Examining Substitution: State Strategies to Limit "Crowd Out" in the Era of Children's Health Insurance Expansions

Publication Date

Dec 8, 1997

By Anna Fallieras, Mary Jo O'Brien, Susanna Ginsburg, and Amy Westpfahl of the The Lewin Group for Office of Health Policy of the Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Health and Human Services, December 9, 1997.

The Project Officers for ASPE were Cheryl Austein and Caroline Taplin

Executive Summary

In recent years, state initiatives to provide health insurance for low-income children have raised a new set of concerns regarding the actual and potential impact of employer- and individual-based substitution. As a number of states elected to expand Medicaid in the late 1980s, researchers began to explore the possibility that Medicaid was substituting for private coverage. By 1992, substitution became a more salient concern as states expanded coverage to all pregnant women and children up to age 6 with incomes up to 133% of the federal poverty level (FPL), and more than half of the states elected to receive federal matching funds to cover all pregnant women and infants up to 185% FPL. As Federal and state governments seek to increase the proportion of people with health coverage, it is critical to understand the nature and extent of substitution. Concerns focus on the potential that if Medicaid and other public expansions are responsible for shifting individuals from employer-sponsored insurance to public programs, the effectiveness of public funds to expand insurance coverage might be limited. As programs minimize the effects of substitution, states will then have the ability to target funding specifically for insuring children who do not have access to affordable health coverage.

Enabled under Title XXI of the Social Security Act, the State Children’s Health Insurance Program (CHIP) is providing $24 billion in funding to states over a five-year period to provide health insurance to uninsured children. The purpose of the law is to assist states in initiating and expanding children’s health assistance programs to uninsured, low-income children. This assistance can be provided through one of three methods: (a) a program to initiate and expand the provision of health care assistance via a separate State Insurance Program; (b) a Medicaid expansion; or (c) a combination of these methods. Eligibles include children under age 19 not eligible for Medicaid with family incomes below 200% of the federal poverty level or 50% above the current state Medicaid limit. State expansions could begin as early as October 1, 1997.

The primary objective of Title XXI, to insure uninsured children, charges states to expand the number of eligible children with health insurance beyond current Medicaid eligibility limits. As research has suggested that substitution is most commonly seen in higher income ranges, substitution of state-funded programs for private insurance coverage has become a substantial concern. Further complicating the issue, families are making difficult decisions which may directly impact individually-based substitution. For example, families may opt to substitute a separate state program for private insurance or for Medicaid. It is important to understand that family decisions are based upon affordability of plans, comparability of the benefit packages, and health status of children. State and national research has not produced conclusive evidence that employers are dropping coverage, yet there is evidence that employer-based coverage is declining. Several states have suggested that the coverage issue, in most cases, is not employers dropping coverage, rather individuals are choosing to opt out of private insurance coverage for state subsidized programs.

Research has attempted to assess the amount of substitution, suggested ways to limit its effects, and weighed the benefits and costs of expanding Medicaid with the potential of crowding out private insurance. Limited data and the complexity of the issue make it difficult to accurately measure the crowd out of public programs for employer-sponsored insurance. Actual substitution is complicated by secular trends including the following: the decline in employer-sponsored insurance; increasing levels of employee premium contributions; the decrease of unionization; and the shift from manufacturing to lower wage service industries. The complexity of measuring substitution and limited state-specific data contributes to conflicting estimates of substitution and makes it challenging for State governments to address substitution in the design and implementation of their children’s health insurance programs. It has also been difficult to apply research focusing on Medicaid expansions to other models of providing low-income children with health coverage. With few estimates indicating the degree of actual substitution and little data on the demographic profile of the "substituters," the discussion on substitution has been primarily based on the perceived effect of crowding out private coverage.

There are two ways that substitution can occur under CHIP. First, there can be substitution of coverage which can occur when families or employers drop existing private coverage to enroll in CHIP. Second, there can be substitution of premiums which can occur when employers reduce their contribution levels of dependent coverage because families are able to obtain subsidies that offset the reduced amount of contributions. To assist states as they address issues related to substitution in their Title XXI plans, this paper focuses primarily on an examination of the experiences of a set of states that have already developed and implemented children’s health insurance programs prior to the creation of Title XXI. Qualitative data, which was collected from state officials through interviews and a round-table discussion, augments previous research by examining states’ perceptions and their actual experiences with substitution.

State Mechanisms To Limit Substitution

State strategies to address substitution include nine mechanisms that either purposefully or inadvertently limit two types of substitution: individual-based and employer-based. Because different dynamics drive employee and employer substitution, states have implemented specific mechanisms to address these two types of substitution. For the most part, state officials agreed that the primary concern is whether, and to what extent, previously insured workers are dropping their employer-sponsored coverage. Substitution of coverage seems to be primarily driven by individual choices rather than employers strategically eliminating coverage. Although states have distinguished between individual- and employer-based substitution and have established mechanisms focused on limiting both types of crowd out, states and researchers also have identified the overlap between the two. Substitution is a result of the dynamic relationship between the types of benefits employers want to provide to employees and the willingness of employees to participate in the programs offered. Both employers, in offering coverage, and employees, in taking up coverage, are forced to make complex decisions which are driven by complicated issues such as secular trends in the economy and family dynamics.

The extent to which states deliberately institute mechanisms limiting substitution varies by program and state. In some cases, substitution appears to be restricted indirectly through mechanisms established in program designs for other purposes. For example, many states require copayments for services rendered which mirror cost-sharing in the private market. Although required copayments are customarily included in program design for financial purposes, when comparable to private market copayments they limit the number of individuals who drop private insurance to enroll in state-subsidized programs.

Mechanisms limiting individual based substitution. State mechanisms to limit individual-based substitution primarily target those dynamics that prompt family decisions to opt out of private coverage such as: the cost comparisons of copayments and premiums; the affordability of private coverage; and the comprehensiveness of benefits. The majority of states’ efforts to curb substitution has focused on the dynamics that drive families to opt out since individual-based substitution is a greater concern than employer-based substitution. The primary mechanisms utilized by states include the following: (1) evaluating affordability of private coverage; (2) requiring periods of uninsurance; (3) providing subsidies; and (4) limiting the scope of benefit packages. These mechanisms are effective ways to curb substitution by making state-sponsored children’s health insurance comparable to employer-sponsored coverage.

Mechanism 1: Identifying the Affordability of Private Coverage: Setting Premiums and Copayments That Do Not Deter Utilization or Enrollment

Setting Premiums. While there are no data that directly link the establishment of premiums as a deterrent to families opting out of employer-based coverage, several states anticipate that this is the case. New York specifically identified that while increases in monthly premiums were not established as a mechanism to discourage substitution, it is anticipated that newly established increases will ultimately result in reducing individual-based crowd out. The state has identified that a $9 to $13 monthly premium per child should discourage families from dropping employer coverage to enroll in the New York Child Health Plus Program. Florida’s Healthy Kids program has set premiums that range from $5-10 per child per month for families below 130% FPL and $13-27 per child per month for families from 131-185% FPL. Families pay 100% of premium above 185% FPL. Florida experimented with raising premiums beyond ten dollars, but found these amounts directly resulted in decreased enrollment. Nevertheless, state program representatives felt that premiums at a lower level (between $5-10) were a viable means of evoking a sense of responsibility in participants by making public insurance comparable to employer-sponsored coverage.

Mechanism 2: Using Periods of Uninsurance, Access to Employer-Sponsored Insurance, and Employer Contributions

Several states have established requirements based on access to employer-sponsored insurance, employer contributions, and/or specific periods of uninsurance as a means to limit substitution. MinnesotaCare has instituted relatively stringent time and cost requirements, described by some as "firewalls" in order to prevent the crowd out of employer-sponsored insurance. For example, MinnesotaCare requires that prior to enrollment in the program, families may not have access to employer-sponsored coverage in which an employer subsidizes 50% or more of the cost of the policy for at least eighteen months preceding application to the program. Exemptions to this include children under 150% FPL and situations where a parent in a family over 150% FPL becomes unemployed, does not qualify for workers compensation, and is unable to pay for COBRA.

Mechanism 3: Use of Subsidies

Some states are designing their programs to work in tandem with employer-sponsored coverage as a means of limiting substitution. Assisting families to purchase private coverage with public funds may deter families from dropping, or not pursuing employer-sponsored insurance beyond their economic reach. While subsidies support families in purchasing insurance in the private market, they may also contribute to substitution. Oregon's 1997-1999 budget proposed the creation of the Family Health Insurance Assistance Program (FHIAP), an insurance subsidy program for families with incomes between 100% and 185% of poverty. Essentially, FHIAP will establish a voucher system for families with access to employer-sponsored insurance, assisting eligible families in buying into employer-sponsored programs. Colorado has also considered subsidizing families’ insurance as a viable means of limiting substitution and is considering the implementation of a voucher program. Colorado has identified this strategy as an approach that will prompt the public and private sector to cooperate in the process to insure children. Florida is also considering a type of voucher mechanism that will assist families in purchasing coverage as a supplement to Healthy Kids and other efforts.

Mechanism 4:Limiting the Scope of the Benefit Package

Three types of mechanisms are being used by states to enhance the affordability of employer-sponsored insurance: While there is no evidence to suggest that employers are strategically eliminating health coverage so that their workers may receive public insurance, several states have been proactive in designing mechanisms to encourage the availability of affordable employer-sponsored insurance.

Mechanism 5: Purchasing Cooperatives

Health insurance purchasing cooperatives, which assist employers in purchasing insurance at more affordable rates, are enabling small employers to offer benefits in Oregon and Rhode Island. The State supported Associated Oregon Industries’ (AOI) efforts to develop a privately sponsored Health Insurance Purchasing Cooperative (HIPC) targeted at small group employers. AOI, Oregon’s largest business organization, shows promise for assisting small employers in maintaining existing or establishing new insurance programs to employees. Rhode Island has four small business purchasing cooperatives that may be deterring crowd out by making the provision of insurance by small businesses affordable.

Mechanism 6: Employer Buy-Ins

Mechanism 7: Reimbursements To Small Businesses

Assisting employers through reimbursement plans is another strategy geared toward helping small employers purchase health coverage. Massachusetts has planned the implementation of an experimental program in the year 1999, entitled the "Insurance Reimbursement Plan" to be administered through the Medical Assistance program. The Plan will be structured to reimburse small businesses (i.e., employers with less than 50 workers) that cover in excess of 50% of the cost of insurance for their employees. Eligible employers will be annually reimbursed $1000 for family coverage, $800 for couples coverage, and $500 for individual coverage. This option, not yet implemented, has been considered quite controversial as it would be paying employers to provide insurance, which in some cases, they may already offer.

Disconnect Between State Experiences and Research

While research has yielded varying estimates of crowding out private coverage, ranging from 0-50%, states providing coverage for low-income children indicate they have seen little evidence of these levels of substitution. Although the majority of these states have not collected sufficient data to determine an accurate estimate of crowd-out, states’ anecdotal evidence pointed to minimal numbers of enrollees with access to affordable private coverage and did not seem to warrant further investigations. States queried about their experience with substitution noted that despite some researchers’ predictions about employers strategically eliminating or reducing their contribution to employee health benefits, real or potential crowd out is primarily the result of families opting to discontinue private insurance for financial reasons.

