Juvenile Out- of- Home Placement

Juvenile Out-of-Home Placement

January 1999

A PROGRAM EVALUATION REPORT

Office of the Legislative Auditor State of Minnesota

Table of Contents
SUMMARY
INTRODUCTION
BACKGROUND
Types of Placements Types of Residential Settings Recent Law Changes
2 PLACEMENT DECISIONS
Placement Criteria County Review of Placement Recommendations Screening, Assessment, and Case Planning Adequacy of Information About Facilities Court Justifications for Placements Out-of-State Placement Placement Considerations Summary
3 PLACEMENT SPENDING AND FUNDING
Data Sources Statewide Spending Statewide Trends County Variation Funding Family Preservation Spending Summary
4 CHARACTERISTICS OF CHILDREN IN PLACEMENT Reasons for Placement Demographic Characteristics Distance Placed From Home Number and Length of Placements Accuracy of State Placement Information Summary
SERVICE NEEDS Non-Residential Service Needs Bed Availability Service Adequacy Program Completion Outcome Measurement Summary
6 DISCUSSION AND RECOMMENDATIONS State Funding Assistance Facility Moratorium Placement Decision Making, Case Management, and Accountability Information Systems Other
AGENCY RESPONSES 111 RECENT PROGRAM EVALUATIONS

vi JUVENILE OUT-OF-HOME PLACEMENT

List of Tables and Figures

Page Table 1 County Officials’ Perceptions About Whether Judicial

Placement Decisions Adequately Consider Key Information xi Table 2 Juvenile Placements and Spending, By Type of Facility, 1997 xii Table 3 Reasons for Out-of-Home Placements, 1997 xv Table 4 Distance of Placements From Home, By Facility Type, 1997 xvi Table 5 Perceived Need for Additional Juvenile Facility Beds xviii Table 1.1 Number of Licensed Beds and 1997 Days of Care in Minnesota Facilities for Juveniles 10 Table 2.1 County Perceptions About the Need for Clearer Placement Criteria 20 Table 2.2 County Officials’ Perceptions About Whether Judicial

Placement Decisions Adequately Consider Key Information 23 Table 2.3 Judges’ Perceptions About Adequacy of County Assessments 24 Table 2.4 County Officials’ Perceptions About the Adequacy of ProgramInformation Available When Making Placement Decisions 27 Table 2.5 Use of Out-of-State Placement, by Minnesota Judicial District, Fiscal Years 1996-97 31 Table 2.6 ‘‘Important Justifications ’’ for Child Placements, According to County Officials 35 Table 3.1 Juvenile Out-of-Home Placement Spending by Type of Facility, 1997 40 Table 3.2 Change in Juvenile Out-of-Home Placement Spending by Type of Facility, 1992-97 41 Table 3.3 1997 Spending for Juvenile Out-of-Home Placements, Selected Counties 43 Table 3.4 Characteristics of 1997 Juvenile Out-of-Home Placements in High, Medium, and Low Spending Counties 43 Table 3.5 Changes in Out-of-Home Placements for Counties with Large, Medium, and Small Spending Increases, 1992-97 46 Table 3.6 Inflation-Adjusted Percentage Change in Out-of-Home Placement Spending per Resident Under Age 18, by Facility Type, 1992-97 47 Table 3.7 Poverty Rate and Placement Spending by Judicial District 48 Table 3.8 1997 Out-of-Home Placement Spending Funded by County Dollars, Selected Counties 53 Table 4.1 Reasons for Out-of-Home Placements, 1997 59 Table 4.2 Percentage of Children in Placement and Average Length of Placement, By Racial and Ethnic Groups, 1997 60 Table 4.3 1997 Days of Care Per Child in Population, By Racial/ Ethnic Category and Region 61

Page Table 4.4 Reasons for Placement, by Racial and Ethnic Groups 62 Table 4.5 Ages of Children in Placement During 1997, by Facility Type 63 Table 4.6 Distance of Placements From Home, by Facility Type, 1997 Table 4.7 Total Time That Individual Children Spent in Placement, 1995-97 66 Table 5.1 County Satisfaction With Availability of Juvenile Services 73 Table 5.2 Judges’ Opinions About the Potential for Reducing Placements Through Improved Non-Residential Services 74 Table 5.3 Number of Rule 8, Rule 5, and Corrections Beds by Judicial District 77 Table 5.4 Statewide Occupancy Rates for Selected Types of Residential Facilities 77 Table 5.5 Estimated Occupancy Rates for Selected Facilities by Judicial District 78 Table 5.6 County Officials’ Perceptions About Recent Trends in Placement Availability, by Facility Type 80 Table 5.7 Perceived Need for Additional Juvenile Facility Beds 81 Table 5.8 County Perceptions About Length of Existing Residential Programs 83 Table 5.9 County Perceptions About Adaptability of Facilities to Children’s Needs 84 Table 5.10 County Perceptions About Facilities’ Efforts to Work with Families 85 Table 5.11 Perceptions About Availability of Aftercare Services 87 Table 5.12 Perceptions About Cultural Sensitivity of Residential Programs, Selected Counties 88 Table 5.13 County Perceptions About Adequacy of Funding for Juvenile Services 89 Figure 1 Extent to Which Judges Clearly Specify Placement Goals, According to County Officials xxi Figure 1.1 Definitions of Delinquent Children and Children in Need of Protection or Services 7 Figure 2.1 County Staff Perceptions About Judicial Consistency in Placement Decisions 20 Figure 2.2 Judges’ Opinions About Whether Minnesota Has Adequate Placement Options For Children With the Most Serious Problems 33 Figure 3.1 1997 Out-of-Home Placement Spending Per Resident Under Age 18, by County 44 Figure 3.2 Minnesota Judicial Districts 48 Figure 3.3 Selected County Strategies for Reducing Placement Costs 51 Figure 3.4 1997 Out-of-Home Placement Funding by Source 52 Figure 5.1 Extent to Which Judges Clearly Specify Placement Goals, According to County Officials 97 Figure 6.1 County Officials’ Opinions About Whether to Extend the Moratorium on Facility Licensure 102

Minnesota Office of the Legislative Auditor Juvenile Out-of-Home Placement
SUMMARY

Each year, thousands of Minnesota children are placed away from their homes for care and treatment. For example, victims of abuse or neglect might be placed with a foster family until they can safely return home or enter other ‘‘permanent’’ living arrangements. Some delinquent juveniles are sent to correctional facilities that aim to hold offenders accountable for their actions, protect the public, or provide therapeutic programs. Juveniles with emotional disturbances or serious drug or alcohol problems may be sent to residential programs that offer treatment.

In recent years, counties have worried about the impact of out-of-home placement costs on their budgets, and state legislators have questioned whether placement facilities adequately serve Minnesota juveniles. The 1998 Legislature placed a temporary moratorium on the development of large new facilities (and large expansions of existing facilities). The Legislature also requested this evaluation, and we asked the following questions:

How do counties and courts decide when to make placements? Is there adequate screening and assessment of children?

What is the total public cost of out-of-home placements in Minnesota, and what trends in placements and spending have occurred? To what extent do counties vary in their placement spending, and what are the reasons for the variation? What strategies have counties used to control placement spending?

What are the characteristics of the children served in out-of-home placements? In what types of facilities are they placed, for how long, and how far away from home? To what extent do children complete the programs they enter, and what are the reasons for non-completion?

Does Minnesota need more beds to serve children placed out of home? If so, what types of beds (or supportive services) does the state need? Does Minnesota have sufficient non-residential alternatives to placement and aftercare services following residential placements?

Is there sufficient information on the performance and operation of Minnesota’s out-of-home placement system?

To answer these questions, we analyzed existing statewide information on child placements and their costs. We also surveyed county corrections supervisors, human services directors, and district court judges throughout the state. We visited seven counties, reviewed case information for more than 250 individual juveniles, and interviewed numerous state, local, and facility staff. We examined placements at various types of residential facilities licensed by the Department of Human Services or Department of Corrections, including family foster homes, ‘‘Rule 5’’ mental health treatment facilities, ‘‘Rule 8’’ group homes licensed to serve ten or fewer residents, chemical dependency treatment facilties, child shelters, detention facilities, and correctional facilities for delinquent juveniles.

Overall, we conclude that Minnesota generally has a more pressing need for additional non-residential services for its juveniles than additional residential services. Minnesota does not appear to face significant statewide shortages of beds (with the possible exception of foster care), although the services in existing residential facilities do not always adequately address the needs of juveniles in placement. Unfortunately, Minnesota has little information on the effectiveness of services for juveniles, and we think that the Legislature and state agencies should take steps to improve information on service outcomes.

PLACEMENT DECISIONS

There are several ways that Minnesota children can be placed in publicly-funded out-of-home care. First, peace officers may temporarily place a child in detention or shelter care. State law requires the court to hold hearings within 72 hours to determine whether the child should remain in custody. Second, the courts may order placement of a child who has been found by the court to be delinquent or in need of protection or services. Third, parents or guardians may enter an agreement with a local social services agency to ‘‘voluntarily’’ place a child--often when the agency is considering asking the court to remove the child from home. County social services and corrections agencies play key roles in selecting placement options, assessing child needs, and advising the juvenile courts.

Court-ordered placements are the most common type of out-of-home placements among Minnesota children. State law requires the courts to articulate in writing the reasons for child placement and for rejecting other possible options, but our review of individual cases suggests that the courts often have not explained their actions thoroughly or clearly. Furthermore, our surveys indicated that:

Sixty-two percent of county human services directors and 32 percent of county corrections supervisors told us that judges were not usually consistent in their decisions about which circumstances justify placement.

There are a variety of possible reasons for inconsistency in the child placement process, both within the courts and within the county agencies that help make placement decisions. First, most counties and judicial districts do not have written criteria that identify specific circumstances that justify out-of-home placement. 1 This may reflect the lack of consensus about which types of children benefit from out-of-home placement. Second, some Minnesota counties involve few county staff in placement decisions. For example, only about half of Minnesota’s county social services agencies have multi-disciplinary ‘‘juvenile treatment screening teams’’ authorized by state law ( Minn. Stat. §260.152, subd. 3) to review cases recommended for placement. In addition, officials from only half of all county corrections agencies told us that their agencies typically involve at least one supervisor or manager in placement decisions.

Minnesota has greater need for additional non-residential rather than residential services.

In addition, we concluded that practices for assessing children prior to placement could be improved. A majority of county officials surveyed told us that judicial placement decisions are generally based on sufficient consideration of children’s needs, but more than one-third of the human services directors said that children’s needs are not considered sufficiently (see Table 1). In addition, state chemical dependency staff told us that many adolescents in residential corrections and mental health facilities have chemical abuse problems that have not been treated. Most judges we surveyed gave high marks to the timeliness and thoroughness of county chemical dependency assessments, and they gave somewhat lower ratings to counties’ assessments of juveniles’ mental health. We were unable to examine the outcomes of mental health screening in a systematic way because few counties have complied with state requirements for reporting this information annually. Finally, assessments of juvenile offenders’ risk of committing new offenses can help the courts and counties determine what services should be provided, but nearly half of Minnesota counties do not formally do such assessments.

To make consistent, appropriate placement decisions, counties and courts not only need information about the child and family, but they also need to understand the strengths and weaknesses of service options available. In our surveys, a large majority of county staff told us that they have sufficient information describing the services offered by various programs. But less than half of the officials we surveyed said they have enough information about (1) recurrence of juveniles’ problems following residential placement, (2) the extent to which children run away from placement facilities, and (3) the extent to which children ‘‘complete’’ the programs in which they are placed.

Table 1: County Officials’ Perceptions About Whether Judicial Placement Decisions Adequately Consider Key Information

Corrections Supervisors (N = 82) Human Services Directors (N = 84) Sometimes, Sometimes, Do judges make dispositions based Usually or Rarely, or Usually or Rarely, or on sufficient consideration of: Always Never Always Never Children’s mental health needs? 83% 13% 62% 37% Children’s chemical dependency problems? 87 10 61 36 Children’s cultural and ethnic backgrounds? 71 21 55 39 Facilities’ ability to meet children’s service needs? 87 10 54 45

NOTE: Percentages of officials who responded "don’t know" are not shown. SOURCE: Program Evaluation Division surveys, August 1998.

Screening and assessment practices could be improved.
SUMMARY xi

1 For instance, the 1994 Legislature directed judicial districts to develop disposition criteria for delinquency cases, but seven of the ten districts identified factors to consider in the case disposi tion process rather than specific guidelines indicating when out-of-home placement might be appropriate.

PLACEMENT SPENDING

Using information we collected from state data sources and surveys of local agencies, we estimate that:

Public agencies spent $225 million in 1997 for children placed out of home, not including education costs.

Table 2 shows spending, placements, and days of service for major categories of residential settings. Foster homes have relatively low average costs per day ($ 35), but they accounted for a third of all Minnesota placement spending because children tend to stay in foster homes for long periods. Correctional facilities accounted for 26 percent of all placement spending, ranging from numerous short-term placements in juvenile detention centers to longer-term placements intended to hold juvenile offenders accountable for their actions. On average, Rule 5 mental health treatment facilities had relatively long stays (168 days) and high costs per day ($ 179), so they accounted for 21 percent of statewide placement spending despite having only 4 percent of the placements.

Table 2: Juvenile Placements and Spending, By Type of Facility, 1997

Percent of Placements

Made in 1997

Estimated Average Length of Stay (Days) a

Percent of Days of Care

Occurring in 1997

Average Cost Per

Day Percent of

Total 1997 Placement

Spending Shelters 21% 25 6% $89 7% Family foster homes 21 285 66 35 34 Rule 8 group homes 4 119 5 99 8 Rule 5 facilities b 4 168 8 179 21 Correctional facilities 45 28 14 129 26 Chemical dependency facilities 4 40 2 135 3 Other c 1 1

Total 100% 100% 100% a 1997 days of care divided by 1997 placements. b Includes placements at Brainerd and Willmar regional treatment centers paid by Medicaid. c Includes placements and days of care that we could not allocate to the categories shown.

SOURCE: Program Evaluation Division analysis of data from the departments of Human Services and Cor rections and June-July 1998 survey of counties.

Placements cost taxpayers more than $200 million annually. xii JUVENILE OUT-OF-HOME PLACEMENT

In 1997, per capita spending for out-of-home placement varied widely among counties, ranging from $25 per county resident under age 18 (Red Lake County) to $322 (Hennepin County). Likewise, placement spending per child in poverty ranged from $156 to $1,954 among Minnesota counties. We found that the group of counties with the highest per capita spending placed more children, for much longer periods, and for slightly higher costs per day than the group of the lowest spending counties. In addition, we found that high spending counties (as a group) had substantially higher spending per capita in each of the six categories of facilities we examined. 2

The widespread variation in placement spending appears to reflect county differences in underlying social conditions as well as placement policies and practices. For example, we found that counties with high poverty rates tended to have high levels of placement spending. But we also found that some counties with very low placement rates have (1) procedures for closely scrutinizing placement recommendations and children already in placement, and (2) strong preferences for using community-based services rather than out-of-home placement.

Counties have expressed concerns about recent growth in placement costs, and we found that:

Adjusted for inflation, statewide placement spending per Minnesota resident under age 18 increased 22 percent between 1992 and 1997.

The highest rates of increase in inflation-adjusted spending were in correctional and chemical dependency facilities, which rose 39 and 37 percent, respectively. There were lower rates of increase in family foster homes (14 percent), Rule 8 group homes (21 percent), and Rule 5 mental health treatment facilities (26 percent).

Among 34 counties that spent more than $1 million for placement in 1997, we found considerable variation in 1992-97 placement trends. At one extreme, Pine County’s inflation-adjusted placement spending increased 126 percent during this period; on the other hand, St. Louis County’s spending decreased 19 percent. As a group, counties with the largest overall spending increases between 1992 and 1997 had above-average increases in spending in all categories of juvenile residential facilities, not just some categories.

Seventy-six percent of local human services directors and 54 percent of corrections supervisors told us that they expect placement spending in their counties to increase faster than inflation during the next three years. However, about half of the local officials told us that there are additional steps that their counties could take to control placement costs without sacrificing service quality.

Placement spending varies widely among counties.

Statewide, placement spending has grown faster than inflation. SUMMARY xiii

2 The group of high spending counties included the 28 counties with the highest placement spending per county resident under 18; the low spending counties included the 28 counties with the lowest per capita spending.

FUNDING SOURCES

About three-fourths of Minnesota’s 1997 placement costs were paid from the budgets of local social services agencies. For these expenditures, property taxes were the main county revenue source, although counties also received general purpose aid from the state. The main sources of federal funding were Title IV-E of the Social Security Act, which paid for a portion of placements at certain types of residential facilities, and a social services block grant funded under Title XX of the Social Security Act. The largest source of state funding was Minnesota’s Community Social Services Act block grant. For all 1997 placements paid for by local social services agencies, we estimate that:

County revenues paid for 59 percent of 1997 placement costs, while federal revenues paid for 20 percent and state revenues paid for 12 percent.

Compared with other states, Minnesota relies much more on local revenues and less on state revenues to pay for social services (including child placement costs).

Funding sources for out-of-home placement vary considerably among counties. For example, the percentage of 1997 spending paid for by county revenues varied from 33 percent (Clearwater County) to 79 percent (McLeod County). Such variation likely reflects differences in counties’ (1) eligibility for (and possibly pursuit of) federal funds, (2) use of facilities eligible for federal reimbursement, and (3) overall levels of placement spending.

About half of the county corrections supervisors and human services directors told us that budget considerations have limited their ability to provide the care and services that children need. When asked whether counties would likely place more children out of home if state or federal funds paid for a larger proportion of placement costs, most human services directors (63 percent) said they would not, while county correctional supervisors were evenly split.

CHARACTERISTICS OF CHILDREN IN PLACEMENT

Children are placed away from home for a variety of reasons. Based on an analysis of all types of Minnesota child placements in 1997, we found that:

Children spent more time in out-of-home placement due to their parents’ actions than their own conduct.

As shown in Table 3, 46 percent of the time children spent in out-of-home care resulted from parents’ conduct, and other parent-related reasons accounted for another 12 percent. A large majority of foster care placements resulted from parent-related reasons, while most delinquency, chemical dependency, mental health treatment, and group home placements resulted from child-related reasons.

Counties pay for most placement costs, and many said that budget considerations have affected service decisions.

Children in out-of-home placement range in age from infants to adolescents. In 1997, nearly half of Minnesota children in family foster homes were under age 10, while correctional facilities, group homes, chemical dependency facilities, and Rule 5 mental health treatment facilities generally served older children. Boys outnumbered girls in all categories of residential facilities, but especially in correctional and Rule 5 facilities.

We found dramatic differences in rates of child placement among various racial and ethnic groups. In particular,

African American and American Indian children had disproportionately high rates of out-of-home placement, compared with children in other racial/ ethnic groups.

Only 4 percent of Minnesota children are African American, but African American children accounted for 22 percent of all Minnesota children in placement in 1997. Likewise, only 2 percent of Minnesota children are American Indian, but American Indians accounted for 12 percent of 1997 children in placement. About 8 percent of Minnesota’s African American and American Indian children were in out-of-home placement at some time during 1997, compared with 1 percent of Minnesota’s white, non-Hispanic children. In addition, African American and American Indian children had longer placements, on average, than white children. Also, African American and American Indian children had at least 12 times as many days in placement per capita in 1997 due to parent-related reasons as did white, non-Hispanic children in placement.

Table 3: Reasons for Out-of-Home Placements, 1997

Reason Percentage of Days of Care Parent misconduct 46.2%

(Child neglect/ abuse, child abandonment, parental substance abuse, incarceration, other) Other parent-related reasons 12.2

(Disability, temporary absence, other) Child misconduct 30.3

(Delinquency, status offenses, substance abuse, behavior problems) Other child-related reasons 5.9

(Disability, other) Family interaction problems 5.4

TOTAL 100.0% NOTE: For placements funded by social services agencies, counties regularly report reasons for placement. For correctional and chemical dependency placements not funded by social services, we assumed that the reason for placement was child conduct.

SOURCE: Program Evaluation Division analysis of DHS placement data, June and July surveys of counties, DHS Consolidated Chemical Dependency Treatment Fund data.

Parent-related reasons account for most days of child out-of-home placement.

Among children who were in a placement that lasted for more than three days during 1995-97, 45 percent had multiple placements of this length during this period. Of children who were in placement on January 1, 1995, 23 percent remained in placement continuously for at least the next three years.

LOCATION OF PLACEMENTS

By definition, children in out-of-home placement live apart from their immediate families. Legislators have questioned whether some children are placed too far from home, making it more difficult for service providers to work with the child’s family and help children successfully return to their home communities. Table 4 shows the distance placed from home, by category of residential facility. We found that:

Statewide, 62 percent of days that children spent in placement during 1997 were at facilities in the children’s home counties. Eight percent of days in placement were at Minnesota facilities more than 100 miles from home, and five percent of days in placement were in facilities in other states.

Juveniles placed in shelters and foster care usually remained in their home counties, but those placed in mental health treatment facilities, group homes, and correctional facilities were more commonly in distant counties.

Table 4: Distance of Placements From Home, By Facility Type, 1997

Percentage of 1997 Days of Care Spent in Placements: Facility Type

Total Days in Care in

1997 Within

Same County

In Border County

In State and Within 100 Miles, But Not in Same or Border

County In State,

But More than 100 Miles Away

In Another

State TOTAL Shelters 183,143 79.8% 13.3% 4.7% 2.2% 0.0% 100% Family foster homes 2,086,280 71.6 13.5 6.5 4.9 3.4 100 Rule 8 group homes 166,279 25.8 26.4 23.3 20.7 3.8 100 Rule 5 facilities 228,579 12.9 17.0 31.1 25.2 13.8 100 Correctional facilities 436,031 45.2 12.2 19.3 13.7 9.7 100

TOTAL 3,100,312 61.6% 14.3% 10.9% 8.3% 4.9% 100% NOTE: For placements in correctional detention facilities that were not paid for by social services agencies, we assumed that the placements occurred in the juvenile’s home county. This is usually true, but we did not have case-s pecific information on the location of these placements.

SOURCE: Program Evaluation Division analysis of Department of Human Services, county placement data , Program Evaluation Division June and July 1998 surveys of counties, and Department of Corrections data on Red Wing and Sauk Centre placements.

Most cases of out-of-state placement have involved children placed in (1) foster homes, or (2) facilities certified by Minnesota’s Department of Corrections (DOC) to serve delinquent juveniles. 3 Among foster care cases, most out-of-state placements have involved Minnesota children who are living with relatives in other states. For delinquent juveniles, counties have used out-of-state facilities for a variety of reasons: for programs that are longer or address specialized needs better than those available in Minnesota; for lower costs; to discourage juveniles from running away; and because out-of-state facilities are closer than in-state facilities for some counties. Two counties (Ramsey and Hennepin) accounted for two-thirds of all 1996-97 out-of-state placements at DOC-certified facilities, and Ramsey County had far more children per capita in such out-of-state placement than other judicial districts in Minnesota. The states whose facilities were used most often for delinquent juveniles were South Dakota, Iowa, and Colorado.

NEED FOR ADDITIONAL SERVICES

It is difficult to assess the need for additional residential services without considering the availability of placement alternatives. In some cases, it might be possible to avoid (or shorten) residential placements if there are appropriate non-residential programs in the juvenile’s home community. In our surveys of county officials, we found that:

Most counties reported that they have a greater need for additional non-residential services for juveniles than additional residential services.

Seventy-one percent of county corrections supervisors and 64 percent of county human services directors said that non-residential services would be a higher spending priority than residential services if additional funds were available. In addition, we asked counties to assess their satisfaction with 25 categories of services, and most counties expressing dissatisfaction with particular services said that their most pressing need in these categories was for non-residential services. Human services directors and corrections officials both identified truancy services as the category of service with which they were least satisfied.

Judges, county human services directors, and county corrections supervisors told us that some out-of-home placements could be avoided with improved non-residential services. For example, more than one-third of judges said there is ‘‘significant potential’’ to reduce placements of truants, runaways, and misdemeanor-level offenders through non-residential services, and a majority of judges said that there is at least ‘‘some potential’’ for placement reductions in all categories of juveniles except ‘‘extended jurisdiction juveniles’’ and felony-level violent offenders. 4

Some out-of-home placements could be avoided.

3 Under state law, the commissioner of corrections must ‘‘certify’’ that out-of-state facilities meet Minnesota facility standards before delinquent juveniles (or preadjudicated delinquents) from Minnesota can be placed there. Many of the DOC-certified facilities are correctional facilities, bu t some are comparable to Minnesota’s Rule 5 and Rule 8 facilities.

To help us assess the need for additional beds in residential facilities, we examined occupancy rates in selected categories of facilities. Although counties sometimes have difficulty finding available beds, we found a considerable amount of unused capacity in several categories of residential facilities. Statewide, we found that 88 percent of beds in secure correctional detention and residential facilities were occupied, compared with 77 percent of non-secure correctional beds (detention and residential), 67 percent of Rule 8 group home beds, and 65 percent of Rule 5 mental health treatment facility beds. 5

We also surveyed county officials about service needs and, as shown in Table 5, we found that:

The greatest need for more beds is in foster care, according to county human services directors, and secure residential facilities, according to county corrections supervisors.

Table 5: Perceived Need for Additional Juvenile Facility Beds

Percentage of Officials Who Said There Is: No Need for New

Beds Some Need

for New Beds

Significant Need for New Beds

Human services directors (N= 84): Shelter care 25% 67% 6% Treatment foster care 15 54 29 Regular foster care 5 46 48 Relative foster homes 14 40 40 Group homes 45 40 6 Rule 5 mental health facilities 52 38 6 Corrections supervisors (N= 82): Secure detention facilities 33% 50% 15% Secure residential facilities 28 45 26 Non-secure correctional facilities 35 51 10 Group homes 43 49 6

NOTE: Percentages of respondents who said "don’t know/ not applicable" are not shown. SOURCE: Program Evaluation Division surveys, August 1998.

Many counties perceive a need for additional foster care. Some types of facilities have a considerable amount of unused capacity.

4 ‘‘Extended jurisdiction juveniles’’ are felony offenders for whom the court has executed a juvenile disposition, along with a stayed adult sentence. The court can maintain jurisdiction over these offenders until they reach age 21, and the adult sanctions may be executed if the offender commits a new offense or violates the conditions of the stayed sentence.

5 We estimated that 45 percent of Minnesota’s licensed family foster home beds were occupied in Fall 1998, but human services officials told us that it would be unrealistic for many foster homes t o serve the maximum number of children allowed by their licenses.

The 1994 Legislature authorized construction grants for secure detention and secure residential beds in all of the state’s judicial districts, and this has helped to address the need for more secure juvenile correctional facilities. Fifty-nine percent of corrections supervisors told us that the availability of secure detention beds in Minnesota improved in the past three years, and 45 percent said that the availability of secure residential (post-disposition) beds improved. Some judicial districts are still constructing or planning their new secure correctional facilities, so reductions in occupancy rates are likely.

Overall, given the relatively low occupancy rates of many facilities, the ongoing construction of additional juvenile corrections beds, and the preference of many counties for improved non-residential services, we concluded that:

There is not a serious statewide shortage of residential beds for juveniles, with the possible exception of foster care.

Individual counties may have occasional difficulties finding specialized residential services, such as correctional services for juvenile offenders with low intelligence or secure correctional beds, and sometimes they cannot find an immediate vacancy in a preferred facility. For example, 50 percent of judges told us that there are not usually sufficient residential options for the ‘‘children with the most serious problems.’’ But, in general, we think that Minnesota’s total number of residential beds is adequate (or nearly adequate), and the beds are distributed quite evenly throughout the state.

SERVICE ADEQUACY

Even if Minnesota has enough of most types of residential beds, it is important to consider whether existing facilities effectively serve children’s needs. Our surveys of county officials revealed various concerns about the adequacy of existing services:

Corrections and human services officials were less satisfied with the availability of short-term placement options (less than three months) than with the availability of longer-term options.

Human services officials cited group homes and corrections officials cited correctional facilities as the types of residential facilities that least adequately tailor services to meet juveniles’ needs.

County corrections and human services officials said that residential corrections facilities have not worked with families of the children they serve as well as other types of juvenile facilities.

In all categories of facilities, county corrections staff reported a need for improved ‘‘aftercare’’ services following residential placements; human services directors said that aftercare services are least adequate following placements in correctional and chemical dependency programs. Counties identified various ways that residential and aftercare services could address children’s needs more effectively.

Among counties in which minority groups comprise at least 5 percent of the population, more than one-third of human services directors and correctional supervisors said that residential facilities are often insensitive to cultural and ethnic differences in the children they serve.

Fifty-five percent of human services directors and 35 percent of corrections supervisors said that residential facilities discharge too many children for violating facility rules.

Staff with the departments of Human Services and Corrections told us that improvements in facility and aftercare services are likely to result from proposed facility rules drafted jointly in 1998, at the direction of the Legislature. In addition, the departments have taken steps recently to foster development of more responsive services--for example, through the revision of programs at the Red Wing correctional facility and the encouragement of more community-based mental health programming.

It would be useful to know more about the effectiveness of Minnesota’s residential programs for juveniles. In recent years, some counties have reduced the length of time that children remain in placement, others have made fewer referrals to residential mental health treatment facilities, and some counties have diverted to non-residential services juveniles who previously might have been placed out of home. For the most part, the results of these changes are unclear. Only 7 percent of county human services directors and corrections supervisors told us that their agencies produced summary information during the past year on the success of children subsequent to out-of-home placements. The Department of Human Services worked with counties during 1998 to identify child welfare performance measures that could be tracked in the future, and this was an encouraging first step.

Many county human services directors and corrections supervisors said that judges often do not clearly specify the intended purpose of the placements they make.

STATEWIDE INFORMATION ON CHILD PLACEMENTS

Not only is there insufficient information on the outcomes of child placements, but there is also incomplete information about the placements, their costs, and the

Few counties have systematically tracked the outcomes of juvenile placements.

But, to properly measure service outcomes, it is necessary to consider the goals of each child placement. For example, depending on a child’s circumstances, the desired outcomes of placements might include law-abiding behavior, sobriety, placement in a permanent home, protection from maltreatment, or other goals. Unfortunately, as shown in Figure 1, characteristics of the children in placement. The Minnesota Department of Human Services (DHS) collects and analyzes information on placements paid for by county social services agencies, but some child placements are paid for by other local agencies. For example, DHS has little information on children placed at county-operated ‘‘home schools’’ for juvenile offenders, including some of Minnesota’s largest juvenile facilities--such as the Hennepin County Home School and Ramsey County’s Boys Totem Town. Also, there has been limited statewide information collected on individuals in juvenile detention because most detention placements are funded by local corrections or law enforcement agencies. It is understandable that DHS has focused its data collection efforts on services paid for by social services agencies, but the information missing from this database has limited its usefulness to policy makers.

In addition, we found various problems with the accuracy of DHS’ county reported data on child placement. The problems included missing cases, inaccurately reported placement discharge dates, duplicate placements, and single placements inaccurately reported as multiple placements. If uncorrected, these problems can result in inaccurate information on individual counties’ number of placements, days of care, and average placement length. For example, among eight counties that we examined in detail, we estimated that DHS overstated the actual days of care by at least 20 percent for four counties in 1996. Given the recent interest of the Legislature, DHS, and counties in tracking the length and outcomes of child placements, we think it will be important for DHS to monitor and correct the types of problems we found. We saw evidence that the department was doing a better job of this in 1997, although some problems remained. In addition, the department is implementing a new information system that is designed to improve the accuracy of placement information in the future.

Figure 1: Extent to Which Judges Clearly Specify Placement Goals, According to County Officials

7 25 36 31 18 29 35 16 0 10 20 30 40 50

Always or almost always

Usually Sometimes Rarely or never Human services directors Corrections Supervisors

NOTE: The survey question was: “In court dispositions that involve child placements, do judges clearly specify the intended purpose of each placement?”

SOURCE: Program Evaluation Division survey of county human services directors (N= 84) and corrections supervisors (N= 82), August 1998.

Percent of county officials

Policy makers need better information about out-of-home placements.

RECOMMENDATIONS

Historically, the state’s role in child placement has been very limited. The courts and counties have considerable discretion about whom to place, and county funds have paid for the largest share of placement costs. State agencies license the facilities in which children are placed, but it is up to the courts and counties to select the facilities that best address children’s needs. We think there are steps the Legislature and state agencies could take to improve Minnesota’s child placement system, while preserving the important roles played by the courts and counties.

We have no recommendation regarding the proper level of state funding for out-of-home placement or child welfare services. On the one hand, counties have considerable discretion about which children to recommend for placement, so a significant local role in placement funding may encourage better decisions and closer scrutiny. Also, some analysts have suggested that Minnesota has not made maximum use of non-state revenue sources for out-of-home placement, such as federal funding and parental fees. On the other hand, a larger state funding role might be justified by (1) inadequacies in some residential and non-residential services, (2) the inability of counties to fully control costs for placements that are often made by the courts, and (3) the burdens that placement costs impose on poor counties, due to the fact that placement and poverty rates are positively related. The 1998 Legislature authorized $30 million in state family preservation aid in 2000--largely in response to county concerns about growing out-of-home placement costs.

