University of Southern California Dissertations and Theses

Mental health service use by children and youth in the child welfare system: a focus on need and predisposing factors and caregiver type

(USC Thesis Other)

Mental health service use by children and youth in the child welfare system: a focus on need and predisposing factors and caregiver type

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

MENTAL HEALTH SERVICE USE BY CHILDREN AND YOUTH IN THE
CHILD WELFARE SYSTEM: A FOCUS ON NEED AND PREDISPOSING
FACTORS AND CAREGIVER TYPE

by

Margarita Villagrana

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(SOCIAL WORK)

December 2008

Copyright 2008 Margarita Villagrana
ii
EPIGRAPH

If I have seen further than others, it is only because
I was standing on the shoulders of giants

~ Sir Isaac Newton

iii
TABLE OF CONTENTS

Epigraph ii

List of Tables v

List of Figures vi

Abstract vii

CHAPTER 1: INTRODUCTION 1
Study Rationale 3
Study Aims 7

CHAPTER 2: REVIEW OF THE LITERATURE 9
The Child Welfare System 9
Mental Health Need of Children in the Foster Care System 10
Mental Health Service Utilization of Children in the Foster Care System 11
The Gateway Provider Model 15
Need and Predisposing Factors 17
Knowledge and Perceptions of Mental Health Services 22
Type of Caregiver 27
Preliminary Studies 34
Conceptual Model 37
Summary of the Literature 39

CHAPTER 3: RESEARCH METHODS 40
Research Questions and Hypotheses 40
Data Source 43
Sample Characteristics 46
Variables and Measures 48
Data Analysis 54
Power Analysis 60

CHAPTER 4: RESULTS 62
Preliminary Analyses 62
Research Question 1 Hypotheses 1A and 1B 65
Research Question 2 Hypotheses 2AI, 2AII, and 2B 68
Research Question 3 Hypotheses 3A1, 3AII, 3BI, and 3BII 77
Overview of Major Findings 89

iv
CHAPTER 5: DISCUSSION 92
Caregivers’ Characteristics and Mental Health Service Use 93
Need and Predisposing Factors and Mental Health Service Use 94
Perceptions and Knowledge of Mental Health Services 97
Caregiver Type 102
Limitations 106
References 110

v
LIST OF TABLES

Table 1. Demographic Characteristics of Caregivers 48

Table 2. Need and Predisposing Factors, and Caregiver Type by Mental
Health Service Use 64

Table 3. Hierarchal Logistic Regression: Covariates, Need Factor, and
Mental Health Service Use 66

Table 4. Hierarchal Logistic Regression: Covariates, Predisposing Factor,
and Mental Health Service Use 68

Table 5. Need and Predisposing Factors and Mental Health Service Use
by Caregiver Type 79

Table 6. Model Comparison Test Results for Need Factor and Perceptions
and Knowledge of Mental Health Services 83

Table 7. Estimated Path Coefficients in the Moderator Model of the Need
Factor and Perceptions and Knowledge of Mental Health
Services by Caregivers 84

Table 8. Model Comparison Test Results for Need Factor and Mental
Health Service Use 85

Table 9. Estimated Path Coefficients in the Moderator Model of the Need
Factor and Mental Health Service Use 86

vi
LIST OF FIGURES

Figure 1. Gateway Provider Service Framework 6

Figure 2. The Gateway Caregiver Model 38

Figure 3. Need and Predisposing Factors and Perceptions and
Knowledge of Mental Health Services 70

Figure 4. The Effect of Need Factor on Perceptions and Knowledge of
Mental Health Services 72

Figure 5. The Effects of Predisposing Factor on Perceptions and
Knowledge of Mental Health Services 73

Figure 6. The Effects of Perceptions and Knowledge of Mental Health
Services to Mental Health Service Use Controlling for the
Need Factor 75

Figure 7. The Effects of Perceptions and Knowledge of Mental Health
Services to Mental Health Service Use Controlling for the
Predisposing Factor 76

Figure 8. Moderator Effect by Caregiver Type Revised Conceptual
Model 81

Figure 9. Moderator Effect of Need Factor and Mental Health Service
Use by Caregivers 85

Figure 10. Moderator Effect of Predisposing Factor and Perceptions and
Knowledge of Mental Health Services by Caregivers 88

Figure 11. Moderator Effect of Predisposing Factor and Perceptions and
Knowledge and Mental Health Service Use by Caregivers 89

vii
ABSTRACT

Studies have documented that children in foster care are in need of mental
health intervention, but not all foster care children receive the mental health
services they need. One area that lacks attention and may help explain the
underutilization of mental health services is the influence that the caregiver has on
a child’s utilization of services. Caregivers serve as gatekeepers for children
while in the child welfare system, but few studies have focused on the factors that
influence a caregiver’s use of mental health services for the children under their
care. This study represents a step toward addressing this gap in the literature by
examining the influence that caregivers have on the utilization of mental health
services for children and youth in the child welfare system. This study examined
whether the caregivers’ perceptions and knowledge of mental health services
mediate need and predisposing factors and the utilization of mental health
services for children under their care. This study also examined caregiver type as
a moderator to understand whether type of caregiver was significant.
Hierarchal logistic regression and structural equation modeling with a
mediating-moderating effect were used to test the hypotheses proposed. Results
indicate that older caregivers were more likely to use mental health services for
their children. Caregivers’ with children with high externalizing behaviors and a
high level of stress were more likely to use mental health services for their
children. No mediation effect was found. However, statistically significant
viii
associations between need and predisposing factors and perceptions and
knowledge of mental health services were found with the higher the need factor
the more negative the perceptions and knowledge of mental health services and
the higher the predisposing factor the more negative the perceptions and
knowledge of mental health services were by the caregiver.
In addition, a moderating effect was found with the higher the need factor
the more negative the perceptions and knowledge of mental health services
among birth parents and relative caregivers. Furthermore, the higher the need
factor the less likely that relative caregivers and foster parents were to use mental
health services for the children under their care.

1
CHAPTER 1: INTRODUCTION
In 2006, there were approximately 3.3 million reports made of children
being abused or neglected. Of these reports, approximately 30% determined at
least one child was a victim of abuse or neglect. The majority of children were
victims of neglect (64.1%), followed by physical abuse (16%), sexual abuse
(8.8%), and emotional maltreatment (6.6%) (DHHS, 2007). The physical and
psychological consequences experienced after an incident of abuse have been
determined to vary depending on the type of abuse a child experiences, age of the
child, developmental status, frequency, duration and severity of the abuse, and the
relationship between the victim and the abuser (Chalk, Gibbons, & Scarupa,
2002). However, research has shown that the consequences of maltreatment in
childhood can have devastating outcomes regardless of the factors that are
involved in the abuse. The long-term consequences of abuse may result in severe
psychological and risk behaviors such as sleeping disturbances, anxiety, ADHD,
conduct disorders, depression, posttraumatic stress disorder, eating disorders,
alcoholism, suicidal ideation, low self-esteem, and psychiatric problems, (Dallam,
2001; Perry, 2001; Kelley, Thornberry, & Smith, 1997; Silverman, Reinherz, &
Giaconia, 1996; Mullen, et al., 1996).
Untreated mental disorders among children in the child welfare system can
also have a serious impact on society, as children in foster care with untreated
mental health disorders are often homeless, in prison, or in psychiatric hospitals as
2
young adults (Courtney et al., 2001). In addition, untreated mental health
problems with children in the child welfare system can place great demands on an
already over burdened system. Children with emotional and behavior problems
often burn out foster parents and social workers, resulting in a high turnover rate
for both (see Kerker & Dore, 2006). As such, the mental health needs and service
use of maltreated children warrants examination.
Children enter the foster care system primarily due to maltreatment. Most
children experience multiple types of maltreatment and often a traumatic
separation from their primary caregiver and adjustment to a new environment
(Leslie et al., 2000; Clausen et al., 1998). The confounding of these extraneous
events will often precipitate or exacerbate mental health problems. Studies
indicate that a significant number of children in foster care exhibit mental health
problems that require mental health intervention (Clausen et al., 1998; Halfon,
Mendonca & Berkowitz, 1995). The prevalence of mental health disorders with
children in foster care are estimated as high as 80% (dosReis et al., 2001;
Landsverk & Garland, 1999; Halfon, Mendonca & Berkowitz, 1995). Although
studies indicate that children in the foster care system utilize mental health
services at a higher rate than non-foster care children on public assistance (Staudt,
2003; dosReis et al., 2001; Halfon et al., 1992), not all foster care children receive
the mental health services they need and little is known of the utilization of
services by this population (Garland et al., 2000; Harman et al., 2000).
3
While studies have generally focused on the child’s characteristics to
attempt to understand mental health service utilization of children in the foster
care system (Burns et al., 2004; Leslie et al., 2004; James et al., 2004; McMillen
et al., 2004; Newton et al., 2000; Garland et al., 2000; Zima et al., 2000; Ezzell,
Swenson, & Faldowski, 1999; Garland et al., 1996; Garland & Besinger, 1997;
Halfon, Berkowitz, & Klee, 1992), few studies have examined contextual factors
that may be significant in the utilization of mental health services for children
while in the child welfare system (Garland & Besinger, 1996; Hurlburt et al.,
2004; Raghavan et al., 2006; Raghavan et al., 2007)). Children in the child
welfare system are generally supervised by multiple sectors that are influential in
the receipt of services, but little attention has been given to the sources that serve
as pathways into mental health services for these children. One area that lacks
attention and may help explain the underutilization of mental health services is the
influence that the caregiver has on a child’s utilization of services. The focus of
this study, therefore, is on the caregiver and the utilization of mental health
services for the children under their care.
Study Rationale
Caregivers serve as gatekeepers for children while in the child welfare
system (Pasztor et al., 2006), but few studies have focused on the factors that
influence a caregiver’s use of mental health services for the children under their
care. The extant literature with caregivers has generally only focused on need and
4
predisposing factors of kin caregivers to predict utilization of mental health
services for children in foster care (Yeh et al., 2003; Ehrle & Geen, 2002; Zima et
al., 2000; Burnette, 1999; Timmer, Sedlar & Urquiza, 2004; Ezzell et al., 1999;
Kolko et al., 1999; Shore et al., 2002). However, research studies with caregivers
in community samples indicate that a caregiver’s perceptions and knowledge of
mental health services is significant in whether children under their care utilize
mental health services (Flisher et al., 1997; Pavuluri, Luk & McGee, 1996;
Richardson, 2001; McKay et al., 1998; Zima et al., 2000; Yeh et al., 2003).
This study represents a step toward addressing this gap in the literature by
examining the influence that caregivers have on the utilization of mental health
services for children and youth in the child welfare system. The study examined
whether caregivers’ perceptions and knowledge of mental health services was
influential in the utilization of mental health services for the children under their
care. This study also examined caregiver type, which includes birth parent,
relative caregiver, and foster parent, as a moderator in understanding whether type
of caregiver was significant.
The conceptual model that was used in the present study is an adaptation
of the Gateway Provider Model (Stiffman, Pescosolido & Cabassa, 2004).
Concepts from the original model were adapted and modified for use with
caregivers with a population of children in the child welfare system. The
Gateway Provider Model was selected, as it is one of the few models to use
5
“gatekeepers” in examining the utilization of mental health services for children.
The Gateway Provider Model, developed from Pescosolido’s Network Episode
Model and Decision Theory, focuses on three central influences that affect the
treatment children receive: 1) the role of the person who first identifies a problem
or sends the child to treatment (i.e., “gateway provider”) 2) the information
needed by the “gateway provider” about the child’s problem and potential
resources, and 3) the provider’s attitudes, impressions of support treatment, and
system burden. The model also posits that need (i.e., presence of disorder,
severity of disorder, comorbid conditions, and impairment), enabling (i.e.,
availability, accessibility, affordability, and acceptability), and predisposing
factors (i.e., demographics and risk and protective factors) also contribute to the
utilization of services. However, it is the gateway provider’s perceptions and
knowledge of service resources and mental illness that ultimately influence the
decision to provide services (see Figure 1).
Figure 1. Gateway Provider Service Framework