States that have collected information on the number of program participants with previous insurance, revealed estimates that were lower than estimates derived from nationally-based research.

· Preliminary data from Colorado’s Child Health Plan estimate that 0.007% dropped employer coverage to enroll in the plan. In addition 5% of their enrollees carry other insurance coverage.

· Florida’s Healthy Kids program estimates that 94% of their children had been uninsured for 6-12 months before enrolling in Healthy Kids.

· New York found that more than 50% of all Child Health Plus enrollees had no insurance at all during their lifetimes; approximately one-sixth had Medicaid prior to Child Health Plus; and approximately one-sixth were underinsured or fully insured.

MinnesotaCare, which began as a state sponsored program in 1992 and switched to a Medicaid waiver program expansion in 1995, is one of the few programs that has evaluated their participants' access to private coverage. MinnesotaCare, offering coverage to families with children under 275% of the federal poverty level, noted that approximately 13% of their enrollees technically had access to other health coverage: 7.1% reported giving up private market insurance and 5.7% of the sample reported dropping coverage from another public program to enroll in MinnesotaCare. Of note, the majority of enrollees who had access to private insurance said they could not afford it.

Implications For Future Expansions Under Title XXI of the Social Security Act

Current research coupled with past experiences of states provides some information for states to consider as they plan and implement programs funded by Title XXI. Four major program design issues were identified as important areas in addressing substitution: the overall design of the program; determining whether to use Medicaid expansions or separate state programs; the development of specific provisions related to employer coverage; and the oversight and evaluation of the expanded program.

One of the most difficult issues facing states is how to design a program that neither limits access to new coverage, nor replaces existing insurance coverage. Among the many programmatic areas that need to be addressed are: the determination of an appropriate benefit package; the implementation of suitable cost-sharing measures; the development of an administratively simple, yet stringent enrollment process; the identification of appropriate eligibility criteria that will encourage enrollment, while limiting substitution; and strategies focused on collaboration with the private market.

As state programs to insure children have evolved, such expansions have raised a new set of concerns about the effectiveness of these state programs in enrolling only their targeted populations. As programs have expanded into higher income ranges, policy makers, program administrators, and researchers have expressed a concern about the potential that these programs have begun to substitute for employer-sponsored insurance. In an era of limited public resources, the erosion of employer-sponsored insurance coverage has taken center stage in the debate over new or expanded state health insurance programs at the federal level and in some states.

States' strategic choices to be evaluated in the light of federal program goals include: (a) reducing the number of uninsured children over time; (b) increasing the participation of children in public programs for which they are eligible; (c) minimizing the potential to cover children who would have otherwise had private insurance coverage; and (d) creating a private/public partnership which increases coverage for children in the private market. Given the goals of the federal program, states need to determine not only how they will develop strategies to limit substitution, but also how they will evaluate the cost of enforcing enrollment barriers of potential eligibles and the overall impact on administrative complexity.

I. Introduction

A. State initiatives to insure children have yielded a new set of concerns regarding the substitution of public coverage for private health benefits.

Enabled under Title XXI of the Social Security Act, the State Children's Health Insurance Program (CHIP) is providing $24 billion in funding to states over a five-year period to provide health insurance to uninsured children. The purpose of the law is to assist states in initiating and expanding children's health assistance programs to uninsured, low-income children. This assistance can be provided through one of three methods: (a) a program to initiate and expand the provision of health care assistance via a separate State Insurance Program; (b) a Medicaid expansion; or (c) a combination of these methods. Eligibles include children under age 19 not eligible for Medicaid with family incomes below 200% of the federal poverty level or 50% above the current state Medicaid limit. State expansions could begin as early as October 1, 1997. Each state will receive a portion of the total amount, and such allotments shall remain available for up to three years.

B. Due to limited data and the complexity of the issue, the discussion of substitution has been based upon the perceivedeffect of crowding out private coverage.

Research has attempted to assess the amount of substitution, suggested ways to limit its effects, and weighed the benefits and costs of expanding Medicaid with the potential of crowding out private insurance. Limited data and the complexity of the issue make it difficult to accurately measure the crowd out of public programs for employer-sponsored insurance. Actual substitution is complicated by secular trends including the following: the decline in employer-sponsored insurance; increasing levels of employee premium contributions; the decrease of unionization; and the shift from manufacturing to lower wage service industries. The complexity of measuring substitution and limited state-specific data contributes to conflicting estimates of substitution making it challenging for State governments to address substitution in the design and implementation of their children's health insurance programs. It has also been difficult to apply research focusing on Medicaid expansions to other models of providing low-income children with health coverage. With few conclusive estimates indicating the degree of actual substitution and very little data on the demographic profile of the "substituters," the discussion on substitution has been primarily based on the perceived effect of crowding out private coverage.

C. Structure of the Report

To assist states as they address the issue of substitution in their Title XXI plans, this paper focuses primarily on an examination of the experiences of a set of states that have already developed and implemented children's health insurance programs prior to the creation of Title XXI. Qualitative data, which was collected from state officials through interviews and a round-table discussion, augments previous research by examining states' perceptions and their actual experiences with substitution.

This report is organized within the following structure. Section I provides the background and purpose of the paper, a brief synopsis of Title XXI of the Social Security Act, and a discussion on individual- and employer-based substitution. Section II contains the methodology and an introduction to the literature. Section III defines substitution and provides a review of the literature including the range of substitution estimates. Section IV describes state mechanisms to limit substitution. Section V discusses estimating substitution, the disconnect between state experiences and research findings, and the impact of secular trends on measuring crowd out. Section VI presents implications for future expansions under Title XXI; evaluation of the nature, scope, and levels of substitution; summary types of data to be collected by states; and the need for additional research. Section VII presents a summary of the findings.

II. Methodology

A. Interviews of selected state children's health program, Medicaid, and MCH directors were conducted in order to obtain specific program information.

The Lewin Group completed in-depth interviews of children's health insurance program directors, Medicaid directors, Maternal and Child Health directors, and related individuals in nine states selected for this study. The nine states were chosen for their experience with children's health programs as well as for geographic diversity. These interviews provided background information on state programs, including state experiences with substitution and states' future issues and concerns. Information collected through the interview process has been organized in a database created for the purpose of collecting children's health insurance program, administrative, and legislative information. In addition, eight states, including representatives of the study states and several others who either seriously considered the issues surrounding substitution or had implemented mechanisms to limit substitution, were invited to participate in a roundtable discussion which also included researchers and selected federal staff.

B. The September 9th roundtable discussion was designed to identify state experiences with and future concerns regarding substitution, especially as it impacts expansions under the provisions of Title XXI.

The purpose of the September 9th discussion entitled "A Roundtable Discussion on State Experiences Regarding Crowd Out and Children's Health Insurance" was to: (1) discuss concerns of states that have been considering and addressing the issues of real and/or potential substitution effects in their efforts to expand health insurance for children; (2) consider the implications of research in the context of such state experiences; and (3) examine and explore these state experiences in order to identify issues and implications for future changes under Title XXI of the Social Security Act.

The following set of questions represents the issues discussed on September 9th. These questions were designed to address the complexity of the issues including those factors that might mediate substitution. The questions were divided into the following four categories:

Defining Substitution
· What are the real program concerns regarding substitution?

· What are the political concerns?

· What is the real problem? (e.g.,, "crowd out," "push out," or "opt out")
Financing and Market Issues
· Is the problem grounded in the concern that programs will not be able to pay for all of the children who have decided to enroll?

· Is the concern that employer-based coverage is going to erode in the market as other state options are becoming available to the public?

· Is there really a different problem between employer-sponsored insurance and individual coverage?
Determining Demographics
· Will programs have the ability to assist children already enrolled in private insurance programs in such a way that it does not replace existing coverage?

· Is substitution really occurring among individuals who have appropriate access to employer-sponsored coverage? (e.g., individuals who have access to affordable family coverage)
Considering Other Concerns
Will this be viewed as a transition program or as a permanent replacement of insurance for eligible children?

The information collected through initial interviews and the September 9th roundtable discussion are the basis for this qualitative analysis which focuses on accurately reporting both state experiences with substitution and presenting additional issues that will need further exploration. This analysis identifies issues states are facing as they plan and implement programs to cover uninsured children at higher income levels. Moreover, this study points out the lack of data surrounding substitution, which limits state and federal officials' ability to evaluate possible solutions. Researchers and federal representatives participating in the roundtable discussion provided the context in which these concerns might be addressed (e.g., the collection of appropriate data, enrollment processes, and market-based issues, etc.).

III. Review of the research on substitution

A number of national research organizations and academic researchers have examined the issue of substitution utilizing the Current Population Survey (CPS) and the Survey of Income and Program Participation (SIPP) data. Such studies have produced varying estimates of substitution and have generated discussion regarding appropriate methodologies in order to estimate accurate levels of substitution. The following section briefly discusses definitions of substitution, estimations of substitution based on existing literature, the lack of consensus in estimating substitution, and the usefulness and generalizability of the current research estimates.

The paper focuses on what has been learned from states' prior experiences with substitution, not on a comprehensive review of the literature. Therefore, we have provided an overview of recent literature that has identified the causes and estimations of substitution based on national data sets.

In addition, the paper examines state research focused on estimating levels of and identifying reasons for substitution within state programs. The data cited are population-based and not organization-focused. While substitution literature to date may not have focused on the economic incentives behind employers' decisions to sponsor health insurance and other benefits, there is a considerable amount of literature that separately examines these issues.

A. Substitution occurs when individuals substitute public benefits for private sector benefits, and it may be defined as (1) individual-based or (2) employer-based.

What is the substitution effect or "crowd out?" The substitution effect occurs when individuals substitute free or reduced-price public benefits for private sector benefits. Substitution can be viewed from two different perspectives. The first, individual-based substitution, or "opt out," is an estimate of the proportion of individuals who choose a government-subsidized program instead of selecting employer-sponsored coverage. The second, employer-based substitution, or "push out," is the effect of employers reducing or eliminating health insurance coverage to workers and their families as part of their employee compensation package with the expectation that a public program will provide needed coverage.

B. There is a range of substitution estimates reported in the literature.

Previous research has examined substitution in the context of Medicaid expansions. Although most researchers agree that substitution occurs at some level in all social insurance programs, it may be difficult to accurately predict future levels of substitution for new state children's health insurance programs based on the current, disparate research findings. The variation of substitution estimates based on current literature has yielded values ranging from 0 to 50%. Thus, the range itself indicates the complexity in estimating substitution with any level of precision.

There have been debates in the research community regarding the appropriate methodology for measuring substitution. Although two of the research teams (David Cutler and Jonathan Gruber and Lisa Dubay and Genevieve Kenney) used Current Population Survey (CPS) data, their approaches differed considerably. Cutler-Gruber focused on the question: what is the reduction in private insurance coverage as a share of the persons who enrolled in Medicaid directly as a result of the expansions? Dubay-Kenney, on the other hand, posed the question: What is the reduction in private insurance coverage as a share of the total increase in Medicaid coverage over this period?