There are a variety of ways that the Legislature could allocate state funding for children in placement or at risk of placement, and we did not conduct an in-depth analysis of alternative measures of county need and fiscal capacity. However, we think that a funding approach that is tied too closely to out-of-home placements could create incentives for placement or penalize counties that have invested in placement alternatives. Thus, we recommend:

The Legislature should not allocate funds to counties based solely or largely on their historical number of out-of-home placements (or levels of placement spending). 6

Likewise, as a general rule, we think that the Legislature should not restrict the use of funds to reimbursement of out-of-home placement costs. Counties expressed a strong desire in our surveys for improved non-residential services. In our view, counties should have the flexibility to use funds to pay for whatever residential or non-residential services will best serve the needs of children and families.

Although we did not find that Minnesota needs large numbers of new beds in residential facilities, we recommend that:

The state’s role in out-of-home placement has been very limited.

6 If the Legislature wants to use historical county spending levels as a measure of service need in a funding formula, we think that it should use a measure of aggregate spending for both residential and non-residential services.

The Legislature should not extend the moratorium on large, new residential facilities (or facility expansions), which is scheduled to expire in mid-1999.

For several reasons, we think that an extension of the moratorium could unfairly constrain placement options for counties (which pay for most placement costs) and courts (which are responsible for making case dispositions that serve the child and protect public safety). First, counties expressed to us some concerns about the quality and cost of residential services now available in Minnesota facilities, and we think that it is important for counties to have a variety of placement options. A moratorium might protect existing facilities from new competition and make them less responsive to the needs of counties and courts. We think that counties and courts are in the best position to judge whether to place their children in new or long-standing facilities, or in large or small facilities--so long as the facilities meet basic licensing requirements that help to ensure quality service. Second, the moratorium was adopted in 1998 largely in response to concerns about additions of correctional beds, but our survey of county corrections officials indicated that more would oppose rather than favor an extension of the moratorium. 7 Third, while some people believe that counties and courts will fill to capacity whatever number of beds Minnesota licenses, this is not currently the case. There are many vacant beds in juvenile residential facilities, and counties have increasingly looked for alternatives to expensive, long-term residential placements. Overall, we do not think that Minnesota has a significant shortage of residential beds for juveniles, but we think that a moratorium could limit the responsiveness of service providers to juveniles’ needs.

An alternative to a moratorium might be a requirement for facilities to demonstrate to state licensing officials that they are ‘‘needed,’’ prior to receiving a license. Some people expressed concerns to us that Minnesota communities may encourage development of new facilities as a way of luring jobs and redevelopment, without sufficient consideration of how these facilities would address the needs of Minnesota children. We share this concern, although we think that it would probably be best to let counties and courts determine which facilities are ‘‘needed’’ through their actual placements, rather than having state regulators try to evaluate the ‘‘need’’ for a facility before it has opened. 8

To address the problem of inconsistency in placement decisions, we considered whether to recommend statewide or county placement criteria that would identify circumstances that justify child placement. However, counties expressed limited support for such a state requirement, and research literature has provided limited insight into which types of children fare best in out-of-home care. As an alternative means of ensuring more consistent, thoughtful decisions on child placement, we recommend:

Minnesota does not need significantly more residential facilities, but a moratorium might limit placement options for counties that pay for most placement costs.

7 Human services directors tended to favor an extension of the moratorium, but DHS has not received a proposal for a facility large enough to be subject to the moratorium for more than 25 years.

8 The ‘‘need’’ for a facility may be difficult to evaluate before it begins to offer services. For example, a facility might be needed if it provides services that other facilities do not, but it mig ht also be needed if it provides duplicative services more effectively or at lower costs than other providers.

The Legislature should require all counties to establish multi-disciplinary juvenile screening teams.

Presently, these screening teams are optional. We think that multi-disciplinary teams should review all placements in treatment facilities and all court-ordered placements potentially exceeding 30 days--including post-dispositional placements in facilities licensed by the Commissioner of Corrections. In our view, an expanded role for juvenile screening teams will enhance accountability, while helping to ensure that juvenile service needs are identified.

In addition, risk assessment (and corresponding needs assessment) can help agencies decide which juveniles need the most attention, and it can also help them to develop service plans. Similar to state requirements for adult offenders, we recommend that:

The Legislature should require each county corrections or court services agency to adopt written policies for classifying the risks and service needs of juvenile offenders.

There is little systematic monitoring of service outcomes for juveniles in placement, partly because the goals of these placements vary widely and are not always well articulated. To supplement the individualized case planning done by counties and service providers, we recommend:

The Legislature should require courts to state in their dispositions the intended outcome( s) of each juvenile placement made under their authority. The Legislature should establish a working group of judicial, legislative, and executive branch representatives to (1) develop a uniform list of possible placement goals from which judges would select, and (2) identify steps (and related costs) required for state agencies to collect summary information on achievement of these goals.

After this working group completes its tasks,

The Legislature should require the departments of Human Services and Corrections to regularly report statewide information on the extent to which the goals of court-ordered placements are met--based on their own analyses or on summaries of information provided by counties or residential facilities.

Screening teams should review lengthy placements.

The courts should clearly specify placement goals.

Many county staff expressed concerns about the adequacy of services for juveniles in placement (and following placement). For example, they cited a need for facility staff to work more effectively with the families of juveniles, and they said they would like better ‘‘aftercare’’ services. We think it is reasonable to expect counties to help develop plans to ensure that these types of services are provided, and many counties do this now. In fact, proposed rules recently drafted by the departments of Corrections and Human Services refer to county ‘‘case plans’’ and ‘‘transitional services plans’’ for each juvenile in certain types of placement. However, current law does not require counties to develop case plans for delinquent juveniles, and the law does not clearly indicate whether counties are responsible for monitoring aftercare services identified in the transitional services plans. We recommend:

The Legislature should require counties to develop juvenile case plans following delinquency dispositions. Consistent with requirements for cases involving children in need of protection or services, the plans should identify any social and other services that will be provided to the child and child’s family, whether in residential or non-residential settings.

To help ensure that juveniles receive the services they need following placements in residential facilities, we recommend:

The Legislature should clarify in law that counties are responsible for monitoring implementation of ‘‘transitional services plans,’’ even if aftercare services are provided by the residential facilities or other providers.

In addition, county human services agencies expressed concerns to us about the absence of clear definitions of ‘‘treatment foster care’’--that is, foster homes that provide in-home therapeutic services. We recommend:

The Department of Human Services should adopt state rules that outline the components of treatment foster care.

There are very high rates of child placement among certain racial and ethnic groups, and many county officials told us that there is room for improvement in residential programs’ sensitivity to cultural differences. Proposed rules drafted by the departments of Human Services and Corrections would require residential facilities to provide ‘‘culturally appropriate care,’’ but we think the departments should provide counties and facilities with practical assistance. We recommend:

The departments of Human Services and Corrections should identify a set of ‘‘best practices’’ for facility and county staff to help them provide culturally appropriate screening, assessment, case management, and direct services.

Finally, we think that state agencies should initiate steps to improve existing information on child placement. In particular, they should find ways to supplement placement and spending information currently collected by DHS. We recommend:

The departments of Human Services and Corrections should establish a work group to identify ways to collect comprehensive statewide information on juvenile placement spending and individual juvenile placements. Case plans should be required for delinquent juveniles, and counties should monitor aftercare services.

To the extent possible, the Department of Human Services should identify and correct errors in its existing juvenile placement database that have resulted (and may continue to result) in misrepresentations of the number of children in placement, the characteristics of those children, and the days spent in placement.

State rules should require facilities to collect program completion information and make it publicly available. The departments of Corrections and Human Services should establish a working group to adopt uniform definitions for measuring program completion rates.

Introduction Children are placed away from their homes for a variety of reasons and in a variety of settings. The residents receiving care and treatment in juvenile facilities range from infants to adolescents. They include delinquents, abused and neglected children, emotionally disturbed children, and children with drug and alcohol problems.

The 1998 Legislature’s omnibus crime bill requested the Legislative Auditor to study out-of-home placements and included a lengthy list of suggested topics for study. 1 The Legislative Audit Commission approved this study in April 1998, and our research posed the following questions:

How do counties and courts decide when to make placements? Is there adequate screening and assessment of children?

Does Minnesota need more beds to serve children placed out of home? If so, what types of beds (or supportive services) does the state need? Does Minnesota have sufficient non-residential service alternatives and aftercare following residential placements?

Is there sufficient information on the performance and operation of Minnesota’s out-of-home placement system?

In our study, we analyzed placement and spending information obtained from state agencies, counties, and residential facilities. Because of the absence of statewide data on placements paid for by county agencies other than social services departments, we collected information on these placements with a survey in June and July 1998. 2 To help us better understand the perceptions of county officials and judges about the placements they make and the adequacy of existing services, we surveyed the human services director and a juvenile corrections supervisor (or lead probation agent) in each county in August 1998. We also surveyed district court judges at that time, limiting our analysis to those judges who had presided over ten or more juvenile cases in the previous year. We received responses from all of the county corrections and human services officials, and 92 percent of the judges. 3 The judgments of these three surveyed groups sometimes differed, perhaps reflecting differences in the types of cases with which they were most familiar or their interpretations about what constituted ‘‘adequate’’ or ‘‘available’’ service.

This study was requested in the 1998 crime bill.

1 Minn. Laws (1998), ch. 367, art. 10, sec. 16.

We also reviewed samples of case files during site visits to seven counties, and we reviewed selected case information that we obtained directly from residential facilities. 4 In addition, we interviewed numerous county and facility staff.

To avoid duplicating research done by others (or previously by our office), there were some topics related to out-of-home placement that we did not evaluate in depth. For example, a 1998 House-Senate task force examined education costs for children in out-of-home placement, so we limited our research in this area. Also, the Legislature requested that our study analyze ‘‘the effectiveness of juvenile out-of-home placement, including information on recidivism, where applicable, and the child’s performance after returning to the child’s home.’’ Based on subsequent discussions with legislators, we did not collect information on juvenile recidivism--which had been a topic of extensive review in an earlier report by our office. 5 But chapters 5 and 6 of this report offer general observations and suggestions on performance measures for children placed out of home.

Our study uses the term ‘‘juveniles’’ to refer to persons under age 18, plus persons 18 or over who continue to receive child welfare services or who remain under the juvenile court’s jurisdiction. In addition, we focused on juveniles whose placements were partly or fully funded by public agencies--not those whose placements were funded entirely by their families or private insurance. Finally, we examined juveniles in a broad range of state-licensed residential ‘‘facilities,’’ ranging from family foster homes to large institutions.

Our methods included surveys, file reviews, interviews, and analysis of existing data.

2 We requested summary information on 1992 and 1997 placements not paid for by social services budgets. We received responses from 100 percent of Minnesota counties. The agencies (other than social services) that most often paid for placements were county corrections and law enforcement agencies.

3 Some agencies in Minnesota’s 87 counties are jointly administered or share staff, so we surveyed 84 human services directors and 82 juvenile corrections supervisors. Some counties contract with the Department of Corrections to provide juvenile probation services, so our survey respondents in some counties were state employees.

4 We reviewed 164 files in Crow Wing, Hennepin, Nicollet, Olmsted, Ramsey, St. Louis, and Washington counties, and we reviewed facility discharge reports for 98 juveniles. The sample sizes were not large enough to ensure that the reviewed cases were representative of all cases within thes e counties or the state as a whole.

5 Office of the Legislative Auditor, Program Evaluation Division, Residential Facilities for Juvenile Offenders (St. Paul, February 1995).

Chapter 1 of this report provides background information on residential placement options and laws governing child placements. Chapter 2 discusses how decisions are made about out-of-home (and out-of-state) placements. Chapter 3 examines the total cost of child placements in Minnesota, plus variation in county costs. Chapter 4 describes characteristics of children in out-of-home placement. Chapter 5 discusses the availability of residential and non-residential services for juveniles in Minnesota, and Chapter 6 offers our recommendations.

INTRODUCTION 3

Background CHAPTER 1

Government intervenes in the lives of juveniles for a variety of reasons, including delinquent acts, protection from abuse or neglect, emotional problems, repeated truancy, and drug and alcohol use. Children usually remain in their own homes when they (or their families) encounter problems, but some receive care and treatment away from home. This chapter provides an overview of the child placement process in Minnesota. We asked:

How are children placed out of their homes in Minnesota, and what guidance does state law provide about the child placement process?

How many residential beds are available for child placements in Minnesota and in what types of settings?

How might recent law changes affect future placement patterns?

Overall, we found that it is difficult to describe a ‘‘typical’’ out-of-home placement in Minnesota. Some residential settings primarily provide care and shelter, some are meant to hold children accountable for their actions, and others aim to change children’s behaviors through therapeutic programs. Some juveniles in out-of-home placement live in family homes; others live in large facilities or behind locked doors. Most long-term, publicly-funded child placements are court-ordered, but others result from voluntary agreements between a child’s guardians and a local social services agency. County agencies often advise the courts about placements, and they sometimes select the facility in which a child will be placed.

TYPES OF PLACEMENTS

There are several ways that Minnesota children are placed in publicly-funded out-of-home care. First, a child may be temporarily placed in detention or shelter care. For instance, police may place a suspected juvenile offender in a detention facility after taking the child into custody, or they may place an abused child in a shelter for temporary protection and care. Depending on the circumstances, state law requires the court to hold a hearing within 24 to 72 hours to determine whether a child placed by a peace officer should remain in custody. 1

1 Minn. Stat. §260.172, subd. 1.

In general, a child placed in detention or a shelter by a peace officer must be returned home unless the court finds reason to believe that this would endanger the child or others, or that the child might run away.

A second type of placement is a voluntary placement. This occurs when the county social services agency and the child’s parents or guardians complete a written ‘‘voluntary placement agreement.’’ Parents often enter into voluntary placement agreements in cases where the county is considering court action to have the child removed from the home. Because the agreement is voluntary, it may be revoked by the parents or guardians. If the agreement is revoked, the child must return home unless a court determines that this would not be in the child’s best interests. Voluntary placements require court approval if they are to last more than 90 days, or six months for placements involving developmentally disabled or emotionally handicapped children. If a child has not returned home six months after a voluntary placement, the county social services agency must file a petition seeking up to six more months in the existing placement, a petition for protection or services, or a petition to terminate parental rights. 2

A third type of placement is a ‘‘court-ordered’’ placement, and this type accounts for most of the days that Minnesota children spend in out-of-home placement. 3 Such a placement may occur if a court, in response to a petition, finds that a child is (a) delinquent, or (b) in need of protection or services (see Figure 1.1). The law provides the court with various options for case ‘‘disposition,’’ including placement away from home. Among its options, the court can transfer legal custody of the child to a social services agency, ‘‘child-placing agency,’’ county home school, commissioner of corrections, county probation officer, or ‘‘reputable person of good moral character.’’ 4

Court-ordered placements account for most days of care.

2 When the court is petitioned to review the foster care status of a developmentally disabled or emotionally handicapped child, it may authorize the voluntary placement to continue for up to another 12 months. Minn. Stat. §257.071, subd. 3 and §260.192.

3 The Department of Human Services maintains county-reported information about placements paid by social services budgets. For 1997, this database showed that 75 percent of placement days were court-ordered, compared with 15 percent for voluntary placements and 9 percent for children in police custody for protective services. This database does not contain records for some delinquency related residential placements and most detention placements--the vast majority of which result from court orders and actions by law enforcement agencies.

4 Minn. Stat. §260.185, subd. 1 and §260.191, subd. 1. ‘‘Child placing agencies’’ are county or private agencies designated or licensed by the Commissioner of Human Services to place children in residential programs, foster care, or adoptive homes.

5 Minn. Stat. §260.012, subd. 2 (a).

State law directs the courts to make judicious use of out-of-home placement. For delinquent children under the court’s jurisdiction, ‘‘it shall be the duty of the court to ensure that reasonable efforts are made to reunite the child with the child’s family at the earliest possible time, consistent with the best interests of the child and the safety of the public.’’ 5 In cases involving children in need of protection or services (or ‘‘CHIPS’’ cases), the paramount consideration is the health, safety, and best interests of the child, according to state law. The law says that children in CHIPS cases should be removed from their parents’ custody ‘‘only when the child’s welfare or safety cannot be adequately safeguarded without removal; and, when removal from the child’s own family is necessary and in the child’s best interests, to secure for the child custody, care and discipline as nearly as possible equivalent to that which should have been given by the parents.’’ 6

State law requires all court disposition orders to specify why the best interests of the child are served by the order, what alternatives were considered, and why the alternative dispositions were rejected. When making dispositions, the courts rely considerably on the recommendations of county corrections and human services agencies, and they often consider the opinions of defense and prosecution attorneys, guardians ad litem, parents, and others.

Figure 1.1: Definitions of Delinquent Children and Children in Need of Protection or Services

A ‘‘delinquent’’ child is one who has:

Violated a state or local law, except for habitual truants, runaways, juvenile petty offenders, certain traffic offenders, and juveniles who have violated laws pertaining to possession of small amounts of marijuana and underage alcohol possession;

Violated a law of another state or the federal government and whose case has been referred to juvenile court; or

Escaped or absconded from a court-ordered stay at a state or local juvenile correctional facility.

A child in need of protection or services (often called ‘‘CHIPS’’) is one who has:

Been abandoned or is without a parent, guardian, or custodian;

Been a victim of abuse or neglect (or is living with abuse victims or perpetrators);

Parents, guardians, or custodians who want to be relieved of the child’s care and custody for good cause;

Been placed for adoption or care in violation of law;

Behaviors, conditions, or environments that may be injurious or dangerous to the child or others;

Committed a delinquent act before age ten;

Run away from home;

Been a habitual truant;

Been found incompetent to proceed or not guilty due to mental health problems in a delinquency, criminal, or petty offense case;

Committed domestic abuse by a minor, been court-ordered to live out of the parent’s home, and has no alternative, safe living arrangement;

A custodial parent whose rights to another child have been involuntarily terminated within the past five years.

SOURCE: Minn. Stat. §260.015, subd. 2a and 5.

Children can be placed due to delinquency, protection from abuse or neglect, or a need for services.

6 Minn. Stat. §260.011, subd. 2 (a).

The 1997 Legislature established a state policy that placements should be selected to meet juveniles’ individual needs. 7 In addition, it adopted a list of factors that should be considered, including: (1) the child’s current functioning and behaviors, (2) the medical, educational, and developmental needs of the child, (3) the child’s history, (4) the child’s religious and cultural needs, (5) the child’s connection with a community, school, and church, (6) the child’s interests and talents, (7) the child’s relationship to caretakers, parents, siblings, and relatives, and (8) the preferences of the child. Also, when children are placed due to imminent risk of abuse or neglect or due to a need for treatment of emotional disorders, chemical dependency, or mental retardation, state law requires social services agencies to determine ‘‘the level of care most appropriate to meet the child’s needs in the least restrictive setting and in closest proximity to the child’s family.’’ 8

A court may order juveniles within its jurisdiction to be examined by physicians, psychiatrists, or psychologists. The court may also request local social services or corrections agencies to prepare reports on juveniles’ personal and family histories. 9 In some cases, the law requires that the courts obtain specific types of child assessments. For example, if a child is found to be delinquent for committing a felony-level offense, the court is required to make a finding regarding the juvenile’s mental health and chemical dependency treatment needs. Likewise, children found by the court to have committed certain sex offenses must be assessed to determine whether sex offender treatment is needed. 10 Prior to placement in a residential chemical dependency treatment facility, a child must be assessed and meet the placement criteria outlined in state rules. 11 And, before a child with a severe emotional disturbance can be placed in a treatment facility, county staff are required to determine whether residential treatment is necessary and appropriate. 12

In CHIPS cases, the court must document ‘‘whether reasonable efforts. . . were made to prevent or eliminate the necessity of the child’s removal and to reunify the family after removal.’’ 13 If the court finds that further preventive or reunification efforts could not permit the child to safely return home, the court may authorize (or continue) removal of the child from the home--even if the social services agency has not yet made reasonable efforts to prevent placement or reunify the family.

Federal and state laws establish special procedures for cases in which placement of American Indian children is being considered--partly in response to high placement rates among Indian children. 14 For example, Minnesota tribal social service agencies must be notified of cases that could result in placement of an American Indian child. The tribal agency has authority to review relevant case records, and official tribal representatives have a right to participate in court proceedings that are subject to the federal Indian Child Welfare Act. In some cases, the law gives tribal courts jurisdiction over child placement proceedings.

Placements should be selected to meet juveniles’ individual needs.

7 Minn. Stat. §260.181, subd. 3.

8 Minn. Stat. §256F. 07, subd. 2. The law also requires the agency to estimate the length of the placement, set a placement goal, and state the anticipated outcome of the placement.

9 Minn. Stat. §260.151, subd. 1.

10 Minn. Stat. §260.185, subd. 1.

11 Minn. Rules 9530.6615-9530.6650.

12 Minn. Stat. §245.4885, subd. 1.

13 Minn. Stat. §260.191, subd. 1a.

14 25 U. S. Code, secs. 1901-1963; Minn. Stat. §§ 257.35-257.3579.

TYPES OF RESIDENTIAL SETTINGS

All children placed out of home in Minnesota live in residences that are licensed by either the Minnesota Department of Human Services (DHS) or Minnesota Department of Corrections (DOC). At some facilities--such as the three facilities operated by DOC--all (or nearly all) of the juveniles have been adjudicated delinquent by a court. In contrast, some DHS-licensed foster homes only serve juveniles who have been placed for purposes of protection or services. But many residential facilities serve a variety of children, and there is considerable overlap in the types of juveniles served by DOC-licensed facilities and those served by DHS-licensed facilities. 15 For example, some delinquent juveniles are placed in residential mental health treatment facilities licensed by DHS, and many juveniles in DOC-licensed facilities have mental health problems. In fact, this overlap was one reason that the 1995 Legislature required DHS and DOC to jointly develop rules which would apply uniform standards to certain aspects of residential settings licensed by the agencies. 16

Table 1.1 provides information on the number of licensed beds in each facility category, plus our estimate of the actual number of days of care in each category in 1997 (discussed further in Chapter 2). As shown in the table,

There are more than 17,000 beds for juveniles in residential facilities licensed by DHS and DOC.

Foster care is the largest single category of residential care for juveniles, representing three-fourths of Minnesota’s total licensed juvenile beds and two-thirds of total days of care in 1997.

It would be useful to know how Minnesota’s total number of beds, children served, or days in residential services compare with other states. Unfortunately, we found no recent, reliable sources of information on the aggregate numbers of children in placement by state. 17

Facilities are licensed by different agencies but serve overlapping populations of juveniles.

15 Office of the Legislative Auditor, Program Evaluation Division, Residential Facilities for Juvenile Offenders (St. Paul, February 1995), 38-40.

16 Minn. Laws (1995), ch. 226, art. 3, sec. 60.

17 Paul Lerman, ‘‘Child Protection and Out-of-Home Care: Systems Reforms and Regulating Placements,’’ in Protecting Children from Abuse and Neglect , ed. Gary B. Melton and Frank D. Barry (New York: Guilford Press, 1995), 353-437. There are state-by-state data on juveniles in certain types of facilities, but the data are ‘‘notoriously tardy’’ and duplicative across systems, an d they cannot be aggregated reliably due to the different methods of counting placements.

In the sections that follow, we briefly describe the general categories of residential services available for juveniles in Minnesota. For each facility category, Chapter 3 provides information on the cost per day, and Chapter 4 discusses the characteristics of juveniles served.

Correctional Facilities

In late 1998, there were a total of 2,175 beds for juveniles in facilities and foster homes licensed and regulated by the Minnesota Department of Corrections (DOC). This included 1,600 ‘‘non-secure’’ beds and 575 ‘‘secure’’ beds. A secure living unit keeps its residents in locked confinement--often to protect public safety or to prevent a suspected or adjudicated offender from fleeing.

A majority of DOC-licensed beds are in ‘‘juvenile residential facilities.’’ 18 For the most part, these facilities serve juveniles who have been adjudicated delinquent by the courts. Some of these facilities primarily provide a consequence treatment (" Rule 5") facilities "Rule 8" group homes DHS 659 174,580 Family foster homes DHS 13,406 2,223,102

Table 1.1: Number of Licensed Beds and 1997 Days of Care in Minnesota Facilities for Juveniles

Licensing Number of Days of Type of Facility Agency Licensed Beds 1 Care, 1997 2

Correctional residential and DOC 1,849 460,880 detention facilities (secure and non-secure) Residential mental health DHS 918 260,533

DOC 326 Not available 3 Chemical dependency facilities DHS Not available 4 54,031 Shelters DHS Not available 5 188,083 Other 6 31,108

TOTAL 17,158 3,392,317 SOURCE: DHS and DOC licensing units, Program Evaluation Division analysis of DHS substitut e care database, DOC data on days of care at Red Wing and Sauk Centre facilities, June-July 1998 Prog ram Evaluation Division surveys of counties, and DHS chemical dependency data.

1 Beds licensed as of May 1998 for foster care and November 1998 for other types of facilities. 2 Includes days of care for Minnesota children placed out of state, but excludes children from o ther states in Minnesota facilities. 3 Days of care in DOC-licensed foster homes are included in the DOC correctional facilities total.

4 Chemical dependency facilities are often licensed to serve both adolescents and adults, so i t is not possible to precisely determine statewide adolescent bed capacity. 5 Shelters are not a separate category for licensing purposes--the number of shelter beds are included in the Rule 5 and family foster home categories. 6 These are days of placement that we were unable to allocate to types of facilities.

Family foster homes are the largest category of residential facilities.

18 Minn. Rules 2935.

State rules require DOC-licensed juvenile residential facilities to have at least one direct service employee per 12 residents during waking hours. Residents must have individualized program plans, and facilities must work with each resident to plan for any needed programs after their discharge. State rules require these facilities to provide ‘‘a comprehensive and continuous ’’ education program for residents. Each facility must also have ‘‘a social services program, such as individual and group counseling, community services, and family services.’’ 19

A variety of public and private organizations operate juvenile residential correctional facilities. Currently, there are 3 facilities operated by the Minnesota Department of Corrections (at Red Wing, Sauk Centre, and Togo), 13 operated by counties, and 13 operated by private organizations. The DOC’s Red Wing facility admits juveniles committed by the courts to the Commissioner of Corrections, and it is the only residential facility in Minnesota that must admit all referred juveniles. (The Sauk Centre facility served committed juveniles through the end of 1998.) 20

Another category of DOC-licensed facilities is secure detention centers, most of which are operated by counties. 21 These facilities primarily confine and care for juveniles who have not yet been adjudicated by the courts. A minimum of two staff must be on duty in these facilities at all times. Most of Minnesota’s secure juvenile beds are in detention centers.

Residential Treatment Centers (or ‘‘Rule 5 Facilities’’)

As of late 1998, the Department of Human Services licensed and regulated a total of 918 beds in juvenile mental health treatment facilities. These facilities are commonly called ‘‘Rule 5 facilities ’’ or ‘‘residential treatment centers, ’’ and they operate under the clinical supervision of a mental health professional. 22 State rules authorize these facilities to provide shelter, food, training, treatment, and other care to children with severe emotional disturbances. Most treatment services at Rule 5 facilities are provided on-site by facility staff.

Some correctional facilities serve juveniles found delinquent by courts; others detain juveniles awaiting court hearings.

19 Minn. Rules 2935.2000, subp. 3.

20 In addition, the 56 counties that did not participate in the Community Corrections Act could place juveniles without charge at Red Wing and Sauk Centre through 1998. Starting in 1999, all counties placing children at the Red Wing and Sauk Centre facilities are billed the actual cost of services, excluding education costs ( Minn. Stat. §242.192).

21 Minn. Rules 2930.

22 The rules governing these facilities are now Minn. Rules 9545.0905-9545.1125.

State regulations require Rule 5 facilities to base admission decisions upon a comprehensive assessment of a child and his or her needs. Following a child’s admission, the treatment program should ‘‘assist the child in returning to the community in a manner that will enable the child to function to the child’s fullest possible extent.’’ 23 If a child requires the services of a psychologist or psychiatrist, the facility must arrange for this. State requirements for staff-to-child ratios vary, depending on the ages of the children. For instance, facilities must have one staff member during waking hours for every three children who are ages four or five; there must be one staff member for every eight children who are 12 to 16 years old. Generally, children under age 16 must attend school within the Rule 5 facility’s school district or at a school within the facility. 24

All of Minnesota’s Rule 5 facilities are privately operated except for two regional treatment centers operated by the Department of Human Services in Brainerd and Willmar. These two state facilities are also licensed as acute care hospitals.

Group Homes

As of late 1998, the Department of Human Services licensed and regulated a total of 659 beds in ‘‘group homes.’’ Sometimes called ‘‘Rule 8 facilities,’’ these homes of ten or fewer children combine aspects of residential treatment facilities and foster homes. 25 The homes have staff who can provide care, supervision, and treatment, but they also rely considerably on community resources for employment, counseling, education, and recreation. All DHS-licensed group homes are privately owned and operated.

Each group home must have ‘‘group home parents ’’ who provide 24-hour-a-day supervision and care in the residence. State regulations recommend a staff-to-child ratio of one staff member for every five children. As needed, the home must employ or contract for social services, medical, dental, psychiatric, and psychological staff. School age children are referred to the local school district for their education.

State rules require group homes to develop a program plan for each resident. In addition, the rules recommend that the homes offer social services to families of the residents ‘‘whenever feasible and in accordance with the planning of the child.’’ 26

Group homes are small residential facilities. "Rule 5" facilities serve emotionally disturbed youth.

23 Minn. Rules, 9545.0915, subp. 4.

24 State rules authorize designated school authorities to exempt certain children under 16 from attending school in the local district. Schools operated within Rule 5 facilities must meet state a nd local district standards. Special plans for education must be developed in cases where children are unable to attend neighborhood schools.

25 Minn. Rules 9545.1400-9545.1480.

26 Minn. Rules 9545.1450, subp. 6.

Foster Care

The Department of Human Services licenses most of Minnesota’s foster care beds. According to DHS rules,

‘‘The purpose of foster care is to provide substitute family or group care for a child while an intensive effort is made to correct or improve the conditions causing placement and to reunite the family or, if the child cannot be returned home, to provide some other permanent plan.’’ 27

The rules identify six categories of foster family homes that provide 24-hour a day care: (1) ‘‘emergency shelter homes,’’ which usually provide care for 30 days or less; (2) ‘‘interim homes,’’ in which children are expected to return home within one year or be placed for adoption within two years, (3) ‘‘permanent homes,’’ which care for children until they turn age 18, (4) ‘‘restricted homes,’’ which can only serve those children named in the license, (5) ‘‘special services homes,’’ which have specialized staff who can provide ‘‘extraordinary care or services,’’ and (6) ‘‘group family foster homes,’’ which can serve up to ten children. 28 For each category, state rules establish limits on the number of children that can be served per home and designate experience and training requirements for service providers.

Some foster homes provide specialized therapeutic services within the residence for placed children (such as counseling or treatment), but most do not. 29 Local social services agencies pay for foster care through a uniform ‘‘maintenance’’ payment based on the child’s age plus a supplemental payment based on the child’s ‘‘difficulty of care.’’ 30

As of May 1998, foster family homes licensed by DHS had a ‘‘licensed capacity’’ of about 13,400 beds. DHS delegates responsibility to county and private social services agencies to accept foster care licensure applications and recommend their approval or denial. These agencies also can inspect and evaluate foster homes, monitor their compliance with state rules, issue ‘‘corrections orders, ’’ and enforce orders of the DHS commissioner. 31

Most foster homes do not provide counseling or treatment.

27 Minn. Rules 9560.0510.

28 Minn. Rules 9545.0010, subp. 7. A family foster home may provide more than one of these types of care.

29 Foster homes are not required by law or rule to develop treatment plans for their residents, although some do. County social services agencies develop a ‘‘placement plan’’ for each child within 30 days of placement.

30 Minn. Rules 9560.0650-9560.0656.

31 Minn. Rules 9543.0030.

32 Minn. Stat. §257.072, subd. 1.

State law requires public and private child-placing agencies to make ‘‘special efforts to recruit a foster family from among the child’s relatives.’’ 32 Within six months of a child’s placement in a residential facility, the local social services agency must identify relatives of the child and notify them about the need for a foster home for the child and the possibility of the child’s out-of-home placement on a permanent basis. 33 A relative of a child who wishes to provide foster care must obtain a DHS license to do so--either an ‘‘emergency license’’ for temporary care or a regular foster home license.

The Commissioner of Corrections is authorized by law to license foster care facilities for delinquent juveniles. 34 As of late 1998, there were 326 beds in DOC-licensed group foster homes, and the homes ranged in size from 1 to 20 beds.

Chemical Dependency Facilities

DHS licenses residential programs that aim to change individuals’ patterns of drug and alcohol use. 35 ‘‘Primary inpatient treatment ’’ must provide at least 30 hours of rehabilitative services to residents per week, ‘‘extended care ’’ programs must provide an average of 15 hours of rehabilitative services per week, and ‘‘halfway houses’’ must provide at least 5 hours of rehabilitative and transitional services per week. State rules establish uniform placement criteria for each category of facility, and decisions about which persons meet the criteria are made by county chemical dependency assessors. On average, juveniles stay in inpatient treatment for shorter periods than other types of chemical dependency treatment, but inpatient care is also the most expensive per day. 36

Since 1988, Minnesota’s Consolidated Chemical Dependency Treatment Fund has pooled local, state, and federal funds to help pay for assessments and services. The state has restricted eligibility in recent years to match the size of the fund, but there has been sufficient funding to pay for adolescents from families with incomes up to 60 percent of the state median. In 1997, the fund paid for about half of the placements of Minnesota children at chemical dependency facilities, and private sources paid for almost all the rest.

