Organization & Financing

SAMHSA Managed Care Initiative Training Sessions

October 1997

Dates: October 20, 1997-Parklawn Conference Center, Rockville, Maryland
October 21, 1997-American Psychiatric Association, Washington, D.C.

Title: Treating Co-Occurring Disorders: Is Managed Behavioral Health Care the Solution or Another Version of the Problem?

Speaker: M.. Susan Ridgely, J.D., M.S.W.
Associate Professor
Florida Mental Health Institute
University of South Florida
Tampa, Florida

Summary

Ms. Ridgely began the meeting by providing an overview of the prevalence of co-occurring disorders, pointing out that these should not be associated only with "special populations." The mental health field tends to think in this regard and has devised specific interventions, such as creating a 10-bed unit or 10-slot program, to meet the problem. In reality, for mental health (MH)/substance abuse (SA) practitioners, co-occurring disorders are the norm rather than the exception. This is borne out by the literature which suggests that, in terms of treated prevalence (i.e., those in mental health treatment), from 25 to 80 or 90 percent also have a co-occurring substance abuse or substance dependence disorder, or are using these substances to their detriment. (The generally acceptable figure is 50 percent- depending upon how the diagnosis was defined, what instrument used, etc.) Although there is less agreement in the substance abuse literature, most would agree that at least 25 percent, and probably closer to 50 percent, of all those receiving substance abuse treatment also display psychiatric symptomatology that may or may not rise to the level of a diagnosis (given the same parameters as described for mental health). Dual-diagnosis does encompass a large crosshatch of patients in mental health and substance abuse services. In treatment, therefore, it is very important to consider the heterogeneity of the population involved. When considering treatment modalities, there are patients in mental health services, as well as patients in substance abuse services, where there is some utility in the concept of a primary diagnosis. But for those patients with co-occurring severe disorders, the assessment of which is the primary disorder is not clinically important-they are disabled by both disorders and need treatment for both disorders. Treatments must be integrated to ensure that clinical messages are clear and consistent.

Ms. Ridgely then turned to the major clinical, economic, and systems issues that need to be addressed in the area of co-occurring disorders. From the mental health perspective, the literature suggests that those with co-occurring disorders exhibit more severe psychiatric symptomatology; higher rates of relapse and rehospitalization; a worse course in treatment; a greater likelihood of missing therapy sessions and discontinuing treatment; more admissions to emergency rooms and noncompliance with treatment interventions; increased acting out; suicidal and criminal behavior; and a decreased ability to carry out daily activities. From the substance abuse perspective, the study of co-occurring disorders has changed the understanding of addictions and has helped to enhance addiction treatment in all areas. For example, it is now generally accepted that substance abuse disorders vary in type, severity, chronicity, and degree of impairment and not all people will respond to traditional treatment. Treating substance abuse problems may not solve other problems and ignoring other problems (e.g., the need for entitlements, housing, and healthcare) may interfere with treating the substance abuse. Research has suggested that the best single predictor of failure in substance abuse treatment is co-occurring psychiatric symptomatolgy. People may need multiple services over extended periods of time-a difficult point when dealing with managed care systems that are predominantly focused on acute, episodic treatment-even in the public sector.

To illustrate the economic impact of co-occurring disorders, Ms. Ridgely cited a study that reviewed Massachusetts Medicaid expenditures for psychiatric treatment under various conditions, and found that the annual cost of psychiatric treatment for the single mental health diagnosis of schizophrenia was $12,350. When co-occurring schizophrenia and substance abuse were present, however, the annual cost of psychiatric care alone almost doubled to $23,169. Such data emphasizes the economic as well as the clinical concerns.

Ms. Ridgely reviewed the preliminary results of a demonstration study conducted in California that identified persons in the public mental health system with co-occurring disorders who were not well-served in their current treatment. Because the research funds available only covered the evaluation and no new service money was allocated, the effort went into retooling existing community mental health programs to begin to address the substance abuse problems. This was accomplished by offering three different interventions: (1) linkage to traditional 12-step programs (e.g., Alcoholics Anonymous), (2) intensive case-management, and (3) behavioral skills training. The average public cost of the substance abuse subjects enrolled in the program was evaluated at different time periods. Before entry into one of the interventions, the combined costs to the system (which included the costs of health care, SA, and MH treatments, legal costs, and costs to the family in terms of support, travel, and time) totaled approximately $12,000 for six months. After 18 months in the program, the average total public costs of these individuals were reduced by 43 percent. This was mainly accomplished through a reduction in psychiatric acute and subacute care and an increase in outpatient psychiatric supportive services.

