Health Affairs, 24, no. 5 (2005): 1344-1355 doi: 10.1377/hlthaff.24.5.1344 © 2005 by Project HOPE	New Online   How Would Obama, McCain Cover The Uninsured?   Debating Cost Of Uninsured   Try Medicare-For-All   HA Blog Top 10

This Article Abstract Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Mitchell, J. B. Articles by Hoover, S. Search for Related Content PubMed PubMed Citation Articles by Mitchell, J. B. Articles by Hoover, S. Related Collections Insurance Market Access To Care Consumer Issues Health Reform Insurance Coverage Insurance Coverage - Children Medicaid State/Local Issues

DataWatch

Premium Subsidy Programs: Who Enrolls, And How Do They Fare?

Abstract

 
Little is known about who enrolls in state premium subsidy programs or enrollees’ experiences. This study surveyed parents of children enrolled in two programs with identical income eligibility requirements: Oregon’s State Children’s Health Insurance Program (SCHIP) and its premium subsidy program (FHIAP). Parents choosing FHIAP were more likely to be employed, to speak English, to have prior experience with premiums and private health insurance, and to perceive insurance as protection against future health care needs. Despite copayment requirements and more restricted benefits in FHIAP, there were few differences in access to care between children enrolled in the two programs.

States can adopt several approaches in implementing such programs, and Congress has given states increased flexibility over time. The Health Insurance Flexibility and Accountability Act (HIFA), passed in 2001, allows states greater latitude in using Section 1115 waivers to cover the uninsured using Medicaid and SCHIP funds. The HIFA initiative requires that states coordinate their waiver programs with private coverage, and several states are complying with this requirement by implementing premium subsidy programs.¹ Thirteen states operate such programs through the Health Insurance Premium Payment (HIPP) program or Medicaid/SCHIP 1115 waivers.²

Most of the policy attention has centered on the administrative difficulties involved in implementing premium subsidy programs.³ A variety of barriers, including lack of employer cooperation and strict federal criteria regarding benefit packages and cost-sharing limitations, have made it difficult for many states to enroll more than a small number of beneficiaries through these vehicles. Little if any attention has been focused on how low-income families view these programs relative to traditional SCHIP and Medicaid, or how the experiences of people enrolled in such programs might differ. Oregon offers a premium subsidy program, whose income eligibility requirements are identical to those of its SCHIP program. This provided us with a unique opportunity to examine why low-income parents might choose the premium subsidy over SCHIP for their children.

The premium subsidy program provides coverage for the entire family (not just the children) through private insurers, rather than through a public program. Its benefits are more restricted than SCHIP’s, and it requires cost sharing, which is nonexistent in Oregon’s SCHIP. Thus, parents choosing one program over the other are making explicit trade-offs between services, the extent and source of coverage, and out-of-pocket payments. The more restricted benefit package and increased cost sharing of most private plans could pose barriers to access for children in premium subsidy programs.⁴ On the other hand, children are more likely to be covered by health insurance when coverage is offered to their parents as well.⁵ There is also some evidence that children have better access to care when enrolled in the same plan their parents have.⁶

In this paper we examine the factors leading parents to choose the premium subsidy program over SCHIP. We also compare the experience of children enrolled in each program with regard to access, use of services, and satisfaction.

SCHIP And Premium Subsidies In Oregon

Top SCHIP And Premium Subsidies... Study Data And Methods Study Findings Conclusions And Policy... NOTES

 
Oregon’s SCHIP. Oregon’s SCHIP is a Medicaid "look-alike" program. It provides coverage to children up to 170 percent of poverty through the Oregon Health Plan (OHP), Oregon’s Medicaid program. Parents cannot distinguish between Medicaid and SCHIP; they only know that they are applying for OHP. Children receive the traditional Medicaid benefit package, including dental care and prescription drugs, subject to OHP’s priority list. Previous research has shown that parents of OHP children have relatively few complaints about service denial because of the priority list.⁷ There are no premiums or copayments for SCHIP children in OHP, and the vast majority of SCHIP children receive their care through capitated managed care plans.

