Skip Navigation
Administration for Children and Families  
ACF
ACF Home   |   Services   |   Working with ACF   |   Policy/Planning   |   About ACF   |   ACF News   |   HHS Home

  Questions?  |  Privacy  |  Site Index  |  Contact Us  |  Download Reader™  |  Print      


Children's Bureau Safety, Permanency, Well-being  Advanced
 Search

Home > Programs & Funding > Profiles of the Title IV-E Child Welfare Waiver Demonstration Projects > Connecticut - Managed Care Payment System

Connecticut

Demonstration Type: Managed Care Payment System1
Approval Date: September 29, 1998
Implementation Date: July 9, 1999
Completion Date: October 20022
Interim Evaluation Report Expected: June 2002
Final Evaluation Report Expected: July 2003
 

Target Population

Children eligible for this demonstration were those between 7 and 15 years of age with significant behavior problems and whose placement in residential care or in a group home had been authorized. Of these children, only those with "moderate" mental health acuity levels were eligible for the demonstration.

Jurisdiction

The demonstration operated in two of the State's five regions. One Lead Service Agency (LSA) served children in the North Central Region of the State, while a second LSA served children in the South Central Region.

Intervention

Connecticut used a managed care model to address the high level of need and costs related to providing services to children with behavioral, mental health, and educational problems. Connecticut contracted with two LSAs to provide a continuum of services in treatment facilities and community-based settings.

For children in the experimental group, the State expected each LSA to place each child in the least restrictive setting possible and to coordinate the provision of comprehensive care using a network of service providers. Services included case management, group care, home-based services, outpatient services, residential treatment, and aftercare. The State and the LSAs agreed that the LSAs would serve a maximum of 30 children at any given time. Children in the control group received standard services through the Department of Children and Families (DCF).

Connecticut paid the LSAs a fixed rate for each referred child, which was equal to the average cost of 12 months of residential placement. Funds cover the full range of services necessary for each referred child and family. The State expected the LSAs to serve children and families for 15 months (including 3 months of aftercare).

For reimbursement, the State and the LSAs negotiated a shared-risk corridor. The LSAs retained savings of up to 10 percent below the fixed rate. However, the LSAs were responsible for costs of up to 110 percent of the fixed rate. The LSAs were responsible for any residential service required during the first 6 months following achievement of the permanency goal, up to the 15-month service requirement. The State paid 25 percent of the rate to the LSAs upon case acceptance, an additional 25 percent of the rate following 60 days of service, and 25 percent of the rate following 180 days of service. The LSAs received the remaining 25 percent upon treatment completion or at the end of 15 months.

Evaluation Design

The evaluation consisted of process, outcome, and cost-effectiveness components. Connecticut used random assignment in its evaluation design. With the implementation of two experimental sites over a five-year demonstration, the State initially expected to enroll approximately 240 children and families in the demonstration (including both experimental and control groups). The State used the following outcome measures: average length of stay in out-of-home care, substantiated allegations of child abuse/neglect, use of less restrictive placements, children's behavioral health, and child and family satisfaction with the Department's services.

Evaluation Findings

The process and outcome findings presented here include excerpts from Connecticut's Final Report, dated July 2003.

Process Evaluation

After three years of implementation, in February 2002, the State decided to discontinue the demonstration due to statewide reform of Connecticut's behavioral health system, which affected the need for the title IV-E waiver. By the beginning of the third year of implementation, referrals were inadequate to sustain the LSA contractors. The State, therefore, modified its contracts with the LSAs such that all cases needing ongoing services were transitioned back to DCF by June or October 2002 (depending on the site).

A total of 157 children participated in the waiver demonstration evaluation3, with 79 children in the experimental group and 78 children in the control group. The North Central Region LSA received a negotiated rate of $50,911 per case, while the South Central Region LSA received $48,000 per case.

  1. Children referred: The demonstration was designed to focus on children who displayed moderate levels of mental health needs. A total of 432 children were evaluated for inclusion in the demonstration. Of these, 263 (61 percent) were ineligible for the demonstration. In two-thirds of these cases, the child's mental health needs were too severe for them to be included in the demonstration.
  2. Service delivery systems: The State reported several differences in the approaches and procedures used by the two LSAs to achieve their goals. These include differences in staffing arrangements, caseload sizes, and service delivery network systems, as described below.
    • In the North Central Region, the LSA was a multi-service agency that included residential treatment services. In the South Central Region, the LSA was a general community hospital.
    • In the North Central Region, the LSA shared financial risk with a coalition of five-regionally based agencies. This LSA also established fee-for-service contracts with providers outside the coalition, when necessary. In the South Central Region, the LSA used child-specific, fee-for-service contracts with six service providers and individual therapists.
    • The LSA in the North Central Region used a care coordinator who managed an average of 15 cases. In the South Central Region, a team of two staff members served an average of 11 cases each. In addition to providing case management services, the team provided counseling and other services when the LSA could not purchase necessary services. (In contrast, DCF workers averaged a 24-family caseload.)
    The State found that these service delivery arrangements differed from the comprehensive service delivery systems that were anticipated. Network partners participated in the care of children; however, only the LSAs provided assessment, case management, quality assurance, and discharge planning services. In addition, the LSAs purchased most of the children's services through child-specific agreements without the creation of new community-level service initiatives. The State offered several explanations for the limited development of community-based, continuum-of-care service systems. The State reported that it was difficult to create and manage a comprehensive service system using a single rate payment system. Reasons given included the diversity of children's and families' needs, the small number of families, and the geographic distribution of these families.
  3. Services to children: Connecticut analyzed service data for 109 children (52 children in the experimental group and 57 children in the control group) through February 2002 (the first 12 months of the program). The State reported significant differences in the services the LSAs provided to children, as compared to traditional services, during the first year of implementation.

