Financing of Drug Treatment Services

Literature Review
December, 1997

- Author's Note
- Executive Summary
- Introduction and Background
- A Financing Framework
- Private Insurance Coverage
- Public Financing
- Benefit and Coverage Design
- Carve-Outs and Risk Contracting
- Risk Sharing
- Performance Issues
- References

Author's Note

The author has drawn on a considerable literature on the financing of drug abuse treatment published since 1991. These works are cited in the bibliography, and they have been the basis for conclusions about the need for future research provided in the text. Works that are cited in the text indicate specific sources of facts, arguments, or viewpoints.

Sections

Executive Summary

This paper presents suggestions for research needed in the financing of drug abuse treatment. The literature on this topic has been reviewed and critically appraised to determine the gaps in theoretical, conceptual, and empirical knowledge. The appraisal considers the way in which financing influences incentives of payers of insurance coverage, providers of treatment, and consumers of services. The issues presented encompass both private- and public-sector financing of drug abuse treatment. The financing dimensions for which future research should be undertaken are mandates, the design of benefits packages, carve-outs, consumer cost sharing, various provider risk-sharing instruments, and program performance.

Sections

Introduction and Background

The purpose of this paper is to present suggestions and observations about future research needs with respect to the financing of drug treatment in the United States. This objective is undertaken through a critical review and synthesis of the theoretical, empirical, and policy explorations found in the literature on drug abuse treatment financing and the changes and status of funding arrangements with a focus on writings since 1991. The paper does not entail an in-depth description of the analysis, discussions, and findings of the literature; rather it provides an assessment of areas in which research is needed to enhance the economic efficiency of drug treatment financing. The commentary on future research directions encompasses two sources of financing for drug abuse treatment: (a) private-sector health insurance, and (b) public-sector health financing through the insurance program of Medicaid and direct delivery government programs for the uninsured. Numerous financing dimensions — financing policies, instruments, mechanisms, and characteristics — are considered for relevance to drug abuse treatment within the context of the two sources of financing. These dimensions are (a) taxation and mandates in private financing, (b) design of benefits packages and the issue of parity coverage, (c) consumer cost sharing and demand, (d) "carve-outs," (e) risk-sharing instruments, and (f) program performance. These financing dimensions are not mutually exclusive, so a discussion of them overlaps for both financing sources.

Sections

A Financing Framework

The financing of drug abuse treatment in the United States is a bifurcated system composed of private- and public-sector funding sources. The private sector, which encompasses spending by private insurance companies and out-of-pocket expenditures by clients, accounts for 30% of drug abuse treatment funding in the United States (Kronson, 1991; Rouse, 1995). The privately financed system has delivered drug abuse services, along with other behavioral health care, mainly to lower middle to high income insured individuals and their dependents. Prior to the 1990s, drug abuse treatment was dominated by an inpatient modality. In recent years, service delivery has shifted to outpatient facilities due to the implementation of managed care in many health plans in which financing and organizational service delivery have been combined. Prior to the 1980s, indemnity insurance plans were the dominant form for private-sector health care financing. Independent providers were paid on the basis of a fee for service — payment for each unit of care — by insurance entities that acted strictly as risk pooling agents.

Public-sector drug abuse treatment has been financed through public insurance —primarily Medicaid, a joint federal and state government program — and direct delivery programs that have been jointly funded by federal government intergovernmental grants and both state and local government expenditures. These programs have been delivered by the latter governments either with their own agencies or, more generally, through contracts with health care providers. The clients and/or target populations of public programs have been low-income and economically disadvantaged individuals and households who sometimes are treated in public hospitals, but mostly are cared for in residential and outpatient facilities. In the past 5 years, managed care has been initiated for Medicaid in a limited number of states through waivers granted by the federal Medicaid oversight authority of the Health Care Financing Agency (HCFA) (Callahan, Shepard, Beinecke, Larson, & Cavanaugh, 1994; Goplerud, 1995).

A central question for financing drug abuse treatment is how to constrain service costs while ensuring ready access to and quality of care services that provide effective outcomes for patients. Such concern has arisen because of considerable growth, on an annual basis, of public and private expenditures for drug abuse treatment services in the past 2 decades (e.g., Government Accounting Office [GAO], 1993; Tommasini, 1994). This increase has paralleled a large rise in medical care financed by both sectors over the same period (GAO, 1993). A major consequence has been that various financial instruments have been proposed or introduced as means that might constrain costs and ensure access to and quality of care. These instruments could affect the supply of and the demand for substance abuse health services. They are expected to be economically beneficial by curbing the incentives of payers, consumers, and providers for excessive utilization that produces high cost for treatment. As a consequence, they could determine the extent to which economic efficiency can be obtained, but in so doing they could redistribute income among the participants involved in the financing of services.

On the demand side, cost sharing by consumers has been employed as a rationing device to inhibit consumer behavior brought about by the "moral hazard" of insurance. A moral hazard occurs because insurance coverage reduces the net price of care to the insured with the result that additional care is demanded. Hence, excessive services are consumed. Cost-sharing instruments — coinsurance, copayments, and deductibles — have been proposed or implemented to raise the price of care to the insured consumer in order to decrease demand for such care. The curtailing of utilization would bring the quantity and thus the costs closer to consumers’ true valuation of services. Therefore, economic inefficiencies would be mitigated.

On the supply side, instruments have been proposed or implemented encompassing benefit limits and design, risk sharing/spreading between payers and providers, and required management techniques to be applied by providers. Benefit package designs, along with payment mechanisms (e.g., prospective payments, capitation, fixed budgets), are expected to influence the incentives to constrain service utilization prescribed by treatment providers, thereby reducing treatment costs. To offset the possibility that some providers may minimize treatment quality in order to maximize profits, management techniques such as utilization review and admission precertification have been instituted.

Partial capitation, risk adjustment, risk contracting, reinsurance, and performance contracting have been suggested as effective risk-sharing instruments. These instruments are expected to deal with adverse selection by which individual choice of health plans results in either low health risks or high health risks being concentrated in particular plans so as to produce respectively large profit margins or financial losses to the insurer or provider given the fixed compensation for service provision. Likewise, these risk-sharing instruments may be effective in preventing or reducing risk selection by insurers and/or providers. That is, these two groups have strong economic incentives to choose consumers who are low health risks, since treatment costs are lower and higher profit margins can be obtained. Finally, performance contracting can be viewed as a mechanism that addresses both adverse and risk selection problems but also establishes provider accountability for service effectiveness. All of the above-mentioned supply side constraints, individually and collectively, have been posited as contributors to the reduction in service levels and thus costs; if so, they could improve efficiency of treatment by a closer alignment of service costs with the value of consumer demand.

