 |
|
Health Insurance
|
|
|
|
Health Insurance Coverage: 2001
Reversing two years of falling uninsured rates, the share of the
population without health insurance rose in 2001. An estimated
14.6 percent of the population or 41.2 million people were
without health insurance coverage during the entire year in 2001,
up from 14.2 percent in 2000, an increase of 1.4 million people.
The estimates in this report are based on the 2002 Current
Population Survey (CPS) Annual Demographic Supplement, conducted
by the U.S. Census Bureau. Respondents provide answers to the
best of their ability, but as with all surveys, the estimates may
differ from the actual values. A copy of the CPS Supplement
questionnaire is available electronically at http://www.census.gov/apsd/techdoc/
cps/cps-main.html.
New Population Controls and Expanded Sample
The estimates in this report are based on the 2000, 2001, and 2002
Current Population Survey Annual Demographic Supplements (CPS ADS)
and provide information for calendar years 1999, 2000, and 2001,
respectively. These estimates use population estimates based on
Census 2000. Earlier reports presenting data for calendar years
1993 through 2000 used population estimates based on the 1990 census.
In 2001, the Census Bureau tested a sample expansion of 28,000
households to the CPS ADS. The sample expansion was officially
implemented in the estimates presented here. It is primarily designed
to improve the reliability of state estimates of children's health
insurance coverage, but the larger sample size also improves the
reliability of national estimates of other topics.
Because results presented in this report from the 2001 survey have
been recalculated based on the expanded sample and the Census
2000-based weights, they may differ slightly from earlier estimates
that did not incorporate the sample expansion and were based on the
1990 census. Appendix B presents more detail on the introduction of
the sample expansion and new population controls based on Census 2000.
All statements in this report have undergone statistical testing, and
all comparisons are significant at the 90 percent confidence level.
Further information on the source and accuracy of the estimates is at
www.bls.census.gov/cps/ads/2002/ssrcacc.htm.
Highlights: /1
The number and percentage of people covered by employment-
based health insurance dropped in 2001, from 63.6 percent to
62.6 percent, the foundation of the overall decrease in
health insurance coverage.
The number and percentage of people covered by government
health insurance programs rose in 2001, from 24.7 percent to
25.3 percent, largely from an increase in the number and
percentage of people covered by medicaid (from 10.6 percent
to 11.2 percent).
The proportion of uninsured children did not change,
remaining at 8.5 million in 2001, or 11.7 percent of all
children.
Although medicaid insured 13.3 million poor people, 10.1
million poor people still had no health insurance in 2001,
representing 30.7 percent of the poor, unchanged from 2000.
Hispanics (66.8 percent) were less likely than non-Hispanic
Whites (90.0 percent) to be covered by health insurance. /2
The coverage rate for Blacks in 2001 (81.0 percent) did not
differ from the coverage rate for Asians and Pacific
Islanders (81.8 percent).
American Indians and Alaska Natives were less likely to have
health insurance than other racial groups, based on 3-year
averages (1999-2001) - 72.9 percent, compared with 80.8
percent of Blacks, 81.5 percent of Asians and Pacific
Islanders, and 90.2 percent of non-Hispanic Whites.
However, American Indians and Alaska Natives were more
likely to have insurance than were Hispanics (67.0 percent)
Among the entire population 18 to 64 years old, workers
(both full- and part-time) were more likely to have health
insurance (83.0 percent) than nonworkers (75.3 percent), but
among the poor, workers were less likely to be covered (51.3
percent) than nonworkers (63.2 percent).
Compared with 2000, the proportion who had employment-based
policies in their own name fell for workers employed by
firms with fewer than 25 employees, but was unchanged for
those employed by larger firms.
Young adults (18 to 24 years old) were less likely than
other age groups to have health insurance coverage -- 71.9
percent in 2001, compared with 83.3 percent of those 25 to
64 and, reflecting widespread medicare coverage, 99.2
percent of those 65 years and over.
More people did not have health insurance in 2001.
The number of people without health insurance coverage rose to
41.2 million (14.6 percent of the population) in 2001, up 1.4
million from the previous year, when 14.2 percent of the
population lacked coverage (see Table 1). Interestingly, the
number of people covered by health insurance also increased in
2001, up 1.2 million to 240.9 million (85.4 percent of the
population). Both increases can be attributed in part to an
overall population growth from 2000 to 2001.
Most people (64.1 percent) were covered by a health insurance
plan related to employment for some or all of 2000, an increase
of 0.6 percentage points over the previous year. The increase in
private health insurance coverage largely reflects the increase
in employment-based insurance, which increased 0.3 percentage
points to 72.4 percent in 2000.
