Health Care Reform Tracking Project:

Tracking State Managed Care Reforms as They Affect Children
and Adolescents with Behavioral Health Disorders and Their Families

2000 State Survey

View the PDF version
You will need Adobe Acrobat Reader to view this file.

Background

The Health Care Reform Tracking Project was initiated in 1995 to track and analyze public sector managed care initiatives as they affect children and adolescents with mental health and substance abuse (referred to as behavioral health) disorders and their families. The methodology of the Tracking Project has involved two major components: surveys of all States to describe State-managed care reforms and in-depth site visits to a select sample of States to analyze the impact of State policy choices and implementation strategies. To date, the Tracking Project completed 3 all-State surveys and 2 impact analyses involving a combined sample of 19 States.¹ Throughout its activities, the Tracking Project has explored and compared the differential effects of carve out designs, defined as arrangements in which behavioral health services are financed and administered separately from physical health services, and integrated designs, defined as arrangements in which the financing and administration of physical and behavioral health care are integrated (even if behavioral health services are subcontracted).

This paper presents highlights of major findings from the 2000 State Survey. All 50 States, plus the District of Columbia, responded to the survey, with most States (42) reporting involvement in publicly financed managed care activity affecting behavioral health services for children, adolescents and their families. There were slightly fewer States reporting managed care activity in 2000 than reported such activity in 1997–98. There were, in addition, reports of eight reforms terminated and three new reforms underway since the 1997–98 survey, indicating continued State and local experimentation with managed care. In response to the 2000 survey, detailed, descriptive data on 35 reforms underway in 34 States were provided; these included 27 carve outs and 8 integrated approaches. The highlights that follow pertain to this sample.

Beth A. Stroul, M.Ed.; Sheila A. Pires, M.P.A.; Mary I. Armstrong, M.B.A., M.S.W.

Findings

Role of State Medicaid Agency

• As they were in 1997–98, State Medicaid agencies continue to be the dominant players in managed care, reportedly having lead responsibility for 88 percent of the integrated reforms and 44 percent of the carve outs—overall 55 percent of all reforms.

Key Stakeholder Involvement

• Continuing a trend noted in 1997–98, key stakeholders (i.e., families, child-serving systems, providers) increased their involvement in managed care planning, implementation, and refinement activities over the past few years. Even with reported gains, however, significant involvement for most stakeholder groups occurs in fewer than half of the reforms. State education staff, the least likely stakeholder group to have involvement, reportedly have significant involvement in only 19 percent of reforms, in spite of the major role schools play in providing and referring children for behavioral health services. Juvenile justice staff reportedly have significant involvement in only 23 percent of reforms, in spite of increased enrollment of the juvenile justice population in managed care. State substance abuse staff are involved significantly in just over a third of reforms, and child welfare staff are involved in 46 percent. Families also reportedly have significant involvement in less than half (48 percent) of reforms—all carve outs—in spite of heightened national attention to amplifying the consumer and family voice in managed care systems.

Covered Populations

• The 2000 survey data suggest acceleration of a trend noted in 1997–98 of States' covering more Medicaid populations in managed care, including those with more serious, costlier service needs. There has been a reported 23 percent increase in the percentage of reforms covering the Supplementary Security Income (SSI) population, a 22 percent increase in reforms covering the child welfare population, and a 42 percent increase in reforms covering the juvenile justice population. These data also are consistent with 2000 survey findings that States, increasingly, are including both acute and extended care services within their managed care reforms as they include more populations requiring extended treatment.

Attention to Special Needs Populations

• Planning for special populations, such as children with serious emotional disorders and those involved in child welfare and juvenile justice systems, reportedly has increased in managed care (10 percent, 26 percent, and 24 percent increases for each population respectively), which might be expected given the increased enrollment of these populations in managed care.
  
• There has been a significant 44 percent increase since 1997–98 in the percentage of reforms that reportedly include special provisions for children with serious and complex behavioral health needs, such as intensive case management, an enhanced benefit package, or interagency coordination protocols. However, only 20 percent of all reforms incorporate risk-adjusted rates for this population (and even fewer incorporate risk-adjusted rates for the child welfare and juvenile justice populations).