The Cutler-Gruber study examined data from 1987 through 1992 of women of childbearing age (15-44) and children made newly eligible via Medicaid expansions. Cutler-Gruber looked across states for differences in Medicaid eligibility and public and private insurance, controlling for economic and demographic factors of the population in each state. They attempted to separate the substitution effect from secular trends in employer coverage, but admitted that there is no definitive way to isolate one from the other. Based on regression analysis, Cutler-Gruber found that the decline in private insurance was approximately 50 % of the increase in Medicaid coverage induced by expansions. This study identified that states with a greater increase in Medicaid eligibility had larger declines in private coverage when compared to states with smaller increases in Medicaid eligibility. The authors considered this to be strong evidence exemplifying substitution. The authors additionally noted in their analysis that opt out occurred largely due to the fact that employees enrolled in employer-sponsored insurance less frequently; however, employers may have encouraged this behavior among their employees by contributing less for their insurance. Therefore, Cutler-Gruber make the important point that the issue of substitution is not merely opt out or push out, but perhaps a complex interaction between both forms of substitution.

Another commonly cited study was conducted by Dubay and Kenney. This study examined the occurrence of substitution as Medicaid expanded into higher income ranges by focusing on poor (incomes below 100% FPL) and near-poor (incomes of 100-133% FPL) populations. Their research identified virtually no substitution of employer-sponsored insurance among poor pregnant women, but approximately 8.5% of poor children under age 11 were substituting coverage. Among near-poor pregnant women and children, the rate of substitution was more pronounced. Dubay and Kenney estimated a substitution rate of 45% for pregnant women and 21% for children under age 11 in the near-poor category. Though these findings suggest the effects of substitution, Dubay and Kenney cited secular changes in the insurance market as additional contributors that may have caused a portion of this effect. Overall, Dubay and Kenney estimated that 14% of the increase in Medicaid coverage for pregnant women and 12% of the increase in enrollment for children under age 11 could be attributed to substitution of private insurance. Both Dubay and Kenney and Cutler and Gruber found substitution to be less of an issue among children than among pregnant women. A study conducted by Lara Shore-Sheppard at the University of Pittsburgh additionally supported the work of Cutler-Gruber and Dubay-Kenney. Shore-Sheppard also utilized CPS data to conclude that declines in private coverage were unlikely to have occurred as a result of recent Medicaid expansions.

A July 1997 study conducted by The Lewin Group developed estimates of substitution consistent with those found by Dubay and Kenney. This analysis used two separate databases, the CPS and Survey of Income and Program Participation (SIPP), and identified that between 20.3% and 24.2% of children enrolled in Medicaid under eligibility expansions could be substituting Medicaid for available employer-sponsored coverage. This study also indicated that rates of substitution are highest among children with incomes above 100% of poverty. SIPP data (1990) were also used by Linda Blumberg, Lisa Dubay and Stephen Norton of The Urban Institute to examine the movement from private coverage to Medicaid under Medicaid expansions. This analysis supports the previous finding, that with respect to children, the displacement of private insurance coverage for Medicaid as a result of recent expansions accounted for only a small portion of enrollment into the Medicaid program.

Researchers have identified the concern of using population-based surveys to determine accurate levels of substitution in state children's health insurance programs. For example, the difficulty in interpreting CPS data to estimate the extent of substitution has been acknowledged by many researchers. Therefore, estimates of the uninsured may reflect only the number of individuals without insurance for the entire year, rather than portraying the true transitory nature of health insurance experienced by many low-income families. Of additional concern is that respondents may report health insurance coverage at the time of interview, rather than their experience throughout the previous calendar year. Studies which have utilized other population-based surveys such as SIPP have yielded comparable conclusions, yet are concerned with similar interpretation and data accuracy issues.

While research identifies a wide range of substitution estimates, it is important to note the vast majority of researchers have identified that there is some existence of substitution within state programs. However, the main weakness of current literature is its limited practical relevance to state programs and their experiences. For example, the definition of substitution used by Cutler and Gruber, Dubay and Kenney, and Shore-Sheppard does not include any concept of affordability, an issue that is of major concern to state policymakers. The nature of the research data required researchers to define substitution as any reduction in private coverage consequent with an increase in Medicaid coverage while controlling for secular trends. The data used did not allow for examination of the burden of private insurance premiums and out-of-pocket payments as a share of family income on the newly eligible, who are by definition low income. Therefore, by definition, any family that refuses an offer of health insurance from an employer, regardless of the expense, has "opted out" of the private health insurance market. State administrators have specifically operationalized the notion of individually-based substitution as the refusal of an offer of "affordable" private insurance. Some states have defined this as an employer subsidy of at least 50% of the total premium (e.g., Minnesota).

C. National research informed the policy debate and raised important issues, yet it has not been focused on addressing state concerns with substitution.

Debates in the 105_th Congress regarding the numerous legislative efforts to insure children were focused on extending health insurance to uninsured children without replacing existing insurance coverage. Substitution was viewed as an inherent concern in the debates, as it was an issue in prior congressional expansions of social insurance programs. Based on existing research which identified substitution effects within the Medicaid expansions of the past decade, legislators and federal officials required that states address the potential of substitution in their plans to expand children's health insurance programs under Title XXI. National research identified the existence of substitution, however it did not necessarily reflect the specific experiences of states that had expanded coverage for children. The overall lack of data and inconsistencies in existing data have contributed to this disconnect between research findings and state experiences. It is important to focus on state experiences in order to better determine the impact of substitution in state expansion efforts under Title XXI. However, it may be premature to reach any conclusions regarding crowd out given the small size of existing programs and the limited period of actual implementation. An accurate assessment of the impact of substitution is likely to require a longer-term analysis.

IV. State Mechanisms To Limit Substitution

This section reviews the nine mechanisms states identified as strategies to address substitution. Exhibit 1 exemplifies the range of state mechanisms used to limit substitution. States investigated various mechanisms that either purposefully or inadvertently limited substitution. States have also emphasized the importance of distinguishing between two types of substitution: individual-based and employer-based. Because different dynamics drive employee and employer substitution, states have implemented specific mechanisms to address these two types of substitution. For the most part, state officials agreed that the primary concern is whether, and to what extent, previously insured workers are dropping their employer-sponsored coverage. Substitution of coverage seems to be primarily driven by individual choices rather than employers strategically eliminating coverage. Although states have distinguished between individual- and employer-based substitution and have established mechanisms focused on limiting both types of crowd out, states and researchers also have identified the overlap between the two. Substitution is a result of the dynamic relationship between the types of benefits employers want to provide to employees and the willingness of employees to participate in the programs offered. Both employers, in offering coverage, and employees, in taking up coverage, are forced to make complex decisions which are driven by complicated issues such as secular trends in the economy and family dynamics.

The information represented in this document is focused on the experiences of nine states that were either interviewed and/or attended the roundtable discussion. Most states represented in this paper identified the issue of individual-based substitution as a major concern. Other states, either in their current initiatives or future initiatives under Title XXI, may be even more concerned about substitution in its various forms as they cover higher income children under Title XXI.

A. States have developed a variety of strategies to limit individual-based substitution.

State mechanisms to limit individual-based substitution primarily target those dynamics that prompt family decisions to opt out of private coverage such as: the cost comparisons of copayments and premiums; the affordability of private coverage; and the comprehensiveness of benefits. The majority of states’ efforts to curb substitution has focused on the dynamics that drive families to opt out since individual-based substitution is a greater concern than employer-based substitution. The primary mechanisms utilized by states include the following: (1) evaluating affordability of private coverage; (2) requiring periods of uninsurance; (3) providing subsidies; and (4) limiting the scope of benefit packages. These mechanisms are effective ways to curb substitution by making state-sponsored children’s health insurance comparable to employer-sponsored coverage.

Exhibit 1: Mechanisms to Limit Individuals from Substituting Private Coverage
STATE PROGRAM	MECHANISMS TO LIMIT INDIVIDUALS FROM SUBSTITUTING PRIVATE COVERAGE
	Evaluating Affordability of Private Coverage		Periods of Uninsurance	Providing Subsidies	Limiting Scope of Benefit Package
	Increasing Premiums	Redefining Copayments
CaliforniaKids					X
Children’s Medical Security Plan (MA)	X			Under consideration^g	X
Colorado Child Health Plan			To be implemented^e	Under consideration^f
Florida Healthy Kids Corporation		X	X^a	Under consideration
MinnesotaCare			X		X^j
NY Child Health Plus	X^b	X^c			X^d
Oregon Health Plan			To be implemented^h	To be implementedⁱ
RiteCare		X
Washington Basic Health Plan Plus

Mechanism 1: Identifying the Affordability of Private Coverage: Setting Premiums and Copayments That Do Not Deter Utilization or Enrollment

Several states have identified the affordability of employer-sponsored insurance as an important consideration in the implementation of mechanisms to limit crowd out. Affordability is usually defined as the percentage of family income that is believed families could or are willing to use to purchase health care. Although states are interested in the affordability of private coverage, there is a lack of research that has identified the basis upon which to determine the affordability of private insurance coverage. A recent study conducted by economists at the Agency for Health Care Policy and Research identified that high employee contribution rates may discourage employees from participating in employer-sponsored plans while rising premiums may also discourage employers from offering coverage. Data from a MinnesotaCare evaluation found that of individuals questioned in 1995, approximately 73% of those who were uninsured stated that they could not afford to purchase insurance. The overall lack of data has led states to experiment with specific strategies to determine effective cost sharing levels in their children's health insurance programs. In developing strategies, states have considered the tradeoffs between encouraging participation and limiting substitution. Specifically, states have identified the importance of establishing cost sharing (e.g., premiums and copayments) that is low enough to encourage participation, yet high enough to limit substitution.

States have also struggled with the issue of identifying individuals who should be limited from entering state programs based solely on their ability to afford employer-sponsored insurance. For example, Rhode Island’s RIteCare queries applicants on their access to and affordability of coverage for individuals above 185% FPL. In such instances, RIteCare has defined affordable insurance premiums as less than $150/month for an individual or $300/month for families. RIteCare does not formally verify this information, and the verification process is based on the "honor system." Although determining affordability is essential in the effort to develop a balance between increasing enrollment and decreasing substitution, elaborate verification processes have been identified as costly, administratively burdensome, and difficult for state programs to conduct.

Moreover, states have experimented with the balance between limiting substitution without discouraging enrollment and participation through the use of premiums and copayments. Through this experimentation, states have been able to identify general threshold levels for cost sharing in their programs.