Sections

Private Insurance Coverage

Approximately 75% of individuals with private health insurance have acquired coverage for drug abuse treatment (Jensen & Morrisey, 1991; Kronson, 1991). Most of this coverage is purchased by employers for their employees and their dependents through employer-sponsored health plans (GAO, 1993; Jensen & Morrisey, 1991). Insurance is generally financed by joint contributions of the employer and the employee. Some of this coverage for drug abuse treatment may stem from state mandates enacted in the 1980s (Jensen, Cotter, & Morrisey, 1995; Kronson, 1991; Scott, Greenberg, & Pizarro, 1992). Although since 1993, 24 states plus Washington, DC, have promulgated employer insurance coverage for alcohol treatment, only 18 states have stipulated drug abuse service insurance coverage, jointly with coverage for alcohol treatment (National Institute on Drug Abuse [NIDA], 1992b; Scott, Greenberg, & Pizarro, 1992). In the plans carried in the 18 states, detoxification is required for nearly 100% of all insured individuals, but outpatient care and rehabilitation treatment are given lower priority (Scott, Greenberg, & Pizarro, 1992). Not only do mandates specify a preference for hospital-based treatment for drug abuse, they also require service delivery by particular providers, generally certified counselors, psychiatrists, and/or psychiatric nurses (Kronson, 1991; Scott, Greenberg, & Pizarro, 1992). In comparison to medical care, however, coverage limits have been more restrictive (e.g., lower in the number of days) and copayments have been higher for similar modalities. Inpatient detoxification for drug abuse, the most prevalent service covered, has been granted higher limits than outpatient and inpatient rehabilitation services (Kronson, 1991; Rouse, 1995).

Many firms have escaped state mandates by becoming self-insured under ERISA (Employee Retirement Income Security Act of 1974). Firms are exempted from state taxation and regulation pertaining to self-insured employer health plans. These plans encompass almost 50% of insured employees of medium and large firms; self-insurance is virtually universal in large firms (Jensen et al., 1995; Larson, Bowden, & Hogan, 1991). Self-insurance plans are both indemnity and managed care plans.

Because the cost of coverage rose during the late 1980s and 1990s, the percentage of wage and salary workers insured through employment dropped from 68.6% in 1988 to 65% in 1993 (Frontin & Snider, 1996). High health insurance costs could be an obstacle to maintaining coverage. For the typical firm, health care costs increased substantially between 1980 to 1990 from 5.8% to 8.5% of total employee wages and salary. Over this period, with wages adjusted for inflation, real wages fell slightly, indicating that health coverage was a substitute for take-home wages and salary.

Cost increases have induced some firms to initiate greater employee cost sharing (e.g., coinsurance), for their health plans as well as to offer alternative and competing health plans, especially managed care. Small firms, whose health care cost increases have been greater than those for larger firms, have dropped coverage (GAO, 1993). Approximately 66% of all uninsured workers are employed in smaller firms (100 or fewer employees) (Thorpe, 1995). These economic conditions are not conducive to the retention or provision of insurance for substance abuse treatment (Frank, Salkever, & Sharfstein, 1991).

There are several major, and intertwined, issues regarding taxation and regulatory policy impacts on insurance for substance abuse treatment. A central thread that runs through them is the demand for drug abuse services.

1. Analysis should be conducted on the role played by the "tax subsidy" of employee benefits compensation in the provision and demand for substance abuse coverage. Health insurance premiums paid by employer and employee are not taxable income. Consequently, the price of insurance is reduced, and a moral hazard is fostered. Consideration must be given to how the progressive tax subsidy (the higher the income, the larger the value of the subsidy) influences the extent and type of coverage and determines the incidence of the cost of insurance. Evaluation of the tax subsidy should encompass the effective price of insurance paid by the firm, taking into consideration that the larger firms obtain economies of scale in the purchase of insurance. Moreover, institutional forces (e.g., workforce characteristics, nature of the industry) that could shape decisions about insurance coverage should be incorporated in the analysis.



2. Studies should be undertaken to determine consumer demand for substance abuse treatment. This effort should focus on several dimensions: (a) the demand for services by individuals/households with and without insurance; (b) the demand for types of services (e.g., inpatient, outpatient detoxification, residential rehabilitation); and (c) demand with the application of cost-sharing mechanisms — deductibles, coinsurance, copayments, stop-loss provisions. These analyses should determine the price elasticity and income elasticity (taking into account different income levels) controlling for the social and economic characteristics of consumers. The analyses also should include the estimation of substitution of different modalities by consumers through a determination of cross-price elasticities.



3. Research is needed to assess the economic implications of employer mandates for substance abuse. As a regulatory instrument, mandates are a "hidden" tax in the form of a required increase in the costs to a firm if its benefits package is less then the package mandated. Mandates are expected to address adverse selection by employees due to differential provision of substance abuse coverage by employers and the failure of the market to provide such coverage. There is a need to measure the efficiency costs of a mandate in the labor market. The costs of the required coverage are expected theoretically to result in losses in jobs and a reduction in wages/salary. Short-run and long-run impacts, especially on real wages, should be examined. Two approaches should be undertaken. The efficiency costs incurred in states that have mandates can be estimated. Estimation of the cost of labor market inefficiencies also can be appraised in terms of a universal, nationwide mandate by the federal government. In both instances, the focus should be on the nature of coverage requirements and the financial size of the mandate(s). Finally, there should be an evaluation of how mandated coverage would affect income distribution, since the financial burden of mandated benefits will vary disproportionately according to income levels of employees

Empirical analysis would be useful for understanding the incentives of employers’ provision of employee coverage.

1. Have state mandates led to self-insurance by firms under ERISA? What types and limits of coverage have self-insured firms adopted? Have firms used self-insurance to escape substance abuse mandates, or have they been motivated by the "cumulative" burden imposed by other mandates? Have firms self-insured because of state fiscal and regulatory policies (e.g., state taxation of insurance premiums and state assessments for insurance pools) or other economic forces?



2. In states with mandated substance abuse coverage, have employers with mandated coverage been plagued with adverse selection and the expected consequence of an immobile work force?



3. When mandates for substance abuse coverage and other types of health care have been initiated, have firms engaged in health care substitution to maintain budget neutrality (i.e., have they downsized their financial commitment to other parts of their health plan), have they chosen to implement cost-saving measures (e.g., modality substitution), or have they offered a managed care alternative?



4. What has been the experience of mandated firms in terms of their costs and client access to drug abuse treatment? A comparison of mandated and nonmandated firms with similar and different health benefit packages would yield useful evidence.

Other potential detrimental "external" effects could occur due to variations in state mandates that warrant examination.

1. In states with mandates, do self-insured firms have a competitive advantage over mandated firms in the selling of their products and labor cost, recruitment, and retention?



2. What is the impact of mandates across states in terms of sales and labor market competition, and on incentives for firm location within a state?

There are two intertwined concerns for financing substance abuse treatment with respect to small firms. One, what financial instruments can be employed to provide coverage for the substantial number of uninsured employees? Two, how can the cost/price of existing insurance be reduced for small firms in the small group market? Common to both these concerns is small firms’ inability to make coverage affordable by obtaining reduced premium prices for the same benefits packages that large firms receive. Although insurance carriers may not aggressively pursue business with small firms because of their (perceived) high-risk workforce, risk selection is also a major problem for firms carrying insurance. Small firms that provide employer-based insurance are likely to be subject to medical underwriting whereby the assessment of health risk of individual employees is required. Such behavior allows insurance carriers to engage in churning (i.e., to discourage renewals of high risks or reduce coverage).