Types of Managed Care Organizations (MCOs)

• There has been a reported 7 percent increase since 1997–98 in the use of for-profit behavioral health organizations, particularly within reforms with integrated designs, perhaps indicating a movement toward the use of specialized entities to manage behavioral health services. There was also a 15 percent increase in the use of government entities as MCOs, found almost exclusively within carve outs, continuing a trend noted in 1997–98. Community-based, nonprofit agencies remain the least likely type of entity to be used as MCOs.

Service Coverage

• There has been a reported 18 percent increase in the percentage of reforms covering a broad array of services, with carve outs far more likely to include broad service coverage than integrated reforms (70 percent versus 13 percent). The services most likely to be covered by managed care reforms reportedly are assessment and diagnosis, outpatient psychotherapy, crisis services, medical management, day treatment, and inpatient services. The services least likely to be covered are therapeutic foster care, respite services, therapeutic group care, and residential treatment.
  
• While carve outs are more likely than integrated reforms to expand coverage of home and community based services (63 percent versus 38 percent), the 2000 State Survey also found that, regardless of the design of the managed care system, these services continue to be provided outside of managed care systems by other child-serving systems, such as child welfare. The services most likely to be provided outside of managed care systems, instead of or in addition to coverage within the managed care system, reportedly are therapeutic foster and group care, residential treatment, school-based mental health services, transportation, family support and education, wraparound services, and respite.
  
• As in 1997–98, few services reportedly are provided to infants, toddlers, preschoolers, and their families through managed care systems, with half of the carve outs and three-quarters of the integrated reforms reportedly providing "few to no" services to this population.

Service Capacity

• Although over half of managed care reforms reportedly have expanded home- and community-based service coverage, in nearly half, there reportedly has been little to no expansion in service availability. Carve outs are far more likely than integrated reforms to have expanded service availability (73 percent versus 12 percent). However, lack of service capacity, a problem that pre-dated managed care, remains an issue in virtually all states. In only 31 percent of carve outs and in none of the integrated reforms was behavioral health service capacity for children and adolescents in the state rated as highly developed or close to highly developed. Yet, respondents also reported a 16 percent decline since 1997–98 in the percentage of reforms that require reinvestment of savings to expand service capacity.

Support for Systems of Care

• Similar to findings in 1997–98, managed care reforms are reportedly supportive of systems of care in the majority of carve outs (88 percent) but in only 29 percent of integrated reforms. Carve outs reportedly have much higher rates of inclusion of system of care values and principles than do integrated reforms. For example, nearly all carve outs (92 percent) incorporate a broad array of services, family involvement, and interagency service coordination principles and protocols, compared to 57 percent of integrated reforms.

Managed Care Financing

• As in 1997–98, the 2000 State Survey found that Medicaid and mental health agencies are the primary sources of financing for managed care systems, with Medicaid agencies contributing in 91 percent of the reforms and mental health agencies contributing in 76 percent of the reforms. There has been a significant (20 percent) increase since 1997–98 in reforms with mental health agency financing, though carve outs are far more likely to involve mental health agency financing than integrated systems (96 percent versus 13 percent).
  
• There has been a 15 percent increase since 1997–98 in the percentage of reforms in which both the Medicaid and behavioral health agencies contribute to the financing pool. However, the 2000 survey found declines in the financial participation of other child serving agencies, and the level of participation of these agencies remains low, with the child welfare system contributing resources in only 21 percent of the reforms, other systems in fewer than 10 percent of the reforms, and the education system in none of the reforms. Carve outs are far more likely than integrated systems to use dollars from other child-serving systems.
  
