U.S. Department of Health and Human Services

Foster Care for Children and Adults with Handicaps: Child Welfare and Adult Social Services

Bradley K. Hill, K. Charlie Lakin, Angela R. Novak and Carolyn C. White

University of Minnesota, Center for Residential and Community Services

October 1987

PDF Version: http://aspe.hhs.gov/daltcp/reports/fosteres.pdf (70 PDF pages)

This report was prepared under grant #90DD0051/2 between the U.S. Department of Health and Human Services, Office of Social Services Policy (now the Office of Disability, Aging and Long-Term Care Policy) and the University of Minnesota. Additional funds were provided by the HHS Administration on Developmental Disabilities. For additional information about the study, you may visit the DALTCP home page at http://aspe.hhs.gov/daltcp/home.shtml or contact the office at HHS/ASPE/DALTCP, Room 424E, H.H. Humphrey Building, 200 Independence Avenue, SW, Washington, DC 20201. The e-mail address is: webmaster.DALTCP@hhs.gov. The ASPE Project Officer was Margaret Porter.

This research was funded by the Office of the Assistant Secretary for Planning and Evaluation (ASPE) as a supplement to the National Study of Specialized Foster Care and Small Group Care Models of Long-Term Care for Persons with Developmental Disabilities, funded by the Administration on Developmental Disabilities (ADD) (Grant #90DD0051/2). Contractors undertaking such projects under government sponsorship are encouraged to express freely their professional judgment. Points of view or opinions stated in this report do not, therefore, necessarily represent the official position of either ASPE or ADD.

TABLE OF CONTENTS

ABSTRACT

I. INTRODUCTION

Background

Purpose

Previous Research

II. METHODOLOGY

Survey Design

Definitions and Limitations

III. FINDINGS

Children's Foster Care

Adult Foster Care

Specialized Foster Care

State-County Population Estimate

IV. DISCUSSION AND RECOMMENDATIONS

REFERENCES

NOTES

APPENDICIES

APPENDIX A. Definitions

APPENDIX B. Type of Handicap by Type of Foster Placement

LIST OF TABLES

TABLE 1. Total Children in Foster Care by Type of Placement: 12/31/85

TABLE 2. Foster Children per 100,000 State Population Age 0 Through 17 in 1985

TABLE 3. Handicapped Children in Foster Care by Type of Placement: 12/31/85

TABLE 4. Handicapped Children in Foster Care by Type of Handicap: 12/31/85

TABLE 5. Approximate Number of Children in Substitute Care: 1961-1985

TABLE 6. Children in Foster Care: 1980 and 1985

TABLE 7. Children's Foster Care: Management Information Systems

TABLE 8. Children's Foster Care MIS: Types of Handicap Recorded

TABLE 9. Children's Foster Care MIS: Types of Placement

TABLE 10. Children's Foster Care: Policies and Practices

TABLE 11. Adult Foster Care Programs Reported by State Social Services Offices

TABLE 12. Number of Adults in Social Services Foster Homes

TABLE 13. Adult Foster Care: Management Information Systems

TABLE 14. Adult Foster Care: Interagency Coordination

TABLE 15. Characteristics of Specialized Foster Care Homes and Group Homes: 6/30/82

TABLE 16. Foster Children and Adults Reported by Selected Counties

TABLE 17. Foster Child Placement Rates by Size of County

TABLE B-1. Mentally Retarded Children in Foster Care by Type of Placement

TABLE B-2. Emotionally Disturbed/Mentally Ill Children by Type fo Placement

TABLE B-3. Specific Learning Disabled Foster Children by Type of Placement

TABLE B-4. Hearing, Sight, or Speech Impaired Foster Children by Type of Placement

TABLE B-5. Foster Children with Physical or Health Handicaps by Type of Placement

TABLE B-6. Other/Multiple/Special Need/Uncategorized Handicap by Type of Foster Placement

Additional copies of this report may be obtained postage-paid for $10.00 per COPY payable to the University of Minnesota.

The recommended citation for this report is: Hill, B.K., Lakin, K.C., Novak, A.R., & White, C.C. (1987). Foster care for children and adults with handicaps: Child welfare and adult social services (Report No. 23). Minneapolis: University of Minnesota, Department of Educational Psychology.

The University of Minnesota is committed to the policy that all persons shall have equal access to its programs, facilities, and employment without regard to race, creed, color, sex, national origin, or handicap.

ABSTRACT

This report summarizes state-by-state data on the number of children and adults in foster homes administered by child welfare or social service agencies. As of December, 1985, there were approximately 261,000 children in out-of-home foster care, including 54,000 handicapped children, of whom 14,000 were mentally retarded. The total number of children in foster care on any one day has decreased only slightly since 1980, whereas the number of children reported to be mentally retarded decreased by approximately 7,500 (35%) to 5.3% of all foster children. Specialized foster homes administered by state mental retardation agencies serve approximately 6,400 additional mentally retarded children and 10,700 mentally retarded adults.

Adult social services foster care programs and data collection efforts regarding children and adults in foster care are also discussed. Many states gather and can report detailed data on both handicapped and nonhandicapped foster children. In other states, only counties can provide data on handicapped children. Data collection efforts would benefit form standardization of terminology used by states.

I. INTRODUCTION

Background

When natural families, for various reasons, cannot or will not provide adequate care for dependent family members, whether for reasons of dysfunction of the family unit, disability or delinquency of the family member, or other reasons, public agencies often assume the role of ensuring that the dependent individual will receive appropriate food, shelter, clothing, and nurturance. A number of specialized institutions have evolved over the decades to implement this public commitment. These include a variety of residential and nursing institutions, group homes, halfway houses and other facilities for persons with dependencies due to age or disability. A primary and extremely important setting for fulfilling the public commitment to dependent persons living outside their natural homes is foster care.