Several financing instruments in the form of risk-pooling mechanisms — purchasing cooperatives, reinsurance — could address the two concerns. Although these instruments could enhance the financing of substance abuse treatment, little substantive knowledge about their contribution is available and thus future research is warranted. Reinsurance is discussed in the section on risk sharing.

Small employers can establish cooperatives to buy health insurance that includes substance abuse treatment coverage. Through risk pooling, health costs are distributed across many employers. Cooperatives can reduce the cost of acquiring insurance for their participant employers in two ways. First, the cooperative arrangement could reduce marketing and administrative costs in the acquisition of insurance plans. Second, because of its size, due to the large number of employees represented, the cooperative can exert bargaining power to negotiate a contract or strongly entice competitive bidding, both of which should reduce the price of insurance. Questions for research include the following: First, how should cooperatives be organized internally (and what criteria should be used) so that costs are equitably allocated? Second, should cooperatives be organized geographically and have monopoly power within a spatial boundary? Third, should membership be compulsory to avoid adverse selection (i.e., the migration of lower/higher risk to them if employers had a choice of cooperative)? Fourth, how much would be realized in cost savings? Finally, would mandates that required substance abuse treatment be needed to achieve such coverage through the cooperative?

Sections

Public Financing

Public financing of substance abuse treatment, which accounts for 49% of such funding in the United States, involves (a) the federally directed Medicaid insurance program and (b) direct delivery programs conducted by state and local governments through either their own agencies or contracts with providers. Medicaid is financed through a federal open-ended grant to state governments that must supply matching funds to conduct the program. The Medicaid program provides health care coverage for people between 18 and 65 years old who are uninsured, low income, at or below the federal poverty level (FPL), generally qualified, or receiving federal and/or state income supports. Where Medicaid programs have been granted a managed care waiver by HCFA, state programs have raised income levels above the FPL for eligibility and allowed exceptions of pregnant women. With both "traditional" (fee for service) and managed care Medicaid programs, coverage for drug abuse treatment has remained very limited; Medicaid finances 9% of all drug abuse treatment in the United States. In effect, the Medicaid program serves mostly women and children in all states, excluding nondisabled, uninsured, low-income, single men. This latter group is protected for substance abuse treatment through the direct delivery programs that have restricted coverage and, like Medicaid, limited funding and capacity characterized by waiting lists for service. These direct delivery programs, which are responsible for 40% of all U.S. spending for drug abuse services, are financed through a fixed federal budget for block grants that are allocated among states by a formula. The states may supply additional funding.

Direct delivery programs are financed through federal alcohol, drug, and mental health block grants. These grants subsume separate setasides for funding mental health and alcohol and drug abuse. Much attention has been focused on the components of the formula that allocate funding among and within states. Empirical studies are needed, however, on the evaluation of the impact of block grants on state fiscal behavior with respect to behavioral health and in particular drug abuse. Economic theory indicates (a) that block grants as lump sum transfers should increase spending from their own funding by less than the amount of the grant, and that the recipient will shift some of its own funding to other activities (fungibility), and (b) that spending increases from the grant will be less than the influence of the state’s growth in personal income. However, the "flypaper" effect, (i.e., "money sticks where it hits") has been found in empirical research on grants. The result is that increases in lump sum grants raise spending more than increases in the personal income of the recipient government. Research on how much block grants affect state spending also should include the evaluation of federal block grant regulations, especially the maintenance of effort requirements. In addition, the leakage of block grant funds into services or taxes can be examined to appraise some of the income distribution impacts of the grants. Moreover, consideration should be directed at whether the type of government agency — a solely independent substance abuse agency or an inclusive behavioral health agency — influences state allocation of grant and state-owned moneys. Finally, analysis should be undertaken to assess the effects of alternative types of grants on state drug abuse activities. Because of their price effects, open-ended and closed-ended matching grants under certain economic conditions should stimulate state spending more than do (lump sum) block grants. If these types of grants were selected as policy instruments, larger amounts of resources could be allocated to substance abuse treatment. This analysis also should involve the evaluation of the appropriate formula to determine the matching ratio for allocation of the grants. A starting point would be the open-ended grant formula that is the funding basis for Medicaid.

Research should be conducted with respect to the Medicaid program. One major objective would be to determine the financial costs of expanding drug abuse treatment coverage for all Medicaid programs. This analysis should provide cost estimates for varying coverage limits and clientele cost-sharing arrangements as well as differences in managed care and non-managed care service delivery. Consideration should be given to various budget constraints that are politically acceptable. Second, an analysis should be undertaken to appraise the long-run strategy of phasing in all the uninsured poor under Medicaid. This "mainstreaming" assessment should include an array of cost estimates beyond those of drug abuse treatment. The evaluation should be based on different coverage limits, client cost-sharing arrangements, politically acceptable budget constraints, adjustments of the grant formula that allocates federal expenditures to states, and the cost savings that would be obtained from the elimination of the block grant program.

If the fiscal support of federal grants and Medicaid were increased for drug abuse treatment financing, the resulting actions likely would involve expansion of both federal and state budgets or contraction of spending on other activities. In either situation, substantial opportunity costs could be incurred. It is imperative to assess the efficiency costs and income distribution effects of the budgetary expansion (or likewise the spending contraction), indicating who will bear the burden of paying for the enhanced coverage of substance abuse treatment (as well as perhaps other behavioral health services). Therefore, after the cost analyses of service expansion are completed, there should be an evaluation of (a) the tax handles that could be employed at each level of government to finance the expanded service levels and (b) the interrelationship of the tax handles of both levels of government.

Sections

Benefit and Coverage Design

The types and scope of services of allowable benefits, as well as the mechanisms chosen to deliver them, determine the costs of substance abuse treatment. The economic efficiency of financing drug abuse treatment requires a cost-effective design of its benefits package, since financing generates the required revenues. A consensus does not prevail, however, on the optimal design of benefits packages for substance abuse treatment. There is a lack of agreement on the components of a benefits package of an efficiently designed benefits structure as well as limited knowledge about the various dimensions of each suggested component.

Four components are necessary for a benefits package: (a) coverage limits; (b) modalities, or alternative delivery mechanisms; (c) consumer cost sharing; and (d) provider cost sharing/shifting. (Consumer cost sharing has been discussed earlier with respect to consumer demand for drug abuse treatment. Issues regarding provider cost sharing are raised in the discussion on risk sharing.) These components encompass both supply and demand incentives in the delivery of treatment and thus can influence the cost of provision. Although the components are interrelated, research on them can proceed separately, but the findings must be synthesized in order to produce an integrated design or designs that allow rationing of access and constraint of costs but also provide effective outcomes for patients. To judge the comparative contributions to economic efficiency, cost estimates of various designs are needed, and simulations should be conducted not only on separate components but also on the integrated components as models of benefit design. These analyses should be conducted within the parameters of clearly acknowledged budget constraints (e.g., budget neutrality or various levels of budget expansions).

