Order Code 97-975 EPW
Updated August 30, 2004
CRS Report for Congress
Received through the CRS Web
Health Insurance Coverage of Children, 2003
Chris L. Peterson
Analyst in Social Legislation
Domestic Social Policy Division
Summary
In 2003, 9.1 million children (11.8%) went without health insurance. This was not
significantly different than in 2002. However, for the third year in a row, the number
of children covered by employment-based health insurance dropped but was more than
offset by the increase in public coverage. In 2003, 1.1 million fewer children had job-
based coverage, compared to 2002, but 1.8 million more children had public coverage.
In 2003, 62% of children had employment-based health insurance at any time during the
year, and 26% had publicly provided health insurance.
Only 8% of non-Hispanic white children were uninsured in 2003, compared to 22%
of Hispanic children. Children in poor or near-poor families were more likely to be
uninsured than those in higher-income families. Children whose parents worked in a
small firm were much more likely to be uninsured (21% in firms with less than 10
workers) than those whose parents worked in a large firm (6% in firms with 1,000 or
more workers). Of uninsured children, more than half (57%) live in a household with
a parent where at least one adult worked full time for the entire year.
This report will be updated annually each fall, when new data are released.
Health Insurance Coverage of Children Over Time
In 2003, the estimated number of uninsured children (under age 19, 9.1 million) did
not change significantly from 2002. From 1998 to 2000, the number of uninsured
children dropped significantly because the number covered through employment-based
health insurance and through public coverage each increased. From 2000 to 2003,
however, the percentage of children covered by employment-based health insurance
dropped significantly from 67% to 62% — a decrease of nearly 2.6 million children. That
decrease was offset by an increase in public coverage of nearly 4.3 million children. More
than one out of four children (26%) had publicly provided health insurance in 2003. In
spite of these changes in the sources of children’s health insurance, the percentage of
children without health insurance remains at its lowest level in more than a decade, as
shown in Table 1.
Congressional Research Service ˜ The Library of Congress

---

CRS-2
Table 1. Health Insurance Coverage of Children, 1991 to 2003
Children <19
Employment
Other
Medicaid/
No Health
(thousands)
Based
Private HI
Other Publica
Insurance

2003
77,607
62.3%
5.3%
26.1%
11.8%
2002
77,276
63.9%
5.3%
23.8%
12.0%
2001
76,559
64.9%
5.0%
22.6%
12.1%
2000
76,321
66.7%
4.9%
20.8%
12.3%
1999
76,303
66.0%
5.6%
20.1%
13.2%
Results Using pre-1999 Questionnaireb
Children <19
Privately
No Health
(thousands)
Insured
Medicaidc
Insurance

1999
76,303
68.6%
18.6%
14.4%
1998
76,004
67.7%
18.6%
15.6%
1997
75,491
67.0%
19.4%
15.4%
1996
74,899
67.4%
21.3%
15.1%
1995
74,766
67.1%
22.8%
14.0%
1994
73,973
66.9%
22.5%
14.4%
1993
73,182
67.6%
23.5%
14.0%
1992
70,376
69.4%
21.2%
12.8%
1991
69,367
70.0%
20.1%
13.0%
Source: CRS analysis of data from the 1992-2004 Current Population Surveys.
Note: Children may be covered by more than one type of insurance; thus, percentages may total to more
than 100. For 1999 onward, population weights were based on the 2000 census.
a. Includes the State Children’s Health Insurance Program (SCHIP), other state programs for low-income
individuals, and Medicare. Excludes military and veterans coverage.
b. The numbers from the new questionnaire are more reliable but go back no earlier than 1999. For more
information on the changes in the CPS questionnaire, see CRS Report RL31275, Health Insurance:
Federal Data Sources for Analyses of the Uninsured, by Chris L. Peterson and Christine Devere.
“Employment based” and “other private health insurance” are grouped into “privately insured”
because changes to the Current Population Survey (CPS) beginning with the March 1995 survey
resulted in a greater proportion of private insurance being classified as employment based.
c. Includes the State Children’s Health Insurance Program (SCHIP). Excludes Medicare, military and
veterans coverage. The percentage covered by Medicaid using the pre-1999 questionnaire does not
include other state programs for low-income individuals.
Health Insurance Coverage and
Demographic Characteristics
As shown in Table 2, children ages 13 to 18 were more likely to be covered by
employment-based coverage in 2003 than younger children. Yet those teenage children
were more likely to be uninsured because rates of coverage for them through Medicaid
and other public programs were much lower than for children 12 and under. Black and
Hispanic children were less likely to have employment-based coverage and were more
likely to be without health insurance, compared to non-Hispanic white children.
However, black and Hispanic children were more than twice as likely as non-Hispanic
white children to receive coverage through Medicaid or some other public program.
Children in the South and the West were more likely to be uninsured than children in the
Northeast or the Midwest.

