little to explaining racial/ethnic access disparities.

In summary, these four studies—specifically designed to quantify the contribution of a variety of factors toward racial/ethnic disparities in access to a usual source of care—are nearly all consistent in the finding that racial and ethnic insurance differences explain a significant and sizable share of these disparities.

Discussion

Top Study Approach Study Results Discussion Policy Implications NOTES

 
The studies identified for this review provide compelling evidence that health insurance accounts for much of the variation in racial/ethnic disparities in access to a usual source of care. The issue is not whether either race or health insurance is independently associated with experiences in the health care system. The preponderance of evidence from credible research has shown that they are. Rather, the issue is whether and to what extent does racial/ethnic background interact with health insurance to affect a person’s access to care.

Of the eight comparisons in this review, seven are consistent in showing that health insurance is a sizable, although not the only, factor in explaining racial/ethnic disparities in having a regular source of medical care. Health insurance explained a statistically significant share of the access gap in all four comparisons between Hispanics and whites and in three of the four comparisons between African Americans and whites.

The one exception to these otherwise consistent findings was the study that also differed from the others in the way it structured its statistical analysis. The Weinick and Zuvekas (2000) study used a single regression equation with interaction terms to separate the effect of racial/ethnic differences in health insurance on access to care. In contrast, the other three studies applied a regression decomposition method with separate regression equations for each racial/ethnic group that allowed them to determine how much of the access disparity could be explained by each of the observed factors versus unobserved factors (such as racial differences in insurance and income that were measured versus those that were not measured, such as help-seeking behavior).¹⁷

Despite the use of fairly sophisticated analytic techniques, disentangling interactions between race/ethnicity and health insurance remains complex. The limitations reflect both the state of the art of the measurement tools and the fact that race/ethnicity continues to be deeply intertwined with life experiences in the United States. As such, it is difficult to sort out the influence of factors such as education and income that are associated with racial/ethnic identity. In fact, the studies in this review also show that other socioeconomic and "unmeasured" factors explain a substantial share of the access gaps. However, many of the other socioeconomic factors are not easily addressed through health policy reforms.

Taken together, the results from these four studies, and their consistency, suggest that racial/ethnic disparities in access to a regular source of care could be greatly reduced by greater equity in health insurance coverage. Our findings show that by equalizing levels of health coverage, the United States could conceivably address roughly one-third (specifically, 23–42 percent, with one exception) of the reason for disparity in a measure of access that is well recognized as improving opportunities for high-quality health care.

Policy Implications

Top Study Approach Study Results Discussion Policy Implications NOTES

 
Recent proposals to expand insurance coverage focus on a combination of public- and private-sector approaches to increase coverage incrementally. Reducing the number of uninsured people in communities of color requires political will as well as effective strategies that are based on knowledge of the characteristics of the uninsured and those at risk of being uninsured.

Given current federal and state fiscal realities, policy options that aim to increase coverage through private-sector approaches are politically attractive but would likely have minimal impact. Of the twenty-three million uninsured in communities of color, approximately three-fourths are in families with incomes below 200 percent of the federal poverty level ($37,620 for a family of four in 2003). Unless private-market insurance options include a sizable public subsidy to make coverage affordable, they are unlikely to have much impact on rates of coverage.

Evidence from a number of studies indicates that public programs are the most effective approach to expanding health insurance coverage.¹⁸ Medicaid and the State Children’s Health Insurance Program (SCHIP) have the administrative means already in place to enroll beneficiaries and pay providers, and they have demonstrated how responsive they can be when unemployment increases and family incomes decrease. Even as Medicaid enrollment has risen with the recent economic downturn, the growth in Medicaid spending has not been as great as that of employer-sponsored insurance premiums.¹⁹ Making public health insurance programs available to low-income populations, as opposed to the current categorical coverage that excludes most adults without dependent children, would be a more equitable public policy than the current system.

Several changes in public policy could have a sizable impact on the health coverage of persons of color. An estimated 74 percent of the twenty-three million uninsured minority Americans could be covered by using Medicaid and SCHIP to (1) expand outreach and enrollment efforts to assure that all children who are eligible—children in families with incomes less than 200 percent of poverty—are enrolled in these programs (approximately 4.6 million low-income minority children); (2) expand coverage to parents of children who are enrolled in these programs (approximately 5.0 million low-income minority parents); and (3) expand coverage to low-income adults without dependent children (approximately 7.5 million).²⁰

We believe that our analysis shows that health insurance is one of many factors that interact with race/ethnicity and the health care system to play a major role in people’s ability to obtain high-quality health care. As such, efforts to maintain and expand health insurance coverage should be a major component of any strategy to reduce racial/ethnic disparities in care.

Editor's Notes

 
Marsha Lillie-Blanton (MLBlanton{at}kff.org) is a vice president in health policy at the Henry J. Kaiser Family Foundation in Washington, D.C. Catherine Hoffman is an associate director of the Kaiser Commission on Medicaid and the Uninsured.