Oregon’s premium assistance program. At the time of our study (2002), the Family Health Insurance Assistance Program (FHIAP) was a state-funded program that subsidized premiums for families with incomes up to 170 percent of poverty. FHIAP is administered by the Insurance Pool Governing Board and not by the state agency responsible for OHP. (Under Oregon’s recently approved HIFA waiver, the state now receives federal matching funds for these subsidy payments for all families with Medicaid- and SCHIP-eligible children.)⁸ The subsidy can be used to purchase either group (employer-sponsored ) coverage or coverage in the individual market through a plan certified to participate in FHIAP. In fact, the majority of families in FHIAP with children (70 percent) at the time of our study did not have access to employer-sponsored coverage and took advantage of the individual market option.⁹ Six carriers are certified to offer individual coverage through FHIAP. Most carriers offer a variety of plan options that include the most popular plans in the state. The size of the premium subsidy varies by family income: Families with incomes below 126 percent of poverty are subsidized for 95 percent of the premium cost; those with incomes between 126 percent and 150 percent of poverty receive a 90 percent subsidy; and the remaining families (up to 170 percent of poverty) receive a subsidy of 70 percent. At the time of our study, 53.5 percent of families received the 95 percent subsidy, 29.8 percent received the 90 percent subsidy, and only 16.7 percent received the 70 percent subsidy. In addition to their shares of the premium, families must also pay the deductible and copayment amounts associated with the plan of their choice. Most individual plans offered through FHIAP (but not necessarily the group insurance plans) offer prescription drug coverage; almost none offer dental coverage.

Study Data And Methods

Top SCHIP And Premium Subsidies... Study Data And Methods Study Findings Conclusions And Policy... NOTES

 
Study sample. Samples of children age seventeen and under were randomly selected from both the OHP and FHIAP enrollment files. The sample of children in OHP was limited to children eligible for SCHIP. We used a two-stage sampling approach, first sampling families and then one child within each sampled family. Because of the relatively small size of FHIAP (the number of enrollees was limited by the state budget), we ended up sampling all families with children in the program. The survey was conducted in both English and Spanish in the spring and summer of 2002, using computer-assisted telephone interviewing techniques. Respondents were almost exclusively the child’s parents, usually the mother. The final sample included 1,206 OHP children and 339 FHIAP children, with response rates of 53.4 and 59.1 percent, respectively. The most common reason for nonresponse was the inability to locate the family because of disconnected phone lines, lack of forwarding addresses, and so forth. These response rates are similar to those achieved in other published surveys of Medicaid and other low-income populations.¹⁰ Response propensity weighting was used to adjust for nonresponse in the analyses.

Key measures. This study was part of the Child Health Insurance Research Initiative (CHIRI), a collaboration of nine research projects funded by the Agency for Healthcare Research and Quality (AHRQ). CHIRI investigators collaborated in developing a common core of key measures, including access, unmet need, and health status variables. These measures were largely drawn from standard, well-validated instruments such as the National Health Interview Survey (NHIS).¹¹

Study Findings

Top SCHIP And Premium Subsidies... Study Data And Methods Study Findings Conclusions And Policy... NOTES

 
Prior knowledge and reasons for choice. Applications for both OHP and FHIAP reference the other program, advising applicants that they may be eligible for either program. We asked parents whether they had heard of the other program. Virtually all FHIAP parents (96 percent) had heard of OHP; in fact, many had been enrolled in OHP at some time in the past. Parents were then asked the main reason why they chose to enroll their child in FHIAP rather than OHP. More than half (52.2 percent) responded that they did not think their child was eligible for OHP. Other major reasons included a preference for private, rather than public, insurance (16.2 percent); the desire to insure the entire family (16 percent); and a wish to keep their current insurance plan or physician (7 percent). By contrast, only 14 percent of OHP parents said that they had heard of FHIAP.

Who enrolls? OHP children were significantly more likely than FHIAP children to be of Hispanic origin (Exhibit 1). Parents of OHP children were also significantly more likely to not speak English. There were no differences by age or sex of children (data not shown). While the responding parents of OHP and FHIAP children were equally likely to be employed full time, FHIAP children were significantly more likely than OHP children to have at least one parent employed (Exhibit 1). This is because parents of FHIAP children were more likely to be married, thereby increasing the number of potential wage earners. Parents of FHIAP children were also significantly better educated; almost all had graduated from high school, compared with only 70.9 percent of parents of OHP children (Exhibit 1).