Using the Gateway Provider Model as a guiding conceptual framework,
the adaptation of this model termed, the Gateway Caregiver Model, highlights the
role of the caregiver as the gateway to service use. Specifically, the model
proposes that perceptions and knowledge of mental health services of the
caregiver mediates the relationship between need and predisposing factors and
mental health service use by caregivers for the children under their care. Need
factors refer to the child’s psychological functioning. Predisposing factors refer
to risk and protective factors, such as social support, level of stress, and
depression of the caregiver. Perceptions and knowledge of mental health services
refer to attitudes toward mental health services, perceptions of barriers to service
use, and perceived need. The model further proposes that type of caregiver (i.e.,

6
7
birth parent, relative caregiver, and foster parent) moderates the mediating effects
of perceptions and knowledge of mental health services (see Figure 2 - Chapter II).
Study Aims

The application of the Gateway Caregiver Model is important to the field
because it provides the opportunity to address several gaps in the literature. For
example, the study conceptually extends what is known regarding use of mental
health services by introducing variables such as perceptions and knowledge of
mental health services and risk and protective factors (i.e., predisposing), concepts
that have been hypothesized to impact the use of mental health services by
caregivers for the children under their care. The present study also includes
perceptions and knowledge of mental health services of caregivers as a mediator.
This is important given that a person’s perceptions and knowledge of mental
health services is influential in seeking services. In addition, the majority of the
literature focuses only on kin caregivers, which is relatively a limited portion of
caregivers. Using type of caregiver as a moderator is an attempt to incorporate
the three most common types of caregivers in the child welfare system. Thus, the
results of this study offer important information for a broader range of caregivers
than what has previously been addressed in the literature. The model also offers
an opportunity to move the knowledge base forward regarding use of mental
health services by caregivers for the children under their care, which has received
little attention in the literature.
8
The specific aims of this study are:
Aim 1. To examine the impact of need and predisposing factors on mental
health service use by caregivers for the children and youth under
their care.
Aim 2. To determine whether caregivers’ perceptions and knowledge of
mental health services mediates the relationships between need and
predisposing factors and mental health service use by caregivers
for the children and youth under their care.
Aim 3. To investigate whether the paths from need and predisposing
factors to perceptions and knowledge of mental health services and
from need and predisposing factors to mental health service use by
caregivers for the children and youth under their care differ among
caregiver type.

9
CHAPTER II: REVIEW OF THE LITERATURE

The Child Welfare System

In September 2006, there were approximately 510,000 children in foster
care. The mean age of children in foster care was 9.8 years old. The majority of
children in foster care were Caucasian (40%), followed by non-hispanic Black
(32%), and Hispanic (19%). The rest were Asian, non-hispanic, unknown, and
two or more non-hispanic. Children appear to be spending an average of a little
over two years (28.3 months) in foster care. Forty two percent of children spent
less than 11 months in foster care 21% had spent less than 23 months, and 37%
had spent 24 months or more. The placement setting of children in foster care
consisted of the majority being in a non-relative placement with 46% in a foster
family home (non-relative), 7% in a group home, and 24% in a foster family
home (relative) (DHHS, 2006). There are no data as to the number of children
with an open child welfare case who remained at home with a parent. However,
to better understand the total number of children in home of parent (HOP) or
children residing at home, one of the largest states in terms of the total number of
children in its child welfare system was examined. California has 18% of the
total number of children in foster care nationwide. Estimates of the total number
of children in its child welfare system (i.e., out of home and children with a
parent) indicate that California consists of 115,622 children with open child
10
welfare cases. Of these children, 24% are in family maintenance cases; in other
words, these children are living with a parent (Needell, et al., 2006).
Mental Health Need of Children in the Foster Care System

Due to traumatic experiences (e.g., neglect and/or abuse) that necessitated
placement, psychological and neurobiological effects of disruptive attachment,
and the need to adjust to a foster care environment (Racusin et al., 2005), children
in the child welfare system exhibit mental health problems that require mental
health intervention (Clausen et al., 1998; Halfon, Mendonca & Berkowitz, 1995).
Children in the foster care system exhibit a range of emotional and behavioral
problems such as conduct and attention disorders, aggressive and self-destructive
behaviors, depression, delinquency, autism, bipolar disorders, and impaired social
relationships (see Kerker & Dore, 2006). In addition, due to the traumatic
separation from their parents, children are often found to experience feelings of
rejection, guilt, abandonment, and shame (Garland et al., 2000; Simms et al.,
2000). It is not surprising, then, that the child welfare literature has consistently
documented that children in the foster care system are at a higher risk for
psychopathology (Staudt, 2003; dosRios et al., 2001; Ezzell, Swenson, &
Faldowski, 1999; Clausen et al., 1998; Garland et al., 1996), than children in the
general population. The prevalence of mental health disorders with children in
foster care are estimated as high as 80% (dosReis et al., 2001; Landsverk &
Garland, 1999; Halfon, Mendonca & Berkowitz, 1995), a high percentage when
11
compared to only one tenth of children in community samples (U.S. Public Health
Service, Office of the Surgeon General, 2000). In addition, children in foster care
are 16 times more likely to have psychiatric diagnoses and 8 times more likely to
be taking psychotropic medication than children in community samples (see
Racusin et al., 2005). Studies have also documented that children in the foster
care system utilize mental health services at a higher rate than non-foster care
children on public assistance (Staudt, 2003; dosReis et al., 2001; Harman, Childs
& Kelleher, 2000). In a study conducted by Halfon, Berkowitz, and Klee (1992)
in California of Medicaid paid claim forms of children under 18 years of age in
and out of foster care, researchers found that although children in foster care
constituted less than 4% of the total Medicaid population, they used a substantial
amount of mental health services (41%). However, not all children in the foster
care system receive the mental health services they need (Burns et al., 2004).
Mental Health Service Utilization of Children in the Foster Care System

Comparatively less is known of the utilization of mental health services
for children and youth in the foster care system. Rates of mental health service
utilization by children in the foster care system vary by study and range from a
low of 19% to a much higher rate of 94% (see Burns et al., 2004). There are
several reasons for the variation in mental health service use by children reported
in studies. One reason is that estimates of mental health service use are difficult
to calculate because there is a variation in definition of mental health services
12
across studies (Landsverk et al., 2002). Mental health services may include
mental health assessments, case management services, outpatient mental health
services, day treatment services, therapeutic foster/group home, inpatient
psychiatric services and psychotropic medication. For example, Ezzell et al.
(1999) assessed the relationship between child, family, and case characteristics on
service utilization, including mental health and medical services, on children who
had been physically abused. The sample included children who were either in
foster care or with a relative caregiver, or who had remained at home with their
birth parent(s). Of the children in the study, 43% had received mental health
services, but it was not clear as to the type, frequency or duration of the services.
In another study conducted by Swenson, Brown, and Sheidow (2003) of children
who had been physically abused, researchers found that 35% of the children who
had scored in the clinical range on the Child Behavior Check List (CBCL) were
receiving mental health treatment, but treatment was considered either receiving
outpatient mental health, psychiatric hospitalization, or psychotropic medication.
Another reason for the variation in mental health service use is that the
methodologies used to collect mental health service use rates also vary, with some
studies using administrative data and others relying on caregiver report of receipt
of services (Garland et al., 1996; Leslie et al., 2000). The reliability of report of
service use by caregivers is unknown, but reports by caregivers may be biased or
unreliable due to placement and funding decisions (Leslie et al., 2000), recall
13
issues, and placement changes (James et al., 2004). In addition, the use of
Medicaid claims to analyze service use limits the rates of service use in that other
non-Medicaid funding sources are unaccounted for such as child welfare agency
and use of victims of crime funding (Ezzell et al., 1999). However, in one of the
only national studies of children in the child welfare system, Burns and colleagues
(2004) investigated need for and use of mental health services among children
investigated by child welfare agencies after a report of maltreatment of a cohort of
3,803 youth between the ages of 2 to 14 years. Researchers found that youth with
mental health need as defined by the Child Behavior Check List (CBCL) were
more likely to receive mental health services, but only 25% of such youth
received any specialty mental health care during the previous 12 months of the
study.
In an effort to better understand the utilization of mental health services by
children in the foster care system, research studies have focused on predictive
variables of mental health service use which generally include demographic and
case characteristics (Burns et al., 2004; Leslie et al., 2004; James et al., 2004;
McMillen et al., 2004; Newton et al., 2000; Garland et al., 2000; Zima et al.,
2000; Ezzell, Swenson, & Faldowski, 1999; Garland et al., 1996; Garland &
Besinger, 1997; Halfon, Berkowitz, & Klee, 1992). Most notably are the child’s
demographic characteristics such as, age, gender, and ethnicity which have been
reported to be influential in the receipt of services with older age, male gender,
14
and Caucasian children more likely to receive services (Burns et al., 2004; Leslie
et al., 2004; James et al., 2004; Zima et al., 2000; Garland et al., 2000; Garland &
Besinger, 1997; Halfon, Berkowitz, & Klee, 1992). Other case characteristics
include type of abuse with children who experience sexual abuse most notably
referred for services (Burns et al., 2004), while children who experience physical
abuse are only referred when externalizing behaviors are displayed (Garland et al.,
1996; Blumberg et al., 1996). Although, the majority of children in foster care are
placed due to neglect (DHHS, 2005), children who have experienced neglect are
the least likely to utilize services (Leslie et al., 2004; Garland et al., 2000; Leslie
et al., 2000; Garland et. al., 1996). Type of placement has also been consistently
documented in the child welfare literature as yielding significant findings in the
utilization of mental health services. More restrictive placement settings, which
include institutions and residential facilities, usually provide a comprehensive
array of services and may include psychiatric care, psychotropic medication, and
psychological services. Foster homes and kinship care placements, the least
restrictive of placement settings, generally provide less comprehensive services or
no services at all. In addition, studies indicate that children in kinship care are the
least likely to receive mental health services (Leslie et al., 2004; McMillen et al.,
2004; James et al., 2004; Garland et al., 2000; Leslie et al., 2000; Garland et. al.,
1996).