• There is little difference between carve outs and integrated reforms in use of Medicaid, TANF, and SCHIP dollars. However, there are significant differences in the extent to which each uses other types of dollars, with carve outs being far more likely to use State general revenue, block grants, and child welfare dollars. Carve outs are far more likely to draw on multiple funding streams from multiple agencies, and integrated systems are more likely to rely almost exclusively on Medicaid and SCHIP dollars contributed by the Medicaid agency.
  
• In virtually all cases (91 percent of reforms), some Medicaid dollars for children's behavioral health are left outside of the managed care system in other child-serving systems. Reportedly, the education system is most likely to be using Medicaid dollars outside of managed care (in 81 percent of reforms), but all other child-serving systems (i.e., child welfare, mental retardation/developmental disabilities, juvenile justice, and child mental health) also are using Medicaid dollars outside of managed care (in 50 percent to 72 percent of reforms, depending on the system). Thus, even though more managed care reforms include coverage for both acute and extended treatment, other child-serving systems still retain responsibilities and funding for behavioral health service provision outside of managed care systems. This reality may create a safety net for children unable to access needed services through the managed care system, but it also perpetuates opportunities for fragmented care and cost shifting.

Cost Shifting

• In over two-thirds of reforms, cost shifting reportedly is occurring. Cost shifting from the managed care system to other child-serving systems is more likely to occur in integrated reforms, and cost shifting from other child-serving systems to the managed care system is more likely to occur in carve outs. Only 16 percent of reforms are reported to be tracking and monitoring cost shifting.

Risk Structuring

• As in 1997–98, most reforms (88 percent) use some type of risk-based financing, with carve outs more likely to use case rates or no risk-based financing and integrated reforms more likely to use capitation. Most reforms (83 percent) reportedly have made changes in rates since 1997–98, with most (80 percent) of the changes being rate increases. None of the integrated reforms requires that a certain specified percentage of the rate be allocated to behavioral health care.
  
• More States than in 1997–98 reportedly are now incorporating risk sharing arrangements with MCOs, with only about one-half of reforms pushing full risk to MCOs (20 percent fewer than in 1997–98).
  
• Slightly more than half of reforms (55 percent) place limits on MCO profits, and about half limit administrative costs. As in 1997–98, carve outs are far more likely than integrated reforms to limit MCO profits. Carve outs also are more likely to tie bonuses or penalties to MCO performance with respect to behavioral health care.

Clinical Decisionmaking Criteria

• Most reforms (82 percent) reportedly now have medical necessity criteria that allow consideration of psychosocial and environmental factors in clinical decisionmaking. However, it also was reported that, in 60 percent of the integrated reforms and nearly 20 percent of the carve outs, MCOs continue to interpret and apply criteria narrowly. Similarly, while two-thirds of reforms reportedly incorporate level of care criteria specific to children's behavioral health, it also was reported that, in two-thirds of the integrated reforms and one-third of the carve outs, use of these criteria have not improved consistency in clinical decisionmaking.

Access

• Initial access to behavioral health services reportedly has improved since implementation of managed care in 76 percent of carve outs but in only half of the integrated reforms, with reports of worse access in one-third of integrated reforms and in 10 percent of carve outs. Access to extended treatment reportedly has improved in 39 percent of carve outs but in only 20 percent of integrated reforms, with reports of worse access in 60 percent of integrated reforms, but in only 4 percent of carve outs.
  
• Initial access to inpatient care was not reported to be problematic, regardless of managed care design. However, inpatient lengths of stay reportedly are shorter in most managed care systems (56 percent of carve outs and 88 percent of integrated systems), and a number of problems were reported in association with reduced length of inpatient stays, particularly within integrated reforms. Premature discharge before stabilization, children discharged without needed services, and placement of children in community programs lacking appropriate clinical capacity reportedly occur in more than 40 percent of integrated reforms and one-quarter of carve outs. Inappropriate use of child welfare shelters was reported in 43 percent of integrated reforms and 8 percent of carve outs, and inappropriate use of juvenile justice facilities was reported in 29 percent of integrated reforms and 13 percent of carve outs. Carve outs were reported to be more likely than integrated reforms to develop alternatives to hospitalization (68 percent of carve outs versus 43 percent of integrated reforms are doing so).

