Mental Health Parity: Federal and State Action and Economic Impact

Contents

• Introduction
• Economic Factors Opposing Parity
• Impact of Managed Behavioral Health Care
• State Mental Health Parity Laws
• Mental Health Parity Act of 1996
• Enforcement of MHPA
• Enforcement Provisions
• Tax Code
• ERISA
• PHSA
• Existing Enforcement
• Federal Employees Health Benefits Program
• Mental Health Parity Legislation
• 110^th Congress
• Legislative History
• Impact of Mental Health Parity on Health Care Costs
• Coverage of DSM-IV Mental Disorders
• Issues for Congress
• Persistent Mental Health Benefit Limitations
• Financial Protection and Access to Quality Care

Summary

In the 110^th Congress, the Senate and House have passed different versions of expanded mental health parity legislation (S. 558 and H.R. 1424). These bills have always been strongly supported by advocates for the mentally ill and have had broad, bipartisan support in Congress. Although employers and health insurance groups opposed the legislation in the past because of concern that it would drive up costs, the provisions in S. 558 now have their support. Expanded parity legislation was introduced in the 107^th, 108^th, and 109^th Congresses, but each time it failed to pass.

Private health insurers often provide less coverage for mental illnesses than for other medical conditions. Historically, health plans have imposed lower annual or lifetime dollar limits on mental health coverage, limited treatment of mental health illnesses by covering fewer hospital days and outpatient office visits, and increased cost sharing for mental health care by raising deductibles and copayments. The lack of parity (i.e., equivalence) in part reflects insurers' concerns that mental disorders are difficult to diagnose, and that mental health care is expensive and often ineffective. However, the 1999 Surgeon General's report on mental health concluded that mental illnesses are largely biologically based disorders, like many other medical conditions. It found that effective treatments exist for most mental disorders.

Differences in insurance coverage of mental illnesses and other medical conditions are also the result of economic factors. Studies indicate that demand response of mental health patients to reduced cost sharing is approximately twice as large as that observed in general medical care. Partly as a consequence, insurers impose higher cost sharing for mental health. Insurers have also restricted mental health coverage to protect themselves against adverse selection (i.e., the tendency for plans with generous mental health coverage to attract patients with mental illnesses that are costly to treat). Health plans frequently subcontract management of the mental health component of their benefits to specialized managed behavioral health care organizations (MBHOs). Recent studies have shown that there is no significant increase in mental health costs to the insurer when parity is implemented in the context of managed care. Despite this finding, introduction of managed behavioral health care, and passage of state parity laws, mental health coverage continues to be subject to more limitations and higher cost sharing than coverage of other illnesses. Some analysts argue that parity alone is not sufficient to guarantee equal access to quality care and equal financial protection for people with mental disorders.

Twenty-eight states have laws that mandate full-parity mental health coverage, though these laws do not apply to self-insured group plans. In 1996, Congress enacted the Mental Health Parity Act (MHPA), which is more limited in scope and does not compel insurers to provide full-parity coverage. For group plans that choose to offer mental health benefits, the MHPA requires parity only for annual and lifetime dollar limits on coverage. Group plans may still impose more restrictive treatment limitations and cost sharing requirements on their mental health coverage. On June 17, 2008, the President signed into law (P.L. 110-245) the extension of MHPA through 2008.

---

Mental Health Parity: Federal and State Action and Economic Impact

Introduction

In the 110^th Congress, there is a renewed push to expand existing federal mental health parity law. Legislation introduced in the House (H.R. 1424) by Representatives Kennedy and Ramstad, and in the Senate (S. 558) by Senators Domenici and Kennedy, have passed in the respective chambers. On March 5, 2008, the House passed H.R. 1424 by a roll call vote.¹ On September 18, 2007, the Senate passed S. 558 with an amendment, by a voice vote. The Congressional Budget Office (CBO) scored S. 558 and H.R. 1424 and estimated that, if enacted, both bills would increase health insurance premiums by 0.4%.²

On April 29, 2002, in a speech on mental health care during which he announced the formation of the New Freedom Commission on Mental Health, President Bush urged Congress to enact legislation that would provide full parity in the health insurance coverage of mental and physical illnesses. The President identified unfair treatment limitations placed on mental health benefits as a major barrier to mental health care. Historically, private health insurers have provided less coverage of mental illnesses compared to other medical conditions. For example, health plans have imposed lower annual or lifetime dollar limits on mental health coverage, limited treatment of mental health illnesses by covering fewer hospital days and outpatient office visits, and increased cost sharing for mental health care services. Under full parity, a plan must use the same treatment limitations and financial requirements in its mental health coverage as it does in its medical and surgical coverage.³

The New Freedom Commission endorsed mental health parity in its final report, issued on July 22, 2003. "The commission strongly supports the President's call for federal legislation to provide full parity between insurance coverage for mental health care and physical health care," the report said, in reference to the President's April 2002 address.⁴

In 1996, Congress enacted the Mental Health Parity Act (MHPA), which established new federal standards for mental health coverage offered by group health plans. However, the MHPA is limited in scope and does not compel health plans to offer full-parity mental health coverage. It requires group health plans that choose to provide mental health benefits to adopt the same annual and lifetime dollar limits on their coverage of mental and physical illnesses. Plans may still impose more restrictive treatment limitations or cost sharing requirements on their mental health coverage. Lawmakers recently reauthorized the MHPA through December 31, 2008.

Senators Domenici and Wellstone introduced full-parity legislation (S. 543) in the 107^th Congress; the measure saw some action but failed to pass. The legislation (S. 486) was reintroduced at the beginning of the 108^th Congress by Senators Domenici and Kennedy. An identical House bill (H.R. 953) was introduced by Representatives Kennedy and Ramstad. No legislative action was taken on either bill. In the 109^th Congress, the MHPA was introduced in the House (H.R. 1402) by Representatives Kennedy and Ramstad. No legislative action was taken on this bill, and no corresponding legislation was introduced in the Senate. For an analysis of the legislative history of federal parity legislation, see CRS Report RL33820, The Mental Health Parity Act: A Legislative History, by [author name scrubbed] and [author name scrubbed].

Patient advocacy groups and health care provider organizations that support mental health parity argue that there is no longer any scientific justification for discrimination in mental health coverage, which they believe only reinforces the stigma that many in society attach to mental illness. Their efforts to combat discrimination received a boost with the release of the 1999 Surgeon General's Report on Mental Health.⁵ The report reviewed the extensive scientific literature on mental health and concluded that mental illnesses were largely biologically based disorders like many other medical conditions. It also found that the efficacy of mental health treatments is well documented, and that effective treatments exist for most mental disorders. Proponents of mental health parity highlight the high costs to society of untreated and undertreated mental illness. The Substance Abuse and Mental Health Services Administration estimated that the direct treatment costs of mental illness in 2001 totaled $85 billion.⁶ This does not include the economic costs of lost productivity due to illness, lifetime lost productivity, and other indirect costs.⁷

Employer and health insurance associations oppose parity legislation because of concerns that it will drive up costs. Parity supporters refute those claims, pointing to recent studies that indicate that full parity can be implemented without substantial cost increases within the context of comprehensively managed behavioral health care.

