Testimony for the Record of the APA Regarding the Subcommittee on Employer-Employee Relations Committee on Education and the Workforce

U.S. House of Representatives Hearing on Mental Health Parity

Wednesday, March 13, 2002

Mr. Chairman and Members of the Committee:

The American Psychological Association (APA), the largest membership association of psychologists with more than 155,000 members and affiliates engaged in the study, research, and practice of psychology, appreciates the opportunity to submit this testimony for the record regarding mental health parity. We are longtime supporters of parity, having strongly advocated for passage of the original Mental Health Parity Act of 1996 (1996 Parity Act). We are grateful to the Chairman and the members of the Committee for holding a hearing to discuss the need to end widespread discrimination in health insurance coverage of mental disorders.

APA strongly urges passage of full mental health parity legislation -- the logical extension of the 1996 Parity Act -- to end this discrimination. Only full mental health parity legislation will provide coverage for mental health benefits equal with that provided for medical and surgical benefits. Only full mental health parity legislation offers the solution originally promised by the partial parity of the 1996 Parity Act.

We emphasize four key points in our testimony today:

First, despite clear evidence of the significant mental health needs of our citizens and despite passage of the 1996 Parity Act, inadequate access to mental health treatment continues. Only enactment of full parity legislation will ensure that employees have nondiscriminatory mental health coverage. This means that full parity must be required for all aspects of coverage, including day and annual visit limits or other limits on duration or scope of treatment, deductibles, co-payments, maximum out-of-pocket limits, or other cost sharing requirements.

Second, all persons with mental illness diagnoses need and deserve the protection of parity. Only broad-based parity legislation will protect people with all forms of mental illness, not just those on a random list of diagnoses such as a list of "severe mental illnesses" (SMI).

Third, the cost of full parity for all persons afflicted with mental illness is very affordable. The Congressional Budget Office (CBO) has estimated that full mental health parity, as provided in the "Mental Health Equitable Treatment Act of 2001" (S. 543) (introduced by Senators Pete V. Domenici and Paul Wellstone), will cause premiums to rise on average by a mere 0.9%. This extremely low cost impact is born out by the actual experience of states that have enacted full parity.

Fourth, there is no reason to believe that the very modest additional costs on the private sector of offering full parity will lead to a significant number of workers losing their mental health benefits or losing their health insurance coverage. Again, actual experience with parity in 34 states demonstrates that parity will not lead to widespread loss of insurance coverage.

I. Only Enactment of Full Parity Legislation Will Ensure that Employees Have Nondiscriminatory Mental Health Coverage Equal to Their Coverage for Medical and Surgical Benefits


Three years ago, the U.S. Surgeon General issued a comprehensive report documenting the mental health needs of our nation. The Surgeon General found that about one in five Americans are affected by mental disorders during any given year. (Office of the Surgeon General, the U.S. Department of Health and Human Services, Mental Health: A Report of the Surgeon General, 1999.) In a subsequent report focusing specifically on children's mental health, the Surgeon General found that one in ten children and adolescents suffer from mental illness severe enough to cause some level of impairment, while the unmet need for services remains as high now as it was 20 years ago. (Office of the Surgeon General, the U.S. Department of Health and Human Services, Report of the Surgeon General's Conference on Children's Mental Health: A National Action Agenda, 2001.) Health insurers and employers, however, traditionally have offered less coverage for mental health services than for medical/surgical services and have imposed more restrictions.

The 1996 Parity Act was an important step forward in recognizing the need for mental health parity coverage in our nation; it provided limited parity in annual and lifetime dollar limits. Nevertheless, despite clear evidence of the need for mental health treatment for many Americans and despite enactment of the 1996 Parity Act, health insurers and employers have continued to discriminate against people with mental illness.

