|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Addressing Coverage Gaps For Low-Income Parents
This paper examines how rates of uninsurance for low-income parents have been changing over time and the extent to which expanding coverage to parents through Medicaid and the State Childrens Health Insurance Program (SCHIP) could help them. We find that uninsurance rates have been rising for low-income parents, especially those living in poverty, and that Medicaid and SCHIP could greatly reduce uninsurance among parents and would likely increase their access to care. Such expansions would still leave many noncitizen parents uninsured and would require reaching and enrolling families whose children have remained uninsured despite being eligible for public coverage.
Even though much of the policy discussion about the uninsured focuses on low-income children, low-income parents are much more likely than their children to be uninsured.1 In 1999, 16 percent of all parents lacked health insurance, and low-income parents were more than five times as likely as high-income parents to be uninsured (35.9 percent versus 6.8 percent).2 Low-income children are much more likely than their parents to qualify for some type of public coverage, since until very recently, expansions of public health insurance programs have been targeted primarily at low-income children. Beginning in the mid-1980s a series of laws first mandated expansions in eligibility for Medicaid under Title XIX of the Social Security Act for children and then allowed states to expand Medicaid eligibility beyond the mandated levels. These expansions were followed in 1997 by the creation of the State Childrens Health Insurance Program (SCHIP) under Title XXI. States that expand coverage to children receive federal matching dollars at higher rates under Title XXI than under Title XIX even if their SCHIP programs are implemented as Medicaid expansions. By September 2000 thirty-three states had Medicaid or SCHIP eligibility levels at or above 200 percent of the federal poverty level for children age eighteen and under.3 Although Medicaid coverage for adults has expanded to include many low-income pregnant women, coverage for parents in most states has historically been limited to those who qualify for cash-assistance programs, which typically restrict income eligibility to levels well below the federal poverty level. Beginning with the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) of 1996, recent changes in federal policy have provided greater access to federal matching dollars to states that want to expand coverage to parents.4 PRWORA gave states the option of using less restrictive Medicaid eligibility standards for low-income families. In addition, policy guidance issued by the Department of Health and Human Services (HHS) in July 2000 provides a new option to use unspent SCHIP funds to expand coverage to parents through SCHIP waivers, and since August 2001 states have even more latitude to expand coverage to parents under waivers through the Health Insurance Flexibility and Accountability (HIFA) initiative. A number of states have made use of these mechanisms to expand Medicaid eligibility to more parents.5 Although eligibility expansions to parents would be expected to reduce uninsurance among low-income parents, many occurred at the same time that federal welfare reform was being implemented, which appears to have contributed to reductions in Medicaid coverage for children and adults.6 In light of growing interest in expanding coverage to parents, this paper examines how coverage was changing for low-income parents in the late 1990s and the extent to which expanding coverage to parents through Medicaid and SCHIP could improve low-income families access to and use of health care services.
Data sources. This analysis draws on the Current Population Survey (CPS) and the National Survey of Americas Families (NSAF). We use the 2002 CPS to provide up-to-date estimates of coverage for parents and children and data from the March 19952001 CPS to obtain consistent estimates of changes in coverage over time. The income data used in the CPS analysis are based on the income of the individuals health insurance unit in relation to census poverty thresholds. Estimates from the 2002 CPS differ from those derived based on the earlier CPS years in two important ways: First, the data used to formulate the control totals for the weights are based on information from the 2000 census; and second, a question was added to confirm the absence of insurance coverage for different household members. Coverage estimates for the earlier years rely on control totals based on the 1990 census and are derived from an insurance sequence that did not include a question confirming the absence of insurance coverage. We use the NSAF to (1) contrast uninsurance rates for low-income parents and children in thirteen states; (2) estimate how many uninsured parents could be covered under Medicaid and SCHIP; and (3) provide a picture of the access situation facing low-income, uninsured parents.7 The NSAF provides nationally representative as well as state-representative estimates for thirteen states.8 Methods. To estimate how far expansions in Medicaid and SCHIP can go to solving the problem of uninsured parents, we first determined the number of uninsured parents in 1999. We then estimated how many uninsured parents had children eligible for Titles XIX and XXI based on the eligibility rules in place for children as of July 2000. We also estimated how many of these parents were noncitizens, and therefore potentially ineligible for Medicaid and SCHIP expansions, and how many parents had children who were enrolled in Medicaid or SCHIP at the time the NSAF survey was fielded in 1999 and, therefore, could readily be enrolled under an expansion. These analyses rely on a detailed Medicaid and SCHIP eligibility simulation model for children, designed to mimic the eligibility determination process faced by families applying for Medicaid or SCHIP.9 Our estimates adjust for the fact that some noncitizen uninsured parents who meet the income and other requirements for Medicaid or SCHIP eligibility still could not qualify for coverage under the programs because of their legal status.10 We use six indicators to compare access to and use of services among uninsured parents of Medicaid- and SCHIP-eligible children with the levels of their counterparts who are covered by Medicaid: having a usual source of care; confidence in ability to obtain care for the family; unmet need; physician visits; dental visits; and breast exams. We conduct both descriptive and multivariate analyses of the six access and use measures. We estimate linear probability models to assess the potential effects of insurance coverage that control for the age, race, work status, education, health status, activity limitations, parents citizenship, and familys income and welfare history.
