Table 1

Characteristics of Mental Health and Substance Abuse Parity Legislation by State

Column Descriptions.

Element of Parity Descriptions

Existing Parity Models. The following models, developed by Mr. Ron Bachman, each contain certain elements of parity that are described above.

Mental Health Parity Act Model. The federal Mental Health Parity Act (MHPA) prevents differing lifetime or annual dollar limitations between mental health and non-mental health insurance coverage. The MHPA does not require or mandate that mental health coverage be provided under any plan. If mental health coverage is provided, however, the plan cannot impose annual or lifetime dollar limitations on mental health coverage that are not imposed on coverage for non-mental health conditions. This model, when enacted at the state level, applies these limitations to whatever population that the individual state selects. The estimated cost, at the national level, for implementation of this model is 0.13 percent of total plan costs.

Limited Parity Model. This model extends the MHPA model to outpatient visits and inpatient days limits. Following the approach of MHPA, there is no mandated coverage. That is, while no mental health coverage is required, if mental health coverage is provided, the plan cannot impose day or visit limits on mental health coverage that are not imposed on coverage for non-mental health conditions. The estimated cost, at the national level, for implementation of this model, is 0.7 percent assuming that all other elements of the mental health coverage remain the same.

Catastrophic Parity Model. This model includes the features of Limited Parity listed above and adds maximum out-of-pocket (OOP) expenses limit. Maximum OOP limits are common features for non-mental health coverages provided by many insurance plans (particularly indemnity plans). This feature protects an individual or family from the financial ravages of a catastrophic cost where even the traditional 20 percent co-insurance paid by the patient can accumulate to an amount that threatens the family’s financial resources. The maximum OOP limit is the total amount of eligible expense cost sharing for which a covered member is responsible. The maximum OOP can be separate and distinct from any non-mental health maximum OOP. Again, as with the MHPA approach, there is no mandated coverage. If mental health coverage is provided, however, the plan cannot impose a maximum OOP limit on mental health greater than any imposed on other coverages. The estimated cost, at the national level, for implementation of this model is 1.1 percent assuming that all other elements of the mental health coverage remain the same.

Significant Parity Model. This model includes all of the elements listed in the MHPA, Limited, and Catastrophic models and extends parity to the co-insurance and co-payment features of an insurance plan design. With the introduction of this model, Mr. Bachman maintains that "anti-selection" begins to play a role in costs. It is at this point, he says, that small employers (if they are included) will begin to quit providing mental health coverage if it is not mandated. The estimated costs of implementation at the national level if coverage is mandated are 1.6 percent, assuming that all other elements of the coverage remain the same.

Financial Parity Model. This model represents the point at which all reimbursement features for existing plan-eligible expenses are made on the same basis as non-mental health eligible expenses. In addition to the features of the previously listed models, this brings parity to the issue of deductibles. The estimated costs of implementation at the national level, again assuming that coverage is mandated, is 2.4 percent, if other elements of the coverage remain the same.

In the previous models, Mr. Bachman also notes that the increases will likely be mitigated significantly by available cost controls within each model and the increased use of managed care practices. This is a realistic expectation according to the U. S. Department of Health and Human Services.

In preparing actuarial reports for the various other states that have enacted legislation, Mr. Bachman has provided cost estimates for three basic models which combine many of the elements of the previously listed models:

Partial Parity. This model is the same as the Limited Parity Model listed above.

Full Parity. This model provides for parity in terms of cost sharing such as deductibles, maximum OOP expenses, and annual or lifetime maximums. In this model, such cost sharing is "separate but equal."

Comprehensive Parity. This model is the same as Full Parity described above, however, the cost sharing features are common. For example, if the plan had a $250 deductible, initial expenses in physical health and mental health would count toward the same deductible.

Inclusion/Exclusion of Substance Abuse. Substance abuse is specifically excluded from the definition of mental illness. Likewise, it is not included in the Mental Health Parity Act of 1996. Some states have elected to include substance abuse in their parity legislation while others have excluded it. Basically, any mental health parity model can accommodate the inclusion of substance abuse. The cost of inclusion of substance abuse, at a national level, has been estimated at between 0.5 and 1.0 percent.

Impact of Delivery Systems on Costs of Parity. When describing models of mental health parity and the associated implementation costs, it is necessary to include a discussion of health care delivery systems since they have a significant impact on costs. Cost estimates are provided in terms of increased cost of claims; not increased cost of premiums and were obtained from PricewaterhouseCoopers’ principal Ron Bachman in his publication Mental Health: Parity Issues and Costs, and from the SAMHSA publication The Costs and Effects of Parity for Mental Health and Substance Abuse Insurance Benefits. The mental health treatment delivery system in Alaska is dominated by fee-for-service and managed indemnity delivery systems.

Fee-for-Service (FFS). FFS delivery systems are the most expensive because there is relatively little systematic emphasis on cost control. Providers of services are paid on a per-service basis for any services they deliver. Enrollees can typically use any provider they choose with the provider being paid for the services either based on charges or on some determination of customary fees. Mr. Bachman estimates costs for implementation of mental health parity are the highest for this delivery model with those costs ranging from 0.9 percent to 1.3 percent for partial parity (see discussion above), and from 2.7 percent to 4.9 percent for comprehensive parity. SAMHSA, using several different models, suggests that FFS systems faced with implementation of mental health parity will evolve and witness a much greater penetration of managed care, which will serve to dampen costs. They cite several experiences where, following implementation of parity and subsequent penetration of managed care, mental health costs actually declined.