Ms. Ridgely summarized the "lessons" of co-occurring disorders. They are: (1) highly prevalent, (2) very expensive if left untreated, (3) extremely treatable, and (4) locally complicated by "system disjunctures." These "disjunctures," she stated, are amenable to mental health and substance abuse interventions in managed care settings.

Ms. Ridgely then discussed some of the system disjunctures found in the "public sector" (i.e., public sector services and Medicaid). These include the lack of a common administrative structure, categorical funding for services, and the history of distrust and philosophical conflicts between agencies. Government funding for substance abuse services tends to target funds for sub-populations such as pregnant women, chronic public inebriants, and those with HIV/AIDS. Channeling funds for these priority groups can make it extremely difficult for local community programs to serve a broader population in need. The result of categorical financing is that local communities have found it necessary to take "bits and pieces" of categorical money targeted for individuals or a specific type of treatment and try to put together a treatment package relevant to what an individual may need.

Ms. Ridgely briefly reviewed the problems of Medicaid, prior to managed care. Medicaid is an open-ended entitlement program providing health services for individuals in poverty and with disabilities-a fact which resulted in unsustainable growth. For example, by 1993, state spending in Florida for Medicaid had surpassed spending for higher education. By 1996, Medicaid expenditures accounted for more than 20 percent of state budgets and, in Florida, had grown at the rate of 28 percent per year since 1988, consuming 50 percent of all general revenue growth in the state budget.

By 1981 (after federal authorization legislation, OBRA), states began to experiment with managed care interventions as a way of dealing with spiraling Medicaid expenditures. Most of the experience in this area has been with Aid to Families with Dependent Children (AFDC) populations. Although Medicaid included the disabled and aged population, these groups were virtually untouched by managed care until the 1990s.

In defining "managed care," Ms. Ridgely reminded the group that the term includes a wide variety of organizational forms, financial strategies, and management practices. For this reason it is important that, as research findings in managed care are disseminated, the exact meaning of these terms is understood. In general, characteristics of managed care arrangements are that they influence treatment decisions through (1) budget constraints and assignment of financial risk to providers; (2) financial incentives for providers; or (3) review of treatment plans against criteria which define appropriate care.

By 1995, enrollment in Health Maintenance Organizations (HMOs) and Preferred Provider Organizations (PPOs) accounted for an estimated 73 percent of all employed insured population of the U.S. By 1995, approximately 111 million people across the U.S. (both public and private insurance) had at least some of their mental health/substance abuse benefit managed by a speciality managed care behavioral health care firm. Today, approximately 21.8 million people are enrolled in "risk-based" carve out contracts.

Ms. Ridgely reviewed the terminology "carve-out," and "carve-in,"observing that these terms fail to capture the many complexities of managed care. She noted that there are arguments in favor of each approach. The argument for "carve-in" mental health and substance abuse benefits (as generally exemplified by an HMO concept) includes the integration of health and mental health care, effectively offering "one stop shopping" for consumers. The arguments for "carve-out" benefits-meaning that Medicaid monies expended for alcohol, drug, and mental health (ADM) care are carved-out, and managed separately by an organization that specializes in behavioral health care-include (1) the assurance that adequate resources will be spent for ADM care, (2) ADM criteria will drive the selection of a managed care organization; and (3) there will be the maintenance of public agencies/"safety net" providers

Arguments against each approach also exist. For example, HMOs under the "carve-in" approach have a poor record of dealing with severe and chronic mental health services. They have a history of neglecting non-medical, yet essential, services, wrap-around services, and they also offer potential for cross-subsidization of health care with ADM dollars. Concerns about "carve-out" benefits include the isolation of ADM from general health care and the lack of data on the relative cost effectiveness of different approaches with different populations (e.g., non-commercially insured markets and disabled persons). Using an example in Florida, Ms. Ridgely illustrated that the term "carve-in" versus "carve-out" alone doesn't necessarily provide any information. Citing two HMOs that provide mental health and substance abuse services to the Medicaid population, she followed the structure of services, demonstrating that the HMOs (which may imply "carve-in") have a capitation arrangement with a "carve-out" organization (for outpatient care), which, in turn, has a capitated "carve-out" arrangement with another organization ( for inpatient care). All inpatient providers are paid on a fee-for-service basis; similarly, outpatient providers-except one-also are paid on a fee-for-service basis. This, she stated, is not a "carve-in" approach to care, nor is it integrated. The lesson to be learned is that it is essential to look at the entire organizational structure and not make assumptions based on broad categorizations.