A decline in employment-based insurance prompted the decrease in
insurance coverage rates. /3
Most people (62.6 percent) were covered by a health insurance
plan related to employment for some or all of 2001, a decrease of
1.0 percentage point from the previous year. The 1.1 percentage
point decline in private health insurance coverage, to 70.9
percent in 2001, largely reflects the decrease in employment-
based insurance.
Although it did not offset the overall decline, health insurance
coverage provided by the government increased between 2000 and
2001. This increase largely reflects the increase in medicaid
coverage, which rose by 0.6 percentage points to 11.2 percent in
2001. /4 Among the entire population, 25.3 percent had government
insurance, including medicare (13.5 percent), medicaid (11.2
percent), and military health care (3.4 percent). Many people
carried coverage from more than one plan during the year; for
example, 7.6 percent of people were covered by both private
health insurance and medicare.
The uninsured rates for the poor and the near poor did not change
between 2000 and 2001.
Despite the medicaid program, 10.1 million poor people, or 30.7
percent of the poor, had no health insurance of any kind during
2001. This percentage -- more than double the rate for the total
population -- did not change significantly from the previous
year. The uninsured poor comprised 24.5 percent of all uninsured
people (see Table 2).
Medicaid was the most widespread type of health insurance among
the poor, with 40.5 percent (13.3 million) of those in poverty
covered by medicaid for some or all of 2001. This percentage did
not change from the previous year. /5
Among the near poor (those with a family income greater than or
equal to, but less than 125 percent of, the poverty level), 26.5
percent (3.3 million people) lacked health insurance in 2001,
unchanged from 2000. Although private health insurance coverage
among the near poor declined in 2001 -- from 40.3 percent to 37.8
percent -- their rate of government health insurance coverage did
not change from 2000 (it was 47.1 percent in 2001).
Key demographic factors affect health insurance coverage.
Age - People 18 to 24 years old were less likely than other age
groups to have health insurance coverage, with 71.9 percent
covered for some or all of 2001. Because of medicare, almost all
people 65 years and over (99.2 percent) had health insurance in
2001. For other age groups, health insurance coverage ranged
from 76.6 percent to 88.3 percent.
Among the poor, people 18 to 64 years old had a markedly lower
health insurance coverage rate (57.7 percent) in 2001 than either
people under 18 (78.7 percent) or 65 years and over (97.3
percent). /6
Race and Hispanic origin - While the uninsured rate rose in 2001
for non-Hispanic Whites - from 9.6 percent to 10.0 percent -
the uninsured rates among Blacks (19.0 percent) and among Asian
and Pacific Islanders (18.2 percent) did not change from 2000. /7
The uninsured rate among Hispanics (33.2 percent in 2001) also
did not change from 2000 (see Table 1). /8
The CPS Annual Demographic Supplement, the source of these data,
obtained interviews from 78,000 households nationwide but is not
large enough to produce reliable annual estimates for American
Indians and Alaska Natives. However, Table 3 displays 3-year
averages of the number of American Indians and Alaska Natives,
their uninsured rate, and 3-year-average uninsured rates for other
race groups. The 3-year average (1999-2001) shows that 27.1
percent of American Indians and Alaska Natives were without
coverage, higher than the 19.2 percent for Blacks, 18.5 percent
for Asians and Pacific Islanders, and 9.8 percent for non-Hispanic
Whites. /9 However, the 3-year-average uninsured rate for Hispanics
(33.0 percent) was higher than the uninsured rate for American
Indians and Alaska Natives.
Comparisons of 2-year moving averages (1999-2000 and 2000-2001)
show that while the uninsured rate fell for American Indians and
Alaska Natives from 27.7 percent to 25.5 percent and for Blacks
from 19.3 percent to 18.9 percent, uninsured rates among non-
Hispanic Whites, Asian and Pacific Islanders, and Hispanics did
not change.
Nativity - In 2001, the proportion of the foreign-born population
without health insurance (33.4 percent) was more than double that
of the native population (12.2 percent). /10 Among the foreign
born, noncitizens were much more likely than naturalized citizens
to lack coverage -- 42.9 percent compared with 17.2 percent.
Educational attainment - Among all adults, the likelihood of
being insured increased as the level of education rose. Compared
with the previous year, coverage rates decreased for those with
no high school diploma, those who are high school graduates only,
and those with some college education but no degree. Coverage
rates did not change from 2000 to 2001 for adults with an
associate degree or higher.
Economic status affects health insurance coverage.
