 |
This Web site is a component of the SAMHSA Health Information Network. |
|
1999 Annual Report to Congress on the Evaluation
of the Comprehensive Community Mental Health
Services for Children and Their Families Program
Back to Report
EXECUTIVE SUMMARY
Over the past three decades, increasing attention has been given to the
needs of youngsters with serious emotional disturbance. Beginning with
the findings from the Joint Commission on the Mental Health of Children
(National Institute of Mental Health, 1969) and substantiated by numerous
subsequent studies, task forces, and reports, children with serious emotional
disturbance were typically found to be underserved or served inappropriately
by a fragmented mental health services system. In response to these findings,
Federal leadership, along with a growing family movement, began to emerge
and create a new paradigm for serving the estimated 4.5 to 6.3 million
children in the United States who have a serious emotional disturbance
(Friedman, Katz-Leavy, Manderscheid, & Sondheimer, 1999). This paradigm
shift has resulted in a system-of-care approach, designed to help build
comprehensive service systems that allow children with emotional disturbance
to receive a comprehensive array of integrated, community-based services.
The system-of-care approach is based on a philosophy built on three hallmark
tenets: (a) mental health service systems are driven by the needs and
preferences of the child and family; (b) services are community based
with their management built on multi-agency collaborations; and (c) the
services offered, the agencies participating, and the programs generated
to meet the mental health needs of the children are responsive to the
cultural context and other characteristics of the populations being served.
To develop a system of care consistent with the theoretical approach
described above, a community must focus its service program activities
at two distinct levels: (a) infrastructure to house, organize, and manage
the integrated program elements; and (b) service delivery to provide services,
treatments, and supports that are offered directly to children and families.
The Center for Mental Health Services (CMHS) within the Substance Abuse
and Mental Health Services Administration (SAMHSA) has had the primary
responsibility for translating this framework into a program of services
and supports that now exists in 45 grant communities around the country.
Beginning with a funding mechanism to support the creation of a service
system infrastructure called the Child and Adolescent Service System Program
(CASSP) in 1986 and culminating with the passage of the Children's and
Communities Mental Health Services Improvement Act in 1992, CMHS has championed
the development of community-based, family-focused, and culturally competent
systems of care. Funding for these systems of care is provided through
CMHS's Comprehensive Community Mental Health Services for Children and
Their Families Program.
This program, now in its sixth year, provides grants to States, communities,
territories, American Indian tribes, and Alaskan Native communities to
improve and expand their systems of care to meet the needs of children
and adolescents with serious emotional disturbance and their families.
These include children and youth with a serious emotional disturbance
from birth to age 21 who currently have, or at any time during the past
year had, a mental, behavioral, or emotional disorder of sufficient duration
to meet diagnostic criteria specified in the Diagnostic and Statistical
Manual for Mental Disorders, Fourth Edition (DSM–IV; American
Psychiatric Association, 1994), that resulted in functional impairment
that substantially interferes with or limits one or more major life activities.
Administered by CMHS's Child, Adolescent and Family Branch, funding for
the initiative has grown from an initial $5 million in its first year
to the current level of $78 million.
In addition to developing systems of care, the CMHS program is mandated
by Congress to demonstrate the effectiveness of these systems. The foremost
method to demonstrate program effectiveness is through evaluation. Beyond
serving as a fiscal watchdog, evaluation—both of specific grant community
programs and of the combined effect of many systems of care—can inform
CMHS about what works best for whom, and under what conditions. As such,
it helps guide new directions for current grant communities and sets a
benchmark for future generations of community-based systems of care.
This report presents data accumulated through August 1999 from 22 grant
communities initially funded in either FY 1993 or FY 1994 (Phase I), and
from 9 grant communities first funded in FY 1997 (Phase II). Therefore,
this evaluation report focuses on 31 grant communities that established
systems of care for approximately 40,029 children and their families.
The 14 grant communities first funded in FY 1998 will begin collecting
data in FY 2000, and therefore do not have data to contribute to this
report.
This Executive Summary presents findings from three types of evaluation
data, described as follows:
- Outcome data based on project site evaluations of
a representative group of children in each community selected at
intake into services and assessed at intake, 6 months, 1 year, and
annually thereafter for as long as they remained in the program.