Managed Indemnity. The FFS system can be modified somewhat by the introduction of managed care principles. This typically involves pre-notification and case reviews of lengthy courses of treatment such as inpatient stays. Because care delivered in the most expensive and/or protracted situations (inpatient or extended outpatient) are subject to greater control by the employer and/or insurer, the costs associated with the system are slightly lower than the FFS system. Mr. Bachman estimates that implementing parity in the managed indemnity delivery system would result in cost increases of between 0.7 percent and 0.9 percent for partial parity and between 2.2 percent and 3.6 percent for comprehensive parity.

Preferred Provider Organizations (PPO) and Point-of-Service (POS) Systems. PPOs and POSs represent a further intensification of managed care that we are told will evolve. These systems involve a much higher degree of cost control with employers and insurers establishing gatekeepers as well as networks of providers who are paid at negotiated prices. Estimates of costs for implementing parity under these types of delivery systems range from 0.3 percent to 0.6 percent for partial parity and from 2.3 percent to 3.8 percent for comprehensive parity.

Health Maintenance Organizations (HMO) . The HMO delivery system, combined with intensive gatekeeping, is the most aggressive delivery system model for which costs estimates were provided. In these systems, the organization is a combination of the insurer and the provider so that there is maximum incentive to decrease costs. Estimates for implementing parity in these systems range from 0.4 percent to 0.5 percent for partial parity and 1.2 percent to 1.8 percent for comprehensive parity.

These four models are presented here as discrete systems. They exist, however, in various combinations, the ratio of which varies from state to state. With the implementation of mental health parity, there would most likely be some movement toward the more rigorously managed systems and away from the FFS system.

General. Mr. Ron Bachman of PricewaterhouseCoopers provided the following discussion and cost estimates in a letter to the Task Force dated January 8, 1999. The cost estimates are based on that firm’s proprietary algorithms. The calculations outline the impact of potential mental health and substance abuse legislation in Alaska. Actual legislative language, if developed, may alter the results up or down. Cost estimates shown are (1) for mental health only and (2) for mental health and substance abuse combined. The cost projections are higher than national averages due to three key factors specific to Alaska. One, the current coverage levels of mental health and substance abuse are low. This is true in spite of the substance abuse mandate that requires more up-front reimbursements through equalized deductible and co-insurance rates. The $9,600 over two consecutive years; $19,200 lifetime limit is less than generally available elsewhere. Second, the current level of managed care in Alaska is very low. Most of the weighted pricing is on managed indemnity. Third, mental health and substance abuse utilization rates are high in Alaska. Expanded benefits will cost more than similar expansions in other states.

Delivery Systems. The four delivery systems assumed are:

Delivery System 1. Indemnity plans with utilization review found on typical medical plans. There is no special mental health or substance abuse focus and the review (pre-admission certification and continued stay review) generally applies only to inpatient care.

Delivery System 2. Indemnity plans with specialized mental health and substance abuse utilization review. The utilization review applies to inpatient care, but may also apply to intensive or lengthy outpatient treatments.

Delivery System 3. Preferred Provider (PPO) and Point of Service (POS) plans that have specialized mental health and substance abuse networks. These are not carve-out programs, but act with similar attention to negotiated rates, utilization controls, and limited provider access. There is no gatekeeper mechanism. Plan design and cost sharing are primarily used to channel members to network providers.

Delivery System 4. HMO/Gatekeeper plans and carve-out mental health and substance abuse programs. Access to mental health and substance abuse providers is through a primary care gatekeeper or other similar intensive utilization controls. Provider reimbursements are highly negotiated. HMO POS plans are also included.

Parity Options. The following is a summary of the three parity options priced within the four delivery systems:

Mental Health Parity Act (MHPA). This is limited parity for mental health and substance abuse for annual/lifetime dollar limits. It is consistent with the federal legislation except that it is applicable to groups with 20 or more employees, includes substance abuse coverage, and mental health and substance abuse coverage is mandated.

Catastrophic Parity. This requires parity for any lifetime/annual/episode limits, day and visit limits, and maximum out-of-pocket provisions. It does not require parity for deductibles, co-payments, or co-insurance (except after any maximum out-of-pocket limit).

Comprehensive Parity. This requires financial parity with mental health and substance abuse benefits reimbursed under health insurance plans on the same basis as medical/surgical benefits.

Cost Estimates. The following are cost estimates for mental health and combined mental health and substance abuse parity. The cost increases are based on the cost of claims. Composite market analyses represent the aggregate of the costs over the various distribution systems before any impact of increased managed care. The net composite information takes into account efforts by employers and carriers to contain costs through a variety of mechanisms including increased cost share, lower overall benefits, or managed care.

*PMPM – "Per member per month"

Table 3, which follows, provides these analyses for a plan that includes both mental health and substance abuse. Certain assumptions were made as to the impact on parity relative to the existing substance abuse mandate. We assumed under the MHPA that a 20-day and 20-visit limit would replace the current dollar limit. We further assumed that a plan’s co-insurance coverage would be decreased to 50%, generally. In spite of these tradeoff limitations, we estimated the cost of the substance abuse coverage to increase slightly.