Prior insurance history. Working parents were asked if they were eligible for employer-sponsored insurance at the time they enrolled in OHP or FHIAP. About one-quarter had access to employer coverage, with no differences between parents of OHP and FHIAP children (Exhibit 2). However, FHIAP parents were significantly more likely than OHP parents to actually be covered by job-based insurance. Considering all sources of health insurance, FHIAP parents were also more likely than OHP parents to be insured. However, the majority of both groups of parents were uninsured at the time of their children’s enrollment.

There were significant differences in the type of health insurance between the two groups of insured children as well. FHIAP children with prior insurance coverage were more likely to have purchased that insurance directly from the insurer, while OHP children were more likely to have had that previous insurance spell covered through OHP.

We hypothesized that parents’ health insurance choices might be at least partially driven by their own or other family members’ prior experiences with different types of coverage. Interestingly, the two groups of parents were equally likely to have had a family member enrolled in OHP in the past. However, FHIAP parents were significantly more likely than OHP parents to have had a family member covered by private health insurance and to have had experience with paying health insurance premiums. This is particularly noteworthy because it shows that low-income families in Oregon could have experience paying premiums, even if they did not have prior private health insurance coverage. OHP includes an eligibility expansion that covers adults up to 100 percent of poverty but requires monthly premiums.¹²

Perceived purpose of health insurance for children. Parents were asked to choose the most important reason for having health insurance coverage for their children, from the following three reasons: (1) to pay for possible future accident or illness; (2) to pay for a current medical condition; or (3) to pay for routine checkups. There were marked differences in parents’ perceptions of health insurance. FHIAP parents were significantly more likely (p <.01) to report that the most important reason for health insurance coverage was to pay for a child’s future accident or illness—68.2 percent, compared with only 45.9 percent of OHP parents. A mere 3.3 percent of FHIAP parents reported that coverage was important to pay for a child’s current medical condition, versus 16 percent of OHP parents; 28.5 percent of FHIAP parents considered coverage important to pay for routine checkups, versus 38.1 percent of OHP parents.

Factors explaining parents’ choice of programs. We estimated a logistic regression to examine the factors leading parents to enroll their children in FHIAP rather than OHP (Exhibit 3). Choice appears to be driven strictly by parents’ characteristics, experiences, and values. Characteristics of the child, including health status as measured by special needs, were not significant predictors of program choice. Families with more highly educated parents, and those in which at least one parent was employed, were significantly more likely to enroll in FHIAP than OHP. Prior family experience with paying premiums and a belief that health insurance is important for protecting against future (as opposed to current) health care needs also significantly increased the odds of choosing FHIAP. Hispanic families per se were not any less likely to enroll in FHIAP, but Hispanic families whose parents did not speak English were significantly less likely to choose FHIAP. The odds of having a non-English-speaking Hispanic parent choose FHIAP were only 15 percent, compared with an English-speaking non-Hispanic parent (based on a joint test of both the Hispanic and the Hispanic-non-English-speaker interaction term parameters).

The majority of children in both programs had seen their primary care physician at least once during the past six months (Exhibit 5). (Primary care visits could include both routine checkups and illness visits.) Relatively few children in either program had been hospitalized during this time period, and OHP children were significantly more likely than FHIAP children to have had emergency room (ER) visits.

Absolute levels of unmet need for dental care were considerably higher than for other types of care for both groups of children, and FHIAP children reported significantly higher rates than OHP children reported. Parents of FHIAP children with unmet need for dental care reported high costs and the lack of dental coverage as the primary reasons for not getting the care. By contrast, parents of OHP children cited their inability to find OHP-participating dentists.

We found no differences in unmet need for prescription drugs or mental health care. The apparent high levels of unmet need for mental health care are misleading, because only about 10 percent of children in either program reported that they needed such care in the first place.

Health status and other differences between OHP and FHIAP children could explain these differences in use and unmet need. To test this, we estimated logistic regressions for each of the measures shown in Exhibit 5, controlling for the child’s health status and demographics, geographic location, and parental characteristics such as education and employment. The regression results (not shown) confirmed the descriptive findings. Compared with OHP children, FHIAP children were 38 percent less likely to have had an ER visit (based on the odds ratio associated with a dummy variable for FHIAP enrollment), 66 percent less likely to report unmet need for primary care, and 87 percent more likely to report unmet need for dental care. There were no differences in any of the other measures (including unmet need for specialty care).

