Skip Navigation  U.S. Department of Health and Human Services  www.hhs.gov
Agency for Healthcare Research Quality www.ahrq.gov
www.ahrq.gov
MEPS Home Medical Expenditure Panel Survey
Font Size:
Contact MEPS FAQ Espanol Site Map
 
S
M
L
XL
 

Research Findings #8: Health Insurance Status of the Civilian Noninstitutionalized Population, 1997

Jessica P. Vistnes, Ph.D., and Samuel H. Zuvekas, Ph.D., Agency for Health Care Policy and Research.


Abstract

This report from the 1997 Medical Expenditure Panel Survey (MEPS) provides preliminary estimates of the health insurance status of the civilian noninstitutionalized U.S. population during the first half of 1997, including the size and characteristics of the population with private health insurance, with public insurance, and without any health care coverage. During this period, 83.2 percent of all Americans were covered by private or public health insurance, leaving 16.8 percent of the population, some 44.6 million persons, uninsured. Among the non-elderly population, 81.1 percent of Americans had either private or public coverage and 18.9 percent of the population (44.2 million persons) lacked health care coverage. The probability that an individual would be uninsured during this period was especially high for young adults aged 19-24 and members of racial and ethnic minorities (especially Hispanic males). Public health insurance continues to play an important role in ensuring that children, black Americans, and Hispanic Americans obtain health care coverage.

^top


Introduction

This report is the second in a series of yearly reports on the health insurance status of the U.S. population. The first report (Vistnes and Monheit, 1997) presented health insurance estimates for the first half of 1996.

The health insurance status of the U.S. population, especially the size and composition of the uninsured population, has become an issue of perennial public policy concern for several reasons.

First, health insurance is viewed as essential to ensure that individuals obtain timely access to medical care and protection against the risk of expensive and unanticipated medical events. Compared to people without health care coverage, insured individuals are more likely to have a usual source of medical care, to spend less out of pocket on health services, and to experience different treatment patterns, quality, and continuity in their health care (Lefkowitz and Monheit, 1991; U.S. Congress, Office of Technology Assessment, 1992).

Second, concern over the population’s health insurance status reflects a variety of equity and efficiency considerations. These include the magnitude and appropriate mix of private and public sector responsibility for financing health care, the impact of health insurance on the efficient use of health care, and the manner in which health insurance affects the distribution of health care among the general population and across groups of specific policy interest.

Third, timely and reliable estimates of the population’s health insurance status are essential to evaluate the costs and expected impact of public policy interventions to expand coverage or to alter the manner in which private and public insurance is financed. Identification of how individual and household demographic characteristics, health status, and economic circumstances are associated with the population’s health insurance status is of critical importance in developing efficient and targeted policy interventions. This is especially relevant given the current emphasis on incremental health care reform that is focused on particular health care markets and population groups.

Finally, comparisons of the characteristics of insured and uninsured populations over time provide information on whether greater equity has been achieved in the ability of specific population groups to obtain health insurance or whether serious gaps remain. In this regard, estimates of the population’s health insurance status from the Medical Expenditure Panel Survey (MEPS), which is conducted annually, provide critical data for evaluating the health insurance implications of recent legislative initiatives: the 1996 Health Insurance Portability and Accountability Act (HIPAA), Public Law 104-191; welfare reform under the 1996 Personal Responsibility and Work Opportunity Reconciliation Act, Public Law 104-193; and the 1997 State Children’s Health Insurance Program (CHIP). A primary goal of HIPAA is to reduce the impact of preexisting health conditions on the continuity of health insurance during employment transitions. Under welfare reform, mandated work requirements and time limitations governing the receipt of public assistance may have consequences for a recipient’s health insurance status. The goal of the CHIP program is to provide health insurance coverage to low-income children who are not eligible for Medicaid.

This report presents preliminary estimates of the number and characteristics of people with private and public health insurance at any time during the first half of 1997, on average. Particular emphasis is directed toward estimating the size of the population that was uninsured throughout the first half of 1997 and identifying groups especially at risk of lacking health insurance.  