Under the catastrophic option, the equalized deductible and co-insurance is replaced with unlimited days and visits and a maximum out-of-pocket cost. This tradeoff is substantial and results in increased costs since many of the inpatient stays will exceed the maximum out-of-pocket costs with the excess covered at 100%. The deductible and co-insurance value of the existing substance abuse legislation, however, is particularly important to those using outpatient and very short-term inpatient (detox) care.

*PMPM – "Per member per month"

The PricewaterhouseCoopers modeling assumes a reasonable, but conservatively low managed care penetration for Alaska. The assumptions were established conservatively to account for the impact of fewer small groups currently using managed care. We did not find data available to determine the split of managed care for behavioral health plans in Alaska solely for insured plans.

The "Net Market Impact" reflects how employers respond to any potential increase in benefit costs in a variety of ways including

• competitively marketing the plan to obtain lower premiums;
• intensively negotiating lower provider costs;
• cutting plan administrative costs;
• increasing plan cost sharing by members;
• increasing premium contributions by members;
• reducing other benefits; and, in the extreme;
• dropping plan coverages and reducing wages (or reducing wage increases).

General. Implementing any level of parity for mental health insurance, with or without substance abuse, will impact organizations and individuals differently. The following discussion is intended to identify those groups that will be impacted.

ERISA. Under the terms of ERISA, non-government employers who offer self-insured health plans will be exempt from any mental health parity legislation. In Alaska, this accounts for a significant portion of the lives covered under health insurance plans. All large private companies who are self-insured would be exempt.

Governmental Entities. Health insurance for state employees as well as federal and local government employees would be exempt.

Policies Written Outside Alaska. Some companies operating in Alaska have their headquarters and offices in other states. Employees’ health plans for these types of operations are sometimes written outside Alaska and, therefore, they would also be exempt.

Small Businesses. A major concern in implementing any changes in health insurance mandates in Alaska is the impact on small businesses. Alaska Statute 45.21.56 defines a small business for certain purposes as those having from 2 to 100 employees. For purposes of this report and the included recommendations, this definition does not apply. The following analyses were conducted assuming an exemption for businesses with less than 20 employees.

A Note on Methodology. There are a number of factors and assumptions that introduce the possibility of error. These factors include uncertainty about the percentage of small businesses providing coverage for their employees, uncertainty regarding the extent to which dependents of small business employees are covered, and timing differences for the data. Some data represents conditions existing at the time of the report while other information is based on 1997 data. Another source of uncertainty is that the number of lives covered is based on a small sample of the insurance companies operating in the state. The different sources of error impact the results in different directions so it is difficult to speculate on whether the estimates are high or low. The following analyses represent our best estimate of the number of impacted employees and lives covered. We recognize that these numbers might vary as much as ten percent in either direction. In many of the following computations, we will employ rounding to the nearest one thousand in recognition that many of the figures are estimates.

Number/Size of Employers. According to the Alaska Department of Labor, there were 11,997 small businesses with less than 20 employees in 1997. This represents 86% of the businesses in Alaska. There were 1,916 firms that employed between 20 and 50 employees. The numbers begin to fall off dramatically for firms with more than 50 employees.

1 – 19 11,997

20 – 49 1,224

50 – 99 395

101 – 249 241

251 – 499 88

Over 500 75

Data Source: Alaska Department of Labor

In viewing the data on the number of employers in each size category, one of the major concerns is how to balance the need to exempt small businesses from onerous mandates against the need to insure that a sufficient risk pool exists to make changes feasible. With most large employers exempt because of ERISA, having policies written outside the state, being governmental entities, and small businesses exempted due to size, the risk pool becomes very small.

To test the impact of changes on concerned groups, the Task Force set a tentative cutoff for small businesses at those with 20 or more employees. Consequently, employers with 19 or less employees would be exempt. Based on this, we determined the number of employees who would be excluded from any count. The following is an analysis based on Department of Labor data.

Estimated 1997 Alaska Private Sector Small Business* Employees:

Size of Firm Total Number of Employees

1 – 4 employees 22,269

5 – 9 employees 26,025

10 – 19 employees 31,659

Total Small Business Employees 79,953

* Small Businesses are defined here as those businesses with less than 20 employees.

Data Source: Alaska Department of Labor

While we can estimate the number of employees working for firms with less than 20 employees, we must still know what percentage of these are covered by health care insurance. Based on reports from five states, the incidence of health insurance for specific size companies are estimated as follows:

1 – 4 employees 40% 37% 40% 40% 30-35%

5 – 9 employees 61% 61% 60% 61% 50-55%

10 – 24 employees 76% 76% 75% 76% Unknown

Data Source: Coopers & Lybrand Actuarial Analyses

These figures represent the percentages of employers within each size category that offer health care insurance. For purposes of this analysis, we assume that the proportion of employers offering insurance is approximately the same as the proportion of small business employees covered by health insurance. For our analyses we used the following proportions: 1 through 4 employees – 40%; 5 through 9 employees – 60%; 10 through 19 employees – 75%. In using these percentages, we also assume that the percentages of covered small business employees in Alaska are comparable to the percentage covered in other states.

Applying these ratios to the numbers of employees in each size category we find:

1 – 4 employees 22,269 40% 8,908

5 – 9 employees 26,025 60% 15,615

10 – 19 employees 31,659 75% 23,744

Total Small Business Employees Covered by Health Insurance: 48,267 or approximately 48,000

This figure will be used to adjust the total number of lives covered to account for the fact that firms with less than 20 employees will be exempted.