Twenty-eight states have passed laws mandating full-parity mental health coverage. Except Wyoming, all other states have enacted legislation that requires health plans to provide certain specified mental health benefits, but not necessarily full parity. However, employers who have self-insured plans (i.e., the employers pay physicians and hospitals directly) are not bound by state insurance regulations.

This report briefly summarizes the economic forces that help explain the persistent limitations on mental health coverage in conventional, fee-for-service (indemnity) health plans. It also discusses issues relating to the feasibility of parity under managed care. In the past decade managed care has transformed the delivery of mental health care services. Employers and health plans now frequently contract out administration of their mental health benefits to specialty managed behavioral health care organizations. The report then reviews state mental health parity legislation and compares the Senate and House versions of the latest full-parity legislation (S. 558 and H.R. 1424) in the 110^th Congress. It concludes with a discussion of some of the key issues in the ongoing congressional debate on mental health parity and discusses why parity may not be enough to provide equivalent financial protection and access to quality care for persons with mental disorders.

Appendix A provides a side-by-side comparison of the key provisions in the engrossed versions of S. 558 and H.R. 1424. Appendix B summarizes state mental health parity laws. Appendix C lists, by committee, the mental health parity hearings since the 106^th Congress. Finally, Appendix D provides websites and annotations of various patient advocacy groups and professional associations that have taken a position on mental health parity.

Economic Factors Opposing Parity⁸

Most mental health care used to be delivered and financed by state-run institutions that provided medical treatment, room and board, and vocational activities for individuals with severe psychiatric disorders. In the 1960s, the role of state governments in mental health began to diminish as alternative forms of outpatient and community-based care gradually replaced institutional care. The mental health system over time started to resemble the general health care system, financed by a combination of private and public insurance. However, private insurance coverage for mental health care no longer included some of the nonmedical services provided by state institutions, such as accommodation and employment. In addition, mental health coverage tended to be more restrictive than the coverage for physical illnesses and surgery and include a higher level of cost sharing. The lack of parity in insurance coverage in part reflected insurers' concerns that the costs of mental health care were high and unpredictable. Insurers argued that mental disorders were difficult to define, and that treatments involving long-term, intensive psychotherapy and extended hospital stays were expensive and often ineffective.

Although stigma has played and continues to play an important role in the mental health care debate, differences in insurance coverage of mental illnesses and other medical conditions are also the result of important economic factors. Studies of indemnity insurance have found that the moral hazard problem is more pronounced for mental health care than it is for general medical care. Moral hazard refers to the tendency for patients to demand more services as the price they pay for those services declines. While health insurance, in general, creates incentives for overuse by insulating patients from the total costs of care, research shows that the demand response to reduced cost sharing in mental health care is approximately twice as large as that observed in general medical care.⁹ The result has been for insurers to impose higher cost sharing for mental health care.

Insurers have also restricted their mental health coverage to protect themselves from adverse selection. Adverse selection refers to the tendency for health plans with generous coverage provisions to attract sick (i.e., high-cost) enrollees. The evidence suggests that adverse selection may be an especially powerful force in mental health care. Studies indicate that individuals with mental illness select health plans that offer more generous mental health coverage.¹⁰ Such behavior can create a strong economic incentive for health plans to reduce their attractiveness to users of mental health care.

While competition among plans to avoid enrolling high-risk individuals may represent a good strategy for an individual plan, health economists argue that it is wasteful and inefficient for the health insurance system as a whole. Competition among indemnity insurance plans is seen as an important factor in reducing the level of coverage for mental health care. During the 1970s and 1980s, this argument was used to justify federal and state mandated benefit laws that required insurers to cover minimum levels of mental health care (see below). Some health economists claim that parity can improve the efficiency of insurance markets by reducing wasteful forms of competition that are the result of adverse selection. These commenters note that requiring parity for mental health benefits establishes a uniform "floor" of mental health coverage across all plans. Supporters of parity maintain that studies show that implementing parity through managed mental health care will not lead to a significant increase in costs to the insurers.

Impact of Managed Behavioral Health Care

The movement to establish parity for mental health care has been fueled by changes in the scientific understanding of mental illness and the rapid increase in managed behavioral health care (i.e., mental health and substance abuse treatment). Recently revised estimates suggest that about 15% of the adult U.S. population (approximately 30 million individuals) are affected by a clinically significant mental disorder in any given year.¹¹ Clinicians are often able to diagnose mental illness with precision, and effective treatments now exist for many psychiatric conditions. Some studies show that the effectiveness of treatments for major mental disorders, which typically involve a combination of medication and psychotherapy, often match or exceed the effectiveness of common treatments for physical illness.¹² As noted earlier, proponents of parity argue that untreated and undertreated mental illness has a major impact on the economy and costs employers tens of billions of dollars annually in lost productivity.

Health plans frequently subcontract, or carve out, to managed behavioral health care organizations (MBHOs) the management of the mental health (and substance abuse) component of their benefits package. Over the past few years, behavioral carve-outs have become central to the delivery and payment of mental health care.¹³ Nearly three-quarters of Americans with health insurance are covered by managed health benefits. SAMHSA has published a report that explores the current design and administration of mental health coverage. The report lays out the necessary components of an adequate mental health benefit by examining the evidence-base for particular benefits; effect on access, utilization, and costs; the components of a cost-effective mental health benefit package; and the effects of benefits administration on effectiveness.¹⁴

Managed care is changing the way in which mental health services are provided. Whereas conventional fee-for-service insurance controls the demand for services primarily through cost sharing (e.g., deductibles, co-payments) and treatment limitations, MBHOs influence the treatment decisions of mental health care providers through a variety of techniques, including financial incentives, greater emphasis on preventive medicine, development of treatment protocols, and prior authorization of certain services. Cost sharing and coverage limits assume less importance under managed care, which seeks to control moral hazard by internal rationing methods, rather than having to rely on demand-side cost sharing.