^top


Overview

During the first half of 1997, on average, 83.2 percent of all Americans in the civilian noninstitutionalized population had some type of private or public health insurance coverage (Table 1). Roughly 68 percent of Americans obtained health insurance from private sources.Another 15.1 percent obtained public sources of coverage, primarily from the Medicare and Medicaid programs. The remaining 16.8 percent of Americans, 44.6 million people, were without health insurance throughout the first half of 1997. Among the non-elderly population, 69.2 percent were covered by private insurance and 11.9 percent by public insurance. Almost a fifth of the non-elderly population (18.9 percent), an estimated 44.2 million people, were uninsured. Table 2 gives more detailed information on the health insurance status of the non-elderly population. Overall, these health insurance estimates do not differ significantly from the 1996 MEPS figures reported in Vistnes and Monheit, 1997.

The data in Tables 1-3 provide estimates of the population’s health insurance status according to selected demographic characteristics, perceived health status, employment status, and residential location. Table 4 provides estimates of the distribution of the uninsured population by selected characteristics. Table A in the Technical Appendix provides estimates of the number of people by health insurance status. Some of the key findings and relationships revealed by these data are discussed below.

Age

MEPS data reveal that, in general, children are more likely than non-elderly adults to have health insurance coverage. The main findings among age groups are described below.

Children

There has been considerable interest in the health insurance status of children. This interest stems from the role health care coverage plays in ensuring that children obtain the medical care appropriate to their specific stage of development. To help ensure such coverage, Congress passed the State Children’s Health Insurance Program in 1997, allocating approximately $24 billion over 5 years to provide health insurance coverage to low-income children who are not eligible for Medicaid. The CHIP program follows on Medicaid expansions beginning in the late 1980s that focused attention on the role of the public and private sectors in financing health care for low-income children.MEPS data indicate that public health insurance covered a substantial proportion of children in the first half of 1997: Nearly 30 percent of children under age 4, one in four children ages 4-6, and close to one in five children ages 7-12 had public coverage, primarily through Medicaid. As a result, children under age 18 were less likely to be uninsured than were non-elderly adults in general. Despite this finding, nearly 11 million children lacked health care coverage.

Adults

Young adults ages 19-24 were the age group most likely to lack health insurance. Over a third of young adults (34.6 percent) were uninsured, twice the rate at which all Americans lacked coverage. Young adults ages 19-24 also had the lowest rate of private health insurance coverage among the non-elderly adult population.On the other hand, 6 out of 10 elderly Americans (60.5 percent) were covered by private health insurance. Nearly 4 out of 10 elderly Americans (38.4 percent) held only public coverage (Medicare alone or in conjunction with Medicaid).

Employment Status

Since most private health insurance in the United States is provided through the workplace, employment status is an important indicator of access to private health insurance. MEPS data reveal the following for the non-elderly population (Table 2):

  • Over three-quarters (78.7 percent) of workers were covered by private health insurance, compared to half (50.3 percent) of individuals who were not employed. 
  • People who were not employed were more likely than those who were employed to be covered by public insurance (23.9 and 3.2 percent, respectively). Workers were less likely than people who were not employed to be uninsured (18.0 and 25.9 percent, respectively).

Race/Ethnicity

MEPS data indicate that significant disparities exist in the rate at which racial and ethnic minorities are covered by private and public health insurance compared to white Americans (Table 1). For example:

  • Less than half of all Hispanic Americans (45.4 percent) and half of black Americans (50.2 percent) were covered by private health insurance, compared to three-quarters of whites (75.2 percent). A third of Hispanics (32.9 percent) and over a fifth of blacks (21.4 percent) were uninsured. In contrast, 13.2 percent of white Americans were uninsured.
  • Among all racial/ethnic groups, Hispanic males were the most likely to be uninsured; 36.9 percent lacked coverage.

Hispanic and black Americans were more likely than white Americans to be covered by public health insurance (21.6 percent and 28.4 percent, respectively, compared to 11.6 percent).