Satisfaction with care. Parents were asked to rate their child’s health care on a scale of 0 to 10. Parents in both programs rated care very highly, although the ratings were higher among FHIAP parents (average 8.8, compared with 8.1 for OHP children). Parents were equally satisfied with the benefits their children received, with 89.9 percent of FHIAP parents and 88.2 percent of OHP parents reporting that they were very or somewhat satisfied (data not shown).

Conclusions And Policy Implications

Top SCHIP And Premium Subsidies... Study Data And Methods Study Findings Conclusions And Policy... NOTES

 
For a number of years, programs such as premium subsidies that help low-income working families purchase health insurance through the workplace have been widely discussed as a prescription for high uninsurance rates. However, if premium subsidy programs are to be successful in enrolling low-income families, the results of our study suggest that these programs may need to be accompanied by efforts to educate these families about the importance of health insurance and how it works.

We found that characteristics of the parent, not the child, were key determinants of whether the family chose the premium assistance program (FHIAP) or the Medicaid option (OHP). In general, more highly educated parents and those with experience paying premiums for private coverage were more likely to choose FHIAP than OHP for their child. Similarly, parents who perceived health insurance as protection against future health care needs were more likely to choose FHIAP.

Some populations may be particularly difficult to attract to premium subsidy programs. The odds that a Spanish-speaking Hispanic parent would enroll a child in FHIAP were only 15 percent of the odds that an English-speaking non-Hispanic parent would do so, even after differences in education, employment, and prior insurance history were controlled for. Although we do not know the reasons behind these differences, one factor could be the provider networks available through private insurance. Many low-income families covered by Medicaid or SCHIP rely on community health centers and other safety-net providers for their care. Non-English-speaking families may be particularly likely to seek out these safety-net providers because they often provide on-site translators and other culturally sensitive services. These providers are often not included in the networks maintained by private plans. As a result, premium subsidy programs could inadvertently disrupt usual sources of care and cause access problems for non-English-speaking families.

Only one-third of low-income uninsured children are estimated to have access to employer-sponsored insurance, and some policymakers see this as a major limitation of premium subsidy programs.¹⁴ Our study confirms these concerns. Although parents who were employed at the time they applied for coverage were significantly more likely to choose FHIAP over OHP, only one-quarter of families in our study had access to employer-sponsored insurance at the time of enrollment in either program.

Oregon’s program is different from those of other states in that it subsidizes health insurance coverage purchased in the individual market, as well as job-based insurance. At the time of our study, 70 percent of families in FHIAP received their coverage through the individual market. However, the lack of an employer contribution to the premiums for individually purchased insurance greatly increases the cost to the state.

Families receiving premium subsidies are responsible for the same deductibles and copayments as any other family enrolled in the same private health insurance program. Given their low incomes, these out-of-pocket payments could pose a burden on families and affect access to care for potentially vulnerable children. We found no evidence that this was occurring. Children enrolled in FHIAP were just as likely to have seen a doctor for routine care during the previous six months as were children enrolled in OHP (which has no cost-sharing requirements). It should be noted, however, that copayment requirements were generally modest. The very generous subsidy offered through FHIAP encourages families to choose high-option plans with low cost sharing. Thus, the level of cost sharing in other states’ premium subsidy programs might be considerably higher than that faced by many FHIAP enrollees.

As private insurance plans increasingly shift costs to employees through higher copayments and deductibles, out-of-pocket payments under premium subsidy programs could pose a greater barrier for low-income families than those that we found in Oregon. Thus, it will be important to monitor how low-income children fare in programs that have greater cost sharing.

Editor's Notes

 
The authors are all with RTI International in Waltham, Massachusetts. Janet Mitchell (jmitchell{at}rti.org) is director of the Division for Health Services and Social Policy Research at RTI; Susan Haber is a senior economist and Sonja Hoover is a health services researcher in the same division.

The authors gratefully acknowledge funding from the Agency for Healthcare Research and Quality (Cooperative Agreement no. HS10463), with cofunding from the David and Lucile Packard Foundation and the Health Resources and Services Administration. The authors also thank staff from Oregon’s Office of Medical Assistance Programs, Family Health Insurance Assistance Program, and Office for Oregon Health Policy and Research.