The introduction of managed mental health care can reduce spending and in some cases increase plan usage. For example, Pacific Bell lowered its mental health expenditures by 13% when it implemented managed behavioral health care in the early 1990s.¹⁵ The cost reduction was not attributable to decreased initial access to care. The number of persons using any mental health care actually increased following the change. Instead, the cost reduction was the result of fewer outpatient sessions per patient, a reduced likelihood of inpatient admission, a reduction in the length of stay for those admitted as inpatients, and significantly lower costs per unit of service delivered. Massachusetts saw a 25% decline in behavioral health care costs for state employees as a result of introducing managed care in 1993.¹⁶

Studies also suggest that MBHOs are able to control the costs associated with mental health parity. In 1996, estimates of the cost of implementing full parity ranged as high as 11% of the total health care premium, which led Congress to limit parity-level benefits in the MHPA. But those estimates did not adequately reflect the impact of managed care and increased use of psychotropic drugs and short-term psychotherapeutics on controlling costs. More recent studies in states that have enacted full-parity laws for mental health coverage provided by managed care plans found that premium increases have been modest.¹⁷ Magellan Health Services, the nation's largest MBHO covering nearly 70 million individuals, reported that it had yet to see a premium cost increase of more than 1% as a result of implementing state mental health parity legislation.

At a 2001 Robert Wood Johnson Foundation workshop on the costs of mental health parity, actuaries, economists, and government officials discussed the assumptions and methods used in calculating parity cost estimates. There was broad agreement among the workshop participants that the baseline level of mental health spending has decreased significantly as a result of changes in clinical practice (e.g., use of psychotropic drugs and short-term psychotherapies) and the growth of managed care. Baseline mental health spending is often represented by the share of the total health insurance premium spent on mental health services without parity. Changes in premium costs that result from parity are then expressed as a percentage change in baseline. Workshop participants were also in agreement that managed care has had an important affect on the impact of parity laws. Managed care plans have responded to the expansion of benefits under parity by tightening their internal controls on the use of mental health services so as to dampen any increase in demand and premiums.¹⁸

State Mental Health Parity Laws¹⁹

States began to address inequities in mental health coverage in the 1970s. More than a dozen states enacted laws requiring health plans operating within the state to offer a specific set of mental health benefits. While these mandated-benefit laws increased coverage, they had important limitations. They seldom provided catastrophic coverage against the financial risk of severe mental illness and they did not apply to self-insured employers, which are exempt from state regulation under the Employee Retirement Income Security Act (ERISA).

In 1991, Texas and North Carolina became the first states to enact mental health parity legislation. Both state laws were limited in their scope and applied only to insurers that covered state and local government employees. By 1996, when federal parity legislation was enacted (see below), a total of seven states had passed laws that required certain specified state-regulated health plans to provide full-parity mental health coverage. Since then, more than a dozen other states have passed similar legislation, bringing to 28 the total number of states that now mandate mental health coverage with full parity.

State laws that mandate full-parity mental health benefits vary in the types of health plans covered and also in the types of mental illnesses they cover. In 15 states, the laws apply both to group health plans and to the individual health insurance market, whereas in another 9 of these states they apply only to group plans. In the remaining four states, the laws apply only to state-employee plans. In 12 of the states with full parity laws, the laws apply to the treatment of all the conditions listed in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV).²⁰ The other parity laws restrict coverage to specified "serious" or "biologically based" mental illness (e.g., schizophrenia, depression, bipolar disorder). About one-third of the state parity laws exempt small employers, typically those with 50 or fewer employees.

In addition to the 28 states that have enacted full parity legislation, 6 states have passed laws mandating a certain minimum level of mental health benefits (but not full parity). Fourteen other states have passed so-called mandated offering laws, under which covered plans that choose to offer mental health coverage must provide a specified minimum level of benefits. The bills currently introduced in the House and Senate are mandated offering laws. See Appendix B for a summary of state parity laws.

New Jersey, which in 1999 enacted a full parity law that covers both group plans and the individual market, recently passed legislation that requires individual carriers to offer a policy with minimum mandated mental health benefits. Those benefits include coverage for 90 days of inpatient treatment with a $500 copayment per inpatient stay, and 30 days of outpatient treatment with a 30% coinsurance. The new law does not replace the existing full parity mandate, but is intended to provide individuals with a less expensive alternative to a policy with full-parity coverage. The aim of the law is to allow individuals who might otherwise not be able to afford a policy with full parity to purchase insurance coverage. Texas has also enacted new legislation that allows for the sale of less expensive health insurance policies without state mandates for the treatment of mental illness. An insurer that offers such a policy must also provide at least one policy with state-mandated health benefits.

Mental Health Parity Act of 1996

The Mental Health Parity Act (MHPA) amended ERISA and the Public Health Service Act (PHS Act) and established new federal standards for mental health coverage offered by group health plans, most of which are employment based.²¹ Identical provisions were later added to the Internal Revenue Code (IRC) by the Taxpayer Relief Act of 1997.²² The MHPA is not a full-parity law. It requires equivalence in only one area: catastrophic coverage. The MHPA prohibits group plans from imposing annual and lifetime dollar limits on mental health coverage that are more restrictive than those imposed on medical and surgical coverage. Group plans may still impose more restrictive treatment limitations or cost sharing requirements on their mental health coverage compared to their medical and surgical coverage.

The MHPA includes several other important limitations. Group plans that choose not to provide mental health benefits are not required to add them, and employers with 50 or fewer employees are exempt from the law. In addition, employers that experience an increase in claims costs of at least 1% as a result of MHPA compliance can apply to the Department of Labor for an exemption.

The MHPA standards apply to private-sector, employer-sponsored group health plans, including fully insured and self-insured plans, but not to the individual (nongroup) health insurance market. They also apply to the Federal Employees Health Benefits Program and to some state and local government health plans. Under provisions included in the 1997 Balanced Budget Act (P.L. 105-33), Medicaid managed care plans and State Children's Health Insurance Programs also have to comply with the requirements of the MHPA.²³ The MHPA does not apply to Medicare.

In 1999, the General Accounting Office (GAO) reviewed the extent to which employers were complying with the MHPA and how they had revised their health plans.²⁴ GAO surveyed 863 employers in 26 states without full parity laws. While 86% of the employers reported compliance with the MHPA, a majority of these plans (87%) restricted their mental health coverage in other ways. For example, about two-thirds of MHPA-compliant plans covered fewer outpatient visits and hospital days for mental health treatment than for other medical treatment. Surveys by the Labor Department and the Centers for Medicare & Medicaid Services found similar results.

Many plans that had to increase annual and lifetime dollar limits to comply with the MHPA reportedly introduced other more restrictive mental health design features to mitigate the financial impact of the law's more generous dollar limits. Despite concerns about the MHPA's effect on claims costs, only 3% of employers surveyed by GAO reported that their costs had increased, and less than 1% of surveyed employers dropped their mental health coverage altogether following the law's enactment. It is difficult to gauge the impact of the MHPA's increased dollar limits, however, because many plans took steps to counter increases in claims costs by restricting mental health coverage in other ways.