Marital Status

Married individuals were more likely than others to have private health insurance (Table 1). Of those who were not married at the time of the survey:

  • Widowed people were the least likely to be uninsured (6.5 percent) because of their higher rate of coverage from public programs (39.4 percent).
  • More than one-quarter of Americans who never married were uninsured (27.4 percent).
  • Almost a third of Americans who were separated were uninsured (31.9 percent).
  • More than a fifth of all divorced persons (21.2 percent) were uninsured.

Residential Location

The type of health care coverage obtained by Americans and the likelihood of being uninsured also varied by region and whether they lived in a metropolitan statistical area (MSA). MEPS data show that: People living in the South and West were less likely than residents of other regions to have  private health insurance (64.2 percent and 65.2 percent in the South and West, respectively,  compared to 70.3 percent and 75.0 percent of residents in the Northeast and Midwest). Nearly  one out of five persons in the South and West were uninsured (19.6 percent and 18.9 percent,  respectively) compared to14.2 and 12.5 percent in the Northeast and Midwest, respectively.

  • People living outside MSAs were less likely than those living within MSAs to be covered by  private health insurance (63.3 percent vs. 69.3 percent). They also were more likely to be  uninsured (18.7 percent vs. 16.3 percent).

Health Status

There is considerable public policy interest in determining whether people with health problems are able to obtain health insurance and, if so, the source of such coverage. MEPS respondents were asked to rate their health and family members’ health as excellent, very good,good, fair, or poor. The data in Table 3 reveal the relationships described below between health status and insurance coverage.

Non-Elderly Persons

More than one in five non-elderly Americans in good health (22.9 percent), fair health (23.4 percent), or poor health (21.2 percent) were uninsured throughout the first half of 1997. Among the non-elderly:

  • People in fair or poor health were less likely than those in better health to have private health insurance. Only 39.3 percent of those in poor health and 54.2 percent of those in fair health had any private coverage.
  • Public insurance helped to reduce the health-related disparities in private coverage. Over 20 percent of people in fair health and almost 40 percent of people in poor health had public coverage.

Elderly Persons

Elderly Americans in fair or poor health were less likely to have private coverage than those in better health. As a result, those in fair or

poor health were more likely to be covered by insurance from public sources only (43.5 and 55.7 percent, respectively) than other elderly Americans. Medicare, either alone or with Medicaid, was the main public source of coverage. 

^top


Characteristics of Uninsured Americans

Previous sections of this report have described the health insurance status of Americans by focusing on demographic, health status, and geographic characteristics associated with the likelihood that particular groups obtained private or public health insurance or were more at risk of being uninsured. To put this discussion in perspective, data displayed in Table 4 characterize the uninsured population by considering the representation of specific groups in the general population of non-elderly Americans relative to their representation among the uninsured population. In this way, one can assess whether certain population groups are disproportionately represented among the uninsured. Such information can be useful in formulating targeted policy interventions on behalf of people without health insurance.

Age

Young adults ages 19-24 composed 9.0 percent of the non-elderly population but 16.4 percent of the uninsured population. Among all age groups, young adults had the greatest risk of being uninsured.

Race/Ethnicity

Racial and ethnic minorities were more at risk of lacking health insurance than white Americans were. As a result, minority representation among the uninsured exceeded their representation among the general population. For example:

  • Although Hispanics represented only 12.0 percent of the non-elderly U.S. population, they accounted for 22.0 percent of the uninsured population.
  • Hispanic males represented only 6.2 percent of all non-elderly Americans but were the racial/ethnic group most likely to be uninsured, comprising 12.6 percent of the uninsured population.
  • Although 7 out of 10 non-elderly Americans were white, whites accounted for less than 6 out of 10 uninsured persons.
  • When the uninsured are categorized by race/ethnicity and sex, white males represent the largest proportion of the uninsured population.

Other Factors

People with specific residential locations and marital status were also disproportionately represented among the uninsured:

  • People living in the South represented about a third (34.8 percent) of all non-elderly Americans but 40.9 percent of all uninsured Americans.
  • People who never married accounted for over a fifth of the non-elderly population but over a third of the uninsured population. 