RECENT LAW CHANGES

Chapters 3 and 4 present 1997 data on the number of children in placement and the amount of time they spend in placement. This represents the most recent statewide information available about child placements in Minnesota.

Some juveniles are placed out of their homes because of drug or alcohol use.

33 Minn. Stat. §257.071, subd. 1d.

34 Minn. Stat. §241.021, subd. 2.

35 Minn. Rules 9530.4100-9530.6810.

36 Among juveniles who entered primary inpatient programs in 1997 that were paid for by the consolidated treatment fund, the average stay was 26 days, and the average cost per day was $205.

There has been increasing concern nationally and in Minnesota about the amount of time children spend in out-of-home placement. Long-term out-of-home placement is expensive, and many observers believe that delays in finding permanent living arrangements may cause children to lose opportunities to develop strong emotional attachment to adults. In part, the U. S. Congress passed the Adoption Assistance and Child Welfare Act of 1980 to prevent children from remaining unnecessarily long in foster care. It required periodic reviews of cases involving children in placement, as well as dispositional hearings to ‘‘determine the future status of the child ’’ within 18 months of placement. 37 The 1993 Minnesota Legislature required that these ‘‘permanency placement determination hearings’’ occur within 12 months of placement, unless the court granted a time extension. 38 Permanent placement dispositions allowed by Minnesota law include returning the child home, terminating parental rights and finding an adoptive home for the child, long-term foster care, and a transfer of custody to a relative. But the courts and counties interpreted Minnesota’s permanency ‘‘timeline’’ in many ways, prompting further legislative concerns about the amount of time that children were in placement. 39

Within the past two years, changes in federal and state law were intended to reduce the amount of time children spend in CHIPS placements before permanent homes are sought or identified. Examples of key changes have included the following:

The 1997 Legislature modified (and clarified) the timing of certain events in the ‘‘permanency planning’’ process. It required courts to conduct hearings to determine the permanent status of children within 12 months of placement, eliminating the option of a timeline extension. The Legislature established procedures for (1) determining when this 12-month period begins, and (2) counting placement time under current and previous placements toward the deadline for permanency planning. 40

The federal Adoption and Safe Families Act of 1997 requires that agencies make ‘‘reasonable efforts ’’ toward reunification of children with their families during only the first 15 months of placement. It also exempted certain cases altogether from ‘‘reasonable efforts ’’ requirements. In addition, the law required that termination of parental rights petitions be filed in cases where children have been in foster care for 15 of the past 22 months. 41

Recent laws aim to speed the process of finding permanent homes for children.

37 42 U. S. Code, sec. 675 (5) (C).

38 Minn. Laws (1993), ch. 291, sec. 21.

39 Final Report of the Minnesota Supreme Court Foster Care and Adoption Task Force (St. Paul, January 1997).

40 Minn. Laws (1997), ch. 239, art. 6, sec. 26.

41 P. L. 105-89.

The 1998 Legislature said that, effective July 1999, permanent placement determination hearings for children under age eight must be held within six months of out-of-home placement, not 12. For voluntary placements exceeding 60 days and for court-ordered placements, the Legislature provided counties with funding for ‘‘concurrent planning ’’--that is, planning for ways to reunite children with their families while simultaneously exploring alternative ways to find permanent homes for the children. The 1998 Legislature also set faster timelines for filing adoption petitions, forfiling termination of parental rights petitions in cases where children have been subject to egregious harm, and for filing disposition orders in CHIPS cases. 42

42 Minn. Laws (1998), ch. 406, arts. 1 and 2.

Placement Decisions CHAPTER 2

State policy requires courts and child-placing agencies to make placements that will meet children’s individual needs. 1 To do this, persons involved in the placement process need information about the children involved (and often their families), and they should know about the strengths and weaknesses of various service options. Also, decisions to place children should be made only when alternatives to placement have been fully explored. We asked:

How do counties and courts decide which children to place out of home? Is there a need for more explicit placement criteria?

Do counties adequately screen and assess children being considered for placement?

Do court dispositions clearly state the reasons for child placements and the alternatives that were considered?

Do counties have enough information for making placement decisions?

To what extent are children placed out-of-state, and what are the reasons?

Overall, we found that counties and courts usually do not have criteria that indicate specific circumstances in which placement may be appropriate, so placement decisions depend considerably on the judgments of the professionals involved. Often these decisions undergo scrutiny from multiple staff and agencies, but in many cases they do not. In addition, some human services officials believe that decisions do not sufficiently consider information about juveniles’ chemical use and mental health. In general, there is evidence that the screening and assessment process could be improved in many counties, and the courts could do a better job of articulating their reasons for placement.

PLACEMENT CRITERIA

In 1990, a Child Welfare League of America task force said that one of the key challenges facing child welfare agencies nationally was the development of criteria to determine (a) when it is in the best interests of a child to be removed from home, and (b) which placement option to select in cases for which out-of-home care is appropriate. 2 Minnesota laws and rules provide limited guidance about the specific circumstances that justify placement of children out of their homes. For example, child protection agencies are authorized to ‘‘seek removal of the child from the home. . . if the child is found in surroundings or conditions which endanger the child’s health or welfare and the child cannot be protected from harm while remaining in the home.’’ 3 In general, state laws and rules give courts and local agencies considerable discretion to determine the specific situations in which out-of-home placement is appropriate. 4

1 Minn. Stat. §260.181, subd. 3.

We examined whether counties or district courts have more detailed policies that outline circumstances that justify child placement. First, in cases involving child maltreatment, we found that:

Most county child protection agencies do not have written criteria that identify circumstances that justify out-of-home placement--aside from the very broad criteria in state law and rules.

In June 1997, we surveyed county human services agencies to obtain criteria that the agencies use to make child protection decisions. We found that only 38 percent of these agencies had screening criteria that helped them determine which allegations of child maltreatment to investigate. 5 These same criteria are used by some of the agencies to determine what constitutes child abuse or neglect. But even the agencies with maltreatment screening criteria usually did not have specific criteria to help them distinguish which children needed to be removed from their families.

Second, we examined district court policies regarding dispositions of delinquent children. In 1994, the Legislature required each of the state’s ten judicial districts to develop and publish written criteria for making juvenile delinquency dispositions. 6 We found that:

Most judicial districts’ disposition guidelines do not indicate specific circumstances that might justify placement, and many county staff told us that the guidelines have had limited impact on placement decisions.

Seven of the ten judicial districts’ guidelines consist of lists of factors that could be considered when making a disposition. For example, the eight factors considered by the Second Judicial District (Ramsey County) are (1) seriousness of the offense, (2) prior offense history, (3) prior interventions, (4) offender’s treatment needs, (5) offender’s acceptance of responsibility, (6) family and community support, (7) offender’s age, and (8) available disposition options. As in most judicial districts, these criteria do not clearly specify how these factors should be used to make decisions or when out-of-home placement may be appropriate.

Most child placements are not based on detailed, explicit criteria.

2 Child Welfare League of America, Out-of-Home Care: An Agenda for the Nineties, Report and Recommendations of the CWLA Task Force on Out-of-Home Care (Washington, D. C., 1990), 11-13.

3 Minn. Rules 9560.0220, subp. 8.

4 The most specific placement criteria are those for various categories of chemical dependency treatment (Minn. Rules 9530.6625-9530.6655). Decisions about which persons meet the placement criteria are made by county chemical dependency assessors.

5 Minnesota Office of the Legislative Auditor, Program Evaluation Division, Child Protective Services (St. Paul, January 1998), 24-29.

6 Minn. Laws (1994), ch. 576, sec. 59.

The other three judicial districts have identified specific disposition options that should be considered, depending on a juvenile’s offense and other characteristics. 7 For example, a point system devised by the Third Judicial District (southeastern Minnesota) offers guidance about which cases are appropriate for in-home services and which merit out-of-home placement. This district’s guidelines also recommend the amount of punishment or service for each disposition. As in other districts, judges are not required to follow the guidelines.

For the most part, corrections officials told us that judicial districts’ disposition criteria have had limited impact on placement decisions. Some county staff told us they were unaware of the guidelines, and some others said that judges (or others involved in the disposition process) have not used the criteria when making placement decisions. Some corrections officials told us that factors such as the juvenile’s demeanor and family circumstances can play an important role in placement decisions, even if disposition criteria exist.

Given the absence of detailed guidelines for deciding when out-of-home placement is an appropriate option, we wondered whether county officials perceive that judges are making consistent decisions about which children to place. In August 1998, we surveyed the human services director and a juvenile corrections supervisor (or lead probation agent) in each county. 8 As shown in Figure 2.1, our surveys indicated that:

Sixty-two percent of human services directors and 32 percent of corrections supervisors said that judges were ‘‘sometimes, rarely, or never’’ consistent in their decisions about which circumstances justify out-of-home placement.

One option for improving consistency might be the implementation of written placement criteria that are more detailed than the broad criteria in state law. We think that explicit criteria could help counties and courts make more consistent decisions, but our review of literature indicated that no consensus on appropriate placement criteria has yet emerged. 9 There is a need for further research to identify the types of children who benefit (and those who do not) from residential placement.

Many county officials think that judicial placement decisions are inconsistent.

7 These districts are the First, Third, and Fourth districts.

8 Some of Minnesota’s counties jointly administer services or share staff, so we sent surveys to 84 human services directors and 82 juvenile corrections supervisors. We received responses from all of them.

9 Brady C. Bates, Diana J. English, and Sophia Kouidou-Giles, ‘‘Residential Treatment And Its Alternatives: A Review of the Literature,’’ Child & Youth Care Forum 26, no. 1 (February 1997), 11-12.

In addition, county officials expressed mixed feelings about the need for placement criteria. As shown in Table 2.1, only 18 percent of county corrections officials and 35 percent of human services directors expressed support for clearer placement guidelines in state laws. Fifty percent of human services directors said they would support the development of more specific county policies on placement, compared with 34 percent of corrections supervisors.

1 36 48 14 13 52 28 4 0 10 20 30 40 50

Always or almost always consistent

Usually consistent

Sometimes consistent

Rarely or never consistent

Human services directors

Corrections supervisors

Percent of respondents who rated judges as... Figure 2.1: County Staff Perceptions About

Judicial Consistency in Placement Decisions

SOURCE: Program Evaluation Division survey of county officials, August 1998 NOTE: The survey question asked: “Are judges consistent in their decisions about which circumstances justify out-of-home placement?”

Table 2.1: County Perceptions About the Need for Clearer Placement Criteria

Corrections Supervisors

(N = 82) Human Services

Directors (N = 84)

There is a need for: Agree Disagree Agree Disagree Clearer placement criteria in state laws 18% 62% 35% 45% Clearer county placement criteria 34 48 50 29

NOTE: Table does not show the percentages of county officials who responded "neither agree n or disagree" or "don’t know."

SOURCE: Program Evaluation Division surveys, August 1998.

County officials generally do not favor statewide placement criteria.

COUNTY REVIEW OF PLACEMENT RECOMMENDATIONS

The Minnesota Family Preservation Act ( Minn. Stat. §256F) requires each county board to establish a ‘‘preplacement procedure’’ to review ‘‘each request for substitute care placement.’’ 10 The law does not specify who should conduct these reviews. Furthermore, DHS staff told us that the law was not intended to apply to cases involving delinquent juveniles.

Another Minnesota law authorizes--but does not require--county social services agencies to establish ‘‘juvenile treatment screening teams.’’ 11 For those counties that choose to establish such teams, the law prescribes the composition of the team and its procedures for making recommendations about child placements. The team must include social workers, juvenile justice professionals, and experts in the treatment of emotional disorders, chemical dependency, and developmental disabilities. It must involve parents or guardians ‘‘as appropriate.’’

If a court proposes to place a child in a residential facility in order to treat an emotional disturbance, chemical dependency, or developmental disability, it must seek input from the screening team, if one exists. The court may accept or reject the team’s recommendations, but the court is required to justify its decision in writing if it decides to place a child over the objection of the screening team. According to our August 1998 survey of Minnesota’s 84 county human services directors,

Fifty-two percent of Minnesota’s county social services agencies have established juvenile treatment screening teams.

Of the agencies with screening teams, 93 percent said that their teams ‘‘usually or always’’ review cases that the court is considering for placement, and 84 percent said that their screening team’s recommendations to the court are ‘‘usually or always’’ followed.

In addition to the review of potential child placements by multi-disciplinary screening teams, some county agencies have established internal processes to review placement options before recommendations are made to a screening team or a judge. Involving multiple staff from an agency in the decision-making process may help to ensure that the decisions are consistent and appropriate. For example, a panel of three senior probation officers reviews all cases being considered for placement by Anoka County’s probation office. If this panel recommends long-term residential placement, the placement is not made unless approved in a meeting of the county’s juvenile corrections supervisors. 12

Half of Minnesota counties have multidisciplinary screening teams.

10 Minn. Stat. §256F. 07.

11 Minn. Stat. §260.151, subd. 3.

12 Involving multiple staff from an agency may be difficult in Minnesota’s smallest probation agencies. For example, there are some individual counties that are served by just one probation officer.

For cases in which local agencies have made recommendations to the court for placements exceeding one month, our surveys indicated that:

Sixty-one percent of county corrections agencies ‘‘usually or always’’ involved two or more probation officers in the decisions;

Eighty percent of county human services agencies ‘‘usually or always’’ involved two or more social services line staff in the decisions;

Ninety-two percent of county human services agencies but only 51 percent of corrections agencies ‘‘usually or always’’ involved at least one supervisor or manager in the decisions.

Given the importance of placement decisions, we think it is usually reasonable to expect that more than one staff person in an agency will be involved--including at least one supervisor. The findings above suggest that some agencies--especially corrections agencies--should improve their scrutiny of placement recommendations.

SCREENING, ASSESSMENT, AND CASE PLANNING

As counties and courts try to make decisions that are consistent with a child’s ‘‘best interests,’’ they should have detailed information about the child. Decision makers should consider the child’s placement history, relationships with family members, psychiatric diagnoses, chemical use, and other factors. Such factors may indicate whether the child should be placed away from home and the types of supportive services the child needs. The law requires that assessments be done in some cases--for example, the court must make findings about chemical dependency and mental health treatment needs for juveniles who have committed felony-level offenses. 13 But in most cases involving child placement, the counties and courts have discretion about which information to gather before making a placement decision. 14

The case files we reviewed during site visits to counties did not always precisely convey the information that decision makers had at the time of their placement decisions, so it was not possible for us to decide whether the placement decisions were based on sufficient information. Also, it would take specialized expertise to tell whether county staff properly interpreted the available information and assessments. However, we used several other approaches to examine the adequacy of screening and assessment, ranging from surveys of staff participants in the placement process to a review of independent data on juveniles’ drug and alcohol treatment needs.

County supervisory staff do not always review placement decisions.

13 Minn. Stat. §260.185, subd. 1.

14 The 1997 Legislature required the state commissioner of health to administer a pilot project grant program for juvenile assessment centers in up to three judicial districts ( Minn. Laws (1997), ch. 203, art. 2, sec. 30). These centers are being established in the Third, Fourth, and Ninth districts. In addition, 15 counties have early intervention mental health screening for children in the court system or children at risk of court involvement.

We asked county human services directors and juvenile corrections supervisors whether court dispositions are based on sufficient consideration of children’s mental health needs, chemical dependency needs, and cultural and ethnic backgrounds, as well as the residential facilities’ ability to meet their service needs. As shown in Table 2.2,

Most county officials said that judges have ‘‘usually or always’’ made dispositions based on sufficient consideration of children’s needs, but many human services directors indicated room for improvement.

For instance, more than one-third of the directors said that dispositions are ‘‘sometimes, rarely, or never’’ based on sufficient consideration of children’s mental health needs and chemical dependency problems. In addition, 45 percent of human services directors said that judges ‘‘sometimes, rarely, or never’’ sufficiently consider the facilities’ ability to meet children’s service needs.

We also asked judges to evaluate whether county staff have conducted appropriate assessments. As shown in Table 2.3,

Most judges said that counties have conducted timely, thorough chemical dependency assessments, but somewhat fewer said that mental health assessments were thorough and timely.

Regarding mental health screening and assessment, state law requires county screening of all children referred for publicly-funded treatment of severe emotional disturbances at residential facilities. The screening must determine whether the proposed residential treatment is necessary, appropriate to the child’s needs, and no longer than necessary to serve the child. Furthermore, this law requires that:

Table 2.2: County Officials’ Perceptions About Whether Judicial Placement Decisions Adequately Consider Key Information

Corrections Supervisors (N = 82) Human Services Directors

(N = 84) Sometimes, Sometimes, Do judges make dispositions based Usually or Rarely, or Usually or Rarely, or on sufficient consideration of: Always Never Always Never

Children’s mental health needs? 83% 13% 62% 37% Children’s chemical dependency problems? 87 10 61 36 Children’s cultural and ethnic backgrounds? 71 21 55 39 Facilities’ ability to meet children’s service needs? 87 10 54 45

NOTE: Percentages of officials who responded "don’t know" are not shown. SOURCE: Program Evaluation Division surveys, August 1998.

PLACEMENT DECISIONS

The county board shall annually collect summary information on the number of children screened, the age and racial or ethnic background of the children, the presenting problem, . . . the screening recommendations[,] the degree to which these recommendations are followed and the reasons for not following recommendations. Summary data shall be available to the public and shall be used by the county board and local children’s advisory council to identify needed service development. 15

In August 1998, we asked county human services directors whether their counties had summary information for children screened for mental health services in 1997. We were especially interested in knowing more about the extent of child screening, the types of mental health problems identified, and (where applicable) the reasons that screening recommendations were not followed. We found that:

Less than 10 percent of human services directors said that their counties produced the summaries of mental health screening information for 1997 required by Minn. Stat. §245.4885.

We also wanted to know whether counties were properly identifying children needing services for drug and alcohol problems. Although many county officials and judges expressed satisfaction with chemical dependency assessments in our surveys, chemical dependency staff in the Minnesota Department of Human Services (DHS) told us that they questioned whether the drug and alcohol problems of adolescents in residential facilities are being adequately identified. In 1997, DHS surveyed 817 adolescents in juvenile corrections, Rule 5, and Rule 8 residential facilities. The department asked questions to determine the extent to which the adolescents met 15 diagnostic criteria used to define substance abuse and dependence. Twenty-seven percent of the surveyed juveniles met five diagnostic criteria and had used drugs or alcohol at least 40 times in the previous year--which DHS staff considered to be a good indicator of a need for chemical dependency treatment. 16 Of the juveniles deemed by DHS to need treatment, only about half had received it at some time. In addition, DHS staff estimated that another 35 percent of the surveyed juveniles at these residential facilities needed early intervention services because they exhibited some alcohol or drug problems but did not yet need treatment. 17

Table 2.3: Judges’ Perceptions About Adequacy of County Assessments

Have County Staff: Usually or Always

Sometimes, Rarely or Never Arranged for timely chemical dependency assessments of juveniles? 87% 11%

Conducted thorough chemical dependency assessments of juveniles? 80 17

Arranged for timely mental health assessments of juveniles? 73 24

Conducted thorough mental health assessments of juveniles? 66 27

NOTE: Percentages of officials who responded "don’t know" are not shown. SOURCE: Program Evaluation Division surveys of district court judges (N = 143), August 1998.

DHS staff think that many juveniles in residential facilities need drug or alcohol treatment.

15 Minn. Stat. §245.4885, subd. 5.

Another category of assessment that may be done prior to child placement is ‘‘risk assessment.’’ Risk assessment is intended to indicate a juvenile offender’s likelihood of committing new offenses or a maltreated child’s likelihood of again being the subject of abuse or neglect. County agencies often use risk assessments--and accompanying ‘‘needs assessments’’--to help identify services that would be most appropriate for the child or family. In recent studies by our office, we determined that:

All Minnesota county child protection agencies assess families’ risk of subsequent maltreatment, and about half of the counties told us they use these assessments to help them decide whether to recommend out-of-home placement. 18

Corrections agencies in 53 percent of Minnesota counties formally assess the reoffense risk of juvenile offenders. 19

State rules require child protection agencies to conduct risk assessments with an instrument approved by the Department of Human Services. 20 There is no state requirement for corrections agencies to classify juvenile offenders’ risk levels, but offender classification is required by state law for adult offenders. 21 National literature has identified the use of valid risk assessment instruments as an important component of effective interventions with juvenile offenders. 22

Risk assessments can help identify juveniles needing placement or other services.

16 DHS decided that juveniles who met 5 of the 15 diagnostic criteria could be assumed to need treatment, based partly on its finding that the average Minnesota adolescent in chemical dependency treatment met 5.5 of the diagnostic criteria.

17 Patricia Ann Harrison, Minnesota Department of Human Services, letter to John W. Patterson, July 21, 1998. DHS defined adolescents needing early intervention services as those who met at least one of the diagnostic criteria.

18 Minnesota Office of the Legislative Auditor, Program Evaluation Division, Child Protective Services (St. Paul, January 1998), 36-38.

19 Minnesota Office of the Legislative Auditor, Program Evaluation Division, Funding for Probation Services (St. Paul, January 1996), survey of corrections agencies.

20 Minn. Rules 9560.0220, subp. 6. B.

21 Minn. Stat. §244.24.

22 For example, see David M. Altschuler and Troy L. Armstrong , Intensive Aftercare for High-Risk Juveniles: A Community Care Model (Washington, D. C.: Office of Juvenile Justice and Delinquency Prevention, September 1994), 4-7.

After juveniles have been screened and assessed, it is important to develop plans that address the issues that have been identified--whether through residential services, non-residential services, or actions by the juvenile and his or her family. When the court determines that children are in need of protection or services (i. e., ‘‘CHIPS’’ cases), or when parents voluntarily place their children in a residential facility, state law requires the local social services agency to develop a case plan for each child within 30 days of placement. 23 Among other things, the plan must identify the reasons for placement, actions that will be taken by the parents to address the reasons for placement, services that will be provided to the family and child, and the date the child is expected to return home. However, we observed that:

Unlike the law governing CHIPS cases, the law governing delinquency dispositions (Minn. Stat. §260.185) does not require development of case plans outlining ways that the child’s (or family’s) needs will be addressed.

This is significant because proposed state rules for residential facilities drafted jointly in 1998 by the state departments of Human Services and Corrections refer to county ‘‘case plans,’’ but local and state corrections staff with whom we spoke acknowledged that such plans are not currently developed in some counties. 24 In Chapter 6, we suggest that the Legislature should clarify county case planning responsibilities.

ADEQUACY OF INFORMATION ABOUT FACILITIES

Choosing the right facility to meet the needs of an individual can be a complicated decision. Counties and courts need information about the child being placed, but they also need to have an understanding of the residential settings available. Many placement staff develop this knowledge through facility visits, meetings with facility staff, and experience with the facilities from prior placements.

We asked county officials whether they think their staff have sufficient information about residential programs. Table 2.4 shows that a large majority of county officials said they have adequate information about the services provided at the facilities and the length of the programs. It is common for larger facilities to distribute brochures that describe their programs and approaches.

On the other hand, most county officials said that they lack enough information about the extent to which residents’ problem behaviors recur following discharge from residential programs. 25 Measures of ‘‘recidivism’’ should be interpreted with care because they may reflect factors other than the success of the residential programs--such as the adequacy of ‘‘aftercare’’ services. Still, counties appear to want better information about juvenile recidivism, as well as information about the extent of ‘‘runaways’’ from facilities and the extent to which children placed in facilities complete the programs.

State law does not require case plans for delinquent juveniles.

County officials want more information about residential facilities.

23 Minn. Stat. §260.191, subd. 1e and Minn. Stat. §257.071, subd. 1. The law requires development of the plan following placement in a ‘‘residential facility,’’ defined as ‘‘any group home, family foster home or other publicly supported out-of-home residential facility, including any out-of-home residential facility under contract with the state, county or other political subdivision, or any agency thereof, to provide ’’ 24-hour-a-day services or care. The plan is defined as ‘‘a written document which is ordered by the court or which is prepared by the social service agency responsible for the residential facility placement.’’

24 In July 1998, the departments drafted rules on facility licensure and program certification related to residential care for children. The rules distinguish a ‘‘case plan’’ (developed by the placing agency) from a ‘‘treatment plan’’ (developed by the licensed residential facility). Although county correction agencies are not required by law to develop case plans, many such agencies assess the service needs of delinquent juveniles--some through standardized instruments, and others informally.

COURT JUSTIFICATIONS FOR PLACEMENTS

In all dispositions related to CHIPS and delinquency petitions, state law requires the court to state (a) why the best interests of the child are served by the disposition, and (b) what alternative dispositions were considered by the court and why such dispositions were not appropriate in this case. 26

We examined 164 case files during site visits to seven counties--among other reasons, to see whether the courts provided clear justifications for the placements they ordered. Some court orders succinctly articulated the reasons for placement, such as the following:

‘‘[ The mother’s] family of origin was considered and found inappropriate [for placement] because of previous abuse issues. The alleged father has not come forward with any possible family home. Therefore, no relatives are available at this time. The older two children are placed in the [name] foster home and this home was considered and rejected because [foster mother’s name] is unwilling to have another infant placed in her home. . . [The child’s mother] is unable to care for children on a consistent basis. [She] is unable to place the needs of her children above her own needs. . . [She] has had over three years to make these changes and is unable to make them.’’

Table 2.4: County Officials’ Perceptions About the Adequacy of Program Information Available When Making Placement Decisions

Corrections Supervisors (N = 82) Human Services Directors (N = 84) Sometimes, Sometimes, Do county staff have Usually or Rarely, or Usually or Rarely, or sufficient information about: Always Never Always Never Services provided by programs? 96% 4% 81% 18% Length of programs? 99 1 79 20 Educational progress of residents while in programs? 89 11 60 38 Percentage of residents who complete programs? 50 50 31 62 Number of residents who run away during placement? 39 59 19 73 Recurrence of residents’ problem behaviors following 37 61 13 82 placement?
NOTE: Percentages of officials who responded "don’t know" are not shown. SOURCE: Program Evaluation Division surveys, August 1998.

PLACEMENT DECISIONS

25 Measures of subsequent delinquency, school attendance, and living arrangements are among the ways that post-discharge behaviors could be evaluated.

26 Minn. Stat. §260.185, subd. 1 and §260.191, subd. 1a.

But we observed that the courts usually were not this explicit about the placement decisions they made. We found that:

Many court disposition orders contained little information about why the placement was ordered and why alternatives to this placement were rejected.

Some court justifications for placement merely restated legal requirements and did not provide specific information that explained why the placement was appropriate in the case at hand. For example:

‘‘Reasonable efforts were made to prevent out-of-home placement, but these efforts were unsuccessful. The Court finds placement outside the home is in the best interests of the child and represents the least restrictive alternative available. Moreover, the Court believes placement outside the home is necessary to return the child to law-abiding behavior.’’

Some disposition orders provided no indication of alternatives that were considered, and some only gave very general justifications for placement (such as ‘‘serious criminal offense,’’ ‘‘assaultive behavior,’’ and ‘‘continued delinquent behavior and beyond parental control ’’). In cases where juveniles were placed outside Minnesota, the courts rarely discussed reasons that Minnesota placement options were rejected--although they sometimes offered reasons for liking the chosen facility. For example:

The court said that an American Indian juvenile’s ‘‘continued problematic and dangerous behavior can best be addressed by confronting his thinking and poor decision making. Grehill Academy [in Iowa] is a program designed to specifically confront thinking patterns and thus the resulting behavior patterns.’’ The court also said that it liked the facility’s security level and its program of graduated privileges and rewards (perhaps reflecting the juvenile’s history of escapes from other facilities), and it noted that this facility was the preferred choice of the juvenile’s tribe. The disposition did not discuss any in-state options that were considered.

Often, we observed that county staff provided the courts with detailed information about the juveniles and useful discussions about placement options. But the courts often receive recommendations from various sources, such as probation agencies, social services agencies, public defenders, and county attorneys. For this reason, it is important for the court to offer a clear written justification for its final decision. In addition, despite some of the useful staff disposition recommendations that we read, judges told us that they often felt they needed better information from staff regarding service options. Forty-one percent of judges we surveyed said that they ‘‘sometimes, rarely, or never ’’ received adequate information from county staff on the advantages and disadvantages of alternative placement and service options. 27

OUT-OF-STATE PLACEMENT

In recent years, legislators have expressed concern about the number of juveniles--especially delinquent juveniles--placed in residential facilities outside of Minnesota. They have wondered whether out-of-state residential facilities can effectively work with a juvenile’s family and arrange for appropriate ‘‘aftercare’’ services in the juvenile’s home community following placement. They have also questioned whether counties’ use of out-of-state facilities reflects service deficiencies among Minnesota facilities. 28 In 1994, the Legislature prohibited courts from placing pre-adjudicated delinquents, adjudicated delinquents, or ‘‘extended jurisdiction juveniles ’’ (EJJs) in residential or detention facilities outside of Minnesota unless the facilities were certified by the Commissioner of Corrections. 29 Specifically, the commissioner must certify that these facilities meet or exceed DHS or DOC standards for Minnesota residential facilities. In 1997, the Legislature adopted the following policy:

It is the policy of this state that delinquent juveniles be supervised and programmed for within the state. Courts are requested, to the greatest extent possible and when in the best interests of the child, to place these juveniles within the state. 30

Legislators expressed to us an interest in knowing more about out-of-state placements, including the reasons for these placements. We examined recent statewide information on the number of juveniles placed out of state, and we talked with county staff about why they use (or do not use) out-of-state facilities. We also reviewed court dispositions for 73 of the 273 juvenile offenders placed out of state in fiscal year 1997 to better understand the reasons for their placements, but (as noted in the previous section) we found that judges often did not explicitly indicate why Minnesota facilities were not selected. We found that:

Legislators have been concerned about the number of out-of-state placements.

27 Twenty-one percent said that they ‘‘always or almost always’’ received adequate information, and 36 percent said they ‘‘usually’’ did.

28 Legislators have also been concerned that children placed in Minnesota from other states can impose financial burdens on Minnesota school districts. The 1998 Legislature required correctional facilities to have agreements with placing counties to pay for educational costs.

29 Minn. Laws (1994), ch. 576, sec. 27. As of early 1998, 34 facilities were certified. ‘‘Extended jurisdiction juveniles’’ (or EJJs) are felony offenders for whom the court has executed a juvenile disposition, along with a stayed adult criminal sentence. The EJJ designation is a sort of ‘‘last chance’’ given by the courts before imposing adult sentences on juvenile offenders. The juvenile court maintains jurisdiction over EJJs until they reach age 21, unless the court terminates its jurisdiction or the sentences expires before that time. If the offender commits a new offense or violates the conditions of the stayed adult sentence, the adult sanctions may be executed.

30 Minn. Laws (1997), ch. 239, art. 9, sec. 14.

Out-of-state placements represented about 5 percent of the total days of care that Minnesota children spent in out-of-home placement in 1997. 31

The vast majority of children placed out of state were in (1) foster homes, or (2) facilities certified by Minnesota’s Commissioner of Corrections to serve delinquent juveniles. 32 Department of Human Services data indicated that 286 placements were in out-of-state foster care during 1997, or 2.2 percent of all Minnesota placements in foster care. About two-thirds of these out-of-state foster care placements were with relatives, and 80 percent were from Hennepin and Ramsey counties.

Our analysis of data collected by the Department of Corrections showed that 422 delinquent Minnesota juveniles or EJJs were in out-of-state placement during fiscal year 1997, at an estimated total cost of $9.0 million (or $116 per day in placement, excluding education costs). 33 On an average day in 1997, there were 211 delinquent juveniles or EJJs from Minnesota in non-Minnesota facilities. 34 The states whose facilities were used most were South Dakota, Iowa, and Colorado.

For each judicial district, we reviewed data on placements at non-Minnesota facilities certified by the Commissioner of Corrections. These facilities mostly served delinquent juveniles and extended jurisdiction juveniles, but they also served some juveniles whose placements were not based on a delinquency adjudication. 35 For fiscal years 1996-97, we found that:

Ramsey County (Judicial District 2), with 10 percent of the state’s juvenile population, accounted for 35 percent of the days spent by Minnesota juveniles at non-Minnesota facilities certified by the Commissioner of Corrections.

Table 2.5 shows several measures of each judicial district’s use of out-of-state placement at facilities certified by the Commissioner of Corrections. Among the districts, Ramsey County (District 2) had the most total days in out-of-state placement (53,994) and the most days in out-of-state placement per 1,000 children in the county (445). In addition, out-of-state placements accounted for 6 percent of all of Ramsey County’s days in placement, more than any other judicial district. A majority of Ramsey County’s out-of-state placements were at South Dakota facilities. Our review of county records indicated that the delinquent juveniles placed by Ramsey County included many who had committed serious offenses or had extensive delinquency histories, as well as some who were placed following first adjudications or short delinquency histories. Ramsey County corrections staff told us that they have considered out-of-state facilities to be simply another placement option--not necessarily a ‘‘last resort’’ or a placement option for only the most serious offenders. Also, Ramsey County has used some programs because they are less expensive than in-state counterparts or offer longer programs. 36 In addition, ‘‘extended jurisdiction juveniles’’--who could have an adult criminal sentence imposed if they violate the terms of their juvenile adjudication--are sometimes placed out of state by Ramsey County to reduce the temptation of running away or associating with friends in their home community. County staff acknowledged that it is difficult to get families to participate in programs that are outside Minnesota, but they said that fostering family involvement can be difficult even when juveniles are placed within Ramsey County or at other Minnesota facilities. 37

Minnesota spent $9 million for out-of-state placement of delinquents in 1997, excluding education costs.