15
The Gateway Provider Model
One recent model developed by Stiffman, Pescosolido, and Cabassa
(2004) of children in community settings has shifted the focus from the child to
the various gateway providers associated with the child. Children usually do not
seek services on their own; they tend to be directed to services by the adults in
their lives that generally constitute their parents, teachers, juvenile justice
authorities, and educators as well as other adults. Generally, these individuals
cannot offer direct mental health services; however, their knowledge and
awareness of services and assessment of children’s symptoms, diagnosis, and
impairment are essential in referring children to services. Even when need is
identified, the gateway providers actual or perceived lack of available mental
health services and treatment can present a barrier to actual service provision (see
Stiffman, Pescosolido, & Cabassa, 2004).
The Gateway Provider Model (see Figure 1) was developed from
Pescosolido’s Network Episode Model (NEM) and Decision Theory to
incorporate the gateway providers perceived need of mental health services by the
child and the service system such as organizational factors (e.g.,
interorganizational relationships within and across service systems), area
characteristics (e.g., availability, accessibility, and cultural behavior), and social
network mechanisms (e.g., linking functions of gateway providers). The NEM
focuses on the kinds of individuals that become involved in the child’s mental
16
health problem. More specifically, it focuses on how members of the community
and the treatment systems work together to either facilitate or hinder the provision
of services and outcomes. Decision Theory, on the other hand, links knowledge
of resources by the individual (or community member) to perceived need in
deciding to provide mental health services (see Stiffman, Pescosolido, & Cabassa,
2004).
The Gateway Provider Model “draws insight from the NEM to understand
what environmental factors may affect provider behavior and from Decision
Theory to suggest what factors providers could consider in making treatment
decisions” (Stiffman, Pescosolido, & Cabassa, 2004 p.192). It focuses on three
central influences that affect the treatment children receive: 1) the role of the
individual, usually not in the mental health system or a formal provider, who first
identifies a need or sends the child to treatment (i.e., gateway provider); 2) the
information needed by the “gateway provider” about the child’s problem and
potential resources; and 3) the provider’s attitudes, impressions of support
treatment, and system burden. All three influences can either hinder or facilitate
the treatment that children receive. The model posits that need, enabling, and
predisposing factors contribute directly to the utilization of services. However,
the gateway provider’s perceptions of these factors mediate their impact on
service use. The gateway provider’s perceptions and knowledge are also
influenced by structural and systemic characteristics of their environment.
17
“Together, the gateway provider’s perceptions, knowledge, and environment
influence their decision making surrounding service provision” (Stiffman,
Pescosolido, & Cabassa, 2004 p.193). Overall, the gateway provider’s
perceptions and knowledge of service resources and mental illness ultimately
influence the decision to provide services.
Need and Predisposing Factors
Using the Gateway Provider Model’s main concept as a premise in
understanding utilization of mental health services for children and youth in the
child welfare system, an area that may provide an explanation in the
underutilization of mental health services for children and youth is the influence
that the caregiver has on a child’s utilization of services. Children generally do
not seek mental health services for themselves; it is the caregiver that serves as
the gatekeeper for children while in the child welfare system (Pasztor et al., 2006).
However, few studies have focused on the caregiver and the utilization of mental
health services for the children and youth under their care in the child welfare
literature. Of these studies, the majority have only focused on mental health
service use by kin caregivers (Leslie et al., 2004; McMillen et al., 2004; Garland
et al., 2000; Leslie et al., 2000; Garland et. al., 1996). Research studies on kin
caregivers and with caregivers in community samples have generally examined
caregiver’s need and predisposing factors, with social support, stress level, mental
health, and a child’s psychological functioning influencing whether children
18
under their care receive services (Ezzell et al., 1999; Kolko et al., 1999; Zima et
al., 2000; Geen, 2002; Shore, 2002; Brannan, Heflinger, & Foster, 2003; Timmer,
Sedlar & Urquiza, 2004; Alegria et al., 2004; Leslie et al., 2006). For example,
predisposing factors such as social support, stress, and the mental health of the
caregiver have been found to be significant, with caregivers displaying a higher
level of stress more likely to seek mental health services (Ezzell et al., 1999;
Kolko et al., 1999; Timmer, Sedlar & Urquiza, 2004) and caregivers with a small
or no support system less likely to utilize services for the children under their care
(Geen, 2002).
Studies have found that caregiver strain is significant in the utilization of
mental health services for the children under their care. In a study conducted by
Brannan and Heflinger (2005) of caregivers of children who were Medicaid
enrollees and who were receiving mental health treatment in Mississippi and
Tennessee, researchers found that the more the stress in disruptive family and
social relationships, problems with neighbors in the community, and interrupted
work and personal time, that the caregiver reported the more services their child
was likely to use. In the same study, researchers also found that report of feelings
of anger, resentment, and embarrassment about the child’s problem decreased the
likelihood of receiving services by 33%. In another study (Brannan, Heflinger, &
Foster, 2003) of a cohort of 574 children receiving mental health services,
researchers found that the lower the report of caregiver worry, guilt, sadness, and
19
fatigue the greater the probably of children only receiving outpatient services,
whereas, the higher the report of worry, guilt, sadness, and fatigue, the greater the
probably of children receiving out patient services plus residential treatment.
Angold and colleagues (1998) conducted a study of a cohort of 1,015 children of
the public school system in 11 counties in western North Carolina to examine
parental burden resulting from children’s and adolescents’ psychiatric disorders.
Researchers found that the most common individual burden reported involved
effects on personal well-being, stigma, and restrictions on personal activities. In
addition, parental perceived burden was a “powerful” predictor of specialty
mental health service use with perceived parental burden having the strongest
effect even after amount or type of symptomotology of the child was entered into
the model.
The level of social support of the caregiver has also been found to be
influential in the receipt of mental health services for the children under their care.
In a study conducted by Zima and colleagues (2000) of children in the child
welfare system, caregivers that sought advice from a formal source of help were
more likely to have used specialty mental health services. In another study
conducted by Geen (2003) of 13 counties in four states, it was found that in
addition to the hardships experienced by kin caregivers, kin caregivers were
unfamiliar with the workings of the child welfare system, and therefore, unable to
access community resources. The study also found that because of their age and
20
often times lack of transportation, some kin caregivers were isolated and without
a support system, whereas non-kin foster parents interact with other foster parents
through training, organizations, and associations.
Another important factor that is influential in the utilization of mental
health services by children is the caregiver’s mental health. In a study conducted
by Farmer and colleagues (1997) of children in a community sample, researchers
found that parents of youth with a psychiatric disorder or impairment who
received mental health services in any setting reported more worries, depression,
feeling of incompetence due to their children’s symptoms than did parents of non-
service using youth. In another study of a cohort of 1,071 youth in a
predominantly rural region of the southeastern U.S. (Farmer et al., 1999),
researchers found that having a primary parent with a history of psychopathology
was positively associated with service use. When psychological need of the child
was added to the model, a history of parental psychopathology continued to
increase the likelihood of service use beyond need. In a study conducted by
Timmer, Sedlar, and Urquiza (2004) of kin and nonkin caregivers in a
predominant sample of children in the child welfare system, researchers found
that caregivers who were most likely to remain and complete mental health
treatment were caregivers who showed a higher level of depressive symptoms.
Need for mental health services as evidenced by externalizing behaviors,
clinical diagnosis or scores on the Child Behavior Checklist (CBCL) has also
21
been shown to be significant in the utilization of services with children displaying
externalizing behaviors more likely to be deemed by caregivers in need of mental
health services (Alegria et al., 2004; Timmer, Sedlar & Urquiza, 2004; Shore et
al., 2002; Zima et al., 2000). Brannan and Heflinger (2005) found that children in
Mississippi and Tennessee who were Medicaid enrollees were more likely to use
services when there was an increase in externalizing behavior problems and an
increase in impairment in social functioning. In another study of children
receiving mental health services in a rural region of the southeastern U.S., Farmer
and colleagues (1999) found that children with more extensive symptomotology
were more likely to use services, and once in services, were more likely to
continue receiving services. In a study of 480 children in foster care, Leslie and
colleagues (2000) corroborated previous research findings of studies in
community samples in that children with a CBCL scores of 60 or greater (clinical
cutoff point for impairment) had a higher number of mental health visits
compared to those with a score less than 60. In a fourth study, McKay, McCadam,
and Gonzales (1996) conducted a study of children either residing with a parent or
a foster parent who sought mental health services and found that children were
more likely to be referred for services due to aggressive behavior.
Additionally, demographic factors of the caregiver such as marital status,
education, ethnicity, income, and age have been examined in relation to mental
health service use for the children under their care, with findings indicating that
22
older, single, poor, minority, and less educated caregivers are less likely to seek
services (Ehrle & Geen, 2002; Zima et al., 2000). Kazdin, Holland, and Crowley
(1997) conducted a study of 242 children and their families referred for treatment
to an outpatient clinic and found that parents who dropped out of treatment were
more likely than those who completed treatment to experience socioeconomic
disadvantage, to be from a minority group, to be younger, and to be single parents.
Educational level of the caregiver appears to be the most prominent and consistent
variable in the children’s mental health literature. Brannan and Heflinger (2005)
found that parents of children who were receiving mental health services who had
not completed high school were more likely to receive 15 fewer service
encounters than parents who had completed high school. In another study
conducted by Farmer and colleagues (1999) of children receiving mental health
services in different sectors of care, researchers found that youth with more highly
educated parents were at increased likelihood of using specialty services.
Knowledge and Perceptions of Mental Health Services
The research literature has primarily focused on need and predisposing
factors of caregivers to predict utilization of mental health services for the
children and youth under their care. However, the mediating effect that
perceptions and knowledge of mental health services plays in the utilization of
mental health services for children and youth in the child welfare system has not
been examined. This unexamined area can further contribute in explaining the
23
under utilization of mental health services by caregivers for the children under
their care, as the mental health research literature has consistently documented
that a person’s perception, knowledge, and attitudes toward mental health and
mental health services are significant in the utilization of mental health services
(Atdjian & Vega, 2005; DHHS, 2001; Alvidrez, 1999;Constantine, 2002; Cepeda-
Benito & Short, 1998; Deane & Todd, 1996; Kelly & Achter, 1995; Delphin &
Rollock, 1995; Diala et al., 2000; Nickerson, Helms & Terrell, 1994). This is an
area that deserves attention as the same perceptions and knowledge that preclude
individuals from utilizing mental health services for themselves may translate into
the possible causes for the under utilization of mental health services for the
children and youth under their care.
To date, there are no research studies in the child welfare literature that
have examined perceptions and knowledge of mental health services as they relate
to the utilization of mental health services by any type of caregiver (i.e., birth
parents, relative caregivers, or foster parents) in the child welfare system.
However, studies in the children’s mental health literature have found that
parental perceptions and knowledge of mental health services are significant in
the utilization of mental health services for their children (Flisher et al., 1997;
Pavuluri, Luk & McGee, 1996; Richardson, 2001; McKay et al., 1998; Zima et al.,
2000; Yeh et al., 2003). Primarily is the parent’s perception of mental health
services. For example, in a study conducted by Richardson (2001) of children in
24
an ethnically diverse community sample, when parents were asked about their
perceptions and knowledge of the mental health services for their children more
than half reported having doubts about the trustworthiness of the mental health
professional; stigma was a concern with parents not wanting relatives to find out
the child was receiving services; parents did not expect medication to be effective;
and more than one third were unaware of the role of the mental health
professional. In addition, African American parents were two times more likely
than Caucasian parents to expect mental health professionals to be untrustworthy
and to provide poor care, and although more African American children had been
identified with a mental health problem, parents reported being less concerned
about it than Caucasian parents. In a similar study conducted by Flisher and
colleagues (1997) results indicate that parents of children with unmet need were
more likely to believe that services would take too much time or be inconvenient
and that the youth would want to solve the problem on their own or would refuse
to attend services. Nock, Phil, and Kazdin (2001) conducted a study of a
community sample of 405 children in an outpatient treatment clinic for child
oppositional, aggressive and antisocial behavior. The primary caregiver in the
sample was identified as the birth mother (92%). Researchers examined parent
expectancies and premature termination from therapy. Results indicate that
parents who did not expect therapy to be effective and thought that therapy was
25
not relevant and too demanding experienced more stressors and obstacles to
treatment and a poorer relationship with the therapist.
Another influential factor in the utilization of mental health services are
the barriers that parents perceive. Barriers to mental health service use appear to
be a significant factor in determining whether parents enter or participate in
treatment for the children under their care. Kazdin, Holland, and Crowley (1997)
conducted a study of 242 children referred to an outpatient clinic and examined
barriers to participation in treatment. Researchers found that perceived barriers
to treatment participation such as stressors and obstacles was associated with
attending treatment, perceptions that treatment was not relevant, and poor parent-
therapist relationship were related to who dropped out of treatment prematurely.
In another study with a sub-sample of children and families from the same study,
Nock and Photos (2006) examined parental motivation to treatment adherence in
a cohort of 76 children attending an outpatient clinic for children with
oppositional, aggressive, and antisocial behavior. The majority of primary
caregivers were birth mothers (91%). Researchers were interested in measuring
parent motivation for treatment. Findings suggest that a decrease in parent
motivation increased perception of barriers to participating in treatment which in
turn lead to poor treatment adherence. In a third study, Owens and colleagues
(2002) conducted a study of a cohort of 579 children in a school based community
sample of a project designed to reduce early risk behaviors for later substance
26
abuse and affective and conduct disorders and examined perceptions and
knowledge of mental health services with different type of barriers. Researchers
examined barriers to care and parental difficulties in accessing care. Barriers to