NOTES

Top SCHIP And Premium Subsidies... Study Data And Methods Study Findings Conclusions And Policy... NOTES

 

1. C. Williams, A Snapshot of State Experience Implementing Premium Assistance Programs (Portland, Maine: National Academy for State Health Policy, 2003).
2. National Conference of State Legislatures, Summary of Employer Premium Assistance Programs, July 2003, www.ncsl.org/programs/health/buyin03.htm (24 May 2005).
3. U.S. Government Accountability Office, Medicaid: Three States’ Experience in Buying Employer-based Health Insurance, Pub no. GAO/HEHS-97-159 (Washington: GAO, 1997); R.E. Curtis and E. Neuschler, "Premium Assistance," Future of Children 13, no. 1 (2003): 214–223[CrossRef][Medline]; E. Neuschler and R. Curtis, Premium Assistance: What Works? What Doesn’t? (Washington: Institute for Health Policy Solutions, 2003); and A. Lutsky and I. Hill, Premium Assistance Programs under SCHIP: Not for the Faint of Heart? Assessing the New Federalism Occasional Paper no. 65 (Washington: Urban Institute, 2003).
4. States offering premium subsidies through HIPP or Section 1115 waivers must provide wraparound coverage for Medicaid/SCHIP–covered services that are not part of the private insurance plan. States must also place limits on the total cost-sharing amount borne by families. However, states with HIFA waivers are not bound to either of these requirements.
5. K.L. Hanson, "Patterns of Insurance Coverage within Families with Children," Health Affairs 20, no. 1 (2001): 240–246[Abstract/Free Full Text]; S. Guendelman and M. Pearl, "Children’s Ability to Access and Use Health Care," Health Affairs 23, no. 2 (2004): 235–244[Abstract/Free Full Text]; A. Davidoff et al., Patterns of Child-Parent Insurance Coverage: Implications for Coverage Expansions, Assessing the New Federalism Series B, no. B-39 (Washington: Urban Institute, 2001); and L. Dubay and G. Kenney, "Covering Parents through Medicaid and SCHIP: Potential Benefits to Low-Income Parents and Children," Background paper prepared for the Kaiser Commission on Medicaid and the Uninsured (Washington: Urban Institute, 2001).
6. K.L. Hanson, "Is Insurance for Children Enough? The Link between Parents’ and Children’s Health Care Use Revisited," Inquiry 35, no. 3 (1998): 294–302.[ISI][Medline]
7. J.B. Mitchell et al., "Children in the Oregon Health Plan: How Have They Fared?" Medical Care Research and Review 59, no. 2 (2002): 166–183.[Abstract/Free Full Text]
8. Oregon’s HIFA waiver was implemented just after our survey in November 2002. In addition, the waiver increased the income eligibility ceiling to 185 percent of poverty for both OHP’s SCHIP and FHIAP.
9. To maximize the number of people receiving subsidies through FHIAP, Oregon’s HIFA waiver stipulated that new enrollment in individual coverage plans would be frozen until enrollment in group plans grew to the point at which spending on subsidies for group and individual insurance was equalized. This allowed the state to cover more individuals for the same budget (as the employer share reduces the state’s premium subsidy payment for group plans). By March 2005, only 37 percent of children and 56 percent of adults were enrolled in individual plans.
10. See, for example, J.A. Brown et al., "Special Issues in Assessing Care of Medicaid Recipients," Medical Care 37, no. 3 Supp. (1999): MS79–MS88[CrossRef][ISI][Medline]; S. Long and T. Coughlin, "Impacts of Medicaid Managed Care on Children," Health Services Research 36, no. 1, Part 1 (2001): 7–23[ISI][Medline]; and J. Mitchell et al., "Impact of the Oregon Health Plan on Access and Satisfaction of Adults with Low Income," Health Services Research 37, no. 1 (2002): 11–31.[ISI][Medline]
11. For more information, see the Child Health Insurance Research Initiative home page, www.ahrq.gov/chiri.
12. OHP’s eligibility expansion originally included children, but all children under 100 percent of poverty who are not eligible under traditional Medicaid rules are now covered by SCHIP.
13. Unmet need is calculated only for those respondents who reported that their child needed a service. This results in relatively higher levels of unmet need compared with statistics that are based on the entire sample, regardless of whether a child needed a service. The disparity between the two methods of calculating unmet need is largest for services such as specialty care and mental health services that are needed by smaller percentages of children.
14. M.S. Marquis and K. Kapur, "Employment Transitions and Continuity of Health Insurance: Implications for Premium Assistance Programs," Health Affairs 22, no. 5 (2003): 198–209.[Abstract/Free Full Text]

                      What's this?