The most common use of foster care is as a setting in which children's basic needs for food, shelter, clothing, and nurturance can be met at times when their natural families are unable to do so adequately. Public Law 96-272, the Adoption Assistance and Child Welfare Act of 1980, partly in response to a growing caution about the use of long term or repeated out-of-home foster care placements, established the concept of permanency planning for children as a national policy. A primary purpose of this Act was to promote a reduction in the number of children in foster care, reduce the duration of care, and improve preventative and family-based support services and case management to promote more stability in the lives of foster children. While it is difficult to obtain statistics that demonstrate the effects of the permanency planning effort, available statistics (discussed later in this report) suggest that although placement duration may have decreased somewhat, the number of children and youth in foster care at any one time did not change appreciably between 1980 and 1985.

Foster care is also becoming increasingly important as a long-term placement for persons with disabilities. Although the foster care model of residential services to persons with disabilities gained considerable attention in the first third of this century, interest waned from the early 1930s until the late 1960s. Since then, as greater stress has been placed on community integration and normalized lifestyles for persons with mental, physical and/or sensory impairments, there has been increasing attention on maintaining persons with disabilities in their natural homes or in the most home-like long-term care placement possible. Obviously in pursuit of this goal, foster care arrangements are particularly attractive.

To examine the role and potential of foster care in responding to the long-term care needs of persons with disabilities, it is important to observe the administrative structure through which long-term care services are provided. The present study emphasizes services to persons with mental retardation/developmental disabilities, but analogies exist in services to other "disability groups" with long-term care needs. For persons with mental retardation/developmental disabilities, the largest long-term care service system is operated, licensed, administered and funded by state departments or divisions of mental health or mental retardation and is primarily comprised of state institutions, large private institutions, relatively small group homes, and specialized foster homes. The state mental retardation residential service system serves approximately 250,000 persons, about 25% of whom are less than 22 years old. Foster homes specially licensed by state mental retardation agencies (referred to in this report as "specialized foster homes") served 6,400 children and youth (0-21 years) and 10,700 adults in 1982, an increase of almost 20% from 5 years earlier (while the total residential service system remained about the same size).

The second largest system of long-term care for persons with developmental disabilities is made up of nursing homes and personal care homes, generally operated under the supervision of state departments of health. The most recent available data on placements in such facilities are provided by the 1977 National Nursing Home Survey (NCHS, 1979). Its estimates indicated that among the 1,303,000 people in nursing homes in 1977 there were about 44,000 persons with a primary diagnosis of mental retardation or a diagnosis of epilepsy with mental retardation as additional handicap.

The third system of long-term care, and the focus of the present study, is operated by state social service agencies. This substitute care is comprised of foster homes, group homes, and other types of residential facilities. Typically, this system is operated through county departments of social services, although some states administer foster care through regional offices. In this report, residential services provided by child welfare or adult social service systems will be referred to as generic foster care to differentiate them from the specialized foster care programs administered by state mental retardation, mental health, or other agencies focused on specific disability groups. According to an Office for Civil Rights Study (OCR, 1981) which will be discussed later in this report, about 267,000 children and youth, including 64,000 children and youth with handicaps, 21,000 of whom were mentally retarded, were in generic substitute care programs in 1980 (includes age birth to 17 years; excludes independent living).

Foster care is generally considered to be a child welfare service, but in some states and counties adult foster care programs have also been established. Nearly two thirds of residents in specialized foster care programs administered by state mental retardation agencies a re adults. Many states have also established roster home programs for adults with mental illness and/or those who are elderly and/or disabled, although data on the number of persons mentally ill and elderly participating in foster care programs do not appear to be available. For the most part, both generic and specialized foster adult care programs for persons with disabilities rely on funding assistance available to participants through the Social Security Act (Supplemental Security Income, Social Security Disability Insurance, Medical Assistance), often supplemented by state programs. Under the Medicaid Home and Community-Based Services waiver, Title XIX funds are also being used to support the care, supervision and training provided to adults living in foster homes. Certainly cost is of major importance as one looks to the future of foster care programs for persons with disabilities. In a recent (1986) sample of specialized foster care homes and small group homes (6 and fewer residents) for people with developmental disabilities, we found that foster care homes had an average daily cost of $23.03 versus $48.60 for small, licensed (but not ICF-MR certified) group homes.

Numerous factors suggest greater utilization of foster care arrangements to meet the long-term care needs of people with disabilities in the future. Among the advantages of foster care are the more normal patterns of daily living they provide residents, their integration within the community and, of course, their low cost. Unfortunately, while data are increasingly available on the nature, size, and quality of specialized foster care programs, even the most basic statistics on generic foster care services are difficult to obtain. Among the most obvious reasons for this lack of comprehensive data is that generic foster care programs are operated by approximately 2,500 separate jurisdictions.