The concept of carve-out in the public sector (Medicaid) is associated with mental health and substance abuse, as well as pharmacy, vision, and a few other speciality health care areas. As of 1996, at least 32 states had approved or pending 1115 and 1915(b) waivers for managed behavioral healthcare. Twenty-one states had "carve outs" that would include both mental health and substance abuse; 10 states have "carve-outs" that include only mental health; and one state has a "carve-out" that covers only substance abuse. Clearly, the need to focus on these areas has encouraged many advocates to endorse the concept of the "carve-out." It has been suggested that there may be no differences in terms of integration of services if mental health and substance abuse are carved out together.

Ms. Ridgely suggested that managed care holds promise in addressing barriers to care for people with co-occurring disorders in the public sector. She suggested that managed care "carve-outs" allow for a unified administrative structure and make use of private sector management technologies (e.g., databases). Most significantly, their capitation arrangement allows for flexibility and promotes the development of comprehensive continuum of services. Putting managed care providers at risk can theoretically improve efficiency, competition, choice, and access.

The critical issue currently being raised in the legislatures, the literature, and by advocates across the country, however, is the risk when individuals with complex, clinical needs are enrolled in a managed care program designed for those with less complex needs. This concern has arisen with the realization that much of managed care organizations' experiences in mental health and substance abuse services are limited to commercial or employed populations. Ms. Ridgely discussed the risks of managed care for vulnerable (co-occurring disorders) populations, e.g.:

Persons with co-occurring disorders have ongoing, long-term needs. Most clinical protocols of managed care companies are oriented to acute, episodic care. Data, such as that supplied by the California demonstration project previously mentioned, demonstrates that a long-term investment in this population will save money in the aggregate, but it is not easy to "sell" to managed care companies.
Some needed services fall beyond the boundaries of "medical necessity" (e.g., supportive services) or are outside the "amount, scope, and duration" benefit limitations of managed care plans.
In most communities, mental health consumers are not sufficiently sophisticated to influence the process whereby purchasers negotiate for plan coverage, or grievance and appeals procedures. Most plans "lock in" the subscriber for months, making it impossible for consumers to "vote with their feet," or to move to another managed care plan.
The focus of managed care on efficiency can easily become "underserving," rather than providing less expensive alternatives.

The lesson which emerges is that the list of "benefits" provided by a managed care plan may not be meaningful in itself, since most benefit plans include a plethora of services. The only real information such a list provides is the services that are excluded-what is NOT covered. Only through careful analysis of the medical necessity criteria and utilization review process will the actual benefits be described.

Despite these risks, Ms. Ridgely emphasized her belief that managed care may represent an opportunity to solve some of the problems associated with treating dually-diagnosed populations, with which the public sector has grappled unsuccessfully for years. Managed care can accomplish this through: (1) aggregation of funds for developing an expanded range of ADM and related supportive services; (2) flexibility in capitation to tailor individual service packages to the need of the individual, rather than arbitrary limits of benefits, and (3) the substitution of less costly for more costly services ("downward substitution") thus freeing monies for the development of new and innovative services.

Ms. Ridgely then provided a comparison of how co-occurring disorders were handled in managed care contracts in two states. She noted that, despite Florida's carve-out of only mental health programs and Massachusetts' carve out of both mental health and substance abuse, with some slight differences in financing (managed care organization payment to providers is fee-for-service in MA and mixed in FL), according to researchers in both states, there were no real differences in organization of care and no integration of services in either state. "Carving-out" mental health and substance abuse together does not necessarily solve the problem of co-occurring disorders.