Though limited in its scope, the MHPA nevertheless appears to have added momentum to the passage of state parity laws. All states, except Wyoming, have passed some form of parity legislation since the federal law was enacted in 1996. Some states passed parity laws that essentially mirrored the MHPA, and later strengthened the laws to exceed the provisions of the federal law.

The MHPA originally sunset on September 30, 2001. In three separate legislative actions, the 107^th Congress extended the MHPA through the end of 2003. Title VII of the FY2002 Labor-HHS-Education appropriations bill (H.R. 3061, P.L. 107-116) reauthorized the MHPA in all three federal statutes through December 31, 2002. Section 610 of the Job Creation and Worker Assistance Act of 2002 (H.R. 3090, P.L. 107-147) further amended the MHPA provisions in the IRC—but not in ERISA or the PHS Act—by extending the authorization an additional year through December 31, 2003. Finally, the Mental Health Parity Reauthorization Act of 2002 (H.R. 5716, P.L. 107-313) reauthorized the MHPA provisions in ERISA and the PHS Act through December 31, 2003.

The 108^th Congress extended the MPHA through the end of 2005. First, the Mental Health Parity Reauthorization Act of 2003 (S. 1929, P.L. 108-197) reauthorized the MHPA through December 31, 2004. The bill amended the MHPA provisions in ERISA and the PHS Act, but not the IRC. Section 302 of the Working Families Tax Relief Act of 2004 (H.R. 1308, P.L. 108-311) reauthorized the MHPA through December 31, 2005. P.L. 108-311 amended the MHPA provisions in all three statutes.

The 109^th Congress extended MPHA through the end of 2007. In the first session of the 109^th Congress, the Employee Retirement Preservation Act (H.R. 4579, P.L. 109-151) extended the provisions requiring mental health parity in ERISA, the PHS Act, and the IRC through 2006. In the second session, Section 115 of the Tax Relief and Health Care Act of 2006 (H.R. 6111, P.L. 109-432) extended the MPHA provisions in all three statutes through 2007.

The second session of the 110^th Congress extended MHPA through the end of 2008. On June 17, 2008, Sec. 401 of the Heroes Earnings Assistance and Relief Tax Act of 2008 (P.L. 110-245, H.R. 6081) passed an extension of MHPA through the end of 2008. The law amends ERISA, the PHS Act, and the IRC.

Enforcement of MHPA

By amending all three federal statutes (i.e., ERISA, the PHS Act, and the IRC), the MHPA standards apply to a broad range of group health plans, as well as state-licensed health insurance organizations. The ERISA provisions apply to most group plans sponsored by private-sector employers and unions. The IRC provisions, which cover ERISA plans plus church-sponsored plans, permit the Internal Revenue Service to assess tax penalties on employers that do not comply with the MHPA requirements. The PHS Act provisions apply to insurers and some public-sector group health plans. Self-insured state and local government health plans may elect exemption from the MHPA

Although states have taken on primary responsibility for the enforcement of many of the mandates as they apply to health insurers, other enforcement actions are available to the Secretaries of the Department of Labor, Department of Health and Human Services, and the Internal Revenue Service. The enforcement provisions that apply through the MHPA (ERISA, Tax Code, and PHSA) are described more specifically below.

Enforcement Provisions

Tax Code

Tax penalties for violations of federal health mandates take the form of excise taxes that are imposed on employers or, in the case of multiemployer plans, on the health plans. The excise tax for violations of certain group health plan requirements of the MHPA is specified in section 4980D of the Code. The taxes are payable to the U.S. Treasury.

In general, the excise taxes are $100 per day of noncompliance for every qualified beneficiary. When violations are discovered after notice of examination, the minimum tax is $2,500, or $15,000 if violations are more than de minimus. The minimum tax does not apply to church plans. When violations are due to reasonable cause and not willful neglect, the maximum tax during a taxable year cannot exceed the lesser of $500,000 or the employer's group health plan expenses for the prior year. In addition, when violations are due to reasonable cause and not willful neglect, the Secretary of the Treasury may waive part or all of the tax if payment would be excessive relative to the failure involved. No taxes apply if failures were not discovered when exercising reasonable diligence or if failures are corrected within certain periods.

Governmental plans are not subject to the excise taxes. In addition, with respect to violations of HIPAA's prohibition against discrimination based on health status, certain small church-sponsored plans are not subject to penalties.

ERISA

Section 502 of ERISA establishes a civil enforcement scheme for violations of the statute. In general, ERISA provides only for the recovery of benefits due to a participant or beneficiary under the terms of a plan, or for declaratory or injunctive relief. Courts have uniformly held that other monetary or damage remedies are not available.

PHSA

Enforcement of group health provisions is described in Section 2722 of the PHSA as established by Section 102 of HIPAA. In cases in which HHS is required, because states have not taken on this responsibility, to enforce group market rules regarding preexisting condition exclusion periods, discrimination, guaranteed availability and renewability, and information disclosure, the Secretary may impose civil money penalties on non-conforming health insurance plans. The penalties available include $100 per day for each day and for each individual when such a failure occurs. In imposing the penalty, the Secretary may consider the previous record of compliance of the entity being assessed. There are limitations on this penalty as well. Penalties cannot be applied for failures that are corrected within 30 days of discovery and under other limited circumstances. In addition, the entity assessed may request an administrative review consisting of an initial hearing and judicial review and appeal.

Existing Enforcement

Little information is available on enforcement actions related to MHPA by the three agencies.

A 2003 publication available from the Pension and Welfare Benefits Administration (PWBA) of the Department of Labor, however, summarizes some of the enforcement activity as part of a 2001 compliance project. The project was undertaken to give the Agency a baseline for assessing compliance with the health mandates. PWBA conducted 1,267 investigations of group health plans during 2000-2001 for compliance with 42 specific requirements of the health laws. At that time, just over one-third of plans were out of compliance with at least 1 of 36 substantive provisions of the new laws. Compliance rates dropped further when six provisions requiring plan sponsors to provide notice to enrollees for various reasons were calculated into the rates. As a result of this work, the department initiated a program to help employer plans come into compliance with the laws. The program included additional publications and educational materials, a Web page devoted to compliance assistance, and live workshops around the country.

The HHS Centers for Medicare and Medicaid Services were unable to provide information on the use of the civil enforcement penalties under the PHSA. A spokesman for the agency pointed out that such information has not been made publically available to date.

Similarly, a contact at the Internal Revenue Service informed CRS that tax penalties are not tracked in a manner that would allow the separate identification of amounts assessed or collected only under sections 4980B and 4980D of the tax code.