The number of children enrolled in each community ranged from 100
to over 3,000, based on sampling determined by the number of children
served. In total, communities collected outcome data for 18,623
children. Outcome measures applied in the evaluation included—but
were not limited to—an assessment of the child's clinical and
social functioning, educational performance, and stability of living
arrangements, coupled with an assessment of family and child ratings
of the services provided. Instruments typically used in the field
of children's mental health, including the Child Behavior Checklist
(CBCL; Achenbach, 1991) and the Child and Adolescent Functional
Assessment Scale (CAFAS; Hodges, 1990), were used to collect these
data.
- Service system and service experience data obtained
through systemwide and family assessments of service delivery in
communities with system-of-care grants and communities without these
grants.
- Descriptive data (e.g., demographic information,
diagnostic status, functional characteristics, and referral sources)
obtained at the time children entered services. Communities collected
descriptive data for 40,029 children.
The format for reporting findings in this Executive Summary is to present
a brief finding followed by its corresponding graph. All findings are
for the number of children with complete data for variables of interest
at each data collection point included in a specific analysis. Findings
of change over time are reported as the proportional change occurring
among children from the time of entry into services to subsequent data
collection times.
This Executive Summary concludes with a list of the grant communities
funded in FYs 1993, 1994, 1997, and 1998 under this congressionally mandated
program.
CHILD OUTCOMES
Preliminary findings show improvements for children who are in services
for up to 2 years.
Behavioral and Emotional Problems Were Reduced
After 2 years of receiving services, 42 percent of the children showed
a significant reduction in severe behavioral and emotional problem symptoms.
An additional 48.3 percent of the children were stabilized. These trends
were measured using the total problem score on the Child Behavior Checklist
(see Figure ES-1).
For over two-thirds of children who improved in their behavior, problem
scores fell below those indicating a need for clinical care (see Figure ES-2).
Clinical Functioning Improved
After 2 years in services, the percentage of children who had scores
below 40 on the Child and Adolescent Functional Assessment Scale (CAFAS)
more than doubled, from 13.5 percent to 29 percent. Scores below 40 indicate
that children are no longer considered to be clinically impaired in their
social functioning (see Figure ES-3).
Boys and girls entered services with similar levels of functional
impairment, and both made improvements. Girls made faster and greater
improvements in functioning than boys (see Figure ES-4).
School Attendance Improved
Regular school attendance increased from 85.9 percent at entry into services
to 89.4 percent after 1 year (see Table ES-1).
School Performance Improved
Grades that were Average or Above Average made a 20 percent gain among
children remaining in services up to 1 year (see Table ES-1).
Children in special education programs also showed improvement
in their grades. Average or Above Average grades improved by 14 percent
over those at intake (see Table ES-2).
Law Enforcement Contacts Were Reduced
Law enforcement contacts were reduced by 25 percent among children who
remained in services after 1 year when compared to similar contacts at
intake (see Figure ES-5).
Children with Reduced Law Enforcement Contacts Improved
in Functioning
Children who had a decrease in contacts with law enforcement made the
greatest improvement in their overall social functioning. The average
functional impairment score on the Child and Adolescent Functional Assessment
Scale (CAFAS) decreased by 19 points for these children. Children whose
contacts with law enforcement stayed the same also made improvements in
social functioning, while children who had increased contacts did not
improve (see Figure ES-6).
Improvements in overall social functioning, as indicated by the CAFAS,
suggest that children are functioning better in their home, school, and
community environments; are behaving more positively with others; are
reducing substance use; and are less impaired by moods or emotions, or
dysfunctional thought processes.
Residential Stability Improved
Establishing stability in the living situations of children who may have
had multiple, short-term placements, ranging from the homes of relatives
to foster homes and residential treatment facilities, is an important
goal of the program. Single residential living arrangements among children
who remained in services for 1 year made a 20 percent gain over similar
placements at intake (see Figure ES-7).
Children with Substance Abuse (SA) Diagnosis Improved in Functioning
About 17 percent of the children had a secondary diagnosis of substance
abuse at entry into services. These children faced greater challenges,
yet made the greatest improvements in functioning 1 year after entering
services. Children with a substance abuse diagnosis showed a 29-point
decrease in their average functional impairment score as measured by the
Child and Adolescent Functional Assessment Scale (CAFAS). Children without
a substance abuse diagnosis showed a 12-point decrease in their impairment
score (see Figure ES-8).