Purpose

This study was developed as a result of conversations held with staff of the Office of the Assistant Secretary for Planning and Evaluation (ASPE) in 1985. At that time the Center for Residential and Community Services (CRCS) was reporting data from its 1982 national census study of residential services for persons with mental retardation. CRCS had also just been funded by the Administration on Developmental Disabilities to conduct a more detailed comparative study of representative samples of specially licensed foster care homes and small group care facilities (6 or fewer residents). The statistics on specialized foster care in the 1992 national census study, and the effort that CRCS was undertaking to expand its basic survey data with a comprehensive data set, brought indirect attention to the lack of similar data on the use of generic foster care for children and adults with developmental disabilities. At about the same time ASPE personnel were involved in a set of activities related to constructing a national foster care data base. Because of the complementary, if not overlapping roster care systems, a meeting was held with CRCS and ASPE staff members to discuss the use of these two models of foster care, particularly as they related to children and youth with disabilities. Within these discussions questions were raised about the extent to which generic foster care was being used to provide residential services to persons with mental retardation, the extent to which the generic and specialized systems served similar subpopulations of persons with mental retardation, and the reasons for state variations in utilization of specialized and generic foster care models. These discussions focused on the availability of CRCS data on specialized foster care, the limited availability of data on generic foster care, and on related statistics regarding changing patterns of residential care for children and youth with handicaps. CRCS data, for example, revealed that the number of children and youth (0-21 years) in mental retardation facilities decreased dramatically between 1977 and 1982, from about 91,000 to 60,000. While CRCS staff argued that this change was probably attributable to the increased availability of public school programs and community-based family support services for children and youth with handicaps, it was conceded that some portion of the change could have resulted from children, who at one time would have been placed in mental retardation facilities, entering child welfare (generic) foster care programs instead. While these discussions raised many questions of importance to policy evaluation and future direction, it was evident that there was relatively little information readily available with which to respond to them. Discussions therefore turned to how a modest survey effort could be carried out to assess the availability of basic generic foster care data and what kinds of procedures would be required to gather them from state and local agencies from which they might be available.

To carry out an exploratory study of the generic foster care system and to gather statistics from states and selected counties, ASPE provided modest supplemental funding to the CRCS study of specially licensed foster care and small group homes. In addition to reporting on state systems with respect to persons with mental retardation and other developmental disabilities, ASPE staff members requested that information and available statistics also be gathered on children and youth with other handicaps in generic foster care, on nonhandicapped children and youth in generic foster care, and on adults with handicaps in generic foster care.

The Center began gathering background data from states in late 1985, gathered data on the number of children in homes as of December 31, 1985 during the first half or 1986, and continued with follow-ups to states throughout 1986. Among the specific purposes of the study were the following: (1) to determine the different types of information gathered and aggregated by different states regarding persons with mental retardation, persons with other handicapping conditions, and persons without handicaps in generic foster care programs; (2) to gather information available from states on the nature of the handicapping conditions of individuals living in the generic foster care homes; (3) to determine changes in the number of total children and handicapped children in the generic foster care nationally (through comparison to previous surveys); and (4) to the extent that desired information was not readily obtainable, to suggest methods by which such data might be gathered in the future.

Previous Research

Since 1960 there have been several national surveys of children in generic foster care. Notably each employed a somewhat different methodology, used somewhat different inclusion criteria, and was operationalized with somewhat different definitions. These studies are briefly described in the following paragraphs.

Jeter, 1961. Data in this 1961 survey of social services (Jeter, 1963) were based on a national sample of 49,838 case records (out of 377,000 children) receiving public social services and 12,368 (of an estimated 125,000) children receiving private social services. Of the total of 502,000 children (all but 700 were age 0-20), an estimated 233,440 were in foster care, excluding placements with parents or relatives (collected, but not included here). This estimate included both public and private (voluntary) social service agencies. Seven percent of children receiving services were mentally retarded. Of all children in foster care, 6.7% were reported to be mentally retarded.

Westat, 1977. The Westat study (Shyne & Schroeder, 1978), based on a national probability sample of 315 local agencies, included 9,597 case records of children less than 18 years in substitute care on a single day during the first quarter of calendar year 1977. This study included children in public or private care for whom the primary state child welfare agency had case management responsibility. Of 1,800,000 children receiving services, 503,000 were in substitute care, excluding those with parents, relatives, emergency shelter, or detention centers (reported separately). Mental retardation was reported for 4.7% of children, of whom between 28% (the proportion for all children) and 67% (the proportion for children with mental retardation as primary problem) were in foster care.

OCR, 1980. The Office for Civil Rights study (OCR, 1981) was a national census that included 2,436 of 2,439 local child welfare agencies (mostly counties) in the U.S.¹ The survey gathered data on all children in the legal custody of child welfare agencies as of January 7, 1980, including the number of children by race, handicap, and type of placement. Children in various types of placements made by or under the supervision of each local child welfare agency numbered 301,943, including "independent living," defined by OCR to include placement with relatives; 267,000 excluding independent living. Of children and youth with mental retardation on agency caseloads, 59% were in foster care. Of all children in foster care, 24% were handicapped; 8% were mentally retarded.

Maximus, 1982. The Maximus, Inc. study (1983) was a national probability survey of 167 child welfare agencies. Data were extracted from 4,225 case records of children less than 18 years old who were in substitute care on December 15, 1982. Of the estimated approximately 202,000 children in foster care, (not counting placement with parents, relatives, final adoptive parents, or independent living which were reported separately) 57,300 were estimated to be handicapped.

For reporting handicapping conditions, defined in a case-specific questionnaire as "physical, psychological, or mental handicap," some agencies may have used their own "agency criteria" which varied widely and in some states included broad "special needs" categories. Over 60 percent of handicapped children were over II years old. Of the children in substitute care who were handicapped, over 60 percent had been in care for more than two years, as compared with only 38 percent of nonhandicapped children. Almost 30 percent of all children with handicaps were living in either group homes or institutions, while only 13 percent of children without handicaps lived in group homes or institutions. There was virtually no difference in the racial distribution between handicapped and non-handicapped children.

VCIS, 1982 and 1983. In 1982, the American Public Welfare Association implemented the Voluntary Cooperative Information Systems (VCIS) to collect annual information about children less than 21 years old in substitute and adoptive care. For fiscal years 1982 and 1983, respectively, primary child welfare agencies in 48 and all 50 states and DC responded with aggregated information on out-of-home placements. although not all states responded to all items. The VCIS reports indicated that 21.1% of the children in fiscal year 1982 and 22.3% in 1983 were reported to have one or more disabling conditions (Tatara & Pettiford, 1983,1985). Types of disabling conditions and the types of placements in which handicapped children lived were not reported.

