Home > Programs & Funding > Profiles of the Title IV-E Child Welfare Waiver Demonstration Projects > Ohio - Flexible Funding - Phase II

Ohio

Demonstration Type:	Flexible Funding - Phase II1
Approval Date:	October 1, 2004
Implementation Date:	October 1, 2004
Expected Completion Date:	September 30, 2009
Interim Evaluation Report Received:	August 30, 2007
Final Evaluation Report Expected:	March 30, 2010

Target Population

The Target Population for Ohio's Phase II waiver demonstration (referred to as ProtectOhio) includes children ages 0–17 who are at risk of or in out-of-home placement, and their parents or caregivers. Both title IV-E-eligible and non-IV-E-eligible children may participate in the demonstration.

Jurisdiction

Phase II of the demonstration is operating in the 14 counties that participated in Ohio's initial five-year waiver demonstration: Ashtabula, Belmont, Clark, Crawford, Fairfield, Franklin, Greene, Hamilton, Lorain, Medina, Muskingum, Portage, Richland, and Stark. Hamilton County began Phase II but temporarily discontinued its participation from October 2005 to October 2007. In October 2006, four additional counties joined the waiver demonstration: Coshocton, Hardin, Highland, and Vinton.

Intervention

Participating counties continue to use title IV-E funds flexibly to prevent the unnecessary removal of children from their homes and to increase permanency rates for children in out-of-home placement. For Phase II, the State selected five distinct "intervention strategies" that are the focus of waiver activities. As their core intervention strategy, all 18 participating counties are implementing Family Team Meetings (FTMs), which bring together immediate family members, social service professionals, and other important support resources (e.g., family, friends, extended family) to jointly plan for and make crucial decisions regarding a child in or at risk of placement. An independent, trained facilitator in each county arranges and supports the FTM process. In addition to FTMs, each participating county is implementing at least one of the following core intervention strategies/supports:

• Structured visitations between parents/caregivers and children in out-of-home placement to promote reunification (eleven counties);
• Kinship support services to facilitate and maintain kinship placements (six counties);
• Managed care strategies, especially related to case rate contracting (one county); and
• Enhanced mental health/substance abuse (MH/SA) assessments and services (five counties).

Evaluation Design

Ohio's evaluation includes process and outcome components, as well as a cost analysis. The State's evaluation is testing the hypothesis that the flexible use of title IV-E funds to provide individualized services to children and families will decrease the frequency and duration of out-of-home placements, increase reunification rates for children in out-of-home care, and decrease rates of re-entry into foster care, while keeping children at least as safe as they would have been without the waiver.

As during the original demonstration, the Phase II evaluation employs a comparison group design with counties serving as the unit of analysis. The same 14 counties that formed the comparison group during Phase I of the demonstration are being used in Phase II: Allen, Butler, Clermont, Columbiana, Hancock, Hocking, Mahoning, Miami, Montgomery, Scioto, Summit, Trumbull, Warren, and Wood Counties. In addition, Guernsey, Morrow, and Perry Counties were added to the comparison group to balance the new counties in the experimental group. In selecting comparison counties, the State considered several relevant demographic and child welfare variables to ensure comparability with experimental group counties, including population size and density, percent of county designated as rural, poverty rates, child abuse and neglect rates, out-of-home placements rates, and median number of placement days.

Process Evaluation Findings

As mentioned above, Phase II of the ProtectOhio waiver demonstration has focused on five distinct intervention strategies that were selected by the State and the participating experimental counties. As such, the Process Implementation Study consists of five distinct research studies, all addressing structural or service delivery changes being implemented systematically in all or some of the experimental counties. The State's Interim Evaluation Report (June 2007) describes findings from the Process Implementation Study on each of the five intervention strategies, as summarized below:

1. Family Team Meetings (FTMs): Beginning in early 2005, all 18 experimental counties started to implement the ProtectOhio FTM model. As of September 2006, the ProtectOhio counties had conducted over 5,440 FTMs, involving over 2,900 children. While the individual experimental counties had fairly low fidelity to the ProtectOhio FTM model² overall, they appear to do better in adhering to certain model components than others. In particular, 75 percent of children in the experimental counties had a first meeting within 35 days and in four counties, well over 90 percent of children have had a second meeting within 100 days of the first one.
2. Kinship Supports: Six counties implemented the enhanced kinship strategy which focused on increasing the identification and recruitment of kinship caregivers, as well as increasing the services and supports available to these caregivers. In general, kinship counties utilized some components of the kinship model more frequently than other experimental or comparison counties. For example, all six kinship counties utilized genograms or another specific form to identify potential kin, compared with only three of seven other experimental counties and seven of fourteen comparison counties. In addition, four of six kinship counties offered either a one-time or per diem cash payment compared to two of seven other experimental counties and one of fourteen comparison counties. Overall, experimental counties appear to be using kinship caregivers in higher proportions than comparison counties; this pattern emerges when looking at children placed with relatives while in public child-serving agency (PCSA) custody and unpaid placement days (an indication of the use of kinship caregivers).
3. Supervised Visitation: Overall, the eight original experimental counties participating in supervised visitation have implemented a visitation program which appears to be an enhancement to the regular visitation program occurring in most other counties. As a group, these eight counties held over 5,000 visits, serving nearly 500 children, during the first year of data collection. However, the eight visitation counties have had varying degrees of success adhering to the ProtectOhio model for supervised visitation. For the components of duration and attendance, all counties have easily met the model standard, with 97 percent of all visits lasting at least one hour and attended by at least one parent. However, counties are performing less well on the components of planned activities and weekly visits. Only 65 percent of visits were recorded as having activities which were planned and at least partially completed, while the average amount of time between visits was 15 days for visits one and two, and 9 days for visits 2–10, exceeding the model standard of seven days.
4. Mental Health/Substance Abuse (MH/SA) Services: The five counties participating in this strategy are seeking to improve the process of obtaining MH/SA evaluations and services for child welfare clients. The study team has completed the evaluation of Lorain County's MH/SA enhancements; it reveals some encouraging patterns. Collection and analysis of case record and outcome data from 93 families suggest that since the implementation of enhanced MH/SA services:
   • More children received assessments;
   • More clients had evidence in their case record of treatment completion;
   • Cases closed more quickly;
   • The time between assessment and case closing was shortened;
   • The time between the start of services and case closing was shortened;
   • More children experienced a substantiated or indicated CAN report while their cases were open or within one year after closing; and
   • Cases opened after implementation of MH/SA services had fewer placement days.
   Many of these results may simply reflect an overall agency trend toward closing cases more quickly. However, evaluation evidence suggests that the efforts of the Lorain PCSA's Alcohol & Drug and Extended Casework Services units have expedited services for families and are making practical differences in case resolution.
5. Managed Care: The managed care study in Franklin County examined the impact of a managed care model on outcomes for children. The Franklin County Children Services agency employed case rate contracting on a sample of cases, with the goal of more effectively and efficiently using limited service resources. The evaluation tested the hypothesis that the use of case rate financing leads to no worsening of outcomes for children. In order to do this, the study team used administrative data to build a hierarchy of case histories and examine the quality of long-term child outcomes. The analysis found no evidence of any significant differences in the quality of child outcomes between cases assigned to private contractors and those assigned to the public agency. However, a high rate of "hold-backs," in which nearly half of the children assigned to private agencies were actually served by the public agency, makes it difficult to detect effects. In analyzing the hold-backs, the team found that among the children assigned to a private agency, those who were recent victims of child maltreatment were more likely to actually be served by the public agency (i.e., never transferred to the private agency) than were children who were not recent victims. The evaluation team notes that this variation does not compromise the validity of the estimates on child outcomes; however, it is an indication of systematic differences between the public and private agency caseloads.

Outcome Evaluation Findings

Utilizing data from Phase I of the ProtectOhio waiver the State's Interim Evaluation Report describes two major analyses of child outcomes: (1) child safety, through examination of case trajectories; and (2) permanency outcomes for children who were in placement at the start of the waiver and movement to less restrictive placement.

1. Child Safety Analysis: The waiver demonstration is designed to shift services away from placement by changing the fiscal mechanisms that support in-home and out-of-home services. In effect, the goal is to level the playing field so that service investments that lower the demand for out-of-home care draw Federal financial participation in the same way that out-of-home care would have. A key assumption underlying the waiver is that services provided in the home offer at least as much protection for the child as placement (i.e., supporting children in the home will not increase the likelihood of a subsequent maltreatment report).
   
   The study team assembled child-level data to examine the initial sequence of events in a case (referred to as a trajectory), and is assessing whether the basic patterns of events have changed over time as a result of the waiver. At this stage the study team is not able to draw definitive conclusions about safety outcomes; however, the results do point in some clear directions. Over time in the experimental counties, children appeared to be more likely to have their cases opened and somewhat less likely to be placed. With respect to safety, the evidence also suggests that over time, children in experimental counties were less likely to have a subsequent report of maltreatment. However, the data show that subsequent reports of maltreatment also declined in the comparison counties and that the rate of change sometimes favored the comparison counties.
2. Analysis of Children in Long-Term Placement: Using data from the first waiver period (Phase I), the evaluation team examined permanency and placement changes to less restrictive environments (step-downs) for children in long-term, out-of-home care. The study team found that:
   • During the first waiver, experimental counties were able to use waiver funds to move children languishing in long-term placements into stable, permanent living arrangements. Without the waiver, fewer of these children would be in permanent settings.
   • There was some evidence that the waiver contributed to stable step-downs for children in congregate care. Without the waiver, fewer children in experimental counties would have been placed into less restrictive settings.

Cost Analysis Findings

The State's cost study compares the experimental and comparison counties for significant differences in foster care maintenance expenditures and non-placement-related expenditures, including the costs of specific child welfare services and interventions. This study uses data from Phase II of the waiver and examines changes in spending patterns over time. The distribution of changes in foster care board and maintenance expenditures during the interim period between the first and second waiver (2003-2004) reflected a somewhat different pattern than those observed during the first waiver (1998-2002). In the first waiver, comparison counties occupied positions at both the low and high end of expenditure change distribution, although they were more clustered at the high end. During the interim period, three experimental counties reduced foster care expenditures significantly while comparison counties continued to dominate in the higher end of the distribution. The pattern suggests that experimental counties were beginning to control expenditures more than comparison counties. When considering the first year of Phase II of the waiver relative to the average of the two interim years, it again appears that expenditure changes in the experimental counties are beginning to diverge from those in comparison counties. Specifically, experimental counties as a group appear to be controlling foster care expenditures more than comparison counties. These findings suggest that the waiver incentive may be beginning to operate as it was intended in experimental counties.

Web Links

All evaluation reports associated with Ohio's demonstration are available at the following Web site: http://jfs.ohio.gov/ocf/pohio.stm

¹ Based on information submitted by the State as of November 2007. Back

² ProtectOhio FTM model components include: (1) all FTMs will be facilitated by independent trained facilitators; (2) an initial FTM will be held within 30 days of case opening, at least quarterly thereafter, and at all other critical points in the life of the case; and (3) FTM participants will include family members, foster caregivers, social service professionals, and other support persons important to the family. Back

Back to Table of Contents