Federal Employees Health Benefits Program

At the White House Conference on Mental Health in June 1999, President Clinton directed the federal Office of Personnel Management (OPM) to implement full parity for both mental health and substance abuse benefits in health plans offered under the Federal Employees Health Benefits Program (FEHBP) beginning in 2001. The FEHBP parity requirement covers medically necessary treatment for all categories of mental illness listed in the DSM-IV. According to the OPM, parity implementation resulted in an average premium increase of 1.64% for fee-for-service plans and 0.3% for HMOs. FEHBP health plans are providing mental health coverage in a variety of ways. Some plans are using the services of managed behavioral health care organizations, while others are managing their own provider networks. Under FEHBP, mental health parity is required only for services provided on an in-network basis. In-network generally refers to a contracted group of providers established by a managed health care organization and/or an insurance carrier. OPM and the Department of Health and Human Services are conducting a three-year evaluation of the FEHBP parity initiative.²⁵

A recently published study comparing FEHB plans with health plans outside FEHBP that did not mandate parity concluded that implementation of parity in insurance benefits for mental health and substance abuse, when coupled with management of care, resulted in little or no significant adverse impact on access, spending, or quality, while providing users of behavioral health care with improved financial protection in most instances.²⁶ The researchers analyzed plan benefits data for seven FEHB plans both before (1999 and 2000) and after (2001 and 2002) the introduction of parity. Changes in access, utilization, and cost were compared to changes over the same time period in a matched set of non-FEHB comparison plans (mostly large, self-insured employers). The analysis indicated that the observed increase in the rate of use of mental health and substance abuse services in FEHB plans after implementation of the parity policy was due almost entirely to a general trend in increased use that was observed in the comparison plans.²⁷ Furthermore, compared to spending trends observed in the non-FEHB plans, the implementation of parity was associated with significant reductions in out-of-pocket spending in five of the seven FEHB plans.²⁸

Mental Health Parity Legislation

110^th Congress

On February 12, 2007, Senators Domenici and Kennedy introduced the Mental Health Parity Act of 2007 (S. 558) to amend and expand the MHPA by requiring employer-sponsored group health plans to impose the same treatment limitations and financial requirements on their mental health coverage as they do on their medical and surgical coverage. This bill would amend ERISA and the PHS Act. The bill was referred to the Senate Health, Education, Labor, and Pensions Committee, where it was marked up on February 14, 2007 (S.Rept. 110-53). The Senate passed S. 558, with an amendment, by voice vote on September 18, 2007. On March 7, 2007, Representatives Kennedy and Ramstad introduced the Paul Wellstone Mental Health and Addiction Equity Act (H.R. 1424), which would amend ERISA, the PHS Act, and the IRC. On July 18, 2007, the House Education and Labor Committee approved H.R. 1424, with an amendment. The measure, as amended, was approved by the House Ways and Means Committee on September 26, 2007, and by the House Energy and Commerce Committee on October 16, 2007 (H.Rept. 110-374). The House passed H.R. 1424 by a roll call vote of 268-148 on March 5, 2008.

H.R. 1424 and S. 558 do not mandate full parity. Like the MHPA, they apply only to group plans that choose to offer mental health coverage. H.R. 1424, which is modeled on the parity requirements in the FEHBP, covers the treatment of all psychiatric conditions listed in the DSM-IV. On the other hand, S. 558's parity provisions apply only to conditions that the health plan chooses to cover. Both bills would extend parity to in-network and out-of-network mental health benefits. Both bills also exempt small employers with 50 or fewer employees. In an effort to address some of the concerns of the health insurance industry, H.R. 1424 includes language permitting employers and health plans to manage mental health benefits and cover only those treatment services that are medically necessary. Finally, the bills require GAO, within two years, to evaluate the impact of the new federal parity standards on access to insurance coverage and on insurance costs.

There are two key differences between the House and Senate parity bills. The first difference is that the Senate version would allow insurance companies to determine which mental illnesses they cover, whereas the House version would require coverage for all mental illnesses (see discussion below). The second key difference is that the Senate version does not explicitly address preemption of state mental health parity laws, whereas the House version explicitly states that it does not preempt state mental health parity laws. For an analysis of the differences between the House and Senate parity bills, see Appendix A. Unlike previous versions of expanded parity legislation, the House and Senate versions in the 110^th Congress include parity for substance abuse treatment services. Another difference between the Senate bill and the previous parity bills is that S. 558 has the support of insurance companies and employers.

Bills that would amend Title XXI of the Social Security Act to provide for equal coverage of mental health services under the State Children's Health Insurance Program passed in the House on August 1, 2007 (H.R. 3162), and in the Senate on August 3, 2007 (H.R. 976).

Legislative History

Senators Domenici and Wellstone first introduced the Mental Health Equitable Treatment Act (S. 543) on March 15, 2001. In its June 2000 report to Congress, the National Advisory Mental Health Council (NAMHC) estimated that full parity similar to that provided by S. 543 would raise premium costs by 1.4%, adding that this figure may overestimate the true cost of parity because the forecasting models did not reflect the most recent changes in managed care. Pricewaterhouse Coopers concluded that S. 543 would result in a 1% increase in costs, or $1.32 per enrollee per month.²⁹ CBO estimated that, on average, S. 543 would increase premiums for group health plans by 0.9%.³⁰

On August 1, 2001, the Senate Health, Education, Labor, and Pensions (HELP) Committee approved a substitute version of S. 543 (S.Rept. 107-61), which retained most of the major components of the original bill including the full-parity requirement. On October 30, 2001, the Senate added S. 543 as an amendment to the FY2002 Labor-HHS-Education appropriations bill (H.R. 3061). The House version of the Labor-HHS-Education appropriations bill did not include any parity language. During the conference on H.R. 3061, House conferees rejected the Senate amendment on a party-line vote. Unable to agree on new federal parity standards, the conference voted to reauthorize the MHPA through December 31, 2002. Conferees added language to the conference report (H.Rept. 107-350) "strongly urging the committees of jurisdiction in the House and Senate to convene early hearings and undertake swift consideration of legislation to extend and improve mental health parity protections during the second session of the 107^th Congress."

The House Education and the Workforce Subcommittee on Employer-Employee Relations held a hearing on mental health parity on March 13, 2002, followed by a House Energy and Commerce Health Subcommittee hearing on July 23, 2002. On March 20, 2002, Representative Roukema introduced the Senate parity legislation in the House (H.R. 4066), but there was no further legislation action taken before the 107^th Congress adjourned.

Representatives Patrick Kennedy and Jim Ramstad, and Senators Domenici and Kennedy introduced full parity legislation on February 27, 2003, which included the same language as S. 543, as reported by committee, in the 107^th Congress. During the second session, Senator Daschle introduced the Paul Wellstone Mental Health Equitable Treatment Act of 2003 on November 6, 2003. This bill bears the name of the late Senator Paul Wellstone, who was killed in a small plane crash on October 25, 2002. No further legislative action was taken on this bill in the 108^th Congress.