Uninsurance rates. Trend data from the CPS indicate that uninsurance rates have been growing among low-income parents, particularly among parents with incomes below the poverty level (Exhibit 1
The picture for parents living below poverty looks bleaker: Their uninsurance rate rose from 34 percent in 1994 to 41 percent in 2000. Declines in Medicaid coverage appear to be driving the seven-percentage-point increase in uninsurance among poor parents (data not shown). Between 1994 and 2000 Medicaid coverage fell for poor parents by twelve percentage points, which was offset only partially by five-percentage-point increases in employer/private coverage. This increase was likely an outgrowth of both federal welfare reform and the strong economy. The fact that Medicaid coverage fell so much for poor parents over this period suggests that the effects of welfare reform may have overwhelmed any expansions in eligibility that were occurring for poor parents.
Coverage gaps between parents and children.
The coverage gap between low-income parents and their children increased over this time period (Exhibit 2
The magnitude of the parent/child coverage differential varies dramatically across states, according to data from the 1999 NSAF. Among the thirteen states with large samples in the NSAF, the difference in the uninsurance rate between low-income parents and children is smallest in Minnesota, where low-income parents are only six percentage points more likely than their children to be uninsured, and greatest in Alabama, where they are twenty-two percentage points more likely than their children to be uninsured (Exhibit 4
Coverage potential of Medicaid and SCHIP. In light of these patterns, expansions of Medicaid and SCHIP to cover more parents may be critical to reducing coverage gaps among parents. It appears that 7.4 million uninsured parents (70 percent of all uninsured parents) could be covered if eligibility levels for parents were raised to the eligibility levels states have set for children. Medicaid expansions alone could cover half of all uninsured parents, potentially providing coverage to roughly 5.6 million uninsured parents, while expansions up to the higher SCHIP eligibility levels could provide coverage to another 20 percent (data not shown).11 Another 1.2 million uninsured noncitizen parents have incomes below the current eligibility levels for children but would not qualify for Medicaid or SCHIP under current federal law because of their immigrant status. How easy it would be to reach these potentially eligible uninsured parents is likely to depend on whether their children are already enrolled in Medicaid or SCHIP. Data from 1999 suggest that at least three million uninsured parents could be readily enrolled in these programs since their children are already participating. Importantly, given the continued growth in SCHIP since the 1999 NSAF, even more uninsured parents are likely to have SCHIP-enrolled children now.12 Enrollment challenges. Enrolling uninsured parents whose children are uninsured presents challenges to Medicaid and SCHIP. Other research suggests that a combination of factors, including knowledge gaps, administrative barriers, and lack of interest, keep eligible children from being enrolled in public programs.13 Data from early 2001 suggest that most low-income parents with uninsured children would like to enroll them in Medicaid or SCHIP, but many either have not heard of the programs, do not know that their child is eligible, or do not perceive that the Medicaid and SCHIP application processes are easy.14 More barriers were identified for children whose parents have less education or who were interviewed in Spanish.
Access problems among uninsured parents.
The uninsured parents who could be targeted by Medicaid or SCHIP expansions appear to experience major access problems. In 1999, 37 percent had no usual source of care or used the emergency room as a usual source of care, and 28 percent experienced some type of unmet need for care (Exhibit 5
Relative to Medicaid-covered parents, these uninsured parents are about twice as likely to have no usual source of care or to rely on the emergency room and to lack confidence in their familys ability to receive needed care. They are also just two-thirds as likely to have had physician visits, dental care, or breast examinations in the preceding year. These access gaps persist even when we control for a number of factors that affect access and use. Other things equal, relative to Medicaid-covered parents, uninsured parents were sixteen percentage points more likely to have no usual source of care or to rely on the emergency room for care in the past year; seven percentage points more likely to have unmet health care needs; nineteen percentage points less likely to have had a physician visit; seven percentage points less likely to have had a dental visit; and eleven percentage points less likely to have received a breast exam. While unobserved factors may contribute to these differentials, these findings suggest that uninsured parents could experience meaningful improvements in their access to care if they were to gain coverage under Medicaid or SCHIP.