Defining Copayments. Many existing children’s health insurance programs have established copayments for physician visits, prescriptions, and emergency room use as mechanisms to mirror the requirements of private coverage. The initial intent behind establishing copayments varied by program. The primary rationales state officials expressed were to offset costs and to instill a sense of responsibility in participants. Some states believe that price sensitivity was a critical aspect of determining copayments and cost-sharing policies. Florida and Rhode Island program officials identified the importance of establishing copayments that do not deter eligible families from enrolling in state programs and utilizing necessary services, yet do not encourage the inappropriate use of services. In Florida, all families pay copayments for certain services at the time they are rendered. While there is no copayment for wellness checks and immunizations, there are copayments for regular doctor office visits, ER, glasses, outpatient behavioral health, etc. Copayments generally range from $3-5 except for ER which is $25 if the use is inappropriate. States believed that families would be less likely to opt out of private coverage if copayments were similar for both public and employer-sponsored insurance. An additional issue regarding the collection of copayments is the administrative burden that will be placed either on the program or its providers to collect a minimal amount of funds. Many programs have shifted the burden of collecting copayments to the providers in order to reduce the administrative costs to the program. However, it has been suggested that, in many cases, providers were either unable or unwilling to collect copayments from beneficiaries who were incapable or unwilling to pay for services rendered. This becomes a concern for providers who collect copayments from patients as a portion of their reimbursement under the program. For these providers, the inability to collect copayments may result in an overall decrease in their total reimbursement by the program.

States emphasized the "the concept of price sensitivity" in determining cost sharing: premiums and copayments must be set at levels considered reasonable and affordable to potential program beneficiaries, yet substantial enough to not encourage individuals from dropping private coverage. Determining this balance between setting cost sharing levels that limit substitution and not limiting enrollment is very difficult to predict and achieve. Although there is limited research on the price sensitivity of low income workers, one might suspect that it is an important issue. Nevertheless, the pressing question is to what extent do states discourage the really poor from enrolling in programs in order to prevent the less poor from dropping private coverage.

Mechanism 2: Using Periods of Uninsurance, Access to Employer-Sponsored Insurance, and Employer Contributions

Several states have established requirements based on access to employer-sponsored insurance, employer contributions, and/or specific periods of uninsurance as a means to limit substitution. MinnesotaCare has instituted relatively stringent time and cost requirements, described by some as "firewalls" in order to prevent the crowd out of employer-sponsored insurance. For example, MinnesotaCare requires that prior to enrollment in the program, families may not have access to employer-sponsored coverage in which an employer subsidizes 50% or more of the cost of the policy for at least eighteen months preceding application to the program. However, some exemptions apply. All children under 150% FPL are exempt from this requirement. Children in families over 150% FPL are exempt from this requirement if a parent becomes unemployed and loses access to employer sponsored coverage.

Additional states have instituted policies that base eligibility on an inability to access employer-sponsored insurance in accordance with minimum employer contribution levels; however, the minimum employer contribution and rates of access vary across programs and states. Overall, states have suggested that eligibility based on access to employer-sponsored insurance is an effective requirement that limits substitution. For example, Florida has found that most of their enrollees are either self-employed or employed in industries where insurance is not available, thus reducing the probability of individuals entering the program for the purpose of substituting for employer-based insurance.

Three states have established periods of uninsurance, whereby individuals enrolling in programs would by definition already be uninsured. Florida initially implemented a six-month period of uninsurance for children whose applications indicated that they had access to employer coverage. However, this restriction was eliminated after hearing testimony from families and identifying evidence that families who otherwise lost their insurance through no fault of their own would possibly postpone seeking services during the six month period. This might result in an increased pent up demand upon their initial enrollment and possibly a more costly health problem for health plan partners to treat. Colorado is evaluating what the "most appropriate" period of uninsurance to limit substitution would be without hampering timely access to coverage. Currently, Colorado requires a three-month period without insurance coverage. In addition, some states have identified underinsurance as an eligibility requirement. MinnesotaCare has liberally defined underinsurance as any child below 150% FPL who lacks both dental and vision benefits; therefore, such children meeting other eligibility criteria may be granted access to the program.

With respect to uninsurance periods, waiting lists have been found to inadvertently limit substitution. Few states have waiting lists, yet for those that do, individuals with employer-sponsored insurance will be deterred from dropping existing coverage in order to be placed on a waiting list. Although this has occurred naturally, states have not purposefully instituted waiting lists as a means to control substitution.

The process of verification of access to employer-sponsored insurance, employer contributions, and periods of uninsurance has been challenging for states. Most states verify eligibility based on self disclosure or the "honor system," as other means of verification are costly and increase the complexity of enrollment processes. Some states have chosen to piggy back on other existing state programs in order to efficiently evaluate eligibility status. The most commonly cited example has been the process used by the Florida Healthy Kids Corporation in which eligible children are required to qualify for the National School Lunch Program. Other states have chosen to utilize existing verification structures of state Medicaid programs to assist in the determination of eligible enrollees. MinnesotaCare has utilized an audit procedure by which enrollees are retrospectively surveyed to determine if they had access at the time of enrollment. In addition, if an individual applies for MinnesotaCare and the program believes that their employer provides insurance, the enrollment office sends an "employer report form" to the employer to verify that no coverage is offered. Anecdotal evidence provided by MinnesotaCare has suggested that most individuals are not purposefully substituting, but rather do not realize they have access to coverage with an employer contribution of at least 50%.

Mechanism 3: Use of Subsidies

Some states are designing their programs to work in tandem with employer-sponsored coverage as a means of limiting substitution. Assisting families to purchase private coverage with public funds may deter families from dropping, or not pursuing employer-sponsored insurance beyond their economic reach. While subsidies support families in purchasing insurance in the private market, they may also contribute to substitution. Oregon's 1997-1999 budget proposed the creation of the Family Health Insurance Assistance Program (FHIAP), an insurance subsidy program for families with incomes between 100% and 185% of poverty. Essentially, FHIAP will establish a voucher system for families with access to employer-sponsored insurance, assisting eligible families in buying into employer-sponsored programs. In an attempt to maintain administrative simplicity while limiting crowd out, FHIAP will use an automated billing system to limit administrative costs. Colorado has also considered subsidizing families’ insurance as a viable means of limiting substitution and is considering the implementation of a voucher program. Colorado has identified this strategy as an approach that will prompt the public and private sector to cooperate in the process to insure children. Florida is also considering a type of voucher mechanism that will assist families in purchasing coverage as a supplement to Healthy Kids and other efforts.

Mechanism 4: Limiting the Scope of the Benefit Package

Limiting the scope or comprehensiveness of the benefit package is another mechanism that seems to deter families from opting out of employer coverage. The CaliforniaKids program, the Massachusetts Children’s Medical Security Plan, and until recently, New York’s Child Health Plus program, declined to offer inpatient care with the rationale that it was better to provide a more modest benefit package to greater numbers of children. Although CaliforniaKids and Child Health Plus were primarily concerned with stretching their funding to meet the basic health care needs of as many children as possible, representatives from both programs believed that excluding inpatient care deterred substitution. These programs were comfortable with offering a limited benefit package, knowing that Medicaid would serve as a safety net for children requiring inpatient care or extended benefits. Because most employer plans include inpatient care, families were less inclined to drop private coverage. Limited benefit packages also reduce the risk of adverse selection within state-funded programs. Families with children having special health care needs may not select out of private coverage or Medicaid in order to obtain coverage under a state program offering a less generous benefit package. In addition to the benefit package, families with such children may also examine the provider network encompassed by the state program before substituting for their existing coverage. As children with special health care needs must often access specialty services, parents are faced with complex choices which relate not only to cost, but substantially to access of health services. Such considerations may limit adverse selection into state programs that provide limited benefit packages and access to specialty services as compared to private insurance plans and Medicaid.

B. Three types of mechanisms are being used by states to enhance the affordability of employer-sponsored insurance.

While there is no evidence to suggest that employers are strategically eliminating health coverage so that their workers may receive public insurance, several states have been proactive in designing mechanisms to encourage the availability of affordable employer-sponsored insurance. Exhibit 2 exemplifies state mechanisms to limit employers from dropping insurance coverage. Because small businesses, usually defined as less than 50 employees, have difficulty purchasing health insurance at reasonable rates, state efforts to encourage employer-sponsored insurance have specifically targeted small businesses. Oregon, Rhode Island, Washington, Colorado and Massachusetts are examples of states that have either already implemented, or are currently considering, mechanisms to enhance the affordability of employer-sponsored insurance.

Exhibit 2: Mechanisms to Limit Employers from Dropping Insurance Coverage
STATE PROGRAMS	MECHANISMS TO LIMIT EMPLOYERS FROM DROPPING INSURANCE
	Purchasing Cooperatives	Buy-Ins	Reimbursements
CaliforniaKids
Children’s Medical Security Plan (MA)			Under consideration
Colorado Child Health Plan		Under consideration
Florida Healthy Kids Corporation	X^a
MinnesotaCare
NY Child Health Plus
Oregon Health Plan	X
RiteCare	X
Washington Basic Health Plan		X

Mechanism 5: Purchasing Cooperatives

Mechanism 6: Employer Buy-Ins

Allowing employers to directly buy into state programs at reasonable costs is another means of encouraging employers to take responsibility for sponsoring workers' insurance. The Washington Basic Health Plan (BHP) established provisions in which employers can buy into the program. As BHP is generally less expensive than private insurance for most employers due to subsidized premiums for low-income workers, there is a positive incentive for employers to participate. The employer buy-in also provides employers the benefit of the larger purchasing power and risk pool of the state. Employers contribute approximately $45 per employee per month and employees contribute the remainder of the premium based on a sliding fee scale. For low income workers, the $45 employer contribution covers the employee share of the monthly premium . However, if a low income worker enrolls in BHP as an individual, the monthly premium is approximately $10 for that individual. Despite the reasonable cost of employer buy-in, Washington state has found that some employers will offer to "subsidize" employees for the $10 premium required for individual enrollment in BHP, rather than contributing $45 as their employer. This has made the buy-in program challenging, as it is administratively difficult for the program to control. Other states have considered similar programs to increase employer participation. A small employer alliance in Colorado has also expressed an interest in participating in a buy-in program with the Colorado Child Health Plan. The Children's Basic Health Plan is currently examining the possibility of implementing an employer buy-in program.

Mechanism 7: Reimbursements To Small Businesses

V. Estimating the Amount of Substitution

Currently, there are minimal data on the levels of substitution resulting from individual states' efforts in providing children's health insurance. A few states have some initial estimates of the effects on their programs, but important data gaps exist.

A. The comparison of benefit packages and program costs drive parental decisions to substitute public programs for private insurance.

MinnesotaCare determined that of the 7.1% substitution of public for private coverage, 2.8% was employment-based and 4.2% was self-insured, with the majority of respondents reporting that they could not afford the insurance to which they technically had access. Of those sampled in MinnesotaCare's evaluation, 86% applied due to cost, 67% due to benefits, and 61% reported that MinnesotaCare was their only way to acquire health coverage. Both of these surveys exclude small employers and are likely to under-represent employers in low wage service industries, such as food service.
New York's Child Health Plus program determined that of their participants who had previous access to insurance, 10% dropped coverage because their income was too low to afford private insurance yet too high for Medicaid; and 12% dropped coverage because private insurance premiums were too high.
Anecdotal information on substitution from CaliforniaKids found that most families which dropped employer coverage did so because it neither provided necessary services nor was affordable. In addition to considering cost and benefits, family decisions to "opt out" of private coverage are also impacted by secular trends in the job market, insurance industry, and general economy, which will be discussed in later sections.