Finally, more than 1 out of 10 uninsured people (10.5 percent of the uninsured population) were in fair or poor health. These individuals are of particular policy concern because of the importance of health insurance in assuring timely access to needed health care services.

^top


Conclusions

Preliminary estimates from the 1997 MEPS reveal that, during the first half of 1997, 68.1 percent of Americans obtained health insurance from private sources, 15.1 percent obtained coverage through public programs, and 16.8 percent of the population (44.6 million people) lacked any health care coverage. Among the non-elderly population, nearly one person in five was uninsured.

The tabulations presented in this report indicate that the health insurance status of the U.S.  population is strongly associated with specific  demographic characteristics, health status, and employment status. Important disparities in health care coverage exist for particular groups. Among the groups especially at risk of lacking health care coverage are young adults ages 19-24 and members of racial and ethnic minorities (especially Hispanic males). Public health insurance continues to play an important role in insuring children, black Americans, and Hispanic Americans. Disparities in rates of insurance coverage also exist by health status, with non-elderly people in good or fair health more likely than people in better health to be uninsured. 

^top


Tables

1. Health insurance coverage and population characteristics - all ages 
2. Health insurance coverage and population characteristics - under age 65
3. Health insurance coverage and perceived health status - all ages
4. Population characteristics - total population and the uninsured, under age 65

Table 1: Health insurance coverage and population characteristics - all ages

Table 1. Health insurance coverage and population characteristics - all ages, continued

a Includes persons with unknown employment status and marital status.

b For individuals age 16 and over.

Note: Percents may not add to 100 because of rounding.

Source: Center for Financing, Access, and Cost Trends, Agency for Health Care Policy and Research: Medical Expenditure Panel Survey Household Component, 1997.

^top


Table 2: Health insurance coverage and population characteristics - under age 65

Table 2: Health insurance coverage and population characteristics - under age 65, continued

a Includes persons with unknown employment status and marital status.

b For individuals age 16 and over.

Note: Percents may not add to 100 because of rounding.

Source: Center for Financing, Access, and Cost Trends, Agency for Health Care Policy and Research: Medical Expenditure Panel Survey Household Component, 1997.

^top


Table 3: Health insurance coverage and perceived health status - all ages

a Includes persons with unknown perceived health status.

* Relative standard error is greater than or equal to 30 percent.

Note: Percents may not add to 100 because of rounding.

Source: Center for Financing, Access, and Cost Trends, Agency for Health Care Policy and Research: Medical Expenditure Panel Survey Household Component, 1997.

^top


Table 4: Population characteristics - total population and the uninsured, under age 65

Table 4: Population characteristics - total population and the uninsured, under age 65, continued

a Includes persons with unknown marital status and perceived health status.

b For individuals age 16 and over. Excludes unknown marital status. As a result, percents do not sum to 100.

Note: Percent distributions may not add to 100 because of rounding.

Source: Center for Financing, Access, and Cost Trends, Agency for Health Care Policy and Research: Medical Expenditure Panel Survey Household Component, 1997.

^top


References

Cohen JW, Monheit AC, Beauregard KM, et al. The Medical Expenditure Panel Survey: a national health information resource. Inquiry 1996;33:373-89.

Cohen S. Sample design of the 1996 Medical Expenditure Panel Survey Household Component. Rockville (MD): Agency for Health Care Policy and Research; 1997. MEPS Methodology Report No. 2. AHCPR Pub. No. 97-0027.

Lefkowitz D, Monheit AC. Health insurance, use of health services, and health care expenditures. Rockville (MD): Agency for Health Care Policy and Research; 1991. National Medical Expenditure Survey Research Findings 14. AHCPR Pub. No. 92-0017.

U.S. Congress, Office of Technology Assessment. Does health insurance make a difference? Background paper. Washington: U.S. Government Printing Office; 1992. Report No. OTA-BP-H-99.

Vistnes, JP, Monheit, AC. Health insurance status of the civilian noninstitutionalized population: 1996. Rockville (MD): Agency for Health Care Policy and Research; 1997. MEPS Research Findings No. 1. AHCPR Pub. No. 97-0030.