• Coverage limits (e.g., types of illness treated, days or units per year, lifetime maximums) have been the traditional way that private insurance and public programs have specified (and restricted) treatment utilization and thus heavily influenced the costs of services. For spending on services to enhance efficiency, the types and amount of coverage for drug abuse should correspond to the clinical nature of drug abuse illness, which is chronic and recurring; has long-term effects, perhaps with lifetime persistence; and is characterized by relapses. The short-term and the long-term effectiveness of treatment may require a continuum of care inclusive of wraparound services (e.g., vocational training, child care). Studies are needed to ascertain the extent to which various mixes (of quantity and types) of services would contribute to the improvement of patient well-being. This effort to assess the impact of services means that patient outcomes must be established. Therefore, research objectives must entail the determination of performance standards that measure the improvement in patient health and even social status if it is determined to be a valid indication of successful progress due to treatment.



• The cost and effectiveness of drug abuse treatment are likely to vary by the type of modality, or alternative service delivery, that is prescribed to patients. Acute inpatient care, residential care, outpatient care, and other modalities not only vary in cost per unit of service, but also could have a differential impact on patient outcome. To the extent they are substitutable, cost offsets could be obtained. Empirical analysis therefore could provide estimates of the effects of different modalities as well as substitution rates among modalities to determine the cost offsets. These analyses would yield knowledge about the more cost-effective design of benefit limits.



• Consumer cost sharing in the form of deductibles, copayments, and coinsurance should influence the design of benefits packages. In part, utilization and consequently costs of drug abuse services depend on the demand for treatment by the consumer/patient. The nature and extent of consumer demand for services can be influenced by cost sharing, which changes the price of services. Studies of demand for drug abuse treatment to determine the price sensitivity of the consumer/patient have been called for above. The results of such work should be extended to link demand with coverage limits, modalities, and patient outcomes in order to formulate a benefits package that takes into account consumer responsiveness; this will minimize excessive consumption due to moral hazard. This analysis should assess the way in which cost-sharing arrangements influence individuals of different income levels and whether flexibility and the size of the financial burden of cost sharing should vary according to different phases and types of treatment.

An issue pertinent to the appropriate benefits package is that of the parity of behavioral health in general, and drug abuse treatment in particular, with medical care financed through private insurance and public programs. The initial focus regarding parity has been on mental health services for both state government and federal proposals. In this respect, a major thrust involving parity has been the passage of the federal government’s Mental Health Parity Act of 1996. The act, effective in 1998, requires that if group health plans offer mental health benefits, the plans should include the same amount of benefits for mental health care as provided for medical and surgical benefits. The annual and aggregate lifetime annual limits must be identical. This parity requirement does not apply to substance abuse or chemical dependency, nor does it apply to small employers (50 or fewer employees), or if expanded coverage results in a 1% increase in the costs of benefits.

The Mental Health Parity Act is representative of a policy approach that can be called quasi-compulsory/voluntary. That is, equality of substance abuse and medical care services of a health plan can be achieved only through the willingness of firms providing insurance protection to continue substance abuse benefits. Because providing substance abuse parity would increase their health care insurance costs, and the parity requirement is contingent on the offering of such services as part of a health plan, firms have a strong incentive to drop substance abuse treatment coverage altogether. A research question agenda for evaluating this quasi-compulsory/voluntary policy regime should explore:

• The extent to which firms would drop their existing substance abuse benefits.



• The extent to which firms would reduce medical care coverage to establish parity. Consideration should be given to health insurance coverage: whether budget neutrality would be pursued or the extent to which firms are willing to bear some increased costs in the substitution of medical care for substance abuse services.



• The effects among firms for labor competition (inclusive of wage impacts) due to differential coverage of substance abuse services.



• Whether adverse selection would worsen or improve as a result of differential parity implementation by firms.



• Estimation of cost shifting for substance abuse, if any, to the public sector, given that some firms would drop substance abuse coverage.

An alternative policy approach to the quasi-compulsory/voluntary regime is implementation of mandates for the parity of substance abuse and medical care of employer-based health insurance. Parity mandates are more likely to be advocated and to occur at the state level, since many state governments have used employer mandates for medical care insurance coverage. (Such mandates also could be spurred by detrimental actions of firms dropping substance abuse coverage or reducing health care coverage under quasi-compulsory/voluntary legislation.) Irrespective of their legal source (federal or state level of government), parity mandates require research. The inquiry should consider whether parity mandates stipulate that equal medical care and substance abuse services are to be provided by all firms under employer-based health insurance coverage, or that firms must match their substance abuse coverage with their medical care coverage, but that benefit level (above a minimum) can be determined by firms. The research issues are:

• Inquiry into the potential economic efficiency losses that could be incurred in labor markets, as was described above.



• Estimates of the increases in financial costs of health insurance not only totally relative to the economy but to firms of different sizes.



• Assessment of whether adverse selection would be mitigated or intensified with mandates.

Another line of inquiry would be to follow the view expressed above with respect to the design of benefits packages. With this perspective, before decisions about coverage parity for drug abuse treatment are made, the effectiveness and costs of such services must be determined. The underlying assumption of this research direction is that the effectiveness of service units may differ among types of illness and therefore different levels of benefits are required for types of care. Such analysis does not obviate the need to address how such coverage should be financed; consideration would still have to address the issues of the efficiency costs of mandates.

Sections

Carve-Outs and Risk Contracting

With the advent of managed care, the establishment of carve-outs (or Managed Behavioral Health Care Organizations [MBHCOs]) has proliferated as a means for treating mental health illness, alcohol abuse, and drug abuse. Purchasers set aside moneys for financing behavioral health care through contracting with vendors that manage — singularly or combined — mental health, alcohol abuse, and drug abuse risks. Carve-outs generally are remunerated with a prospective payment system, many with a risk-sharing basis. Behavioral health services also can be conducted through indemnity plans, managed care plans such as HMOs and PPOs, and integrated service delivery systems (which are providers organized into networks to manage and finance behavioral health care). In the private sector, many carve-outs are used in concert with Employee Assistance Programs (EAPs). Carve-outs have been publicly financed mainly through the Medicaid program, with the hiring of either private-for-profit or private-nonprofit MBHCOs.

Opposition to carve-outs is based on the view that with a comprehensive health care system, the primary care physician is pivotal in the diagnosis of drug abuse and can prescribe unified, coordinated health care. Support for carve-outs stems from (a) use of specialists for particular illnesses so that higher quality care can be provided, (b) mitigation of adverse selection due to availability of choices among plans delivered through integrated care, (c) mitigation of risk selection by providers and insurers who seek to avoid patients with chronic illnesses, and (d) better management of moral hazard because of knowledge of specialty providers.

Several areas of research on carve-outs are needed:

• There are very limited empirical analyses on the performance of carve-outs. Studies should be conducted to assess the impact of carve-outs on costs, utilization, and access. These studies should encompass a comparison with the performance of other managed care alternatives (e.g., HMOs, PPOs, and indemnity plans). In addition, where carve-outs produce treatment for more than one behavioral health illness (comorbidity), the existence of economies of scope should be determined. Furthermore, private- and public-sector carve-out experiences should be compared. This effort must consider (or control for) the differences in clientele that participate in the programs of each sector. Finally, analyses are warranted for evaluating the impacts on costs, utilization, and access that could occur in the public sector because private-for-profit and private-nonprofit MBHCOs have different incentives.



• Studies should be undertaken of public-sector carve-outs to determine whether cost shifting occurs with respect to the Medicaid program. Costs could be shifted within parts of the Medicaid budget, or costs could be shifted to other social services if carve-outs are employed.