Income - The likelihood of being covered by health insurance
rises with income. Among households with annual incomes of less
than $25,000, the percentage with health insurance was 76.7
percent; the level rises to 92.3 percent for those with incomes
of $75,000 or more. Compared with the previous year, coverage
rates decreased at every level of household income.
Work experience - Of those 18 to 64 years old in 2001, full-time
workers were more likely to be covered by health insurance (84.0
percent) than part-time workers (78.0 percent), and part-time
workers were more likely to be insured than nonworkers (75.3
percent). /11 However, among the poor, nonworkers (63.2 percent)
were more likely to be insured than part-time workers (54.0
percent), who were more likely to be insured than full-time
workers (49.7 percent).
Firm size - Of the 142.6 million workers in the United States who
were 18-64 years old, 56.3 percent had employment-based health
insurance policies in their own name. The proportion increased
with the size of the employing firm from 31.3 percent for firms
with fewer than 25 employees to 69.6 percent for firms with 1000
or more employees. (These estimates do not reflect the fact that
some workers were covered by another family member's employment-
based policy). Compared with the previous year, the proportion
who had employment-based policies in their own name decreased for
workers employed by firms with fewer than 25 employees, but were
unchanged for those employed by larger firms.
The uninsured rate for children did not change between 2000 and
2001.
The percentage of children (people under 18 years old) without
health insurance did not change in 2001 (see Table 1), remaining
at 8.5 million or 11.7 percent. A decline in employment-based
health insurance coverage of children was offset by an increase
in coverage by medicaid or the State Children's Health Insurance
Program.
Among poor children, 21.3 percent (2.5 million children) had no
health insurance during 2001, unchanged from the previous year
(see Table 2). For this group, employment-based coverage
decreased from 20.1 percent to 18.6 percent, while government
health insurance coverage increased from 60.9 percent to 63.3
percent. Poor children made up 29.3 percent of all uninsured
children in 2001.
Among near-poor children (those in families whose income was
greater than or equal to, but less than 125 percent of, the
poverty level), 21.6 percent (0.9 million children) were without
health insurance in 2001, unchanged from 2000. /12 For this group,
private health insurance coverage decreased from 39.8 percent to
36.4 percent, but government health insurance coverage did not
change.
The likelihood of health insurance coverage varies among
children.
Children 12 to 17 years old were more likely to be uninsured
than those under 12 -- 13.1 percent compared with 11.0
percent.
The uninsured rate declined in 2001 for Hispanic children --
from 25.3 percent to 24.1 percent. The uninsured rates for
non-Hispanic White children (7.4 percent), Black children
(13.9 percent), and Asian and Pacific Islander children
(11.7 percent) were unchanged from 2000.
While most children (68.4 percent) were covered by an
employment-based or privately purchased health insurance
plan in 2001, nearly one in four (22.7 percent) was covered
by medicaid.
Black children had a higher rate of medicaid coverage in
2001 than children of any other racial or ethnic group -
38.3 percent, compared with 34.9 percent of Hispanic
children, 18.0 percent of Asian and Pacific Islander
children, and 15.3 percent of non-Hispanic White children.
Children living in single-parent families in 2001 were less
likely to be insured than children living in married-couple
families - 84.3 percent compared with 90.4 percent.
Some states had higher uninsured rates than others.
The proportion of people without health insurance ranged from 7.2
percent in Rhode Island to 23.2 percent in New Mexico, based on
3-year averages for 1999, 2000 and 2001 (see Table 4). Although
the data presented suggest that New Mexico had the highest
uninsured rate, its rate was not statistically different from the
rate for Texas. Similarly, although the data suggest that Rhode
Island had the lowest uninsured rate, its rate was not
statistically different from the rate for Minnesota.
Comparisons of 2-year moving averages (1999-2000 and 2000-2001)
show that the proportion of people without coverage fell in 14
states: Alaska, Arizona, Idaho, Louisiana, Massachusetts,
Montana, Nevada, New Mexico, North Dakota, South Carolina, South
Dakota, Virginia, West Virginia and Wisconsin. Meanwhile, the
proportion of people without coverage rose in nine states:
Arkansas, Georgia, Indiana, Missouri, Ohio, Oklahoma,
Pennsylvania, Rhode Island, and Texas.
Accuracy of the Estimates
Statistics from surveys are subject to sampling and nonsampling
error. All comparisons presented in this report take sampling
error into account and meet the Census Bureau's standards for
statistical significance. Nonsampling errors in surveys may be
attributed to a variety of sources, such as how the survey was
designed, how respondents interpret questions, how able and
willing respondents are to provide correct answers, and how
accurately answers are coded and classified. The Census Bureau
employs quality control procedures throughout the production
process including the overall design of surveys, the wording of
questions, review of the work of interviewers and coders, and
statistical review of reports.