care included: 1) structural barriers such as lack of availability of providers, long
waiting lists, lack of insurance or inadequate insurance coverage, inability to pay
for services, transportation problems, and inconvenient services; 2) barrier related
to perceptions about mental health problems such as parents’, teachers’, and
medical providers’ inability to identify children’s need for mental health problems,
and belief that the problem can be handled without treatment; and 3) barriers
related to perceptions about mental health services such as lack of trust in or
negative experience with mental health providers, lack of children’s desire to
receive help, and stigma related to receiving help. Findings indicate that more
than one third of parents with school aged children reported a barrier to mental
health services. All three barrier types were equally enforced. In addition,
parents who reported barriers perceived more difficulties with parenting their
child compared with parents who did not report any barriers. In examining barrier
enforcement, 50% of parents who reported a barrier also reported barriers to entry
into the mental health system. Parents who reported barriers to entry into the
system were less likely to report structural barriers (i.e., lack of availability of
providers, long waiting lists, lack of insurance or inadequate insurance coverage,
inability to pay for services, transportation problems, and inconvenient services)
27
and more likely to report barriers related to perceptions of mental health problems
(i.e., parents’, teachers’, and medical providers’ inability to identify children’s
need for mental health problems, and belief that the problem can be handled
without treatment).
A third influential factor is the parent’s perceived need for mental health
services for the children under their care. It is important to note that perceived
need differs from psychological symptomotology of the child such as
externalizing behaviors as evidenced from clinical diagnosis or scores on the
CBCL. Perceived need refers to the caregiver’s perception that the child is in
need of mental health services regardless of whether there is a clinical diagnosis.
In a study conducted by Zima and colleagues (2000) of 255 children in foster care
in Los Angeles County, researchers examined the help-seeking steps and service
use patterns of school aged children in foster care. The study focused on
perceived need of mental health services as identified by the child’s caregiver.
Results indicate that caregivers who detected a problem and perceived a need for
specialty mental health services were more likely to have used specialty mental
health services in the previous year than those children of caregivers who had not
detected a problem or perceived a need.
Type of Caregiver
As previously noted, approximately 54% of children in the child welfare
system are in non-relative placement (either foster home or group home) and 24%
28
of children are in a foster family home with a relative (DHHS, 2006). No data
exists as to the number of children in the child welfare system that are with a
parent (i.e., returned or remained home). However, in examining the total number
of children in the child welfare system nationwide California has a total of 18%,
the state with the highest number of children in its child welfare system. Of these
children, 24% are living with a parent (Needell, et al., 2006).
Before a discussion of the different types of caregivers in the child welfare
system is presented. A brief introduction as to the various placement settings that
are available to children once removed from their home is warranted. Once the
decision to remove a child from the home is made, the child can be placed in a
number of settings. The most common and less restrictive settings are in foster
family homes (non-relative) and kin care (relative). Other alternative placement
settings are group homes and residential facilities or institutions, which are more
restrictive and generally provide a comprehensive array of services. It is
important to note, however, that not all children are removed from their homes
once a child welfare case is open. Often, children are left with a non-offending
parent when the offending parent is removed from the home, either through
incarceration or court mandates. Throughout the course of a case, the main goal
is to have children reunified with their parent(s) if they have been removed from
the home; thus, some children with open child welfare cases are placed in their
home, making an additional placement setting of home of parent (HOP).
29
Caregivers of foster family homes and kin care are given the same
responsibilities, to provide the child with a safe and nurturing environment in the
absence of the parent(s). It is then that these substitute care providers become
“gatekeepers” to a child’s well being. Both are provided with monetary
compensation for the care of the child, and both are supervised by the child
welfare system. However, there are significant differences in the two placement
settings, which make it substantially crucial in a child receiving mental health
services.
Since most children will experience trauma due to the abuse and
separation from their parents, kin care has often been thought of minimizing the
trauma, as children are placed in a familiar environment. Although, children may
need to adjust to their new surroundings, being with a family member was thought
to ease the transition. Foster care on the other hand, may exacerbate the trauma,
as children need to deal with the separation from their parents, possibly from their
siblings, and adjust to new caregivers that may have children of their own or other
foster care children in the home. Due to this belief and to various system and
legislature initiatives, there has been a trend in recent years to place children in
kin foster care (Grogran-Kaylor, 2000; Berrick, 1998; Rittner, 1995).
There has been some debate as to whether children in kin care experience
and/or exhibit less behavior or emotional problems than children placed in non-
kin placement settings. However, regardless of whether children are placed in kin
30
foster care or nonkin foster care, there appears to be a proportional need for
mental health services, as studies have found that children placed in kin care are
just as likely than children placed in non-kin care to experience mental health
problems (Landsverk, Garland, and Leslie, 2002; Pilowsky, 1995).
As previously noted, few studies have focused on the caregiver and the
utilization of mental health services for the children and youth under their care in
the child welfare literature. Of these studies, most have focused on kin care and
have found that children in kin care are less likely to receive mental health
services than children placed in non-kin care (Timmer, Sedlar, & Urquiza, 2004;
McMillan et al., 2004; James et al., 2004; Leslie et al., 2000; Blumberg et al.,
1996). For example, in a study conducted by Leslie and colleagues (2000) of
children in the foster care system in San Diego, CA, researchers found that 41.5%
of children had at least one outpatient visit while in out of home care during an 18
month period. Children in non-kin care averaged twice as many visits during the
18 month period as children in kin care. One explanation for this may be partly
due to kin caregivers experiencing unusual hardships. On average, kin caregivers
are older than non-kin caregivers with up to 21% being over 60, have health or
mental problems, are experiencing financial hardship, are single and less educated
(Erhle and Geen, 2002). It is also important to note that some kin caregivers
avoid any involvement with the child welfare system and are generally in a crisis
situation when asking for help (Geen, 2003).
31
Although studies have consistently documented that non-kin caregivers
are more likely to seek and receive mental health services for children under their
care, it is unclear whether children in non-kin care in fact display more behavior
problems and therefore their caregivers are more likely to seek services or
whether kin foster care parents perceive the children under their care as having
fewer problems. It is possible that non-kin caregivers are less tolerant of their
children’s behavior and therefore more likely to report their children’s behavior
problems to caseworkers and increase the likelihood of receiving services
(Timmer, Sedlar, & Urquiza, 2004). In a study conducted by Timmer, Sedlar, and
Urquiza (2000), researchers found that foster parents tended to rate children under
their care with higher levels of externalizing behaviors and more frequent
behavior problems than kin caregivers and thus were more likely to seek services.
In another study that used two national datasets (i.e., National Survey of Child
and Adolescent Well-being and Long Term Foster Care), researchers found that
children placed in foster homes evidenced higher levels of behavioral problems
than children placed in kinship care or those in birth family homes as reported by
their caregivers. However, researchers found that from the perspective of
teachers, children in kin care evidence higher behavioral problems than children
placed in nonkin foster homes (Rosenthal & Curiel, 2006).
While previous research studies have found an underutilization of mental
health services by kin caregivers for the children under their care, these studies
32
did not examine the factors that may contribute to the differences between kin and
non-kin caregivers in service use. Only one study to date has examined the
differences between non-kin and kin caregivers. Timmer, Sedlar, and Urquiza
(2004) conducted a study of 259 kin and non-kin foster parents and their foster
children who had been referred for Parent-Child Interaction Therapy (PCIT).
Children were referred to the clinic primarily due to trouble adjusting to
placement and/or were displaying acting out and disruptive behaviors. The
researchers used the framework of Social Exchange Theory which proposes that
relationships are maintained when the benefits of the relationship to the individual
outweigh the costs. In this case, foster parents who were willing to take the time
to participate in therapy with their children were more likely invested in making
the placement work as were kin caregivers. Results showed that kin caregivers
were significantly more likely to complete treatment than non-kin caregivers,
suggesting that kin caregivers were possibly more invested in the children under
their care than non-kin caregivers. It is important to note, however, that this study
examine entry and completion of treatment as part of a dyad technique, where the
caregiver and child participated in treatment. Previous research studies have
examined the utilization of mental health services by only the child.
To date, no study has examined the role that birth parents in the child
welfare system play in the utilization of mental health services for their children.
However, there is evidence to suggest that birth parents also play an important
33
part in the utilization of mental health services for children in the child welfare
system. In a national study conducted by Burns and colleagues (2004) of children
and youth investigated for maltreatment, researchers found that children who
remained at home versus those who were placed out of home were significantly
less likely to have received mental health services. Although, researchers did not
specifically identify children placed with a birth parent, approximately 90% of the
sample was living at home with their permanent primary caregiver. In another
study conducted by Swenson, Brown, and Sheidow (2003) of physically abused
children either in foster care or with a parent, researchers found that children were
more likely to be receiving mental health treatment if the maltreating caregiver
was not living in the household. McKay, McCadam, and Gonzales (1996)
conducted a study of a cohort of 450 children residing either with a parent or a
foster parent. Researchers were interested in examining the number of children
who actually attended their scheduled intake appointment for mental health
services. Researchers found that children who were living with their birth
mothers were the least likely to attend their initial intake appointment, whereas
children being parented by a foster caregiver were more likely to attend their
scheduled appointments. Becker, Jordan, and Larsen (2006) also found that
children were more likely to attend mental health services before placement than
after they were returned to their parents. These findings indicate that type of
34
caregiver may act as a moderator in examining the caregiver’s influence on
mental health service use for the children and youth under their care.
Preliminary Studies
The present study builds on previous work done by the Child and
Adolescent Services Research Center (CASRC) in San Diego, California which
has for over a decade examined mental health service use patterns in relation to
child welfare system variables. Three research studies, which used a sample of
children from five public service sectors, are particularly relevant to this project.
The present study used data from the same sample, but only analyzed one public
service sector, the child welfare system. While the three research studies had
different foci, they all shed light on the variables of interest to this study.
Yeh and colleagues (2003) examined parental endorsement of barriers to
care among a sample of 1,338 at risk youth with mental health needs across five
sectors of care. The researchers were particularly interested in examining the
racial/ethnic differences among parents in the enforcement of barriers to mental
health service use. Results corroborated previous findings of ethnic minority
children having higher levels of unmet need compared to Caucasian children.
Results also indicated that parents of ethnic minority youth reported fewer
barriers to service use than Non-Hispanic Whites with a persistent pattern across
barrier types. Researchers also found barrier endorsement was unrelated to the
child’s unmet need for mental health services. Regarding perceptions and
35
knowledge of mental health services by their caregivers, findings were
unexpected in that parents who reported fewer barriers to service use were less
likely to utilize mental health services. The researchers recommended that “future
research on barriers to mental health care should consider cultural influences upon
barrier endorsement” (p.73) as minority groups from countries that generally do
not provide mental health services may perceive any type of service adequate or
some may not have the desire to utilize services due to cultural beliefs or biases,
and therefore, do not perceive any barriers. This indicates that ethnicity and the
child’s need for services does not explain parental endorsement of barriers,
suggesting that other individual factors related to the caregiver may explain help
seeking of services and thus, deserve further examination.
In a follow up sample, Yeh and colleagues (2004) examined racial/ethnic
differences in parental beliefs about the causes of their children’s mental health
problems. The same sample of children used in the previous study (2003) was
used for this study. Eleven etiological belief areas were identified with five
biopsychosocial (i.e., physical causes, personality, relational issues, familial
issues, and trauma), four sociological (i.e., friends, American culture, prejudice,
and economic problems), and two related to spiritual/disharmony issues (i.e.,
spiritual causes and nature disharmony). Respondents were asked to answer
yes/no to a question of whether they believed their child’s problems were caused
by issues in each of the eleven areas. The researchers found that parents of ethnic
36
minority children were less likely to endorse etiologies consistent with
biopsychosocial beliefs about mental illness and more likely to endorse etiologies
consistent with sociological beliefs, such as American culture and experience of
prejudice. Although, researchers did not examine mental health service use,
findings suggest that parent’s help seeking of mental health services for their
children may be significantly related to their beliefs about the causes of mental
illness.
A third study by Yeh et al., (2005) investigated the role of parental beliefs
about the causes of child problems in predicting later mental health service use.
The researchers used the same sample of children previously used in the two
studies presented (2003 & 2004) with the present study using the same measure of
eleven etiological belief areas (2004). Results indicate that Asian Pacific
Islanders and Latino youth were less likely than non-Hispanic Whites to use
specialty mental health services. Results also indicate that parental beliefs of the
child’s problems that were due to physical causes and trauma had a greater
likelihood of mental health service use while parental beliefs of the child’s
problem that were due to friends had a lesser likelihood of service use. This lends
support to the notion that parental beliefs about mental illness are related to
mental health service use, and as such, this area deserves further attention.
It is important to note that the three studies presented used a much larger
sample of children than the present study and consisted of five service sectors of
37
care. Only 23% of the sample in the three studies was from the child welfare
system and the majority of caregivers (73%) were birth parents, thus lending
support to the studies presented in the literature review of children in community
samples.
Conceptual Model

