State Mental Health Policy: States' Early Experience in Improving Systems-Level Care for Persons With Co-occurring Disorders

State mental health and substance abuse authorities historically have functioned in silo-like isolation, with separate administrations and funding streams. A considerable body of research has found a high coincidence of mental and substance use disorders, suggesting that these authorities concomitantly treat many of the same individuals ( 1 , 2 ). Research also has shown that individuals with co-occurring mental and addictive disorders have special treatment needs and that treating both disorders in a coordinated fashion contributes strongly to satisfactory outcomes ( 3 , 4 ).

There are two ways in which this linkage can happen: services integration, where mental health and substance abuse services are integrated, and systems integration, where mental health and substance abuse treatment systems are merged to form behavioral health systems. Both approaches have led to improved outcomes among individuals with co-occurring disorders. In practice, however, many of these individuals receive limited treatment or inappropriate treatment, partly because of the fragmentation of services between and across mental health and substance abuse authorities ( 5 , 6 ). In light of this research, state mental health and substance abuse authorities have sought to develop strategies to better link and coordinate their services.

COSIG initiative

In 2003 the Substance Abuse and Mental Health Services Administration (SAMHSA) created the Co-occurring State Incentive Grant (COSIG) initiative, in recognition of the urgent need to improve the service linkages across mental health and substance abuse treatment authorities, as outlined in an influential 2002 report to Congress ( 7 ). This initiative was designed to stimulate states to develop innovative approaches that would increase their capacities to provide accessible, effective, comprehensive, coordinated, and evidence-based services to persons with co-occurring disorders. SAMHSA limited the eligibility for COSIG grants to the office of the governor because that office had the greatest potential to provide the multiagency leadership necessary to fulfill the purpose of the program.

In September 2003 SAMHSA awarded grants to seven states (from 37 applicants) to participate in the first cohort of this initiative: Alaska, Arkansas, Hawaii, Louisiana, Missouri, Pennsylvania, and Texas. New cohorts of states were added from 2004 to 2006, bringing the total number of COSIG states to 17. States participating in COSIG receive five years of funding (approximately $1 million annually for the first three years and declining amounts for years 4 and 5).

RAND implementation evaluation

In early 2003, SAMHSA contracted with the RAND Corporation to develop and conduct an implementation evaluation of the first cohort of seven states participating in COSIG during the start-up phase of their projects, defined as the first two years of funding. The goal of this evaluation was to document the implementation processes of each state and emphasize the lessons that state administrators learned in launching their projects.

After an initial group meeting with representatives from all seven states, key informant interviews were conducted with project leadership every six months for the first two years of the project. Interviews were 1.5 hours long and involved two key informants from each of the seven states during each interview cycle (14 total key informants per cycle). The key informants were selected on the basis of their centrality in the project's implementation and for their strong working knowledge of all aspects of the project implementation. All states had their principal investigators or project directors serve as key informants, and some included other research or clinical staff as key informants. The interview questionnaire was semistructured and focused on three core domains: progress toward project goals, implementation barriers and challenges, and lessons learned.

States varied dramatically in their ability to achieve the project goals they outlined for the first two years of their project. Goals ranged from recruiting specified numbers of community partners to conducting kickoff meetings and other activities by specified dates. By the end of the second year of the projects, two states reported only a few delays in reaching their interim project goals (for example, being less than three months behind schedule or spending more than 75% of project funds by the end of year 2), three states reported moderate delays (being three to six months behind schedule or spending 25%–75% of the project funds available by the end of year 2), and two states reported significant delays (they were six or more months behind schedule or had spent less than 25% of project funds available by the end of year 2). Reasons for project delays fell into one of three broad categories: planning challenges, bureaucratic challenges, and consensus challenges.

Planning challenges

All COSIG states developed detailed work plans that included timelines for tasks such as hiring staff, recruiting participants, creating committees, conducting meetings with key stakeholders, and developing deliverables and products. An examination of these plans as well as discussions with project leadership revealed that some states based their timelines on knowledge or experiences from similar or related projects, whereas other states set goals and timelines based on less direct information. States that based their plans on previous experience consistently missed fewer milestones and goals than those that had less direct plans.