The Current Population Survey weighting procedure uses ratio
estimation whereby sample estimates are adjusted to independent
estimates of the national population by age, race, sex, and
Hispanic origin. This weighting partially corrects for bias due
to undercoverage, but biases may still be present when people who
are missed by the survey differ from those interviewed in ways
other than age, race, sex, and Hispanic origin. How this
weighting procedure affects other variables in the survey is not
precisely known. All of these considerations affect comparisons
across different surveys or data sources.
For further information on statistical standards and the
computation and use of standard errors, contact Jeffrey Stratton
of the Demographic Statistical Methods Division on the Internet
at dsmd.source.and.accuracy@census.gov.
Technical Note
This report presents data on the health insurance coverage of
people in the United States during the 2001 calendar year. The
data, which are shown by state and selected demographic and
socioeconomic characteristics, were collected in the 2002 Annual
Demographic Supplement to the Current Population Survey (CPS).
Treatment of major federal health insurance programs
The Current Population Survey (CPS) underreports medicare and
medicaid coverage compared with enrollment and participation data
from the Centers for Medicare and Medicaid Services (CMS),
formerly the Health Care Financing Administration (HCFA). /13 A
major reason for the lower CPS estimates is that the CPS is not
designed primarily to collect health insurance data; instead, it
is largely a labor force survey. Consequently, interviewers
receive less training on health insurance concepts.
Additionally, many people may not be aware that they or their
children are covered by a health insurance program if they have
not used covered services recently and therefore fail to report
coverage. CMS data, on the other hand, represent the actual
number of people who enrolled or participated in these programs
and are a more accurate source of coverage levels.
Changes in medicaid coverage estimates from one year to the next
should be viewed with caution. Because many people who are
covered by medicaid do not report that coverage, the Census
Bureau assigns coverage to those who are generally regarded as
"categorically eligible" (those who received some other benefits,
usually public assistance payments, that make them eligible for
medicaid). Since the number of people receiving public
assistance has been dropping, the relationship between medicaid
and public assistance has changed, so that the imputation process
has introduced a downward bias in the most recent medicaid
estimates.
After consulting with health insurance experts, the Census Bureau
modified the definition of the population without health
insurance in the March 1998 Current Population Survey, which
collected data about coverage in 1997. Previously, people with
no coverage other than access to the Indian Health Service were
counted as part of the insured population. Beginning with the
1997 Health Insurance Coverage report, however, the Census Bureau
counted these people as uninsured. The effect of this change on
the overall estimates of health insurance coverage was
negligible.
Contact:
Robert J. Mills
301-763-3213
1. Confidence intervals for estimates are provided in Table 1.
These measures of sampling error should be used when comparing
estimates.
2. Because Hispanics may be of any race, data in this report for
Hispanics overlap slightly with data for the Black population and
the Asian and Pacific Islander population. About 10.9 percent of
White households, 3.0 percent of Black households, 2.0 percent of
Asian and Pacific Islander households, and 13.1 percent of American
Indian and Alaska Native households are maintained by a person of
Hispanic origin.
3. Employment-based health insurance is coverage offered through
one's own employment or a relative's.
4. Some of this increased medicaid coverage may be due to
expansions in the State Children's Health Insurance Program. For
further discussion, see Charles T. Nelson and Robert J. Mills; U.S.
Census Bureau; "The Characteristics of People Reporting State
Children's Health Insurance Program Coverage in the March 2001
Current Population Survey;" August 2002; at
www.census.gov/hhes/hlthins/asa02.pdf.
5. Changes in year-to-year medicaid estimates should be viewed with
caution. For more information, see the Technical Note on page 14.
6. The health insurance coverage rates of poor people 18-64 years
old and poor people 35-44 years old were not different.
7. Data users should exercise caution when interpreting aggregate
results for the Asian and Pacific Islander (API) population because
the API population consists of many distinct groups that differ in
socio-economic characteristics, culture, and recency of immigration.
In addition, the CPS does not use separate population controls for
weighting the API sample to national totals.
8. Data users should exercise caution when interpreting aggregate
results for the Hispanic population because this population consists
of many distinct groups that differ in socio-economic characteristics,
culture, and recency of immigration.
9. Data users should exercise caution when interpreting aggregate
data for American Indians and Alaska Natives (AIAN) because the AIAN
population consists of groups that differ in economic characteristics.
