Introduction

bout one year ago, the Balanced Budget Act of 1997 was enacted. It contained the largest expansion of publicly supported insurance coverage in three decades. The act created the State Children’s Health Insurance Program (CHIP) and authorizes about $40 billion over the next 10 years for it. Although there was widespread bipartisan Congressional support for expanding health insurance coverage to children, less consensus existed as to the best approach to achieving this expansion. The resulting program reflects a compromise among competing approaches. Block grants are available to states with which they have the option of expanding Medicaid eligibility, establishing new state programs that subsidize the provision of private insurance to children (under a new Title XXI), or a combination of the two approaches. Although these grants come with strings attached and require state matching funds, states are provided considerable latitude in designing their programs.

Most states have enacted their CHIP programs. As of June, 1998, 36 states plus the District of Columbia have submitted plans to HCFA. Of these, 20 have been approved. Other state legislatures are still considering their plans and only two have decided not to expand coverage for children at this time. At present, it appears that about 20 states are pursuing Medicaid expansions, 14 states are creating new Title XXI programs, and 25 are combining the two approaches. For the most part, states that combine the two approaches are relying mostly on new Title XXI programs, only expanding Medicaid to coverage teenagers who would have been covered anyway under previous legislation by the year 2002. Because some states rushed to design a program in time to qualify for first year funding, it is likely state approaches to implementation of CHIP will continue to evolve.

The ultimate success of CHIP does not primarily depend on whether states pursue public or private insurance options, but rather on the specific details of these programs and how they are implemented. In particular, success in attracting children into these new programs will likely rest on the ease of enrollment, cost sharing requirements, whether the insurance programs are packaged in a way that reduces stigma associated with their use, and how effective publicity and outreach efforts are. The final key to success is whether states provide themselves the flexibility to tailor their programs to different areas of the state and to different target populations. Although the problem of uninsured children is often thought of as a national problem, the scope and nature of the problem varies tremendously at the local level. Table 1 provides estimates of uninsurance rates among children in twelve randomly selected metropolitan areas. ¹ A fivefold difference is found between Lansing, Michigan, the site with th owest rate at 4.4%, and Miami, Florida, the site with the highest rate at 22.4%.

These findings suggest that approaches to cover uninsured children need to tailored to the size and characteristics of the target population and the reasons they are uninsured. This study uses new data to explore the children lack health insurance. These reasons relate to why children lack access to coverage from employer-sponsored health plans, why their parents might decline offered coverage, gaps in the coverage of public health insurance programs, and the failure of families to enroll their children in public insurance programs. Implications of these results for the design and implementation of new state programs are explored.

Data and Methods

Our data are from a new source, the Community Tracking Study’s Household Survey. This survey is one component of the Community Tracking Study (CTS), a major initiative of the Robert Wood Johnson Foundation to track changes in the health care system over time and to gain a better understanding of how health system changes are affecting both consumers and providers.²

Data on insurance coverage and attributes, demographic characteristics, health status, access to and utilization of health care, and satisfaction with the health care system were obtained from telephone interviews conducted between July 1996 and July 1997, supplemented by in-person interviews of families in households lacking a telephone. As such, the survey provides the most recent data available.

The survey contains observations on nearly 33,000 families, about 11,600 of whom have children. Detailed information was gathered on all adults in each family and one randomly chosen child. Altogether, the survey has information on over 60,000 individuals. The data are weighted to be representative of the U.S. population in the contiguous 48 states.

Characteristics of the population of uninsured children derived from the CTS household survey closely mirror those found in other national surveys, such as the Current Population Survey or the Medical Expenditure Survey.³ These are shown in the second numerical column of Table 2. Although uninsured children represent a diverse group, in general we can characterize them as largely children of the working poor. Nearly four fifths have parents in the work force. Despite this, 36% of uninsured children live in poverty and another 39% live in families with incomes between 100% and 200% of poverty. Two thirds come from intact, 2 parent families.

A majority of uninsured children are white, although minorities are disproportionately represented. This is particularly true for Hispanic children, who make up 31% of all uninsured children. Over one Hispanic child in four lacks health insurance, almost twice the rate among Black, non-Hispanic children, and two and a half times the rate of white, non-Hispanic children. About 7% of uninsured children are in fair or poor health. Although not shown in the table, this rate falls between the rate for children with public insurance (10.0%) and the rate found among children with private insurance (3.2%). The differences, which are statistically significant, probably reflect differences in income, special access to Medicaid by disabled and seriously ill children, along with differences in access to care attributable to insurance status. Nevertheless, among children in fair or poor health, one in six lacks insurance coverage.

¹ The 12 metropolitan areas were selected among all metropolitan statistical areas with populations over 200,000 in the continental United States. Data sources are described later in the text.

² A description of the CTS can be found in P. Kemper, et. al., "The Design of the Community Tracking Study: A Longitudinal Study of Health System Change and Its Effects on People," Inquiry (Summer 1996): 195-206.

³ We estimate that there are 8.8 million children aged 18 and younger who lack health insurance in the 48 contiguous states and the District of Columbia. This is lower than estimates from the Current Population Survey and the Medical Expenditure Survey. For instance, the CPS estimates that about 11.1 million children are uninsured (R. Bennefield, Health Insurance Coverage, 1996. Washington, DC: Census Bureau, Current Population Reports, P60-199, September, 1997). Reasons for the difference are likely attributable to variations in sampling universes, survey methodology and instrumentation. Each survey has its relative strengths and weaknesses so that the actual number of uninsured children remains uncertain.