FAMILY RATINGS OF SERVICES
Caregivers Rated Service Activities at a High Level
- Caregivers were satisfied with services. Over
three-fourths of the caregivers rated services as "good or excellent"
- Caregivers were satisfied with their child's progress.
Over two-thirds of the caregivers were satisfied with their child's
progress after 1 year in services (see Figure ES-9).
- Caregivers reported having a choice in making service
decisions. Over three-fourths of the caregivers felt that they were
always given a choice in decisionmaking related to their children's
services (see Figure ES-9).
- Caregivers reported being asked their opinions about
their child's treatment. Three-fourths of the caregivers indicated
that they were always asked for their ideas and opinions (see Figure ES-9).
- Caregivers received unconditional, high-quality
care. Three-fourths of the caregivers felt they would receive care
unconditionally and believed that the quality of their care would
be high (see Figure ES-9).
SYSTEM-OF-CARE SERVICE FEATURES
The evaluation of service system implementation was based on eight principles.
The eight system-of-care principles are family focused, culturally competent,
interagency, community based, accessible, collaborative/coordinated, individualized,
and least restrictive. In each community, approximately 20 to 25 system
participants were interviewed, including agency representatives, administrators,
providers, and family members. These assessments were useful for describing
how developed systems of care differed from communities that provided
services in the traditional manner.
Communities Delivered Services According to
System-of-Care Principles
In East Baltimore, a system-of-care grant community, service delivery
scores on each principle were higher than in West Baltimore, a community
that did not have a system-of-care grant (see Figure ES-10).
Because each community's service system is unique and impacted by a variety
of local factors, the characteristics presented here for the East and
West Baltimore communities are an example specific to these communities
and do not necessarily reflect the characteristics of other funded system-of-care
or non-funded non-system-of-care communities.
Families Experienced Services According to System-of-Care
Principles
Based on data from the family-centered interview study, families in Stark
County, Ohio, a system-of-care community, found their direct service experiences
to be more child and family focused, community based, and culturally competent
compared to families who received services in Youngstown, Ohio, a community
that did not have a system-of-care grant. The families in Stark County
also felt that services they received had a greater impact. For purposes
of the study, impact was defined in terms of the appropriateness of services
for the child and family and the level of the child's improvement. Thus,
families served in Stark County felt their services were more appropriate
and their children showed greater improvement compared to families served
in Youngstown. The family-centered interview study involved rating the
extent to which a family's service experience embodied system-of-care
principles in each of four domains: child and family focused, community
based, culturally competent, and impact (see Figure ES-11).
Because each community's service system is unique and impacted by a variety
of local factors, the characteristics of service experience presented
here for Stark County and Youngstown, Ohio, are an example specific to
these communities and do not necessarily reflect the characteristics of
other funded system-of-care or non-funded non-system-of-care communities.
FUNDING
A goal of the program is to create systems of care but also to expand
capacity of communities to use the system-of-care approach. To do so,
expansion of funding as well as new funding sources are important factors.
In fact, the grant required recipients to develop mechanisms for matching
Federal dollars that slowly declined over the 5-year funding period. Evidence
of change in funding for children's mental health services comes in part
from the evaluation of specific communities.
Total Dollars Contributed to Children's Mental Health Budget Increased
In Milwaukee, Wisconsin, expansion as well as changes in funding are
evidenced by the dramatic change in contributions from sources other than
the program grant. The total dollars contributed to the children's mental
health budget by sources other than the CMHS grant rose by $28 million
in one community (Wraparound Milwaukee) (see Figure ES-12).
This increase resulted, in part, from a significant expansion of the
program's target population to include children from child welfare and
children's court who were currently in or at risk of placement in residential
treatment and to the fact that services were being reimbursed by Medicaid
at a capitated rate in Wisconsin. While this dramatic increase in funding
from sources other than the grant cannot be directly attributed to the
presence of the grant program, program staff viewed the expansion as
a credit to the program for providing effective services at a reduced
cost.
CHILD AND FAMILY DEMOGRAPHICS
Children with Serious Emotional Disturbance Were Disproportionally Poor,
Male, and in Living Situations Other Than Two-Parent Homes
- Almost two-thirds were boys. About 62 percent
of the children were boys and 38 percent were girls (see Table ES-3).