Qualitative data derived from the September 9th meeting and interviews with state officials revealed limited evidence of employer-based substitution. Available data from employer surveys confirms states' experiences: 1,100 employer plans questioned by the Hay/Higgins survey in 1992 through 1996 indicated that they all provide coverage for dependents; and in the KPMG 1996 survey of employer health plans, all but 1% provide dependents coverage on an optional basis. One state representative commented that fears regarding employer substitution wrongly assume that employers are familiar with state policies and savvy enough to correlate their workers' salaries with eligibility levels of state programs. The Director of the Florida Healthy Kids Program noted that the majority of their participating families are self-employed, employed in service and tourist industries, or employed in small companies with no access to SSI or affordable private coverage. In addition, a recent Florida evaluation found that approximately 38% of parents whose children are enrolled in Healthy Kids are employed part time. Hence, there has been little concern about employers dropping coverage. Likewise, the MinnesotaCare representative theorized that the advent of MinnesotaCare did not cause employers already offering insurance to drop coverage. However, it was possible that MinnesotaCare may be providing new or existing companies who did not offer insurance prior to MinnesotaCare less of an incentive to now offer coverage. In addition, there is concern that although employers may not be dropping coverage altogether, they may be dropping the amount of their contributions toward individual and dependent coverage.

B. There is a disconnect between state experiences and research.

The discrepancy between state experiences and research findings may be attributed to an overall lack of data and inconsistencies in existing data. Information collected from the states reveals a disconnect between state experiences with substitution and research findings. While research has yielded varying estimates of crowding out private coverage, ranging from 0-50%, states providing coverage for low-income children indicate they have seen little evidence of these levels of substitution. Although the majority of these states have not collected sufficient data to determine an accurate estimate of crowd-out, states' anecdotal evidence pointed to minimal numbers of enrollees with access to affordable private coverage and did not seem to warrant further investigations. States queried about their experience with substitution noted that despite some researchers' predictions about employers strategically eliminating or reducing their contribution to employee health benefits, real or potential crowd out is primarily the result of families opting to discontinue private insurance for financial reasons.

States that have collected information on the number of program participants with previous insurance, revealed estimates that were lower than estimates derived from nationally-based research.

Preliminary data from Colorado's Child Health Plan estimate that 0.007% dropped employer coverage to enroll in the plan. In addition 5% of their enrollees carry other insurance coverage.
Florida's Healthy Kids program estimates that 94% of their children had been uninsured for 6-12 months before enrolling in Healthy Kids.
New York found that more than 50% of all Child Health Plus enrollees had no insurance at all during their lifetimes; approximately one-sixth had Medicaid prior to Child Health Plus; and approximately one-sixth were underinsured or fully insured.

C. Substitution may also be linked to secular trends in the job market, insurance industry, and the general state of the economy.

While research has shown varying estimates of substitution, few investigations have examined the motivating factors behind employer- and individual-based decisions regarding health insurance. At least some of the apparent substitution of public coverage may be linked to secular trends in the job market, insurance industry and the general economy. These secular trends may affect employer and individual decisions which, in turn, may affect rates of substitution. This section will focus on the impact of economic and political factors, managed care, types of employment, and employee contributions on substitution.

1. Economic and Political Factors

As the marketplace varies by region throughout the United States, it is plausible that substitution also varies based on the unique economic and political environment within each state. For example, wage and cost-of-living differences in the North and South translate into different earning power for families at 200% of poverty. States with a proclivity to more service-oriented industries are likely to have a greater percentage of workers without access to health insurance. The Healthy Kids program, situated in Florida where tourism and farming drive the marketplace, found that the majority of their participants were from families employed in service-related jobs that do not typically offer employee coverage, i.e., waitressing, landscaping, the fast food business, hotel maintenance staff, and small businesses geared toward tourism. Similarly, the RIteCare representative theorized that many of their participants come from Rhode Island's textile, jewelry, and fishing industries, which are not likely to offer coverage.

Other market conditions such as the nature of unemployment may also affect insurance levels. For example, the economy in Oregon is relatively strong which has tightened the competition for employees. This has resulted in increased benefits and improved incentives being offered by employers. In other cases, where there is less competition for labor, employers may attempt to limit costs, including benefit costs, resulting in better product pricing. Such conditions are likely to affect the nature of insurance offered by employers, and the costs to employees, especially for dependent coverage.

2. Managed Care

The increased use of managed care may potentially reduce substitution effects. Managed care penetration, which impacts the affordability of insurance, varies by region and also may impact substitution. Oregon, which has actually witnessed an increase in employer-sponsored insurance, has a relatively high-managed care penetration rate of approximately 60% or higher within the state's insured population. Managed care options such as health maintenance organizations (HMOs), preferred provider organizations (PPO), and point-of-service plans (POS)

are appealing insurance options for employers. A strong state economy and a competitive labor market that requires employers to provide insurance coverage as a means of attracting workers affect Oregon's high rates of employer-sponsored insurance. Costs to small employers for health insurance are also relatively low in Oregon: the average premium for a 2-50 person business is approximately $425-450 per month for family coverage, which is 15-25% less than the national average.

3. Types of Employment, Firm Size, and Employee Status

Another major factor underlying the declining rates of employer coverage is the shift of employment from high coverage industries such as manufacturing to lower coverage industries such as service industries. Job market trends toward small business employment, part-time work, and lower-wage industries contribute to the increasing numbers of employees without health insurance. For example, a Minnesota study reported that although most employers contributed to the cost of employee insurance, over 50% stated that they either had not worked long enough or did not work enough hours to be eligible. Nationally, the percentage of workers employed in firms with fewer than 25 employees has increased from 28.5% in 1989 to 30.3% percent in 1994 (29.3% in 1996). In addition, the percentage of workers employed part-time increased from 17.7 % in 1989 to 19.3% in 1994. There has also been an increase in lower wage employees, with the percentage of workers earning less than $250 per week increasing from 22.3 % in 1989 to 25.3% in 1994.

4. Employee Contributions

The affordability of insurance influences families' decisions to maintain or opt out of employer-sponsored insurance. The increasing amount of employee premium contributions may be encouraging families to substitute public coverage. From 1992 to 1996, the average percentage of the premium paid by workers for individual coverage increased from 17.5% to 21.7%. The percentage of the premium paid by workers for family coverage increased from 23.7% in 1992 to 32% in 1995, but dropped to 29.5% in 1996. Similarly, increases in patient cost sharing requirements may influence individual-based substitution. Although patient cost sharing has remained relatively stable for managed care plans, there have been significant increases in indemnity plans: the median maximum out-of-pocket amount for individuals increased from about $1,200 in 1994 to $1,300 in just one year and the percentage of indemnity plans with an individual out-of-pocket maximum for individuals in excess of $1,500 per year increased from 35% in 1993 to 48% in 1995.

VI. Implications For Future Expansions Under Title XXI of the Social Security Act

A. Implications for the design of Title XXI programs by states

Decisions regarding how and to what extent to expand coverage for uninsured children will be made in the context of broader state policy objectives and existing state programs. The issue of substitution and how states will address it is likely to be influenced by specific views regarding the role of the public and private (employer) sectors in providing affordable health care coverage for their citizens. There are two ways that substitution can occur under CHIP: (1) substitution of coverage, which can occur when families or employers drop existing private coverage to enroll in CHIP, and (2) substitution of premiums, which can occur when employers reduce their contribution levels for dependent coverage because families are able to obtain subsidies that offset the reduced amount of contribution. Existing and potential policies regarding insurance reform, efforts to assist employers in providing coverage, and state social and economic policies are likely to be examined in the overall planning for expanded children's coverage and the effect of possible substitution of private coverage. Current research coupled with past experiences of states provides some information for states to consider as they plan and implement programs funded by Title XXI. Four major program design issues were identified as important areas in addressing substitution: the overall design of the program; determining whether to use Medicaid expansions or separate state programs; the development of specific provisions related to employer coverage; and the oversight and evaluation of the expanded program.

1. The overall design of the program needs to address factors that do not limit access to new coverage or replace existing insurance coverage

It is essential for children's health insurance programs to adequately design programs that target children not eligible for Medicaid, affordable employer-sponsored, or other health insurance programs. Implementation of well-defined state programs that conduct outreach to enroll targeted children while controlling for substitution, may be a vehicle through which states can actually increase participation in all forms of available insurance. States also need to consider how the implementation of their programs will affect and may potentially enhance the private market. Programs may utilize strategies such as employer buy-ins as a collaborative effort with the private sector to augment the enrollment of eligible children and increase the sustainability of state programs in the long-term.

2. Implementing a Medicaid expansion versus a separate state insurance program

States deciding whether to implement a separate state insurance program or expand Medicaid will need to focus on differences between the potential for substitution of a Medicaid program versus a separate state program. States may consider the following questions with respect to understanding the complexity of substitution: What are the tradeoffs? What will influence family choice? Are there certain situations that foster substitution in one type of program versus another? Employer-sponsored plans have the ability to provide varying benefit packages and impose premiums, deductibles, and coinsurance which may have an impact on individuals at certain income levels. However, Medicaid programs have a comprehensive benefit package with minimal cost sharing for beneficiaries. In some cases, Medicaid has the potential for significant substitution due to these features, especially for populations such as children with special health care needs. Title XXI programs, can select from a range of "benchmark" or "benchmark equivalent" plans and have more flexibility on cost sharing. There are other considerations states may also take into account in deciding between a Medicaid expansion and a separate state insurance program. For example, there is a stigma attached to social insurance programs such as Medicaid that may deter families from substituting public for private programs.

3. Issues related to employer coverage

Employee participation may increase as a result of additional support for dependent coverage as provided by small employers, including employer buy-in programs or other alternatives that lower the cost to employees. Employee participation may also be enhanced as a result of increased support for dependent coverage provided by small employers. Most insurance companies require a minimum level of employee participation in employer-sponsored plans in order to reduce the threat of adverse selection. It is often difficult for employers to obtain the required minimum level of participation for dependent health insurance programs due to the high cost of such plans. Therefore, assistance in lowering cost to employees may increase total participation in small employer health plans.

4. Evaluation of program expansions should provide the data necessary to monitor possible substitution.

States need to consider how they plan to implement strategies that might limit substitution. There are two major issues that may affect compliance with and evaluation of expanded programs which attempt to monitor substitution. First, several states have suggested that ensuring compliance with existing strategies to limit substitution is a cumbersome and expensive process. Second, the establishment of strategies to limit substitution, or "firewalls," may increase the complexity of the enrollment process, thus impeding the enrollment of eligible children. However, there are some examples of verification processes that are effective and efficient. Florida has utilized the National School Lunch Program as a method to verify eligibility of enrollees, and other state programs have used the verification processes of state Medicaid programs.

B. Despite past difficulties in quantifying the substitution effect, states must evaluate the nature, scope, and levels of substitution as required under Title XXI.