Data from the 1990 census show that economic characteristics of those
American Indians and Alaska Natives who live in American Indian and
Alaska Native areas differ from the characteristics of those who live
outside these areas. In addition, the CPS does not use separate
population controls for weighting the AIAN samples to national totals.
See Accuracy of Estimates on page 13 for a further discussion of CPS
estimation procedures.
10. Natives are people born in the United States, Puerto Rico, or an
outlying area of the United States, such as Guam or the U.S. Virgin
Islands, and people who were born in a foreign country but who had at
least one parent who was a U.S. citizen. All other people born
outside the United States are foreign born.
11. Workers were classified as part time if they worked fewer than 35
hours per week in the majority of the weeks they worked in 2001.
12. The health insurance coverage rates of poor children and near-poor
children were not different.
13. CMS is the federal agency primarily responsible for administering
the medicare and medicaid programs at the national level.
Appendix.
Sample Expansion and Introduction of Census 2000-Based Population
Controls
The 2001 Current Population Survey (CPS) served as a tool for testing
a sample expansion of the Annual Demographic Supplement and as a
bridge to introduce new Census 2000-based population controls. The
following section discusses the effects these methodological changes
had on health insurance coverage rates.
Sample Expansion
In 2001, the Census Bureau tested an expansion in the sample for the
CPS Annual Demographic Supplement. The original sample size of
approximately 50,000 interviewed households increased to approximately
78,000 households. The primary goal of the sample expansion was to
produce more reliable state estimates of the number of low-income
children without health insurance to use in the funding formula for
the State Children's Health Insurance Program (SCHIP), through reduced
variances. Although the SCHIP sample expansion was specifically
targeted toward producing better children's health insurance estimates
at the state level, other state estimates, as well as national
estimates, improved.
The effect of the sample expansion on major national and state
uninsured estimates in 2000 was minimal. Nationally, the difference
between the original and expanded samples in the estimated percentage
of people without health insurance is not statistically significant
(see Table B-1). There are also no significant differences between
original and expanded sample uninsured rates by age or by major race
and ethnic groups, except for Blacks, whose expanded sample rate
(18.9 percent) is significantly higher than the original sample rate
(18.4 percent).
For most states and the District of Columbia, the uninsured rates for
the expanded sample are not statistically different from the original
sample (see Table B-2). Although health insurance coverage estimates
were significantly different in 21 states, the differences were
moderate (no state had a change greater than 2 percentage points).
The uninsured rate increased in 7 states and decreased in 14 states,
with differences extending from a 1.8 percentage point increase for
Connecticut to a 2.0 percentage point decrease for Vermont.
Further information on the effects of the CPS ADS sample expansion on
health insurance estimates is available at
www.census.gov/cps/ads/adsmain.htm.
Introduction of Census 2000-Based Population Controls
The procedure used in developing estimates for the entire civilian
non-institutionalized population from the Current Population Survey
(CPS) involves the weighting of sample results to independent
estimates of the population by sex, age, race, and Hispanic/
non-Hispanic categories. These independent estimates are developed by
using civilian noninstitutional population counts from the decennial
censuses and projecting them forward to current years using data on
births, deaths, and net migration. Beginning with the 2002 CPS Annual
Demographic Supplement, the independent estimates used as control
totals for the CPS are based on civilian noninstitutional population
benchmarks established by Census 2000.
Table B-3 shows two sets of data for 2000 to show the effect of
introducing new population controls -- one using new Census 2000-based
population controls and the other using controls based on the 1990
census. Following is a brief discussion of the effects of the new
population controls on health insurance uninsured rates.
The effect of new population controls on major national uninsured
estimates in 2000 was minimal. Nationally, the difference between the
Census 2000-based and the Census 1990-based samples in the estimated
percentage of people without health insurance is not statistically
significant. Use of the new Census 2000-based controls raised the
uninsured rate for males by 0.3 percent, but the rate for females did
not change. While the new controls left the uninsured rate for most
of the major race and ethnic groups unchanged, the uninsured rate for
Whites rose by 0.2 percent and the uninsured rate for Hispanics
increased by 0.8 percent. Similarly, while the uninsured rate for most
age groups did not change, the percentage of uninsured people 18 to 24
or 25 to 34 years old each increased by 0.7 percent.
Go to Health Insurance 2001
Contact the Demographic Call Center Staff at 301-763-2422 or 1-866-758-1060 (toll free) or visit
ask.census.gov for further information on Health Insurance Data.
Source: U.S. Census Bureau, Housing and Household Economic Statistics Division
Last Revised: December 07, 2004