This research was supported by the Henry J. Kaiser Family Foundation. The opinions expressed in this paper are solely those of the authors and do not reflect the positions of the foundation. The authors gratefully acknowledge Diane Rowland and Linda Wright Moore for their thoughtful comments on earlier drafts of the paper, and Alicia Carbaugh, Sonia Ruiz, and Kinite Holt for their assistance in the research and preparation.

NOTES

Top Study Approach Study Results Discussion Policy Implications NOTES

 

1. U.S. Department of Health and Human Services, "HHS Secretary Tommy G. Thompson Launches Third Annual ‘Take a Loved One to the Doctor Day’," Press Release, 12 July 2004, www.hhs.gov/news/press/2004pres/20040712.html (16 December 2004).
2. Kaiser Commission on Medicaid and the Uninsured, Health Insurance Coverage in America: 2003 Data Update (Washington: Kaiser Commission, November 2004).
3. 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); and Institute of Medicine, Coverage Matters: Insurance and Health Care (Washington: National Academies Press, 2001).
4. See,, for example, J.L. Hargraves, "The Insurance Gap and Minority Health Care, 1997–2001," Center for Studying Health System Change Tracking Report no. 2 (Washington: HSC, June 2002); M. Lillie-Blanton, R.M. Martinez, and A. Salganicoff, "Site of Medical Care: Do Racial and Ethnic Differences Persist?" Yale Journal of Health Policy, Law, and Ethics 1 (Spring 2001): 15–32; and J.Z. Ayanian et al., "Unmet Health Needs of Uninsured Adults in the United States," Journal of the American Medical Association 284, no. 16 (2000): 2061–2069.[Abstract/Free Full Text]
5. Hargraves, "The Insurance Gap." The gap in the percentage of African American and white adults with a regular provider among the uninsured (36 percent versus 51 percent) was twice as large as the gap among the insured (71 percent, 78 percent). Findings were similar for the Latino-white gap between uninsured adults with a regular provider (31 percent, 51 percent) and insured adults (69 percent, 78 percent).
6. L. Ku and T. Waidmann, "How Race/Ethnicity, Immigration Status, and Language Affect Health Insurance Coverage, Access to Care, and Quality of Care among the Low-Income Population" (Washington: Kaiser Commission on Medicaid and the Uninsured, August 2003).
7. Smedley et al., eds., Unequal Treatment; IOM, Coverage Matters; and Agency for Healthcare Research and Quality, National Healthcare Disparities Report (Washington: U.S. Government Printing Office, 2003).
8. J.S. Haas and N.E. Adler, "The Causes of Vulnerability: Disentangling the Effects of Race, Socioeconomic Status, and Insurance Coverage on Health," Background Paper prepared for the IOM Committee on the Consequences of Uninsurance, October 2001; and H.J. Geiger, "Racial and Ethnic Disparities in Diagnosis and Treatment: A Review of the Evidence and Consideration of Causes," Background Paper for IOM, Unequal Treatment, 417–454.
9. We reviewed 285 studies (145 identified by the PubMed/MEDLINE search of articles published from 1995 to the present).
10. The four studies are (1) R.M. Weinick, S.H. Zuvekas, and J.W. Cohen, "Racial and Ethnic Differences in Access to and Use of Health Care Services, 1977 to 1996," Medical Care Research and Review 57, Supp. 1 (2000): 36–54 (Weinick and colleagues analyzed data on 20,793 people under age sixty-five from a nationally representative survey in 1996)[Abstract/Free Full Text]; (2) T.A. Waidmann and S. Rajan, "Race and Ethnic Disparities in Health Care Access and Utilization: An Examination of State Variation," Medical Care Research and Review 57, Supp. 1 (2000): 55–84 (the authors analyzed 1997 data from 45,000 households drawn to be representative of each of thirteen states and the rest of the country)[Abstract/Free Full Text]; (3) S.H. Zuvekas and G.S. Taliaferro, "Pathways to Access: Health Insurance, the Health Care Delivery System, and Racial/Ethnic Disparities, 1996–1999," Health Affairs 22, no. 2 (2003): 139–153 (the study included a sample of 25,565 adults from a nationally representative 1998 survey)[Abstract/Free Full Text]; and (4) J.L. Hargraves and J. Hadley, "The Contribution of Insurance Coverage and Community Resources to Reducing Racial/Ethnic Disparities in Access to Care," Health Services Research 38, no. 3 (2003): 809–829 (this study analyzed data on 96,414 adults under age sixty-five from a nationally representative survey conducted in 1996–97 and 1998–99).[CrossRef][ISI][Medline]
11. Hargraves and Hadley defined their access measure of having a usual source of care somewhat more precisely: as seeing the same health care provider at each visit to a regular source of care. While their measure is more precise than the "having a usual source of care" measure of the other studies we reviewed, relative to other means of measuring access to care, it is substantially comparable and relevant to our purposes here.
12. N. Lurie, "Measuring Disparities in Access to Care," in IOM, Guidance for the National Healthcare Disparities Report, ed. E.K. Swift (Washington: National Academies Press, 2000).
13. See, for example, R.M. Weinick and K.M. Beauregard, "Women’s Use of Preventive Screening Services: A Comparison of HMO versus Fee-for-Service Enrollees," Medical Care Research and Review 54, no. 2 (1997): 176–199[Abstract/Free Full Text]; and C.M. Sox et al., "Insurance or a Regular Physician: Which Is the Most Powerful Predictor of Health Care?" American Journal of Public Health 88, no. 3 (1998): 364–370.[Abstract/Free Full Text]
14. J.M. Lambrew et al., "The Effects of Having a Regular Doctor on Access to Primary Care," Medical Care 34, no. 2 (1996): 138–151[CrossRef][ISI][Medline]; and J.E. DeVoe et al., "Receipt of Preventive Care among Adults: Insurance Status and Usual Source of Care," American Journal of Public Health 93, no. 5 (2003): 786–791.[Abstract/Free Full Text]
15. Zuvekas and Taliaferro, "Pathways to Access."
16. Waidmann and Rajan, "Race and Ethnic Disparities."
17. When these four studies analyzed data on other measures of access (percentage with medical visit, percentage with breast exam, and unmet health needs), findings on the contribution of health insurance to the racial/ethnic access gap varied more widely across studies and differed for Hispanics versus African Americans. For example, health insurance explained 10–80 percent of the access gap in the percentage with a medical visit in a year. Based on discussion with at least one of the authors, we think that the wider variation reflects a number of factors, including differences in the study years, the age range of participants, and the study methods. Also, insurance coverage appeared to explain less of the Hispanic-white gap than the black-white gap in the percentage with a medical visit. The variability in findings for the two racial/ethnic minority groups could be due to factors such as language and even heterogeneity among the groups identified as Hispanic.
18. J. Feder et al., "Covering the Low-Income Uninsured: The Case for Expanding Public Programs," Health Affairs 20, no. 1 (2001): 27–39[Abstract/Free Full Text]; J. Gruber, "Evaluating Alternative Approaches to Incremental Health Insurance Expansion," Unpublished data presented at the Alliance for Health Reform briefing, Washington, D.C., 16 June 2004; and J. Feder et al., "Assessing the Combination of Public Programs and Tax Credits," in Covering America: Real Remedies for the Uninsured (Washington: Economic and Social Research Institute, December 2003).
19. Henry J. Kaiser Family Foundation and Health Research and Educational Trust, Employer Health Benefits: 2004 Annual Survey (Washington: Kaiser/HRET, 2004); and V. Smith et al., States Respond to Fiscal Pressure: A Fifty-State Update of State Medicaid Spending Growth and Cost Containment Actions (Washington: Kaiser Commission on Medicaid and the Uninsured, January 2004).
20. The estimate of low-income children that would qualify for Medicaid or SCHIP does not consider any other state, citizenship, or other categorical requirements that may affect eligibility. Based on our calculation, this estimate excludes the approximately one million uninsured minority children with incomes of 200 percent or above poverty. With regard to extending insurance coverage to low-income, minority parents, in 2003 only sixteen states covered parents with incomes at 100 percent of poverty.