31 Based on an analysis of placements paid for by social services agencies, which accounted for 91 percent of days of placement in 1997.

32 Some DOC-certified out-of-state facilities are inspected by the Department of Human Services, not by DOC.

33 During 1997, there were another 48 non-delinquent juveniles placed at facilities certified by Minnesota to serve pre-adjudicated delinquents, adjudicated delinquents, or extended jurisdiction juveniles. The 1997 costs to serve these juveniles totalled about $761,000.

34 In addition, there was an average daily population of 16 non-delinquent juveniles in nonMinnesota facilities certified by the Commissioner of Corrections.

35 Of the Minnesota juveniles in placement at non-Minnesota facilities certified to serve delinquents, 10 percent were non-delinquents, and 14 percent were extended jurisdiction juveniles.

Table 2.5: Use of Out-of-State Placement, by Minnesota Judicial District, Fiscal Years 1996-97

Judicial District (and Location in Minnesota) a

Total Days in Out-of-State

Placement, 1996-97

District’s Percentage of the State’s Total

Out-of-State Placement Days, 1996-97

Out-of-State Placement Days

Per 1,000 Children in Population,

1996-97 Days in Out-of-State Placement as a Percentage of District’s Total Placement Days, 1997

Average Distance Placed From Home (in miles), 1996-97

Out-of-State Placements b

First (southeast) 4,140 2.7% 24 1.0% 302 Second (Ramsey County) 53,994 35.3 445 6.2 344 Third (southeast) 6,917 4.5 63 2.4 222 Fourth (Hennepin County) 49,132 32.1 201 2.7 590 Fifth (southwest) 21,980 14.4 300 4.8 97 Sixth (northeast) 1,519 1.0 26 0.4 562 Seventh (west-central) 5,926 3.9 51 0.9 286 Eighth (west-central) 4,932 3.2 109 2.8 176 Ninth (northwest) 1,287 0.8 15 0.3 360 Tenth (east-central) 3,021 2.0 14 0.5 463

STATE 152,848 100.0% 122 2.5% 351 a See map on page 48.

b Placement distances were straight-line distances between the county placement agency and the ou t-of-state facility. The averages are based on all placements that were active during some part of fiscal years 1996 or 1997. SOURCE: Program Evaluation Division analysis of Department of Corrections out-of-state pla cement data. Total placement days are based on DHS placement data, DOC data on Red Wing and Sauk Centre placements, and Program Evalua tion Division June and July 1998 surveys of counties.

PLACEMENT DECISIONS

36 According to information collected by the Department of Corrections, a placement at South Dakota’s Chamberlain Academy or Springfield Academy cost $75 per day during fiscal year 1997, excluding education costs. Ramsey County had 150 juveniles at these two facilities for at least a portion of fiscal years 1996 or 1997. Among out-of-state facilities used by Minnesota counties, $75 was the lowest per diem, and few residential corrections facilities in Minnesota had rates below $100 per day.

37 Partly at the urging of Ramsey County staff, the parent company of the facilities the county most often uses in South Dakota opened a facility in southern Minnesota in 1998. It is licensed to have 150 beds, and its program and costs are similar to those of the South Dakota facilities. The development of this facility might enable Ramsey County to reduce out-of-state placements in the future.

Table 2.5 also indicates the average distance from home for juveniles placed in non-Minnesota facilities certified by the commissioner of corrections. During fiscal years 1996-97,

Juveniles from Hennepin County (Judicial District 4) were placed farther from home, on average, than juveniles from Minnesota’s other judicial districts.

Hennepin County has not placed juvenile offenders out of state as often as Ramsey County, but it has tended to place them farther away. Hennepin officials told us that they use out-of-state facilities primarily to meet specialized service needs--for example, for offenders with serious mental health problems, strong gang influences, developmental disabilities, or histories of serious violence. The majority of Hennepin County’s out-of-state placements have been in Colorado and Pennsylvania facilities. We found that the county probation office’s 1996 disposition guidelines did not designate any Minnesota facilities as placement options for offenders who committed the most serious offenses and were considered to be high risks for new offenses. However, Hennepin County staff told us they have increasingly used the Department of Corrections’ Red Wing facility for serious offenders in recent months, due to improvements in that facility’s program and security level.

Table 2.5 indicates that judicial districts other than Hennepin and Ramsey counties have varied considerably in their use of out-of-state placement. We observed that:

Judicial District 5 (southwestern Minnesota) placed more juveniles at certified out-of-state facilities than other Greater Minnesota judicial districts, but its average out-of-state placement was less than 100 miles from home.

Judicial District 5 includes 15 counties in southwestern Minnesota. These counties have frequently used out-of-state placement, but they have mostly used nearby South Dakota or Iowa facilities. For example, there were six juveniles from Rock County who were in placement at DOC-certified facilities outside of Minnesota during 1996-97, but all of them were in Sioux Falls, South Dakota residences about 30 miles away. 38 In some cases, out-of-state placement options may be closer to home than available options in Minnesota, perhaps increasing the possibility of family visitation.

Thirty-six of Minnesota’s 87 counties had no juvenile delinquents in out-of-state placement at any time during fiscal years 1996-97. Some large counties--such as Anoka, Dakota, Olmsted, and St. Louis--have placed very few children out of state in recent years. Staff in these counties told us that this reflects a preference for keeping juveniles closer to their families or staff’s perception that in-state (and often local) services could address the needs of even the most serious offenders.

Some out-of-state placements are closer than in-state placement options.

38 As noted in the previous footnote, the 1998 opening of a large facility in southern Minnesota might affect the extent of out-of-state placement from this region in the future.

Our surveys indicated some concern by county officials and judges with the ability of Minnesota residential facilities to meet the needs of children with the most serious problems. As shown in Figure 2.2, about half of the judges said that there were ‘‘sometimes, rarely, or never ’’ sufficient residential options for these children within Minnesota. During our review of juvenile offender court dispositions that resulted in out-of-state placements, the courts often did not appear to consider the option of placing offenders at Minnesota’s ‘‘last resort’’ facilities--the state-operated facilities at Red Wing and Sauk Centre--as alternatives to out-of-state placements. 39 However, the Department of Corrections has taken steps to improve programming at these two facilities during the past two years, based on a statewide juvenile services survey and meetings with corrections and court staff in all judicial districts. It remains to be seen whether improvements to these and other programs will reduce county reliance on out-of-state placement.

Overall, there is considerable variation in the use of out-of-state placement by counties, but the large majority of out-of-state placements have been made by Hennepin and Ramsey counties. Some out-of-state placements are very far away from juveniles’ homes, while others are quite close. Even if Minnesota expanded residential program options within the state, some counties would probably continue to prefer placing certain juveniles in settings that are far removed from delinquent peers and family problems. Also, some counties would likely continue to prefer out-of-state facilities if they continue to perceive that these facilities provide services comparable to Minnesota facilities at lower costs.

10 35 40 11 0 10 20 30 40 Always or almost always

Usually Sometimes Rarely or never Percent

Figure 2.2: Judges’ Opinions About Whether Minnesota Has Adequate Placement Options For Children With the Most Serious Problems

NOTE: The survey question asked: “Within Minnesota, are there sufficient residential options to enable you to adequately place and serve the children with the most serious problems?”

SOURCE: Program Evaluation Division survey of district court judges, August 1998 (N= 143).

Judges often have problems finding Minnesota placement options for difficult juveniles.

39 Through 1998, neither of these facilities could refuse to admit juveniles committed to the Commissioner of Corrections. In January 1999, state law required the Sauk Centre facility to cease admitting juveniles committed to the commissioner.

PLACEMENT CONSIDERATIONS

People we interviewed early in our study told us that they believed that factors other than child or family characteristics contributed to out-of-home placements. For instance, county officials told us that placements may sometimes be made to get problem children ‘‘out of the hair ’’ of local officials for a while. A former human services director told us that his staff used to prefer making out-of-home placements because placements could be arranged with a single phone call, while it was more difficult to find and coordinate a package of community-based, non-residential services.

If such considerations do play an important role, it would be difficult to determine this through a review of official case records. We chose instead to directly ask county human services directors and juvenile corrections supervisors about the importance of various factors that may affect the placement process. We recognize that officials whose agencies are responsible for making placements might not have been fully candid in our surveys about the reasons for placement, but we think their survey responses provide a useful starting point. Table 2.6 shows the extent to which officials said that various placement rationales ‘‘were important justifications ’’ for placements in which their agencies were directly involved.

Most of the respondents considered residential settings to be ‘‘more structured ’’ than non-residential settings and said that this helped influence their placement decisions. Few agency officials said they usually considered residential services to be attractive for cost reasons or the ease of making placements. Consistent with the populations of children they serve, human services directors said they were more apt to make placements to protect children from abuse and neglect, while corrections supervisors were more likely to make placements aimed at protecting the community and sending a message to the child about his or her behaviors.

Counties like the structured environments that facilities provide but not the costs of residential services.

40 Annie Woodley Brown and Barbara Bailey-Etta, ‘‘An Out-of-Home Care System in Crisis: Implications for African American Children in the Child Welfare System,’’ Child Welfare 76, no. 1 (January-February 1997), 66.

Another factor that some people believe may be a consideration in placement is race and ethnicity. Non-Hispanic whites comprise 88 percent of children under age 18 in Minnesota, but we note in Chapter 4 that non-Hispanic whites comprised 61 percent of the children in out-of-home placement in 1997. Furthermore, only 43 percent of Minnesota juveniles in out-of-state placement in 1997 were non-Hispanic whites, and only 18 percent of ‘‘extended jurisdiction juveniles’’ (EJJs) placed out-of-state were non-Hispanic whites. In our view, the possible causes of these patterns are complex and difficult to judge conclusively. For example, one recent analysis concluded that ‘‘an array of problems--including racism, poverty, inadequate housing, substance abuse, HIV/ AIDS, teenage pregnancy, incarceration, lack of appropriate social support systems, and violence--all combine to account for the growing number of African American children needing out-of-home care.’’ 40 Some people told us that disproportionate placement of minority children may reflect bias or a lack of cultural awareness on the part of staff who work in the placement process. Our surveys indicated that:

Ten percent of judges, 8 percent of human services directors, and 9 percent of juvenile corrections supervisors said that they believe that minority children are ‘‘sometimes, usually, or always’’ more likely to be placed out-of-home (or recommended for placement) than Caucasian children with identical circumstances. 41

Figure 2.6: "Important Justifications" for Child Placements, According to County Officials

Percentage of county officials who said this was an important reason for placements in which their agency was involved last year County Human Services Directors (N= 84) County Corrections Supervisors (N= 82) Residential placement: Usually orAlways Sometimes Usually or Always Sometimes Protected people in the community from harm 46% 43% 87% 13% Protected the child from abuse or neglect 68 31 17 45 Got the child away from a dysfunctional home or neighborhood 24 50 40 48

Sent a message to the child about the seriousness of his/ her behaviors 24 56 76 20

Provided a more structured setting for learning/ treatment than non-residential services

50 44 90 10 Was more readily available than non-residential services 10 37 24 29

Required less time to arrange than non-residential services 6 14 2 10

Got the child "out of the hair" of local officials 6 25 5 24 Was more cost-effective than non-residential services 4 15 17 28

NOTE: Percentages of officials who responded "rarely or never" and "don’t know" are not shown. SOURCE: Program Evaluation Division surveys, August 1998.

Most judges and county officials do not think the placement process is biased.

41 Sixty-eight percent of human services directors and 77 percent of corrections supervisors responded ‘‘rarely or never.’’ Twenty-three percent of human services directors and 13 percent of corrections supervisors responded ‘‘don’t know.’’

Seven percent of human services directors and 4 percent of corrections supervisors said that prejudice in the placement process is an important reason for the disproportionate placement of minority children. 42

While the majority of county officials said that racism has rarely directly affected placement decisions, some of them told us that racism may have affected placements in other ways. For example, some officials told us that that long-term discrimination may have contributed to higher poverty among minority groups, and placement rates tend to be positively related to poverty levels (see Chapter 3).

SUMMARY

The courts and counties have considerable discretion to determine which circumstances merit child placement, and this may contribute to a perception of inconsistency in placement decisions reported by many county officials. Usually counties and courts do not have written criteria that indicate when placement may be appropriate, and sometimes placement decisions are made with little or no review from multi-disciplinary screening teams or agency supervisors. Furthermore, many court dispositions provide little discussion of the factors that persuaded the court to favor out-of-home placement or about the reasons that other options were rejected.

Many county officials and judges believe that placement decisions are based on a sufficient consideration of children’s needs, but there appears to be room for improvement. County staff usually have what they consider to be sufficient information on services offered at residential programs, but they usually do not think they have enough information about program outcomes.

42 Seventy-four percent of corrections supervisors disagreed with a statement that prejudice is an important reason, 7 percent neither agreed nor disagreed, and 16 percent responded ‘‘don’t know.’’ Sixty-five percent of human services directors disagreed that prejudice is an important reason, 11 percent neither agreed nor disagreed, and 18 percent responded ‘‘don’t know.’’

Placement Spending and Funding CHAPTER 3

M any counties have expressed concerns to legislators about growth in out-of-home placement costs. In particular, they have observed that county property taxes pay for a significant share of child placements ordered by the courts. Meanwhile, however, legislators have had incomplete information about out-of-home placement costs and spending trends. We asked:

How much do Minnesota public agencies spend on juvenile out-of-home placements and family preservation services?

How has placement spending changed over the last five years?

How much does placement spending vary by county, and what accounts for this variation?

What levels of government pay for out-of-home placements, and how does the financing vary by county?

Overall, we found that Minnesota public agencies spent more than $200 million for out-of-home placements in 1997, and the costs increased considerably faster than inflation over the last five years. However, placement spending per resident under age 18 varied dramatically among counties. In addition, spending increased considerably over the last five years in some counties but declined significantly in others. While some variation in county spending is related to the needs of each county’s population, some of it is explained by the child placement philosophies and practices of the counties and courts.

DATA SOURCES

In some counties, the department of social services pays for almost all placements (including correctional placements); in other counties, a combination of departments (social services, corrections, and others) pay for the placements. In 1997, county social services budgets paid for 76 percent of Minnesota’s total placement spending (excluding education costs). We analyzed this portion of spending by examining data from the Department of Human Services’ Social Services Expenditure and Grant Reconciliation (SEAGR) information system.

Statewide information about out-of-home placement spending is incomplete.

In addition, to develop complete statewide information on child placement spending, we obtained information from four other sources. First, we sent surveys to each county in June and July 1998 to collect information on spending by corrections, law enforcement, and other non-social services agencies. In about half of the counties, non-social service agencies pay for certain correctional placements, such as juveniles who are placed in a facility run by the county.

Second, we obtained information on placement costs paid directly by the Minnesota Department of Corrections (DOC). 1 Until December 31, 1998, the state paid for the placement of children committed to the Commissioner of Corrections from the 56 counties that do not participate in Minnesota’s Community Corrections Act (CCA). Furthermore, the per diems paid by CCA counties for children committed to the commissioner only covered 81 percent of the placements’ total cost in 1997 (excluding education), and the state paid the difference. 2 Third, we obtained Department of Human Services (DHS) information on spending for residential chemical dependency treatment through Minnesota’s Consolidated Chemical Dependency Treatment Fund, which pools local, state, and federal funds. Finally, Medicaid paid for some children’s mental health services at two state-run acute care hospitals which are also licensed as Rule 5 facilities, the Brainerd and Willmar regional treatment centers. We included the Medicaid spending in our statewide analysis but not the county-by-county analysis. 3

Our estimates of county and state out-of-home placement spending do not include education costs. In general, a child’s home school district remains responsible for the cost of the child’s education, even if the child is placed in another district. There are no statewide data that indicate whether (and to what extent) the cost of educating students in residential placements is greater than education costs incurred while they live at home, but some school districts have expressed concerns about higher costs for children in placement. A House-Senate task force examined placement-related education costs in 1998, so we chose not to duplicate their efforts.

In addition, our estimates of placement spending do not include the cost of services that were not covered by the per diem billing used by the residential facilities. If a facility provided or contracted mental health or chemical dependency treatment and the cost was included in its per diem, we included it in our spending estimates. If a child was placed in a foster home and received counseling on an outpatient basis in the community, we did not include the cost in our spending estimates.

We collected data from several sources.

1 Based on the number of days of care provided to each juvenile committed to the Commissioner of Corrections by the counties, we allocated Red Wing’s and Sauk Centre’s costs (excluding education costs and hold/ detention placements) to each county. For counties paying per diems, we subtracted their payments from the allocation.

2 Program Evaluation Division analysis of data from the Department of Corrections. To be consistent with other facilities, we applied all of the county per diem payments toward noneducation costs.

3 We did not collect data from acute care hospitals that are not licensed under Rule 5, such as the Wilson Center.

The data presented in this chapter generally do not distinguish between the various types of correctional facilities, such as pre-disposition detention, post-disposition residential, secure, and non-secure facilities. Existing county-reported data on correctional facility costs, days of care, and number of placements have not adequately distinguished between these types of categories. As we note later, it is important to consider that the statewide averages we present for correctional facility length of stay and cost per placement are based on a wide variety of placements, ranging from single day detention stays to multi-month residential stays.

STATEWIDE SPENDING

Using information from state data sources and our surveys of local agencies, we estimated that:

Minnesota public agencies spent $225 million in 1997 for children placed out of their homes, not including education costs.

As Table 3.1 shows, most of the spending was for foster homes ($ 78 million), correctional facilities ($ 59 million), and Rule 5 residential mental health treatment facilities ($ 47 million). (It should be noted that these figures refer to types of facilities and not the reason for a placement. A juvenile delinquent placed by the corrections system can be sent to a corrections or Rule 5 facility or even a foster home licensed by DHS.)

Three factors determine these levels of spending: 1) the number of children who were placed, 2) the length of the stay, and 3) the cost per day of the placements. 4 Across all types of facilities, there were over 36,000 placements in Minnesota in 1997 with an average length of stay of 93 days and an average cost of $66 per day. Even though foster homes accounted for only 21 percent of the placements and cost an average of only $35 a day, they made up 34 percent of total placement spending because the average length of stay was long (285 days).

Correctional facilities made up 45 percent of all placements and cost an average of $129 a day, but they accounted for only 26 percent of all the spending because many of the stays were short, such as pre-adjudication detention (which often lasts less than 72 hours) and short-term consequence programs (some lasting less than a month). However, post-adjudication residential placements often cost $10,000 to $40,000 each and last for several months. Unfortunately, available data did not allow us to separate the costs, placements, and days of care of detention stays from longer-term correctional placements. As a result, the average length of stay and cost per placement shown in Table 3.1 are not representative of ‘‘typical’’ corrections placements. For example, data that we collected from facilities run by Hennepin and Ramsey counties indicate that the average detention stay in these facilities was seven days. In contrast, the average post-adjudication stay was 115 days.

Out-of-home placements cost taxpayers more than $200 million in 1997.

4 For our analysis, we combined data from several sources, the largest being DHS’ Social Services Expenditure and Grant Reconciliation (SEAGR) reports and substitute care database. The SEAGR reports classify placement spending into five categories --shelter, foster care, Rule 8, Rule 5, an d corrections. The substitute care database classifies placements and days of care into categories th at are somewhat similar to the SEAGR categories but not identical. While we tried to match spending with placements and days of care as closely as possible, errors still remain.

Finally, the average Rule 5 placement was relatively long (168 days) and expensive ($ 179 per day). As a result, Rule 5 placements accounted for 21 percent of the spending even though they accounted for only 4 percent of all placements.

STATEWIDE TRENDS

We used data from state and county sources to examine spending trends between 1992 and 1997. We found that:

Adjusted for inflation, public spending for out-of-home placements per resident under age 18 increased 22 percent between 1992 and 1997.

Table 3.1: Juvenile Out-of-Home Placement Spending by Type of Facility, 1997

Type of Facility Total Spending

Number of Placements Made in 1997

Estimated Average Length of Stay

(Days) a Days of Care

Occurring in 1997

Average Cost per

Day b Estimated

Average Cost per Placement b

Shelters c $16,792,531 7,521 25 188,083 $89 $2,233 Family foster homes 77,512,966 7,811 285 2,223,102 35 9,924 Rule 8 group homes 17,303,784 1,463 119 174,580 99 11,828 Rule 5 facilities d 46,705,553 1,551 168 260,533 179 30,113 Correctional facilities 59,389,793 16,211 28 460,880 129 3,664 Chemical dependency 7,309,808 1,346 40 54,031 135 5,431 facilities

Other e 502 31,108 TOTAL $225,014,435 36,405 93 3,392,317 $66 $6,181 a Program Evaluation Division estimate based on the number of days of care in 1997 divided by the number of placements made in 1997.

b These figures may slightly overestimate the actual cost per day and per placement because we could not allocate some placements and days of care to the facility types. c We included shelter placements at Rule 5 facilities in the shelter category rather than in th e Rule 5 category.

d Includes placements paid by Medicaid at Brainerd and Willmar regional treatment centers for state fiscal year 1997. e Placements and days of care that could not be allocated to the facility types.

SOURCE: Program Evaluation Division analysis of data from the Minnesota Department of Huma n Services’ Social Service Expenditure and Grant Reconciliation reports, substitute care database, Consolidated Chemical Depend ency Treatment Fund, and unpublished Medi caid data; Department of Corrections’ data on Red Wing and Sauk Centre placements and spendin g; and the Program Evaluation Division’s surveys of counties (June and July 1998).

During this period, the number of placements per resident under age 18 increased by 7 percent, the average length of stay decreased by 3 percent, and the average cost per day after adjusting for inflation increased by 15 percent. 5

As Table 3.2 shows, the different types of facilities experienced different trends. 6 While placement spending at correctional facilities increased by 39 percent, spending for shelter facilities declined by 3 percent. Of the three types of placements that account for most of the spending in the state --foster homes, Rule 5, and corrections --spending at Rule 5 and corrections facilities increased by more than the statewide total. It would be useful to examine trends in placements, lengths of stay, and cost per day for each of the six placement types, but the available data did not allow us to do this. For example, we do not know whether the statewide decline in average length of stay reflected shorter average stays in the individual categories of facilities or an increased reliance by counties on certain short-term placement types (such as detention). 7

Table 3.2: Change in Juvenile Out-of-Home Placement Spending by Type of Facility, 1992-97

Type of Facility Percentage Change in

Inflation-Adjusted Spending per Resident Under Age 18

Shelters -3% Family foster homes 14 Rule 8 group homes 21 Rule 5 residential treatment facilities a 26 Chemical dependency facilities 37 Correctional facilities 39

Total 22% NOTE: Spending is adjusted for inflation using the Chain-Type Price Index --State and Local Consumption Expenditures and Gross Investments.

a Includes Medicaid payments at Brainerd and Willmar Regional Treatment Centers. SOURCE: Program Evaluation Division analysis of data from the Minnesota Department of Human Services’ Social Service Expenditure and Grant Reconciliation reports, Consolidated Chemical Dependency Treatment Fund, and an unpublished table titled "DHS Funding for Mental Health Services-Children Only;" Department of Corrections’ data on Red Wing and Sauk Centre placements and spending; and the Program Evaluation Division’s surveys of counties (June and July 1998).

Spending for placements has grown faster than inflation, especially in correctional facilities.

5 While the 22 percent increase in spending included all sources of funding and all counties, the figures for placements, average length of stay, and cost per day excluded placements paid by Medicaid (because data on the number of placements and days of care were not available for 1992) and placements made by 12 counties (because complete data for 1992 were not available.)

6 It should be noted that facilities sometimes change their licenses, for example from a Rule 5 facility to a corrections facility or a Rule 8 home. Some of the change in spending for the facilit y types is explained by a change in licenses rather than a change in placement policies. In addition, the implementation of higher standards of care for Rule 5 facilities in 1996 may explain some of the increase in Rule 5 spending.

7 It would be interesting to know more about the year-to-year changes in placement spending, but comprehensive statewide data are not available. For the largest category of placements --those paid for by social services --we found that most of the spending increase occurred in 1992-94, not in 1994-1997.

When we asked human services directors and corrections supervisors about future spending, we found that:

A majority of county officials believe that their county’s placement spending will increase in the next 3 years.

Among county human service directors, 76 percent anticipate an increase in their agency’s out-of-home placement spending in the next three years (not including inflation), 10 percent expect a decrease, and 13 percent think spending will stay the same. Among county corrections supervisors, 54 percent expect an increase in their agency’s placement spending, 5 percent anticipate a decrease, and 28 percent think that spending will stay the same. 8

Some county officials believe it is difficult to control increases in placement spending because the courts do not have to pay for the placements they order. In our surveys, 77 percent of county human services directors disagreed with the following statement: ‘‘Judges fully understand the costs of the placements and services they order.’’ 9

COUNTY VARIATION

Table 3.3 shows counties with the highest and lowest 1997 placement spending per resident under age 18. We found that:

Spending for out-of-home placements varied widely among counties in 1997. On average, counties spent $173 per resident under the age of 18, but individual counties ranged from $25 to $322.

To examine the variation more closely, we separated the 84 county social service agencies into high, medium, and low spending categories, each with 28 counties. 10 As Table 3.4 shows, low spending counties placed proportionally fewer children for significantly shorter periods of time at less expensive facilities than high spending counties. High and medium spending counties were largely distinguished by the number of placements made per capita, not the length of placements or the average cost per day. The main difference between medium and low spending counties was the average length of stay in placement. In addition, we found that high spending counties (as a group) had substantially higher spending per capita than low spending counties in each of the six categories of facilities that we examined. 11

County spending for placements varies widely. Many county officials expect their placement spending to increase.

8 Program Evaluation Division survey of county human service directors and corrections supervisors, August 1998 (N= 84 human services directors and 82 corrections supervisors). One director and 11 supervisors said they had no basis for making a reasoned estimate. The median expected spending increase was 11 to 15 percent for human service directors and 6 to 10 percent for corrections supervisors.

9 Ten percent of the directors agreed with the statement. Among corrections supervisors, 45 percent agreed with the statement and 37 percent disagreed.

10 Lincoln, Lyon, and Murray counties operate under a single social service administration, and Faribault and Martin counties operate under another. For these two groups of counties, we also aggregated their spending from non-social services sources.

Table 3.3: 1997 Spending for Juvenile Out-of-Home Placements, Selected Counties

1997 Spending per Resident Under Age 18 1997 Spending per

Child in Poverty

High spending counties

Hennepin $322 $1,696 Cass 302 1,064 Polk 282 1,225 Watonwan 274 1,654 Jackson 269 1,652

Low spending counties

Washington $64 $1,121 Scott 50 945 Dakota 49 712 Traverse 45 230 Red Lake 25 158

Statewide average $173 $1,171

SOURCE: Program Evaluation Division analysis of data from the Minnesota Department of Human Services’ Social Service Expenditure and Grant Reconciliation reports and Consolidated Chemical Dependency Treatment Fund; Department of Corrections’ data on Red Wing and Sauk Centre placements and spending; and the Program Evaluation Division’s surveys of counties (June and July 1998).

Table 3.4: Characteristics of 1997 Juvenile Out-of-Home Placements in High, Medium, and Low Spending Counties

Number 1997

Spending per Resident Under Age 18

1997 Placements per 1,000 Residents under Age 18

Estimated Average Length of Stay (days) a

Average Cost per

Day Counties with high spending rates b 28 $265 39 103 $66 Counties with medium spending rates c 28 141 21 104 63 Counties with low spending rates d 28 80 20 68 60

Statewide average 84 $173 29 93 $64 a Program Evaluation Division estimate based on the number of days of care in 1997 divided by the numb er of placements made in 1997. b Counties that spent $188.0 or more per resident under age 18. c Counties that spent $108.7 or more per resident under 18 but less than $188.0 per resident under 18. d Counties that spent less than $108.7 per resident under 18.

SOURCE: Program Evaluation Division analysis of data from the Minnesota Department of Human Services ’ Social Service Expenditure and Grant Reconciliation reports, substitute care database, and Consolidated Chemical Dependency Treatment Fund; Department of Corrections’ data on Red Wing and Sauk Centre placements and spending; and the Program Evaluation Division’s surveys of counties (June and July 1998).

PLACEMENT SPENDING AND FUNDING

Figure 3.1 shows counties with low, medium, and high spending per resident under age 18. Two counties (Hennepin and Ramsey) accounted for 69 percent of the spending among the 28 ‘‘high spending’’ counties, and St. Louis County accounted for another 7 percent. Eight Twin Cities suburban counties (Anoka, Carver, Chisago, Dakota, Scott, Sherburne, Washington, and Wright) accounted for 64 percent of the spending in the ‘‘low spending’’ category, and Olmsted and Stearns counties accounted for another 18 percent. Later in this chapter we discuss the relationship between a county’s socio-economic makeup and its level of spending.

Figure 3.1: 1997 Out-of-Home Placement Spending Per Resident Under Age 18, by County

SOURCE: Program Evaluation Division analysis of data from the Department of Human Services , Department of Corrections, and counties.

Spending per Capita $188.0 to $322.0 $108.7 to $187.9 $25.0 to $108.6

Twin Cities suburban counties have had relatively low spending levels per capita.

11 The difference was $23 to $5 for shelters, $93 to $32 for family foster homes, $19 to $6 for Rule 8 group homes, $49 to $14 for Rule 5 facilities, $75 to $20 for correctional facilities, and $6 to $ 3 for chemical dependency facilities.

Variation in County Trends

Among the 34 counties that spent more than $1 million for child placements in 1997, the change in inflation-adjusted spending per resident under age 18 between 1992 and 1997 ranged from a 19 percent decrease in St. Louis County to a 126 percent increase in Pine County. Even though statewide placement spending per child increased by 22 percent between 1992 and 1997, we found that:

Five counties that spent more than $1.0 million for placements in 1997 decreased their placement spending per child between 1992 and 1997.

These counties were Anoka, Dakota, Olmsted, and St. Louis and the joint administration of Lincoln, Lyon, and Murray counties. 12 Seven other counties decreased their spending, but we excluded these counties from extensive analysis because they had small placement budgets and a few expensive placements can significantly affect their spending. For example, eliminating a year-long placement at a facility with a $250 per diem would decrease a $500,000 placement budget by 18 percent.

To examine county trends in more detail, we divided the 34 counties that spent more than $1 million on placements and for which we had complete 1992 and 1997 data into three groups: 1) those with a spending increase of less than half the statewide increase (i. e., less than 11 percent), 2) those with a spending increase of more than twice the statewide increase (i. e., more than 44 percent), and 3) those with a spending increase between these two thresholds.

As shown in Table 3.5, the nine counties with small increases (or decreases) in placement spending had smaller increases (or larger decreases) in the number of placements made, length of stay, and cost per day of care than counties with large or medium spending increases. However, the counties with large overall spending increases surprisingly had a significant decline in their average length of stay (9 percent). As discussed above, a decline in the average length of stay is difficult to interpret. It is quite possible that this decline reflects a change in the mix of the placements rather than a policy of these counties to decrease the length of placements. As Table 3.6 shows, of the six types of facilities, the ones with relatively short average lengths of stay --shelter, corrections, and chemical dependency --had the biggest increases in spending for the 13 counties with large overall spending increases.

In addition, as displayed in Table 3.6, we found that counties with small overall spending increases (or decreases) had much smaller spending increases than counties with large overall spending increases in all categories of facilities. Declines in Rule 5 and Rule 8 spending were the main contributors to the cost containment efforts of counties with small overall increases as a group. However, the primary source of cost containment varied for the individual counties. For example, Anoka County reduced its spending at Rule 5 facilities by 32 percent, Rule 8 homes by 26 percent, and corrections facilities by 11 percent. On the other hand, Olmsted County increased its Rule 5 spending by 88 percent but decreased its Rule 8 spending by 97 percent and chemical dependency spending by 76 percent.

In the last five years, some counties significantly decreased their spending, while other counties more than doubled spending.

12 These three counties jointly administered social services but not corrections. Underlying Causes of County Variation

Table 3.5: Changes in Out-of-Home Placements for Counties with Large, Medium, and Small Spending Increases, 1992-97

Percentage Change in: Number

Inflation-Adjusted Spending per Resident Under

Age 18 Number of

Placements Made per Resident Under

Age 18 Average

Length of Stay a

Days of Care per Resident Under Age 18

InflationAdjusted Cost per Day Counties with large spending increases b 13 68% 43% -9% 31% 28%

Counties with medium spending increases c 12 22 7 1 8 13

Counties with small spending increases (or spending decreases) d

9 -12 -8 -11 -18 8 Statewide average e 72 20% 7% -3% 4% 15% NOTE: We excluded from the small, medium, and large categories 38 counties that spent less than $1 million on placements in 1997 and the 12 counties for which we did not have complete 1992 data.

a This estimate is based on the number of days of care divided by the number of placements made. b Counties that spent more than $1 million on placements in 1997 and had a 1992-97 spending increase o f more than 44 percent. c Counties that spent more than $1 million on placements in 1997 and had a 1992-97 spending increase o f between 11 and 44 percent. d Counties that spent more than $1 million on placements in 1997 and had a 1992-97 spending increase o f less than 11 percent. e This is the average for the 72 counties for which we had complete placement and days of care data fo r 1992 and 1997, including the 38 counties that spent less than $1 million on placements in 1997. We essentially had complete 1992 an d 1997 spending data for all counties, and the statewide increase was 22 percent, rather than the 20 percent shown for the 72 cou nties in this table.