The guiding conceptual framework for this study is drawn from the
Gateway Provider Model previously discussed in the review of the literature. The
original model focuses on the provider (gatekeeper) associated with the child, and
the provider’s perceptions and knowledge of service resources and mental illness
in the decision to provide services. In the adaptation of the model used in this
study termed, the Gateway Caregiver Model, the caregiver is seen as the gateway
to service use. The model was developed to examine the caregiver as gatekeeper
of mental health service utilization for children in the child welfare system. The
model proposes that need and predisposing factors and perceptions and
knowledge of mental health services are influential factors in the utilization of
mental health services by caregivers for the children under their care. The model
further proposes that type of caregiver is also influential (Figure 2).
Specifically, the model proposes that perceptions and knowledge of
mental health services of the caregiver mediates the relationship between need
and predisposing factors and mental health service use by caregivers for the
children under their care. Need factors refer to the child’s psychological
functioning. Predisposing factors refer to risk and protective factors, such as
social support, level of stress, and depression of the caregiver. Perceptions and
knowledge of mental health services refer to attitudes toward mental health
services, perceptions of barriers to service use, and perceived need. The model
further proposes that type of caregiver (i.e., biological parent, relative caregiver,
and foster parent) moderates the mediating effects of perceptions and knowledge
of mental health services.
Figure 2. The Gateway Caregiver Model