The Paul Wellstone Mental Health Equitable Treatment Act of 2005 was again introduced by Representatives Patrick Kennedy and Jim Ramstad on March 17, 2005. No legislative action was taken on this bill during the 109^th Congress, and no corresponding legislation was introduced in the Senate.

For a more detailed legislative history of these bills, see CRS Report RL33820, The Mental Health Parity Act: A Legislative History, by [author name scrubbed] and [author name scrubbed].

Impact of Mental Health Parity on Health Care Costs

Federal full-parity legislation has staunch support among patient advocates and mental health provider organizations, who see it as an important step in eliminating what they characterize as inappropriate discrimination in private health insurance coverage of mental illness. But groups representing employers and the health insurance industry strongly oppose the legislation on the grounds that it will add significantly to the dramatically rising costs of health care. They argue that employers cannot afford to spend more money on health insurance coverage for their employees in the current economic climate. They contend that parity costs would likely take the form of increased cost sharing for all covered benefits, reductions in other health care coverage, and/or the elimination of health coverage entirely, which would lead to an increase in the number of uninsured.

Proponents of parity legislation counter that full parity does not significantly increase costs under managed care. They argue that parity can in fact reduce costs to employers by improving productivity and reducing absenteeism. Furthermore, they claim that full-parity coverage lowers overall health care expenditures by eliminating the need for medical care and emergency room visits that result if mental illnesses are left untreated. Some large employers have reported that parity in mental health benefits has had a net positive financial impact. As an example, they cite Delta Airlines. Delta increased mental health benefits for its 69,000 employees in 1994, when it switched to managed care. Use of mental health services increased but costs remained flat. Spending in other areas of health care declined and employees missed less work.³¹

The Congressional Budget Office (CBO) scored S. 558 and H.R. 1424 and estimated that, if enacted, both bills would increase health insurance premiums by 0.4%.³²

Coverage of DSM-IV Mental Disorders

Employers and health insurers are especially concerned about the broad definition of mental illness in H.R. 1424. They believe that federal parity legislation should cover only serious mental illnesses or illnesses that have been shown to be related to the biological functioning of the brain (e.g., schizophrenia, bipolar disorder), as do many state laws. S. 558, which covers only mental illnesses that are specifically included by the health plan, has the support of employers and health insurers. Critics of H.R. 1424 claim that extending coverage to all the mental disorders listed in the DSM-IV opens the door to dubious complaints of less serious problems by the "worried well." They object to providing coverage for many of the conditions that are classified as mental disorders in the DSM-IV (e.g., academic skills disorders, sexual desire disorders) because they are not seen as medically significant.

Parity supporters view opposition to providing coverage of all DSM-IV disorders as stemming, in part, from stigma and the mistaken belief that mental illness does not have a physiological basis. They claim that restricting mental health coverage to a few specified psychiatric conditions is no different than having medical benefits that cover only serious physical disease such as cancer and heart disease. They argue that covering all the DSM-IV disorders is unlikely to lead to abuse or inflated costs for two reasons. First, H.R. 1424 does not prevent plans from managing mental health benefits through such practices as utilization review, preauthorization, the application of medical necessity and other appropriateness criteria, and through the use of provider networks. Second, the DSM-IV establishes a threshold for diagnosis by requiring evidence of "clinically significant impairment or distress." Any claims for treatment of a patient with a mental health condition that was not serious enough to meet that threshold could be excluded on the basis of medical necessity. Advocates of mental health parity also assert that restricting coverage to a few major mental illnesses is penny-wise and pound-foolish. They point out that milder forms of emotional illness often worsen into more serious psychiatric disorders, if left untreated.

Issues for Congress

Persistent Mental Health Benefit Limitations

National employer survey data indicate that despite the passage of state parity laws and changes in the delivery of mental health services, mental health coverage is still not offered at a level comparable to coverage for other medical conditions. A recent analysis of the 2002 Kaiser Family Foundation/Health Research and Educational Trust (KFF/HRET) Employer Health Benefits Survey found that overall 98% of workers with employer-sponsored health insurance had coverage for mental health care.³³ However, 74% of those covered workers were subject to an annual outpatient visit limit, and 64% were subject to an annual inpatient day limit. The proportion of covered workers subject to annual mental health day and visit limits appears to have increased over the past few years. In contrast, the survey found only 22% of covered workers had higher cost sharing (i.e., copayment or coinsurance) for mental health benefits. This suggests that health plans are relying less on higher cost sharing as a means of limiting the use of mental health services.

The 2002 KFF/HRET survey data indicate that about one-third of workers with employer-sponsored health insurance receive their mental health care through carve-outs. Surprisingly, the investigators found relatively little difference in the nominal mental health benefits (i.e., the treatment limitations and cost sharing requirements spelled out in the insurance contract) under carve-outs versus integrated health plans. Carve-outs and integrated plans had similar limitations on the number of inpatient days and outpatient visits. There was also no significant difference in the percentage of covered workers with higher cost sharing for outpatient mental health services in carve-outs compared to integrated plans.

Given that MBHOs incorporate supply-side utilization controls rather than relying solely on cost sharing and benefit limits to lower demand, one might expect them to expand mental health benefits while maintaining control over costs. But the KFF/HRET survey data indicate that carve-outs continue to impose special limits and substantial cost sharing on mental health. Some researchers hypothesize that a lack of employer education about the cost advantages of behavioral mental health care management, minimal risk sharing under many carve-out contracts, or a single-minded focus on cost containment could explain why mental health benefit limitations persist. Lingering concerns about adverse selection could also play a role in the persistence of benefit limits.

Overall, the 2002 KFF/HRET survey findings suggest that mental health parity may be difficult to establish without broader (i.e., federal) parity laws. State parity laws have a limited impact because they do not cover self-insured plans. ERISA exempts self-insured plans from state regulation. About 52% of covered workers are in a self-insured plan, according to the KFF/HRET survey.³⁴

Financial Protection and Access to Quality Care

Mental health analysts often argue that parity laws are an important step in improving the efficiency and fairness of insurance coverage for mental illness. But many are concerned that parity in nominal benefits for mental health care, by itself, is not sufficient to guarantee equal access to high-quality care and equal levels of financial protection for people with mental disorders. For one thing, many mental health services do not have any counterpart in general medical care and are, therefore, unaffected by parity legislation because they do not have to be included in covered benefits. Private insurance usually does not cover day-hospital care, psychosocial rehabilitation, or residential treatment programs, all of which are seen as effective components of mental health care. Moreover, health plans do not cover supervised housing or employment for patients with chronic mental health conditions. Proponents of parity maintain that taking a broader view of access to quality mental health care means encompassing a variety of social-welfare services.