SOURCE: Program Evaluation Division analysis of data from the Minnesota Department of Human Services ’ Social Service Expenditure and Grant Reconciliation reports, substitute care database, and Consolidated Chemical Dependency Tre atment Fund; Department of Corrections’ data on Red Wing and Sauk Centre placements and spending; and the Program Evaluation Division’s surveys of counties (June and July 1998).

The wide variation in placement spending appears to reflect county differences in both underlying social conditions and placement policies and practices. First, some counties have more families needing social services (including placement services) than other counties. These families frequently have very limited resources and a lot of problems to address --such as financial stress, domestic violence, drug and alcohol abuse, and homelessness. We found that:

There is a positive relationship between county placement spending and child poverty levels. 13

As Table 3.7 shows, the First and Tenth judicial districts --which include some relatively wealthy suburban counties around the Twin Cities --had the two lowest child poverty rates and the two lowest spending rates in 1997. On the other hand, the rural Ninth District of northwestern Minnesota and the urban Second (Ramsey County) and Fourth (Hennepin County) districts had the three highest poverty rates and three of the four highest spending rates. Figure 3.2 displays the counties in each of the ten judicial districts.

Counties do not have immediate or direct control over the demand for social services, but they have significant influence over placement decisions, either directly or through their recommendations to the court. Based on our site visits and phone interviews with officials from 14 counties, we found that:

Among the counties with the lowest placement spending are ones that (1) expressed a strong philosophy that out-of-home placements should be used only after all appropriate community-based options have been exhausted, and (2) have subjected placement recommendations to considerable scrutiny. 14

Table 3.6: Inflation-Adjusted Percentage Change in Out-of-Home Placement Spending per Resident Under Age 18, by Facility Type, 1992-97

Number Shelter Spending Foster Care

Spending Rule 8

Group Home Spending

Rule 5 Residential

Treatment Facility Spending

Correctional Facility Spending

Chemical Dependency

Facility Spending

Counties with large spending increases a 13 99% 34% 76% 60% 102% 185%

Counties with medium spending increases b 12 -8 15 62 52 29 -37

Counties with small spending increases (or spending decreases) c

9 3 6 -61 -44 15 -8 NOTE: We excluded from the small, medium, and large categories 38 counties that spent less than $1 million on placements in 1997 and the 12 counties for which we did not have complete 1992 data.

a Counties that spent more than $1 million on placements in 1997 and had a 1992-97 spending increase o f more than 44 percent. b Counties that spent more than $1 million on placements in 1997 and had a 1992-97 spending increase o f between 11 and 44 percent. c Counties that spent more than $1 million on placements in 1997 and had a 1992-97 spending increase o f less than 11 percent.

SOURCE: Program Evaluation Division analysis of data from the Minnesota Department of Human Services ’ Social Service Expenditure and Grant Reconciliation reports and Consolidated Chemical Dependency Treatment Fund; Department of Corrections’ data on Red Wing and Sauk Centre placements and spending; and the Program Evaluation Division’s surveys of counties ( June and July 1998).

A county’s placement spending is related to its economic conditions and placement philosophy.

13 The correlation coefficient between 1997 county placement spending per resident under age 18 and poverty levels was 0.52.

14 We contacted Crow Wing, Dakota, Hennepin, Jackson, Nicollet, Olmsted, Pine, Polk, Ramsey, Scott, St. Louis, Traverse, Washington, and Winona counties.

Table 3.7: Poverty Rate and Placement Spending by Judicial District

Judicial District Percentage of Children in Poverty in 1993

1997 Placement Spending per Resident

Under Age 18 First 8% $70 Second 21 211 Third 12 126 Fourth 19 322 Fifth 14 173 Sixth 19 210 Seventh 17 146 Eighth 17 139 Ninth 22 200 Tenth 9 93

Statewide average 15% $173 SOURCE: The poverty and population figures are from a Program Evaluation Division analysis of data from the United States Bureau of the Census. The spending numbers are from a Program Evaluation Division analysis of data from the Minnesota Department of Human Services’ Social Services Expenditure and Grant Reconciliation reports and Consolidated Chemical Dependency Treatment Fund; Department of Corrections’ data on Red Wing and Sauk Centre placements and spending; and the Program Evaluation Division’s surveys of counties (June and July 1998).

Figure 3.2: Minnesota Judicial Districts

Districts with low poverty rates also had low levels of placement spending.

Earlier in this chapter, Table 3.3 showed the highest and lowest spending counties. Examples of the impact of county philosophy and practice include the following:

Scott County social services staff told us that the county’s low poverty rate (5.7 percent and the second lowest of all counties) and high proportion of stable, intact families contributed to its low rate of spending, but they also said that the county’s strong belief in community-based, family-oriented programs helped. For example, the county hired mental health workers with advanced degrees to work with families in their homes. When the county decides to make mental health placements, it tries to restrict them to three or four months (compared with a six-month state average).

Dakota County social services staff told us that they have one of the lowest spending rates because they are very skeptical about the effectiveness of residential treatment. Therefore, they generally reserve these expensive placements only for children with a clear need for intensive supervision -for example, suicidal children. In addition, Dakota County social services has a very structured placement review process. A committee reviews placement recommendations not only prior to placement but again within 30 days after placement.

Traverse County had a 1993 child poverty rate of 19 percent but the second lowest placement spending rate of all the counties in 1997. In fact, we learned that the county did not have a Rule 8, Rule 5, or corrections placement in 1997 and has not had a corrections placement in the last five years. (In fact, Traverse County only made ten placements of any sort in 1997.) According to county staff, out-of-home placements are truly considered an option of last resort, and they have instead emphasized community-based programs, such as specialized services for American Indians and in-home counseling for juvenile delinquents.

Even though St. Louis County’s spending per resident under age 18 was higher than the statewide average in 1997, the county had one of the largest declines in spending between 1992 and 1997. According to social services officials, the staff psychologist has screened more children away from Rule 5 facilities and tried to scrutinize the lengths of these placements more closely. As a result, St. Louis County reduced its spending at Rule 5 facilities from $2.6 million in 1992 to $0.8 million in 1997 (both figures are in 1997 dollars), accounting for the greatest share of the county’s reduction in spending.

Jackson County’s spending per resident under age 18 was one of the highest in 1997, and its spending per resident under age 18 increased by 78 percent between 1992 and 1997 (after adjusting for inflation). Until recently, the county had no social service supervisors, and county officials told us that the lack of adequate training and support from the agency for social workers made it difficult for them to resist community pressure to place problem children. In addition, county staff told us that placement recommendations were not subject to formal, structured reviews internally.

Some counties use community based programs as an alternative to placements.

Counties also differ in the way they handle cases involving parent-child conflict. For example, we reviewed county records of a case in which a court placed a 17-year-old girl in a group home and foster care because she did not get along with her father’s live-in girlfriend. In other cases, parents ‘‘are simply not interested in raising their children,’’ according to one human services director, and they seek out-of-home placement as a way to make their lives easier.

Some county staff told us that they recommend placements in cases involving parent-child conflict when they believe that the situation will deteriorate further and require greater intervention if placement does not occur. But other counties told us that they have tried to entirely eliminate placements resulting from parent-child conflicts. For example, Dakota and Scott counties have 24-hour crisis teams that try to respond to domestic conflicts without making placements. Dakota County offers families up to ten free in-home counseling sessions, in the hope of avoiding more expensive long-term placements.

Without adequate outcome data, we cannot know for sure whether counties with low spending rates are making good management and policy decisions or are sacrificing the best interests of children. Based on a survey of county officials throughout Minnesota, we found that:

Fifty-five percent of human service directors and 40 percent of corrections supervisors said that there are additional steps that their counties could take to control out-of-home placement costs without sacrificing service quality. 15

We also asked the directors and supervisors to list up to three innovative practices that they have used in recent years to control out-of-home placement spending. Figure 3.3 lists many of these practices.

FUNDING

Counties pay for juvenile out-of-home placements with federal and state funds, county property taxes, parental fees, and other funding sources. We examined revenue sources for placements paid for by county social service agencies (which accounted for 76 percent of all placement spending in 1997). Unfortunately, there is no comprehensive information on revenue sources for all placements paid for by other agencies. Our analysis did not include placements paid by county agencies other than social services, most placements at the DOC’s Red Wing and Sauk Centre facilities, and chemical dependency placements. We found that:

Counties funded 59 percent of placement costs paid by social service agencies in 1997, while the federal government funded 20 percent and state government funded 12 percent. Other funding sources covered the remaining 9 percent. The state funds a relatively small portion of placement costs. Many county officials believe they can limit placement spending without sacrificing service quality.

15 Program Evaluation Division survey of county human service directors and corrections supervisors, August 1998. Two directors and three supervisors did not answer this question or responded ‘‘don’t know.’’

When comparing Minnesota with other states, we found:

Minnesota’s social services system is funded with county revenues more than most states’ systems.

Figure 3.3: Selected County Strategies for Reducing Placement Costs

Having pre-placement panels or committees review and screen placement recommendations.

Increasing interagency (social services, corrections, schools, law enforcement, etc.) cooperation through collaboratives.

Reducing caseloads for social workers and probation officers.

Using community-based programs, such as intensive supervision, mental health counseling, and mentoring.

Accessing family-based programs, such as in-home family counseling and parental education.

Creating school-based programs, such as having social workers and probation officers based in the schools.

Developing truancy programs, including alternative schools for truants.

Making culturally specific programs available.

Using electronic home monitoring.

Finalizing adoptions as quickly as possible.

Making services available to children and families after the children return home.

Using treatment foster care as an alternative to residential treatment facilities.

Placing children for shorter periods of time.

Aggressively pursuing parental fees and federal foster care funding.

Making available to each social worker or probation officer monthly statistics about the placements they made --the number of placements, days of care, and cost.

SOURCE: Program Evaluation Division survey of county human services directors and corrections supervisors, August 1998.

16 American Public Welfare Association, A Statistical Summary of the VCIS Social Services Block Grant (SSBG) Data for Fiscal Year 1990 (Washington, D. C., 1994), 27.

According to data gathered in 31 states by the American Public Welfare Association, local funds accounted for only 13 percent of social service expenditures in 1990, state funds accounted for 41 percent, and federal funds accounted for 46 percent. 16 Furthermore, a recent survey of 38 states by the Child Welfare League of America indicated that Minnesota was one of only seven states in which local revenues accounted for more than 20 percent of child welfare spending. 17 In most other states, state agencies (not the counties) directly provide child welfare services.

The primary source of county funding of out-of-home placements is property tax revenue, but county revenues also include general purpose aid provided by the state. Most of the federal funding comes from the foster care program under Title IV-E of the Social Security Act. This program helps pay for maintenance costs (food, shelter, clothing, daily supervision, school supplies, insurance, and other incidentals) and county administrative costs for children who (1) meet the eligibility requirements of the former Aid to Families with Dependent Children program, and (2) are placed in public institutions with up to 25 children, private facilities, or family foster homes. In Minnesota, the federal government pays for 52 percent of the maintenance payments and 50 percent of administrative costs, and the counties pay for the rest. 18 As Figure 3.4 shows, federal Title IV-E paid for $24 million in out-of-home placements costs in 1997. Between 1992 and 1997, federal Title IV-E funding for out-of-home placements per resident under age 18 increased by 41 percent after adjusting for inflation.

Figure 3.4: 1997 Out-of-Home Placement Funding by Source

SOURCE: Department of Human Services’ Social Services Expenditure and Grant Reconciliation information system.

County Revenue $102 Million Other

$9 Million CSSA Funds $18 Million

Title XX Funds $7 Million

Title IV-E Funds $24 Million

Service Charges $11 Million

County Social Services Budget Total = $171 Million

Counties pay for most out-of-home placement costs.

17 Michael Petit and Patrick Curtis, Child Abuse and Neglect: A Look At the States: The 1997 CWLA Stat Book (Washington, D. C.: Child Welfare League, 1997), 159. The survey reported spending (excluding Medicaid) for FY 1996.

18 42 U. S. Code, sec. 670 et seq. (1998).

19 42 U. S. Code, sec. 1397 et seq. (1998).

In addition, the social services block grant under Title XX of the Social Security Act makes federal funds available to states based on their population and has no requirement for states to provide matching funds. 19 The state of Minnesota then allocates the funds to the counties based on their share of the state’s welfare caseload and population. 20 Counties have wide discretion about how to spend these funds. In 1997, this grant paid for $7 million of out-of-home placement costs in Minnesota. Between 1992 and 1997, Title XX funding for all social services per resident under age 18 declined by 36 percent after adjusting for inflation.

The primary source of state funding is the Community Social Services Act (CSSA) block grant. The state allocates one-third of the funds to counties based on their share of the state’s welfare caseload, another third based on county population, and the final third based on the number of residents age 65 or older. 21 In 1997, CSSA grants paid for $18 million of out-of-home placement costs and accounted for over 90 percent of state funding in cases in which county social services agencies paid for the placement. Between 1992 and 1997, CSSA funding for all social services per resident under age 18 increased by 4 percent after adjusting for inflation --well below the 22 percent increase in overall placement spending.

Like the variation in spending levels, there was wide variation among counties in funding sources. As Table 3.8 shows:

In 1997, county revenues accounted for as little as 33 percent of total placement funding (Clearwater) and as much as 79 percent (McLeod).

Table 3.8: 1997 Out-of-Home Placement Spending Funded by County Dollars, Selected Counties

Percentage of 1997 Placement Spending Funded by County Dollars

Highest five counties

McLeod 79% Wilkin 76 Dakota 75 Waseca 73 Sherburne 70

Lowest five counties a Blue Earth 42% Cottonwood 42 Traverse 42 Freeborn 35 Clearwater 33

a The lowest county was actually Red Lake County at -41 percent. Because the data is reported on a cash basis and some revenues are distributed to counties in lumps (rather than continuously), the Social Services Expenditure and Grant Reconciliation reports can produce county revenue numbers that appear to not make sense, especially for small counties.

SOURCE: Department of Human Services’ Social Services Expenditure and Grant Reconciliation (SEAGR) reports.

The county share of placement funding varies considerably.

20 Minn. Stat. §256E. 06.

21 Minn. Stat. §256E. 01 - §256E. 15.

This wide variation reflects several factors. First, counties vary in their ability to access federal Title IV-E and XX funds and state CSSA funds. Because the distribution of these funds is partially dependent on the level of poverty in each county, counties with more people in poverty will receive more funds from the federal and state governments. Some people also told us that counties might not always pursue Title IV-E funding when a placement is eligible to receive it. Second, if a county places a lot of children who are not in poverty, it will finance a higher proportion of the spending with local revenues because these placements are not eligible for Title IV-E funding. Third, counties vary in the types of placements that they make. If a county places children in public institutions with more than 25 children, the placement is not eligible for Title IV-E funds. In addition, Title IV-E does not cover the treatment component of a placement. Finally, if a county has a high overall level of placement spending, it may finance a higher proportion of placement costs because Title XX and CSSA are fixed block grants and not open-ended entitlements.

FAMILY PRESERVATION SPENDING

Counties have developed family preservation programs (1) to help them prevent out-of-home placements and (2) as alternatives to placements. Therefore, to provide context for our analysis of placement spending, we examined family preservation spending. 22 According to the data reported by counties to DHS:

Minnesota spent $158 million on family preservation services in 1997.

County family preservation spending varied widely, ranging from $37 per resident under age 18 (Houston) to $236 (Hennepin). We found that counties with high family preservation spending generally had high levels of out-of-home placement spending, too, as well as high levels of poverty. 23 Thus, county differences in family preservation spending often appeared to reflect differences in underlying service needs. However, the level of family preservation spending sometimes also reflects a county’s philosophy and practices about providing family preservation services. For example, Olmsted County strongly emphasizes family preservation over out-of-home placements. As a result, its family preservation spending per resident under age 18 was 15 percent above the statewide average in 1997 while its placement spending was 39 percent below average.

Like placement spending, family preservation spending varies significantly by county.

22 We used data from DHS’s Social Services Expenditure and Grant Reconciliation reports and a broad definition of family preservation programs which includes, among others, community education and prevention, child protection assessment, group counseling, family-based counseling services, general case management, and mental health day treatment. These are the services that counties may support with revenue from the state’s Family Preservation Fund. This list of services goes beyond the core family preservation programs of family-based crisis, counseling, and life management services to include 18 other services. (See Department of Human Services Bulletin #98-32-6, ‘‘DHS Issues Information on the Closing of EA-IFPS and on the New Family Preservation Fund Remittance Advice,’’ April 1, 1998.) However, we excluded from this list the three services for people with developmental disabilities. Also, we did not include services funded by Medicaid.

23 The correlation coefficient between 1997 county family preservation spending per resident under age 18 and poverty level was 0.40, and the coefficient between 1997 county family preservation and placement spending per resident under age 18 was 0.53.

On the funding side, we found that:

In 1997, counties provided 43 percent of the funding for family preservation, while the federal government provided 39 percent and the state provided 16 percent.

Thus, counties contributed a significant share of family preservation revenues, but not as much as they contribute to out-of-home placement revenues. 24

SUMMARY

Overall, out-of-home placements cost about $225 million in 1997, and inflation-adjusted placement spending increased by 22 percent per resident under age 18 between 1992 and 1997. However, placement spending and trends vary widely among counties. Out-of-home placement spending per resident ranged from $25 (Red Lake County) to $322 (Hennepin County). Some counties reduced their placement spending between 1992 and 1997, while costs in some other counties more than doubled. Some of the variation is related to the service needs of the counties, and some of it is explained by the placement philosophies and practices of the counties. Correctional facilities experienced the highest growth in spending between 1992 and 1997, while foster care spending grew much more slowly. We also found that county revenues pay for most out-of-home placement costs in Minnesota, which appears to differ from practices in most other states.

24 The portion of family preservation funding provided by the counties ranged from 0 percent (Cass and Clearwater) to 60 percent (Stevens).

Characteristics of Children in Placement CHAPTER 4

M innesota residential facilities serve a variety of types of children. In fact, it is difficult to identify a ‘‘typical’’ child in placement because they

vary considerably in age, reasons for placement, and the nature of their placement settings. We asked:

To what extent are children placed away from home because of their own behaviors, their parents’ behaviors, or other reasons?

What are the demographic characteristics of children placed away from home in Minnesota? How far from home are children placed?

How common is it for juveniles to have multiple placements away from home?

Does the state’s primary information system on child placement have accurate, reliable data?

Overall, we found that the reason reported by counties for a majority of child placement days is a parent-related reason, not a child-related one. We also found that African American and American Indian children are more likely to be placed out of home than other racial/ ethnic groups. In addition, we found that most children placed in shelters and foster care live within their home counties, but children placed in Rule 5 mental health treatment facilities, group homes, and correctional facilities are more likely to be placed farther away.

For some analyses in this chapter, we profiled Minnesota’s placement population by integrating information we obtained from counties, the Department of Human Services (DHS), and the Department of Corrections (DOC). In other cases, however, we found that this would be difficult or impossible with existing information systems, so we based these analyses solely on placements paid for by county social services agencies. 1 Placements funded by social services agencies accounted for 91 percent of the days that children spent in placement in 1997. Residential facilities serve juveniles with various needs.

1 These analyses did not include placements for juveniles in chemical dependency treatment facilities that were paid for by the state’s considated treatment fund. The analyses also did not include most juveniles served at the Department of Corrections’ Red Wing and Sauk Centre facilities and placements paid by county agencies other than social services. Statewide, social services budgets only paid for 33 percent of spending at correctional facilities.

As we discuss later in the chapter, we found various problems in the data we obtained from DHS on placements paid for by social services agencies. The DHS data are county-reported, but we found many cases where counties had more complete, accurate information than what had been reported to DHS. As a result, we obtained comparable placement data directly from eight counties (which accounted for 62 percent of the state’s 1997 total days that children spent in placement), and we often used the county data to adjust or correct the DHS data. 2

REASONS FOR PLACEMENT

Children are placed away from home due to a variety of underlying problems. Sometimes the immediate reason for placement is a child’s behavior--such as when a child breaks the law or abuses alcohol or drugs. In other cases, placements may occur due to parental behavior--such as child abuse or neglect.

For each placement paid for by social services budgets, counties report to DHS the reason for placement. The vast majority of placements not paid for by social services budgets are delinquency and chemical dependency cases, and we assumed that the reason for placement in each of these cases was the child’s behavior. By combining DHS data with information we collected from counties and DOC, we found that:

Children spent more time in out-of-home placement in 1997 due to their parents’ actions than their own conduct.

Table 4.1 shows the reasons for child placements--by 1997 days of care and by the number of 1997 child placements. Because many of the 1997 placements were very short-term placements in detention centers, we think that the days of care column in Table 4.1 presents a better measure of the reasons for placement. Among children in placement in 1997, 46 percent of days of care resulted from parents’ conduct, and other parent-related reasons accounted for another 12 percent. In contrast, children’s conduct was the cause of 30 percent of the 1997 days of care, and other child-related reasons accounted for another 6 percent. Family interaction problems accounted for 5 percent of the 1997 days of out-of-home care.

We also looked at the reasons that children were placed in various categories of facilities. Seventy-one percent of children in foster care during 1997 were placed there because of a parent’s conduct or other parent-related reasons. In contrast, a large majority of children in Rule 8 group homes, Rule 5 mental health treatment, and correctional facilities in 1997 were placed due to child behaviors or other child-related reasons. 3

Out-of-home placements often result from the actions of parents.

2 Hennepin, Ramsey, Anoka, Dakota, Washington, St. Louis, Crow Wing, and Blue Earth counties.

3 Child behavior and other child-related reasons accounted for 71 percent of group home residents, 75 percent of Rule 5 residents, and 92 percent of correctional facility residents whose placements were funded by social services budgets.

Some juveniles who have been placed away from home because of their own conduct have previously been placed for their parents’ conduct. Among children placed away from home for delinquency during 1997, we found that about 6 percent also had at least one placement in 1995-97 for which the reason was a parent’s conduct. Undoubtedly, the percentage of delinquent children with prior placements for child protection reasons would be higher than 6 percent if their placement histories could be tracked for longer than three years. 4

Table 4.1: Reasons for Out-of Home Placements, 1997

Percentage of: Average 1997 Placements Days of Care Days of Care Reason for Placement (N= 35,705) (N= 3,327,066) Per Placement 1

Parent conduct Neglect/ abandonment 8.9% 18.7% 195 Physical abuse 4.2 8.2 180 Substance abuse 2.9 8.0 257 Termination of parental rights 1.0 4.8 422 2 Sexual abuse 1.4 2.8 187 Parent cannot cope 1.3 2.5 172 Parent incarceration 1.4 1.7 107 SUBTOTAL 21.2 46.2 203

Other parent-related reason Illness/ disability 2.1 4.0 181 Temporary absence 1.1 1.4 120 Other 4.0 6.7 157 SUBTOTAL 7.2 12.2 158

Child conduct Delinquency, status offense 51.8 21.4 38 Substance abuse 3.9 1.8 42 Other behavior problem 7.3 7.1 91 SUBTOTAL 63.1 30.3 45

Other child-related reason Child disability 1.3 4.0 286 Other 1.9 1.9 96 SUBTOTAL 3.2 5.9 175

Family interaction problems 5.4 5.4 93 TOTAL 100.0% 100.0% 93 1Total days of care during 1997 for children in this category (regardless of the year the placement occurred) divided by the number of placements in this category in 1997.

2 For cases involving termination of parental rights, the 1997 days of care per 1997 placement exceeded 365 days because many of these cases’ days of care were for placements made prior to 1997.

SOURCE: Program Evaluation Division analysis of DHS substitute care database and Consolidated Treatment Fund data; DOC data on Red Wing and Sauk Centre placements; and Program Evaluation Division’s June-July 1998 surveys of counties. Excludes cases where reason for placement was unknown and cases where children were discharged on the same day they entered placement.

CHARACTERISTICS OF CHILDREN IN PLACEMENT

4 According to information collected by the Minnesota Council of Child Caring Agencies, 32 percent of children discharged from member agencies’ correctional programs during 1997 had previously been victims of documented abuse or neglect, and many others were suspected victims.

DEMOGRAPHIC CHARACTERISTICS

The 1998 Legislature requested that we summarize the demographics of children in placement, including information on race, gender, and age. 5 We developed profiles of children in placement and examined differences in placement patterns, primarily using data on 1997 placements paid for by social services agencies.

Race/ Ethnicity

Our review of 1997 placement patterns indicated that:

African American and American Indian children had disproportionately high rates of out-of-home placement, compared with children from other racial/ ethnic groups.

As Table 4.2 shows, African American children comprised 4 percent of Minnesota’s population under age 18 in 1997, but they comprised 22 percent of the children in placement. Likewise, only 2 percent of Minnesota children are American Indian, but American Indians accounted for 12 percent of children in placement in 1997.

4.2: Percentage of Children in Placement and Average Length of Placement, By Racial and Ethnic Groups, 1997

Racial/ Ethnic Category

Percentage of State Population Under Age 18 1

Percentage of Children in Out-of-Home

Care 2 Average 1997

Days of Care Per Child in Placement 3

White, non-Hispanic 87.8% 60.8% 162 African American 4.1 22.3 194 American Indian 1.9 11.6 212 Hispanic 2.2 3.4 168 Asian American 4.0 1.9 152

TOTAL 100.0% 100.0% 175 NOTE: This analysis was based solely on placements funded by local social services budgets. It doe s not include most detention and chemical dependency placements, plus some longer correctional placements. It excludes cases where the racial/ ethnic category was unknown.

1 In 1997, the estimated population under age 18 was 1,250,685. 2 In 1997, the number of children in out-of-home care totalled 18,142. 3 Total days of care during 1997 (regardless of the year the placement occurred) for each child who wa s in placement at some time during 1997.

SOURCE: Program Evaluation Division analysis of Department of Human Services and county placement data.

Some minority groups have had disproporTable tionately high rates of out-of-home placement. 60 JUVENILE OUT-OF-HOME PLACEMENT

5 Minn. Laws (1998), ch. 367, art. 10, sec. 16.

Table 4.2 also shows that the average African American and American Indian child spent more time in placement during 1997 than the average child in other racial/ ethnic categories. For example, American Indian children who were in placement during 1997 spent an average of 212 days in placement that year, compared with 162 days for white, non-Hispanic children.

We also found disparities in the statewide percentage of each racial/ ethnic group’s children who were in placement at some time during 1997. Specifically, 8.1 percent of Minnesota’s African American children and 8.8 percent of the state’s American Indian children were in out-of-home placement at some time during 1997, compared with 1.0 percent of Minnesota’s white, non-Hispanic children, 2.2 percent of Hispanic children, and 0.7 percent of Asian American children. Thus, African American and American Indian children were eight times as likely to have been in out-of-home placement in 1997 as white, non-Hispanic children.

We observed some regional differences in patterns of placement among racial/ ethnic groups, as shown in Table 4.3. In all racial and ethnic groups, suburban Twin Cities counties, as a group, had below average days of out-of-home care per child in the population. For instance, African American children from suburban counties had about 6 days of care per African American child in the population, compared with a statewide total of 15 days of care per African American child. Hennepin County’s total days of care per child in the population exceeded the state average by 60 percent (4.0 vs. 2.5), but its days per child were below the state average for children who were (1) white, non-Hispanic, (2) Hispanic, or (3) Asian American. Ramsey County’s total days of care per child in the population exceeded the state average by 44 percent (3.6 vs. 2.5), but its days per child were below the state average for children who were (1) American Indian, or (2) Hispanic. The total days of care for counties in Greater Minnesota (as a group) was near the state average (2.3 vs. 2.5), but Greater Minnesota had days of care per child that were above the state average for children who were (1) white, non-Hispanic, (2) Hispanic, or (3) Asian American.

Table 4.3: 1997 Days of Care Per Child in Population, By Racial/ Ethnic Category and Region

Days of Care Per Child in the Population Population Twin Cities Twin Cities Under Racial/ Hennepin Ramsey Suburban Metro Area-Greater State Age 18 Ethnic Category County County Counties Total Minnesota Total (in 000s)

White, non-Hispanic 1.3 2.1 1.0 1.3 1.8 1.6 1,098 African American 16.4 15.9 6.2 15.3 12.5 15.2 51 American Indian 28.2 11.5 5.1 20.9 15.5 17.9 24 Hispanic 1.2 3.2 2.0 2.2 6.2 3.7 28 Asian American 0.7 1.7 0.7 1.1 1.4 1.1 50

TOTAL 4.0 3.6 1.1 2.6 2.3 2.5 1,251 NOTE: This analysis was based on placements funded by social services budgets, plus placements at t he Hennepin County Home School and Ramsey County Boys Totem Town. Excludes days of care for which racial/ ethnic category wa s unknown.

SOURCE: Program Evaluation Division analysis of data from DHS substitute care database, selected c ounties, and the U. S. Census Bureau.

Twin Cities suburban counties have below average placement rates in all racial/ ethnic categories.

In 1997, children from racial/ ethnic minority groups accounted for at least half of children in placement in 6 of Minnesota’s 87 counties (Mahnomen, Beltrami, Hennepin, Clearwater, Sibley, and Ramsey). 6 In some counties, these high rates reflected large numbers of minority children in the general population and high rates of placement within certain minority groups. For example, Mahnomen County had (1) the state’s highest percentage (36 percent) of minority children in its county population under age 18, mainly American Indians, and (2) a high percentage of its American Indian child population in placement during 1997 (13 percent, compared with 9 percent statewide).

Counties’ stated reasons for placement differed considerably among the racial and ethnic groups, as shown in Table 4.4. For example, compared with white, non-Hispanic children, we found that American Indian children had more than 16 times as many days of placement per capita in 1997 that were due to parent conduct or other parent-related reasons, and African American children had 12 times as many. Thus, it is possible that differences among racial/ ethnic groups in their respective levels of family dysfunction explained some of their differences in rates of child placement. Also, compared with white, non-Hispanic children, American Indian children had six times as many days of placement per capita for child conduct or child-related reasons; African American children had five times as many.

Table 4.4: Reasons for Placement, by Racial and Ethnic Groups

1997 Days of Care Per Child in Population That Were Attributed To:

Racial/ Ethnic Category

Population Under 18

(in 000s) Parent Conduct or Parent-Related

Reasons Child Conduct or Child-Related

Reasons Family

Interaction Problems Total

White, non-Hispanic 1,098 0.8 0.7 0.1 1.6 African American 51 11.9 3.3 0.3 15.5 American Indian 24 13.6 3.9 0.8 18.3 Hispanic 28 2.2 1.3 0.3 3.8 Asian-American 50 0.5 0.7 0.0 1.2

TOTAL 1,251 1.5 0.8 0.1 2.5 NOTE: This analysis was based on placements funded by local social services budgets and the Consoli dated Chemical Dependency Treatment Fund, plus placements at Hennepin and Ramsey County correctional facilities. We excluded cases where the racial/ ethnic category and reasons for placement were unknown.

SOURCE: Program Evaluation Division analysis of Department of Human Services and county placement d ata.

American Indian and African American children are far more likely to be placed for parent-related reasons.

6 The percentage of placed children who were from racial/ ethnic minority groups in these counties was 88, 73, 72, 62, 60, and 60 percent, respectively.

Finally, we looked at juvenile placements in chemical dependency treatment programs publicly funded by Minnesota’s Consolidated Chemical Dependency Treatment Fund. We found that about 1.2 percent of Minnesota’s American Indian population under age 18 entered a residential chemical dependency program during 1997, compared with 0.2 percent of Hispanic juveniles, 0.1 percent of white, non-Hispanic juveniles, 0.1 percent of African American juveniles, and less than 0.1 percent of Asian American juveniles. 7

Age

Children in out-of-home placement represent a wide range of ages. For example, children born to chemically dependent mothers are sometimes placed in foster care at birth. Occasionally, children continue to receive foster care services or remain in correctional placements after their eighteenth birthdays.

Statewide, the median age of children in placement in 1997 was 13.7 years. 8 However, this varied considerably by type of facility. Children in foster care had a lower median age (10.7 years) than children in other types of facilities. Children in correctional facilities (16.0 years) and group homes (15.5 years) had the highest median ages. Table 4.5 shows the percentage of children in various age groups by facility type. 9

The median age of African-American children in placement (9.0 years) was considerably lower than the median ages of children from other racial/ ethnic categories. This apparently reflected the greater likelihood of an African-American child being placed for child protection reasons.

