Trends in Health Insurance Coverage and Access Among Black, Latino and White Americans, 2001-2003

Tracking Report No. 11
October 2004
J. Lee Hargraves

Overall health insurance rates changed little among nonelderly black, Latino and white Americans between 2001 and 2003, according to new findings from the Center for Studying Health System Change (HSC). But sources of coverage shifted—especially for Latinos—from employment-based insurance to public coverage, suggesting the economic downturn took a greater toll on Latinos. Low-income Latinos and whites were particularly hard hit by declines in employer coverage. Shifting sources of coverage had little effect on access to medical care. With the sole exception of decreased access to specialists among blacks, access to care did not change between 2001 and 2003. Significant gaps in access to care among Latinos, blacks and whites persisted, with Latinos and blacks consistently reporting lower levels of access than whites.



Shifting Insurance Coverage

he ongoing gap in the proportion of uninsured Latino, black and white Americans essentially remained unchanged between 2001 and 2003, with one in three Latinos, one in five blacks and one in 10 whites under age 65 lacking health insurance in 2003 (see Table 1). While overall coverage rates remained fairly constant, coverage sources shifted—particularly for Latinos, who saw a marked decline in employer-sponsored coverage and a commensurate increase in public coverage.

All Americans saw a drop in access to employer-sponsored health insurance between 2001 and 2003, but the decline was especially severe for Latinos. Less than 65 percent of nonelderly Latinos had access to health insurance from employers in 2003, compared with more than 71 percent in 2001 (see Table 2). Moreover, the proportion of eligible Latinos who took up an offer of employer coverage dropped from 79.5 percent in 2001 to 72.3 percent in 2003. In comparison, more than 80 percent of whites and blacks had access to employer coverage in 2003, while 92.2 percent of whites and 83.6 percent of blacks took up offers of employer-sponsored insurance—rates that have remained virtually unchanged since 1997.

Back to Top



Table 1
Insurance Coverage Among Nonelderly Blacks, Latinos and Whites, 1997-2003

 
1997
1999
2001
2003
Employer-Sponsored Health Insurance
Black
53.5%
56.0%
56.3%
53.9%
Latino
47.1
48.5
47.8
40.4*#
White
73.9
74.3
75.1
73.5*
Public Health Insurance 1
Black
19.3
19.8
20.5
22.6#
Latino
13.7
14.7
15.6
22.5*#
White
4.8
5.3
5.9
8.1*#
Other Health Insurance 2
Black
7.2
5.9
5.2
4.7#
Latino
5.5
5.4
4.9
4.3
White
8.9
8.7
8.4
7.5*#
Uninsured
Black
19.9
18.3
18.0
18.8
Latino
33.7
31.5
31.7
32.8
White
12.5
11.8*
10.6*
11.0#
1 Public coverage includes Medicaid, SCHIP and other state programs.

2 Other coverage includes private nongroup insurance, private insurance obtained from someone outside the family, Indian Health Service and other miscellaneous coverage. Military insurance and Medicare for disabled persons are excluded from this analysis.

* Change from previous survey is statistically significant at p <.05.

# Change from 1997 to 2003 is statistically significant at p <.05.

Note: Bold text shows statistically significant differences from whites.

Source: Community Tracking Study Household Survey



Table 2
Employer-Sponsored Health Insurance Among Nonelderly Blacks, Latinos and Whites, 1997-2003

 
1997
1999
2001
2003
Nonelderly Persons in Working Families1
Black
70.3%
72.5%
72.3%
67.9%*
Latino
74.8
77.7*
74.5
70.3
White
85.0
85.8
84.9
83.3*#
People in Families with Access to Employer-Sponsored Insurance2
Black
80.8
81.5
84.4
81.3
Latino
69.0
70.7
71.4
64.9*
White
85.9
86.1
87.1
86.0
People with Access to Employer Insurance who Take Up Insurance3
Black
83.8
86.1
83.8
83.6
Latino
81.1
80.0
79.5
72.3*
White
92.8
93.2
93.3
92.2
1 A working family is defined as one in which total number of hours worked by all adult members of the family is 20 or more hours per week. Dependents of adults on active military duty are included while families in which all adult members are self-employed and have no paid employees are excluded.

2 Access rate is defined at the family level. As long as one member of the family has access to employer coverage, all members of that family have access, excluding people with health insurance from someone outside of the family.