Advocates for the mentally ill worry that behavioral health carve-outs may not provide patients with all the appropriate and medically necessary care. While managed behavioral health care has proved effective at controlling the costs of full parity, patient advocates are concerned about management decisions that may result in across-the-board reductions in treatment without regard to clinical circumstances. MBHOs are under pressure to contain costs. The internal management processes that they use to ration treatment are difficult to regulate. Even under federal and state parity laws, MBHOs still retain wide latitude to manage coverage and control access to mental health care in order to achieve cost-control goals. Managed care contacts, with their complex internal rationing devices, are more remote from regulation than the traditional fee-for-service contracts.

MBHOs maintain that by lowering costs and offering parity-level benefits, patients have greater access to treatment at an earlier point in the development of their illness. This, in turn, they hold, results in less suffering and lower costs associated with that treatment. Moreover, they point to studies that have shown that early, effective treatment of mental illness leads to lower medical costs generally, lower disability costs, and less absenteeism in the workplace. But critics of behavioral carve-outs contend that the managed care tools employed by MBHOs are widening the gap between a plan's nominal benefits and the care actually received by patients. In contrast to using a primary care physician as the gatekeeper to more specialized care, which is a model commonly employed in managed care, MBHOs use a larger range of techniques to manage mental health care (e.g., concurrent utilization review by clinical care managers) and use a different mix of providers and services.³⁵

The American Psychiatric Association and the American Medical Association (AMA) have criticized carve-outs as discriminatory because they separate behavioral health care from "mainstream" health care rather than integrating the two, thus reinforcing the notion that behavioral health is somehow different from other medical conditions.

Results of the 1996-1998 Health Care for Communities (HCC) national survey relate to analysts' concerns about the impact of parity on access to quality mental health care. The HCC survey concluded that state parity laws have had no discernible impact on the overall use of mental health services. The survey suggested that utilization of mental health care was no higher in parity states than in states without such laws. HCC researchers said their survey supports the view that the insurance market has responded to parity laws by increasing the management of care in order to control costs. They analyzed the self-reported unmet needs among respondents seeking treatment for mental health and substance abuse problems. When unmet needs was defined as delays in receiving treatment or receiving less treatment than desired, significantly more respondents in managed care reported unmet needs than those enrolled in indemnity insurance. However, when unmet needs was defined as obtaining no care, those in managed care reported unmet needs less often. According to the HCC researchers, results of the survey reinforce concerns about the impact of parity on access to quality health care.³⁶

A recent study of the implementation of Vermont's 1998 parity law also found that the increased use of managed care, while helping make health care more affordable, may have reduced access and utilization for some services and beneficiaries.³⁷ The study examined the experiences of the state's two largest health insurers—Kaiser/Community Health Plan (Kaiser/CHP) and Blue Cross Blue Shield of Vermont (BCBSVT)—which together covered nearly 80% of Vermont's privately insured population at the time the parity law was implemented. Vermont's parity law, one of the nation's most comprehensive, covers both mental health and substance abuse treatment services.

As a result of the law, both plans made changes to their management of mental health and substance abuse (MH/SA) services. Managed care was an important factor in controlling costs following implementation of parity. Before the parity law took effect, BCBSVT provided MH/SA services mainly through indemnity contracts. After parity, most BCBSVT members received those services through a managed care carve-out and reported a decline both in the likelihood of obtaining mental health treatment and in the average number of outpatient visits. Kaiser/CHP, which had managed care prior to the parity law, increased the use of partial hospitalization treatment and group therapy and reduced the use of inpatient treatment. Overall, MH/SA spending fell by 8-18% after parity was implemented, despite lower consumer out-of-pocket costs and higher limits on the use of MH/SA care.

Finally, parity helps only people who have health insurance. It does not address the larger questions concerning the 17.5% of the U.S. population with no health insurance.³⁸

Appendix A. Comparison of Key Provisions in the Mental Health Parity Act of 2007 (S. 558) and the Paul Wellstone Mental Health and Addiction Equity Act of 2007 (H.R. 1424)

Appendix A.

Appendix B. State Mental Health Parity Laws

Appendix C. Mental Health Parity Hearings

Appendix D. Mental Health Parity Websites

Patient Advocacy

• National Alliance for the Mentally Ill http://www.nami.org
• National Mental Health Association http://www.nmha.org
• Bazelon Center for Mental Health Law http://www.bazelon.org
• Suicide Prevention Action Network USA http://www.spanusa.org

Professional Associations: Health Care Providers

• American Psychiatric Association http://www.psych.org
• American Psychological Association http://www.apa.org
• American Medical Association http://www.ama-assn.org
• American Managed Behavioral Healthcare Association http://www.ambha.org

Professional Associations: Employers and Health Plans

• American Health Insurance Plans http://www.ahip.org
• ERISA Industry Committee http://www.eric.org