States are required under Title XXI to evaluate the effectiveness of their efforts to limit substitution. States must design and implement plans to evaluate how well their programs are reducing the number of uninsured children without substituting for private coverage. With any assessment it is necessary to identify the actual impact of substitution directly related to specific measures versus other unique state-specific issues (e.g., income, cost of living, structure of the labor market, wage levels, characteristics of the insurance market). Through expanded state efforts under Title XXI funding, states will be required to evaluate and estimate the nature of substitution, the scope of the problem, and the actual levels of substitution among employees and employers. The collection of state-specific information will assist state program directors, researchers, and federal agencies in measuring the actual impact of employee- and employer-based substitution, and if consistently measured, will allow for comparison across states.

C. There are various types of data states may need to collect in order to measure substitution.

Most states did not initially identify substitution as a major concern in current efforts to provide coverage to uninsured children. As a result, states did not conduct initial surveys or evaluations to quantify rates of substitution and the long-term impact of employees and employers substituting for private coverage. In order to assist states in determining types of data to collect, the following is a list of information that may be useful in effectively quantifying substitution:

What are the characteristics of the target population covered by state's current programs? Income levels? Differences at various levels (150 vs. 200)? Access to employer coverage? Health status? Other characteristics?
What has the states' enrollment experience been: at different income levels; different age groups? For different employment status; other characteristics influencing enrollment?
From where do program participants come and where do they go when they leave the program? From or to Medicaid? From or to private coverage? From or to no coverage?
There are various program features that may relate to either increasing or decreasing the likelihood of crowd out. What are the issues around crowd out that were considered in designing the state's program? Who raised these issues? How were the issues addressed in the design? How did actual experience match up with expectations? How was the program modified to reflect these experiences?
Do states perceive crowd out to be different between Medicaid coverage and other types of government supported coverage?

D. States have identified the need for additional research.

The implementation of state strategies to expand coverage for children has been too recent to fully assess the actual impacts on individual- and employer-based substitution. Therefore, states have identified a number of areas in need of additional research:

Factors associated with individual (family) decisions regarding participation in employer-based or public insurance. States have identified individuals opting out of employer-sponsored insurance coverage as a greater concern than employers pushing individuals out of coverage. Therefore, states have suggested the importance of collecting information that will describe the factors that influence the complex decisions families are making with respect to purchasing insurance coverage for their children.
Whether and how the availability of public programs affect employers dropping coverage. Little data currently is available to describe why employers decide to drop coverage and the affect the availability of public programs may have on those decisions. Therefore, state specific information focused on where, when and why some employers drop coverage would assist in the understanding of employer-based substitution.
How different mechanisms work to encourage enrollment of individuals in state programs, yet limit crowd out. Much of the current information on state strategies to limit substitution are based on anecdotal evidence. The effectiveness of mechanisms may be confined to the demographics, economics, and general characteristics of a particular state. However, an understanding of why and how certain mechanisms work in specific situations is necessary in order to identify the usefulness of these or variations of these mechanisms in other settings.
The insurance status of individuals that leave state programs. States have identified the need to acknowledge where individuals go to receive health insurance after they have disenrolled from state programs. It is important to understand whether or not individuals are leaving state-subsidized programs for private insurance or whether they are again becoming uninsured. States need to understand whether or not they are meeting the goals of their programs. In addition, this information will assist in shaping Title XXI programs into either long-term or transition programs based on the overall needs of the target population.
Issues related to the affordability of state programs and the implications of price sensitivity among potential and current enrollees. As indicated earlier, most research does not address this issue, although a recent article by economists at the Agency for Health Care Policy and Research identified the importance of examining the affordability of private coverage. However, states have also identified the importance of affordability in state programs to insure children. Several states have found that increasing monthly premiums by a few dollars resulted in increased disenrollment rates. As states are focused on insuring larger numbers of uninsured children, affordability is an important issue that directly impacts the enrollment of eligible children in some income ranges. This issue is related to the complex decisions families make that are substantially based on cost.

VII. Discussion

Substitution is a complex issue and is further complicated by the lack of state-specific data collected on both individual- and employer-based substitution. Actual estimations of substitution are dependent upon the definition itself, i.e., employer- versus individual-based substitution. In lieu of actual data, states have made policy decisions based upon anecdotal information provided by individuals and employers regarding substitution. States have also identified the complex factors affecting consumer choice, acknowledging that families make difficult decisions regarding health insurance coverage for their children, based upon many factors including affordability of the coverage, the types of policies and cost sharing requirements provided by employers, and the availability and accessibility of public programs. The new infusion of federal funding for state programs and the potential impact such funds may have on substitution needs to be considered both in light of past state experiences as well as future program decisions. The sheer volume of children entering the market via state programs may change the type of health plans offered in the market. Health plans may begin to offer employee-only coverage or children-only policies as a result of the availability of new public money targeted to insure low income children. There may also be cost implications for current employer-based coverage as children move to public programs. As children are relatively inexpensive in terms of health care expenditures, displacing them from the risk pool may cause an increase in costs, directly impacting individuals in small employer plans with existing dependent coverage.

The purpose of this paper was to report state experiences with substitution and the strategies states employed to limit the effects of that substitution on the market. Title XXI requires states to take responsibility for the effect of substitution on their new programs. States must recognize that individual state strategies may be uniquely effective within the structure and confines of that market and state. Although some strategies may not be effective in some environments, states can learn from and develop alternatives based on other states' experiences. This will require states to consider the full range of alternatives when deciding to implement strategies that are meaningful to their own political and market climate.

Appendix A: Bibliography

Blumberg, L., Dubay, L., Norton, S. (June 1997). "Did the Medicaid Expansions for children Displace Private Insurance?" The Urban Institute.

Center for Studying Health System Change. (1996, October). Medicaid Eligibility Policy and the Crowding-Out Effect: Did Women and Children Drop Private Health Insurance to Enroll in Medicaid? (Issue Brief No. 3). Washington, D.C.

Cooper, Philip F., Steinberg Schone, Barbara. (November 1997). More Offers, Fewer Takers For Employment-Based Health Insurance: 1987 And 1996. Health Affairs. 16 (6). 142-149.

Cutler, D., Gruber, J. (May 1996). Does Public Insurance Crowd Out Private Insurance? The Journal of Economics.

Cutler, D., Gruber, J. (1997, January/February). Medicaid and Private Insurance: Evidence and Implications. Health Affairs. 16 (1), 194-200.

Dubay, L., Kenney, G. (Spring 1996). The Effects of Medicaid Expansions on Insurance Coverage of Children. The Future of Children, 6 (1), 152-161.

Employee Benefit Research Institute. (1997, May). Trends in Health Insurance Coverage (Issue Brief No. 185). Washington, DC.

Holahan, J. (1997, January/February). Crowding Out: How Big a Problem? Health Affairs. 16 (1), 204-206.

The Lewin Group. (October 21, 1996). Recent Trends in Employer Health Insurance Coverage and Benefits (Report to the American Hospital Association). Fairfax, VA: Sheils, J., Alecxih, L.

The Lewin Group. (June 16, 1997). Exploring the Question of Substitution: Will Expansions of Health Insurance Coverage for Children Result in Shifts from Private to Public Coverage? Fairfax, VA: Sheils, J.

Lurie, N., Pheley, A., Finch, M. (1995, October). Is MinnesotaCare Hitting its Target? Institute for Health Services Research, University of Minnesota School of Public Health and Hennepin County Medical Center.

The Rochester Child Health Studies Group. (April 1996). Evaluation of Child Health Plus in New York State. Rochester, New York: University of Rochester.

Shore-Sheppard, L. (July 1996). The Effect of Expanding Medicaid Eligibility on the Distribution of Children's Health Insurance Coverage. University of Pittsburgh.

Thiede Call, K., Lurie, N., Jonk, Y., Feldman, R., Finch, M. (October 8, 1997). Who Is Still Uninsured in Minnesota? Journal of the American Medical Association. 278 (14), 1191-1195.

Appendix B: Participants in the September 9th Roundtable Discussion

Doris Barnette, ACSW, Acting Principal Advisor to the Administrator
Health Resources and Services Administration

Jennie Bonney, Social Science Research Analyst
Health Care Financing Administration(now known as Centers for Medicare and Medicaid Services(CMS))

Cheryl Austein-Casnoff, Director, Public Health Policy
Office of the Assistant Secretary for Planning and Evaluation

Deborah Chollet, Associate Director
Alpha Center

Rick Curtis, President
Institute for Health Policy Solutions

Pam Dickson, Child Health CoordinatorThe Robert Wood Johnson Foundation

Denise Dougherty, Special Coordinator
Agency for Health Care Policy and Research

Lisa Dubay, Senior Research Associate
The Urban Institute

Margaret Edmunds, Study Director
Institute of Medicine

Judith Feder, Professor of Public Policy
Georgetown University, Institute for Health Care Research and Policy

Paul Fronstin, Research Associate
Employee Benefits Research Institute

Sue Ginsburg, Vice President
The Lewin Group

Ron Hook, Federal Reform Coordinator
MinnesotaCare

Genevieve Kenney, Senior Research Associate
The Urban Institute

Trisha Leddy, Director of the Office of Managed Care
RIteCare

Larry Levitt, Director, Changing Health Care Marketplace and California Grants Program
Henry J. Kaiser Family Foundation

Steve Long, Senior Economist
RAND

Jim Mays, Vice President
Actuarial Research Corporation

Faith McCormick, Assistant to the Director for Intergovernmental Affairs
Department of Health and Human Services

Representative John McDonough
The Commonwealth of Massachusetts

Rose Naff, Director
Florida Healthy Kids Corporation

Mary Jo O’Brien, Vice President
The Lewin Group

Jim Reschovsky, Senior Health Researcher
Center for Studying Health System Change

Christy Schmidt, Deputy to Deputy Assistant for Health Policy
Office of the Assistant Secretary for Planning and Evaluation

John Sheils, Vice President
The Lewin Group

Marjorie Shoffer, Policy Associate
National Conference of State Legislatures

Christopher Snow,Policy Analyst
Office of Human Services Policy

Barney Speight, Administrator
Oregon Health Plan

Caroline Taplin, Senior Policy Advisor
Office of the Assistant Secretary for Planning and Evaluation

Rebecca Weiss, Senior Health Policy Analyst
Colorado Department of Health Care Policy and Financing

Joy Johnson Wilson, Senior Health Committee Director
National Conference of State Legislature

Appendix C: Overviews of State Programs

California Kids

HISTORY:

California Kids is a non-profit organization sponsored by private donations that was founded July 30, 1992. The program was initially based in Los Angeles, and has since expanded to 33 counties. CaliforniaKids offers children a limited health insurance product that focuses on preventive care. The program coordinates with MediCal, the state Medicaid program, to obtain care for children who need inpatient treatment or other serious care that CaliforniaKids does not provide. CaliforniaKids partners with community organizations to identify and enroll eligible children. Children enrolled in CaliforniaKids pay no monthly premiums, but are responsible for small co-payments for their care. CaliforniaKids has provided coverage for 17,000 children since its inception in 1992.