3 The take-up rate is defined at the person level, since it is possible for some family members to be covered by an employer, while others are uninsured or have other coverage.

* Change from previous survey is statistically significant at p <.05.

# Change from 1997 to 2003 is statistically significant at p <.05.

Note: Bold text shows statistically significant differences from whites.
Source: Community Tracking Study Household Survey


Low-Income Latinos and Whites Lose

hile employer coverage for low-income blacks remained fairly constant between 2001 and 2003, low-income Latinos and whites—defined as income below 200 percent of the federal poverty level, or $36,800 for a family of four in 2003—saw dramatic declines in employer-sponsored insurance (see Table 3). Employer coverage for low-income Latinos dropped from 28.3 percent in 2001 to 22.9 percent in 2003, while low-income whites with employer coverage dropped from 46.3 percent to 41.8 percent during the same period.

The proportion of low-income Latinos with access to employer coverage dropped 10 percentage points between 2001 and 2003, from 58 percent to 48 percent. The take-up rate among low-income Latinos with access to employer coverage also declined significantly from 64.5 percent in 2001 to 53.6 percent in 2003 (see Table 4).

Low-income white Americans also saw a significant decline in access to employer coverage, with the proportion eligible for employer coverage dropping from 72.6 percent in 2001 to 64.8 percent in 2003. Unlike Latinos, however, low-income whites’ take-up rates remained fairly constant with almost 77 percent taking up employer coverage in 2003.

Back to Top



Table 3
Insurance Coverage Among Low-Income Nonelderly Blacks, Latinos and Whites, 1997-2003

 
1997
1999
2001
2003
Employer-Sponsored Health Insurance
Black
32.7%
32.2%
35.9%
32.1%
Latino
31.8
29.4
28.3
22.9#
White
48.9
44.3*
46.3
41.8*#
Public Health Insurance 1
Black
32.8
34.8
33.8
37.7
Latino
20.1
23.7
24.8
33.4*#
White
15.2
18.2*
20.5
27.1*#
Other Health Insurance 2
Black
8.1
7.8
5.4
5.1#
Latino
5.2
4.3
4.1
2.6*#
White
9.6
10.5
9.1
7.6#
Uninsured
Black
26.4
25.2
24.9
25.1
Latino
42.9
42.6
42.8
41.1
White
26.3
27.0
24.1
23.6#
1 Public coverage includes Medicaid, SCHIP and other state programs.

2 Other coverage includes private nongroup insurance, private insurance obtained from someone outside the family, Indian Health Service and other miscellaneous coverage. Military insurance and Medicare for disabled persons are excluded from this analysis.

* Change from previous survey is statistically significant at p <.05.

# Change from 1997 to 2003 is statistically significant at p <.05.

Notes: Bold text shows statistically significant differences from whites. Low income is defined as having a family income less than 200 percent of the federal poverty level.

Source: Community Tracking Study Household Survey



Table 4
Employer-Sponsored Health Insurance Among Low-Income Nonelderly Blacks, Latinos and Whites, 1997-2003

 
1997
1999
2001
2003
Nonelderly Persons in Working Families 1
Black
56.3%
58.1%
59.1%
54.1%
Latino
66.6
71.1*
65.6
62.5
White
73.2
73.0
70.9
68.4#
People in Families with Access to Employer-Sponsored Insurance 2
Black
68.4
65.5*
73.4
69.5
Latino
57.6
57.1
58.0
48.0*#
White
73.0
68.4*
72.6*
64.8*#
People with Access to Employer Insurance who Take Up Insurance 3
Black
71.9
71.1
70.9
64.4
Latino
69.5
64.8
64.5
53.6*#
White
82.1
79.2
79.6
76.8#
1 A working family is defined as one in which total number of hours worked by all adult members of the family is 20 or more hours per week. Dependents of adults on active military duty are included while families in which all adult members are self-employed and have no paid employees are excluded.

2 Access rate is defined at the family level. As long as one member of the family has access to employer coverage, all members of that family have access, excluding people with health insurance from someone outside of the family.

3 The take-up rate is defined at the person level, since it is possible for some family members to be covered by an employer, while others are uninsured or have other coverage.

* Change from previous survey is statistically significant at p <.05.