To determine the number of non-ERISA lives covered in Alaska, we contacted individual insurance companies and requested estimates from them. The four firms listed below represent four of the top five companies in terms of market share. Market share, as used in this report, refers to the share of the dollar amount of premiums written. By using the percentage market share in conjunction with the estimates of lives covered, we can arrive at an approximation of the total number of lives covered in the state.

Principal Mutual Life Insurance Co. 7.73% 7,378

Employers Health Insurance Company 3.53% 3,500

Guardian Life Insurance Company 3.52% 5,785

Great-West Life & Annuity Insurance Co. 2.12% 1,800

Total of Sample 16.9% 18,463

The above figures were obtained from two different sources. The market share information was obtained from the State of Alaska Division of Insurance while the number of lives covered was obtained from individual carriers based on their records or estimates. There are a number of mechanisms that present sources of error. First, there is not a perfect correlation between market share (based on the amount of premiums) and the number of lives covered. Second, there are obvious timing differences. The market share information covers 1997 while the lives covered represents current data. For this reason, extrapolating a total number of covered lives based on the individual companies’ data will each yield different results. We used four of the top five carriers in the aggregate in an attempt to minimize the error.

Total Lives = (Sample Lives Covered / Market Share of Sample) + Blue Cross Lives Covered

Total Lives Covered = (18,463/16.9%) + 60,000

Total Lives Covered = 109,248 + 60,000

According to Blue Cross officials, the company believes that they have 38% of the non-ERISA lives covered. If we use this figure with the 60,000 lives covered by Blue Cross, we compute total lives covered as:

Total Lives Covered = Blue Cross Lives / Blue Cross Market Share

Total Lives Covered = 60,000 / 38%

Total Lives Covered = 157,894 or 158,000

Averaging the two, we end up with an estimate of total non-ERISA lives covered in Alaska of approximately 163,000.

Data Source: Key Informant Interviews 11/98

We next needed to reduce this figure to account for businesses with less than 20 employees. As stated before, there is uncertainty with regard to the actual percentage of small business employees who are insured as well as the extent to which dependents are covered under small business health insurance policies. Using the data derived in the analysis of small businesses, we note that approximately 48,000 small business employees are likely to be covered by health insurance. Adjusting for this in the total number of lives covered we find:

Total Lives Covered: 163,000

Less Small Business Employees: - 48,000

Total Impacted Lives Covered 115,000

Impact on State of Alaska Employees Health Insurance. The health plan offered to the State of Alaska employees would not be legally subject to parity mandates recommended by this Task Force. In discussions with state officials, however, we learned that it is the policy of the State to implement any mandates placed on private insurance plans. In reviewing the various state employees’ health plans, we noted that all elements of parity are already in place for current employees as a function of the characteristics of the coverage. For example, the co-insurance (80%-20%) is the same for mental health and physical health. The deductibles for physical and mental health are a combined deductible. The State employees’ health plans have already removed the annual and lifetime maximums. The State employees’ plan currently places no limits on the number of visits and hospital days for mental health. The Alaska Division of Retirement and Benefits confirmed our observations.

In light of the foregoing, it appears that, even if the State were to adopt any mental health parity mandate approved by the legislature, there would be no impact on costs for the State employees’ health plans. The same is not true, however, for State employee retirees’ plans, which do not have the same features as current employees’ plans. The most immediate impact would likely be in the area of co-payments and co-insurance since retirees’ plans require different co-payments for mental health and physical health. There is no clear indication, however, of the extent to which the retirees’ plan would be modified to comply with mental health parity mandates since there is no legal requirement to do so.

General. The following is a summary of the concerns of major stakeholder groups based on key informant interviews, written comment, and testimony given before the Task Force.

Consumers. Key informant interviews and testimony provided to the Task Force repeatedly advanced the concern that individuals and families are frequently brought to the brink of financial ruin because of inadequate mental health and substance abuse benefits. They cited benefit levels that were widely disparate with the medical and surgical benefits in their plans. Once they exhaust their coverage, they revert to public treatment systems. This often means changing providers and ending treatment relationships that have, to that point, been productive and helpful. Among those consumers who closely monitor their benefit utilization, some indicated that they have sometimes chosen not to receive treatment when needed in order to conserve benefits. This has unnecessarily led to emergencies and more intensive treatment than would have otherwise been needed. Mental illness and substance abuse disorders are treatable. With early and appropriate treatment, people who experience these illnesses can continue to function as productive members of society. Many find, however, that when their insurance coverage is exhausted, they are forced to quit their jobs in order to qualify for Medicaid or other publicly funded care.

Treatment Providers. The key concern expressed by providers of mental health and substance abuse treatment was that, faced with the exhausted insurance benefits and high costs, patients and families of patients frequently elect to forego needed treatment. Long-term, however, electing not to receive care can lead to crises and the requirement for care at a much more intensive level such as hospitalization. They also cite concerns that patients who exhaust their benefits frequently revert to the public treatment system ending productive relationships and placing additional pressure on an already overburdened system. As with consumers, treatment providers told the Task Force that mental illness and substance abuse disorders are treatable. Providing parity between mental health/substance abuse coverage and medical/surgical coverage in health insurance will help to assure that people receive early and appropriate treatment.