Two years ago, the General Accounting Office (GAO) issued a study of the 1996 Parity Act's effects. GAO found that most employers have been offsetting the narrow lifetime and annual dollar limit parity requirement by instead restricting other mental health benefits design features, such as outpatient visit limits, inpatient days, or co-payment requirements. As GAO reported:

Although most employers' plans now have parity in dollar limits for mental health coverage, 87 percent of those that comply contain at least one other plan design feature that is more restrictive for mental health benefits than for medical and surgical benefits . . . . [M]any employers may have adopted newly restrictive mental health benefit design features since 1996 specifically to offset the more generous dollar limits they adopted as a result of the federal law. About two-thirds of these newly compliant employers changed at least one other mental health benefit design feature to a more restrictive one compared with only about one-fourth of the employers that did not change their dollar limits. (GAO, Mental Health Parity Act: Despite New Federal Standards, Mental Health Benefits Remain Limited, Pub. No. HEHS-00-95, 2000)

During debate on the passage of the 1996 Parity Act, APA and other mental health parity advocates hoped that employers would implement its lifetime and annual dollar parity requirement forthrightly due to the projected very low impact on claims costs. CBO had predicted a rise in employers' share of premiums of a mere 0.16%. Despite the very low cost associated with the 1996 lifetime and annual dollar limit parity requirement, many employers reacted by simply restricting other components of mental health coverage. By reducing any part of an employee's benefit to meet a parity requirement, Congress' intent of ending unfounded discrimination against persons needing mental health services through their private coverage is thwarted. APA, therefore, strongly supports legislation that offers full parity in all mental health benefit design features.

S. 543, for example, provides for full parity for in-network services with respect to outpatient visits and inpatient days, co-payments, deductibles, and maximum out-of-pocket requirements, in addition to parity for annual and lifetime dollar limits as currently required by the 1996 Parity Act. An employee's mental health coverage, when offered and when in-network, would have to be equal in every respect to the employee's medical/surgical coverage. Full parity would mean that employers could not simply reduce certain design features so that employee access to mental health benefits does not improve in the aggregate.

II. All Persons with Mental Diagnoses Listed in the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) Need and Deserve the Protection of Parity, Not Just Those on a Random List of Diagnoses Commonly Known as "Severe Mental Illnesses" (SMI)

The issue of whether mental health parity legislation should be narrowed to cover only those conditions considered to have a biological basis was raised at the hearing. APA believes that it is simply wrong and bad public policy to condition health insurance coverage on the cause of the illness, and we strongly urge the Subcommittee to reject this strategy. When a person has cancer or heart disease, insurance coverage does not turn on the cause; our society just provides the needed care. We must do the same for mental illness.

Many people suffer debilitating mental health disorders, some of which are thought to have a biological basis (most commonly considered to be only Schizophrenia, Schizoaffective Disorder, Bipolar Disorder, Major Depression, Panic Disorder, and Obsessive-Compulsive Disorder) and some of which are not. Since science cannot say with certainty which mental illnesses are completely biologically-based, an SMI list is necessarily a random list. To adopt such a causality standard is to enshrine - in Federal law - a policy that officially condones inferior insurance coverage for many patients, including:

  • Women suffering from mental and emotional problems stemming from rape or assault;

  • Children suffering from pervasive developmental disorders, such as Autism; Pica (a serious eating disorder); Tic disorders; Attention Deficit/Hyperactivity Disorder; and many more disorders;

  • Adolescents and adults suffering from eating disorders, such as Anorexia Nervosa and Bulimia; Schizophreniform Disorder (a mild form of Schizophrenia that most often sadly progresses to Schizophrenia or Schizoaffective disorder); Trichotillomania (recurrent pulling out of one's hair, often beginning in childhood); Multiple Personality Disorder; Dementia due to Parkinson's Disease; and many more debilitating, even devastating, disorders.

Moreover, there is no scientific justification for covering only SMI conditions, but not other mental illnesses. Success rates for treatment of some non-"biologically based" mental disorders have been shown to surpass those for some "biologically based" mental disorders (see attached chart "Treatment Efficacy (Early Treatment Outcome) for Eight Mental Disorders and Two Cardiovascular Surgical Procedures"). Treatment success rates for Post-Traumatic Stress Disorder (65%), Dysthymic Disorder (a chronic low-level depression) (65%), and eating disorders such as Anorexia Nervosa and Bulimia (78%) all equal or exceed those for Schizophrenia (60%), Major Depression (65%), and Obsessive-Compulsive Disorder (60%). Indeed, treatment success rates for these mental disorders, as well as Panic Disorder (80%) and Bipolar Disorder (80%), all greatly exceed treatment success rates for two common surgical procedures for heart disease (45-50%).