# Change from 1997 to 2003 is statistically significant at p <.05.

Notes: Bold text shows statistically significant differences from whites. Low income is defined as having family income less than 200 percent of the federal poverty level.

Source: Community Tracking Study Household Survey


Public Coverage Increases

ublic insurance—primarily Medicaid and the State Children’s Health Insurance Program (SCHIP)—filled insurance coverage gaps for many nonelderly Americans, especially Latinos. In 2003, slightly less than one in four Latinos (22.5%) had public insurance, compared with one in six (15.6%) in 2001.1 The proportion of nonelderly whites with public coverage increased from 5.9 percent in 2001 to 8.1 percent in 2003, while the rate of blacks with public coverage in 2003 was 22.6 percent—statistically unchanged from 2001.

The increase in public coverage among Latino children was especially striking (see Supplementary Table 1). In 2001, nearly one in three (29.7%) Latino children was covered by public insurance, compared with more than two in five (43.6 %) in 2003. Between 2001 and 2003, both Latino adults and children saw increases in public insurance and declines in employersponsored insurance, and the pattern was similar, yet not as sharp among non-Latino whites.

Among low-income people, shifts toward public coverage were even more pronounced among nonelderly Latinos and whites. Low-income Latinos with public coverage increased from 24.8 percent in 2001 to 33.4 percent in 2003, while, at the same time, public coverage rates for low-income whites increased from 20.5 percent to 27.1 percent.

Back to Top



Access Gaps Remain

hifting forms of insurance coverage—from employer coverage to public insurance—had little effect on access to medical care among nonelderly blacks, Latinos and whites. In assessing minority health care disparities, four measures of access among whites, blacks and Latinos were examined:

With the sole exception of decreased access to specialists among blacks, access to care did not change from 2001 to 2003. Moreover, gaps in access between Latinos, blacks and whites persisted. In tracking access to medical care between 1997 and 2003, nonelderly blacks and Latinos consistently reported lower levels of access to care than whites.

Reduced access to care can result in delayed diagnosis and treatment and contribute to well-documented disparities in minority health.2 Many of the chronic diseases that contribute to racial and ethnic health disparities require early detection and monitoring. Having a regular health care provider who knows patients’ individual history and health care needs, along with periodic contact with a physician, can help build trust and rapport between caregivers and patients. Seeing physicians in hospital emergency departments for nonurgent care contributes to problems with continuity and coordination of care. Finally, disparities in access to specialist care can create additional problems for patients with complex conditions.

Back to Top



Access to Care

onelderly whites continue to be more likely to have a regular health care provider than either blacks or Latinos. Between 1997 and 2003, less than two-thirds of blacks and somewhat more than half of Latinos reported having a regular provider, compared with three-quarters of whites (see Table 5). Latinos and blacks also are less likely than whites to have seen a doctor in the past 12 months. With the exception of a slight increase in 1999, less than three in four nonelderly blacks and slightly more than three in five Latinos saw a doctor, compared with four in five whites.

Blacks’ use of emergency rooms to obtain care continued to decline, from 10.4 percent of all doctor visits in 1997 to 9.1 percent in 2003. However, blacks continue to use emergency rooms in greater proportions than whites. In 1997 and 1999, Latinos tended to use emergency rooms for doctor visits in similar proportions as whites. However, between 2001 and 2003, Latinos made significantly more of their health care provider visits in emergency rooms than whites.

Nonelderly whites’ access to specialists remained virtually unchanged between 1997 and 2003, with about 27 percent of whites’ most recent doctor visits occurring with specialists. Blacks and Latinos were much less likely to have the same level of access to specialists as whites. In fact, blacks access to specialists declined significantly. Between 2001 and 2003, the percentage of blacks whose last doctor visit was to a specialist dropped from 24.4 percent to 19.8 percent.

Gaps in access to care also persisted among insured and uninsured blacks, Latinos and whites (see Supplementary Tables 2 and 3). For example, 80.1 percent of insured blacks in 2003 reported seeing a doctor in the past year, compared with 48 percent of uninsured blacks. Similarly, 74.8 percent of insured Latinos and 82.6 percent of insured whites in 2003 saw a doctor, compared with 38.5 percent of uninsured Latinos and 50.3 percent of uninsured whites.