In the studies outlined above, as well as in the present study, there have been problems in defining "foster care" (e.g., should children in detention homes be included; should children placed with relatives be included), in defining types of foster care placement (e.g., foster family home, foster group home), and in defining types of handicaps (e.g., what is a "special need"). While efforts were made in the present study to use definitions that permit comparison to earlier data, and indeed even contemporaneous comparison among states, some incongruencies remain. Even when decisions are made with respect to what should be considered to be foster care, it becomes readily apparent that some states have difficulty providing estimates of the total number of children in various types of substitute care, much less the number of children with handicaps. In summary, then, the survey attempted to identify and structure as best it could the contemporary organization and status of data systems on children and youth in foster care, including children and youth with various handicaps. It encountered many problems in attempting to rectify state data systems and their various data elements and operational definitions. Despite these problems, existing state data bases have quality with contemporary utility and which, with modest modifications, could provide even more useful statistics to inform foster care policy. This report, in addition to tabulations of the statistics gathered as part of the survey, discusses areas in which promise exists for enhanced, policy-relevant foster care data collection without undertaking major agency surveys such as those identified earlier in this report.

II. METHODOLOGY

A series of surveys were carried out in order to understand state policies on generic foster care use, to identify and gather current statistics available from states, and to examine alternatives for cost effective, ongoing collection of basic foster care data. These included a telephone interview and three questionnaire surveys directed to various state agencies, and a questionnaire survey of selected county social service agencies. These are described below.

Survey Design

State agency surveys. In the fall of 1985, a preliminary telephone survey of all 50 states and the District of Columbia was conducted by Project interviewers prior to mailing the state agency questionnaires. The purpose of the telephone contacts was to identify knowledgeable individuals in each of three state agencies (child welfare, adult services and mental retardation/developmental disabilities) from general agency listings, and to determine through these individuals the types of foster care information available from each state agency and the most knowledgeable "key contact(s)" in each agency regarding policy and population statistics. This preliminary survey was essential to the development of a standardized set of questions and definitions for gathering data across states.

In the spring of 1986, three questionnaires were mailed to each of the 50 states and the District of Columbia. One questionnaire was mailed to each state's primary child welfare agency. It requested information on the number of children and youth in foster care by type of residential settings and by type of handicap. It also included questions about the number and type of data elements contained in the state's substitute care management information system (MIS), as well as questions regarding statewide foster care policies, interagency cooperation/coordination issues, and differences between social services and mental retardation/developmental disabilities agencies regarding foster care practices.

A second questionnaire was mailed to key contact persons in adult services offices of the state social services agency. It asked questions about the use of adult foster care in the state, including the number of adults in foster care, whether programs were generic or whether they were operated and licensed for specific target populations. Other questions were asked about data elements in the agency's MIS and interagency co-involvements and cooperation in adult foster care.

A third questionnaire, mailed to key contact persons in state mental retardation/developmental disabilities agencies, paralleled those sent to social services agencies. It asked about specialized foster care for children and adults with mental retardation/developmental disabilities, about information maintained on recipients of specialized foster care, and about interagency cooperation and program coordination.

County surveys. Based on previous national foster care studies, described above, it was anticipated that much of the information desired in this survey would not be obtainable on the state level. Because the focus of this study was not only on gathering available data, but also on reporting the status of data collection by public agencies, a second component was designed to identify and gather data available from selected counties. The purpose of the county survey was to determine the extent to which information not available on the state level could be gathered from counties and to suggest sampling strategies for gathering county data. To do this three counties were sampled in each of 10 states. Two states were selected from each of the four census regions of the country. These states were selected as being generally representative in terms of size and population. In addition, California and New York were selected in order to reflect as large a proportion of the total foster care population as possible.

Three counties were selected in each of these ten states. For the purposes of selection, counties were rank-ordered by total population and divided into three groups--large (the five most populous counties in the state), medium-size (at least 75,000 people) and small (under 75,000 people). The second largest county in the "large" group, the middle-ranked county in the "middle" group, and the middle-ranked county of the "small" group were selected as the three representative counties. State child welfare contact people were asked to confirm that selected counties were not unrepresentative of their category of counties either demographically or in terms of their child welfare system. In the rare instance where the state key contact indicated that the county was atypical, the next lower ranking county was selected to replace it.

The survey sent to county informants asked for the same basic information as the state agency children's foster care survey, with minor adaptations to reflect the county focus. The county survey also asked about the existence of both generic and specialized adult foster care in the county, and if there was generic adult foster care, how many persons, with which types of disabilities, were in such care.

Definitions and Limitations

A major difficulty in gathering and comparing aggregated statistics on foster care derives from variations in terminology. The three areas where these are most notable are type of residence, age, and type of handicap. Examples of the problems encountered in these areas are identified below.

Type of placement. In the present study, substitute care was defined to include the following categories of residences for children and youth in foster care: family foster home, including placement with relatives who are reimbursed and pre-finalized adoptive foster homes; group home (20 or fewer residents); residential treatment/institution (21+ residents); emergency shelter; secure facility; independent living; placement with family or relatives (not considered foster care in this study); and other. (Definitions of these residential categories are included in Appendix A.).

There is considerable ambiguity in the "foster care" status of many children and youth. In this study the following distinctions were made in operationally defining roster care status. Children placed with relatives who were not licensed or reimbursed were not considered to be in foster care. Children placed with relatives who were licensed and/or reimbursed were considered to be in foster care. Children returned to their natural families, regardless of continued monitoring by a social services agency, were not considered to be in foster care. Children in secure facilities (reformatory, detention center, locked group home) were counted if they were under the case management responsibility of a state's child welfare agency, even if the placement was made by or coordinated through a Department of Corrections. Children in private agency foster placement, but for whom the state social service agency had supervisory or review status were also included. Some states include some adopted children in their foster care data bases, but finalized adoptions were excluded from the present study.