Table 4.5: Ages of Children in Placement During 1997, By Facility Type

Type of Facility Number of 1997 Placements Funded by Social Services Median Age (in Years) of Children in Placement Percentage of Placed Children Who Were Ages:

0 to 4 5 to 9 10 to 14 15 and up Total Shelters 7,765 13.1 15.8% 18.5% 36.2% 29.5% 100% Family foster homes 13,728 10.7 22.4 24.4 28.7 24.5 100 Rule 8 group homes 1,912 15.5 0.2 2.0 37.0 60.8 100 Rule 5 facilities 2,105 14.9 0.9 3.8 46.5 48.8 100 Correctional facilities 3,468 16.0 0.0 0.0 27.0 72.9 100

TOTAL 28,978 13.7 14.9% 16.9% 32.4% 35.8% 100% NOTE: This analysis was based solely on placements funded by local social services budgets. We exa mined all such cases in which children were in placement at some time during 1997, including some in which the placements started before 1997. For placements starting in 1997, we examined each child’s age at the time of placement. For placements starting be fore 1997, we examined each child’s age as of January 1, 1997. We excluded cases where children were discharged on the same day they en tered placement.

SOURCE: Program Evaluation Division analysis of Department of Human Services and county placement d ata.

Children in foster homes tend to be younger than children in other facilities.

CHARACTERISTICS OF CHILDREN IN PLACEMENT

7 These are not unduplicated counts; a juvenile who entered a program twice in 1997 would be counted twice.

8 For children who were placed during 1997, we computed their ages on the date of placement. For children whose 1997 placement started before 1997, we computed their ages as of January 1, 1997.

9 Table 4.5 only shows placements paid for by social services agencies. We obtained information on children who entered placements in 1997 paid for by Minnesota’s Consolidated Chemical Dependency Treatment Fund, and it showed that 83 percent of the children placed were ages 15 or older.

Gender

Statewide, 1.5 percent of Minnesota children were in out-of-home placement at some time during 1997. 10 We found that 1.6 percent of Minnesota boys were in placement during 1997, and 1.3 percent of girls were in placement. The average number of days spent in placement during 1997 was nearly identical for boys and girls (172 days for boys and 176 days for girls). However, we found that:

Boys were much more likely than girls to be placed in correctional, Rule 5, and chemical dependency facilities.

Boys accounted for 80 percent of the 1997 days of care in correctional facilities and 74 percent in Rule 5 facilities. We did not have information on the days of care children spent in chemical dependency facilities, but we did find that boys represented 65 percent of all 1997 placements paid for by Minnesota’s consolidated chemical dependency treatment fund. In contrast, boys accounted for 56 percent of days of care in group homes, 51 percent of foster home days of care, and 52 percent of shelter facility days of care.

Using information on placements paid for by social services agencies, we also examined the relationship between child gender and the reasons for placement. We found that boys accounted for 68 percent of the 1997 days of care that were attributed to child-related reasons and 50 percent of days of care attributed to parent-related reasons.

DISTANCE PLACED FROM HOME

The Legislature requested that our study include ‘‘a summary of the geographic distance between the juvenile’s home and the location of the out-of-home placement.’’ 11 Some people have expressed concerns that placement far from home discourages family participation in facility programs and makes it more difficult for a county or facility to plan for supportive services following a juvenile’s return to the community.

For children in placement during 1997, we determined the distance between the county seats of the child’s home county and the county in which the residential facility was located. In addition, we examined whether each child was placed at facilities in his or her own county or in bordering counties. As shown in Table 4.6, Statewide, 62 percent of days that children spent in placement during 1997 were at facilities in the children’s home counties. Eight percent of days in placement were at Minnesota facilities more than 100 miles from home, and 5 percent of days in placement were in facilities in other states. Children spend most days of out-of-home care in their home counties.

10 This is based solely on placements funded by county social services budgets, which in 1997 represented 91 percent of all days of care and three-fourths of placement spending.

11 Minn. Laws (1998), ch. 367, art. 10, sec. 16.

Juveniles placed in shelters or foster homes usually remained in their home counties. Statewide, 80 percent of time spent in shelters occurred in the juveniles’ home counties, and 72 percent of time spent in foster homes occurred in juveniles’ home counties. In both of these categories of facilities, it was relatively rare for juveniles to be placed more than 100 miles from home or out-of-state.

Placements in Rule 5 mental health treatment facilities, correctional facilities, and group homes were the most likely to be at locations far from children’s homes. Only 30 percent of the time spent in Rule 5 facilities was in juveniles’ home or neighboring counties, and 39 percent of the time was in another state or at a Minnesota facility more than 100 miles from the child’s home.

As we discussed in Chapter 2, placements with relatives accounted for most of the out-of-state foster care placements, and delinquent juveniles accounted for most of the out-of-state placements in the Rule 5 and correctional facility categories.

NUMBER AND LENGTH OF PLACEMENTS

To better understand the frequency and length of child placements, we examined data on all children who were in placements funded by social services agencies at some time during 1995-97. For the three-year time period, we found that:

Of children who were in at least one placement that lasted more than three days, 45 percent had multiple placements of this length.

Table 4.6: Distance of Placements From Home, By Facility Type, 1997

Percentage of 1997 Days of Care Spent in Placements: Facility Type

Total Days in Care in 1997 Within Same County In Border County In State and Within 100 Miles, But Not in Same or Border County In State, But More than 100 Miles Away In Another State TOTAL Shelters 183,143 79.8% 13.3% 4.7% 2.2% 0.0% 100% Family foster homes 2,086,280 71.6 13.5 6.5 4.9 3.4 100 Rule 8 group homes 166,279 25.8 26.4 23.3 20.7 3.8 100 Rule 5 facilities 228,579 12.9 17.0 31.1 25.2 13.8 100 Correctional facilities 436,031 45.2 12.2 19.3 13.7 9.7 100

CHARACTERISTICS OF CHILDREN IN PLACEMENT

Of children who were in at least one placement that lasted more than 30 days, 39 percent had multiple placements of this length.

Of children who were in at least one placement that lasted more than 90 days, 28 percent had multiple placements of this length.

Although most children did not have multiple placements during this three-year period, some had many placements. For example, one child had 21 placements that each exceeded three days. Another child had 11 placements that each exceeded 30 days. 12

We examined the total amount of time children spent in placement during 1995-97, as shown in Table 4.7. Some children had placements that lasted a long time. Of all children in placement on January 1, 1995, 23 percent remained in placement continuously for at least the next three years. Children who remained in placement for this full three-year period comprised 6 percent of all children who were in placement at some time during 1995-97. Most of the children who were in placement for large portions of this three-year period lived in foster homes and had been placed for parent-related reasons.

Table 4.7: Total Time That Individual Children Spent in Placement, 1995-97

Total Time in Placement Percentage of Children Who Were in Placement at Some Time During 1995-97 (N= 33,852) Percentage of Children in Placement on January 1, 1995

(N= 8,720) 1 day 5.4% 0.2% 2 to 7 days 15.3 0.9 8 to 30 days 11.5 3.5 At least 1 month but less than 3 months 11.8 7.2 At least 3 months but less than 1 year 27.6 28.0 At least 1 year but less than 2 years 14.8 19.6 At least 2 years but less than 3 years 7.8 17.9 3 years 5.9 22.8

100.0% 100.0% NOTE: Analysis excluded placements in which children were discharged on the same day they entered placement.

SOURCE: Program Evaluation Division analysis of Department of Human Services and county placement data.

Some juveniles live in out-of-home care continuously for years, while others are in placement only for brief periods.

12 We obtained information from eight individual counties that enabled us to examine the specific facilities in which children were placed (DHS placement data do not identify individual facilities). In these counties, the largest number of separate facilities in which a child was placed during 1995 97 was 16. One child was placed in 14 separate facilities that each cared for the child more than 3 days. Another child was placed in 9 separate facilities that each cared for the child more than 30 days.

Table 4.7 also shows that many children were in out-of-home placement for relatively short periods during 1995-97. Of all children who were in placement at some time during this period, 32 percent were in placement for a total of 30 days or less.

We found variation in the average number and length of placements among various racial/ ethnic groups. Among children in placement during 1995-97, Asian American children were less likely to have multiple placements than children in other racial/ ethnic groups, and American Indian children were the most likely to have multiple placements. 13 Over this three-year period, the average American Indian juvenile in placement spent a total of 362 days in residential settings, which was higher than the averages for children in other racial/ ethnic categories. 14 Of American Indian children in placement during 1995-97, 11 percent remained in placement for the full three years. 15

In general, young children tended to spend more time in placement during 1995-97 than older children. Among children who were in placement at some time during 1995-97, children under age ten at the beginning of 1995 spent an average of 337 days in placement during the three-year period, compared with 256 days for children ages 10 to 15. 16 We also found that children who were ages 10 to 13 at the beginning of 1995 were somewhat more likely to have multiple placements during 1995-97 than children in other age groups. 17

ACCURACY OF STATE PLACEMENT INFORMATION

Recently, Minnesota policy makers have shown increasing interest in measuring trends in placements, lengths of stay, and outcomes for different types of placements. The Department of Human Services’ information system for out-of-home placements is important because it has the most comprehensive statewide data available to examine placement trends and child characteristics. This information system is based on data reported to DHS by county social services agencies.

DHS collects placement information from counties.

13 This was true when we examined placements of (1) all lengths, (2) more than 3 days, (3) more than 30 days, and (4) more than 90 days.

14 Other averages included 334 days for African American juveniles, 245 days for Asian American juveniles, 250 days for Hispanic juveniles, and 259 days for white, non-Hispanic juveniles.

15 The comparable percentages for other racial/ ethnic groups were 9.4 percent for African Americans, 2.8 percent for Asian Americans, 4.5 percent for Hispanic juveniles, and 4.4 percent for white, non-Hispanic juveniles.

16 Related to our age findings, we found that children first placed in 1995-97 for parent-related reasons tended to spend more time in placement (an average of 338 days) over this period than children first placed for child-related reasons (an average of 216 days).

17 Among children with at least one placement, children ages 10 to 13 were the most likely to have two or more placements during this three-year period. This was true when we examined placements of (1) all lengths, (2) more than 3 days, (3) more than 30 days, and (4) more than 90 days.

To examine the accuracy of the state’s placement information, we first examined the DHS data from 1992 through 1997. 18 Second, we reviewed a sample of county case files during site visits and interviewed county officials. Finally, we compared DHS data with data that we independently collected from eight counties that have placed large numbers of children away from home, and we followed up with county staff to resolve significant differences. 19 We found that:

The state’s main database of records on individual out-of-home placements has shortcomings that, if uncorrected, could hinder certain analyses and county comparisons.

As described below, the DHS placement data has many problems--missing placements, inaccurate discharge dates, duplicate or overlapping placements, and single placements inappropriately reported as multiple placements. The magnitude of these errors varies greatly among the counties, and we were not able to assess the exact magnitude of the problems in counties for which we did not independently collect and review county placement data. We think that the DHS database provides fairly accurate statewide estimates for many measures, including days of care, reasons for placements, demographic characteristics of children in placement, and type of facilities in which children are placed. For example, the cumulative effect of errors we found would have changed the total statewide days of care by about three percent for both 1996 and 1997--partly because errors of underreporting helped offset errors of overreporting.

However, we think that the data problems could hinder analyses of county by county trends in a variety of important measures, including days of care, number of placements, and placement length. For example, we estimate that DHS’ analysis of county-reported data overstated days of care during 1996 by more than 20 percent for four of the eight counties we examined (Crow Wing, Dakota, St. Louis, and Washington), by 9 to 15 percent for two counties, and by less than 2 percent for two counties. Also, the data used by DHS were missing at least 10 percent of total 1996 placements in two of the eight counties we examined (Crow Wing and Ramsey).

We found similar problems in the 1997 data, although the size of the errors was not as large as in 1996. Our analysis of 1997 data reported by counties to DHS indicated that days of care would be overstated by 20 percent for Crow Wing County, by 9 to 12 percent for four counties, and by less than 2 percent for three counties. The DHS data on total placements for 1997 differed from our estimates by more than 10 percent for one county (Crow Wing). These reductions in error rates reflect improvements by the department in its review of the 1997 data.

The DHS database overstates days of care for a variety of reasons, including inaccurate placement discharge dates and duplicate or overlapping placements.

There have been significant errors in some placement data, but DHS has been trying to improve accuracy. 68 JUVENILE OUT-OF-HOME PLACEMENT

18 Our discussion of DHS data accuracy is based on our review of data edited by DHS following its submission by counties. For 1997 data, we reviewed data that were edited by DHS as of December 1998.

19 The counties were Hennepin, Ramsey, Dakota, Washington, St. Louis, Anoka, Crow Wing, and Blue Earth. They accounted for 62 percent of Minnesota’s days of placement in 1997.

The most common problem with placement discharge dates is that some counties fail to enter these dates in a timely way. As a result, counties have reported to DHS that placements were open at the end of the reporting period even though the children were actually discharged months or even years prior to this time. 20 The DHS database contains placements that duplicate or overlap other placements of the same individual. While counties may hold open a bed for a child staying in a different facility for a short time, records of many placements in the DHS database duplicate or overlap other placements for several months. For example, the DHS 1996 database contains over 300 duplicate records in 20 counties. Most of these placements appear to be county corrections to placement data that were previously submitted to DHS. The average length of the duplicate placements exceeded six months. Four large counties (Ramsey, St. Louis, Dakota, and Washington) had duplicate records that increased the reported days of care by between 7 and 14 percent. Three smaller counties had duplicate records that inflated their total days of care by 5 to 7 percent.

The main reason that placement counts were inaccurate was that some counties submitted incomplete placement records to DHS because of delays in entering placement data into their own information systems. Several counties told us that their information systems do not contain complete placement information for a reporting period at the time they are required to report it to DHS (within 15 days of the end of each six-month reporting period). For example, Ramsey County staff said that their placement database is not substantially complete until approximately three to six months after the reporting period.

Another reason that the database contains inaccurate placement counts is that some counties divide continuous placements in a single foster home or residential facility into multiple placements when certain case circumstances change. For example, counties told us that this may occur due to changes in (1) the social worker assigned to the case, (2) the facility’s vendor number, (3) the case’s legal status, or (4) the funding source. In our analysis of 1997 placement data from the eight counties, we found about 600 placements that were inappropriately divided into multiple placements.

Another problem with the placement data is that counties do not consistently report unique identifiers that would allow DHS to reliably track placement histories for children who move among Minnesota counties. Presently, social security numbers are the only unique identifiers reported to DHS that can be used to track placements in different counties, but we found that these numbers were reported for only half of the children. While counties do not collect social security numbers for all cases, we think DHS could obtain social security numbers for a higher percentage of children in placement. For example, by obtaining social security numbers directly from Hennepin County, we increased the database’s percentage of children with social security numbers to over 70 percent.

Problems with placement data include inaccurate discharge dates, overlapping placements, and missing placements.

20 For example, we examined 1997 placements that DHS indicated were more than 90 days. In three of eight counties, the DHS data overstated the actual placement length by at least 90 days in more than 10 percent of the cases.

We also found that some counties assign more than one case number to a child in certain circumstances, including cases involving termination of parental rights and children under 18 bearing their own children. In these cases, DHS cannot easily track the child’s complete placement history, even within a county. For example, we found that Ramsey County changed case numbers during the past year for about 80 children who were in placement during 1997.

Currently, the department is in the process of implementing its new Social Services Information System, and this may address some of the problems we found. For example, the new system is designed to reduce duplicate records by requiring one placement to be ended before another placement is entered for the same child. Also, social workers will directly enter data into this system, and DHS hopes this will reduce delays in data entry. In addition, DHS designed the system so that counties can use it for their own management purposes, and this may contribute to improved data accuracy and timeliness. Implementation of this system will continue into 1999, so it is too early to evaluate how much the system will improve the accuracy of out-of-home placement data. In general, however, we think that the magnitude of problems with the existing database indicates that DHS should carefully monitor the quality of data produced with the new system. In addition, as recommended in Chapter 6, we think that DHS should make corrections in its existing information system so that these data can provide reasonably accurate benchmarks for trend analyses.

SUMMARY

Most of the time spent by juveniles in out-of-home placement in Minnesota results from parent-related reasons, not the conduct of the children. In particular, many juveniles are placed out of home due to a parent’s abuse, neglect, or chemical dependency. Juveniles placed out of home for parent-related reasons tend to remain in placement for longer periods than juveniles placed because of their own conduct. Juveniles placed away from home for their own conduct tend to be placed farther from home than juveniles placed for parent-related reasons.

African American and American Indian children are eight times more likely to be placed out of home as white, non-Hispanic children, and the average African American and American Indian child spends more time in placement than the average child from other racial/ ethnic groups. Although we do not fully understand the reasons for high placement rates among some racial/ ethnic groups, the disproportionate rates of placement underscore the need for placing agencies and residential service providers to aim for culturally appropriate services.

A new information system may help to address data problems.

Service Needs CHAPTER 5

The 1998 omnibus crime bill imposed a temporary moratorium on new juvenile residential facilities (and facility expansions) of 25 or more beds. 1

In part, this reflected legislative concerns about two large private facilities that were seeking licensure (and have since opened). Some legislators questioned whether there was a need for additional beds and whether Minnesota’s facilities were adequately addressing juveniles’ service needs. The crime bill also requested that our office study juvenile service needs, and we asked:

Do counties believe that they have sufficient alternatives to residential placement? If not, what types of non-residential services are most needed?

Does Minnesota need additional residential beds for juveniles? Do existing facilities meet the needs of juveniles, and is there sufficient aftercare?

What are the reasons that some children do not complete the programs in which they are placed, and what is the extent of non-completion?

Is there sufficient information about the outcomes of residential placements?

Overall, we found a more pressing need for improved availability of nonresidential services (including aftercare) than residential services. With the possible exception of foster care, most counties do not perceive significant needs for additional residential beds. Facility occupancy rates in most regions of the state have not been high enough to cause serious placement difficulties. Some counties would prefer residential services to be more responsive to juvenile needs--for example, through shorter programs, greater efforts to work with families, and more culturally sensitive programming.

Many of our findings in this chapter are based on surveys of county human services directors, county juvenile corrections supervisors, and judges. The judgments of these three surveyed groups sometimes differed, perhaps reflecting differences in the types of cases with which they were most familiar or their

1 Minn. Laws (1998), ch. 367, art. 10, sec. 15.

interpretations about what constituted ‘‘adequate’’ or ‘‘available’’ service. We asked county officials to base their responses on the types of cases for which their agency has been directly responsible, and we assumed that human services officials might sometimes have direct knowledge of placements in correctional facilities, just as corrections officials might sometimes have direct knowledge of placements in human services facilities. 2 In general, human services directors expressed less satisfaction with existing services for juveniles than did corrections officials.

NON-RESIDENTIAL SERVICE NEEDS

It is difficult to assess the need for residential placements without considering the availability of placement alternatives. In some cases, it might be possible to avoid (or shorten) a residential placement if there are appropriate non-residential programs in the juvenile’s home community. For this reason, we used surveys to document the service preferences of county human services directors and juvenile corrections supervisors throughout Minnesota. We found that:

Most counties reported a greater need for additional non-residential than residential services for juveniles.

We asked county officials what their top spending priority would be if they received additional funding to pay for out-of-home placements or non-residential services for at-risk children. According to our surveys, 71 percent of county corrections supervisors and 64 percent of county human services directors said that their top spending priority would be non-residential services. 3 In addition, 44 percent of corrections supervisors and 38 percent of human services directors said that their counties ‘‘sometimes, rarely, or never’’ have sufficient alternatives to residential placements. 4

Most counties would rather invest new funds in non-residential than residential services.

2 The surveys instructed respondents to mark ‘‘don’t know’’ if survey questions asked about types of cases or facilities with which the respondents had little or no experience.

3 Among corrections officials, 18 percent said that paying for out-of-home placement would be their top spending priority; 11 percent said ‘‘don’t know.’’ Among human services directors, 31 percent said that out-of-home placement would be their top spending priority; 5 percent said ‘‘don’t know.’’

4 Among juvenile corrections supervisors, 16 percent said that they ‘‘always or almost always’’ have sufficient alternatives, and 37 percent said they ‘‘usually’’ do. Among human services directors, 17 percent said that they ‘‘always or almost always’’ have sufficient alternatives, and 45 percent said they ‘‘usually’’ do.

We asked additional questions to determine county satisfaction with the availability of 25 specific types of services--most of which could be provided in either residential or non-residential settings. Table 5.1 shows the percentage of respondents who said they were not satisfied with the availability of these services. More human services and corrections officials said they were dissatisfied with the availability of truancy services than any other service, and many human services and corrections officials also identified services for ‘‘low-functioning’’ (i. e., low intelligence) juvenile offenders as another category of concern. Human services officials said they would like additional foster care arrangements that are designed to serve entire families (including the parents). Corrections officials said they would like to see the parents of troubled juveniles receive parenting training, and they would like to see more instruction in daily living skills for juveniles.

In each case that county officials expressed disatisfaction with service availability, we asked whether they would prefer (1) additional residential beds, (2) additional services in existing residential facilities, or (3) additional non-residential services. For all but one of the services (foster care for entire families), human services directors identified non-residential services as the most pressing need. For all but five of the services (consequence programs; services for emotionally disturbed children, female juvenile offenders, extended jurisdiction juveniles, and low functioning offenders), county corrections supervisors said that non-residential services were the most pressing need. There were no service categories for which county officials said that additional residential beds was their top priority, although some told us that they would like more beds in addition to more non-residential services.

Table 5.1: County Satisfaction With Availability of Juvenile Services Percentage Not Satisfied With Service Availability Service Human Services Directors (N= 84) Corrections Supervisors (N= 82)

Truancy services 70% 59% Foster care for entire families 68 * Services for children who are high risks of running away 67 44 Services for "low functioning" juvenile offenders 61 55 Services for offenders under age 12 60 41 Treatment/ services for serious emotional disturbances 58 49 Treatment/ services for juvenile sex offenders 54 30 Services for children who are high risks of reoffending 54 43 Instruction in critical thinking skills 49 39 Parenting skills (for parents of troubled children) 49 55 Treatment programs for entire families 49 43 Chemical dependency treatment/ services 46 41 Instruction in daily living skills 46 54 Vocational training 46 49 Intensive probation for juveniles * 30 Services for children with limited English skills * * Services for female juvenile offenders 43 43 Services for child abuse victims 40 * Suicide prevention services 39 * Services for child neglect victims 37 * Short-term "consequence" programs for offenders 35 27 Intensive, in-home case management 26 30 Services for children in permanent foster care 23 * Family counseling services 18 18 Services for extended jurisdiction juveniles * 21

* indicates categories for which at least 30 percent of respondents said "don’t know/ not applicable. "SOURCE: Program Evaluation Division surveys, August 1998.

County officials were least satisfied with truancy services.
SERVICE NEEDS

Improved non-residential services would probably not entirely eliminate the need for out-of-home placement. 5 There will always be occasions when children must be removed from home for their own protection or to protect public safety. But county officials and judges told us that effective non-residential services could help them avoid making some long-term placements:

Among both county human services and corrections officials, more than half of our survey respondents said that at least 15 percent of out-of-home placements exceeding one month could have been avoided in the past year if the county had an adequate system of non-residential services. 6

As shown in Table 5.2, most judges said there is little potential for reducing the number of out-of-home placements for extended jurisdiction juveniles and felony-level violent offenders. For all other categories of children we asked about (such as truants and misdemeanants), most judges said there is ‘‘some’’ or ‘‘significant’’ potential for reducing placements through improved community services.

Table 5.2: Judges’ Opinions About the Potential for Reducing Placements Through Improved Non-Residential Services

"For each of the following categories of children, how much potential is there in your judicial district to reduce out-of-home placements from current levels through the development of additional non-residential services?"

Percentage of Judges Who Said: Significant Potential Some Potential Little

Potential Extended jurisdiction juveniles 3% 23% 51% Felony-level violent offenders 2 23 62 Felony-level property offenders 15 52 20 Misdemeanor-level offenders 36 39 12 Truants and runaways 36 38 15 Children with emotional or mental health problems 16 51 21 Children who have been victims of abuse 20 45 23 Children who have been victims of neglect 18 50 21 Chemically-dependent children 21 50 19

NOTE: Percentages of respondents who said "don’t know/ not applicable" are not shown. SOURCE: Program Evaluation Division survey of district court judges (N= 143), August 1998.

Some residential placements could be avoided.

5 In 1996, the Minnesota Department of Human Services set a goal of reducing the number of children in out-of-home placement statewide by 50 percent by the year 2005 ( 1996 Performance Report (St. Paul, December 1996), 29). This goal was not based on a systematic review of service needs but on the belief that too many children were being placed out of home unnecessarily.

6 Eleven percent of human services directors and 12 percent of corrections officials estimated that 50 to 100 percent of their placements could have been avoided through improved non-residential services.

There have been various states and other jurisdictions that have tried to reduce out-of-home placements by enhancing home-based or community-based services. The term ‘‘wraparound services’’ has been used to describe efforts to tailor services to the needs of children and their families--often with small caseloads for social workers or probation officers and funding pooled from multiple programs. For example, Alaska significantly reduced its juvenile out-of-state placements in the late 1980s by making greater use of local outpatient care, home-based services, day treatment, and other community-based services. Massachusetts closed its state-operated ‘‘training schools’’ for delinquent youth in 1972 and developed community-based residential and non-residential services in their place. In 1987, Iowa pooled 30 categories of child welfare funds and gave counties flexibility about whether to use the funds for residential or non-residential services. In the early 1990s, Vermont enhanced local services and mandated interagency service delivery to help children with severe emotional disturbances return home from placement (or avoid placement altogether). 7

While there is general agreement among researchers and practitioners that these types of reforms are promising, there has been limited research about the effectiveness of ‘‘wraparound’’ programs and other community-based services. Some recent literature reviews have concluded the following:

‘‘There is still a critical need for a consensus to emerge regarding the existence of a best practice program (or preferably programs) that have demonstrated effective ness and can be transportable, defensible, and implementable in new locales.’’ 8

‘‘There is little evidence that community-based alternatives are any more effective in reducing recidivism than training schools. A few experiments that set out to demonstrate that small community-based programs or in-home supervision pro grams were more effective than traditional training schools failed to do so. On the other hand, meta-analysis of numerous juvenile program evaluations suggested that treatment programs that employ multimodal methods, including cognitive-be havioral and skill-oriented techniques, were more effective when run in the com munity rather than institutional settings.’’ 9

‘‘A literature is accumulating with respect to the effectiveness [of family preserva tion services], and it, too, is inadequate. . . . Studies that have included more rigor ous controls suggest that family preservation services may reduce child placement in the short run but that their effectiveness diminishes over time.’’ 10

Community based programs hold promise, but more research about their results is needed.

7 Paul Lerman, ‘‘Child Protection and Out-of-Home Care: System Reforms and Regulating Placements,’’ in Protecting Children from Abuse and Neglect , ed. Gary B. Melton and Frank D. Barry (New York: Guilford Press, 1994), 353-437; James Yoe, Suzanne Santarcangelo, Margaret Atkins, and John D. Burchard, ‘‘Wraparound Care in Vermont: Program Development, Implementation, and Evaluation of a Statewide System of Individualized Services,’’ Journal of Child and Family Studies 5, no. 1 (1996), 23-39.

8 Abram Rosenblatt, ‘‘Bows and Ribbons, Tape and Twine: Wrapping the Wraparound Process for Children with Multi-System Needs,’’ Journal of Child and Family Studies 5, no. 1 (1996), 105.

9 Elizabeth Piper Deschenes and Peter W. Greenwood, ‘‘Alternative Placements for Juvenile Of fenders: Results from the Evaluation of the Nokomis Challenge Program, Journal of Research in Crime and Delinquency 35, no. 3 (August 1998), 270.

10 Kathleen Wells, ‘‘Family Preservation Services in Context: Origins, Practices, and Current Issues,’’ in Home-Based Services for Troubled Children , ed. Ira M. Schwartz and Philip AuClaire (Lincoln, NE: University of Nebraska Press, 1995), 18-19.

‘‘In sum, although [wraparound] programs are an interesting concept and evalu ation strategies are becoming more rigorous, there is currently a paucity of valid and reliable empirical data to support their effectiveness.’’ 11

Overall, Minnesota county officials expressed to us a preference for serving more of their most troubled children through non-residential services. These services are not appropriate for all children in out-of-home placement, but they may enable counties to serve some children in less restrictive settings. This complicates the task of estimating the future need for residential services, which we discuss next.

BED AVAILABILITY

We used several methods to evaluate whether there is a need for additional beds in Minnesota residential facilities. First, we compared each region’s number of licensed residential beds to its number of children (and, alternatively, to its number of children in poverty). Second, we examined the ‘‘occupancy rates’’ of existing facilities to determine the extent to which there is unused capacity in the residential services system. Because occupancy data are not regularly collected for all facilities, we relied on (1) information on occupancy collected by a Department of Corrections consultant for December 31, 1997, and (2) similar, one-day occupancy information that we collected for additional facilities in Fall 1998. 12 Third, human services directors and juvenile corrections supervisors in each county provided us with their opinions about service needs in our August 1998 surveys.

Beds Per Capita and Occupancy Rates

11 Brady C. Bates, Diana J. English, and Sophia Kouidou-Giles, ‘‘Residential Treatment and its Al ternatives: A Review of the Literature,’’ Child & Youth Care Forum 26, no. 1 (February 1997), 43.

12 The University of Minnesota Institute for Criminal Justice surveyed facilities that were serving juveniles in court-ordered placement for delinquency. We supplemented these efforts by contacting Rule 5, Rule 8, and detention facilities that were not contacted for the Institute’s survey. As tim e permitted, we contacted some of the larger facilities surveyed by the Institute to update informatio n on juvenile populations. Facilities that we contacted had somewhat higher occupancy levels overall than those surveyed by the Institute, perhaps reflecting seasonal variations in occupancy or differences in the facilities contacted.

13 Regional comparisons should be made with care because some facilities serve children from throughout Minnesota. For example, 39 percent of the beds in the First District were at the state-r un correctional facility at Red Wing.

To assess regional variation in bed capacity, we began by examining each judicial district’s total number of beds in correctional facilities, Rule 5 facilities, and group homes. Table 5.3 shows that there were 30 licensed beds statewide per 10,000 children, and this ranged from 19 to 63 in individual judicial districts. But this disparity became smaller when we compared each district’s number of beds with the number of children in poverty. Specifically, the total number of licensed beds per 1,000 children living in poverty ranged from 13 (in Judicial Districts 2 and 4--Ramsey and Hennepin counties) to 29 (in the Ninth District--northwestern Minnesota). 13 The two districts with the lowest rates are the state’s smallest judicial districts in geographic area, and both are adjacent to districts with much higher numbers of beds per 1,000 children in poverty. Thus, many children from Districts 2 and 4 could be placed outside the districts but remain somewhat close to home.

As shown in Table 5.4, there are about 3,700 beds licensed as correctional facilities, Rule 5 facilities, and group homes. We found that:

Secure correctional facilities had relatively high occupancy rates (88 percent statewide), but the occupancy rates of other types of facilities were considerably lower.

Table 5.3: Number of Rule 8, Rule 5, and Corrections Beds by Judicial District

Judicial District Licensed Beds per 10,000 Children Under Age 18 Licensed Beds per 1,000 Children Living in Poverty

First 19 27 Second 28 13 Third 20 17 Fourth 25 13 Fifth a 37 25 Sixth 51 26 Seventh 38 23 Eighth 37 21 Ninth 63 29 Tenth 24 28

Statewide 30 20 a A new non-secure corrections facility opened in Elmore, Minnesota in 1998. Once the facility is completely open, it will provide 150 beds. At the time of our survey, it only provided 75 beds, so our analysis did not include the 75 beds not yet open.

SOURCE: Program Evaluation Division analysis of data from the Department of Human Services’ and Department of Corrections’ licensing units and the United State Bureau of the Census.

Table 5.4: Statewide Occupancy Rates for Selected Types of Residential Facilities

Facility Type Number of Licensed Beds Estimated Number

of Residents Estimated Occupancy Rate Rule 8 Group Home 659 441 67% Rule 5 Residential Treatment Facilities 918 596 65 Non-Secure Correctional Facilities a 1,600 1,228 77 Secure Correctional Facilities 575 507 88

Total 3,752 2,775 74% NOTE: The estimated number of residents and occupancy rates are based on survey data which included all but 99 of Minnesota’s 3,752 beds.

a A new non-secure corrections facility opened in Elmore, Minnesota in 1998. Once the facility is com pletely open, it will provide 150 beds. At the time of our survey, it only provided 75 beds, so our analysis did not include the 75 beds not yet open.

SOURCE: The Department of Human Services’ and Department of Corrections’ licensing units, Program Ev aluation Division phone survey of facilities (August -October 1998), and University of Minnesota’ s Institute for Criminal Justice survey of facilities (as of December 31, 1997).

There is a fairly balanced distribution of beds throughout Minnesota’s regions.

In the DHS-licensed group homes and treatment facilities, about one-third of licensed beds were unoccupied on the days we contacted them. In non-secure correctional facilities, nearly one-fourth of licensed beds were unoccupied.

Table 5.5 shows facility occupancy rates by judicial district. Several districts with shortages (or near shortages) of correctional space are anticipating increases in the number of licensed correctional beds within the district’s boundaries. In the First District, the Red Wing facility expects to increase its licensed juvenile capacity from 128 to 206 as it converts adult prison space to juvenile facility space. 14 The Second and Fourth districts (Ramsey and Hennepin counties) have been exceeding their licensed detention capacities because they have not yet opened new detention beds with the bonding money authorized by the 1994 Legislature. Hennepin County has been debating the construction of a new facility for several years, and Ramsey County started construction of a detention center addition in 1998. 15 In the Sixth District, a large private correctional facility recently opened in Buhl and is not reflected in our tables (it was not open at the time of our survey). The facility plans to have 186 beds (non-secure and secure) available in early 1999, which will more than double the number of licensed correctional beds in the district. 16 These additions will not ensure that correctional beds will always be available when needed, but they will reduce occupancy rates in the affected districts. In addition, some correctional facilities use detention and residential space interchangeably, which gives them more flexibility for accommodating placements.