---

CRS-3
Table 2. Health Insurance Coverage and Demographic
Characteristics of Children under 19, 2003
Population Employment
Other
Medicaid/
CHAMPUS
(thousands)
Based
Private HI Other Publica
or VA
Uninsured
Age
Under 6
23,834
58.7%
4.8%
31.9%
2.9%
10.3%
6 to 12
28,243
62.8%
5.2%
26.3%
2.6%
11.0%
13 to 18
25,531
65.1%
5.9%
20.4%
2.8%
14.0%
Race/ethnicity
White
46,044
73.6%
6.6%
17.4%
3.1%
7.7%
Black
11,570
46.2%
2.9%
43.1%
2.8%
14.9%
Hispanic
14,518
40.0%
2.8%
39.9%
1.4%
21.9%
Asian
2,854
66.2%
7.6%
19.2%
2.1%
12.7%
Other
2,620
52.9%
4.7%
34.8%
4.5%
12.6%
Region
Northeast
13,734
68.5%
4.1%
24.2%
1.4%
9.4%
Midwest
17,339
70.4%
5.3%
22.6%
2.0%
8.3%
South
27,995
57.3%
5.2%
28.1%
3.6%
14.3%
West
18,539
57.6%
6.5%
27.7%
3.2%
13.0%
Total
77,607
62.3%
5.3%
26.1%
2.7%
11.8%
Table 3. Health Insurance Coverage and Family Characteristics
of Children under 19, 2003
Population Employment
Other
Medicaid/
CHAMPUS
(thousands)
Based
Private HI Other Publica
or VA
Uninsured
Child not with parent
3,755
25.4%
1.4%
41.0%
1.4%
34.9%
Child with parent
73,852
64.2%
5.5%
25.3%
2.8%
10.6%
Family type
Two-parentb
52,513
72.6%
6.2%
17.2%
3.3%
8.9%
Single dad
4,077
48.8%
6.0%
32.7%
2.1%
18.9%
Single mom
17,262
41.9%
3.1%
48.4%
1.5%
13.8%
Poverty status
Under 100%
12,246
17.1%
3.0%
67.1%
1.8%
18.3%
100% to 149%
7,923
37.9%
3.7%
48.7%
2.4%
18.0%
150% to 199%
7,707
54.5%
5.1%
31.5%
3.2%
16.9%
200% to 299%
13,592
71.9%
6.2%
18.5%
3.2%
9.7%
300% and higher
32,383
87.4%
6.7%
5.1%
3.0%
4.7%
Parent’s insurance
Employment based
50,316
91.8%
3.0%
11.1%
2.4%
2.3%
Other private HI
2,951
4.6%
86.2%
15.6%
1.8%
2.1%
Medicaid/other pub.
7,802
3.3%
0.1%
96.7%
1.4%
2.7%
CHAMPUS or VA
697
2.6%
0.2%
7.7%
98.9%
0.3%
Uninsured
12,087
6.4%
0.2%
41.8%
0.2%
52.7%
Source: Tables are from CRS analysis of data from the March 2004 Current Population Survey.
Note: Children may have more than one type of insurance, so percentages can sum to more than 100%.
a. Includes the State Children’s Health Insurance Program (SCHIP), other state programs for low-income
individuals, and Medicare. Excludes military and veterans coverage.
b. Includes stepparent.