38
39
Summary of the Literature
Consistent with the research literature, this study focuses on the caregiver
as a vital source in the identification and utilization of mental health services for
the children under their care. Little is known as to the role that birth parents,
relative caregivers, and foster parents play in the utilization of mental health
services for the children in the child welfare system. This study is intended to
shed light on the influence that caregivers’ have on a child’s mental health service
use, and therefore, have implications for improving service delivery and mental
health outcomes for the large population of underserved children in the child
welfare system.

40
CHAPTER III: RESEARCH METHODS

The current study builds on studies conducted by the Child and
Adolescent Services Research Center (CASRC) in San Diego, California. This
study extends the Center’s previous research studies of a cohort of 1,715 youths
who were active in one or more public service sectors of care during the latter half
of the 1996-1997 fiscal year in San Diego County and who were part of the
Patterns of Youth Mental Health Care in Public Service Systems study (POC –
Patterns of Care). The current study utilized secondary data analyses with only
one public service sector, the child welfare system.
Research Questions and Hypotheses
Due to the gap in the research knowledge on perceptions and knowledge
of mental health services of caregivers in the child welfare system, this study
aimed to provide the groundwork necessary for this area of research. In line with
the specific aims of this study the following three main research questions were
proposed under the theoretical framework of associations and mediation-
moderation effects.
As previously noted, need factors refer to the child’s psychological
functioning. Predisposing factors refer to risk and protective factors, such as
social support, level of stress, and depression of the caregiver. Perceptions and
knowledge of mental health services refer to attitudes toward mental health
41
services, perceptions of barriers to service use and perceived need, and caregiver
type includes biological parents, relative caregivers, and foster parents.
Aim 1. To examine the impact of need and predisposing factors on mental health
service use by caregivers for the children and youth under their care.
Research Question 1: Do need and predisposing factors impact mental health
service use of caregivers for the children and youth under their care, adjusting for
covariates? (Marital status, education, ethnicity, income, and age of the caregiver
will be used as covariates).
Hypothesis 1A: The higher the need factors the more likely that caregivers
will use mental health services for the children under their care.
Hypothesis 1B: The higher the predisposing factors the more likely that
caregivers will use mental health services for the children under their care.
Aim 2. To determine whether caregivers’ perceptions and knowledge of mental
health services mediates the relationships between need and predisposing
factors and mental health service use by caregivers for the children and
youth under their care.
The following question examined the mediating effect of perceptions and
knowledge of mental health services on need and predisposing factors and mental
health service use.
Research Question 2: Do need and predisposing factors influence perceptions and
knowledge of mental health services of caregivers, which in turn, influence
42
mental health service use by caregivers for the children and youth under their
care? Specific hypotheses are formulated corresponding to the two separate paths
of the mediation framework. The first path is from the need and predisposing
factors to perceptions and knowledge of mental health services. The second path
is from perceptions and knowledge of mental health services to mental health
service use.
Hypothesis 2AI: Caregivers of children with a high need will have negative
perceptions and knowledge of mental health services.
Hypothesis 2AII: Caregivers with a high predisposing factor will have
negative perceptions and knowledge of mental health services.
Hypothesis 2B: Caregivers with positive perceptions and knowledge of mental
health services will be more likely to use mental health services for the
children under their care.
Aim 3. To investigate whether the paths from need and predisposing factors to
perceptions and knowledge of mental health services and from need and
predisposing factors to mental health service use by caregivers for the
children and youth under their care differ among caregiver type.
The final question examined the moderator effect of caregiver type on the model’s
paths.
Research Question 3: Is there a difference among caregiver type in the paths from
need and predisposing factors and perceptions and knowledge of mental health
43
services and from need and predisposing factors and mental health service use by
caregivers for the children and youth under their care?
Hypothesis 3AI: Children with a high need factor will have biological parents
and relative caregivers with negative perceptions and knowledge of mental
health services.
Hypothesis 3AII: Children with a high need factor will have: a) foster parents
more likely to use mental health services for the children under their care and,
b) relative caregivers less likely to use mental health services for the children
under their care.
Hypothesis 3BI: Biological parents and relative caregivers’ with a high
predisposing factor will have negative perceptions and knowledge of mental
health services.
Hypothesis 3BII: The higher the predisposing factors the more likely that
biological parents and relative caregivers’ will use mental health services for
the children under their care.
Data Source
The Patterns of Care study consisted of participants randomly selected
from an enumeration of all youths active in one or more of five San Diego County
public sectors of care (N = 12,662) during fiscal year 1996-1997. Public sectors
of care included: 1) child welfare – records indicated that the youth was a
dependent of the juvenile court because of protective issues with 33% of the total
44
sample; 2) alcohol and drug abuse – youth received any county contracted service
from an alcohol/drug treatment provider with 4% of the total sample; 3) juvenile
justice – youth was under jurisdiction of the juvenile court as a ward because of a
true finding of a criminal charge with 30% of the total sample; 4) mental health –
youth received any service through a county mental health service program with
55% of the total sample; and 5) public schools special education services for
youth with SED – youth was designated having the education disability condition
“serious emotional disturbance” as defined by federal law with 16% of the total
sample. It is important to note that youth active in multiple sectors were counted
in each sector and therefore there was an overlap in children when examining the
percentages in each sector. Of the 12,662, a total of 3,417 youth were randomly
selected for recruitment. The sample was stratified by service sector affiliation,
race/ethnicity, and level of restrictiveness of care. The final total sample
consisted of 1,715 youth. One of the aims of the original study was to examine
racial/ethnic differences among service sectors, so the sample was ethnically
diverse and included non-Hispanic whites, Latinos, African American, Asian
Americans/Pacific Islanders, and other. Race/ethnicity was self-identified for
youth ages 11 years and older and was parent identified for children ages 6-10
years. Ages ranged between 6-17 years. Two-thirds of the participants were male.
The mean age was 13.7 years (SD=3.3). Most of the parent/caregivers were birth
parents (72%). Others included adoptive or foster parents, step-parents, and a
45
small number of professional caregivers. The majority of the caregivers in the
sample (94%) were women. Using a range of established standardized measures,
parents/caregivers and youth were interviewed individually about the youth’s
mental health use, needs, and other factors potentially associated with mental
health service use. The POC dataset is the only dataset to examine perceptions
and knowledge of mental health services by caregivers. Data were collected at
baseline (on average, one year after the youth’s presence in a service sector was
recorded), six, twelve, eighteen and twenty-four months. The study attrition rate
was only 9% from baseline to twenty-four months.
The current study consists of only the child welfare data where records
were identified as the youth being a dependent of the juvenile court because of
protective issues. Of the total POC sample, 23% (N=430) of the children and
youth were identified as having an active case in the child welfare sector.
Children in the current study also had active cases in other sectors of care.
Although there is no data as to the total number of children who overlapped in
several sectors of care, the following percentages are based on the number of
children with an active child welfare case who were also identified as having an
active case in other public service sectors, not mutually exclusive. The majority
of children in the current study also had an active case in the mental health sector
with 45%, followed by the school SED sector with 13%, 3% in the alcohol and
drug abuse sector, and 1% in the juvenile justice sector.
46
Sample Characteristics
Due to the focus of the present study being on the caregiver, the data
focuses on the caregivers with the exception of mental health service utilization
which was taken from use of services by the child under the caregiver’s custody.
Table 1 presents descriptive data by caregiver type (group) and on the total
sample of 430 caregivers for the current study. For the total sample the mean age
of caregivers was 44 years (SD=12). Birth parents were the youngest with 37
years (SD=7.85) and relative caregivers the oldest with 51 years (SD=13.04).
Foster parents were somewhere in between with 44 years (SD=11.86). The
majority of caregivers for the total sample were Caucasian with 40%, African
American and Latino were slightly different with 24% and 23% respectively, and
Asian/other with the smallest percentage at 10%. When caregiver type was
examined there were slight differences from the total sample with Caucasians
being the majority with most caregiver types, birth parents 44%, foster parents
41% and relative caregivers 33%. African Americans were the least among birth
parents with 14%, followed by foster parents with 28% and relative caregivers
with the majority at 34%. Latinos were equal among birth parents and relative
caregivers with 26%. Foster parents were the smallest group with 16% among
Latinos. Since Asians were the smallest group in the total sample, they were also
the smallest group among all caregiver types with birth parents having 16%,
relative caregivers 7%, and foster parents with 6%. The majority of caregivers
47
had a high school education with 47%, caregivers with a college education
followed with 29%, and caregivers with no degree with 24% of the sample.
Among caregivers, foster parents were the most educated with 44% having a high
school and college degree and only 12% having no degree. Relative caregivers
followed with 32% having a college degree, 37% with a high school degree, and
31% with no degree. The majority of birth parents had a high school degree with
55% followed by no degree with 31%, and a college degree with 13%. Most of
the caregivers in this sample were not married with 54% and 43% being married.
Foster parents had the highest number married with 56%, followed by relative
caregivers with 48%, and biological parents with 27%. Income appeared to be
almost evenly distributed with 24% of caregivers having an income of >$13,000,
27% having an income of <$25,000, 19% having an income of <$45,000, and
23% with an income of >$45,000. Among caregiver type, foster parents had the
highest income with 43% having an income of >$45,000 and 27% with an income
of <$45,000. Birth parents had the lowest income with 50% having an income of
<$13,000. Relative caregivers were in the median range with the majority having
an income of <$25,000 with 36%.

Table 1. Demographic Characteristics for Caregivers (N = 430)

Characteristics

Caregiver Type

Birth Relative Foster
Parent Caregiver Parent
(n=167) (n=110) (n=153)

Total Sample

N (%) or Mean
(SD)

Age

37.15 (7.85)

50.79 (13.04)

46.54(10.83)

43.93 (11.86)
Ethnicity
Caucasian
African American
Latino
Asian/Other

74 (44.3)
23 (13.8)
43 (25.7)
26 (15.6)

36 (32.7)
37 (33.6)
29 (26.4)
8 (7.3)

63 (41.2)
42 (27.5)
25 (16.3)
10 (6.5)

173 (40.2)
102 (23.7)
97 (22.6)
44 (10.2)
Education
No Degree
High School
College

52 (31.1)
92 (55.1)
21 (12.6)

34 (30.9)
41 (37.3)
35 (31.8)

18 (11.8)
67 (43.8)
67 (43.8)

104 (24.2)
200 (46.5)
123 (28.6)
Marriage
Yes
No

45 (26.9)
122 (73.1)

53 (48.2)
57 (51.8)

86 (56.2)
67 (43.8)

184 (42.8)
230 (53.5)
Income
<$13,000
<$25,000
<$45,000
>$45,000

83 (49.7)
52 (31.1)
15 (9.0)
8 (5.4)

15 (13.6)
39 (35.5)
25 (22.7)
25 (22.7)

6 (3.9)
24 (15.7)
41 (26.8)
65 (42.5)

104 (24.2)
115 (26.7)
81 (18.8)
98 (22.8)

Variables and Measures
In order to assess the influence between need and predisposing factors as
well as knowledge and perceptions of mental health services, variables in these
domains were used from baseline data. Mental health service use was used as the
outcome variable and taken from report of service use at twelve months.
Predictor Variables/Constructs
1) Need Factor: This variable measured the youth’s psychological
symptomotology as reported by parent/caregiver. Data for this variable were
collected using the Child Behavior Checklist (Achenbach, 1991). The Child
48
49
Behavior Checklist (CBCL) is a parent/caregiver report questionnaire that
provides age normed comparisons of behavior/emotional problems for
children ages 2 to 18 with established reliability and validity for both English
and Spanish speaking samples (Achenbach, 1991a, 1991b). The CBCL
produces overall total problem behavior scores and broadband indices of
internalizing behavior and externalizing behavior problems occurring during
the previous six months. Total competency, internalizing behavior, and
externalizing behavior were used to measure need for this study. Total
competency contains sections addressing the area of social competence in
order to determine which reported competencies discriminate between those
children who are adapting successfully and those who are not. The CBCL/4-
18 contains 20 competence items grouped into 3 scales (Activities, Social, and
School). Total competency is based on the higher the score the better the
child’s competence overall. Internalizing problems combines the Social
Withdrawal, Somatic Complaints, and Anxiety/Depression scales, while
Externalizing combines the Delinquents Behavior and Aggressive Behavior
scales. T scores on internalizing and externalizing problems less than 60 are
considered in the normal range, 60-63 represent borderline scores, and scores
greater than 63 are in the clinical range. The Cronbach’s alpha for the entire
scale for the current sample was .95.