- Average age was in early adolescence. The
children's average age across grant communities was 12.2 years.
Among the children, 21 percent were aged 8 years or younger, 17
percent were aged 9 to 11 years, 26 percent were aged 12 to 14 years,
and 36 percent were aged 15 years or older (see Table ES-3).
- A diverse racial-ethnic population was served.
Among the children, 55 percent were Caucasian, 15 percent were African-American,
25 percent were Hispanic, 2 percent were American Indian or Alaskan
Native, 2 percent were Asian or Pacific Islander, and 1 percent
were of other ethnicities. The representation of Hispanic and African-American
families was higher than expected when contrasted to the general
population of the United States. This is partially due to the geographic
distribution of CMHS-funded system-of-care grant communities and
partially due to the provision of system-of-care services within
the public sector (see Table ES-3).
- High rates of single-parent families were served.
Among the children, 53 percent lived in single-parent homes, 24
percent lived in two-parent homes, 7 percent lived with a guardian,
10 percent were wards of the State, and 4 percent had other living
situations. Higher-than-expected rates of children residing in single-parent
families were reported for children entering systems of care. In
comparison, in 1997, 24 percent of all children in the United States
resided in single-parent families (DHHS, 1999) (see Table ES-3).
- The majority of families were poor. Sixty-one
percent of the children's families reported incomes below $15,000.
This level of poverty is well above rates reported for the general
population, which indicate that approximately 20 percent of all
children under age 18 live in poverty (see Table ES-3).
- Mother-maintained households had the highest
poverty rates. Children living in mother-maintained households
were more likely to live in poverty. Of the responding families,
72 percent of the children in mother-maintained households lived
in poverty, 47 percent of the children in father-maintained households
lived in poverty, and 43 percent of the children in two-parent households
lived in poverty (see Figure ES-13). These rates of poverty were greater than national averages. Among mother-maintained households
with children under age 18, 41 percent were reported to live in
poverty (U.S. Census Bureau, 1999a).
REFERRAL SOURCES
An important purpose of the program is to develop partnerships among
major child-serving agencies. Referrals to program services from multiple
sectors in the community are one indicator that agencies are collaborating
to best serve children in their community.
Multiple Referral Sources Indicate Interagency Partnership in Systems
of Care
Because the program has had more than one funding phase, and because
service trends may change over the years, comparing information from the
distinct funding phases may reveal interesting trends. Changes in referral
sources may indicate, for example, greater engagement of specific child-serving
sectors, or greater awareness of available services throughout the community.
When examining referrals to system-of-care services, mental health and
school referrals have remained relatively stable over the two phases.
More referrals were made by social service agencies among communities
funded in 1993 and 1994 (Phase I), while more referrals were made from
courts and correctional institutions and from parents among communities
funded in 1997 (Phase II) (see Figure ES-14).
CHILD RISK FACTORS
Risk factors are events that occur at the child, family, and environmental
levels that increase the probability of diagnosis or the severity of a
serious emotional disturbance (e.g., physical abuse, sexual abuse, family
violence, and drug-alcohol abuse; family history of mental illness, violence,
or drug-alcohol abuse). Risk factors in this evaluation were determined
as lifetime risk, and were assessed independent of each other so that
percentages do not total to 100 percent.
The Majority of Children Had Been Exposed to at Least One Risk Factor
Of the responding families, 60 percent reported that their child had
at least one risk factor for serious emotional disturbance. The most frequently
reported child risk factors included history of physical abuse (33 percent),
previous psychiatric hospitalization (25 percent), history of running
away (25 percent), and sexual abuse (24 percent) (see Figure ES-15).
FAMILY RISK FACTORS
Over Three-Fourths of the Families Had Experienced at Least One Family
Risk Factor
Of the responding families, 79 percent identified at least one family
risk factor, with 39 percent indicating three or more risk factors. Twenty
percent identified one family risk factor, and 20 percent identified two
risk factors. The highest reported risk factors were history of substance
abuse (62 percent), history of violence (54 percent), and history of mental
illness (45 percent) (see Figure ES-16).