TARGET POPULATION:

Uninsured children ages 2-18 who are unmarried, ineligible for coverage under any federal or state health benefit program, not currently enrolled under a private health care contract, and enrolled and attending school (if school-age) are eligible for CaliforniaKids if their family income is below 200% of the poverty level. In August 1997, a pilot program was initiated in San Diego that will expand income eligibility up to 300% of the poverty level in that county.

BENEFIT PACKAGE:

Benefits covered by CaliforniaKids include primary and preventive care, immunizations, prescriptions, a 24-hour nurse hotline, hearing screening and aids without co-payment, emergency treatment with a $25 co-payment, outpatient care and doctor visits with a $5 co-payment each, and dental and vision with a $10 co-payment each.

PROVIDER NETWORKS AND REIMBURSEMENTS:

CaliforniaKids utilizes a managed care, capitated system to maintain cost-effectiveness, minimize risk and unnecessary administrative expense due to excessive claims processing, and provide a "medical home" for the child. The network, administered by Health Affairs International, maintains choice and is sensitive to access, culture, language, gender, and age of child. Over 15,000 health care providers are contracted to provide services throughout the State and a 24-hour nurse hotline is available to all members to provide health care information and re-direct care from the emergency room to a lower cost, appropriate alternative. Plans donate their direct administrative expenses. CaliforniaKids partners with medical groups and IPAs, which offer wellness education to its members. Services are administered by Blue Cross of California, Delta Dental, Vision Service Plan, Wellpoint Pharmacy, and Access Health.

FINANCING:

CaliforniaKids is financed through private donations. CaliforniaKids does not collect premiums, but charges nominal co-payments for care. Costs are kept low by offering a limited benefit package and minimizing administrative costs.

Colorado Children's Health Plan

HISTORY:

The Colorado Child Health Plan (CCHP) was established in 1992 as a community-based health care reimbursement program for low-income children. CCHP operates as a health maintenance organization that is administered by the University of Colorado Health Sciences Center. The program initially targeted rural counties with a low number of safety net providers, and built its own network of physicians. Recent legislation established the Children’s Basic Health Plan (CBHP), a program which will expand eligibility under CCHP up to age 18 and extend the network into every county in the state. CBHP will be a full benefit program administered by private HMOs. The CBHP will be funded in part by the Title XXI block grant. The Colorado Child Health Plan will provide the basic infrastructure for the establishment of the Children’s Basic Health Plan which will be implemented in March 1998. The Colorado Child Health Plan will sunset on June 30, 1998.

TARGET POPULATION:

The target population of CCHP is children age 18 and under with incomes up to 185% of the poverty level who are not eligible for Medicaid. Enrollment in FY1996 was 4,893 children. In FY1997 the target is 6,217 children, and in FY1998 the target enrollment in the CCHP/CBHP is estimated to be 14,000 children.

BENEFIT PACKAGE:

The benefit package was modeled after the Blue Cross/Blue Shield Caring Programs. Benefits include most preventive services such as well child care, immunizations, laboratory tests, visits for illness and injury, chronic care and outpatient surgery; but do not include inpatient hospital care, eyeglasses, hearing aids or dental care. There is a maximum annual benefit of $10,000 per child. The program expansion will include inpatient and mental health care.

PROVIDER NETWORK AND REIMBURSEMENT:

CCHP is administered by the state of Colorado through the University of Colorado Health Sciences Center. The program has a statewide network of pediatric physicians. Primary care physicians are reimbursed on a capitated basis and bear risk. Medical specialists are reimbursed on a fee-for-service basis.

FINANCING:

Funding is obtained for CCHP through a combination of state appropriations and private donations. Funding comes from several sources: General Fund; intergovernmental transfer cash reserves and interest paid on those reserves; private donations; and enrollment fees. The enrollment fee is a $25 per child yearly payment in lieu of premium contributions. Co-payments are required on doctor visits, health screenings, and prescriptions. Blue Cross/Blue Shield HMO donates all claims processing services

Florida Healthy Kids Corporation

HISTORY:

The Florida Healthy Kids Corporation was established in November 1990 to create a comprehensive insurance product for school children using Florida school districts as a grouping mechanism. A HCFA(now known as CMS)-funded demonstration project of Healthy Kids was initiated in Volusia County in 1992 and was completed in 1995. In late 1993, Healthy Kids began expanding into additional counties utilizing both state and local government funds. Healthy Kids currently serves 24 of Florida’s 67 counties, which contain over half the state’s uninsured children. The program is designed to provide affordable access to health insurance for all Florida children. Children who are not on the school lunch program pay the entire cost of their insurance, while children who are on the school lunch program have access to subsidized insurance. Participation by Florida counties in the Healthy Kids program is strictly voluntary.

TARGET POPULATION:

Once a school district elects to participate in the program, qualified children become eligible for Healthy Kids. Children must be between the ages of 5 and 19, uninsured (i.e. not enrolled in Medicaid or private insurance), and enrolled in school up to the 12th grade. Counties may expand eligibility to allow children of other age groups the ability to participate (e.g. children who are less than school age whose families are enrolled in Healthy Kids).

BENEFIT PACKAGE:

Program benefits are extensive, including inpatient care, surgery, emergency services and transportation, eyeglasses, hearing aids, physical therapy, mental health services, pre-natal care and delivery, and organ transplants. Co-payments are required for certain benefits. Counties may add additional services such as dental care to the package. Healthy Kids has a $1,000,000 lifetime cap on services received.

PROVIDER NETWORK AND REIMBURSEMENT:

Twelve managed care networks serve Healthy Kids enrollees. Insurers are selected on the basis of five variables: price, benefits, provider network, reporting capabilities, and general criteria such as accreditation and solvency. Healthy Kids aggregates state, local and family funds to pay premiums to commercial health plans that assume the insurance risk.

FINANCING:

All individuals enrolled in the Healthy Kids program are required to pay some portion of the cost of their insurance. Sliding-scale premiums are based on reported family income required for participation in the school lunch program. The three cost-sharing levels are "Free Lunch", "Reduced Lunch", and "Not on Lunch Program". Premium levels vary by county. Funding for the demonstration project in Volusia county came from a HCFA(now known as CMS) waiver. All subsequent funding has been granted through a combination of state, local and family contributions.

Children’s Medical Security Plan of Massachusetts

HISTORY:

Authorizing legislation for the Children’s Medical Security Plan (CMSP) was passed in 1993 to provide access to preventive and primary care services for Massachusetts' uninsured children. The plan originally covered uninsured children age 12 or younger. The impetus for the program was initially the national attention to health care reform and access to health care for all individuals. The passage of the Health Access Law in July 1996 expanded CMSP program eligibility to include adolescents up to age 18. In 1997 administration of the plan was transferred to the state Department of Public Health.

TARGET POPULATION:

The target population for CMSP is uninsured children under the age of 19 who are residents of Massachusetts and who are not eligible for Medicaid. The number of projected children eligible for CMSP is 76,000.

BENEFIT PACKAGE:

The Children's Medical Security Plan offers a range of benefits designed to cover the services most frequently used by children. Routine well-child check-ups, immunizations, and smoking prevention services are covered without co-payment. All other covered benefits require a co-payment of are $1, $3 or $5 based on family income guidelines.

Limited coverage is offered for emergency care, prescription drugs, durable medical equipment, and outpatient mental health services. CMSP does not pay for OTC drugs, ambulance transport, inpatient care, dental care, or early intervention. However, children enrolled in CMSP may be eligible to receive inpatient care in any of the state hospitals or community health clinics through the state free care pool.

PROVIDER NETWORK AND REIMBURSEMENT:

The Community Health Plan (CHP) and John Hancock administer the health insurance program. The Community Health Plan serves three rural counties with approximately 3,000 participants. John Hancock serves the remainder of the state, with roughly 34,000 enrollees. John Hancock reimburses its providers on a fee-for-service basis, while CHP reimburses both through capitation and fee-for-service arrangements.

FINANCING:

There are three sources of funding for the program: tobacco taxes, general funds, and family premium contributions. Families earning 200% or less of the federal poverty level receive the insurance free of charge. Those families earning below 400% of the poverty level are charged a reduced premium rate of $10.50 per child per month, with a $32.50 maximum per family per month. Families with income over 400% of poverty are charged the full premium, which is currently set at $52.50 per month.

Minnesota Care

HISTORY:

MinnesotaCare began in 1992 as part of a package of legislation aimed at reducing the number of uninsured in the state of Minnesota. With the advent of the MinnesotaCare demonstration project, all Minnesotan enrollees were transitioned to mandatory managed care. Phase I of the demonstration extended Medicaid coverage to families with children under 275% FPL and with no insurancae coverage. Under the demonstration, childdren and pregnant women receive all benefits available to traditional Medicaid enrollees. In May 1997, a bill was approved to expand MinnesotaCare eligibility for adults without children with income from 135% FPL to 175% FPL. The legislation also reduced provider taxes and added non-preventive dental coverage for adults in families below 175%.

TARGET POPULATION:

MinnesotaCare is available to families with children who have incomes of less than 275% FPL and individuals who have incomes of less than 175% FPL. TO become eligible for MinnesotaCare, one must have been uninsured for at least four months and have had no access to employer subsidized coverage for 18 months or more. An asset test will be implemented when the amended demonstration waiver is approved by HCFA(now known as CMS). Under the asset test, families cannot have more than $30,000 in assets, and individuals cannot have more than $15,000 in assetss to be eligible for the plan.

BENEFIT PACKAGE:

The program offers a comprehensive benefit package including inpatient hospital benefits. However, inpatient benefits are capped at $10,000 annually for adults without children and for parents with incomes greater than 175% FPL.

PROVIDER NETWORK AND REIMBURSEMENT:

Enrollees have the opportunity to choose a health plan upon enrollment into the program. Provider reimbursement was originally structured on a fee-for-service basis but now is a capitated payment system.

FINANCING:

MinnesotaCare was originally funded through a cigarette tax, a provider tax, and family contributions. The cigarette tax is no longer used for MinnesotaCare support; it was only applied from July 1992 through January 1994. In July 1995, the state 1115 demonstration waiver was implemented, and the federal match was added to the funding mix. In 1997 a reduction was made in the provider taxes due to a surplus in MinnesotaCare funds.

New York Child Health Plus

HISTORY:

The Child Health Plus program was passed by the New York State legislature in 1990, and by August of 1991, children began receiving coverage under the program. New York’s Child Health Plus is the largest of 13 non-Medicaid, taxpayer-funded child health insurance programs in the country. As of July 1, 1997, the program was providing coverage to over 135,000 children. The program has estimated target enrollment over the next few years to total 250,000; this number may increase significantly due to recent federal legislation. The overall intent of the Child Health Plus Program is the following:

To improve the health status of children participating in the program by providing a "medical home"
To provide primary, preventive, outpatient and inpatient health insurance coverage to low income children by removing financial barriers to purchasing such coverage through an individual subsidy program
To increase children’s access to primary comprehensive, preventive and inpatient health care services
To reduce and more effectively target bad debt and charity care expenditures in the state of New York

TARGET POPULATION:

Children considered eligible for the program must be either uninsured or underinsured, residents of New York State, under the age of 19, and not eligible for state Medicaid benefits. Children under age 19 in families with an income below 222% FPL are eligible for an income-variable premium subsidy. Children in families above 222% FPL are permitted to buy into the program without any premium subsidy.