50
2) Predisposing Factors (Risk & Protective Factors):
A) Stress: Data for this variable were collected using the Caregiver Strain
Questionnaire (Brannan et al., 1997) to measure caregiver strain. The
Caregiver Strain Questionnaire is a 21 item continuous scale designed to
measure the impact of caring for a child with emotional and behavioral
problems in six areas: economic burden, impact on family relations, disruption
of family activities, impact on psychological adjustment of family members,
stigma, anger, and worry/guilt. Responses are rated on a five point Likert
scale and measure each area during the past six months. Scores were the sum
of all responses divided by the number of items, and had a potential range of
1-5 with the higher the score the greater the strain. The Cronbach’s alpha for
the sample was .91.
B) Depression: Data for this variable were collected using the Center for
Epidemiological Studies – Depression Scale (Radloff, 1977). The CES-D was
completed by parents/caregivers to assess parental/caregiver depressive
symptomotology. The questionnaire consists of 20 items that asks
participants how often they experienced symptoms of depression during the
past week. The maximum score is 30; high scores on the CES-D indicate high
levels of distress. A score ≥ 16 suggests a clinically significant level of
psychological distress. Responses are rated on a four point Likert scale. The
51
CES-D has established reliability and validity (Radloff & Teri, 1986). The
Cronbach’s alpha for the sample was .92.
C) Social Support: Data for this variable were collected using the Social
Provision Scale (Cutrona & Russell, 1987). The Social Provision Scale is a
12 item continuous scale that measures global perceived social support. The
questions ask respondents’ perception of available social support in their lives
in general and are not tied to a specific time period. Responses are rated on a
five point Likert scale. Scores were the sum of all responses divided by the
number of items, and had a potential range of 1-5 with higher scores
indicating a greater social support. Cronbach’s alpha for this sample was .83.
Covariates
Demographics of the caregiver:
1) Age – Continuous variable, reflecting the caregiver’s age at the time of
baseline interview
2) Ethnicity – Categorical variable (Caucasian, African American, Latino,
and Asian/Other)
3) Marital Status – Dichotomized variable (married, not married)
4) Income – Categorical variable - total family income was reported by
parent/caregiver on an incremental scale of annual income from ≤$1,000
to ≥$200,000. For this study income was divided as follows: 1) ≤$13,000,
2) ≤$25,000, 3) ≤$45,000 and 4) <$45,000.
52
5) Education – Categorical variable (no degree, high school, and college)
Gender was not used as a covariate as 94% of the sample was female.
Mediating Variable
Perceptions and Knowledge of Mental Health Services: Data were collected using
the Measurement of Parental Endorsement of Barriers developed by Yeh and
colleagues from the Child and Adolescent Services Research Center (CASRC) in
San Diego, California. The questionnaire was developed as a supplement to the
Services Assessment of Children and Adolescents (SACA) structured interview to
identify potential barriers to services as suggested by the literature, expert
consultation, and prior research (Yeh et al., 2003).
The questionnaire consisted of a total of 54 items that were used to create
eight scales classified by Yeh and colleagues (2003) as representing different
service barrier areas: 1) content of services (6 items relating to what might occur
during services; α=.76 – for the current sample α=.77 ); 2) helpfulness of services
(8 items about the perceived effectiveness of services; α=.73 – for the current
sample α=.75); 3) provider characteristics (5 items relating to the respondent’s
beliefs about service providers; α=.69 – for the current sample α=.73); 4) effects
of services (10 items about the potential influence of services upon the child’s
family or social situation; α=.73 – for the current sample α=.74); 5)
economic/financial constrains (5 items involving finances or insurance issues;
α=.69 – for the current sample α=.62); 6) accessibility (14 items delineating
53
practical problems related to service use; α=.78 – for the current sample α=.82);
7) language problems, (2 items addressing language barriers; α=.79 – for the
current sample α=.74) and; 8) lack of need for services (4 items related to the
child’s need for services; α=.60 – for the current sample α=.62). Items were
answered yes/no.
Moderating Variable
Caregiver Type: Three caregiver types were identified for use in this study: 1)
birth parent consisted of only parents who were biologically related to the youth
(n=167), 2) relative caregiver (n=112) included all relatives (grandparents, uncles,
siblings, and step-parent), and 3) foster parent (n=121) consisted of solely those
identified as foster parents.
Outcome Variable/Construct
Mental Health Service Use: Data for utilization of mental health services were
measured by the National Institute of Mental Health (NIMH) Service Assessment
of Children and Adolescents (SACA; Horowitz et al., 2001).
This structured interview consists of both parent/caregiver and youth
reports. The interview assesses utilization of mental health and substance abuse
services by the youth. It obtains lifetime and past year use of school based,
outpatient and inpatient mental health, and substance abuse services. Mental
health service utilization in this study is defined as any use of specialty outpatient
mental health or inpatient mental health services during the past year. Since past
54
year use was examined, the SACA was examined at twelve month follow-up.
The variable was dichotomous to include use of specialty outpatient/inpatient
mental health service or no use of any specialty mental health service.
Data Analysis
Two data analyses techniques were used in this study, logistic regression
and structural equation modeling. Logistic regression was employed to examine
the paths between need and predisposing factors and mental health service use by
caregivers by the children under their care for the first research question. Logistic
regression was found to be the best data analysis technique as the dependent
variable is binary and the use of latent variables was not examined in this question.
Structural equation modeling (SEM) was used to test the mediating and
moderating effect on the conceptual model for the second and third research
questions (Figure 2). SEM was chosen as the best method of data analysis
because it allows complete and simultaneous tests of all the relationships between
one or more independent and dependent variables and has an advantage over a
series of regression equations (Ullman, 2001). SEM also gives an overall
indication of the fit between the proposed theory and data (Aaron & Aron, 1999).
In addition, the relationships that are tested by SEM are free of measurement error,
since measurement error is estimated and removed (Ullman, 2001). The use of
several fit indices also allows for the assessment of several qualities of the model.
55
For structural equation modeling (SEM), data were analyzed in two parts:
measurement models and structural models. Relations between observed
variables (indicators) and their underlying factors (latent variables) are the focus
of the measurement model (Byrne, 1994). Kline (1998) asserts that researchers
should test the measurement model first, and if the fit of the measurement model
is found acceptable, then to proceed to testing the structural model by comparing
the fit with that of different structural models. The measurement model in this
study consisted of latent variables which are the three variables in the conceptual
model (i.e., need factor, predisposing factor, and perceptions and knowledge of
mental health services). Like factor analysis, SEM utilizes latent variables, each
of which is a combination of specific measured variables. Thus, a latent variable
is not directly measured, but stands for a weighted combination of several
variables that it makes up. Each latent variable in this study consisted of
indicators which were the sub-scale scores. The indicators used for the need
factor consisted of internalizing and externalizing behaviors and total competency.
The indicators for the predisposing factor consisted of stress, depression, and
social support. The indicators for the perceptions and knowledge factor were to
include perceived need, attitudes toward service use, and perceptions of barriers
to services use. However, after consultation with the primary author, Dr. May
Yeh, of the Measurement of Parental Endorsement of Barriers from the Child and
Adolescent Services Research Center (CASRC) in San Diego, California, it was
56
determined that the most appropriate method was to use the measure as it had
been previously used in Yeh and colleagues (2003) research study. Therefore, the
eight scales, content of services, helpfulness of services, provider characteristics,
effects of services, economic/financial constrains, accessibility, language
problems, and perceived need, previously used in research studies were used as
indicators for the perception and knowledge of mental health services factor.
The following represents an outline of the analytic strategy used for each
research question and hypothesis: All hypotheses in the three research questions
were evaluated in the following statistical methods: 1) hierarchal logistic
regressions were used for hypotheses 1A and 1B in the first research question, 2)
path analysis were used to examine the hypothesized mediating effects
(hypotheses 2AI, 2AII, and 2B) in research question two, and 3) a multiple group
approach was added to the path analysis in order to test hypothesized regressions
(hypothesis 3AI, 3AII, 3BI, and 3BII) among caregiver type in research question
three. Missing data were handled using the full information maximum likelihood
estimation.
Aim 1 and Hypotheses 1A & 1B: To examine whether need and
predisposing factors impact mental health service use of caregivers for the
children and youth under their care, adjusting for covariates (marital status,
education, ethnicity, income, and age of the caregiver). Hypotheses 1A and 1B
57
were separately tested using hierarchical logistic regressions with the five
covariates controlled.
Aim 2 and Hypotheses 2AI/2AII & 2B: Aim 2 examined whether
perceptions and knowledge of mental health services mediated the relationship
between need and predisposing factors and mental health service use. A path
model was constructed in the following order: A path model using need factor,
perceptions and knowledge of mental health services, and mental health service
use was examined to test the mediation effect of perceptions and knowledge of
mental health services. Predisposing factors were then added to the path model
and mixed mediation effect of perceptions and knowledge of mental health
services on the relationships between need and predisposing factors on mental
health service use were tested.
Mediation effects were evaluated according to four standard criteria
(Baron & Kenny, 1986; MacKinnon & Dwyer, 1993; MacKinnon, Krull, &
Lockwood, 2000). Criterion one, the independent variable must predict the
mediator. Criterion two, the mediator must predict the outcome variable when the
independent variable is controlled. The mediation effect is calculated as the
product of the two regression coefficients from the independent variable to the
mediator in criterion one and from mediator to the outcome variable adjusting for
the independent variable in criterion two. Criterion three, the independent
variable must have a significant effect on the outcome variable that is expressed
58
as total effect. Criterion four, the effect of the independent variable on the
dependent variable shrinks upon the addition of the mediator to the model (Sobel,
1982).
Aim 3 and Hypotheses 3AI/3AII & 3BI/3BII: Aim 3 examined whether
there were differences in caregiver type in the paths from need and predisposing
factors and perceptions and knowledge of mental health services and from need
and predisposing factors and mental health service use by caregivers for the
children under their care. Hypothesis 3AI – 3BII aimed to evaluate the moderator
effect of caregiver type on the four paths. This was tested using multiple group
structural equation modeling. Multiple group SEM was conducted by means of
comparing the following two models: 1) a base model was set up to be the final
path model. All parameters in the base model were constrained to be equal across
the three caregiver types, and 2) three group difference models were set up to free
up the four path coefficients one by one across the three caregiver types. While
each path was freed, all other parameters in the model were constrained to be
equal across the three caregiver types. Finally, the base model and the three
group difference models were compared in terms of their goodness of fit indices.
To determine the best fit model among the four models, the three group difference
models were evaluated against the base model using chi-square difference
statistics. Once the final model was selected with acceptable model fit index
values, the significance test of the four path coefficients was evaluated using z
tests. Significant path coefficients, which varied among the three caregiver types
in the final group difference model, indicated the significant moderator effect of
caregiver type.
SPSS 15 was used for the logistic regression analyses and AMOS 7.0 to
conduct all SEM analyses. In addition, in order to determine the fit of a model in
SEM several fit indices need to be assessed. Using multiple indices insures that
the fit of the model is not dependent on just a few features of the model, but
evaluates several different aspects of the model. Each of the indices assesses a
different aspect of fit including goodness of fit, badness of fit, the fit of the model
to the sample, and comparisons of the sample and population to the best and worst
fitting models (Kline, 2005). For this study, the following fit indices were used to
evaluate the fit of a model: x
2
statistic, Comparative Fit Index (CFI), and Root
Mean Square Error of Approximation (RMSEA). The x
2
statistic is a “badness of
fit” index because the higher its value, the worse the model’s correspondence to
the data. As such, the x
2
statistic needed to be non-significant (Kline, 2005).
The Comparative Fit Index (CFI) assesses the relative improvement in fit of the
researcher’s model compared with a baseline model. A rule of thumb for the CFI
is that values greater than .90 may indicate reasonably good fit of the researcher’s
model (Hu & Bentler, 1999). The Root Mean Square Error of Approximation
(RMSEA) is a “badness of fit” index in that a value of zero indicates the best fit
and higher values indicate the worse fit. The rule of thumb is that RMSEA ≤ .05
59
60
indicates close approximate fit, values between .05 and .08 suggest reasonable
error of approximation and RMSEA ≥ .10 suggest poor fit (Browne & Cudeck,
1993).
Power Analysis
The statistical power involved in structural equation modeling (SEM) is
quite complicated compared to that of regression modeling (Bentler & Chou,
1987). Several factors needed to be considered in computing the power of
goodness-of-fit statistics to evaluate the appropriateness of a SEM model. These
factors were: the sample size (N), the selected significance level ( α), the size of
the model as determined by the degree of freedom (df), test statistics of the null
hypothesis (F0), and test statistic of the alternative hypothesis (Fa). The statistical
power for fitting a model increases with the sample sizes and the size of the
model. Impacts of test statistics of alternative hypothesis on power have been
discussed in Cho and Bentler (1990). Based on the paper and the power analysis
program provided by MacCallum, Browne, and Sugawara (1996), the statistical
power for SEM models can be estimated with the hypothesis-testing framework
for Root-Mean- Square Error of Approximation (RMSEA) as a vehicle. With the
α=0.05, null hypothesis RMSEA of 0.05 and alternative hypothesis RMSEA of
0.10, our current sample size (N=430) given the size of the proposed model (i.e.,
the degree of freedom) was sufficient enough to reach 80% statistical power.
According to the recent empirical simulations by Fritz and MacKinnon (in press
61
in Psychological Science), the sample size of 430 was adequate to detect the
mediation effect when Cohen’s small to medium effect size were assumed in
either of the two mediation paths.