The problem of uninsurance appears to be growing among low-income parents. The most recent data indicate that more than a third of all low-income parents lack health insurance and that the problem is even more pronounced for poor parents, who lack coverage at rates exceeding 40 percent. Our analysis shows that raising Medicaid and SCHIP eligibility standards for parents to the levels that prevail for children could reduce the group of uninsured parents to a fraction of its current size. Of the 10.6 million parents who are uninsured, 7.4 million could qualify for coverage if the Medicaid/SCHIP eligibility thresholds were expanded for parents to the levels states have adopted for children. Such expansions also appear likely to provide benefits to children as well, because family coverage policies appear to lead to lower uninsurance rates among eligible children and to improve access among children who are already covered.16 Federal policy changes made in the past several years mean that Medicaid programs, which could cover as many as half of all uninsured parents, could easily be expanded to cover more poor and many nonpoor parents.17 However, expanding Medicaid and SCHIP is one of many policy options available to reduce uninsurance among parents. For instance, a number of states have developed premium-assistance programs aimed at making more efficient use of public dollars to support insurance coverage, and tax credits have been proposed at the federal level.18 Importance of Medicaid. Alternatives such as premium-assistance programs and tax credits would not likely be as effective as Medicaid expansions at closing coverage gaps for poor parents, for a number of reasons. Uninsured parents living below poverty have little access to employer-sponsored coveragefewer than 17 percent have any type of offer of health insurance through an employer, and more than half reported that they had had trouble paying housing or utility bills or experienced or worried about food shortages in the previous year (data not shown). Thus, most uninsured parents living below poverty would require substantial subsidies to afford nongroup coverage. Without Medicaid expansions, it is unlikely that parents living in poverty will gain health insurance coverage. Moreover, since few poor parents (21 percent in 2000) have employer-sponsored coverage, there is limited potential for expansions in Medicaid to crowd out employer-sponsored coverage among poor parents. Outreach challenges. While increasing eligibility thresholds for parents wont automatically provide coverage to the parents who have children enrolled in Medicaid or SCHIP, the policy problem is the relatively straightforward one of enrolling the parents of children who are already covered. But many uninsured parents have children who are uninsured despite being eligible for Medicaid or SCHIP. Reaching both the children and parents in these families is more challenging and may require additional policy innovation and investments in a number of areas related to raising awareness and understanding of the programs and their benefits, facilitating enrollment and reenrollment in programs, and addressing barriers related to other program dimensions.19 Covering noncitizens. Even if all states expanded Medicaid and SCHIP eligibility to parents, an estimated 1.2 million would still lack Medicaid/SCHIP coverage, because federal law prohibits use of federal dollars to cover undocumented aliens and certain legal aliens. Since noncitizens constitute more than a quarter of all uninsured parents who meet the Medicaid/SCHIP thresholds for children, current law represents a barrier to eliminating uninsurance among low-income parents. Even among noncitizens who qualify for coverage under existing rules, anecdotal evidence strongly suggests that fears around public-charge issues act as a deterrent to applying for Medicaid, as well as for other government programs to which they are entitled.20 Costs of expansion. Ultimately, states willingness to expand Medicaid or SCHIP to cover more parents will depend on their political, budgetary, and economic situations and on the availability of federal resources to help defray the cost of expansions. States now have considerable flexibility to expand coverage to parents, although it is unlikely that any new federal funds, beyond what is available in the Title XXI block grant, will be made available for such expansions.21 Importantly, states that have come close to using their SCHIP allotment, such as Maryland, New Jersey, New York, and Rhode Island, will face constraints on tapping SCHIP funds for maintaining coverage for parents, while states with large unspent SCHIP allotments will be more able to draw on SCHIP funds.22 Expanding Medicaid and SCHIP to cover parents with incomes below 200 percent of poverty would likely cost more than $10 billion per year.23 The worsening of states budget situations is making it difficult for states to allocate state funds to sustain their existing programs, let alone expand to new groups such as parents. In fact, faced with budget problems, states such as Missouri and New Jersey have rolled back eligibility expansions for parents, and others, including California, which had planned one of the most ambitious expansions, have proposed cuts. It appears that addressing coverage gaps among low-income populations generally and among parents in particular through Medicaid and SCHIP may require the federal government to play a larger role in financing coverage.24 If states fail to expand coverage because of fiscal constraints or pull back on expansions, outreach efforts, or enrollment simplifications, the coverage situation for low-income parents will likely deteriorate further. This would have adverse effects on both low-income parents and their children and may have negative ramifications for communities as well.25
Funding for this project was provided by the Kaiser Commission on Medicaid and the Uninsured as part of the Urban Institutes Assessing the New Federalism Project. The opinions expressed are those of the authors and do not reflect those of the Urban Institute or its funders. The authors are grateful to John Holahan, Jocelyn Guyer, and Barbara Lyons for their insightful suggestions and assistance on earlier drafts; Randy Capps and Jeff Passell for running special tabulations on the Current Population Survey; Felicity Skidmore for her skillful editing; and Jennifer Haley, Nirmala Ramalingam, and Alexandra Tebay for their careful and patient assistance throughout the entire process. Lisa Dubay (ldubay{at}ui.urban.org) is a senior research associate at the Urban Institute in Washington, D.C., and Genevieve Kenney (jkenney{at}ui.urban.org) is a principal research associate there.
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||