• Because some Medicaid programs select several MBHCOs to deliver behavioral health, evaluation should be made of whether adverse selection and risk selection prevail.



• Studies need to determine what the costs and financial risks are if MBHCOs subcontract with providers who receive capitation payments or fees for services.

A central issue in carve-outs is risk-sharing contracts. Within the context of carve-outs, risk contracting is a legal agreement between a payer and the vendor (MBHCO) on the extent to which financial risk for managing the behavioral health care of the insured (health risk) will rest, in part or in toto, with the carve-out vendor (MBHCO). The contractual specification determines the incentives of vendors to provide the quantity and quality of treatment. Vendors can be awarded a capitation payment (an amount per person per year) where all the financial risk rests with the vendor. Or, the vendors can receive a partial (or soft) capitation payment whereby the financial risks of service delivery are shared between the payer and the vendor. Partial capitation entails establishing a risk corridor that specifies a target capitation value with the vendor receiving more if total costs incurred are above the targeted amount or less if total costs incurred are below the targeted amount. An additional incentive to provide quality treatment may be achieved through building penalties and rewards for obtaining service goals into contract specifications. The following research topics should be undertaken with respect to risk-sharing contracting:

• Which designs of risk-sharing contracts are cost-effective; what criteria should be employed to judge such designs; and what is the impact on vendor incentives for service delivery? Empirical analyses should be conducted to verify the propositions derived.



• How should risk corridors be determined, and what are the estimates? What are the client utilization and cost implications of different types and amounts of risk corridors?

Sections

Risk Sharing

Purchasers of private health insurance have sought to control adverse selection in the choice of health plans. With the array of available plans, high-risk individuals, such as drug abusers seeking treatment, and low-risk individuals would participate in separate health plans with the consequence that some plans would incur financial losses or limited profits, and others would earn a high profit margin. Likewise, insurance carriers can engage in risk selection, or "cream skimming"; they can charge uniform premiums to all (for the same policy) but pursue lower risks to reduce payouts and thus enhance profits. In both the public and private sectors where capitation is employed as a risk-sharing payment mechanism for providers, two results are likely. One, the provider can be put at financial risk due to the adverse selection of clients. Two, although capitation could induce the provider’s incentive to constrain treatment costs, it also could create an incentive for the provider to undertake risk selection to increase its profits.

Several financial risk assessment instruments have been developed to address these problems. Partial (or soft) capitation, carve-outs, purchasing cooperatives, and risk contracting have been discussed above. Two other instruments — risk adjustment and reinsurance — are pertinent. The literature points to a need for three general evaluations. First, determination should be made of whether there is sufficient adverse selection and risk selection to warrant the employment of risk instruments. Second, there is a need to evaluate the complementarity and compatibility of all these risk-sharing instruments. Research should indicate whether the instruments or their components are inconsistent with each other. Third, because much of the concern for risk assessment with respect to providers stems from the use of capitation, the impact of this payment mechanism on access, utilization, and the quality of drug abuse treatment should be explored empirically.

Considerable research has been conducted on risk adjustment. Risk adjustment is to ensure either fair premiums to an insurer for a health plan or capitation payments to a provider for bearing the financial risk of services. Payment should not be overly generous so as to generate large profits, nor should it be insufficient to cover costs. Central to this risk approach is the determination of the appropriate payments by estimation of econometric models that explain the variation in health expenditures. Numerous analyses of empirical model testing have been undertaken to find accurately predictive risk adjusters that measure the risks of medical conditions; however, the research has not been very fruitful. Nevertheless, additional work on development should be continued with consideration given to the improvement of data sets and model specification. Even if this long-run effort succeeds, given the random nature of health expenditures, a large amount of variation will not be explained. Predictive premiums and capitation payments will not be sufficiently accurate, so adjustments to them will be required.

The literature does not explain how the adjustment would occur. Research should be directed to how compensation mechanisms of risk adjustment would work. Analysis is needed of the administrative process and institutional arrangements of the organization that would regulate and allocate the risk adjustment of premiums or provider capitation payments and activities. In addition to the internal institutional concerns, investigation of the external conditions for conducting risk adjustment should be undertaken. More specifically, there should be analyses of the market forces that could shape the effectiveness of the risk-adjustment process. These analyses would encompass the spatial scope (e.g., regional, national) of the risk adjustment process, the rules of participation, and the role of government. Finally, evaluation should be made regarding whether compensation through risk adjustment is more or less costly than the enrollment of drug abusers in specialized care such as carve-outs. Moreover, where risk adjustment has been implemented, empirical studies should be conducted regarding its effect on access, costs, quality, and outcomes of drug treatment.

As a risk-sharing instrument, reinsurance allows insurance carriers to place high-risk individuals or groups in separate pools so that additional coverage is extended to them. Consequently, reinsurance lowers insurer incentive to engage in risk selection. With mandatory reinsurance, the financial costs of high risk would be distributed across carriers. Reinsurance also could be used in the small group market as a complement to a purchasing cooperative. The cooperative could allocate the high risk across all participating small firms. It would appear that reinsurance does not reduce financial costs directly for its participants, but it may facilitate access to lower-cost insurance than individual firms would obtain, and it mitigates a basis for insurance carriers to engage in cream skimming.

Several research issues should be undertaken given the activities encompassed by reinsurance.

• What rules should be adopted that define the allocation of risk to the reinsurance pool so as to minimize risk selection?



• How are participants assessed for including their risks in the reinsurance pool?



• What are the criteria that should be operational to distribute the financial losses of health risks across pool participants?



• Are mandates for participation required to make the reinsurance pool function more efficiently, and what form should the mandates take?



• Empirical analyses should be conducted on the performance of cooperatives that do and do not have reinsurance.

Sections

Performance Issues

Recently, the performance of drug abuse treatment programs has been a major financing issue. This issue is manifested in the development of performance contracting and performance standards for federally financed programs. The federal "Government Performance and Responsibility and Accountability Act (GPRA) of 1993" requires federal programs to demonstrate measurable results. The Department of Health and Human Services has proposed making public health programs financed by block grants (inclusive of behavioral health funding) into Performance Partnership Grants (PPGs) that would have an outcome assessment requirement. However, this transfer has not yet been implemented. With performance contracting, states would monitor outcomes to assess the performance of the programs for which providers have been contracted to deliver. Providers that meet performance standards within the contract amounts would be rewarded. If standards were not realized, then penalties would be applied. Thus, enhanced performance by providers would be stimulated by economic incentives.

These developments reflect an application of the concepts of program budgeting and the belief that drug abuse treatment programs should be outcome based. The thrust of performance orientation is to link levels of program expenditures with quality of care. The latter can subsume both effectiveness (i.e., client outcomes) and technical efficiency in the form of unit cost of service delivery. Analyses of effectiveness and technical efficiency should provide the bases of budgetary allocation among providers. By doing so, some financial risk is shifted to providers, and they are accountable for service provision. In effect, if performance requirements are implemented, there is potential for the more efficient financing of drug abuse treatment.