^top


Technical Appendix

Derivation of Insurance Status Information
Health Insurance Edits
Population Characteristics
Sample Design and Accuracy of Estimates
Rounding
Comparisons With Other Data Sources
Population and Standard Error Tables

The data in this report were obtained in the third round of interviews for the Household Component (HC) of the 1996 Medical Expenditure Panel Survey (MEPS) and the first round of interviews from the 1997 MEPS HC. MEPS is cosponsored by the Agency for Health Care Policy and Research (AHCPR) and the National Center for Health Statistics (NCHS). The MEPS HC is a nationally representative survey of the U.S. civilian noninstitutionalized population that collects medical expenditure data at both the person and household levels. The focus of the MEPS HC is to collect detailed data on demographic characteristics, health conditions, health status, use of medical care services, charges and payments, access to care, satisfaction with care, health insurance coverage, income, and employment. In other components of MEPS, data are collected on the use, charges, and payments reported by providers; residents of licensed or certified nursing homes; and the supply side of the health insurance market.

The sample for the MEPS HC was selected from respondents to the National Health Interview Survey (NHIS), which was conducted by NCHS. NHIS provides a nationally representative sample of the U.S. civilian noninstitutionalized population and reflects an oversampling of Hispanics and blacks. The MEPS HC collects data through an overlapping panel design. In this design, data are collected through a precontact interview that is followed by a series of five rounds of interviews over 212 years. Two calendar years of medical expenditure and utilization data are collected from each household and captured using computer-assisted personal interviewing (CAPI). This series of data collection rounds is launched again each subsequent year on a new sample of households to provide overlapping panels of survey data which, when combined with other ongoing panels, will provide continuous and current estimates of health care expenditures. The reference period for Round 3 of the 1996 MEPS HC (Panel 1) was from the Round 2 interview date to the Round 3 interview date. The reference period for Round 1 of the 1997 MEPS HC (Panel 2) was from January 1, 1997, to the date of the Round 1 interview. Interviews for Panel 1 (Round 3) and Panel 2 (Round 1) were conducted from March to July 1997.

Derivation of Insurance Status Information

The household respondent was asked if during the interview period anyone in the family was covered by any of the sources of public and private health insurance coverage discussed in the following paragraphs. For respondents in their third MEPS interview (Panel 1), previously reported insurance was reviewed to determine whether it was still in effect and when changes in insurance status had occurred. Although Panel 1 survey respondents were asked about their insurance information for part of 1996 as well as 1997, the insurance information in this report refers to coverage only in 1997. Medicare and CHAMPUS/CHAMPVA coverage were measured at the time of the interview. (CHAMPUS and CHAMPVA are the Civilian Health and Medical Programs for the Uniformed Services and Veterans’ Affairs.) All other sources of insurance were measured for any time between January 1997 and the interview. Persons counted as uninsured were uninsured throughout this time period.

Public Coverage

For this report, individuals were considered to have public coverage only if they met both of the following criteria:

  • They were not covered by private insurance.
  • They were covered by one of the public programs discussed below.

Medicare

Medicare is a federally financed health insurance plan for the elderly, persons receiving Social Security disability payments, and most persons with end-stage renal disease. Medicare Part A, which provides hospital insurance, is automatically given to those who are eligible for Social Security. Medicare Part B provides supplementary medical insurance that pays for medical expenses and can be purchased for a monthly premium.

CHAMPUS/CHAMPVA

CHAMPUS covers retired members of the Uniformed Services and the spouses and children of active-duty, retired, and deceased members. Spouses and children of veterans who died from a service-connected disability, or who are permanently disabled and are not eligible for CHAMPUS or Medicare, are covered by CHAMPVA. In this report, CHAMPUS or CHAMPVA coverage is considered to be public coverage. When persons covered by CHAMPUS/CHAMPVA reach age 65, their coverage generally ends and enrollees are eligible for Medicare.

Medicaid

Medicaid is a means-tested government program jointly financed by Federal and State funds that provides health care to those who are eligible. Program eligibility criteria vary significantly by State, but the program is designed to provide health coverage to families and individuals who are unable to afford necessary medical care.