Table 5.5: Estimated Occupancy Rates for Selected Facilities by Judicial District

Judicial District Rule 8 Group Homes

Rule 5 Residential

Treatment Facilities

Non-Secure Correctional

Facilities Secure

Correctional Facilities Total

First 66% 67% 114% 69% 89% Second 75 55 85 110 77 Third 53 78 56 77 62 Fourth 72 39 85 105 71 Fifth a 76 84 77 17 b 76 Sixth 55 96 90 96 87 Seventh 56 82 58 66 64 Eighth 80 70 70 87 77 Ninth 65 73 77 70 72 Tenth 73 42 71 97 73

Statewide 67% 65% 77% 88% 74% NOTE: The estimated occupancy rates are based on survey data which includes all but 99 of Minnesota’ s 3,752 beds. a A new non-secure corrections facility is opening in Elmore, Minnesota. Once the facility is complet ely open, it will provide 150 beds. At the time of our survey, it only provided 75 beds, so our analysis did not include the 75 beds not ye t open. b The Fifth District only has 12 secure correctional beds; only two were occupied at the time of our phone survey.

SOURCE: The Department of Human Services’ and Department of Corrections’ licensing divisions, Progra m Evaluation Division phone survey of facilities (August -October 1998), and University of Minnesota’ s Institute for Criminal Justice survey of facilities (as of December 31, 1997).

New facilities will be opening in some districts with bed shortages.

14 On the day we contacted Red Wing, it had 148 juvenile residents, although its licensed capacity was still 128.

15 Ramsey County’s addition will result in a net increase of 40 secure beds; Hennepin County officials told us they are considering adding 36 secure beds.

It is much more difficult to objectively assess what portion of Minnesota’s family foster home ‘‘capacity’’ is being used. As of mid-1998, Department of Human Services records indicated that foster homes inspected by counties and private agencies could legally serve up to about 13,400 children. These included a variety of types of foster care, ranging from emergency beds to treatment-oriented foster care for children requiring special services. Based on surveys of foster home licensing agencies we conducted in August and September 1998, we estimated that about 45 percent of the licensed capacity was in use. 17 But human services officials cautioned us that the ‘‘licensed capacity’’ of foster homes may significantly overstate the actual number of children that could be served at a given time. For example, counties often license foster homes for more beds than either the county or foster family prefers to have in use. Also, some foster homes meet requirements for licensure, but counties prefer to rarely (if ever) make referrals to them.

County Staff Opinions About Residential Service Needs

We surveyed county human services directors and corrections supervisors to determine their perceptions about recent trends in bed availability and areas of current need. As shown in Table 5.6,

Fifty-nine percent of county juvenile corrections supervisors said that the availability of secure detention beds in Minnesota improved in the past three years, and 45 percent said that the availability of secure residential (post-disposition) beds improved.

Fifty-one percent of county human services directors said that the availability of Minnesota foster homes operated by relatives of children improved during the past three years, but 40 percent of the directors said that non-relative foster care options declined.

The availability of secure correctional beds has improved.

16 Our data for District 5 (southwestern Minnesota) do include a new correctional facility in Elmore that had 75 beds open by October 1998.

17 The Department of Human Services does not collect information on utilization of foster care beds. We collected information from licensing agencies that represented about 88 percent of the system’s capacity, but some were unable to provide information about the number of children in subcategories of foster care.

The improved availability of secure detention and residential beds largely reflects the 1994 Legislature’s approval of $20 million in bonding authority for juvenile detention center construction grants, which were allocated to each judicial district. 18 According to the Department of Corrections, the grant-funded projects that have been completed or are in progress will double the state’s number of secure detention beds and triple the number of secure treatment beds. Some judicial districts’ projects are still underway or in the planning stages, so not all counties have benefitted from the bed expansions so far.

In contrast, many counties said they have had increased difficulty recruiting foster parents in recent years. Some said this reflects the improved economy and the increase in two-earner families. Counties would prefer to have several viable foster care options for each child being placed, but some staff told us that they have increasingly placed children in the first foster home opening that appears to fit the child’s needs and characteristics.

For various categories of residential facilities, we asked county staff whether they perceive a need for additional beds. In all categories except Rule 5 treatment facilities and Rule 8 group homes, a majority of county staff said that they think there is at least ‘‘some need’’ for additional beds. But much smaller numbers of county staff said there is ‘‘significant need’’ for new beds in these categories, as shown in Table 5.7. We found that:

County human services directors most often identified foster care and corrections supervisors most often identified secure residential facilities as the types of residential care with ‘‘significant need’’ for more beds.

Table 5.6: County Officials’ Perceptions About Recent Trends in Placement Availability, By Facility Type

During the past three years, the county’s ability to find timely, appropriate, in-state placements. . .

Percentage of Officials Who Said: Improved Stayed the

Same Declined Human services directors (N= 84):

Shelter care 38% 52% 10% Treatment foster care 29 57 14 Regular foster care 7 52 40 Relative foster homes 51 43 6 Group homes 13 77 4 Rule 5 mental health facilities 5 85 10

Corrections supervisors (N= 82): Secure detention facilities 59% 33% 5% Secure residential facilities 45 40 9 Non-secure correctional facilities 33 55 10 Group homes 11 76 10

NOTE: Percentages of respondents who said "don’t know/ not applicable" are not shown. SOURCE: Program Evaluation Division surveys, August 1998.

Many counties have had increased difficulty recruiting non-relative foster parents. 80 JUVENILE OUT-OF-HOME PLACEMENT

18 Minn. Laws (1994), ch. 643, secs. 7 and 79.

Despite the fact that half of the state’s judges, county corrections supervisors, and human services directors told us on our surveys that they often have difficulty finding Rule 5 mental health treatment beds within a reasonable distance from children’s homes, relatively few county staff said that there is a significant need for new Rule 5 beds. 19 Even in the Sixth Judicial District--which had a 96 percent occupancy rate at its Rule 5 facilities--just one of the district’s four counties (Carlton) indicated a ‘‘significant need ’’ for additional Rule 5 beds in our surveys.

Conclusion

Overall, county staff have had increasing difficulty finding appropriate non-relative foster care for children. Also, counties have occasional difficulty finding openings in the larger residential facilities at the times they need them, and (as noted in Chapter 2) they sometimes have problems finding placements for the most difficult children on their caseloads. But, particularly in light of most counties’ stated preferences for improved non-residential services,

We do not think that there is evidence of a serious statewide shortage of residential beds for juveniles, with the possible exception of foster care.

Table 5.7: Perceived Need for Additional Juvenile Facility Beds Percentage of Officials Who Said There Is: No Need for New Beds Some Need for New Beds

Significant Need for New Beds

NOTE: Percentages of respondents who said "don’t know/ not applicable" are not shown. SOURCE: Program Evaluation Division surveys, August 1998.

Except for foster care, counties generally do not have a significant need for new residential beds.

19 We found that 54 percent of human services directors, 50 percent of corrections supervisors, and 50 percent of judges said that they are ‘‘sometimes, rarely or never’’ able to make timely, appropriat e placements within a reasonable distance from the child’s home (emphasis in original question). Among facility types, this was the highest level of dissatisfaction regarding placement distance.

DHS-licensed residential treatment facilities and group homes have had relatively low occupancy rates. DOC-licensed facilities have had higher occupancy rates, but many new beds have recently been added or are in the planning stages. In general, there is not a wide disparity in the availability of residential beds by region, and bed expansions underway or already funded by the Legislature should address some of the remaining bed shortages.

A final consideration regarding future bed needs is the size of the juvenile population. Although Minnesota’s population of persons under age 20 has grown slightly in recent years, the population is expected to decline gradually in the near future. The state demographer’s office projects that the population of persons under age 20 will decline by 13 percent between 2000 and 2025.

SERVICE ADEQUACY

Even if Minnesota has enough of most types of residential beds, it is important to consider whether existing facilities effectively serve children’s needs. Our study focused on the adequacy of Minnesota’s system of residential care and did not evaluate the quality of care in individual residential facilities or the content of individual programs. However, we used results from our surveys of county officials to help us consider service adequacy, as discussed below.

Residential Program Length

Some county staff told us that they have been trying to find facilities with shorter programs for the children they are placing--partly out of concern about growing placement costs, but also because they are unconvinced that longer child placements produce better results. In some cases, counties have tried to convince facilities to shorten the length of their ‘‘standard’’ programs. Other counties have simply decided to increase their use of facilities with short programs and decrease their use of facilities with longer programs.

Our surveys of county staff indicated that:

Corrections and human services officials are less satisfied with the availability of short-term placement options (less than three months) in Minnesota than with the availability of longer-term options.

Table 5.8 shows that only about half of corrections officials and one-fourth of human services directors expressed satisfaction with the number of Minnesota residential placement options that are less than three months in length. In contrast, about two-thirds of corrections and human services officials said that there are enough placement options that are more than six months long.

The table also shows that 69 percent of human services directors expressed a desire for more placement options in which the length of stay is dictated by the child’s needs. Most corrections officials told us that there were enough such

The size of Minnesota’s juvenile population is expected to decline.

options available. The responses of corrections officials might reflect the fact that delinquent juveniles are more likely than other juveniles to be placed in programs of pre-determined length--sometimes as a way of imposing a consequence for their actions.

Programs Tailored to Juveniles

Minnesota’s array of placement options includes a mix of public and private facilities that vary widely in size, philosophy, and program content. Many facilities prefer to work with certain types and ages of children, and some facilities have staff with specialized areas of expertise. Some residential facilities operate treatment programs, while others are largely intended to provide a safe place for children to live. Within this ‘‘marketplace ’’ of facilities, it is up to counties and courts to find the best settings for individual children. In our surveys, 77 percent of corrections supervisors and 48 percent of human services directors said that residential programs in Minnesota are ‘‘usually or always’’ well-suited to the types of children their agencies try to place. 20

Although many facilities have adopted unique or specialized approaches to care and treatment, county placement staff usually expect facilities to show some willingness to tailor their programs to meet the unique needs of individuals. We asked county officials to rate various categories of facilities on their adaptability to individual needs. As shown in Table 5.9,

Treatment foster care programs--that is, foster homes that offer in-house supportive services or treatment--received the best ratings from human services and corrections officials for their efforts to tailor programs to meet children’s needs.

Table 5.8: County Perceptions About Length of Existing Residential Programs

Corrections Supervisors (N= 82) Human Services Directors (N= 84) There are enough residential placement options: Percentage

Who Agree Percentage

Who Disagree Percentage

Who Agree Percentage

Who Disagree

Less than three months 49% 43% 26% 58% Three to six months 60 27 35 43 More than six months 70 21 67 15

In which length of stay is guided by child’s needs 59 32 15 69 NOTE: The percentages of those who responded "neither agree nor disagree" and "don’t know" are not s hown. SOURCE: Program Evaluation Division surveys, August 1998.

SERVICE NEEDS 83

20 Twenty-three percent of corrections supervisors and 48 percent of human services directors said that residential programs are ‘‘sometimes, rarely, or never’’ well-suited to the children they try to place.

For each category of facilities, a majority of corrections supervisors said that facilities were ‘‘usually or always’’ sufficiently flexible in their approaches. Corrections officials’ greatest dissatisfaction was with post-disposition correctional facilities; 45 percent said that these facilities ‘‘sometimes, rarely, or never’’ adequately tailored their programs. Human services directors gave relatively low ratings to all categories of facilities except for treatment foster care. State agency staff told us that they hope that proposed rules drafted jointly by corrections and human services officials in 1998 will help to ensure that children’s needs are adequately served by residential facilities.

Working with Families

By definition, children in out-of-home placement are living apart from their parents or guardians (although some live in foster care with relatives). Some children live in residential facilities in their home communities, but others live in facilities that are far away from parents or other close relatives. The Minnesota Family Preservation Act establishes a state policy opposing unnecessary parent-child separations and favoring reunification of families, when appropriate. 21 In cases involving children needing protection or services, one of the stated purposes of state law is to ‘‘preserve and strengthen the child’s family ties whenever possible and in the child’s best interests.’’ 22

Table 5.9: County Perceptions About Adaptability of Facilities to Children’s Needs

Percentage of: "To what extent do the following types of residential facilities adequately tailor their programs to meet the individual needs of the children they serve?"

Corrections Supervisors (N= 82) Human Services Directors (N= 84) Usually or Always Sometimes, Rarely or Never Usually or Always Sometimes, Rarely or Never

"Treatment" foster care 67% 21% 61% 36% Group homes 62 38 17 77 Correctional residential facilities 54 45 5 68 "Rule 5" mental health facilities 55 18 38 60 Chemical dependency treatment facilities 56 40 25 63

NOTE: The percentages of those who responded "don’t know" are not shown. SOURCE: Program Evaluation Division surveys, August 1998.

State agencies hope that rules they have drafted for residential facilities will improve services.

21 Minn. Stat. §256F. 01.

22 Minn. Stat. §260.011, subd. 2 (a).

Some researchers have concluded that children who are more ‘‘connected’’ to family members are less likely to experience emotional distress, have suicidal thoughts, use drugs or alcohol, and engage in violent behavior--things that might result in out-of-home placement or that could make reunification more difficult. 23 Regarding parental contacts in out-of-home care, one summary of research reported that:

Study findings consistently demonstrate an association between the frequency of parental visiting and shorter lengths of stay in out-of-home care, suggesting that parent-child contacts play a role in the child’s functioning and development while in care. . . . A primary purpose of parent-child contacts during placement is the preservation of family relations to meet the child’s need for continuity in relation ships. This is just as important in cases in which the child may not be reunified with the birth family. 24

Our surveys indicated that:

Among various types of residential facilities, corrections facilities received the lowest ratings from county officials for their efforts to work with the families of the children they serve.

According to Table 5.10, most county corrections officials said that facilities other than correctional facilities ‘‘usually or always’’ made sufficient efforts to work with children’s families. Human services directors gave all types of facilities lower ratings than did corrections officials, with the lowest ratings for correctional facilities and group homes.

Table 5.10: County Perceptions About Facilities’ Efforts to Work with Families

"To what extent do staff in the following types of residential facilities make sufficient efforts to work with the families of the children they serve?"

Percentage of: Corrections Supervisors (N= 82) Human Services Directors (N= 84) Usually or Always Sometimes, Rarely or Never Usually or Always Sometimes,

Rarely or Never

"Treatment" foster care 72% 18% 51% 44% Group homes 74 26 33 57 Correctional residential facilities 37 63 4 73 "Rule 5" mental health facilities 59 16 56 40 Chemical dependency treatment facilities 78 20 46 42

NOTE: The percentages of those who responded "don’t know" are not shown. SOURCE: Program Evaluation Division surveys, August 1998.

County officials think that correctional facilities could make greater efforts to work with families. SERVICE NEEDS 85

23 R. W. Blum and P. M. Rinehart, Reducing the Risk: Connections That Make A Difference in the Lives of Youth (Minneapolis: University of Minnesota Division of General Pediatrics and Adolescent Health, 1998). Recently, some theorists have suggested, however, that parents have less influence on children’s behavior than do peers. For example, see Judith Rich Harris, ‘‘Where Is The Child’s Environment? A Group Socialization Theory of Development,’’ Psychological Review 105, no. 3 (July 1995), 458-489.

24 Inger P. Davis and Elissa Ellis-MacLeod, ‘‘Temporary Foster Care: Separating and Reunifyin g Families,’’ in When There’s No Place Like Home: Options for Children Living Apart From Their Natural Families, ed. Jan Blacher (Baltimore: P. H. Brooker Publishing, 1994), 142-3.

Aftercare

Research has shown ‘‘almost universal agreement that the level of children’s in-treatment adjustment is not predictive of their level of postdischarge functioning.’’ 25 In other words, even children who make considerable progress during their time in residential care often do not sustain this when they return home. For this reason, researchers and practitioners usually endorse ‘‘aftercare’’ services following residential placements, although research has provided little solid evidence about which specific types of aftercare may be most effective.

Some facilities provide their own aftercare services for juveniles who have completed their programs. Also, county probation officers and social workers often continue to work with children discharged from residential settings. The 1997 Legislature appropriated $130,000 in each year of the current biennium to improve aftercare services for juveniles released from facilities operated by the Department of Corrections, and it required the Commissioner of Corrections to design a ‘‘juvenile support network’’ in local communities to provide aftercare services for juvenile offenders. 26 For example, the department requires that all juveniles committed to its Red Wing facility participate in an ‘‘extended furlough’’ in a community-based group home or other facility prior to returning home. In addition, some juveniles committed to the department’s facilities have participated in meetings with people from their home communities intended to help the juveniles understand the harm they have caused.

We asked county officials and judges to evaluate the aftercare services available for children discharged from various types of facilities, and their perceptions varied. As shown in Table 5.11, we found that:

For all facility types, a minority of county juvenile corrections supervisors reported to us that adequate, appropriate aftercare services are ‘‘usually or always’’ available.

Human services directors reported that adequate aftercare services are least available for corrections and chemical dependency facilities; for other types of facilities, 50 percent or more said that adequate aftercare services are ‘‘usually or always’’ available.

For all facility types, judges who said that adequate aftercare services are ‘‘usually or always’’ available outnumbered those who said that such services are ‘‘sometimes, rarely or never’’ available. Aftercareservices are needed to help juveniles following their discharge from residential facilities.

25 Bates and others, ‘‘Residential Treatment and Its Alternatives,’’ 16.

26 Minn. Laws (1997), ch. 239, art. 1, sec 12, subd. 3. The legislation required counties, communities, and schools to develop and implement the network. It also mandated the commissioner to require CCA counties to incorporate aftercare programs into their CCA plans.

Cultural Sensitivity

Eight-eight percent of Minnesota’s 1997 population under age 18 was white (not including persons of Hispanic ethnicity). In most Minnesota counties, non-Hispanic white persons accounted for at least 95 percent of the total population under age 18. However, Chapter 4 showed that large percentages of children placed out of home in Minnesota are members of racial or ethnic minority groups.

Whatever the reason for the large number of minority children in placement, service providers’ success in serving children’s interests may depend on whether they can ‘‘connect’’ with the children. This may require an understanding of cultural differences among the children in placement and, when necessary, a willingness to tailor services. State law requires juvenile courts to ensure that reasonable efforts-- ‘‘including culturally appropriate services’’--have been made to prevent placement or reunite a child and family in cases where children have been determined to need protection or services. 27 In addition, the Minnesota Family Preservation Act requires social services agencies to ‘‘strive to provide culturally competent services’’ in their efforts to prevent unnecessary parent-child separations and encourage family reunifications. 28

As shown in Table 5.12, our surveys indicated that:

There is room for improvement in the sensitivity of facility staff to cultural and ethnic differences in the children they serve.

Table 5.11: Perceptions About Availability of Aftercare Services

"To what extent do children have access to adequate and appropriate community-based services after they have completed the following types of residential placement?"

Percentage of: Corrections Supervisors (N= 82) Human Services Directors (N= 84) Judges (N= 143) Usually or Always Sometimes, Rarely or Never Usually or Always Sometimes, Rarely or Never Usually or Always Sometimes, Rarely or Never

"Treatment" foster care 41% 51% 67% 30% 46% 38% Group homes 43 56 50 43 59 28 Correctional residential facilities 37 60 27 54 49 38 "Rule 5" mental health facilities 34 49 57 39 50 37 Chemical dependency 40 59 37 56 63 27 treatment facilities

NOTE: The percentages of those who responded "don’t know" are not shown. SOURCE: Program Evaluation Division surveys, August 1998.

Facility staff should be conscious of children’s cultural differences.

27 Minn. Stat. §260.012. The law does not contain comparable language for child delinquency cases.

28 Minn. Stat. §256F. 01.

Among counties in which racial/ ethnic minorities comprised more than 5 percent of the total population under age 18, about one-third of the corrections and human services officials we surveyed said that facility staff are ‘‘sometimes, rarely, or never’’ sufficiently sensitive to cultural and ethnic differences.

Budget Considerations

As discussed in Chapter 3, county revenues pay for the majority of out-of-home placement costs. In some counties, growth in out-of-home placement has triggered county budget crises. For instance, Hennepin County officials proposed significant cuts in social services programs in 1997 to address a multi-million dollar projected budget shortfall due to growth in placement spending. In 1998, members of the Mahnomen County board expressed concerns to legislative and executive branch leaders that the county faced a fiscal crisis, partly due to the county’s high rates of children in out-of-home placement. Many county officials have suggested that the state should provide additional financial support for out-of-home placements.

Our surveys asked county officials and judges whether budget considerations have affected service decisions. We found that:

About 22 percent of judges said that budget constraints have ‘‘usually or always’’ limited their ability to provide children with the care and services they need; another 42 percent said that this has ‘‘sometimes’’ occurred. 29

Table 5.12: Perceptions About Cultural Sensitivity of Residential Programs, Selected Counties

Are programs at the following facilities "sufficiently sensitive to cultural and ethnic differences in their resident populations?"

Percentage of Officials in Counties Where Minority Groups Comprise More than 5 Percent of the Child Population:

Corrections Supervisors (N= 31) Human Services Directors (N= 32) Usually or Always Sometimes, Rarely or Never Usually or Always Sometimes, Rarely or Never

Facilities or group homes for children with emotional disturbances 52% 35% 53% 38%

Facilities or group homes for children in need of protective services Not Not 53 34

asked asked Juvenile correctional facilities 55 35 Not Not

NOTE: The percentages of those who responded "don’t know" are not shown.
SOURCE: Program Evaluation Division surveys, August 1998.

Budget constraints sometimes affect placement decisions.

29 Twenty-nine percent said that this has ‘‘rarely or never’’ occurred, and 8 percent responded ‘‘don’t know.’’

About half of the county corrections supervisors and human services directors said that budget considerations have limited their ability to provide the care and services that children need. 30

Thirty-nine percent of corrections supervisors and 19 percent of human services directors said that some of their counties’ children needing out-of-home placement have not been placed due to budget considerations. 31

In addition, we asked county officials whether aggregate funding levels from federal, state, and county sources have provided sufficient overall funding for the services shown in Table 5.13. As a group, corrections officials expressed somewhat greater satisfaction with funding for residential services than with funding for various categories of non-residential services. But the human services directors--whose agencies pay most of the costs of out-of-home placement--overwhelmingly said that funding was inadequate in all of the categories.

When asked whether counties would likely place more children out of home if state or federal funds paid for a larger proportion of placement costs, most human services directors (63 percent) said that they would not. 32 County corrections supervisors were split, with 41 percent saying that increased subsidies would result in more placements, and 41 percent saying that they would not. 33

Table 5.13: County Perceptions About Adequacy of Funding for Juvenile Services "From state, local, and federal sources combined, there is a sufficient overall level of funding for the following services in our county:"

Percentage of: Corrections Supervisors

(N= 82) Human Services Directors (N= 84) Agree Disagree Agree Disagree Residential services 32% 50% 11% 85% Prevention and early intervention 16 68 19 79 Aftercare services 12 79 12 80 Other non-residential, community-based services 13 63 12 80

NOTE: The table does not show the percentages of those who said "neither agree nor disagree" or "don’t know." SOURCE: Program Evaluation Division surveys, August 1998.

Most counties said that funding levels for juvenile services were inadequate.

30 Among corrections officials, 52 percent agreed with a statement that budget considerations have limited their ability to provide needed services, 33 percent disagreed, and 13 percent neither agreed nor disagreed. Among human services directors, 50 percent agreed, 39 percent disagreed, and 10 percent neither agreed nor disagreed.

31 In a 1997 survey, 42 percent of human services directors said that budget considerations have caused them to limit the number of child protection cases recommended for out-of-home placement. See Minnesota Office of the Legislative Auditor, Program Evaluation Division, Child Protective Services (St. Paul, January 1998), 92.

32 Of the human services directors, 23 percent agreed with a statement that increased subsidies would result in more placements, and 14 percent neither agreed nor disagreed.

PROGRAM COMPLETION

Completing a residential treatment or correctional program is not a guarantee that a child will successfully return to the community. But there are several reasons to measure program completion rates. First, program completion rates can be a first (but not sufficient) indication of whether the goals of the placement have been met. 34 If too many children in placement do not complete programs prior to discharge, this could reflect poor initial placement decisions, ineffective facility programs, intolerance by facility staff, or impatience by the placing agency. Second, out-of-home placements are expensive, and low program completion rates may reduce the cost-effectiveness of a county agency’s placement strategies. Third, many children in placement have experienced repeated problems in school, family life, and previous social or correctional services, so it is preferable if they can be placed in programs that provide them with a realistic opportunity for success--perhaps avoiding further disruption and additional placements.

The Legislature requested that our study examine the program completion rates of juveniles in out-of-home placements and analyze the reasons for noncompletion. 35 Our surveys of county officials indicated that:

Fifty-five percent of human services directors and 35 percent of corrections supervisors believe that residential facilities discharge too many children for violating facility rules. 36

Sixty-seven percent of human services directors and 41 percent of corrections supervisors said that Minnesota residential facilities ‘‘sometimes, rarely, or never ’’ satisfactorily address the needs of resistive, aggressive, or difficult-to-control juveniles they admit. 37

Although many county officials believe that facilities discharge too many juveniles prior to program completion, several county staff told us that this does not occur as often as it did a few years ago. They said that some facilities have become more tolerant of residents who ‘‘act out’’ as their occupancy rates have declined.

Some counties think that facilities discharge too many juveniles before they complete their programs.

33 Another 16 percent of corrections supervisors neither agreed nor disagreed with a statement that increased subsidies would result in more placements.

34 The Minnesota Council of Child Caring Agencies analyzed data for 568 juveniles who were discharged from member facilities in 1995. It found that the average program completer lived in a less restrictive living setting six months after discharge than the average non-completer--presumabl y indicating better child outcomes and lower public costs.

35 Minn. Laws (1998), ch. 367, art. 10, sec. 16.

36 Among human services directors, 19 percent disagreed with the statement that ‘‘residential facilities discharge too many children for violating facility rules,’’ and 8 percent neither agreed n or disagreed. Among corrections officials, 34 percent disagreed with the statement, and 27 percent neither agreed nor disagreed.

37 Twenty-six percent of human services directors and 56 percent of corrections supervisors said that facilities ‘‘usually or always’’ address the needs of these juveniles.

The 1995 Legislature required the commissioners of corrections and human services to jointly adopt rules for residential facilities that would include ‘‘a no-eject policy by which youths are discharged based on successful completion of individual goals and not automatically discharged for behavioral transgressions.’’ 38 Under the proposed ‘‘no-eject policy’’ drafted by the departments in 1998, residents who have not reached case plan goals could not be discharged unless a review by the facility and ‘‘other interested persons’’ indicates that the discharge is warranted. 39 Thus, the proposed policy would not prevent facilities from discharging juveniles prior to program completion, but it would establish a procedure for considering alternatives.

Examples of ‘‘Non-Completers’’

To help us better understand the circumstances surrounding cases of program non-completion, we obtained facility records for a random sample of 98 children identified by their facilities as ‘‘non-completers’’ and who were discharged in 1997. 40 In about one-third of the cases we reviewed, we discussed the circumstances of the child’s termination from the program with county staff (and sometimes with facility staff, too).

The following are examples of cases in which a facility initiated a child’s discharge. Sometimes this occurred because the child refused to participate in the program, made threats, disrupted the facility’s learning environment, committed crimes, or simply did not show improvement while in the program:

During a one-month stay at a residential facility, a juvenile alternated between a pleasant demeanor and angry outbursts. The ‘‘last straw’’ occurred when she refused to take shelter during a tornado warning. When staff tried to restrain her, she kicked, spat upon, and yelled obscenities at them. The girl’s county social worker told us that she thinks that the discharge was probably in the girl’s best interests, although she was concerned about the abruptness of the discharge.

A juvenile was placed in a residential program for sex offenders for more than a year, but he made no progress. According to his county probation officer, the juvenile was on ‘‘orientation status’’ at the time of discharge--a phase that most residents complete during the first two weeks at the facility. County staff told us that the juvenile completely ignored anything that staff told him; facility staff described him as unmotivated, stubborn, oppositional, defiant, and disrespectful. County staff subsequently placed him at a program that did not focus exclusively on his sexual behavior, and he completed that program.

Proposed rules would require facilities to consider alternatives prior to early discharges of juveniles.

38 Minn. Laws (1995), ch. 226, art. 3, sec. 60, subd. 2.

39 The draft rules do not clearly indicate who would have the final authority to determine whether a discharge is warranted.

40 For 17 private and public service providers, we obtained and reviewed records for a random sample of 25 percent of children discharged from their programs in 1997 prior to program completion. The sample size is not large enough to ensure that it is representative of all placemen ts within the individual facilities or the state as a whole. We collected information from member agencies of the Minnesota Council of Child Caring Agencies, plus several large public and private residential facilities that have collected information on program completion.

A juvenile lived with a foster parent for nearly two years, during which time a strong relationship developed. While in treatment for chemical dependency problems, the juvenile stole from the foster parent to pay for drugs. Upon learning this, the foster parent suggested that it would be best to find a different foster home for the child.

Within three weeks of entering a residential facility, a juvenile refused to participate in group sessions, allegedly committed a sex offense against another resident, and then ran from the facility. The facility initiated the discharge because it thought the juvenile needed to face the consequence of a detention stay before being reconsidered for this program. County staff told us that another consideration was that the alleged victim of the sex offense was still at the facility.

Following more than four months of placement, a juvenile was discharged from a residential facility after she ‘‘destroyed the facility’s group culture’’--through bullying, threats, running away, and obstinance. She was also prosecuted for assaulting a staff member. County staff told us that the facility made appropriate efforts to work with this juvenile prior to her discharge.

In other cases we reviewed, it is possible that the placements ‘‘failed’’ because the child was put in the wrong program or did not receive services at the facility that addressed his or her needs:

A juvenile offender was discharged from a Rule 5 facility six months after placement. The child ran away twice and was not apprehended after the second run. The child’s probation officer told us that this facility had been selected at the urging of the public defender; the probation agency thought that a correctional setting was better suited to this offender’s level of sophistication. In addition, the probation officer said that the facility was unable to provide culturally appropriate services for the child (who is Asian American), and she said that the program’s inability to provide effective services for the child was apparent long before the escape.

At the time a juvenile was placed in a correctional facility, there were no indications of mental health problems. He had been examined by a psychiatrist and psychologist, and he was not in special education. During less than three months in placement, the juvenile destroyed property, was physically and verbally abusive, did not follow directions, and exhibited ‘‘out-of-control behaviors.’’ The county probation officer thought that the facility staff did their best to help the juvenile, but the juvenile was discharged so that that he could receive treatment for emotional/ behavioral problems.

Some early discharges are initiated by facility staff.

A juvenile offender with a history of drug use was placed in a correctional facility. His parents had wanted him placed in a chemical dependency treatment facility instead, but their preferred facility did not have a vacancy at the time. The boy’s mother had been assaulted by him just prior to placement and, for her own protection, she did not want to wait for the preferred facility to have an opening. The boy ran from the correctional facility, and following his return he did not comply with program requirements. Several weeks after this placement was made, the chemical dependency program originally preferred by the family had an opening, and the boy was transferred there.

In some cases, children were removed from a facility by the county--but against the wishes of facility staff. Some facility staff expressed concerns to us that this is increasingly common. The following is an example of such a placement:

Over the objection of facility staff, a county removed a juvenile from a facility following a five-month stay. The county social worker told us that the juvenile was making progress at the facility and had not completed the facility’s treatment plan at the time of removal. But the county philosophy was to try to bring kids back to the community when their behaviors become more manageable--partly as a way to save money. The juvenile was placed in a group home, which was unable to provide the amount of structure he needed. The group home requested his removal, and the county placed him back in the original residential facility. The social worker told us that, in hindsight, it was probably a mistake to have removed him from the original facility so quickly.

We reviewed some cases in which county staff removed children from placement due to concerns about a facility’s programs or supervision. For example:

A county removed 11 children from a facility at one time. County staff concluded that the facility had inadequate supervision, which enabled children to engage in consensual and non-consensual sex. The county requested staffing changes at the facility, but the facility did not make changes that satisfied the county.

A county social worker thought that a facility’s psychiatrist had prescribed an inappropriate combination of medications for a juvenile on her caseload, and she also had concerns about the amount of restraint being used by the facility and its unwillingness to modify its programs. The juvenile told the social worker that one of the facility staff bit her, and she wanted to leave the program. The social worker told us that the facility was unresponsive to her concerns, and she requested removal of the child.

Some early discharges are initiated by county staff.

Finally, we reviewed cases in which voluntarily-placed children failed to complete programs because their parents or guardians chose to remove them from a facility. For example:

Four months after a girl was voluntarily placed at a residential facility for emotional/ behavioral problems, her parents removed her so that they could have a church conduct a ‘‘spiritual intervention’’ for her problems. County staff considered whether to seek a court order to continue the placement but decided to respect the family’s religious views.