This, Ms. Ridgely stated, points out the need for significant attention, at the local, state, and federal level, to the inadequate and ineffective treatment services for co-occurring disorders. Although there is a substantial body of knowledge on the effective treatment of co-occurring disorders, this information is not routinely utilized by clinicians in treatment agencies, nor is it receiving attention from managed care organizations. Managed care companies still are focused on acute models of care. Information about effective mental health and substance abuse treatment needs to be translated into examples which managed care companies can use.

Ms. Ridgely emphasized that she believes that managed care is the opportunity that the field has been looking for and encouraged the audience to think about ways to make this opportunity work. There already has been discussion about pooling funding, parity of benefits, and the recognition that inpatient and residential treatments may not be the most cost effective treatments. HMOs can solve many of the problems of co-occurring disorders, if the incentives are there and the contract is written correctly. To illustrate, she listed the key aspects of managed care design and the ways to address co-occurring disorders in each area:

Enrollment: The issue is adverse selection (i.e., those with a high risk gravitating toward one plan). Systems must be in place to track the migration of high risk populations and to ensure the distribution of the patient load across plans (where there are several plans). This should be accomplished by supplementing or adjusting the rates of the most favored plans-rather than by restricting consumer choice. Vendors must be prohibited from "cherry picking" the members they want to enroll.
Benefit Design: Most benefits need to be settled at the Request for Proposals (RFP) design stage, before the contract is awarded. Examples include ensuring that medical necessity criteria and utilization review procedures are appropriate to chronic relapsing disorders and that performance indicator criteria are the basis for treatment-not the financial "bottom line." It is essential to ensure that appropriate supportive services are included in the benefit plan. One idea is to require vendors to screen for and identify co-occurring disorders through the implementation of specific clinical protocols and have protocols for referring clients to public sector agenices, if the managed care benefit package does not provide the appropriate service.
Service Network Development: There is a question as to whether MCOs should be required to include public agencies/safety net providers as part of the managed care service network. If the benefit package is correctly designed (i.e., making certain types of services mandatory) the result will be that the people/agencies who have the expertise to provide such services are brought into the network.
Payment and Risk Arrangements: More research is needed on capitation rates. It is essential to have realistic rates that will cover the types of services required to treat co-occurring disorders. The major challenge is risk-adjustment of capitation rates. Diagnosis is a poor predictor of utilization rates, as is the designation of SSI versus AFDC. Performance standards of managed care organizations need to recognize issues concerning co-occurring disorders. The quality standards of Health Plan Employer Data and Information Set (HEDIS) as performance standards address only single disorders-they do not address co-occurring disorders. There are ways of establishing screening standards and using co-occurring disorders as a "tracer" condition to evaluate how the program is operating; if a program is meeting the needs of those with co-occurring disorders, it can most likely meet the needs of those with a single disorder.
Relationship with Other Health and Human Services Sectors: Managed care does not exist in a vacuum; it has become virtually impossible to determine the boundaries of the managed care system. The Federal government and the states contribute significant amounts of public money to managed care. It is a matter of public policy to monitor whether MCOs push costs into juvenile justice, criminal justice, or state mental health facilities and thus avoid paying for high-risk patients.

In summary, Ms. Ridgely acknowledged that there are multiple ways of dealing with any of the problems in the area of co-occurring disorders. The challenge, however, is to deal with the problems. Whatever the method, there is a need to focus on the issue of co-occurring disorders to ensure that managed care does not become just another version of the problem.

References

-Dickey B and Azeni H. (1996). Persons with dual diagnoses of substance abuse and major mental illness: Their excess costs of psychiatric care. American Journal of Public Health. 86:973-977.

-Treatment Interventions By Phase of Recovery and Type of Dual Diagnosis, In: Minkoff K and Rossi, A, (authors) Co-Occurring Mental Health and Substance Disorders Panel of the Center for Mental Health Services Managed Care Initiative: Clinical Standards and Workforce Competencies Project. May,1996.

-Jerrell, J.M. and Ridgely, M.S. (1995). Comparative effectiveness of three approaches to serving people with severe mental illness and substance abuse disorders. Journal of Nervous and Mental Disease, 183, 566-576.

-Jerrell, J.M., Hu, T-W., and Ridgely, M.S. (1994). Cost effectiveness of substance disorder interventions for the severely mentally ill. Journal of Mental Health Administration, 21, 281-295.