It is important to recognize that although it is possible and, indeed, important to make semantic distinctions in defining various types of foster care placements, states and counties generally operate with their own operational categories. While efforts were made to produce reports of agency data with maximum congruence to the standard definitions developed for this survey, such efforts obviously have had some degree of error in estimation. Generally issues relating to classification of substitute care facilities more often involve the problem of defining "foster care" than the problem of defining type of foster placement.

Age. A "foster child" may be 0-17 years old in some states, but 0-20 years old in others. In this report, foster children are defined as age 17 or younger, or children age less than 21 who entered foster care before the age of 18 and remain in school full time. While some reporting problems would seem likely in the 18-20 year range, the actual number of foster children who are 18-20 years old is small (estimated to be 3-5% of all foster children) and problems incurred by states in reporting according to our request probably had little impact on the overall findings.

Handicap. The two major problems in gathering information about handicaps among foster children relate to (1) definition, and (2) reporting practices. Some states use a designation called "special needs" which may include individual disabilities, but which in some states also includes special placement needs, such as being placed with a sibling, or being placed in a certain geographic area. Special needs are not therefore always the equivalent of handicaps. Even where states employ indicators of disabilities, problems are evident. Some states use functional descriptors such as physical, emotional, or mental handicap, others use standard diagnostic descriptors with more or less standard definitions (e.g., mental retardation, visual impairment), and others use very general indicators (e.g., "learning problem" or "psychological handicap") which lack objective definitions and which make it impossible to differentiate handicaps such as mental retardation, specific learning disabilities, or emotional disturbance. While the ideal criterion for being considered handicapped in this study might have been formal clinical determination of disability by a qualified diagnostician, clinical as well as functional descriptions were frequently reported by respondents to be informally applied by case workers, foster parents, or natural parents.

Despite recognition of variability among agencies in the methods and meanings in their categories and definitions of handicaps, in this study a basic set of diagnostic descriptors of disability was used. It included mentally retarded; seriously emotionally disturbed/mentally ill, specific learning disabled; hearing, speech or sight impaired; physically or health handicapped; and other, unclassified or multiple handicaps. The definitions employed for these handicaps corresponded to those used in the Office for Civil Rights 1980 Children and Youth Referral Survey (OCR, 1981), except that physical/health handicaps were combined with "other" in the OCR study. The definitions provided to respondents for each category are provided in Appendix A. Throughout this study efforts were made to establish congruence between the definitions within state data systems and those of this study. If no congruence could be established, handicaps were considered by be "non-classified." "Special needs" were not considered to be handicaps unless the state's definition was restricted to handicaps. Hearing, speech, or vision problems were considered to be handicaps if they were defined within MIS systems (or judged based on the MIS definitions) to be serious enough to adversely affect educational or vocational performance.

As will be noted in this report, data on foster children with handicaps are not universally available. Not all state management information systems provide for the coding of handicaps. Some states' data gathering forms have a place to code handicap for each child as a demographic descriptor, but allow optional use of the code. Other states include handicap not as a client descriptor but as one of a list of possible "reasons for placement" (permitting several reasons to be coded). Reports of number of individuals placed "because of" handicap would probably significantly undercount children and youth with handicaps in foster care because of the significant number of children with handicaps for whom placement is not primarily related to an identified disability. But even in cases where a child's disability may play a major role in the placement, the coded reason for placement (e.g., parent cannot care for child) may make it impossible to determine the role of the handicap in the placement decision. The importance of caution in assuming that the "presence" of a handicapped condition implies that condition as the cause of placement can be seen in the 1983 Voluntary Cooperative Information System (Tatara & Pettiford, 1985) which reported that among reporting states 22% of children in foster care had a disabling condition, but that only 2.6% of all foster care children were reported to have been placed in foster care because of a disabling condition.

III. FINDINGS

Children's Foster Care

The ability of states to report number of foster children by type of foster placement, number of handicapped children, or types of handicaps varies considerably. At the most elementary level all states were able to report the total number of foster care children as of approximately 12-31-85. This section reports on numbers of foster children and on the number of foster children with handicaps. It also compares December 1985 data gathered in this study with data from previous studies, and discusses statewide management information systems that are available to provide aggregations of foster care data. It concludes with a discussion about interagency cooperation in the out-of-home care of children and youth with handicaps.

Number of Children in Foster Care

All fifty states and the District of Columbia were asked to report their number of children in foster care by type of placement and by type of handicap on or as close as possible to December 31, 1985. Table 1 shows that states reported a total of 297,069 children and youth age 0 through 20 in foster care, if independent living and placement in their own family homes were included; 261,314 if they were not. Independent living and placement with family or relatives (as defined in Appendix A) are included in tables of this report that summarize specific placement types, but unless otherwise noted, are excluded from accompanying tables. Placement with relatives are counted as roster care only if the relatives were licensed or reimbursed.

Not all states were able to provide data by type of placement, but in states that could, and excluding independent living and placement with family, approximately 77% of children in substitute care were in family foster homes, 10% in group homes (20 or fewer residents) and 13% in large residential facilities (21 or more residents). Only eight states were unable to report the number of their total substitute care population in foster family care as a specific placement type.

In Table 2 foster care placements are indexed by state population. Nationally in 1985, there were 411 children in foster care per 100,000 children age 0 through 17 years old, with a range of from 96 per 100,000 in Texas to 1,603 per 100,000 in the District of Columbia. While notable "outliers" were identified, 42 of the 51 states fell between 250 and 650 foster care placements per 100,000 children and youth.

Table 2 also reports the proportion of all children in 1980 who were not living with either of their parents -- 4.8 percent of all children -- approximately 12 times as large as the number of children in licensed foster care in 1985. There is not a strong relationship between the number of children not living with parents and the number in foster care. For example, although the District of Columbia has both the largest proportion of children not living with parents and the largest rate per 100,000 children in foster care, Mississippi and several other states show the opposite relationship.