Virtually no conceptual and empirical work has been conducted on performance contracting and performance outcomes pertaining to drug abuse treatment, although there is considerable literature on program budgeting developed since the 1960s to draw upon as a framework. As a result, considerable research is needed. At a minimum, the following are required:

• Development of outcome measures to determine the effectiveness of drug abuse treatment. This work should involve the determination of measures for both short-run and long-run results. The research should pursue consensus among scholars and practitioners regarding the validity of likely measures of outcomes. An empirical analysis should be conducted on the relationship between outcomes and program activities so as to establish the functional impact of treatment characteristics and various outcomes. In turn, these relationships can also establish knowledge about the parameters of provider contributions to favorable outcomes. Such information can be used in the implementation of performance contracts.



• Development of various designs of performance contracts. This effort should include (a) appraisal of the rewards and penalties as incentives to influence the performance of providers, and (b) how these incentives would be operationalized.



• Consideration of the rules, regulations, and incentives that would facilitate (a) the adoption of outcomes measures by the public and private sectors, and (b) the reporting of treatment outcomes by these organizations.



• Development of the criteria for successful performance. This work should involve consideration of whether the criteria should be based on (a) overall performance of a program (e.g., average gains for all clients), (b) the outcomes of individual clients, or (c) both. In any case, the scale of effectiveness in performance must be delineated.



• Development of decision rules for budgetary allocation of drug abuse treatment resource when budget decisions are based on program performance.

Sections

References

Alpha Center. (1997). Risk adjustment: A special report. The Robert Wood Johnson Foundation.

Anderson, D. F. (1989). How effective is managed mental health care? Business & Health, 7 (9), 34-35.

Arons, B. S., Frank, R. G., Goldman, H. H., McGuire, T. G., & Stephens, S. (1994). Mental health and substance abuse coverage under national health reform. Health Affairs, 13 (1), 192-205.

Berlant, J., Trabin, T., & Anderson, D. (1994). The value of mental health and chemical dependency benefits: More than meets the eye. In E. Sullivan (Ed.), Driving down health care costs: Strategies and solutions. Frederick, MD: Panel Publishers.

Bevilacqua, J. J., Elias, E., & Navon, M. (1995). Perspectives. Health Affairs, 14 (3), 45-49. The above refers to "states’ embrace of managed care" by Essock and Goldman, which appears in the same issue of Health Affairs.

Boyle, P. J., & Callahan, D. (1995). Managed care in mental health: The ethical issues. Health Affairs, 14 (3), 7-22.

Brindis, C., Pfeffer, R., & Wolfe, A. (1995). A case management program for chemically dependent clients with multiple needs. Journal of Case Management, 4 (1), 22-28.

Browne, M. J. (1989). Evidence of adverse selection in the individual health insurance market. Journal of Risk and Insurance, 56, 13-33.

Burke, A. C., & Rafferty, J. A. (1994). Ownership differences in the provision of outpatient substance abuse services. Administration in Social Work, 18 (3), 59-91.

Burton, W. N., Hoy, D. A., Bonin, R. L., & Gladstone, L. (1989). Quality and cost-effective management of mental health care. Journal of Occupational Medicine, 31 (4), 363-367.

Callahan, J. J., Shepard, D. S., Beinecke, R. H., Larson, M. L., & Cavanaugh, D. (1994). Evaluation of the Massachusetts Medicaid Mental Health/Substance Abuse Program. Report submitted by the Heller School for Advanced Studies in Social Welfare, Brandeis University.

Callahan, J. J., Shepard, D. S., Beinecke, R. H., Larson, M. L., & Cavanaugh, D. (1995). Mental health/substance abuse treatment in managed care: The Massachusetts Medicaid experience. Health Affairs, 14 (3), 173-184.

Cantor, J. C., Long, S. H., & Marquis, M. S. (1995). Private employment-based heath insurance in ten states. Health Affairs, 14 (2), 199-211.

Chollett, D. (1992). Guest editorial: Building a private-public partnership that works. Inquiry, 29, 116-119.

Chollett, D. (1994). Employer-based health insurance in a changing work force. Health Affairs, 13 (1), 317-326.

Committee on Child Health Financing and Committee on Substance Abuse. (1995). Financing of substance abuse treatment for children and adolescents. Pediatrics, 95 (2), 308-310.

Commons, M., & McGuire, T. G. (in press). Some economics of performance-based contracting for substance abuse treatment services. Health Services Research.

Commons, M., McGuire, T. G., & Riodan, M. H. (in press). Performance contracting for substance abuse treatment. Health Services Research.

Daley, J., & Shwartz, M. (1994). Developing risk adjustment methods. In L. I. Iezzoni (Ed.), Risk adjustment for measuring health care outcomes. Ann Arbor, MI: Health Administration Press.

D’Aunno, T., Sutton, R. I., & Price, R. H. (1991). Isomorphism and external support in conflicting institutional environments: A study of drug abuse treatment. Academy of Management Journal, 34 (3), 636-661.

D’Aunno, T., Sutton, R. I., & Price, R. H. (1992). The responses of drug abuse treatment organizations to financial adversity: A partial test of the threat-rigidity hypothesis. Journal of Management, 18 (1), 117-131.

D’Aunno, T., & Vaughan, T. E. (1995). An organizational analysis of service patterns in outpatient drug abuse treatment units. Journal of Substance Abuse, 7, 27-42.

DiMonaco, J. S., & DiMonaco, D. (1989). The systems approach to managing mental health and chemical dependency treatment—synergism at work. Employee Benefits Journal, 14 (1), 12-16.

Docherty, J. P. (1990). Myths and mystifications of managed care. Journal of Mental Health Administration, 17 (2), 138-143.

Edelman, B., & Phelan, A. (1992). Health maintenance organizations and the treatment of substance abuse. Journal of Ambulatory Care Management, 15 (1), 56-67.

Edlund, M., Wheeler, J. R. C., & D’Aunno, T. A. (1992). Payment systems and payment incentives in outpatient abuse treatment. Public Budgeting and Financial Management, 4 (1), 107-123.

England, M. J., & Vaccaro, V. A. (1991). New systems to managed mental health care. Health Affairs, 10 (4), 129-137.

Enthoven, A. C., & Singer, S. J. (1996). Managed competition and California’s health care economy. Health Affairs, 15 (1) 39-57.

Essocks, S. M., & Goldman, H. H. (1995). States embrace managed mental health care. Health Affairs, 14 (3), 34-44.

Feldman, R., Wholey, D., & Christianson, J. (1996a). Effect of mergers on health maintenance organization premiums. Health Care Financing Review, 17 (3), 171-189.

Feldman, R., Wholey, D., & Christianson, J. (1996b). Economic and organizational determinants of HMO mergers and failures. Inquiry, 33 (summer), 118-132.

Foreman, S. E., Wilson, J. A., & Scheffler. (1996). Monopoly, monopsony, and contestability in health insurance: A study of Blue Cross plans. Economic Inquiry, 34, 662-677.

Frabotta, J. (1989). How to weigh drug options. Business & Health, 7 (2), 37-38.

Frank, R. G. (1985). Competitive effects of HMOs: A review of the evidence. Inquiry, 22 (2), 148-161.

Frank, R. G. (1992). Research on organization and financing of care for individuals with severe mental illness. Administration and Policy in Mental Health, 19 (2), 465-468.