Service Coordination

• The 2000 State Survey suggests some improvement in coordination between physical and behavioral health care; improvement was noted in 60 percent of reforms, with little difference between carve outs and integrated reforms. Improved coordination seems to be associated with specific efforts, rather than the design of the reform per se. The 2000 survey also shows some improvement in coordination between mental health and substance abuse, with improvements noted in slightly over half (52 percent) of reforms, again with little difference between carve outs and integrated reforms. Interagency coordination across child-serving systems reportedly has improved in 65 percent of reforms, with this being more likely the case in carve outs than in integrated reforms (71 percent versus 43 percent) and attributed to the need to address and solve problems and challenges resulting from managed care implementation.

Cultural Competence

• While most reforms (85 percent) include general requirements related to cultural competence, specific strategies for enhancing cultural competence were reported with greater frequency for carve outs than for integrated reforms. For example, nearly half of the carve outs reportedly include special services needed by culturally diverse populations, in addition to translation/interpretation services, compared to none of the integrated reforms, which tend to provide translation/interpretation services only. Nearly half of the carve outs reportedly track utilization and outcomes by culturally diverse groups, compared to none of the integrated reforms.

Family Involvement

• Over two-thirds of carve outs reportedly incorporate various strategies to involve families at the system and service levels in managed care, such as using family advocates, covering family supports, requiring family involvement on system advisory bodies, and requiring family involvement in service planning. In contrast, nearly 70 percent of integrated reforms reportedly have no family involvement strategies with respect to families who have children with behavioral health disorders.
  
• The 2000 State Survey found, much like previous surveys, that managed care reportedly has had no impact, one way or the other, on the practice of families' having to relinquish custody to obtain behavioral health services for their children.

Provider Issues

• The majority of both carve outs and integrated reforms reportedly include culturally diverse providers, certified addictions counselors, and school-based providers in their provider networks. However, while about two-thirds of carve outs also include traditional child welfare providers, paraprofessionals, and student interns within their networks, only 13 percent of integrated reforms do so. The use of family members as providers was reported for 42 percent of the carve outs, but for none of the integrated reforms.
  
• Provider reimbursement rates reportedly are lower under managed care than under the previous fee-for-service system in 57 percent of the integrated reforms, but in only 25 percent of the carve outs. In less than a quarter of reforms, mostly carve outs, were rates reported to be higher. Although provider reimbursement rates were found to be either lower or unchanged in most reforms, administrative burden for providers reportedly has increased in nearly two-thirds of reforms, regardless of design; administrative burden was reported to be less under managed care in only 12 percent of reforms.

Accountability

• More than half of the integrated reforms and over one-third of the carve outs reportedly do not have adequate data for behavioral health care decision making, attributed to inadequate management information systems, lack of encounter data, and lack of staff capacity to analyze data.
  
• Only 21 percent of integrated reforms reportedly track the total cost of children's behavioral health services, while nearly all (96 percent) of the carve outs do.
  
• Continuing a trend noted in 1997–98, more reforms reportedly are measuring clinical and functional outcomes, with nearly all (90 percent) reforms reportedly incorporating such measures. However, in nearly half of these reforms, outcome measurement was described as being in early stages of development, with only about one-quarter of reforms—all carve outs—having results from the measurement of clinical and functional outcomes.

Impact Findings

• Not surprisingly, with the lack of adequate data, most managed care systems do not as yet know what impact they are having on children's behavioral health care. In 41 to 46 percent of reforms, the impact of managed care reforms on penetration rates for children's behavioral health care, service utilization, quality, cost, and family satisfaction was unknown. In 63 percent of reforms, impact on clinical and functional outcomes was unknown. In the very small sample in which results were known, most were in a positive direction, with the exception of cost control, where mixed results were reported. Decreased aggregate costs were reported in 19 percent of reforms, increased costs in 25 percent, and no change in 16 percent.