DIAGNOSTIC CHARACTERISTICS
Diagnoses are determined at intake using criteria from the Diagnostic
and Statistical Manual for Mental Disorders, Fourth Edition (DSM–IV).
Children Presented with a Wide Range of Primary Diagnoses
The children had a wide range of diagnoses, with many having disruptive
behavior disorders, including conduct disorder, oppositional defiance
disorder, and attention-deficit/hyperactivity disorder. Other major primary
diagnoses were depressive disorders and anxiety. Disorders such as adjustment,
substance abuse, eating, somatic, speech problems, enuresis, poor self-concept,
phobia, and psychosis were less commonly reported as primary diagnoses.
Many more children have been diagnosed with disruptive behavior disorders
among communities funded in 1997 (Phase II) than in those funded in 1993
and 1994 (Phase I), while fewer children have been diagnosed with depression
or anxiety (see Figure ES-17). These trends may indicate changes in clinical attention to and community awareness of children's emotional and behavioral
problems, means of access to available services, or other service-related
factors.
Approximately One-Fourth of the Children Had a Secondary Diagnosis
More than one-fourth of the children (29 percent) had two distinct diagnoses.
It is worthwhile to consider differences in diagnoses between the two
funding phases, as these may indicate emerging trends in the mental health
field. Among children with two diagnoses, a secondary diagnosis of a conduct-related
disorder was more common among children in communities funded in 1997
(Phase II) (see Figure ES-18).
In summary, more children with disruptive behavior disorders and fewer
children with depression or anxiety entered systems of care. The number
of children with multiple diagnoses was similar to the number reported
in other studies of children with serious emotional disturbance (Greenbaum
et al., 1996).
EDUCATIONAL STATUS
Parents reported on their children's educational placement, school attendance,
and school performance at entry into system-of-care services.
Most School-aged Children Had an Individualized Education Plan (IEP)
Over half of the school-aged children (55 percent) had an IEP at intake,
and most of the children had IEPs related to the emotional disturbance
designation (62 percent). About another 5 percent of children had IEPs
pending (see Figure ES-19). Children in grades 1 and 2 were less likely to have an IEP than children in higher grade levels.
In 1997, just over 11 percent of all children enrolled in public schools
were receiving services under the Individuals with Disabilities Education
Act (IDEA). Only 8.6 percent of children served through IDEA were classified
with emotional disturbance (OSEP, 1997), which indicates that many children
with serious emotional disturbance are not receiving services within the
educational system.
Approximately One-Half of the Children Were Placed in Regular Education
Classrooms
More than half (59 percent) of the children were in regular classrooms,
20 percent were in special education classes, and 2 percent were suspended
or had formally dropped out (see Figure ES-20). Thirty-six percent of the children were assisted by classroom aides, while 64 percent had no
classroom aides (see Figure ES-21).
JUVENILE JUSTICE STATUS
Caregivers reported the following juvenile justice information about
their children:
Contacts with Youth Authorities
During the 12 months before entering system-of-care services, approximately
29 percent of the children had experienced some contact with law enforcement
(see Figure ES-22).
Adjudicated Convictions
In the 12 months before entering system-of-care services, 13 percent
of the sample had one or more adjudicated misdemeanors, 6 percent reported
an adjudicated felony, and 15 percent had one or more arrests that led
to convictions (see Figure ES-22).
According to the Office of Juvenile Justice and Delinquency Prevention's
statistical briefing book, in 1997 law enforcement agencies made an estimated
2.8 million arrests of persons under the age of 18, of whom 32 percent
were under the age of 15. There was also a 14 percent increase in the
total number of juvenile arrests from 1993 to 1997. The overall arrest
rate for youth in 1997 was 9,200 per 100,000, or approximately 9 percent
(OJJDP, 1998). As expected, the numbers of law enforcement contacts and
arrests are higher for children entering systems of care than for the
general population.
SUBSTANCE USE STATUS
Self-reported use of substances collected from youth 11 years of age
or older receiving services was a new assessment added to the evaluation
of communities funded in 1997 (Phase II). Over half of these youth had
tried alcohol and cigarettes at least once, and nearly half had tried
marijuana at least one time (see Figure ES-23).
Grant Communities Funded through the Comprehensive Community Mental Health Services Program for Children and Their Families (1993–1998) (see Table ES-4).
CB-E199E
|
|