Back to Top



Table 5
Access to Medical Care Among Nonelderly African Americans, Latinos and Whites, 1997-2003

 
1997
1999
2001
2003
Regular Health Care Provider
Black
63.9%
65.5%
64.4%
65.8%
Latino
59.6
56.2*
55.4
55.4#
White
74.8
73.5*
75.2*
76.3#
Doctor Visit in the Last 12 Months
Black
74.6
77.4*
74.1*
73.8
Latino
62.0
65.5*
62.2*
62.8
White
77.6
78.4*
79.1
79.0#
Proportion of Doctor Visits in the Emergency Room
Black
10.4
10.7
9.6*
9.1#
Latino
7.4
6.8
7.8
8.4
White
6.8
6.8
6.6
6.1#
Last Doctor Visit with a Specialist
Black
26.0
23.4*
24.4
19.8*#
Latino
23.2
25.1
23.3
24.3
White
27.5
27.7
27.7
27.5
* Change from previous survey is statistically significant at p <.05.

# Change from 1997 to 2003 is statistically significant at p <.05.

Note: Bold text shows statistically significant differences from whites.

Source: Community Tracking Study Household Survey



Implications

hile overall health insurance rates changed little among nonelderly blacks, Latinos and whites between 2001 and 2003, sources of coverage shifted—especially for Latinos—from employment-based insurance to public coverage.

In 2003, about 23 percent of both Latinos and blacks relied on public coverage—a far different situation than in 1997, when 13.7 percent of Latinos and 19.3 percent of blacks had public coverage. Increased reliance on public coverage can be viewed either as an encouraging development—a result of expanded eligibility and increased outreach—or a worrisome one—minorities disproportionately losing employment coverage as a result of job market changes. Either way, as states wrestle with tight budgets, Medicaid and other state coverage programs are particularly vulnerable to budget cuts, leaving Latinos and blacks at risk for losing coverage.

Additionally, blacks and Latinos are more likely than whites to be disconnected from the health care system. For example, they are less likely than whites to have a regular caregiver, less likely to have seen a physician and more likely to see physicians in emergency rooms. When they visit physicians, blacks and Latinos are less likely than whites to see a specialist. The decline in access to specialist care among blacks is particularly troubling.

The access gap has changed little since 1997 and in some cases has increased. As long as access problems for racial and ethnic minorities persist, it is unlikely that health disparities will diminish significantly.

Back to Top



Notes

1. Public coverage includes Medicaid, SCHIP and other state coverage. Private coverage includes nongroup insurance and private insurance obtained through someone outside the family. Persons with military coverage and people enrolled in Medicare because of disability were excluded from this analysis.
2. Institute of Medicine, Coverage Matters: Insurance and Health Care, National Academy Press, Washington, D.C. (2001).
Back to Top



Data Source

This Tracking Report presents findings from the This Tracking Report presents findings from the HSC Community Tracking Study (CTS) Household Survey, a nationally representative telephone survey of the civilian, noninstitutionalized population conducted in 1996-97, 1998-99, 2000-01 and 2003. For discussion and presentation, we refer to a single calendar year for the first three surveys (1997, 1999 and 2001). Data were supplemented by in-person interviews of households without telephones to ensure proper representation. The first three rounds of the survey contain information on about 60,000 people, while the 2003 survey contains responses from about 47,000 people. Response rates ranged from 60 percent to 65 percent for the first three rounds and were 57 percent in 2003. The estimates in this report are representative of people under age 65 in three The estimates in this report are representative of people under age 65 in three racial or ethnic groups. Black refers to all non-Latino blacks, and white refers to all non-Latino whites. Insurance status reflects coverage on the day of the interview.

Back to Top



Supplementary Tables

Supplementary Table 1: Insurance Coverage Among Nonelderly Blacks, Latinos and Whites, 2001-2003

Supplementary Table 2: Access to Medical Care Among Insured Nonelderly Blacks, Latinos and Whites, 1997-2003

Supplementary Table 3: Access to Medical Care Among Uninsured Nonelderly Blacks, Latinos and Whites, 1997-2003




ISSUE BRIEFS are published by the Center for Studying Health System Change.

600 Maryland Avenue, SW, Suite 550
Washington, DC 20024-2512
Tel: (202) 484-5261
Fax: (202) 484-9258
www.hschange.org

President: Paul B. Ginsburg
Vice President: Len M. Nichols