Over the past two decades, it has been increasingly recognized that individuals with co-occurring psychiatric and substance use disorders constitute a difficult and diverse clinical population with poorer outcomes in multiple dimensions and higher costs in multiple settings. During that time, research data have accumulated that identify evidence-based model programs for treating particular subpopulations of individuals with dual diagnoses (1). These programs include intensive integrated case management team models, such as the continuous treatment team model for individuals with severe and persistent mental illness and substance use disorders who are difficult to engage in traditional services (2), and many other types of model programs.
However, despite the value of evidence-based demonstration models, recent research has indicated some limitations in their potential applicability. First, accumulating epidemiologic data from the 1980s and 1990s (3,4) indicate that comorbidity is so common that dual diagnosis should be expected rather than considered an exception. Consequently, the application of best practices cannot be restricted to small subpopulations but rather must be extended to the development of models that apply to the entire system of care and that require integrated system planning involving both mental health and substance abuse treatment agencies.
Second, recent research on the implementation of clinical demonstration projects indicates that better outcomes are related to system-level changes that establish formal structures for interprogram collaboration at the administrative level that, in turn, support the work of the frontline clinicians at the program level. Thus the success of a particular demonstration model depends on the extent to which this support at the system level is provided (5).
Third, managed care initiatives require that service planning, quality monitoring, and outcome evaluation for all populations be implemented at the system level, whether by contracting with outside managed care organizations or through internal implementation of managed care strategies.
Thus a focus on best practices at the program level is being replaced by a greater interest in the need for best practices at the system level. This paper describes a national consensus report that builds on program-level data to propose national standards for best practices for treatment of patients with dual diagnoses at the system level. These consensus standards can, in turn, guide future research on best practices for implementation of comprehensive, integrated systems of care for individuals with co-occurring disorders.
In recognition of the importance of system-level best practices in the design of managed care structures, the Substance Abuse and Mental Health Services Administration funded the Managed Care Initiative in 1995 to develop standards of care for the treatment of various populations in managed care systems. The Managed Care Initiative, coordinated through the University of Pennsylvania Center for Mental Health Policy and Services Research (CMHPSR), attempted to accomplish its goal by funding national expert consensus panels to focus on standards of care for specific populations. I was appointed chair of the panel on co-occurring disorders in 1996, and, with the assistance of CMHPSR, developed an expert consensus panel that was multidisciplinary, was geographically and culturally diverse, included consumers and family members, and incorporated providers from both the addiction and mental health treatment systems.
The panel's first task was to conduct a comprehensive literature review and develop an annotated bibliography (6) that included both published and unpublished material related to the successful treatment of individuals with co-occurring disorders in managed care systems and in all other systems. It became clear not only that there were no standards of care for treating co-occurring disorders in managed care systems but also that there were no such standards in any system. Developing these standards became the focus of the panel's work.
In January 1998 the panel issued a consensus report, titled Co-occurring Disorders in Managed Care Systems: Standards of Care, Practice Guidelines, Workforce Competencies, and Training Curricula, based on the material in the bibliography (7). Despite the apparent philosophical incompatibility between the mental health field and the substance abuse treatment field, consensus was reached relatively easily. The report relies on an integrated conceptual framework that makes sense from both a mental health and an addiction treatment perspective. This framework permits the development of practice guidelines for individualized clinical matching and for identifying program elements of a comprehensive system of care in a way that values and validates the contribution of mental health and addiction systems, programs, clinical interventions, and clinicians.
The report is organized into five interconnected sections. The first section, on consumer and family standards, is based on the principle that best practices should first be defined in relation to the needs of consumers. The definition should answer the question, "If I were an individual with a co-occurring disorder, how would I want the system of care to be organized to best meet my needs?" The section identifies five basic standards: the system must be welcoming, accessible, integrated, continuous, and comprehensive. The rest of the report addresses the designing of systems, programs, and interventions to meet these standards.
The second section of the report emphasizes standards for system design. Such standards are related both to the needs of a purchaser of services who is negotiating with a managed care organization about quality measures for a contract, and to the needs of a public entity intending to organize and manage the system itself. The section defines the importance of creating a structure to oversee the development of the system and that is empowered to promote integrated system planning and design. It recommends adoption of a consensus mission statement incorporating a coherent set of principles on which system design will be based, embodying an integrated philosophy that is acceptable to both mental health care and substance use treatment providers (8). These principles are as follows:
• Comorbidity should be expected, not considered an exception. Consequently, the whole system must be designed to be welcoming and accessible to patients with all types of dual diagnoses.
• Psychiatric and substance use disorders should be regarded as primary disorders when they coexist, each requiring specific and appropriately intensive assessment, diagnosis, and treatment, in accordance with established practice guidelines.
• Serious psychiatric and substance use disorders are chronic, relapsing illnesses that can be conceptualized by using a disease and recovery model, with parallel phases of treatment or recovery.
• Within each subtype of the treatment population, consumers are in different phases of treatment and at different stages of change with regard to their illness. Thus a comprehensive array of interventions that are phase and stage specific is required.
• Whenever possible, treatment of persons with complex comorbid disorders should be provided by individuals, teams, or programs with expertise in mental health and substance use disorders.
• The system should promote a longitudinal perspective on the treatment of patients with dual diagnoses, emphasizing the value of continuous relationships with integrated treatment providers, independent of participation in specific programs.
