In 1997 the state of New Mexico received a 1915 (b) Freedom of Choice waiver to implement a fully capitated Medicaid managed care system for physical and mental health services. The system is named Salud, the Spanish word for health. The state has been tweaking this system ever since its creation (1,3).
Two recent state and national reports highlighted the adverse impact of fragmentation of behavioral health services and the simultaneous need for comprehensively planned systems to deliver care (4,5). In response, on September 12, 2003, New Mexico's Governor, Bill Richardson, announced a major restructuring of Salud and other state-funded behavioral health services. All state agencies that finance mental health and substance abuse services are to establish an "interdepartmental behavioral health purchasing collaborative" to coordinate, administer, and oversee these services. This reform will remove behavioral health care from Salud and enable the purchasing collaborative to contract with one "statewide entity" to manage public mental health and substance abuse services, including those financed under Medicaid.
The proposed approach is full of promise, intended to decrease time-consuming and duplicative administrative processes and to redirect financial resources from bureaucracy to services. The system's design process offers the opportunity to increase payment rates for services by reducing program administration costs. The selection of a nonprofit or public organization to manage the system would minimize such costs, freeing funds for provider payments. The increased payment rates would facilitate the rebuilding of a tattered behavioral health care infrastructure that is inadequate to meet the needs of persons with serious mental disorders (1,34,5,6,7). For example, one local provider coalition documented statewide declines in child and adolescent services ranging from 20 to 65 percent under Salud (6). Another provider association, in a survey of its members, determined that 34 percent of licensed psychologists accepted Medicaid patients, compared with 86 percent before Salud was implemented (7).
In this column we review the essential design issues that will determine whether this latest and more ambitious effort will prove successful. Numerous relevant lessons can be drawn from the experience with Salud. These lessons relate to the rapidity with which Salud was conceived and implemented, basic program design, and system monitoring. These lessons must not be forgotten when multiagency collaboration and other matters unique to this pioneering effort are addressed.
Allowing time for design and implementation
The planning process that led to the establishment of Salud occurred at an accelerated pace over the course of four months. Under the current implementation plan, the Richardson administration allows only one year (September 2003 to September 2004) to complete the bulk of the design process and then initiate the competitive bidding process to select the statewide entity. Therefore, the administration should build on existing Medicaid planning initiatives in New Mexico, which involve state officials, consumers, providers, and advocates. The four committees undertaking these initiatives have spent numerous months considering how to redesign the Medicaid system. The work of these groups should not be abandoned in favor of yet another planning process. Furthermore, the administration must ensure that enough time is allotted for broad public input to be collected and integrated into the design of the new system.
The hurried transition to Salud engendered problems for providers of physical and mental health care. Providers experienced the transition as chaotic and stressful. Lack of training and basic information about Salud exacerbated the effects on providers and staff. Providers did not receive operational manuals until weeks after implementation, and telephone systems that were intended to supply providers and their patients with information about the program's rules were often inoperable or busy. Significant changes in workload occurred as a result of staff turnover and frequently changing policies of the managed care organizations, which impeded the development of adequate knowledge to cope with complicated administrative demands (3). To avert start-up problems, providers and staff must receive adequate education about their contractual relationship to the statewide entity and system processes, including service authorization, utilization review, and payment request procedures, all of which were problematic during the early years of Salud (3). Such instruction is essential before the implementation of the new system, scheduled for July 1, 2005—a date that many providers fear is too soon given the complex transition issues.
Administrative services only versus at-risk contract
At-risk contracts work best when the funding level can support services and administration. Because New Mexico has a history of funding Medicaid at extremely low levels (8), leaving little room for administration expenditures and profit, the planning process should consider whether a public administrative-services-only approach is warranted. Under this approach, an organization is responsible for processing service authorization and payment requests and completing other administrative functions but is not placed at risk for the cost of the services provided. This approach eliminates monetary incentives to withhold services and divert capitations to administration and profit because the state retains financial risk. Alternatively, the state can continue on the path of privatization, entrusting a corporate managed care organization to tend to the treatment needs of its most vulnerable populations. The Richardson administration's decision to replace the multiple managed care organizations that now provide behavioral health services under Salud with one entity may reduce costs and bureaucracy but may result in excessive market power being held by this same entity.