Insurance Industry. The insurance industry expressed a general opposition to parity mandates for several reasons. First, a parity mandate carries a cost that, from the insurance industry point of view, will ultimately be passed on to the purchaser of the plans. One of the response options for employers would be to discontinue health care benefits rather than absorb the premium increases. This leads to fewer people insured. A second concern is that employers are forced to purchase coverage that they may not otherwise need or choose. Spending money on an unwanted or unneeded option reduces their ability to purchase other needed or wanted options.

Employers. The Task Force provided a draft copy of the report to several organizations representing employers including the National Federation of Independent Businesses (NFIB) and the Alaska State Chamber of Commerce. The feedback received was very similar to that from the insurance industry. Specifically, they stated that mental health parity in Alaska would unduly target small business owners who can least afford to pay the cost increase. They further indicated that, although the costs associated with this particular issue might seem small, they become a part of overall cost increases that, in difficult economic times, puts undue burden on small business owners. They argue, as did the insurance industry, that placing these types of mandates on small business owners could result in an overall decrease in health care coverage for small business employees since many owners might choose to discontinue coverage rather than to absorb the cost of the mandates. In cases where employers choose to continue coverage, they could pass the additional costs along to employees in the form of salary reductions or reductions in other benefits; a tradeoff that employees might not want.

The Task Force will continue to work with stakeholder groups to gather and consider further input.

General. The Task Force, having studied the various issues and elements of mental health parity, initially developed a set of three different models for consideration. In developing the recommendations, the Task Force carefully considered key policy issues that needed to be addressed regardless of which, if any, of the parity models were to be recommended:

Organizational Applicability.

Type of Policy. One of the first questions addressed was that regarding the organizations that would be impacted by any parity recommendations. State legislatures do not have the authority to regulate insurance plans covered by ERISA, the federal legislation that deals with self-insured firms. Likewise exempted from state regulatory control are the plans for state, local, and federal government employees. Another group that is exempt are those employees working for firms with headquarters outside Alaska where the health policies are written outside the state.

Size of Organization. The federal mental health parity legislation exempted firms with 50 employees or less. Adopting that criteria would create a risk pool so small that any parity legislation would be impractical. In examining past legislative action, the Task Force noted those firms with less than 20 employees were exempted from similar substance abuse mandates in 1988. After gathering data from the Alaska Department of Labor that described the employer and employee population, the Task Force chose to exempt firms with less than 20 employees.

Diagnostic Criteria. Different states have approached the issue of diagnostic criteria differently with some choosing highly restrictive models while others elected to be more inclusive. There were several key considerations to be addressed in this area:

Diagnosis Creep. Much of the literature addressing parity issues cautioned against overly restrictive diagnostic criteria because of what they called diagnosis creep. This is a phenomenon whereby some treatment providers can display a tendency to over-diagnose clients in an effort to insure that benefits are available. This not only serves to negate any cost savings that might have been realized by the restrictive policy, but also saddles individuals with inappropriate diagnoses that follow them in future years.

Manageability. A few states have elected to specifically identify those diagnoses that would be covered under parity. This approach saddles the legislature with the responsibility of "re-inventing the wheel." Insurance carriers and treatment providers both currently use a pre-determined set of diagnostic criteria that are listed in the Diagnostic and Statistical Manual IV (DSM IV). Most carriers cover all disorders except those assigned "V" Codes. This system of defining criteria is well understood by both groups and using a different definition merely adds confusion to the process.

In considering the issue of diagnostic criteria, the Task Force decided to recommend the use of the existing system recognizing all disorders listed in the DSM IV with the exception of those assigned "V" codes.

Inclusion or Exclusion of Substance Abuse. Although substance abuse disorders are included in the DSM IV, the federal government and all other states have chosen to specifically indicate whether or not substance abuse was covered under parity legislation. The federal legislation specifically excludes substance abuse, as does the legislation for seven states. Five states have included substance abuse in their legislation. In those states that included substance abuse, this element accounted for a relatively small part of the projected costs. Alaska, however, has two offsetting considerations to examine. First, Alaska already has some level of parity for substance abuse insurance so inclusion in a mental health parity mandate would be less expensive than starting from nothing. Second, Alaska has a much higher utilization rate for substance abuse services than the national average. As a result, the projected cost of including substance abuse in the most comprehensive of parity models is approximately 1.1%.

Aside from projected costs, there is also the consideration that there is a high incidence of substance abuse among the mentally ill and, unless both disorders are treated, positive outcomes for either are unlikely. For these reasons, the Task Force elected to include substance abuse in the recommended parity mandate.

Mandatory versus Optional Mental Health Coverage. The final consideration before actually selecting a recommended parity model is that of mandatory versus voluntary mental health coverage. Federal legislation applies only to those policies that include mental health coverage; it does not provide any mandate for health insurance policies to include mental health coverage. Likewise, there is no current mandate in Alaska for health insurance policies to include mental health coverage. The central issue in consideration of mandatory coverage is the phenomenon of what Mr. Bachman calls "anti-selection." Anti-selection is the tendency of companies to drop mental health coverage for employees to avoid the increased costs of parity. This serves to decrease the size of the risk pool making parity even more expensive. This is particularly critical in Alaska with our already small risk pool. According to Mr. Bachman, anti-selection becomes a factor when any level of parity greater than the catastrophic model is chosen. Considering this, the Task Force decided to recommend that coverage be mandated for impacted policies.

Level of Parity: The Models Considered. The Task Force initially considered three different models or levels of parity and obtained actuarial data for all three.