BENEFIT PACKAGE:

The program began providing coverage to children in August of 1991, initially providing subsidized primary and preventive outpatient care to children under age 13. In 1997, the program extended benefits to include inpatient care (excluding inpatient services for mental health and substance abuse) and extended coverage to include children through age 18.

PROVIDER NETWORK AND REIMBURSEMENT:

The provider network currently consists of 15 insurance plans contracted to provide health services to enrollees, of which 11 are MCOs and 4 are traditional indemnity plans. As of October 1, 1997, 24 insurance plans, with almost all providing a managed care product, will participate in Child Health Plus. Additionally, a marketing and outreach organization has been contracted to promote the plan throughout the state. This will be complemented by an in-house statewide marketing effort.

FINANCING:

The program is financed through a provider surcharge established under the New York Health Care Reform Act of 1996. The premium contribution from families participating in the program is an additional funding source.

Oregon Health Plan

HISTORY:

Between 1989 and 1995, the Oregon Legislature passed a series of laws to reform the state health care system. These laws provided the basis for a major overhaul of the method of providing and paying for health care to Oregon's needy populations and the establishment of the Oregon Health Plan. A coalition including consumers and health care industry representatives undertook the weighty task of ranking health care procedures according to their public good. Those items deemed most beneficial are covered by the Oregon Health Plan. The Oregon Health Plan went into effect in February 1994 and enrolled nearly 120,000 new members in its first year. A high risk insurance pool is also available for those denied private coverage because of pre-existing medical conditions, with rates limited to a maximum of 125% of private insurance rates.

TARGET POPULATION:

The Oregon Health Plan is available to all Oregon residents whose income is less than 100% of the poverty level. In addition, pregnant women and children ages 0-5 are eligible for services if their family income is less than 133% of the poverty level. All enrollees must be citizens or legal aliens who reside in Oregon. Coverage is also available to people who were refused health coverage by private insurers because of pre-existing medical conditions through the high risk pool.

BENEFIT PACKAGE:

The legislation enabling the Oregon Health Plan established an eleven-member Health Services Commission that was charged with the task of prioritizing health care services for coverage based on the benefit to the entire population of the state. The ranking of health services was conducted through an analysis that rated health services and treatment according to cost, duration of benefit, physician opinion on the effectiveness of treatment and prevention of death, and consumer opinions on the seriousness of specific health conditions. Coverage is provided for services above a certain threshold on this list. The exact threshold is determined by the legislature each year. All services frequently utilized by children have historically been covered.

PROVIDER NETWORK AND REIMBURSEMENT:

Most individuals enrolled in the Plan are required to join a health maintenance organization; however, the frail elderly, the disabled, and certain others (approximately 1/5 of total enrollment) are permitted to receive services on a structured fee-for-service basis through a designated case manager.

FINANCING:

The central principle of the Oregon Health Plan is that eligibility for publicly sponsored coverage can be expanded if costs are contained through managed care and prioritization of benefits. By increasing the number of covered individuals, it is expected that charity care and cost shifting will decline. The program is funded by the state general fund, federal match, and a newly-passed initiative raising Oregon’s tobacco taxes.

RITECARE (Rhode Island)

HISTORY:

RIteCare is an 1115 Waiver program that was implemented August 1, 1994. Rhode Island's statewide initiative seeks to increase access to primary and preventive health care services for all Aid to Families with Dependent Children (AFDC) beneficiaries and certain low-income women and children through a fully capitated managed care delivery system. RIteCare expands Medicaid eligibility and provides an enhanced set of primary care and preventive benefits through a fully capitated managed care system. Unlike the traditional Medicaid program, RIteCare requires cost sharing for individuals with family incomes between 185% and 250% of the poverty level.

TARGET POPULATION:

The Rhode Island Department of Human Services estimates that approximately 7,000 previously uninsured persons will find themselves able to access care through RIteCare. RIteCare provides coverage to pregnant women and children up to 8 years of age with family incomes at or below 250 percent of the FPL; in addition, women who would otherwise lose Medicaid eligibility post-partum remain eligible for family planning services. RIteCare currently serves 73,500 people. Seventy thousand of these are AFDC and AFDC-related Medicaid beneficiaries. The remaining 3,500 are children up to age 8 and pregnant women with incomes of 185-250% of the federal poverty level. On May 1, 1997, coverage was made available to all uninsured children up to age 18 in families up to 250% of the poverty level.

BENEFIT PACKAGE:

Each health plan offers medical, dental and mental health benefits, and enhanced outreach services such as provision of bus passes. Participating health plans are required to offer a package of services intended to overcome non-financial barriers to care, such as outreach, home visits, nutrition counseling, childbirth education, and parenting classes.

PROVIDER NETWORK AND REIMBURSEMENT:

Four managed care plans are currently serving RIteCare enrollees: Harvard Pilgrim Health Care, Coordinated Health Partners, Neighborhood Health Plan of Rhode Island, and United HealthCare of New England. Rhode Island has made significant strides in improving access to primary care through RIteCare. Primary care physician participation in Medicaid has increased from 350 to over 800 physicians. The average number of physician visits per enrollee increased from two per year to five per year in RIteCare's first year (1995). Both hospital utilization and emergency room visits fell by more than one third from 1993 to 1995.

FINANCING:

RIteCare is funded through a mix of Medicaid dollars, state funds and individual premiums. Individuals with family incomes between 185% and 250% of the poverty level contribute to the cost of their insurance through monthly premium payments.

Washington Basic Health Plus

HISTORY:

In 1993, Washington State created the Basic Health Plus program for children under 200% of the federal poverty level through a Medicaid expansion. These children were previously eligible for health insurance through the full state-subsidized Basic Health Plan established in 1988, under which adults below 200% of the federal poverty level still receive coverage. The Basic Health Plus program was established to take advantage of the federal funding available through Medicaid. Children covered under the program receive a comprehensive benefit package at no cost to their family. Provider networks are aligned between the Basic Health Plus program and the Basic Health Plan so that families can visit the same physicians.

TARGET POPULATION:

Children ages 0-19 whose families earn less than 200% of the poverty level are the target population of Basic Health Plus.

BENEFIT PACKAGE:

The Washington Basic Health Plus program has the same rich benefit package traditionally provided under the state Medicaid program. This includes all preventive services, plus benefits such as inpatient care, dental care, eyeglasses and hearing aids. Basic Health Plus provides routine screenings of up to $125 per child per calendar year without co-payment (this does not apply to well-baby care). Immunizations and well-child visit are covered up to a maximum of $400 per calendar year for children ages 1-4. Children receive a 50% discount on glasses and a 25% discount on contact lenses. There is an unlimited inpatient stay, but care must be authorized within 48 hours of admission.

PROVIDER NETWORK AND REIMBURSEMENT:

There are currently fourteen managed care plans contracting with Basic Health Plus. There is almost a complete overlap between the provider network utilized by Basic Health Plus and that used by the Basic Health Plan.

FINANCING:

The program is funded through a combination of state and federal funds through the Medicaid match. State funds are obtained primarily through state taxes on cigarettes and alcohol.

Topics

Child Welfare

Populations

Children

Examining Substitution: State Strategies to Limit "Crowd Out" in the Era of Children's Health Insurance Expansions

Executive Summary

State Mechanisms To Limit Substitution

Mechanism 1: Identifying the Affordability of Private Coverage: Setting Premiums and Copayments That Do Not Deter Utilization or Enrollment

Mechanism 2: Using Periods of Uninsurance, Access to Employer-Sponsored Insurance, and Employer Contributions

Mechanism 3: Use of Subsidies

Mechanism 4:Limiting the Scope of the Benefit Package

Mechanism 5: Purchasing Cooperatives

Mechanism 6: Employer Buy-Ins

Mechanism 7: Reimbursements To Small Businesses

Disconnect Between State Experiences and Research

Implications For Future Expansions Under Title XXI of the Social Security Act

I. Introduction

A. State initiatives to insure children have yielded a new set of concerns regarding the substitution of public coverage for private health benefits.

B. Due to limited data and the complexity of the issue, the discussion of substitution has been based upon the perceivedeffect of crowding out private coverage.

C. Structure of the Report

II. Methodology

A. Interviews of selected state children's health program, Medicaid, and MCH directors were conducted in order to obtain specific program information.

B. The September 9th roundtable discussion was designed to identify state experiences with and future concerns regarding substitution, especially as it impacts expansions under the provisions of Title XXI.

III. Review of the research on substitution

A. Substitution occurs when individuals substitute public benefits for private sector benefits, and it may be defined as (1) individual-based or (2) employer-based.

B. There is a range of substitution estimates reported in the literature.

C. National research informed the policy debate and raised important issues, yet it has not been focused on addressing state concerns with substitution.

IV. State Mechanisms To Limit Substitution

A. States have developed a variety of strategies to limit individual-based substitution.

Mechanism 1: Identifying the Affordability of Private Coverage: Setting Premiums and Copayments That Do Not Deter Utilization or Enrollment

Mechanism 2: Using Periods of Uninsurance, Access to Employer-Sponsored Insurance, and Employer Contributions

Mechanism 3: Use of Subsidies

Mechanism 4: Limiting the Scope of the Benefit Package

B. Three types of mechanisms are being used by states to enhance the affordability of employer-sponsored insurance.

Mechanism 5: Purchasing Cooperatives

Mechanism 6: Employer Buy-Ins

Mechanism 7: Reimbursements To Small Businesses

V. Estimating the Amount of Substitution

A. The comparison of benefit packages and program costs drive parental decisions to substitute public programs for private insurance.

B. There is a disconnect between state experiences and research.

C. Substitution may also be linked to secular trends in the job market, insurance industry, and the general state of the economy.

1. Economic and Political Factors

2. Managed Care

3. Types of Employment, Firm Size, and Employee Status

4. Employee Contributions

VI. Implications For Future Expansions Under Title XXI of the Social Security Act

A. Implications for the design of Title XXI programs by states

1. The overall design of the program needs to address factors that do not limit access to new coverage or replace existing insurance coverage

2. Implementing a Medicaid expansion versus a separate state insurance program

3. Issues related to employer coverage

4. Evaluation of program expansions should provide the data necessary to monitor possible substitution.

B. Despite past difficulties in quantifying the substitution effect, states must evaluate the nature, scope, and levels of substitution as required under Title XXI.

C. There are various types of data states may need to collect in order to measure substitution.

D. States have identified the need for additional research.

VII. Discussion

Appendix A: Bibliography

Appendix B: Participants in the September 9th Roundtable Discussion

Appendix C: Overviews of State Programs

California Kids

Colorado Children's Health Plan

Florida Healthy Kids Corporation

Children’s Medical Security Plan of Massachusetts

Minnesota Care

New York Child Health Plus

Oregon Health Plan

RITECARE (Rhode Island)

Washington Basic Health Plus

Connect with Us