---

CRS-4
Insurance Coverage and Family Characteristics
Children’s insurance coverage also differs by family structure. As shown in Table
3, 35% of children not living with a parent lacked health insurance, compared to 11% of
children living with at least one parent. Among children living with a parent, family
structure still had an impact on health insurance coverage. Less than 10% of children in
a two-parent family were uninsured in 2003. Compared to children living with their
single mothers, children living with their single fathers were more likely to have
employment-based health insurance but were less likely to have insurance overall because
they were much less likely to be enrolled in Medicaid or other public coverage.
Among children in poverty,1 17% had employment-based coverage, while two-thirds
were covered by Medicaid or other public coverage; 18% were uninsured. As the income-
to-poverty ratio increases, so does the likelihood of children having employment-based
coverage. Among children in families with income three or more times the poverty level,
87% had job-based coverage; 5% were uninsured.
A child’s source of health insurance is strongly affected by the insurance coverage
of his or her parents. Approximately 92% of children who had one or more parents with
employment-based coverage in 2003 also had employment-based coverage.2 Likewise,
about 97% of children in families in which the parent(s) had Medicaid or other public
coverage were also in Medicaid or other public coverage. In 2003, of children in families
in which the parent(s) was uninsured, 53% had no insurance coverage and 42% were
covered by Medicaid or some other public program.
Insurance Coverage and Parents’ Employment
In 2003, of children in families where at least one parent worked year-round and full-
time, 74% had job-based coverage, 16% were enrolled in Medicaid or other public
coverage, while about 9% were uninsured, as shown in Table 4. In families where at least
one parent worked but only part-year or part-time, 37% of children had job-based
coverage, while half (51%) were covered by public coverage, and 15% were uninsured.
Where children lived with at least one parent but no parent was working, 71% had public
coverage, and 19% were uninsured.
Employment-based health insurance coverage is less common for workers in small
firms than in larger ones. Rates of job-based coverage were lowest and lack of health
insurance coverage was highest among children in families where the primary worker was
employed by a firm with fewer than 10 employees. In four industry groups, less than half
of workers’ dependent children had employment-based coverage. However, less than
10% of workers’ dependent children were uninsured in several other sectors.
1 Among children living with at least one parent. In 2003, the poverty threshold for a family with
two adults and two children was $18,660.
2 When a parent had more than one source of coverage, the following hierarchy was used to
determine “primary” coverage: employment based, private, Medicaid/Medicare, CHAMPUS or
VA, and other public. Then the parent with the “highest” coverage was used to classify both
parents’ insurance coverage. Thus, if one parent had employment-based coverage and the other
had private insurance, the parents’ coverage was classified as employment based.