Model 1: Federal Legislation Extended. The first model considered was the extension of the terms of MHPA of 1996 (federal legislation) to Alaska firms with 20 or more employees. This would involve equating annual and lifetime dollar limits between mental and physical health. As noted above, it would also mandate that mental health coverage be provided in health care plans provided by firms subject to the mandate. Although the federal legislation did not include substance abuse, the Task Force elected to include it for purposes of analysis. The main reason for rejecting this model is the same reason that states have elected to enact their own legislation despite the presence of the federal law. The essence of the federal legislation is the elimination of disparity in annual and lifetime dollar limits between mental health and medical/surgical benefits. It allows, however, disparity in the limits on annual visits or inpatient days between mental health and medical/surgical benefits. This, in effect, allows carriers to re-define the limits using visits and day limits instead of dollar limits while providing no protection for consumers.

Model 2: Catastrophic Model. The next model would add, in addition to the elements of MHTA noted above, parity with regard to days and visit limits and maximum out-of-pocket expenses. As the title indicates, this model is designed correct the inequities of the federal legislation with regard to days and visits limits and to help families avoid financial ruin that can accompany massive out-of-pocket expenses. While correcting this problem, it does, however, leave another gap. It does not address disparity of deductibles, co-payments, and co-insurance between mental health and medical/surgical benefits. This is important because these elements represent the first set of barriers to receiving treatment. Medical/surgical deductibles, co-payments, co-insurance are set at levels that encourage appropriate utilization of benefits. Setting these cost-sharing elements at the same level for mental health benefits would serve the same purpose. For this reason, the Task Force also rejected this model.

Recommended Model: Financial Parity. This model would add, in addition to the elements of the two models noted above, parity with regard to co-insurance, co-payments, and deductibles. In short, it merely requires that the benefit levels for mental health coverage be equal with those for medical and surgical benefits.

Benefits of Mental Health Parity. According to a report commissioned by the National Institute of Mental Health published in 1997, the intended benefits of parity legislation are:

1. To overcome discrimination against people with mental illness based on artificial and scientifically untenable distinctions between mental and physical disorders;
2. To make parity mandatory for every health plan so that no plan suffers the "adverse selection" of being preferred by people with severe and costly illnesses;
3. To lessen out-of-pocket expenses for people with severe mental illness and their families;
4. To reduce disability through appropriate access to effective treatment; and
5. To increase the productivity and social and economic contributions of people with treated mental illnesses – contributions that can yield a national net economic return amounting to billions of dollars yearly.

The first four intended benefits listed above are obvious and are targeted primarily to consumers. Benefit number five, however, suggests that additional overall economic benefits may result from implementation of mental health parity. There are a number of different studies, some government-sponsored and others sponsored by private organizations, which seem to support this premise. In examining this concept, we will start with the assumption that parity mandates will lead to better access to appropriate mental health care by consumers. This is basically what parity is intended to accomplish.

In a study conducted by UNUM Life Insurance Company of America published in 1998, D. Salkever noted that employers with health plans having high deductibles for mental health expenses experienced substantially higher rates of psychiatric disability claims and decreased likelihood of employees returning to work than firms with lower deductible plans. The savings realized by lower premiums for high deductible plans were more than offset by losses due to disability claims and employee turnover.

In an unpublished study at Yale University (Rosenheck et al), researchers examined the impact of limiting specialty mental health care in a large national corporation over a three-year period. They found that, when the company decreased mental health services by 44%, there were three unintended results:

1. Reduced work performance (down by 5.1%);
2. Increased absenteeism (sick leave up by 21.9%); and
3. Increased general health expenses (up by 36.6%).

The savings generated by reducing mental health care was, again, more than offset by decreased productivity and increased spending for general health care. The employer was able to reduce costs in the short run by purchasing cheaper insurance that limited mental health care, however, increases in other costs negated that savings.

In a 1995 study reported in the Journal of the American Medical Association, investigators compared treatment patterns, effectiveness, and costs of treatment for depression by primary care physicians and mental health clinicians. In treatment of depression, they found that psychiatrists produced better functional outcomes than did primary care physicians, at greater cost, but overall with greater cost-effectiveness. They concluded that providing reduced care or more non-specialized care may incur less direct costs for treatment, but given the generally worse outcomes, tended to be less cost-effective in the long run.

Two studies in 1995 and 1996 examined the economic consequences of not treating mood disorders. Rupp (1995) concluded that there is a net return of one dollar for each dollar spent on treating the most severely mentally ill. He also found that the current market conditions offer no incentives for private firms to provide adequate mental health coverage because they risk adverse selection by attracting those workers or their family members who have more serious mental problems. Zhang (1996) found that each dollar spent on treating mood disorders yields between three and nine dollars net return. The differences in the net returns noted in the two studies may be explained by the fact that Rupp used national level epidemiological, clinical, and economic data while Zhang used a small community sample from a primarily rural southern state.

Study after study has confirmed the positive economic benefits from appropriate and timely access to mental health and substance abuse treatment. The implementation of parity between mental health benefits and medical/surgical benefits is intended to increase access to such care.

Unless otherwise noted, the following definitions are taken from the U. S. Department of Health and Human Services publication The Costs and Effects of Parity for Mental Health and Substance Abuse Insurance Benefits.

Annual Limits. The term "annual limits" refers to the maximum amount of covered health care expenses paid by an insurance policy for covered care each year (definition developed by Task Force).