Recognizing there is no need and no good reason to offer the protection of parity only for biologically-based mental disorders, fifteen states have enacted broad-based parity laws for all mental illness diagnoses (i.e., Alabama, Arkansas, Connecticut, Georgia, Indiana, Kentucky, Maryland, Minnesota, Mississippi, New Mexico, North Carolina, Rhode Island, Tennessee, Utah, and Vermont). Two of these states, Connecticut and Rhode Island, just recently expanded their original SMI parity laws to broad-based parity covering all mental illness diagnoses.

Broad-based parity means equal coverage for all mental diagnoses listed in the DSM-IV and/or the International Classification of Diseases, Ninth Edition, Clinical Modification (ICD-9-CM), a publication of the World Health organization. The DSM-IV requires that all disorders be required to have clinically significant impairment (in one or more important areas of functioning) or distress (painful symptoms). The DSM-IV is the insurance industry standard for determining mental diagnoses for purposes such as claims reimbursement. DSM codes are used by the Centers for Medicare and Medicaid Services. DSM codes are in agreement with the ICD-9-CM. The Federal Employee Health Benefits Plan (FEHBP) provides parity coverage for all DSM-IV diagnoses. Eight broad-based state parity laws expressly cover all diagnoses listed in the DSM-IV and/or the ICD-9-CM (i.e., Alabama, Arkansas, Connecticut, Georgia, Kentucky, Tennessee, Utah, and Vermont). Many other broad-based state parity laws do not expressly reference the DSM-IV, but the DSM-IV still applies as the industry standard used to judge which conditions should be given the protection of parity under the statute (i.e., Indiana, Maryland, Minnesota, Mississippi, New Mexico, North Carolina, and Rhode Island).

Although the DSM-IV contains the full range of possible diagnoses, the enactment of broad-based parity legislation will not require coverage for every possible DSM-IV diagnosis. Insurers today do not cover all diagnoses, even in states that offer broad-based parity for all DSM-IV diagnoses. Rather, insurers currently employ a variety of managed care techniques, including determinations of medical necessity, to evaluate whether to cover a particular diagnosis. Under S. 543, managed care techniques and medical necessity determinations still will govern eligibility for coverage. In fact, a Senate Committee amendment to S. 543 specifically allows insurers to use their own standards for determining medical necessity.

Moreover, Congress can provide broad-based parity for all diagnoses for about the same cost as providing parity only for SMI. In 1996, CBO estimated that parity for a typical SMI diagnosis-based list costs 90% as much as parity for all diagnoses. The Substance Abuse and Mental Health Services Administration (SAMHSA) subsequently confirmed CBO's cost estimate. In considering leading parity cost studies, SAMHSA found that expenses for this type of diagnosis-based list represent 89% of the increase for all mental health diagnoses due to parity. (Merrile Sing, et al., SAMHSA, U.S. Department of Health and Human Services, The Costs and Effects of Parity for Mental Health and Substance Abuse Insurance Benefits, 1998). Thus, parity can be provided on an equitable basis for all persons with mental diagnoses with little additional cost impact.

III. The Cost of Full Parity for all Persons with Mental Illness Is Very Affordable

CBO estimates that the cost of full mental health parity, as provided in S. 543, will cause premiums to rise by a mere 0.9%. (CBO, Cost Estimate for S. 543, Mental Health Equitable Treatment Act of 2001, 2001.) This cost increase is shared between the employer and employee, with the employer typically paying 0.36% of the total. Actuarial analysis by PricewaterhouseCoopers agrees, and shows that S. 543 would cost the typical plan only four and one-half cents per covered person per day.

In addition to the CBO score for S. 543, we now know from the actual experience of states with full parity laws, that the cost is about 1%, or even well below 1% in some states. For example:

  • Vermont enacted a full mental health parity law that became effective on January 1, 1998. Vermont's full parity law is similar to that offered FEHBP. As insurers filed their proposed rates for 1998 with the State Division of Insurance, insurers identified the proportion of their rate increases attributable to parity. Blue Cross Blue Shield of Vermont estimated the impact at 0% for the Vermont Health Partnership (a managed care plan product), 1-3% for its comprehensive plan, and 2% for its base plan. Kaiser Permanente/CHP estimated a 2.07% increase attributable to parity. (Elizabeth R. Costle, Report of the Department of Banking, Insurance, Securities and Health Care Administration on Mental Health and Substance Abuse Parity (Act 25) to the Vermont General Assembly, 1999.)