For example, one year after project implementation, several states that had based their timelines on indirect estimates were required to completely revise their project plans. One project leader expressed disappointment on realizing that the work plan, which was developed with "best guess" timeline estimates, was "wildly ambitious" and that only a fraction of what was planned would be accomplished.

Bureaucratic challenges

States that experienced fewer bureaucratic challenges within the first two months of receiving their COSIG funds reported achieving more interim project goals and milestones during the first two years of their projects than states experiencing large numbers of these types of organizational and process challenges. Examples of these challenges included cumbersome paperwork to purchase supplies, inefficient processes for hiring staff, and challenges with administrative and financial departments.

For example, in one state, all project staff had to go through a lengthy approval process by the governor's office before they could be hired. In another state, project leadership reported that they were unable to pay project staff promptly because of a complicated reimbursement process. Yet another state reported an extensive process for getting approval for purchases for critical project tools like laptop computers. In nearly all instances these types of barriers could have been anticipated and planned for before project implementation.

Consensus challenges

All COSIG projects involved the creation of committees and teams to help in project efforts. Some projects, however, became bogged down in developing multiple committees or other complex structures, and the resources required to manage and maintain these committees impeded the project's progress. In addition, gaining consensus across these committees was extremely challenging and had the unforeseen effect of halting progress.

For example, one state reported that after working on the project for one year the most important accomplishment was the development of committees. Establishing, running, and gaining consensus across these committees came at the sacrifice of nearly every other first-year project goal, including important goals for screening and assessing patients with co-occurring disorders. Better planning at the onset of the project regarding the appropriate number, composition, and goals of these committees could have limited such problems.

Successes

The most successful states were those that planned carefully on the basis of experience, anticipated and planned around bureaucratic barriers, and gained early consensus from a limited number of key stakeholders. We observed that states with prior experience with similar projects were better able to achieve their project goals. We believe this finding may be due to two factors: pragmatic experience of successful states enabled them to determine more realistically what they could accomplish with their project, and the fact that these states had better resources and infrastructure in place at the project's outset helped them to accomplish their project goals.

For example, one state developed its COSIG project from a series of pilot projects that were already ongoing in the state. This state was able to leverage the knowledge gained and the infrastructure developed from these early pilot projects to successfully start up its COSIG project. Another state based its COSIG project on a treatment model for co-occurring disorders that was developed in one region of the state before the COSIG project began. The COSIG project involved the broader dissemination of this model across the state.

Summary

States across the country are working to improve the linkages and coordination of services between mental health and substance abuse treatment systems. Before embarking on these projects, states would benefit by first studying the successes and failures of other states already working on these issues and related issues. Because SAMHSA specifically opened the COSIG initiative to states at any level of infrastructure development, relatively less advanced states can productively draw on the experiences of more advanced states in the design and implementation of infrastructure development projects. This can help these states to develop realistic planning assumptions and to better anticipate unforeseen barriers.

Since the conclusion of this process evaluation, SAMHSA has supported two activities that foster planning and collaboration among COSIG grantees. SAMHSA sponsored three national policy academies on co-occurring substance abuse and mental disorders, which have been attended by all but one COSIG state, to engage state leadership and local representatives in the development of strategic plans for addressing co-occurring disorders. The national Co-occurring Center on Excellence provided consultation and technical assistance to help states implement their strategic plans. Through conference calls and grantee meetings, this center has fostered the development of an ongoing peer-to-peer learning community that allows states to consult with COSIG colleagues in implementing the infrastructure goals of their grants.

Acknowledgments and disclosures

Preparation of this article was assisted by contracts 04-M000030 and 03-M0002980-1D from the Substance Abuse and Mental Health Services Administration (SAMHSA) and by a grant from the Robert Wood Johnson Foundation Substance Abuse Policy Research Program. Views and opinions are those of the authors and do not necessarily reflect those of SAMHSA or the Department of Health and Human Services.

The authors report no competing interests.