Other Public Hospital/Physician Coverage

Respondents who did not report Medicaid coverage were asked if they were covered by any other public hospital/physician coverage. These questions were asked in an attempt to identify Medicaid recipients who might not have recognized their coverage as Medicaid. In this report, all coverage reported in this manner is considered public coverage.

Private Health Insurance

Private health insurance was defined for this report as insurance that provides coverage for hospital and physician care (including Medigap coverage). Insurance that provides coverage for a single service only, such as dental or vision coverage, was not counted. Private health insurance could have been obtained through an employer, union, self-employed business, directly from an insurance company or a health maintenance organization (HMO), through a group or association, or from someone outside the household.

Uninsured

The uninsured were defined as persons not covered by Medicare, CHAMPUS/CHAMPVA, Medicaid, other public hospital/physician programs, or private hospital/physician insurance (including Medigap coverage) during the period from January 1997 through the time of the interview. Individuals covered only by noncomprehensive State-specific programs (e.g., Maryland Kidney Disease Program) or private single-service plans (e.g., coverage for dental or vision care only, coverage for accidents or specific diseases) were not considered to be insured.

Health Insurance Edits

For the Round 1 (Panel 2) sample, minimal editing was performed on sources of public coverage and no edits were performed on the private coverage variables. For Round 3 (Panel 1), most of the insurance variables were logically edited to address issues that arose during Rounds 2 and 3 when reviewing insurance reported in earlier rounds. The health insurance data were edited as described below.

Medicare

Medicare coverage was edited for persons age 65 and over but not for persons under age 65. Persons age 65 and over were assigned Medicare coverage if they met one of the following criteria:

  • They answered “yes” to a follow-up question on whether they had received Social Security benefits.
  • They were covered by Medicaid, other public hospital/physician coverage, or Medigap coverage.
  • Their spouse was age 65 or over and covered by Medicare.
  • They were covered by CHAMPUS/CHAMPVA.

Medicaid

A small number of cases reporting Aid to Families with Dependent Children (AFDC) or Supplemental Security Income (SSI) coverage (questions included in the MEPS health insurance section for editing purposes) were assigned Medicaid coverage. Since this report does not distinguish among sources of public insurance, no further edits were performed using the other public hospital/physician coverage variables. Other public hospital/physician coverage was included, however, when considering whether an individual was covered only by public insurance.

CHAMPUS/CHAMPVA

Respondents age 65 and over who reported CHAMPUS/CHAMPVA coverage were instead classified as covered by Medicare.

Private Health Insurance

Private insurance coverage was unedited and unimputed for Round 1 (Panel 2). For Round 3 (Panel 1), most of the insurance variables were logically edited to address issues that arose during Rounds 2 and 3 when reviewing insurance reported in earlier rounds. One edit to the private insurance variables corrected for a problem concerning covered benefits when respondents reported a change in any of their health insurance plan names. Additional edits addressed issues of missing data on the time period of coverage.

Individuals were considered to be covered by private insurance if the insurance provided coverage for hospital/physician care. Medigap plans were included. Individuals covered by single-service plans only (e.g., dental, vision, or drug plans) were not considered to be privately insured. Sources of insurance with missing information regarding the type of coverage were assumed to contain hospital/physician coverage.

It should be noted that these data were generally reported by a single household respondent, who may not have been the most knowledgeable source for other family members. The employers and insurance companies of household respondents are being contacted in a follow-up survey as part of the MEPS data collection effort designed to verify and supplement the information provided by the household respondents.

Population Characteristics

Place of Residence

Individuals were identified as residing either inside or outside a metropolitan statistical area (MSA) as designated by the U.S. Office of Management and Budget (OMB), which applied 1990 standards using population counts from the 1990 U.S. census. An MSA is a large population nucleus combined with adjacent communities that have a high degree of economic and social integration within the nucleus. Each MSA has one or more central counties containing the area’s main population concentration. In New England, metropolitan areas consist of cities and towns rather than whole counties. Regions of residence are in accordance with the U.S. Bureau of the Census definition.