Frank, R. G., & Dewa, C. (1992). Insurance system structure and the use of mental health services by children and adolescents. Clinical Psychology Review, 12, 829-840.

Frank, R. G., Goldman, H. H., & McGuire, T. G. (1992). A model mental health benefit. Health Affairs, 11 (3), 98-117.

Frank, R. G., Goldman, H. H., & McGuire, T. G. (1994). Who will pay for health reform? Consequences of redistribution of funding for mental health. Hospital and Community Psychiatry, 45 (9), 908-910.

Frank, R. G., Huskamp, H. A., McGuire, T. G., & Newhouse, J. P. (1996). Some economics of the mental health carve-outs. Archives of General Psychiatry, 53 (10), 933-937.

Frank, R. G., & McGuire, T. G. (1990). Mandating employer coverage for mental health services. Health Affairs, 9 (1), 31-42.

Frank, R. G., & McGuire, T. G. (1992). A new look at rising mental health insurance costs. Health Affairs, 10 (2), 116-124.

Frank, R. G., & McGuire, T. G. (1995). Estimating costs of mental health and substance abuse coverage. Health Affairs, 14 (3), 102-115.

Frank, R. G., & McGuire, T. G. (in press). Savings from a carve-out program for mental health and substance abuse in Massachusetts Medicaid. Psychiatric Services.

Frank, R. G., McGuire, T. G., Regier, D. A., Manderscheid, R., & Woodward, A. (1994). Paying for mental health and substance abuse care. Health Affairs, 13 (1), 337-342.

Frank, R. G., McGuire, T. G., & Salkever, D. S. (1991). Benefit flexibility, cost shifting and mandated mental health coverage. Journal of Mental Health Administration, 18 (3), 264-271.

Frank, R. G., Salkever, D. S., & Sharfstein, S. S. (1991). A new look at rising mental health insurance costs. Health Affairs, 10 (2), 116-124.

Freeman, M. A., & Trabin, T. (1994). Managed behavioral healthcare: History, models, key issues, and future course. Rockville, MD: Substance Abuse and Mental Health Services Administration.

French, M. T., Dunlop, L. J., Galinis, D. N., Rachal, J. V., & Zarkin, G. A. (1996). Health care reforms and managed care for substance abuse services: Findings from 11 case studies. Journal of Public Health Policy, 17 (2), 181-203.

French, M. T., & Fairbank, J. A. (1992). Patterns of drug use, criminal activity, and employment status among patients in federally-funded drug abuse programs. Substance Abuse, 13 (1), 37-47.

Frontin, P., & Snider, S. C. (1996). An examination of the decline in employment-based health insurance between 1988 and 1993. Inquiry, 33 (4), 317-325.

Fuller, M. G. (1994). A new day: Strategies for managing psychiatric and substance abuse benefits. Health Care Management Review, 19 (4), 20-24.

Garnick, D. W., Hendricks, A. M., Dulski, J. D., Thorpe, K. E., & Horgan, C. (1994). Characteristics of private-sector managed care for mental health and substance abuse treatment. Hospital and Community Psychiatry, 45 (12), 1201-1205.

Given, R. S. (1996). Economies of scale and scope as an explanation of merger and output diversification activities in the health maintenance organization industry. Journal of Health Economics, 15, 684-713.

Glazner, J., Braithwaite, W. R., Hull, S., & Lezotte, D. C. (1995). The questionable value of medical screening in the small-group health insurance market. Health Affairs, 14 (2), 225-233.

Gold, M., & Hurley, R. (1997). The role of managed care "products" in managed care "plans." Inquiry, 34, 29-37.

Gold, M., Hurley, R., Lake, T., Esnor, T., & Berenson, R. (1995). Arrangements between managed care plans and physicians: Results from a 1994 survey of managed care plans. Washington, DC: Mathematica Policy Research.

Goldfield, N. (1991). Introduction: Measurement and management of quality in managed care organizations: Alive and improving. Quality Review Bulletin, 17 (11), 343-348.

Government Accounting Office. (1993). Medicaid: States turn to managed care to improve access and control costs. Washington, DC: Author.

Hartwell, T. D., Steele, P., French, M. T., Potte, F. J., Rodman, N. F., & Zarkin, G. A. (1996). Aiding troubled employees: The prevalence, costs and characteristics of employee assistance programs in the United States. American Journal of Public Health, 86 (6), 804-808.

Harwood, H. (1992). A review: Substance abuse policy, economics, and services provision. Current Opinion in Psychiatry, 5, 441-445.

Harwood, H. J., Thonson, M., & Nesmith, T. (1994). Healthcare reform and substance abuse treatment: Costs of financing under alternative approaches, Final Report [Report, program plan]. Fairfax, VA: Lewin-VHI, Inc.

Hasan, M. K., & Shellhorse, M. (1996). Mental health reform: A practical approach. West Virginia Medical Journal, 92 (1), 14-17.

Hellinger, F. J. (1995). Selection bias in HMOs and PPOs: A review of the evidence. Inquiry, 32, 135-142.

Horgan, C. (1995). Financing and reimbursement for provision of alcohol services. NIAAA Advisory Council, Subcommittee on Health Services Research, Panel on Financing and Organization.

Iezzoni, L. I. (Ed.). (1994). Risk adjustment for measuring health care outcomes. Ann Arbor, MI: Health Administration Press.

Jensen, G. A., Cotter, K. D., & Morrisey, M. A. (1995). State insurance regulation and employers’ decisions to self-insure. Journal of Risk and Insurance, 62 (2), 185-213.

Jensen, G. A., & Morrisey, M. A. (1991). Employer-sponsored insurance coverage for alcohol and drug abuse treatment, 1988. Inquiry, 28, 393-402.

Kronson, M. E. (1991, April). Substance abuse coverage provided by employer medical plans. Monthly Labor Review, 4-10.

Kunnes, R., Niven, R., Gustafson, T., Brooks, N., Levin, S. M., Edmunds, M., Trumble, J. G., & Coye, M. J. (1993). Financing and payment reform for primary health care and substance abuse treatment. Journal of Addictive Diseases, 12 (2), 23-42.

Langenbucher, J. W. (1996). Socioeconomic analysis of addictions treatment. Public Health Reports, 111 (2), 135-137.

Larson, M. J., Bowden, M. L., & Horgan, C. (1991). Private insurance for substance abuse services: National overview. Addiction and Recovery, 11, 21-26.

Lewis, B. A., & Phelan, A. (1992). Health maintenance organizations and the treatment of substance abuse. Journal of Ambulatory Care Management, 15 (1), 56-67.

Lo, A., & Woodward, A. (1993). An evaluation of freestanding alcoholism treatment for Medicare recipients. Addiction, 88 (10), 53-68.

Luft, H. S. (1996). Modifying managed competition to address cost and quality. Health Affairs, 15 (1), 23-38.

Lynch, F. (1992). Financing family-centered services for adolescents. In W. Snyder & T. Ooms (Eds.), Empowering families, helping adolescents: Family-centered treatment of adolescents with alcohol, drug abuse, and mental health problems [Technical Assistance Publication Series No. 6] (pp. 135-144). Rockville, MD: Office of Treatment Improvement.