Baseline Benefits Package. For each type of health plan, the baseline benefit package is the benefit package that has the highest percentage of enrollees (the statistical "mode"). Also referred to as the typical benefit package.

Benefit Package. Services covered by a health insurance plan and the financial terms of such coverage. These include cost sharing, limitations on the amounts of services, and annual or lifetime spending limits.

Chemical Dependency. Physiological or physical dependence on a psychoactive substance.

Co-Insurance. This is a type of cost sharing where the insured party and insurer share payment of the approved charge for covered services in a specified ratio after payment of the deductible. Most fee-for-service plans require a 20 percent co-insurance for covered inpatient and outpatient medical/surgical services.

Co-Payment. This is the type of cost sharing where the insured party is responsible for paying a fixed dollar amount per covered service. For example, an HMO could require a $10 co-payment for every visit to a network physician.

Cost Increase. The term "cost increase" as referred to in this study, means the increase in the cost of claims to the insurance carrier (experienced or anticipated). This does not equate to an increase in the cost of premiums to employers (definition developed by Task Force).

Cost Sharing. A health insurance policy provision that requires the insured party to pay a portion of the costs of covered services. Deductibles, co-insurance, and co-payment are types of cost sharing.

Coverage Decision. This is a decision by a health plan whether to pay for or provide a medical service for particular clinical conditions.

Deductible. The type of cost sharing where the insured party pays a specified amount of approved charges for covered medical services before the insurer will assume liability for all or part of the remaining covered services.

ERISA. Employee Retirement Income Security Act of 1974 (ERISA). Health plans that are self-insured are exempt from state regulation under ERISA provisions.

FFS. Fee-for-Service. A type of health care plan where health care providers are paid for individual medical services rendered.

Financial Requirements. The term "financial requirements" refers to co-payments, deductibles, out-of-pocket contributions, fees, annual limits, and lifetime aggregate limits imposed on covered individuals.

Gatekeeper. A primary care physician in a managed care plan (such as HMO or POS plan) who oversees the care of enrollees in the plan.

HMO. Health maintenance organization. A type of managed care plan that acts as both insurer and provider of a comprehensive set of health care services to an enrolled population. Services are furnished through a network of providers (such as physicians and hospitals).

Health Plan. An organization that acts as insurer for an enrolled population. Types of health plans include fee-for-service (FFS), preferred provider organization (PPO), point-of-service (POS), and health maintenance organizations (HMO).

Lifetime Limits. The term "lifetime limits" refers to the maximum amount of covered health care expenses paid by an insurance policy for covered care over the life of the policy (definition developed by Task Force).

Managed Care. A system of health care delivery where the health plan attempts to control or coordinate the use of health services by enrolled members to contain health care expenditures and/or improve quality. Types of managed care plans include HMOs, point-of-service (POS) plans, and preferred provider organizations (PPOs).

Maximum Out-of-Pocket Expenses. The maximum amount, including deductible and co-payments/co-insurance that an insured is required to pay before the insurance policy begins to pay all costs for covered care.

Mental Illness. The term "mental illness" includes mental disorders defined in the Diagnostic and Statistical Manual IV (DSM IV) or subsequent editions published by the American Psychiatric Association, except those codes defining substance abuse disorders (291.0 to 292.9 and 303.0 to 305.9) and the "V" codes.

Mental Health Benefits. Benefits with respect to mental health services, as defined under the terms of the plan or coverage (as the case may be), but does not include benefits with respect to treatment of substance abuse or chemical dependency.

Parity. The term "parity" as used in this study refers to the various levels of equality in coverage between mental health and physical health. The range of levels can span from complete lack of equality to comprehensive parity in which all elements of mental health and physical health coverage are provided equally. (Definition developed by Task Force)

POS. Point-of-service. Point-of-service plans are managed care plans that cover both in-network and out-of-network services. To encourage use of network providers, patient out-of-pocket costs are higher when non-network providers are used. POS plans generally manage in-network services more tightly than PPOs because POS plans use gatekeepers.

PPO. Preferred provider organization. A PPO is a managed care plan that contracts with providers to furnish services to plan enrollees. PPO providers are paid according to a discounted fee schedule. Enrollees may lower out-of-pocket costs when they use network ("preferred") providers. Services they receive from non-network providers, however, are also covered. Enrollees pay higher out-of-pocket costs when they use non-network providers for covered services.

Premium. The amount an insurer charges for a health insurance policy. The premium amount is computed to pay for the expected costs of all health insurance expenses. Health insurance expenses include medical/surgical services, mental health and substance abuse services, and administrative costs and profits.

Primary Care Physician. Primary care physicians generally include physicians with the following specialties: general medicine, family practice, internal medicine, obstetrics/gynecology, and pediatrics.

Severe Mental Illness. The National Advisory Mental Health Council defines serious mental illness (SMI) to include disorders with psychotic symptoms such as schizophrenia, schizoaffective disorder, and autism, as well as severe forms of other disorders such as major depression, panic disorder, and obsessive compulsive disorder.

Self-Insured Plan. Employer-provided health insurance in which the employer, rather than an insurer, is at risk for its employees’ medical expenses.

Service Limits. Limits on the amount of services covered by a health plan. For example, a health plan can limit the number of covered outpatient visits or inpatient hospital days.

Substance Abuse. Use of alcohol, other drugs, or inhalants in a way that is illegal or deviates from medically accepted use.