  • Maryland implemented full parity in 1995. Shortly after, a small increase was observed in the number of inpatient admissions, but that increase was more than offset by a more significant decrease in the average length of inpatient stays. For one insurer, the proportion of the total medical premiums attributable to its mental health benefit actually decreased by 0.2%. A second managed care company with extensive experience in the state subsequently confirmed that its average expense per member per month increased by less than 1% during the first seven months after implementation of full parity. Additional data received in 1997 indicated that, after an initial increase following parity implementation, costs receded towards pre-parity base line levels. (Harold Varmus, M.D., National Institute of Mental Health, U.S. Department of Health and Human Services, Parity in Coverage of Mental Health Services in an Era of Managed Care: An Interim Report to Congress by the National Advisory Mental Health Council, 1997.)

  • Minnesota's full parity law became effective on August 1, 1995. To date, there have been no recognized cost concerns or exodus of insured plans to ERISA status in order to avoid the state's parity mandate. The Minnesota Department of Commerce, which regulates indemnity insurance, estimated costs of 1% of total premium dollars for mental health parity. Medica, an independent consulting organization, estimated state costs for mental health parity at 26 cents per member per month. (Varmus, Interim Report to Congress.)

Actual state experience indicates that full parity coverage has had a minimal impact on claims costs, and in fact closely mirrors the CBO projection. SAMHSA confirmed this finding, documenting that due to the ability of managed behavioral health care organizations to control costs state parity laws have had only a small effect on insurance premiums. (Sing, The Costs and Effects of Parity). The benefit gained by employees through full parity, however, far outweighs its very low cost. With enactment of full parity, employees are assured that their insurance coverage provides nondiscriminatory mental health benefits should they need these benefits.

IV. The Enactment of Full Parity Will Not Lead to a Mass of Employers Dropping Their Mental Health Benefits or Health Insurance Coverage

There is no reliable data or experience showing that full mental health parity legislation will cause large numbers of employers to drop their health insurance coverage overall, or even their mental health benefit coverage. In the 34 states that have enacted parity laws, parity for mental health coverage has not led to widespread loss of insurance coverage anywhere, and no state that has enacted a parity law has ever later narrowed or repealed it. Indeed, as discussed earlier, states' experiences with parity have led some to expand upon their original parity laws, most recently Connecticut and Rhode Island. Further, in studying possible employer reactions to state parity laws, SAMHSA also found that employers neither suddenly decided to self-insure to circumvent state parity laws, nor did employers pass on the full cost of parity to employees. (Sing, The Costs and Effects of Parity.).

Moreover, amendments in the Senate markup of S. 543 already have addressed many concerns raised by employers. Under S. 543, parity for substance abuse is not required and small employers of 50 or fewer employees are exempted from the bill. S. 543, like the FEHB program on which it is modeled, proposes parity only for in-network services, leaving employers and health plans free to use preadmission and other managed care review techniques to deliver cost-effective mental health care.

There is no reason to believe that the very modest additional costs on the private section associated with S. 543 will lead to any significant number of workers losing their health insurance coverage. The Office of Personnel Management (OPM), in weighing the claim that parity would result in fewer people having insurance coverage, concluded that this argument appears to be a myth. (OPM, Mental Health & Substance Abuse Parity Questions and Answers, 2000).

There is also a significant cost to employers of not providing parity for mental health benefits. The 1999 Surgeon General's report on mental health estimated that workers with untreated or undertreated mental illness add some $70 billion annually to employer costs through absenteeism, turnover and retraining expenses, lower productivity, and increased medical costs. (Office of the Surgeon General, Mental Health: A Report of the Surgeon General.)


The time has come for Congress to build upon the 1996 Parity Act and end discriminatory mental health coverage by passing full mental health parity legislation, such as S. 543. CBO has stated, and actual state experience has shown, that completely banning discrimination will result in an average premium increase of less than 1%. The benefit from this small impact on premiums will be enormous. Employees will be assured that their mental health benefits are available, like their medical/surgical benefits, when they need them, and families will no longer be devastated by the financial implications of the illness of a loved one.