Comprehensive benefit package
Current behavioral health services for Medicaid enrollees are clustered around the extremes of inpatient care and outpatient therapy. Under Salud, expenditures for inpatient services increased while expenditures for community-based services remained the same (4). A less costly and more effective package would include in-home services, mobile crisis services, and psychosocial rehabilitation. New Mexico currently receives approximately three dollars for every dollar it spends on the Medicaid system. The reform may enable the state to leverage further funds from the federal government—funds that could be used to develop a broad continuum of community-based services.
Improved program evaluation
Salud initially gave scant consideration to program evaluation and monitoring and to the development of appropriate performance measures. Quality and contract monitoring were inadequate, because the state government lacked technical expertise—expertise that was required to collect and analyze data. The lack of a coherent quality monitoring system for behavioral health services made it difficult for the state government to be responsive to complaints from both consumers and providers that Salud had a negative impact on access to and quality of care.
Mechanisms for ensuring accountability under the reformed system must be clearly delineated, well-funded, and sustained. The Richardson administration should advance an early-warning monitoring system for behavioral health services that is modeled after systems that have proven to be effective in Vermont, Pennsylvania, and the District of Columbia. The basic elements of an early-warning system include the application of performance measures, availability of performance data in real time, community stakeholder involvement in the entire evaluation process, and public dissemination of all monitoring information. Such a system would illuminate the experience of individuals receiving services and facilitate the rapid identification and rectification of service delivery problems (9).
Attention to the safety net
Because the system reform will consolidate public funding for behavioral health services, the impact on safety net institutions—organizations that historically have had a commitment to providing services to persons with low incomes—is of utmost concern. Given that New Mexico is a poor and rural state, with 20 percent of the population insured through Medicaid, reductions in payments under Salud caused financial hardship for safety net institutions (3). These institutions have come to depend on a variety of non-Medicaid funding sources, entering into contractual arrangements simultaneously with several state agencies to remain solvent. With the advent of the collaborative purchasing approach, efforts must be made to ensure that current services can be maintained and expanded and that adequate funding is allocated to safety net institutions.
The discretion to enter into service delivery contracts with safety net institutions should not rest with the statewide entity only. An equitable, open process for awarding contracts is of paramount importance. Because this reform will affect all financing for public behavioral health services—not simply Medicaid—prudence must be exercised when this new system is rolled out, and a vigilant eye must be focused on safety net institutions and the vulnerable populations they serve.
The Richardson administration's reform can constitute a positive development for behavioral health policy in New Mexico if the program design is conceptualized and implemented cautiously, with close consideration to monitoring, especially with regard to how vulnerable populations and safety net institutions fare. The time devoted to a comprehensive planning process that builds on existing initiatives and is attentive to transition issues will give rise to a well-designed model that is supported by state agencies, consumers, providers, and advocates. It is imperative to nurture the development of community-based services in order to address access problems that have been encountered under Salud. An administrative-services-only contract would be more appropriate than an at-risk approach for a rural state that has few behavioral health providers and a history of low Medicaid funding.
This work was supported by grant 1-R03-MH-65564 from the National Institute of Mental Health and grant 1-R01-HS-09703 from the Agency for Healthcare Research and Quality.
Dr. Willging is affiliated with the Behavioral Health Research Center of the Southwest, 612 Encino Place, Albuquerque, New Mexico 87102 (e-mail, firstname.lastname@example.org). Mr. Semansky is with Westat in Rockville, Maryland. Howard H. Goldman, M.D., Ph.D., is editor of this column.