Race/Ethnicity

Classification by race and ethnicity was based on information reported for each household member. Respondents were asked if their race was best described as American Indian, Alaska Native, Asian or Pacific Islander, black, white, or other. They were also asked if their main national origin or ancestry was Puerto Rican; Cuban; Mexican, Mexicano, Mexican American, or Chicano; other Latin American; or other Spanish. All persons who claimed main national origin or ancestry in one of these Hispanic groups, regardless of racial background, were classified as Hispanic. Since the Hispanic grouping can include black Hispanic, white Hispanic, and other Hispanic, the race categories of black, white, and other do not include Hispanic.

Employment Status

Persons were considered to be employed if they were age 16 and over, and had a job for pay, owned a business, or worked without pay in a family business at the time of the Round 1 or Round 3 interview.

Sample Design and Accuracy of Estimates

MEPS is designed to produce estimates at the national and regional level over time for the civilian noninstitutionalized population of the United States and some subpopulations of interest. Each MEPS panel collects data covering a 2-year period, with the first two MEPS panels spanning 1996-97 and 1997-98, respectively. In this report, data from the 1997 portion of the third round of data collection for the MEPS Panel 1 sample are pooled with data from the first round of data collection for the MEPS Panel 2 sample (shaded portion of Figure A).

The statistics presented in this report are affected by both sampling error and sources of nonsampling error, which include nonresponse bias, respondent reporting errors, interviewer effects, and data processing misspecifications. For a detailed description of the MEPS survey design, the adopted sample design, and methods used to minimize sources of nonsampling error, see Cohen (1997) and Cohen, Monheit, Beauregard, et al. (1996). The MEPS person-level estimation

 

 Figure A: Overlapping panel design of the MEPS

Source: Center for Financing, Access, and Cost Trends, Agency for Health Care Policy and Research. 

weights include nonresponse adjustments and poststratification adjustments to population estimates derived from the March 1997 Current Population Survey (CPS) based on cross-classifications by region, MSA status, age, race/ethnicity, and sex.

Tests of statistical significance were used to determine whether the differences between populations exist at specified levels of confidence or whether they occurred by chance. Differences were tested using Z-scores having asymptotic normal properties at the 0.05 level of significance. Unless otherwise noted, only statistical differences between estimates are discussed in the text.

Panel 1

At its beginning in 1996, MEPS Panel 1 consisted of a sample of 10,639 households, a nationally representative subsample of the households responding to the 1995 National Health Interview Survey (NHIS). The 1995 NHIS sampled households with Hispanic members and households with black members at approximately 2.0 and 1.5 times the rate of other households, respectively. These oversampling rates are also reflected in the MEPS sample of households.

The overall MEPS Panel 1 response rate at the end of Round 3 (which collects data for the first part of 1997) was 70.2 percent. This overall rate reflects response to the 1995 NHIS interview and the MEPS interviews for Rounds 1-3.

Panel 2

At its beginning in 1997, MEPS Panel 2 consisted of a sample of 6,281 households, a nationally representative subsample of the households responding to the 1996 NHIS. Like Panel 1, the Panel 2 sample reflects the oversampling of Hispanic and black households in NHIS. However, the sample design for Panel 2 differed from that for Panel 1 because the following policy-relevant groups (classified based on 1996 NHIS data) were also oversampled to produce more reliable estimates for these groups:

  • Adults (age 18 and over) with functional impairments (difficulty with one or more activities of daily living).
  • Children (under age 18) with limitations in activity.
  • Individuals aged 18-64 expected to incur high medical expenditures in 1997.
  • Individuals predicted to reside in low-income households (below 200 percent of poverty level).
  • Adults (age 18 and over) with health limitations other than functional impairments (difficulty with one or more instrumental activities of daily living).

The overall MEPS Panel 2 response rate at the end of Round 1 (when data were collected for the first part of 1997) was 77.9 percent. This overall rate reflects response to both the 1996 NHIS interview and the MEPS Round 1 interview.

Combined Panel Response

Each panel was given approximately equal weight in the development of sampling weights to produce national estimates. Therefore, a pooled response rate for the survey respondents in this data set can be obtained by taking an average of the panel-specific response rates. This pooled response rate for the combined panels is 74.1 percent.