Typical Benefits Package. For each type of health plan, the typical benefit package is the benefit package that has the highest percentage of enrollees (the statistical "mode"). Also referred to as the baseline benefit package.

Appendix A: Task Force Membership

Appendix B: Resources and Suggested Readings

Appendix C: Senate Concurrent Resolution 14

Appendix D: Written Comments Submitted to Task Force

1. Advisory Commission on Consumer Protection and Quality in the Health Care Industry, Consumer Bill of Rights and Responsibilities, 1997
2. American Managed Behavioral Healthcare Association, Performance Measures for Managed Behavioral Healthcare Programs, 1995
3. Bachman, Ronald E., FSA, MAAA, Mental Health: Parity Issues and Costs, 1998
4. Bush, S., "Important Milestones on the Path to Comprehensive Parity," Mental Health Issues Today, 1998
5. Caldwell, B., "Mental Health Advocates Turn to ADA, Courts to Address Shortfall in Parity Law," Employee Benefit Plan Review, 1998
6. Christianson, J.B., Wholey, D., & Peterson, M.S., "Strategies for Managing Service Delivery in HMOs: An Application to Mental Health Care," Medical Care Research and Review, Vol. 54, No. 2, 1997
7. Congressional Budget Office, CBO’s Estimate of the Impact on Employers of the Mental Health Parity Amendment in H.R. 3103, 1996
8. Frank, R.G. & McGuire, T.G., "Mandating Employer Coverage for Mental Health Care," Health Affairs, 1990
9. Frank, R.G., McGuire, T.G., Bae, J.P., & Rupp, A., "Solutions for Adverse Selection in Behavioral Health Care," Health Care Financing Review, 1997
10. Gately, J., "Enforcing the Mental Health Parity Act of 1996," National Mental Health Association: State Advocacy Update, 1998
11. Goldman, W., McCulloch, J., & Sturm, R., "Cost and Utilization of Mental Health Services Before and After Managed Care," Health Affairs, 1998
12. Hay/Huggins Company, Inc., "Hay/Huggins Benefits Report," 1996
13. Health Policy Tracking Service, Behavioral Health: Parity, 1997
14. Heiser, N., Smolkin, S., Maxfield, M., "Parity Study Background Report #1: State Parity Laws," Draft report submitted to the Substance Abuse and Mental Health Services Administration, 1998
15. Hill, S., Sing, M., Smolkin, S., "Parity Study Background Report #2: Case Studies," Draft Report submitted to the Substance Abuse and Mental Health Services Administration, 1998
16. Institute of Medicine, Managing Managed Care: Quality Improvement in Behavioral Health, Vols 1 and 2, 1996
17. Jensen, G.A., Morrisey, M.A., Gaffley, S., & Liston, D., "The New Dominance of Managed Care: Insurance Trends in the 1990s," Health Affairs, 1997
18. Mercer, W.M., Case Studies: A Guide to Implementing Parity for Mental Illness, 1997
19. National Advisory Board on Mental Health Council, "Health Care Reform for Americans with Severe Mental Illness: Report of the National Advisory Mental Health Council," American Journal of Psychiatry, 150:10, 1993
20. O’Grady, M., "CRS Report for Congress: Mental Health Parity: Issues and Options in Developing Benefits and Premiums," Congressional Research Service, 1996
21. Rodgers, J., Analysis of the Mental Health Parity Provisions in S.1028, 1996
22. Salkever, D., "Psychiatric Disability in the Workplace," Insight, 1998
23. Scott, J.E., Greenberg, D., & Pizarro, J., "A Survey of State Insurance Markets Covering Alcohol and other Drug Treatment," The Journal of Mental Health Administration, Vol 19, No. 1, 1992
24. Sing, M. & Hill, S., "Parity Study Background Report #3: Actuarial Assumptions," Draft report submitted to the Substance Abuse and Mental Health Services Administration, 1998
25. Sing, M. & Hill, S., "Parity Study Background Report #4: Cost Estimates," Draft report submitted to the Substance Abuse and Mental Health Services Administration, 1998
26. Sturm, R., McCulloch, J., & Goldman, W., Sturm, R., McCulloch, J., & Goldman, W., Mental Health and Substance Abuse Parity: A Case Study of Ohio’s State Employee Program, Working, 1998
27. Sturm, R. & Wells, K.B., "How can Care for Depression become more Cost-Effective?" Journal of the American Medical Association, 1995
28. U. S. Department of Health and Human Services, National Institute of Mental Health, 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
29. U. S. Department of Health and Human Services, National Institute of Mental Health, Parity in Financing Mental Health Services: Managed Care Effects on Cost, Access, and Quality, 1998
30. U. S. Department of Health and Human Services, National Institute of Mental Health, Mental Illness in America: The National Institute of Mental Health Agenda, 1998

 

31. U. S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, The Costs and Effects of Parity for Mental Health and Substance Abuse Benefits, 1998
32. White, R., "Employers fine-tune Plans to Comply with Parity Law," Business Insurance, 1998
33. Zuvekas, Samuel, et al., "Mental Health Parity: How Large are the Gaps in Coverage?" Achives of General Psychiatry, 1997

The following listing of informational Internet web sites is not meant to be exhaustive but rather to give the reader a number of well-designed, informative sites that also contain further links to valuable sites.

1. www.athealth.com Mental Health Resources, @Health

2. www.mentalhealth.com Internet Mental Health

The following documents are the written testimony provided to the Task Force.