• Admission criteria should not be designed to prevent consumers from receiving services but rather to promote acceptance of consumers at all levels of motivation and readiness and with any combination of comorbid disorders.
• The service system should not begin or end at the boundaries of formal treatment programs; rather, it should include interventions to engage the most detached individuals—for example, those who are homeless.
• The fiscal and administrative operation of the system should support the accomplishment of the system's mission and the implementation of these principles.
In addition, the second section of the report recommends that each system identify quality and outcome measures—structure, process, and outcome measures; program standards and competencies; assessment tools—including those for level-of-care assessment and utilization management; practice guidelines; workforce competencies; and training materials in order to implement the system's mission and philosophy.
The report acknowledges the limitations of current methods for utilization management and level-of-care assessment for individuals with co-occurring disorders but refers to two existing documents that address this issue (9,10). The criteria of the American Society of Addiction Medicine are being replaced to more extensively address the assessment and placement of individuals with co-occurring disorders in treatment for substance use disorders (11).
In addition, the expected high prevalence of co-occurring disorders in almost all mental health and addiction treatment settings indicates the need for all such programs to demonstrate competency in dealing with this issue. Consequently, in the program standards section, each program type is listed—for example, crisis service, inpatient, and detoxification—with a description of the criteria for both standard program competency and specialized program competency.
The third section of the report describes practice guidelines for assessment, treatment and rehabilitation, and psychopharmacology for clinicians to follow in order to meet the standards of the system and the needs of consumers and their families. Examples of practice guidelines are as follows:
• Assessment for either disorder should begin as early as possible, without the imposition of arbitrary waiting periods of sobriety and without a requirement of psychiatric stabilization, on the basis of data collection for an integrated, longitudinal history.
• For each disorder, assessment should include a definition of the stage of change or level of motivation.
• When mental illness and a substance use disorder coexist, each disorder should be considered as primary, and integrated dual primary treatment should be provided; the treatment for each disorder should be matched to the diagnosis and the stage of change.
• Medication for known serious mental illness should never be discontinued on the grounds that the patient is using substances.
• Benzodiazepines are not recommended in the ongoing treatment of patients with known substance dependence with or without a comorbid psychiatric disorder. If a prescriber believes that an exception is warranted, this belief should be considered an indication for peer review, expert consultation, or a second opinion.
The fourth section of the report defines competencies in the areas of attitudes, values, knowledge, and skills that clinicians should acquire in order to implement the practice guidelines. Specific competencies were added for psychopharmacology providers. Examples of competencies are attitudes—the belief that all consumers deserve to be treated with respect and dignity, even when noncompliant, decompensated, or intoxicated; values—the belief that addiction and mental illness should be viewed as no-fault, no-blame diseases; knowledge—familiarity with the integrated conceptual framework, parallel phases of treatment, and the application of the disease and recovery model to both; and skills—the ability to identify indications for psychopharmacological assessment of patients with substance use disorders or dual diagnoses.
The final section of the report identifies training curricula—models for designing competency-based curricula using the competencies in the report as well as six sample curricula that the panel recommended—for training clinicians to achieve the competencies needed for implementing the practice guidelines to meet the standards of the system and the needs of consumers and their families.
The purpose of the report was to go beyond general recommendations for expanded integrated treatment and increased availability and flexibility of funding, to develop a range of specific tools and materials that could be adapted by systems of any size to initiate a process of change, even without access to additional resources. Systems can use this material to establish consensus on a systemwide mission and set of principles; develop program standards, practice guidelines, and required competencies; and adapt curricula for a systemwide training plan. Since the release of the report in 1998, an increasing number of large systems have begun to use the report in just this manner.
For example, Massachusetts has used this material as part of a statewide continuous quality improvement-oriented consensus-building process for developing a comprehensive, continuous, integrated system of care for individuals with severe and persistent mental illness and co-occurring substance use disorders (12). Pennsylvania initiated a statewide dual diagnosis task force that recommended adoption of many of the program standards, practice guidelines, and clinician competencies into statewide regulations for treatment of adults and adolescents with dual diagnoses (12). New Mexico has initiated a statewide dual diagnosis services enhancement initiative that uses a regional managed care system for non-Medicaid state dollars. Arizona, like Massachusetts and Louisiana, obtained funding for a community action grant to develop consensus on the material; it then adapted the consensus to create statewide practice guidelines, including a statewide assessment protocol for identifying patients with severe and persistent mental illness among substance-abusing applicants.
It is hoped that as these systems begin to disseminate information about the success and limitations of their efforts to change, more data will become available on how to design best-practice models for system change. Other systems are encouraged to explore the applicability of the material in the report to their own systems' efforts to change and to document their change strategies as a means of contributing to the evolving literature on this important issue.
The report is available at the CMHPSR Web site, www.med. upenn.edu/cmhpsr—click on Publication, then Managed Care Consensus Reports. It can also be ordered from CMHPSR by telephone (215-662-2886) for $19.95.
Dr. Minkoff is medical director of Choate Health Management in Woburn, Massachusetts, and assistant clinical professor of psychiatry at Harvard Medical School in Boston. Send correspondence to him at 12 Jefferson Drive, Acton, Massachusetts 01720 (e-mail, firstname.lastname@example.org). William W. Glazer, M.D., is editor of this column.