More detailed Bureau of Census data are available on a national (not state-by-state) basis (Bureau of Census, PL 80-2-4B). Excluding children who were themselves heads of households, 4.5% did not live with either parent in 1980. This includes 3.0% for whom a relative was head of household, 1.1% (709,682) who were not related to the household head, 0.3% (167,306) in institutions, and 0.1% (93,119) in other group quarters. The Census Bureau defines foster children as "non-relatives (under 18 years old) of the householder in households with no non-relatives 16 years old and over (who might be parents of the non-relatives under 18)." In 1980, the Census Bureau estimated that of the 709,682 children not related to the head of household, 281,053 (defined by Census as foster children) were in households that could not have also included their parents. This number, 281,053 in 1980, is somewhat larger than the 173,000 children reported by states in 1985 to be in licensed foster homes.

Foster Children with Handicaps

Type of placement. Thirty two states reported statistics on the number of children and youth with handicaps in foster care. These included 25 states that could provide (or reasonably estimate) placements of all children and youth with handicaps by type of substitute care (Table 3) and 30 states that provided numbers of children and youth in substitute care by type of handicap (Table 4). Tables B1 to B6 in Appendix B provide crosstabulations of type of handicap by type of foster placement for those states able to provide such breakdowns.

When compared with the foster care placements of all children, children with handicaps were less likely to be living in family foster care settings. Excluding placements in unclassified residences, in independent living or in parents' or unreimbursed relatives' homes, about 75% of all children and youth and about 64% of handicapped children and youth were in family foster care. Children with handicaps were considerably more likely to be in group residences with 21 or more residents than were nonhandicapped foster children (24% versus 14%, excluding unclassified residences, independent living, and relatives' homes).

Type of Handicap. About two thirds of states were able to provide breakdowns of their populations of children and youth with handicaps into three or more categories of disability, although most diagnoses were reported as "other" or unclassified (18,110 with 12,324 coming from New York and California). Among the approximately 25,000 children and youth with handicaps in states that reported type of handicap, approximately 8,400 (33%) were categorized as mentally retarded and 9,600 (38%) as emotionally disturbed. Perhaps more notable than the number of individuals in the different categories was the wide variability among states. This variability appears due to two primary factors. First, differences in classification procedures clearly exist. States differ in whom they report in categories such as emotional disturbance and specific learning disabilities which in turn accounts for significant differences in prevalence among states. Second, the number of children and youth reported in a specific disability category was clearly affected by the state-county relationship in out-of-home residential placements. For example, Minnesota, in which the state mental retardation system has been exclusively committed to starred group residences, relies heavily on child welfare services to identify and make foster care placements for children and youth. Consequently Minnesota has an extremely high number of "generic" foster care placements of mentally retarded children. Michigan on the other hand has a regional mental retardation program that includes considerable use of specialized family foster care arrangements for children as well as adults with mental retardation. Because of the development or its specialized foster care program, Michigan has relatively low utilization of "generic" foster care for children and youth with mental retardation.

Longitudinal Trends in Foster Care

Table 5 presents basic longitudinal statistics from the previous national studies of foster care utilization. These are discussed below in terms of the total population of children and youth in foster care, the placements of those children and youth, and the populations of children and youth with handicaps in substitute care.

Total population. As noted earlier in this report there have been several previous research efforts focussed on the number and, in some instances, the characteristics of children and youth in generic foster care settings. These studies, which span the approximately 25 years from 1961 to December 31, 1985 (the present study), have considerable usefulness in longitudinal analyses of changing patterns of foster care utilization. There are, however, limitations to their direct comparison. The first or these limitations is that the studies are based on three substantially different methodologies. Three of the studies (Jeter, 1963; Westat, 1978; & Maximus, 1983) drew nationally representative samples of local agencies and case records. One of the surveys (Office for Civil Rights, 1981) gathered aggregated client data from virtually every county in the U.S. The Voluntary Cooperative Information System (VCIS [Tatara & Pettiford, 1985]) and the present study gathered aggregated data from state agencies. In addition to methodological differences, each study also used slightly different definitions or placement types. Finally, as noted above, there are differences among states in defining foster children. Therefore, some caution must be used in examining the summary of statistics from research on foster care presented in Table 5.

In general it appears that the number and rate of placement of all children and of handicapped children in foster care has been fairly stable over the past quarter century. Two "outlier" studies that challenge this conclusion, Westat and Maximus, were sample studies based on samples of local welfare agencies (Westat sampled 315 of 2,439 local child welfare agencies; Maximus 167 of 2,439). The major challenge to these studies is simply that comprehensive surveys of social services agencies conducted at nearly the same time (OCR and VCIS) indicate very different numbers of children in foster care settings. The Westat estimate of the national population of children and youth in foster care was 503,000. This compares with the Office for Civil Rights survey of 2,436 of the 2,439 local jurisdictions which indicated barely half that number in foster care (266,600). It seems likely that while the Westat local jurisdictions were "representative" with respect to the factors by which they were selected, they were probably not representative with respect to foster care utilization.

The second outlier study was the 1982 Maximus study, also based on a sample of local child welfare agencies. The 201,847 children estimated by Maximus to be in foster care in 1982 is significantly less than the 266,584 reported by the Office for Civil Rights two years earlier and the 251,000 reported for essentially the same time period (the beginning of FY 1983) in the VCIS survey. (Note that unless otherwise specified, data from all studies cited have been adjusted to exclude foster children living at home [with natural parents], independently, or with unpaid relatives.) Therefore, it seems reasonable to assume that Maximus significantly underestimated the number of children and youth in foster care. Based on participating 167 local agencies that reported 17.6% fewer children than reported to OCR two years earlier, Maximus estimated a similar decrease in the 2,269 nonparticipating agencies (with about 70% of all foster children in the OCR survey). It would appear that this procedure underestimated the number of children in foster care. One other variation between VCIS, the present study, and the Maximus survey was the inclusion in the former two of some children age 18, 19, and 20 years old who had been placed in foster care before age 18. The 1983 VCIS survey estimated 7,000 such youth.