McCrady, B. S., & Langenbucher, J. W. (1996). Alcohol treatment and health care system reform. Archives of General Psychiatry, 53 (8), 737-746.

McGuire, T. G. (1994). Predicting the costs of mental health benefits. Milbank Quarterly, 22 (1), 3-23.

Mechanic, D., Schlesinger, M., & McAlpine, D. D. (1995). Management of mental health and substance abuse services: State of the art and early results. Milbank Quarterly, 73 (1), 19-55.

Mental Health Parity Act of 1996, S. 2301 (Incorporated into H.R. 3666 [P.L. 104-204]).

Miller, N. A. (1992). An evaluation of substance misuse treatment by an employee assistance program. International Journal of the Addictions, 27 (5), 553-559.

Miller, R. H. (1996a). Competition in the health system: Good news and bad news. Health Affairs, 15 (2), 107-120.

Miller, R. H. (1996b). Health system integration: A means to an end. Health Affairs, 15 (2), 92-106.

Mirin, S. M., & Seder, L. I. (1994). Mental health care: Current realities, future directions. Psychiatric Quarterly, 65 (93), 161-175.

Morrisey, M. A., Alexander, J., Burns, L. R., & Johnson, V. (1996). Managed care and physician/hospital integration. Health Affairs, 15 (3), 62-73.

National Institute on Alcohol Abuse and Alcoholism. (1996). Final report : Panel on financing and organization (June 4). National Advisory Council on Alcohol Abuse and Alcoholism. Rockville, MD: Author.

National Institute on Drug Abuse. (1992a). Drug abuse services research series: Background papers on drug abuse financing and service research. Rockville, MD: Author.

National Institute on Drug Abuse. (1992b). Drug abuse services research series: Extent and adequacy of insurance coverage for substance abuse services. Rockville, MD: Author.

Oss, M. (1994). Managed behavioral health market share in the United States, 1993: An open minds publication. Gettysburg, PA: Behavioral Health Industry News, Inc.

Pauly, M. V. (1991). Financing treatment for substance abuse. In W. S. Cartwright & J. M. Kaple (Eds.), Economic costs, cost-effectiveness, financing and community-based drug treatment [NIDA Monograph 113] (pp.165-174). Rockville, MD: National Institute on Drug Abuse.

Public policy statement on addiction medicine and health insurance reform. (1993). Journal of Addictive Diseases, 12 (2), 161-163.

Rawson, R. A., Obert, J. L., McCann, M. J., Marinelli-Casey, O., & Suti, E. (1991). Outpatient chemical dependency treatment and managed care system: An unrealized symbiosis. Journal of Ambulatory Care, 14 (4), 48-59.

Renz, E. A., Chung, R., Fillman, T. O., Mee-Lee, D., & Sayama, M. (1995). The effect of managed care on the treatment outcome of substance abuse disorders. General Hospital Psychiatry, 17 (4), 287-292.

Robinson, J. C. (1996). The dynamics and limits of corporate growth in health care. Health Affairs, 15 (2), 156-169.

Robinson, J. C., & Casalino, L. P. (1996). Vertical integration and organizational networks in health care. Health Affairs, 15 (1), 8-22.

Rogowski, J. A. (1992). Insurance coverage for drug abuse. Health Affairs, 11 (3), 133-148.

Rogowski, J. A. (1994). Private versus public sector insurance coverage for drug abuse. Drug Policy Research Center, Rand.

Rouse, B. A. (Ed.). (1995) Substance abuse and mental health statistics sourcebook. (pp. 107-126). Rockville, MD: Substance Abuse and Mental Health Services Administration.

Scott, J. E., Greenberg, D., & Pizarro, J. (1992). A survey of state insurance mandates covering alcohol and other drug treatment. Journal of Mental Health Administration, 19 (1), 96-118.

Shaferstein, S. S., Stoline, A. M., & Goldman, H. H. (1993). Psychiatric care and health insurance reform. American Journal Psychiatry, 150 (1), 7-18.

Shaffer, E. R., Cutler, A. J., & Wellstone, P. D. (1994). Coverage of mental health and substance abuse services under a single-payer health care system. Hospital and Community Psychiatry, 45 (9), 916-919.

Shortell, S. E., Gilles, R. R., & Anderson, D. (1994). The new world of managed care: Creating organized delivery systems. Health Affairs, 13 (4), 46-64.

Smith, G. R., & Burns, B. J. (1993). Recommendations of the Little Rock working group on mental and substance abuse disorders in health-care reform. Journal of Mental Health Administration, 20 (3), 247-253.

Strosahl, K., & Quirk, M. (1994, July). The trouble with carve outs: Separate behavioral health plans drive up costs because most patients seek mental health care from primary doctors. Business and Health.

Thorpe, K. E. (1992). Expanding employment-based health insurance: Is small group reform the answer? Inquiry, 29, 128-136.

Tommasini, N. R. (1994). Private insurance coverage for the treatment of mental illness versus general medical care: A policy inequity. Archives of Psychiatric Nursing, 8 (1), 9-13.

U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health, National Institute on Alcohol Abuse and Alcoholism, National Advisory Council on Alcohol Abuse and Alcoholism, Subcommittee on Health Services Research. (1996). Final report panel on financing and organization, June 4.

Wallack, S. (1992). Managed care: Practice, pitfalls, and potential. Health Care Financing Review [1991 Annual Supplement], pp. 27-34.

Welch, W. P., Hillman, A., & Pauly, M. (1990). Toward new typology for HMOs. Milbank Quarterly, 68 (2), 221-243.

Wells, E. A., Fleming, C., Calsyn, D. A., Jackson, T. R., & Saxon, A. J. (1995). User of free treatment slots at community-based methadone clinic. Journal of Substance Abuse Treatment, 12 (1), 13-18.

Wholey, D., Feldman, R., & Christianson, J. B. (1995). The effect of market structure on HMO premiums. Journal of Health Economics, 14, 81-105.

Wholey, D., Feldman, R., Christianson, J. B., & Enberg, J. (1996). Scale and scope economies among health maintenance organizations. Journal of Health Economics, 15, 757-784.

Wickizer, T. M., & Feldstein, P. J. (1995). The impact of HMO competition on private health insurance premiums, 1985-1992. Inquiry, 32 (3), 241-251.

Wickizer, T. M., Lessler, D., & Travis, K. M. (1996). Controlling psychiatric utilization through managed care. American Journal of Psychiatry, 153 (3), 339-345.

Woodward, A. (1992). Managed care and case management of substance abuse treatment. In R. S. Ashery (Ed.), Progress and issues in case management [NIDA Monograph 127] (pp. 34-53). Rockville, MD: National Institute on Drug Abuse.

Young, N. K. (1996). Alcohol and other drug treatment: Policy choices in welfare reform. Washington, DC: National Association of State Alcohol and Drug Abuse.

Zarkin, G. G., Galinis, D. N., French, M. T., Fountain, D. L., Ingram, P. W., & Guyett, J. A. (1995). Financing strategies for drug abuse treatment programs. Journal of Substance Abuse Treatment, 12 (6), 385-399.

Zwanziger, J., & Melnick, G. A. (1996). Can managed care plans control health care costs? Health Affairs, 15 (2), 185-199.

Sections