---

CRS-5
Table 4. Health Insurance Coverage and Parents’ Employment,
Among Children Living with at Least 1 Parent, 2003
Other
Medicaid/
Population Employment Private
Other
CHAMPUS
(thousands)
Based
HI
Publica
or VA
Uninsured
Custodial parents’ work status
At least 1 worked
57,288
74.4%
5.6%
16.0%
2.9%
9.0%
year-round, full-time
None full time; at least 1
11,199
36.9%
6.1%
51.2%
2.4%
14.8%
part-year or part-time
Did not work
5,364
11.3%b
2.9%
71.2%
2.2%
18.7%
Firm size
Under 10 workers
11,141
35.7%
14.7%
33.3%
1.4%
21.0%
10 to 24
5,819
50.2%
7.3%
29.0%
1.3%
18.5%
25 to 99
8,542
62.1%
5.2%
26.9%
1.1%
12.0%
100 to 499
9,202
75.2%
3.3%
20.4%
1.5%
7.5%
500 to 999
3,490
75.3%
3.7%
19.6%
1.4%
7.1%
1,000 or more
26,561
80.6%
2.8%
15.5%
4.9%
5.5%
Not applicablec
9,096
46.2%
4.2%
47.6%
3.1%
11.0%
Industry
Agriculture, forestry,
1,153
24.8%
12.3%
37.3%
1.1%
28.7%
fishing and hunting
Construction
5,939
49.4%
7.5%
28.3%
1.3%
20.6%
Arts, entertainment,
recreation, food
3,898
36.6%
5.1%
43.5%
1.0%
18.9%
services, accomm.
Other services
2,611
41.3%
6.9%
39.7%
1.4%
17.7%
Mining
408
75.7%
5.3%
11.5%
1.3%
12.2%
Wholesale, retail trade
8,238
63.4%
6.8%
25.1%
1.5%
11.1%
Professional, manager,
6,263
64.1%
9.2%
21.4%
1.4%
10.5%
administrative services
Transportation, utilities
3,692
75.5%
3.8%
18.2%
2.1%
9.1%
Finance, insurance, real
4,630
76.8%
8.1%
11.9%
1.2%
8.7%
estate, rental, leasing
Educational, health, and
12,242
72.4%
4.7%
21.7%
2.0%
7.4%
social services
Manufacturing
9,829
80.0%
3.1%
16.6%
1.0%
6.3%
Information
1,772
85.0%
4.0%
10.2%
1.3%
5.0%
Public administration
3,379
89.8%
2.0%
10.1%
6.2%
2.9%
Armed forces, military
702
42.1%
2.2%
5.1%
100.0%
0.0%
Not applicablec
9,096
46.2%
4.2%
47.6%
3.1%
11.0%
Total
73,852
64.2%
5.5%
25.3%
2.8%
10.6%
Source: CRS analysis of data from the March 2004 Current Population Survey.
Note: Children may have more than one type of insurance, so percentages can sum to more than 100%.
a. Includes the State Children’s Health Insurance Program (SCHIP), other state programs for low-income
individuals, and Medicare. Excludes military and veterans coverage.
b. Child’s employment-based coverage may be through a parent’s former employer, from someone outside
the household (e.g., noncustodial parent), or from coverage in the child’s name (e.g., the child’s own
job).
c. No firm-size or industry information is provided because the parent did not work (5.5 million), or
coverage is from outside the household (3.2 million) or in the child’s own name (0.4 million).

---

CRS-6
Characteristics of Uninsured Children
In the preceding discussion, children were grouped according to various
characteristics, with their rates of health insurance coverage compared. For example, in
2003, 8% of non-Hispanic white children were uninsured, compared to 22% of Hispanics.
However, because the United States has many more non-Hispanic white children (46
million) than Hispanic children (15 million), the population of uninsured children
actually consists of more non-Hispanic white children than Hispanic children, as shown
in Figure 1. This apparent paradox — that the group least likely to be uninsured makes
up the greatest portion of the uninsured — exists when looking at other characteristics
as well. The children of full-time, full-year workers are least likely to be uninsured (9%),
compared to the other workforce-attachment categories, yet comprise 57% of the
population of uninsured children. Children in two-parent families are least likely to be
uninsured (9%), compared to the other family types, yet make up more than half of the
population of uninsured children. This makes it more difficult for those developing
proposals to help uninsured children. For example, should proposals be targeted at
uninsured children in two-parent families because they are more numerous, or at other
uninsured children because they are more likely to be uninsured?
Figure 1. Characteristics of Uninsured Children Under Age 19, 2003
(9.1 million)
Source: CRS analysis of data from the 2004 Current Population Survey.
Note: Totals may not sum to 100% due to rounding. “Full time, full year” means at least one parent living
with the child was a full-time, full-year worker. “Part time or part year” means that at least one parent
living with the child worked but not full time and full year. Hispanics may be of any race. In this chart,
whites, blacks and Asians are non-Hispanic and report only one race. Among non-Hispanics, individuals
who report any other single race (e.g., American Indian) or multiple races are categorized as “other.”

---