We conclude that the number of children in foster care at any one time has been fairly stable since 1961. Substitute care placements of children and youth increased from about 233,400 (362 per 100,000 children birth to 17 years old) in 1961 to 266,584 (419 per 100,000) in 1980. Since 1980 there has been considerable stability in placement rates with state surveys showing 251,000 (400 per 100,000) in October 1982, 254,700 (407 per 100,000) in September 1983, and 261,300 (411 per 100,000) in December 1985.²

The number of handicapped children in foster care requires some estimation because, as noted earlier, not all states record handicaps and those that do often do so under different definitions. What is more, for the sake of longitudinal comparisons, the earliest data on total placements of children with handicaps in foster care is 1980. The Jeter and Westat studies (prior to 1980) did gather information on handicaps, but did not present crosstabulations of handicap by type of service/placement.³ Based on the best data states were able to provide and proportional estimates for missing data, the number of children with handicaps in generic foster care decreased from 63,643 to 53,566 between 1980 and 1985. It appears that much of this decrease was related to decreasing numbers of children and youth with mental retardation in generic foster care, from about 21,400 in 1980 to approximately 13,900 in 1985. While the decrease in the reported number of mentally retarded children and youth in foster care between 1980 and 1985 appears particularly dramatic, the amount of decrease (35%) was virtually identical to the 34% decrease in mentally retarded children and youth in mental retardation residential facilities over the five year period between 1977 and 1982 -- from 91,000 on June 30, 1977 to about 60,000 total on June 30, 1982 (Hill, Lakin & Bruininks, 1984). Some of this decrease may correspond to the fact that during this time the public school system has labelled fewer children as mentally retarded and more as learning disabled. But irrespective of the particular type of handicap ascribed to individuals, the general decrease in the number of all handicapped children in foster care (all diagnoses) must also be attributed to improved efforts to provide support and needed services to handicapped children and youth and their natural families.

State-by-State Changes in Foster Care Placements

For the purposes of establishing a comparative base for examining the data gathered in the present study, the 1980 Office for Civil Rights foster care study was selected. The OCR study included mandatory participation by all local child welfare agencies in the country, it used a uniform set of acceptable definitions after which the present study's definitions were modeled, and it defined special needs as a "clinically diagnosed handicap" (emphasis in original). Also of importance to comparability, OCR presented its data on a state-by-state basis.⁴ Table 6 compares OCR and CRCS data by state, for total number of children, for children who are mentally retarded, and for children with any reported handicap. Although these data are presented for comparative purposes, caution is warranted. The CRCS survey collected its statistics from state agencies, whereas OCR gathered data from local agencies. Indeed as part of the CRCS state child welfare agency survey the 1980 OCR numbers for each state were supplied to respondents who were asked whether they felt the number had increased or decreased by 1985. Several state respondents indicated that the OCR statistics for 1980 seemed too high or too low. Additionally, the data from the CRCS survey included a small but unknown number (probably a few thousand nationally) of youth aged 18 to 20 who were placed before age 18 and who remain in foster care; OCR statistics include age 0-17 only.

In Table 6 estimates for states which were unable to provide 1985 data on mentally retarded and handicapped populations were imputed in proportion to the change from 1980 in reporting states. As Table 6 indicates, the total number or children in all three categories decreased nationally, but there was considerable variation from state to state, with 23 states reporting an increase, and 27 states and DC reporting decreases.

Regarding all children in foster care, between 1980 and 1985 most states (29) showed decreases. Most noticeable among states with decreasing foster care populations was New York, where the number of foster children decreased by 10,000 over the 5 year period. Net decreases in a number of states since 1980 were largely counterbalanced by California, which reported an increase of nearly 20,000 children in substitute care. State and federal officials expressed confidence in the accuracy of California's data, but are unsure of the reason for the increase. Both California and Georgia, another state reporting an increase, speculated that greater attention to and reporting of child abuse, and, (in California) a backlog in the court system regarding parent custody and adoption proceedings, may account for some of the increases.

The number of handicapped children and mentally retarded children also decreased between 1980 and 1995. Including estimates of children and youth with mental retardation in states unable to provide actual figures, 5.4% of all roster children were estimated to be mentally retarded. Of the 33 states actually reporting mentally retarded populations, 5.9% of all foster children were reported to be mentally retarded. Variations among states were substantial, ranging from 17.7% in Minnesota to less than 1% (0.8) in Florida. As noted earlier in another context, such variations are often administrative in nature. Minnesota until recently had no specialized foster care (i.e., licensed or funded by its Division of Mental Retardation) so that all children with retardation were included in the current study. Florida, on the other hand, has developed a carefully coordinated state program of specialized foster homes where nearly all foster children with mental retardation are placed; therefore less than 1% of children in child welfare foster homes in Florida are mentally retarded. Statistics reported by states on children with all types of handicaps varied considerably in both the types of handicaps included and the populations of children in foster care with handicaps. The national average prevalence of handicaps among foster children was 20.5%, with a range of from about 3% in Florida to 58% in Nevada. Eleven of 35 states supplying data indicated that 33% or more of the children and youth in foster care were handicapped. Ten states indicated less than 15% of the children in foster care were handicapped. Generally then, the prevalence of handicaps among children and youth in foster care is estimated to be about double that reported among the school age population as a whole.