Footnotes

1.	House Roll Call Vote 101: 268-148, March 5, 2008.
2.	Congressional Budget Office (CBO), Cost Estimate, H.R. 1424 Paul Wellstone Mental Health and Addiction Equity Act of 2007, September 2007, available at http://www.cbo.gov/ftpdocs/86xx/doc8608/hr1424.pdf. Congressional Budget Office (CBO), Cost Estimate, S. 558 Mental Health Parity Act of 2007, March 2007, available at http://www.cbo.gov/ftpdocs/78xx/doc7894/s558.pdf.
3.	Treatment limitations include restrictions on the number of visits or days of coverage, or other limits on the duration and scope of treatment. Financial requirements include deductibles, coinsurance, co-payments, and other cost-sharing requirements, as well as annual and lifetime limits on the total amount of coverage.
4.	New Freedom Commission on Mental Health, Achieving the Promise: Transforming Mental Health Care in America. Final Report. DHHS Pub. no. SMA-03-3832 (July 2003). Available online at http://www.mentalhealthcommission.gov.
5.	U.S. Dept. of Health and Human Services, Mental Health: A Report of the Surgeon General. Available at http://www.surgeongeneral.gov/library/mentalhealth/home.html.
6.	http://www.samhsa.gov/spendingestimates/chapter3.aspx.
7.	The estimated indirect economic cost of mental illness in 1994 included $88.3 billion in lost productivity due to illness; $16.5 billion in discounted (at 6%) lifetime productivity losses as a result of premature death; and $7.8 billion in other related costs, including those associated with crime and incarceration, social welfare administration, and family care giving.
8.	For a detailed discussion of the economics of mental health, see Richard G. Frank and Thomas G. McGuire, "Economics and Mental Health," in Handbook of Health Economics, v. 1B, ed. Anthony J. Culver and Joseph P. Newhouse (Amsterdam: Elsevier, 2000).
9.	J.P. Newhouse and the Insurance Experiment Group, Free for All? Lessons from the RAND Health Insurance Experiment (Cambridge, MA: Harvard University Press, 1993).
10.	For example, see P. Deb et al., "Choice of Health Insurance by Families of the Mentally Ill." Health Economics, v. 5, 1996, pp. 61-76.
11.	William E. Narrow et al., "Revised Prevalence Estimates of Mental Disorder in the United States," Archives of General Psychiatry, v. 59, 2002, pp. 115-123.
12.	The National Institute of Mental Health http://www.nimh.nih.gov estimates the following success rates for treating major mental disorders: schizophrenia (60%); clinical depression (70%-80%); and panic disorder (70%-90%).
13.	More information may be found on the website of the American Managed Behavioral Healthcare Association, whose members are responsible for managing mental health and substance abuse services in the public and private sector for more than 110 million individuals, at http://www.ambha.org.
14.	Sethi R. et al., "Designing Employer-sponsored Mental Health Benefits," SAMHSA Center for Mental Health Services, 2006.
15.	William Goldman et al., "Costs and Use of Mental Health Services Before and After Managed Care," Health Affairs, v. 17, 1998, pp. 40-52.
16.	Ching-to Albert Ma and Thomas G. McGuire, "Costs and Incentives in a Behavioral Health Carve-Out," Health Affairs, v. 17, 1998, pp. 53-69.
17.	Studies in Vermont, Maryland, and Minnesota show that the cost impact of full parity is 1-2%. Details of the studies are available at the American Psychological Association's website at http://www.apa.org/practice/parity_cost.html.
18.	Robert Wood Johnson Foundation, Estimating the Costs of Parity for Mental Health. Available online at http://www.rwjf.org/publications/publicationsPdfs/parity_report.pdf.
19.	Information on state mental health parity laws is based on data compiled by the National Conference of State Legislatures' Health Policy Tracking Service http://www.ncsl.org.
20.	The DSM, produced by the American Psychiatric Association, is a comprehensive system of diagnosis for psychiatric conditions. The fourth and current edition was published in 1995 and is available at http://www.psych.org/research/dor/dsm/index.cfm.
21.	P.L. 104-204, Title VII, codified at 29 U.S.C. 1185a and 42 U.S.C. 300gg-5.
22.	P.L. 105-34, Section 1531(a)(4), codified at 26 U.S.C. 9812.
23.	42 U.S.C. 1396u-2(b)(8); 42 U.S.C. 1397cc(f)(2).
24.	U.S. General Accounting Office, Mental Health Parity Act: Despite New Federal Standards, Mental Health Benefits Remain Limited, GAO/HEHS-00-95, May 10, 2000.
25.	Additional information on FEHBP's implementation of mental health parity may be found on the OPM's website at http://www.opm.gov/insure/health/consumers/parity.asp.
26.	Howard H. Goldman, Richard G. Frank, Audrey Burnam, Haiden A Huskamp, Susan Ridgely, Sharon-Lise T. Normand, Alexander S. Young, Colleen Barry, Vanessa Azzone, Alisa B. Busch, Susan T. Azrin, Garrett Moran, Carolyn Lichtenstein, Margaret Blasinsky, Behavioral Health Insurance Parity for Federal Employees, New England Journal of Medicine, 354; 13, March 30, 2006, pp. 1378-1386.
27.	Samuel H. Zuvekas, Agnes E. Rupp, Grayson S. Norquist, The Impacts of Mental Health Parity and Managed Care in One Large Employer Group: A Reexamination, Health Affairs, v. 24, no. 6, 2005, pp. 1668-1671.
28.	Colleen L. Barry, Richard G. Frank, Thomas G. McGuire, The Costs of Mental Health Parity: Still an Impediment?, Health Affairs, v. 25, no. 3, 2006, pp. 623-634.
29.	Ronald E Bachman, An Actuarial Analysis of S. 543: Mental Health Equitable Treatment Act of 2001. Prepared for the American Psychological Association (July-August 2001).
30.	CBO's estimates are at http://www.cbo.gov/showdoc.cfm?index=3013&sequence=0.
31.	Grazier K, Pollack H, "Viewing employer-based behavioral health services research in the context of public policy," paper presented at 128th American Public Health Association Annual Meeting, November 2000.
32.	Congressional Budget Office (CBO), Cost Estimate, H.R. 1424 Paul Wellstone Mental Health and Addiction Equity Act of 2007, September 2007, available at http://www.cbo.gov/ftpdocs/86xx/doc8608/hr1424.pdf. Congressional Budget Office (CBO), Cost Estimate, S. 558 Mental Health Parity Act of 2007, March 2007, available at http://www.cbo.gov/ftpdocs/78xx/doc7894/s558.pdf.
33.	Colleen L. Barry et al., "Design of Mental Health Benefits: Still Unequal After All These Years," Health Affairs v. 22, 2003, pp. 127-137. The 2002 KFF/HRET survey sampled 2,014 randomly selected public and private employers with three or more workers.
34.	In a self-insured plan, the employer assumes direct financial responsibility for the costs of the workers' medical claims and pays the physicians and hospitals directly. Self-insurance is common among large employers with many workers over which to spread the risk of costly claims.
35.	Recent articles on the history and development of parity legislation and the impact of managed behavioral health care include (i) Kevin D. Hennessy, and Howard H. Goldman, "Full Parity: Steps Toward Treatment Equity for Mental and Addictive Disorders," Health Affairs, v. 20, 2001, pp. 58-67; (ii) Daniel P. Gitterman et al., "Toward Full Mental Health Parity and Beyond," Health Affairs, v. 20, 2001, pp. 68-76; and (iii) Richard G. Frank et al., "Will Parity in Coverage Result in Better Mental Health Care?" New England Journal of Medicine, v. 345, 2001, pp. 1701-1704.
36.	Several RAND studies have analyzed the HCC data to see how parity legislation is affecting insurance coverage and access to care for people with mental illness. Details of those studies are available online at http://www.rand.org/health.
37.	Margo Rosenbach et al., Effects of the Vermont Mental Health and Substance Abuse Parity Law, DHHS Pub. no. (SMA) 03-3822. The study was conducted for SAMHSA by Mathematica Policy Research Inc. and released in September 2003. It is available online at http://www.mentalhealth.samhsa.gov/publications/allpubs/sma03-3822/default.asp.
38.	Fronstin, Sources of health insurance and characteristics of the uninsured: analysis of the March 2000 current population survey, Washington DC, Employee Benefits Research Institute, issue brief no. 228, 2000.