Rates of Program Completion

As reported in Chapter 2, less than half of the county officials we surveyed said that they ‘‘usually or always’’ have sufficient information about rates of program completion by facility residents. Facilities are not required by state law or rule to track information on program completion, and some of them do not systematically track completion rates. Facilities that do monitor program completion have varying ways of defining ‘‘completion.’’ For instance, some facilities consider children to be ‘‘completers’’ if they remain in a program for a certain period of time--such as the standard length of the program, or the length of time required by the court. Other facilities define ‘‘completion’’ based on the child’s progress rather than the time spent in the facility--that is, ‘‘completers’’ are children who have successfully accomplished most or all of the goals established at the time they were placed.

Although the definitions of program completion vary somewhat, we reviewed existing program completion data for a variety of private and public facilities. 41 The most comprehensive data on program completion were available from a consortium of private residential facilities (including Rule 5 facilities, correctional facilities, group homes, and treatment foster care homes). 42 These data indicated that no single factor accounted for the majority of instances of program non-completion. Runaways accounted for 25 percent of the 1992-97 non-completers. 43 In addition, 43 percent of discharges were initiated by the facility or foster parents, 21 percent were initiated by county social workers and probation officers, and 7 percent were initiated by the courts. 44 In each category of residential facilities, we found that African American and American Indian juveniles had lower rates of program completion than white, non-Hispanic juveniles.

Definitions of program completion vary.

41 The Minnesota Council of Child Caring Agencies (MCCCA) is a voluntary association of private therapeutic service providers, and the council has regularly collected a wide variety of information, including program completion data. In addition to reviewing 1992-97 MCCCA data, we collected 1997 program completion data for eight public correctional programs, three private correctional programs, all out-of-state delinquency placements in residential facilities, and all juveniles discharged from chemical dependency residential placements. Except for chemical dependency treatment programs, we limited our analysis to programs exceeding one month in length.

42 Minnesota Council of Child Caring Agencies.

43 Many counties have policies about how long they will continue to pay for an unoccupied bed at a facility, such as one or two weeks. If the child does not return during this time--and sometimes even if he or she does--the placement is terminated.

44 These percentages are based only on cases for which a reason for discharge was indicated; 18 percent of cases did not indicate the reason. In some cases, facilities identified more than one re ason for the discharge.

Among all public and private programs we examined, those in facilities licensed by the Department of Corrections tended to have the highest completion rates--10 of 15 reported rates that exceeded 80 percent. Similarly, 77 percent of delinquents placed outside of Minnesota in fiscal year 1996 completed their programs. 45 In part, correctional facilities have high completion rates because the courts sometimes expect offenders to, at a minimum, remain in placement for a certain period of time as a consequence for their actions. 46 Four programs had rates below 60 percent, but three of these were lengthy programs for difficult offenders (such as sex offenders and repeat violent offenders). The director of one program with a relatively low completion rate said that it wants its offenders to ‘‘do the program, not do time’’--but some counties have routinely removed offenders from the facility at the end of six months.

Rule 5 mental health treatment facilities tended to have somewhat lower completion rates--usually between 50 and 75 percent. This may reflect the fact that Rule 5 programs are, on average, longer than correctional programs. Also, discharge from Rule 5 facilities usually depends on the achievement of individual therapeutic goals, while some other types of facilities have programs of pre-determined length.

Statewide, chemical dependency placements tended to be shorter than most of the other types of placements we reviewed, yet their completion rates were relatively low. For example, the average juvenile who completed an inpatient chemical dependency program stayed for less than one month, but only 68 percent of juveniles who entered these programs completed them. Chemical dependency halfway houses and extended care programs averaged just under four months for program completers, but the completion rates were 42 percent for halfway houses and 65 percent for extended care. Several county staff told us that they have had particular concerns about the extent to which chemical dependency programs discharge children prior to program completion--perhaps because they are unequipped to handle disruptive children or because they have waiting lists and need not tolerate uncooperative residents.

Correctional facilities reported the highest rates of program completion.

45 Excludes juveniles still in programs at the end of fiscal year 1997.

46 We did not collect information from the Anoka County Juvenile Center or the Department of Corrections facilities at Red Wing and Sauk Centre--all of which have guidelines that indicate presumptive lengths of stay based on offenses committed and prior history. Staff in these facilitie s said that it would be extremely rare for an offender to leave the facility before this amount of tim e elapses and program goals are met.

‘‘Treatment foster care’’ programs reported the lowest rates of program completion--usually between 33 and 50 percent. This may partly reflect the relatively long average periods of time that residents tend to remain in such settings, as well as the more subjective definition of program ‘‘completion’’ that has been used by some of the treatment foster care facilities. For example, one program director told us that her agency has computed completion rates based on judgments about whether the child and his or her family have a healthy relationship at the end of the foster care placement. 47

Overall, variations in program completion rates may reflect differences in the types of children served, program goals, program length, program effectiveness, and the extent to which counties and others have removed children from programs prior to completion. But, although these differences make it difficult to compare the completion rates of individual facilities, we think that placement agencies should be able to consider program completion rates (and the reasons for program non-completion) at the time they are considering child placements. In Chapter 6, we recommend that facilities be required to track information about the extent of and reasons for program non-completion.

OUTCOME MEASUREMENT

It would be useful to know not only whether juveniles complete the residential programs they enter, but also whether the programs have the intended long-term effects. Juveniles are placed for different reasons, so there may be a variety of desired outcomes, such as law-abiding behavior, placement in a permanent home, sobriety, and protection from subsequent maltreatment. This complicates the task of measuring placement outcomes.

We found that:

There is very limited information available regarding the effectiveness of Minnesota’s juvenile residential services.

There are no statutory requirements for counties to report information on placement outcomes, but we thought that some agencies might do so for purposes of making better placement decisions. According to our surveys of county officials, however, only 7 percent of human services directors and 7 percent of county corrections supervisors said their agencies had produced summary information during the past year on the success of children subsequent to out-of-home placements.

In 1998, the Minnesota Department of Human Services convened a task force of county and state officials to develop consensus about the most appropriate outcome measures for child welfare services. The task force identified 19 measures of outcomes that should be tracked regularly. For example, the task force recommended measuring the percentage of children who are in permanent living arrangements within 12 months of out-of-home placement. The department hopes to begin analyzing statewide performance on these measures in 1999.

Counties and courts would benefit from improved information on program outcomes.

47 This may be the biological or adoptive family, a relative with whom the child is living, or a permanent foster family.

Sixty-three percent of human services directors (but only 32 percent of juvenile corrections supervisors) told us that residential service providers should be held more accountable than they now are for the success or failure of the children they serve. 48 And, as discussed in Chapter 2, a majority of county corrections and human services officials told us they would like more information about the extent to which residents’ problem behaviors recur following discharge from residential programs. Some service providers have made useful efforts to monitor service outcomes, while others have done little formal follow-up with discharged residents. 49

To properly measure placement outcomes, it is necessary to know the goals of placements. The goals of placements are diverse, reflecting a diverse population of children in a wide variety of residential settings. But, as shown in Figure 5.1, our surveys indicated that:

A majority of county human services directors and juvenile corrections supervisors said that judges ‘‘sometimes, rarely, or never’’ clearly specify the intended purpose of the placements they make.

Figure 5.1: Extent to Which Judges Clearly Specify Placement Goals, According to County Officials

7 25 36 31 18 29 35 16 0 10 20 30 40 50

Always or almost always

Usually Sometimes Rarely or never Human services directors Corrections Supervisors

NOTE: The survey question was: “In court dispositions that involve child placements, do judges clearly specify the intended purpose of each placement?”

SOURCE: Program Evaluation Division survey of county human services directors (N= 84) and corrections supervisors (N= 82), August 1998.

Percent of county officials

The courts often do not clearly specify placement goals.

48 Of the human services directors, 44 percent said that courts should be held more accountable, and 35 percent said tht counties should be held more accountable. Of the corrections supervisors, 7 percent said the courts should be held more accountable, and 16 percent said the counties should be more accountable.

49 For more than ten years, the private facilities who are members of the Minnesota Council of Child Caring Agencies have collected information about each child who is a resident in their programs. For instance, these facilities monitor changes in children’s living arrangements from jus t prior to facility admission to six months after discharge. The council’s public reports include summary information about various categories of facilities, but not about individual facilities.

In addition, 62 percent of human services directors and 23 percent of corrections supervisors said that they believe that state law should be amended to require the courts to explicitly identify the purpose of each placement made.

In our view, improved data on juvenile outcomes could help to inform many decisions that are now made blindly. For example, some counties told us that they have started trying to place children for shorter time periods. Corrections staff in one county expressed concerns to us that its court makes nearly all placements in ‘‘consequences’’ types of programs and that it rarely places children in the therapeutic programs that it once used more often. Some counties have reduced their use of Rule 5 mental health facilities by placing more children in group homes that are less expensive and closer to home. Due to budget constraints, one county has increasingly placed sex offenders in short-term and outpatient programs, rather than the longer programs it used to use. Without careful monitoring, it is unclear whether these changes will better serve the interests of the children involved or protect public safety. As one analysis of recent reforms suggested,

Current evaluations [of alternative service systems] tend to emphasize out-ofhome placement reductions and fiscal savings per capita served, but there is still insufficient evidence that youth are really ‘‘better off’’ for having been clients of the referred systems. . . . Much more is needed to yield convincing evidence that families function more smoothly, youth are less deviant, and presenting psychiat ric symptoms have been reduced. 50

SUMMARY

County officials generally perceive a greater need for additional non-residential services for juveniles than for additional beds in residential facilities. Occupancy rates in some categories of residential facilities are not particularly high, and many beds have recently been added (or are in the planning stages) for the categories of facilities that have had the highest occupancy rates (secure and non-secure correctional facilities). Many counties would like juvenile facilities and aftercare services to respond more directly to the needs of juveniles and their families. Regardless of the types of services counties use, there is a need for improved measurement of program outcomes.

Without better outcome data, it is hard to judge whether placement practices are serving children’s best interests.

50 Paul Lerman, ‘‘Child Protection and Out-of-Home Care: Systems Reforms and Regulating Place ments,’’ in Protecting Children from Abuse and Neglect: Foundations for a New National Strategy , ed. Gary B. Melton and Frank D. Berry (New York: Guilford Press, 1994), 385.

Discussion and Recommendations CHAPTER 6

Historically, the state’s role in child placement has been very limited. State laws provide broad guidance about when placement may be appropriate,

leaving these decisions largely to the discretion of courts and counties. State funding pays for some out-of-home placement and family preservation costs, but counties have provided much larger amounts of funding. State agencies license the facilities in which children are placed, but it is up to the courts and counties to select the facilities that best address the needs of children.

In this chapter, we offer recommendations about what the Legislature and state agencies should do to improve services for some of Minnesota’s most troubled children, while preserving the important roles played by the courts and county agencies. Most of our recommendations relate to existing state and county responsibilities, and we hope that most could be accomplished with existing resources--perhaps reallocated from less pressing activities. In one case (development of better information on placement outcomes), we have asked state agency staff to estimate the cost of information collection before the Legislature mandates the agencies to prepare statewide reports.

STATE FUNDING ASSISTANCE

County officials have expressed concerns to legislators about the burden of out-of-home placement expenditures on local property taxes. As we discussed in Chapter 3, local revenues pay for the majority of placement costs, and Minnesota state government directly funds a smaller proportion of child welfare costs than most states.

The Legislature has made some recent efforts to help local governments pay for out-of-home placement costs, aside from state block grants for community corrections and social services. In 1994, the Legislature created a category of local government aid called ‘‘family preservation aid.’’ The Legislature appropriated $1.5 million for the first year and declared that the purpose of the aid ‘‘is to reduce the rate of increase in the costs of out-of-home placement of children and concomitant increases in county property taxes.’’ 1 The 1998 Legislature authorized an additional $30 million in family preservation aid in 2000. 2

The state’s role in out-of-home placement has been very limited.

1 Minn. Laws (1994), ch. 587, art. 3, sec. 13.

2 Minn. Laws (1998), ch. 389, art. 4, sec. 8 and art. 2, sec. 20.

Department of Revenue staff told us that the main intent of the aid was to provide property tax relief, rather than funding new services. By law, aid has been allocated to counties based on their number of children in out-of-home placement and their income maintenance caseloads. The Legislature required the commissioners of Revenue and Human Services to recommend a new formula for distributing aid, starting with aids payable in 2000.

In 1997, the Legislature appropriated $4.0 million for juvenile residential treatment grants to counties for the first six months of 1999. 3 These grants were largely intended to offset higher costs that counties could incur if they place juvenile offenders at the state-operated Red Wing and Sauk Centre facilities. Starting in 1999, the Legislature required counties to pay the full per diem costs for placements at these two facilities (excluding education costs); previously, counties that did not participate in the CCA were not charged for these placements. Counties may use the new grants to pay for placements at Red Wing and Sauk Centre, but they can also use the grants to pay for other placements.

We offer no recommendations regarding the proper level of state funding for out-of-home placement or child welfare services. More state funding might be justified by inadequacies in some residential and non-residential services (Chapter 5) or by Minnesota’s relatively low state financial contribution to these services compared with other states (Chapter 3). Also, state assistance could be justified by the fact that counties do not have the ability to fully control placement spending. Many placement decisions are made by the courts, and the number of children in placement in counties is positively related to broad measures of social well-being, such as poverty levels. Finally, some relatively poor counties have large numbers of children in placement but limited local funding to pay for them. On the other hand, counties have considerable discretion about which children to recommend for placement, so a significant local role in placement funding may encourage better decisions and closer ongoing scrutiny. Also, some analysts have suggested that Minnesota has not made maximum use of non-state revenue sources for out-of-home placement, such as federal funding and parental fees. 4

There are a variety of ways that the Legislature could allocate state funding for children in placement or at risk of placement, and we did not conduct an in-depth analysis of alternative measures of county need and fiscal capacity. However, we think that a funding approach that is tied too closely to out-of-home placement could create incentives for placement or penalize counties that have invested in placement alternatives. In addition, we have concerns about the accuracy of the Department of Human Services’ historical data on the number of children in placement in individual counties. Thus, we recommend:

Allocating funds based on the number of children in placement could reward counties that have relied too heavily on out-of-home placement.

3 Minn. Laws (1997), ch. 239, art. 1, sec. 12, subd. 4.

4 For example, F. C. Valentine and Associates, Children’s Initiative Project (draft), Nov. 9, 1998; Minnesota Department of Human Services, Report to the Minnesota Legislature on a Uniform Relative Contribution Schedule to Reimburse Costs Associated with Out-of-Home Placement (St. Paul, December 1997).

The Legislature should not allocate funds to counties based solely or largely on their historical number of out-of-home placements (or levels of placement spending). 5

Likewise, as a general rule, we think that the Legislature should not restrict the use of funds to reimbursement of out-of-home placement costs. Counties expressed a strong desire in our surveys for improved non-residential services, and most told us that non-residential services are a higher spending priority than residential services (Chapter 5). In our view, counties should have the flexibility to use funds to pay for whatever residential or non-residential services will best serve the needs of children and families.

Although we think that counties should have ‘‘flexible funds,’’ this does not mean that counties should have less accountability for results. Later, we offer suggestions for better ways to monitor placement outcomes.

FACILITY MORATORIUM

The recent licensure of two large, privately-operated correctional facilities in Minnesota led the Legislature to pass a temporary moratorium on further licensure or expansion of juvenile residential facilities that would add 25 or more beds. Legislators were concerned that communities in pursuit of new jobs and local revenues have used public subsidies to entice facilities. Legislators also questioned whether new facilities would: (1) add too many beds to the juvenile service system, (2) duplicate existing services, rather than addressing unmet service needs, or (3) ‘‘warehouse’’ juveniles in large, impersonal settings.

Our study indicates that Minnesota counties would, for the most part, prefer to expand non-residential programs before expanding residential services, and we did not find evidence of a need for significant numbers of new beds statewide in residential facilities (except, perhaps, in foster care). Partly as a result of these findings, Department of Corrections officials told us they would favor an extension of the moratorium, or at least an approach that would ensure that proposed new facilities demonstrate that they are ‘‘needed’’ before receiving a license. In addition, DOC expressed concern that allowing additional beds might encourage courts and counties to make residential placements, rather than seeking less restrictive alternatives. Also, DOC officials said that a moratorium would help to protect the investments made by facility operators.

Although the state may not need more residential facilities, a moratorium might inhibit improvements in service quality and cost.
DISCUSSION AND RECOMMENDATIONS

5 If the Legislature wants to use historical county spending levels as a measure of service need in a funding formula, we think that it should use a measure of aggregate spending for both residential and non-residential services.

For several reasons, however, we think that an extension of the moratorium could unfairly constrain placement options for counties (which pay for most placement costs) and courts (which are responsible for making case dispositions that serve the child and protect public safety). First, counties and courts expressed to us some concerns about the quality and cost of residential services now available in Minnesota facilities, and we think that it is important for them to have a variety of placement options. A moratorium might protect existing facilities from new competition and make them less responsive to the needs of counties and courts. We think that counties and courts are in the best position to judge whether to place their children in new or long-standing facilities, or in large or small facilities--so long as the facilities meet basic licensing requirements that help to ensure quality service. In addition, there are categories of juveniles--such as juvenile offenders with low intelligence--who are not adequately served by existing facilities, sometimes resulting in out-of-state placements.

Second, the moratorium was adopted in 1998 largely in response to concerns about additions of correctional beds, but our survey of county corrections officials indicated that more would oppose rather than favor an extension of the moratorium, as shown in Figure 6.1. Human services directors tended to favor an extension of the moratorium, but DHS has not received a proposal for a facility large enough to be subject to the moratorium for more than 25 years.

Third, while some people believe that counties and courts will fill to capacity whatever number of beds Minnesota licenses, this is not currently the case. There are many vacant beds in juvenile residential facilities, and counties have increasingly looked for alternatives to expensive, long-term residential placements. In contrast to some other types of facilities that have been subject to a statewide bed limit, such as nursing homes, most of the costs of juvenile out-of-home placement are borne by the counties involved in the placement decisions.

Overall, we do not think that Minnesota has a significant shortage of residential beds for juveniles, but we think that a moratorium could limit the responsiveness of service providers to juveniles’ needs. We recommend:

Figure 6.1: County Officials’ Opinions About Whether to Extend the Moratorium on Facility Licensure

24 35 40 46 32 21 0 10 20 30 40 50

Favor Oppose No opinion Percent Corrections Supervisors Human services directors

SOURCE: Program Evaluation Division survey of county corrections supervisors (N= 82) and human services directors (N= 84), August 1998.

A moratorium could constrain the choices of local agencies that pay for most placement costs.

The Legislature should not extend the moratorium on large, new residential facilities (or facility expansions), which is scheduled to expire in mid-1999.

An alternative to a moratorium might be a facility ‘‘need certification’’ process, which would allow new facilities to be licensed if they could demonstrate a need for their services. Presently, there is such a process in Minnesota for facilities serving chemically dependent and developmentally disabled persons. 6 In both cases, the Commissioner of Human Services must determine whether there is a need for the program prior to licensure, based on a recommendation of the board from the county in which the facility would be located. 7 But state rules governing this process have not clearly defined the concept of ‘‘need’’--for example, whether the home county is supposed to determine only whether its needs are addressed by the facility (and not statewide needs), and whether facilities can be ‘‘needed’’ if they duplicate other facilities but offer lower costs or better service. In fact, we question whether it would be possible to develop a meaningful need certification process that would truly distinguish ‘‘needed’’ from ‘‘unneeded’’ facilities. 8 While we are concerned that communities could promote the development of new facilities for reasons of economic development, we think that it would probably be best to let counties and courts determine which facilities are ‘‘needed’’ through their actual placements, rather than having state regulators try to evaluate the ‘‘need’’ for a facility before it has opened.

With or without a moratorium (or need certification process), the departments of Corrections and Human Services should continue their efforts to encourage development of residential or non-residential services that respond to unmet service needs. For example, the Department of Corrections has revised the programs at its Red Wing facility and created a community-based prevention program at Camp Ripley, and it plans to add a mental health program at its Sauk Centre facility. The Department of Human Services has fostered the development of ‘‘wraparound services’’ in 22 mental health collaboratives throughout Minnesota. Although the service preferences of courts and counties may change over time, we think the departments deserve credit for such efforts to address service needs.

State agencies should continue their efforts to foster improved services.

6 See Minn. Rules 9525.0036 and 9530.6800.

7 In addition, existing programs serving developmentally disabled persons must undergo the needs determination process every four years.

8 If the certificate of need process made the county responsible for submitting recommendations to a state commissioner, this might sometimes impede development of facilities that would serve a statewide need. For example, a county’s residents might object to the development of a non-secure facility specializing in juvenile sex offenders, even if counties throughout the state believe that there is a need for more residential beds for this population.

PLACEMENT DECISION MAKING, CASE MANAGEMENT, AND ACCOUNTABILITY

Many counties told us that interagency placement screening teams have helped them to make more consistent placement decisions, contain placement costs, and identify needed services for individual juveniles. State law prescribes who should be on these teams and how their recommendations should be used, but the law does not require that each county have such a team. Presently, about half of Minnesota’s counties have juvenile screening teams consistent with Minn. Stat. §260.151, subd. 3. We recommend:

The Legislature should require all counties to establish multi-disciplinary juvenile screening teams.

In counties where these teams exist, current law only requires the teams to evaluate cases involving juveniles being placed for treatment of chemical dependency, emotional disturbances, and developmental disabilities. The law does not require review of correctional placements, among others, which has been one of the fastest growing categories of placement spending in recent years. We think that counties should be required to have screening teams review all placements in treatment facilities and all court-ordered placements potentially exceeding 30 days--including post-dispositional placements in facilities licensed by the Commissioner of Corrections. In our view, an expanded role for juvenile screening teams will enhance accountability, while helping to ensure that juvenile service needs are identified.

In addition, we think that there is room for improvement in the assessment of juvenile offenders’ risks of reoffending. Such assessments not only can help counties decide which children need the most intensive services (perhaps including placement), but they can help counties to identify specific services that address juveniles’ individual needs. Also, studies have shown that classifying offenders using formal instruments is more accurate than approaches that rely on the intuition of corrections staff. 9 Minnesota counties use a variety of approaches to assess offender risk and needs, but some counties do not formally assess risk or needs at all. We recommend:

The Legislature should require each county corrections or court services agency to adopt written policies for classifying the risks and service needs of juvenile offenders.

This requirement would be similar to existing statutory requirements for adult offenders. Department of Corrections officials told us that they would favor implementation of uniform classification practices throughout the state. Presently, however, there are a variety of assessment instruments used by local correctional agencies. It might be difficult to get consensus on a single assessment approach, but the Department of Corrections should at least consider ways to help county staff better understand juvenile risk and needs assessment through training programs or discussions of existing county assessment practices.

Screening teams should review lengthy placements.

9 See discussion in Office of the Legislative Auditor, Funding for Probation Services (St. Paul, January 1996), 76-77. Formal instruments often consider factors such as prior offenses and placements, age at first offense, substance abuse, school behavior and attendance, family stability, and peer relationships.

In our view, there is a critical need within the juvenile services system for improved information about service outcomes. Presently, counties and courts are trying a variety of strategies to reduce placement costs, protect the public, and serve children--but there is little systematic monitoring about the impact of these efforts on the children or the public. Part of the reason that it is difficult to measure placement results is because there is such a variety of children in placement, and the goals of these placements vary widely and are not always well articulated. As described in Chapter 5, many county officials told us that court dispositions do not typically provide a clear statement of a placement’s goal( s), and many said they would support changes in state law to requires courts to clearly state the purpose of each disposition. We recommend:

The Legislature should require courts to state in their dispositions the intended outcome( s) of each juvenile placement made under their authority. We recommend that the Legislature establish a working group of judicial, legislative, and executive branch representatives to (1) develop a uniform list of possible placement goals from which judges would select, and (2) identify steps required (and related costs) for state agencies to collect summary information on achievement of these goals.

For example, we anticipate that the court’s goal for many delinquent offenders will be for the offenders to be law-abiding--regardless of whether they are placed in treatment or punitive programs. For children with this goal, it would make sense to examine rates of new offenses following their placement. Likewise, the court might set a goal of sobriety for children with drug or alcohol problems, a goal of permanency for children placed in foster homes, and a goal of safety from abuse and neglect for children placed out of home for their own protection. We think that the working group should aim to develop a list of goals for which outcomes could be measured regularly and consistently. 10 The goals set by the courts should not substitute for individualized, detailed goals that counties and facilities should continue to develop during case planning.

We recognize that tracking the outcomes of placement goals may be complicated, perhaps involving information from multiple data systems. For this reason, we suggest having the working group explore how such tracking would be done, whether it would entail additional costs, and when it could be accomplished. After this group completes its tasks,

The courts should specify the goals of each placement, and progress toward those goals should be monitored.

10 The Legislature could consider broader requirements--such as (1) requiring the courts to identify intended outcomes for each court disposition, not just each court-ordered placement, and (2) requiring county agencies to identify intended outcomes for each voluntary placement they arrange. Under such requirements, DHS and DOC would have to monitor the outcomes of additional cases, including some juveniles in non-residential services. But we think that it may be more manageable to begin on a smaller scale by first improving outcome measurement for children in placement.

The Legislature should require the departments of Human Services and Corrections to regularly report on the extent to which the goals of court-ordered placements are met.

Another issue that we think the Legislature should address is case planning. In Chapter 2, we observed that state law requires development of case plans for children found by the court to need protection or services, but it does not require this for court-adjudicated delinquent children. Nevertheless, the ‘‘umbrella rule’’ on residential facilities drafted by the departments of Human Services and Corrections in 1998 makes numerous references to juveniles’ county case plans and case plan managers. We think this issue needs clarification. We recommend:

The Legislature should require counties to develop juvenile case plans following delinquency dispositions. Consistent with requirements for CHIPS cases, the plans should identify any social and other services that will be provided to the child and child’s family, whether in residential or non-residential settings.

Also, the draft ‘‘umbrella rule ’’ requires the facility and placing agency to develop a ‘‘transitional services plan ’’ for each child discharged from a residential program. The draft rule does not indicate who is responsible for monitoring compliance with this plan, and there has sometimes been confusion about the respective responsibilities of counties and residential facilities in arranging for aftercare services. To help ensure that juveniles receive the services they need following placements in residential facilities, we recommend:

The Legislature should clarify in law that counties are responsible for monitoring implementation of ‘‘transitional services plans.’’

Under this recommendation, counties would not necessarily provide the aftercare services, but they would be responsible for seeing that aftercare services were arranged for and provided. 11

We considered whether to recommend that the Legislature require development of statewide or county criteria for out-of-home placement. In Chapter 3, we said that there is considerable variation in county rates of out-of-home placement, partly reflecting varying county philosophies. In Chapter 2, we noted that most county officials we surveyed opposed the idea of statewide placement criteria, although they were somewhat more receptive to development of county criteria. We are unaware of other jurisdictions that have developed detailed criteria for all categories of child placements that might serve as a model for Minnesota or individual counties. In fact, the research literature has provided very limited insight into which types of children fare best in out-of-home care. We think that counties should, at their own initiative, think of ways to improve their out-of-home placement criteria. For now, however, we do not recommend that the Legislature mandate each county to develop placement criteria. Instead, we hope that the involvement of various professionals on each county’s juvenile screening team will help to ensure that placement decisions are appropriate and consistent.

Case plans should be required following all dispositions.

11 Juveniles who complete the ‘‘Prepare’’ program at the Department of Corrections’ Red Wing facility are placed in community-based group homes or facilities prior to returning home. We think that it is appropriate for the department, not counties, to determine how long to keep children in placement (including transitional living arrangements) who have been committed to the commissioner’s custody.

Finally, as counties continue to explore less restrictive ways to serve juveniles, we think that one of the less restrictive placement settings needs a clearer definition. Although county staff often refer to ‘‘treatment foster care’’ or ‘‘therapeutic foster care,’’ this category of service is not defined in state law or rule. In general, treatment foster care homes provide services in-house that are not provided by other foster homes (or that are provided through community services). Sixty percent of county human services directors told us that the components of treatment foster care should be clarified in law or rule. 12 We recommend:

The Department of Human Services should adopt state rules that outline the components of treatment foster care.

INFORMATION SYSTEMS

One of the reasons we were asked to conduct this study was that existing state information systems could not answer some very basic questions about Minnesota’s out-of-home placements. For example, legislators have been unable to determine the overall cost of out-of-home placements in Minnesota because DHS’ information system contains data on only a portion of counties’ correctional placements, and the remaining placements are not reported to any statewide information system.

To address the need for a comprehensive database on child placement (and related spending), we recommend:

At a minimum, we think that counties should be required to annually report to DHS summary spending data for various types of correctional placements that are not paid for by social services budgets. These data should be available from various county departments--usually corrections or law enforcement. 13

12 Seventeen percent disagreed with the following statement: ‘‘State laws or rules should be amended to clearly define the components of ‘treatment foster care.’’’ Twenty-three percent neither agreed nor disagreed.

13 In two counties (Rice and Washington), the placement costs of all delinquent juveniles are paid for by the corrections budget, and it would be useful for DHS to collect summary information from these counties on the types of facilities used for these placements.

In addition, the work group should identify ways for county corrections agencies to report information on individual placements that can be integrated with the DHS placement data. If feasible, we think this should include all placements, including pre-adjudication detention as well as post-adjudication placement. A common personal identifier--such as a social security number--would be required to fully integrate corrections and human services placement data. We found that DHS’ statewide placement database had social security numbers for only about half of the 1997 cases. Furthermore, the department’s new Social Services Information System will rely considerably on a new system of personal identification numbers, which will be different from identifying numbers used by agencies other than social services. To help ensure that it is possible to integrate information from social services, corrections, and other agencies to determine a complete and unduplicated placement history of each juvenile, DHS and DOC should encourage local agencies that pay for placements to collect social security numbers whenever feasible for juveniles in placement (in addition to any other identifiers used). 14

We found that DHS’ database of county-reported information on individual child placements has had many problems--reflecting county errors, the design of the information system, and an inadequate level of scrutiny by DHS. Specific problems included misreported discharge dates, duplicate cases, missing cases, and the absence of facility identifiers. We were able to correct the database for some, but not all, of these problems in our evaluation.

The accuracy of the DHS placement database is important for several reasons. First, it is the database that DHS intends to use to monitor some key measures of the child welfare system’s performance--including the number of placements, their length, and cumulative time in placement. At a time when legislators are trying to reduce the amount of time that children spend in placement prior to permanency decisions, it is important to have reliable benchmark data on the amount of time spent in placement. Second, DHS placement data have been used to help allocate state ‘‘family preservation aid’’ to each county since 1995. 15 Finally, the DHS placement information system is the only statewide database that can be used to analyze placement trends by individual counties.

Some of the problems we found might be corrected in future years through implementation of the department’s new Social Services Information System. We think the department should carefully monitor the accuracy of information in this system, and we also recommend:

14 DHS staff told us that it is not always possible to collect social security numbers in cases involving emergency placements or cases in which parents or caretakers are not cooperative. Still, DHS staff told us that it should be possible to obtain social security numbers for a large majority of cases.

15 By law, one-half of the aid is paid to each county based on its proportion of the state’s total number of children in out-of-home placement. in 1991 and 1992. See Minn. Stat. §477A. 0121.

For example, we think that some of the problems with existing data could be addressed if DHS asks counties in 1999 to provide updated data spanning the last several years of placements, rather than data for just the most recent six months.

We also recommend that:

DHS should add codes to its placement database that identify the specific facilities in which children have been placed.

Among other reasons, having information on individual facilities will enable DHS to detect cases in which a continuous placement at a facility is reported by counties as more than one placement. We found (and adjusted the data for) many instances in which this occurred.

OTHER Program Completion Rates

We recognize that the extent to which juveniles complete their residential programs is affected by the characteristics of the juveniles, the length of the program, the actions of the placing counties and courts, and other factors. We also recognize that program completion rates are not direct measures of the effectiveness of programs. Nevertheless, we think that it is reasonable for courts and counties to obtain information about program completion rates at the time they select a program for a child placement. We recommend that:

Cultural Sensitivity

In Chapter 3, we reported that patterns of child placement differ markedly by racial and ethnic groups. Of particular note, there are very high rates of child placement among African Americans and American Indians. We do not fully understand the reasons for these patterns, but they might include poverty rates, levels of family dysfunction, prejudice, or other factors.

Regardless of the reasons for this pattern, many county officials told us that there is room for improvement in ensuring that residential programs are sufficiently sensitive to cultural differences among residents. If disproportionate numbers of children of color are in placement, it makes sense for service providers to make special efforts to understand their needs and communicate effectively with them. State law already establishes a goal of ‘‘culturally competent’’ social services and requires that social services agencies ensure that ‘‘culturally appropriate services’’

State agencies should help identify ways to provide culturally appropriate services.

have been offered to eliminate the need for placement. 16 Proposed rules drafted by the departments of Human Services and Corrections in 1998 would require facility license holders to document that they are providing culturally appropriate care. In addition, state agencies have developed and funded staff training programs in cultural competence. These are good first steps, but defining and implementing culturally appropriate services remains a challenge. We recommend:

The departments of Human Services and Corrections should identify a set of ‘‘best practices’’ for facility and county staff--to help them provide culturally appropriate screening, assessment, case management, and direct services.

16 Minn. Stat. §256F. 01, §260.012.