Statewide Management Information Systems for Children

Based on in this survey, it seems that data on foster care populations can be systematically gathered from states at relatively low cost. However, in making comparisons between reports of these populations from states and those from counties as reflected in the OCR data, questions of reliability of are obviously raised. Certainly the fact that several states surveyed directly questioned the OCR aggregation of local social service agency reports raises further questions. In Kansas in 1980, for example, counties reported that 3,394 of 3,941 children in foster care (86%) were handicapped. Most counties reported that all foster children were handicapped. In 1985, the Kansas computerized Child Tracking System indicated that only 44% of foster children were handicapped, a more plausible figure because it is unlikely that there were only 547 nonhandicapped Kansas foster children in 1980.

Although at the present time there is insufficient standardization of definitions of handicaps and of placement types, the present survey has suggested potential for recurring aggregation and comparison of state data. Special analyses of data on handicapped foster children may already be possible from MIS data elements currently maintained by over half the states. In a few more states additional handicap categories could adapt the existing system for such a use. Given the current data collected in about 20 states it would be necessary to gather information on handicapped children directly from counties. This section examines state foster care data reporting abilities and also looks at alternatives for improving or supplementing the existing potential of states.

States differ tremendously in the capabilities of the systems they have developed to meet the data collection requirements of P.L. 96-272. At one extreme, one state (Idaho) accomplishes such reporting by monthly regional reports, manually tabulated. Three states use their foster care payment system to monitor the numbers of children in foster care. At the opposite extreme, there are states that have comprehensive on-line computerized client tracking and billing/reimbursement systems through which all counties and/or state regional offices operate.

Table 7 presents summary information on the status of statewide management information systems (MIS) for children and youth in foster care. Although 43 of the 51 states (including the District of Columbia) reported a computerized information system for tracking children and 43 of 48 states reporting the periodicity of case updates indicated they occurred at the time of program changes, the types of information contained in state MIS's vary tremendously. Two of the most important information components for the purpose of this study regarded the way in which handicaps were recorded.

Handicaps. Following preliminary interviews which identified the general manner in which states categorize handicaps, each state was surveyed to gather specific information on whether its MIS treated handicap as a simple "yes/no" for all children, as a diagnostic item with subcategorizations, or as a reason for placement, and whether this information was expected to be universally reported or was optional. Eight of the 47 states for whom MIS characteristics for recording handicaps are summarized in Table 7 reported that they gather no data on handicapping conditions. Twenty-eight states record handicap as a standard demographic item, (i.e., type of handicap, if any, is a data element collected on all foster children regardless of reason for placement). Fourteen of these 28 states indicated that when handicap is recorded, the specific diagnosis is included. Three states indicated that recording of handicapping condition is optional, or required only in certain circumstances. New Jersey, for instance, indicates that recording of handicaps is optional except when the information is needed because of a specific funding source or special program.

Eight states reported that "handicap" or type of handicap is recorded as being among one or more reasons for placement/service, or as relevant to the type of service needed; however in only four of these states would this be the only circumstance under which handicaps were recorded. Recording handicap only as a "reason for placement" underestimates the incidence of handicaps among children in foster care because although a child may be handicapped, the actual reason for placement in foster care might be coded as "abuse," "neglect," "family dysfunction," etc.

A third way in which a few states allude to handicaps is to code broad categories such as "special needs," or "learning problem," which again may be treated either as a demographic item or as a reason for placement. These data are extremely difficult to interpret because there are no standard definitions regarding what types of handicaps, other conditions or life circumstances constitute a special need or learning problem.

Table 8 reports the type of handicapping conditions coded in the MIS of the various states. Twenty-five of the 27 states that record specific categories of handicap include mental retardation or developmental disabilities. Of these 25, 6 distinguish among levels or degrees of retardation. Twenty-one states' MIS forms ask about emotional disturbance or mental illness, 5 about nature or degree. Twenty-five inquire about physical handicaps, 6 about nature or degree. Eight states ask about substance abuse or chemical dependence, 6 about speech impairment, 11 about a vision or hearing impairment, 11 about a neurological disorder or learning disability, 8 about multiple handicaps, and 11 have other categories. The number of types of handicapping conditions, if recorded, varies from one ("special needs") in Georgia to 16 in Wisconsin.

Type of placement. Foster care is a generic term that usually includes at least four basic types of out-of-home care: family foster homes, group homes, group treatment facilities, and child caring institutions. Table 9 lists the types of substitute care that are used in various states' MIS reporting programs. Forty of the 41 states for whom this information was obtained use "family foster home" as a basic type recorded. (Arizona, the lone exception, codes placement only by the facility ID number, not by facility type.) The variety of different codes for different types of placement is large, varying from simply "foster home" to systems that include categories such as foster home with relatives, one-parent foster home, two-parent foster home, licensed foster home, unlicensed foster home, foster home with non-relatives, and others. Thirty-six states code some type of group home, 35 some form of residential treatment or institutional care, and 19 states "child caring institutions." Other frequently occurring types of care that are coded include: emergency or shelter care (28 states); secure facility or detention (16 states); own home/parents' home (16 states); relative's home (26 states); independent or semi-independent living (18 states); nursing home (13 states); facility in another state (6 states); and runaway (13 states). Twenty-seven states report one or more categories other than those listed.

The two placement situations which represent the greatest source incongruity among state data collection efforts appear to be in the reporting of children who are returned to their parents' home but receive post-placement follow-up, and children who live with relatives. Some states include all or some of these children in their foster care information systems, others do not. Generally, however, it was observed in this survey that almost all states that included natural family placements in their MIS programs were able to exclude such placements from their foster care client counts when requested to do so.

Finally, to evaluate the flexibility of the state MIS programs, states were requested to provide crosstabulations of number of children by type of handicap by type of placement. Approximately 50% of states were able to do so in varying degrees of completeness (see Appendix B). In states in which local social services agencies forwarded aggregated data in both categories to the state MIS system, making the local agency the unit of analysis, "client-level" crosstabulations were not positive.