                      What's this?

This article has been cited by other articles:

				R. J. Blendon, T. Buhr, E. F. Cassidy, D. J. Perez, T. Sussman, J. M. Benson, and M. J. Herrmann Disparities In Physician Care: Experiences And Perceptions Of A Multi-Ethnic America Health Aff., March 1, 2008; 27(2): 507 - 517. [Abstract] [Full Text] [PDF]

				P. Neuman, J. Cubanski, K. A. Desmond, and T. H. Rice How Much 'Skin In The Game' Do Medicare Beneficiaries Have? The Increasing Financial Burden Of Health Care Spending, 1997 2003 Health Aff., November 1, 2007; 26(6): 1692 - 1701. [Abstract] [Full Text] [PDF]

				M. L. Beach, A. B. Flood, C. M. Robinson, A. N. Cassells, J. N. Tobin, M. A. Greene, and A. J. Dietrich Can Language-Concordant Prevention Care Managers Improve Cancer Screening Rates? Cancer Epidemiol. Biomarkers Prev., October 1, 2007; 16(10): 2058 - 2064. [Abstract] [Full Text] [PDF]

				J. Hadley, P. Cunningham, and J. L. Hargraves Would Safety-Net Expansions Offset Reduced Access Resulting From Lost Insurance Coverage? Race/Ethnicity Differences Health Aff., November 1, 2006; 25(6): 1679 - 1687. [Abstract] [Full Text] [PDF]

				T D Simon, C Bublitz Emsermann, L M Dickinson, and S J Hambidge Lower rates of emergency department injury visits among Latino children in the USA: no association with health insurance. Inj. Prev., August 1, 2006; 12(4): 248 - 252. [Abstract] [Full Text] [PDF]

				Y.-C. T. Shih, L. Zhao, and L. S. Elting Does Medicare coverage of colonoscopy reduce racial/ethnic disparities in cancer screening among the elderly? Health Aff., July 1, 2006; 25(4): 1153 - 1162. [Abstract] [Full Text] [PDF]

				B. Starfield Insurance and the U.S. Health Care System N. Engl. J. Med., July 28, 2005; 353(4): 418 - 419. [Full Text] [PDF]