    While both DCF and the LSAs provided an array of services to children, the State found statistically significant differences in the percentage of experimental group children receiving the following services as compared to control group children: (1) crisis stabilization, (2) day treatment, (3) family therapy, (4) family preservation, (5) family support services, (6) behavioral aide services, (7) respite care, and (8) transportation services. The State also reported differences in the number of service units provided to families each month. The LSAs provided experimental group children with more frequent case management, family support, and transportation services than DCF provided to control group children. The LSAs provided children with less frequent medication/treatment monitoring, residential treatment, and inpatient hospitalization than DCF provided to children.

    The analysis also considered differences between the first and second years of treatment. When compared to the first 12 months of treatment, the State reported that, during the second year, children participated in fewer services, both in DCF and in the LSAs. The LSAs provided more intensive services than DCF in the areas of case management, family support, and transportation. DCF, on average, provided more units of service for residential treatment, inpatient hospital stays, and medication monitoring. Those receiving services in the LSA programs (45.2 percent) were less likely to report placement in a residential treatment center as compared to children receiving services through DCF (65.2 percent). During the second year, the difference in the placement within residential treatment facilities was approaching significance.
  4. Role tension: According to the State, both LSAs noted that the most difficult children to serve where those children legally committed to DCF. The need to have two agencies involved in separate but related sets of issues created a certain level of role tension. While the contracts with the LSAs delineated the roles and responsibilities of the LSAs and DCF, the State found that there was a need to better define the roles of frontline case managers. Staff interviews and case records indicated that authority to set case focus and treatment direction was not always consistent, and the dual approaches were sometimes a source of confusion.
  5. Discharge criteria: The State reported a lack of clarity regarding how the LSAs applied the clinical discharge criteria as defined by the State. Most often, discharge appeared to be connected with the end of the 15-month service period. Seventy-three percent of children assigned to the LSAs were discharged within the 15-month period. The remaining 27 percent were discharged within the next six months. A factor complicating discharge criteria was that the majority of children discharged from the demonstration were not in fact discharged from DCF once they left the LSAs. It is therefore recommended that future system-of-care efforts seek to better define the term "discharge" and to specify when it is appropriate to discharge children from mental health care when their substitute care needs have not been met.

Outcome Evaluation

By February 2002, the State had conducted structured interviews with 118 children and caregivers (54 children and caregivers in the experimental group and 64 children and caregivers in the control group) regarding their experiences in the first 12 months since program entry. The State reported the following outcome findings on data from 109 of the interviews:

  1. Custody changes: At the 12-month interval, the State found small, statistically insignificant differences between the experimental and control groups with respect to the percentage of children who experienced changes in custody. While children in both groups spent most of their first 12 months in out-of-home placement, children served through the LSAs were found to have more family placements than residential treatment placements, both during and at the end of the first 12 months. However, for the time period between 12 and 24 months, similar rates of family placement (e.g., with parents, relatives, or in adoptive homes) were found across both groups.
  2. Placement type and placement days: The State reported that, on average, both groups of children spent the predominant amount of their time in residential treatment centers. However, less time was spent in residential treatment centers during the 12- to 24-month period than during the initial 12 months for both the control and experimental groups.

    During the first year, the differences in the percentage of days between those at DCF and at the LSAs were significant. For the children served in the control group, 64 percent of all days were spent in residential treatment centers, compared to 45 percent of all days for children served at the LSAs. The LSAs were more successful at returning children home faster. At 12 months, 36 percent of children in the experimental group and 11 percent of children in the control group were in in-home placements. At the time of the 24-month interview, the gap began to close. Forty-four percent of the children who received services in the LSAs were then in in-home placement, while 37 percent of the children who received services through DCF were in in-home placement.
  3. Mental health status: The State reported that both experimental and control group children improved significantly from program entry to 12 months using three measures of clinical mental health symptoms: reduction in clinical mental health symptoms, decreases in level of functional impairment, and increases in strengths. Results of the mental health indicators at the 24-month interval revealed that children continued to improve. Rates of improvement in clinical symptoms were above 50 percent for both experimental and control group children. Strength levels continued to improve after 24 months for 40 percent of all children. Levels of improvement between control and experimental groups were not significantly different.
  4. The relationship between placement status and mental health: According to the State, the data suggest that there is a strong association between placements and mental health outcomes, and that many children experienced improvements in their mental health status, resulting in less restrictive placements. Specifically, children maintained in in-home settings showed the most improvement in behavior and functioning. While improvements in clinical symptoms and in-home placement were highly related, the type of service program (i.e., DCF versus LSA) did not appear to have a significant effect on this relationship.

Cost Analysis

Overall, the State reported that services delivered by LSAs which were paid at the case rate were cost neutral. The average 15-month expenditure per child was $49,310 for the LSAs, compared to the estimated State residential costs of $62,000 for the same time period.



1 Based on information submitted by the State as of July 2003. Back
2 Connecticut's demonstration project was originally a five-year project; the State terminated the project early, due to a lower than expected number of referrals and statewide mental health care system reform. Back
3 Two additional children participated in the program but did not consent to the evaluation. Back

Back to Table of Contents