Rounding

Estimates presented in the tables were rounded to the nearest 0.1 percent. Standard errors, presented in Tables B-F, were rounded to the nearest 0.01. Population estimates in Tables 1-4 and Table A were rounded to the nearest thousand. Therefore, some of the estimates presented in the tables for population totals of subgroups will not add exactly to the overall estimated population total.

Comparisons with Other Data Sources

Other Surveys

Because of methodological differences, caution should be used when comparing these data with data from other sources. For example, CPS measures persons who are uninsured for a full year; NHIS measures persons who lack insurance at a given point in time--the month before the interview. The CPS interview that contains information on the health insurance status of the population is conducted annually, and NHIS collects insurance data on a continuous basis each year. In addition, unlike MEPS, CPS counts as insured military veterans whose source of health care is the Department of Veterans Affairs. CPS also counts children of adults covered by Medicaid as insured. For these preliminary estimates, MEPS did not consider these children insured unless their families reported them as such.

1996 MEPS Data

Users interested in comparing the 1996 and 1997 MEPS health insurance estimates should be aware that the standard errors presented in the tables do not account for the fact that the estimates are not independent. By design, the 1996 MEPS panel respondents are present in both the 1996 and 1997 sample populations. Users should also be aware of questionnaire wording differences in the Rounds 1 and 3 MEPS interviews. The questionnaire for the Round 3 interviews includes reviews of previously reported sources of health insurance coverage. Note that the 1996 estimates presented in MEPS Research Findings Number 1 (Vistnes and Monheit, 1997) are based on data obtained during the 1996 MEPS Round 1 interview, while the 1997 estimates are based on interviews conducted in Round 1 of the 1997 MEPS panel as well as Round 3 of the 1996 MEPS panel.

Population and Standard Error Tables:

 

Table A: Health insurance coverage by population characteristics

a Includes persons with unknown employment status.

b For individuals age 16 and over.

c Sample size too small to produce reliable estimates.

Source: Center for Financing, Access, and Cost Trends, Agency for Health Care Policy and Research: Medical Expenditure Panel Survey Household Component, 1997.

^top


Table B: Standard errors for Table 1

Table B: Standard errors for Table 1, continued

a Includes persons with unknown employment status and marital status.

b For individuals age 16 and over.

Source: Center for Financing, Access, and Cost Trends, Agency for Health Care Policy and Research: Medical Expenditure Panel Survey Household Component, 1997.

^top


Table C: Standard errors for Table 2

a Includes persons with unknown employment status and marital status.

b For individuals age 16 and over.

Source: Center for Financing, Access, and Cost Trends, Agency for Health Care Policy and Research: Medical Expenditure Panel Survey Household Component, 1997.

^top


Table D: Standard errors for Table 3

a Includes persons with unknown perceived health status.

* Relative standard error is greater than or equal to 30 percent.

Source: Center for Financing, Access, and Cost Trends, Agency for Health Care Policy and Research: Medical Expenditure Panel Survey Household Component, 1997.

^top


Table E: Standard errors for Table 4

Table E: Standard errors for Table 4, continued

a Includes persons with unknown marital status and perceived health status.

b For individuals age 16 and over.

Source: Center for Financing, Access, and Cost Trends, Agency for Health Care Policy and Research: Medical Expenditure Panel Survey Household Component, 1997.

^top


Table F: Standard errors for Table A

a Includes persons with unknown employment status.

b For individuals age 16 and over.

– Sample size too small to produce reliable estimates.

Source: Center for Financing, Access, and Cost Trends, Agency for Health Care Policy and Research: Medical Expenditure Panel Survey Household Component, 1997.

^top



Suggested Citation:
Vistnes, J. P. and Zuvekas, S. H. Research Findings #8: Health Insurance Status of the Civilian Noninstitutionalized Population, 1997. July 1999. Agency for Healthcare Research and Quality, Rockville, MD. http://www.meps.ahrq.gov/data_files/publications/rf8/rf8.shtml