State Child Health Insurance Program (SCHIP)

• The 2000 State Survey found that about half the States have implemented SCHIP as a Medicaid expansion, and about half as a separate program. SCHIP behavioral health benefits were more likely to be reported as broad within Medicaid expansions and limited within separate SCHIP programs.
  
• Where SCHIP was implemented as a separate program, there was a high level of coordination reported between SCHIP and Medicaid in general. However, less than half (43 percent) of the separate SCHIP programs reportedly incorporate strategies to identify and refer children with behavioral health needs.
  
• More than half of the states (57 percent) reported having no data available on the impact of SCHIP on children's behavioral health services. Where data were available, about one-third of the States reported that more children are receiving behavioral health services as a result of SCHIP.

Child Welfare Special Analysis

A specific focus on child welfare issues has been incorporated into all aspects of the Tracking Project. The 2000 State Survey explored a number of issues related to the impact of behavioral health managed care reforms on children and adolescents involved with the child welfare system and their families. Most States now include children in the child welfare systems in their behavioral health managed care systems; 82 percent of the reforms analyzed for the 2000 State Survey reportedly include this population, increased from only 45 percent that included the child welfare population in 1995.

Findings from the 2000 State Survey demonstrate that many concerns of child welfare stakeholders are being addressed. For example, most reforms have special provisions for children in the child welfare systems (87 percent), most provide training for the child welfare system about managed care (72 percent), more than half of the reforms train MCOs about the unique needs of children and families in the child welfare system (52 percent), and most are able to track the use of behavioral health services for children in the child welfare systems (74 percent). On the other hand, a number of concerns reportedly have not as yet been addressed, such as: in 54 percent of the reforms, the child welfare system is not significantly involved in planning, implementing, and refining the behavioral health managed care system; only 11 percent of the reforms have an enhanced capitation or case rate for children in the child welfare system; fundamental services used by children in the child welfare system are not covered in nearly half of the managed care systems; and child welfare providers are not included in provider networks in 47 percent of the reforms.

The special child welfare analysis included in the full report also compares findings from the 2000 State Survey to findings from a survey conducted by the Child Welfare League of America in order to further understand issues related to children and families involved with the child welfare system.

Concluding Observations

Publicly financed managed care continues to present opportunities and challenges for children's behavioral health care. States and counties are becoming more sophisticated designers and purchasers, and, particularly in the case of carve outs, often incorporate specifications that would seem to benefit children needing behavioral health care. However, managed care implementation continues to lag behind policy intentions. Broad benefit designs are hampered by lack of service capacity; broad medical necessity criteria designed by states are rendered meaningless by narrow interpretation at the MCO level. Greater attention is paid to special needs populations in planning and policy as States increasingly enroll these populations in managed care, but training of MCOs about these populations and changes in financing to guard against underservice do not necessarily follow.

The design and implementation of managed care is a developmental process. States are beginning to turn their attention to implementation problems, and most have made mid-course corrections in policy and design. Particularly in States with integrated designs, children with behavioral health disorders typically have not emerged initially as a priority population. Through the efforts of family members, advocates, and researchers, and as a result of their own quality improvement activities, States are beginning to look more closely at how this population experiences managed care and at approaches that may be more effective. As part of this developmental process, the Health Care Reform Tracking Project, while it will continue to survey managed care developments, also is moving to a new phase of identifying promising approaches and features of managed care reforms. The 2000 State Survey captured (and describes in the report) a number of promising approaches that States are undertaking on behalf of children with behavioral health disorders and their families. The Tracking Project will be studying and reporting on promising practices in greater depth in future activities.

¹ Reports are available from the University of South Florida, Louis de la Parte Florida Mental Health Institute, Office of Training Support/Technical Publications, 13301 N. Bruce B. Downs Boulevard, Tampa, FL 33612, 813-974-4484. Special analyses specific to the child welfare system are available from the National Technical Assistance Center for Children's Mental Health, Georgetown University Child Development Center, 3307 M Street NW, Washington DC 20007, 202-687-5000.

MC-02-83