62
CHAPTER IV: RESULTS

T-tests and chi-square tests were conducted as preliminary analyses to
determine if the variables of interests were associated with the outcome variable
of mental health service use. Next, hierarchal logistic regression results are
presented, followed by the mediating effect analyses using structural equation
modeling. Since one of the main variables of interest was caregiver type,
preliminary analyses by caregiver groups using chi-square and ANOVA statistical
techniques are presented before answering research question three of the
moderating effect by caregiver type.
Preliminary Analyses
Bivariate analyses were conducted to examine differences between mental
health service use and no mental health service use by caregivers for the children
under their care with the caregiver’s demographic characteristics, need and
predisposing factors, and caregiver type. Results provided a glimpse as to the
associations with the variables of interest and mental health service use. For the
need factor which includes total competency, internalizing and externalizing
behaviors using t-tests there were differences found between children who used
mental health services and those that did not. Children who did not use services
had significantly lower scores than children who used mental health services.
For the predisposing factor which includes the caregiver’s level of stress, level of
depression, and level of social support, t-tests were also used and only level of
63
stress was statistically significant. Children who did not use mental health
services had caregivers with a lower level of stress than children who used mental
health services. Differences were also found with age of the caregiver and
caregiver type between mental health service use and no service use by caregivers
for the children under their care. Results indicate that children who did not use
mental health services had younger caregivers than children who did use services.
Chi-square tests were run on the caregiver’s demographic characteristics with the
exception of age, all variables were found to not be statistically significant. In
addition, caregiver type was found to be statistically significant with foster
parents were more likely to use mental health services than children placed with
birth parents (see Table 2).
Table 2. Need
1
and Predisposing
2
Factors, and Caregiver Type by Mental Health
Service Use
3

Variables of Interest Service Use No Service Use Test

Internalizing Behavior

55.29 (11.58)

51.75 (11.37)

t (377) = -3.003**

Externalizing Behavior

59.81 (12.69)

53.72 (12.92)

t (377) = -4.622**

Total Competency

38.95 (8.18)

41.57 (8.92)

t (300) = 2.647**

Stress

1.96 (.69)

1.66 (.62)

t (373) = -4.408**

Depression

11.01 (9.84)

10.00 (10.75)

t (378) = .95

Social Support

4.23 (.52)

4.22 (.54)

t (378) = -.286

Age

45.77 (11.80)

42.00 (11.08)

t (381) = -3.219**

Caregiver Type
Birth Parent
Relative Caregiver
Foster Parent

62
53
86

87
50
56

χ
2
(2)

= 10.46**

Education
No Degree
High School
College

46
91
63

50
94
48

χ
2
(2)

= 2.08

Income
<$13,000
<$25,000
<$45,000
>$45,000

45
48
37
48

52
57
38
40

χ
2
(3)

= 1.796

Marital Status
Not Married
Married

112
89

118
75

χ
2
(1)

= 1.190

Ethnicity
Caucasian
African American
Latino
Asian

77
49
45
19

83
42
44
22

χ
2
(3)

= .992
* p ≤ .05, **p ≤ .001
Note: Results are in means and standard deviations. Caregiver Type, Education, Income and
Marital Status are presented as percentages. Due to missing data the sample size is 300.
1
The need factor which includes total competency, internalizing and externalizing behaviors is the
child’s symptomotology as measured by the caregiver.
2
Predisposing factors which include stress, depression and social support pertain to the caregiver.

3
Mental health service use is by the children under the caregiver’s care.
64
65
Research Question 1- Hypotheses 1A and 1B

To determine whether need and predisposing factors impact mental health
service use of caregivers for the children and youth under their care, adjusting for
covariates, two hierarchical logistic regression models were used in order to
control for education, ethnicity, income, marital status, and age.
Hypotheses 1A - The higher the need factors, the more likely that caregivers will
use mental health services for the children under their care.

As displayed in Table 3 the only statistically significant association found
with the caregiver’s characteristics and mental health service use by the child was
age of the caregiver with older caregivers more likely to use mental health
services for the children under their care (OR = 1.032, p<.05). In other words, for
every one year increase in the caregiver’s age there was a 3% increase in the
likelihood of mental health service use for the children under their care.
Externalizing behaviors were also significant with children displaying higher
externalizing behaviors more likely to use mental health services (OR = 1.049,
p<.05). For example, for every one unit increase in the Child Behavior Checklist
there was a 5% increase in the likelihood of mental health service use. No other
statistically significant results were found among the need factors. Hypothesis 1A
of the higher the need factors the more likely that caregivers will use mental
health services for the children under their care held true only for externalizing
behaviors.

66
Table 3. Hierarchical Logistic Regression: Covariates, Need Factor
1
, and Mental
Health Service Use
2

Variables

Odds Ratio
95% Confidence Interval
Lower Upper
Marriage (Married
vs. Not Married

1.511

.829

2.751
Education (vs. No
Degree)
High School
College

.803
.914

.370
.486

1.742
1.709
Ethnicity (vs.
Caucasian)
African Am.
Latino
Asian/Other

.718
.815
.970

.298
.314
.377

1.731
2.112
2.494
Income (vs.
>$45,000)
<$13,000
<$25,000
<$45,000

1.183
.770
.797

.521
.346
.355

2.687
1.715
1.792
Age 1.032* 1.009 1.056
Total Competency .973 .944 1.004
Internalizing
Behavior
.989 .959 1.020
Externalizing
Behavior
1.049* 1.019 1.080
* p<.05
Note: Due to missing data the sample size is 300.

1
The need factor which includes total competency, internalizing and externalizing behaviors is the
child’s symptomotology as measured by the caregiver. All other variables pertain to the
caregiver’s characteristics.
2
Mental health service use is that of the children under the caregiver’s custody.

Hypotheses 1B – The higher the predisposing factors the more likely that
caregivers will use mental health services for the children under their care.

A second logistic regression model tested the second hypothesis of the
higher the predisposing factors the more likely that caregivers will use mental
health services for the children under their care. Again, the only statistically
significant association found with the caregiver’s characteristics and mental
67
health service use by the child was that of the caregiver’s age with older
caregivers more likely to use mental health services for the children under their
care (OR = 1.034, p<.05). The same was true from the previous findings in which
one year increase in the caregiver’s age increased the likelihood of mental health
service use for the children under their care by 3%. The level of stress of the
caregiver was also significant with caregiver’s having a higher level of stress
more likely to use mental health services for the children under their care (OR =
2.346, p<.05). Results indicate that level of stress had a 134% increase in the
likelihood of mental health service use for the children under the caregiver’s care
for every one unit increase on the score of the Caregiver Strain Questionnaire. No
other statistically significant results were found among the predisposing factors.
Hypothesis 1B only held true for the predisposing factor of the caregiver’s level
of stress (see Table 4).

68
Table 4. Hierarchical Logistic Regression: Covariates, Predisposing Factor
1
, and
Mental Health Service Use
2

Variables

Odds Ratio
95% Confidence Interval
Lower Upper
Marriage
(Married vs. Not
Married

1.293

.781

2.141
Education (vs.
No Degree)
High School
College

.716
.818

.355
.468

1.441
1.428
Ethnicity (vs.
Caucasian)
African Am.
Latino
Asian/Other

.700
1.024
1.135

.324
.453
.503

1.512
2.316
2.560
Income (vs.
>$45,000)
<$13,000
<$25,000
<$45,000

.901
.762
.792

.422
.379
.396

1.925
1.535
1.588
Age 1.034* 1.013 1.056
Stress 2.346* 1.604 3.430
Depression 1.010 .984 1.036
Social Support 1.223 .763 1.958
** p<.05
1
Predisposing factors pertain to the caregiver.
2
Mental health service use is that of the children under the caregiver’s custody.

Research Question 2- Hypotheses 2AI, 2AII, and 2B
Testing the Measurement Model
The second research question aimed to examine whether caregivers’
perceptions and knowledge of mental health services mediated the relationships
between need and predisposing factors and mental health service use by
caregivers for the children under their care. Data analyses were conducted using
structural equation modeling with a two step approach as suggested by Anderson
69
and Gerbing (1988) and previously described in Chapter 3. The first step was to
test the measurement model to ensure that the hypothesized latent constructs fit
into the data. The measurement model was comprised of three latent variables
measured by three to eight indicators each (see Figure 3). This measurement
model with a total of 14 indicators (i.e., observed variables) plus error terms for
each indicator and the three latent variables and error terms for each variable
represents the measurement model in this study. Testing the measurement model
revealed that the model fit well into the data ( χ
2
= 249.790, df = 74, CFI = .91,
RMSEA = .074). Since the measurement model was appropriate, proceeding with
the second step of testing the structural models was conducted.

Figure 3. Need and Predisposing Factors and Perceptions and Knowledge of
Mental Health Services

Perception of MH
.72
contserv
e1
.69
helpserv
e2
.63
prvchara
e3
ChiSquare=249.790
DF=74
RMSEA=.074
CFI=.914
.48
effcserv
e4
.32
economic
e5
.56
.42
access
e6
.22
need
e7
.00
language
e8
Predisp
Need factor
.07
socsupp e9
.27
.18
depress e10
-.42
.74
stress e11
-.86
.80
externlt
e12
.90
.61
internlt
e13
.78
.10
totcomt
e14
.13
.64 .47 .05 .69 .79
.85
.83
-.32
-.83
-.22

Testing the Structural Models
The second step consisted of testing the structural models to examine
whether the paths among latent variables (i.e., need, predisposing, and perceptions
and knowledge of mental health services) and the outcome variable of mental
health service use at one year met the four criterion of the mediation effect
(previously noted in Chapter 3). Each criterion was tested separately with its
corresponding hypothesis for each of the paths.
70
71
Hypothesis 2AI – Caregivers of children with a high need will have negative
perceptions and knowledge of mental health services.

The first criterion consisted of the independent variables predicting the
mediator which were the two paths between the need and predisposing factors and
perceptions and knowledge of mental health services. The first path tested was
the path between the need factor and perceptions and knowledge of mental health
services (Figure 4). Results indicate that the path coefficient was significant ( β
= .14, C.R. = 2.47, p < .05) when the model fit the data well ( χ
2
= 122.413, df = 43,
CFI = .953, RMSEA = .066). Therefore, Hypothesis 2AI of caregivers of
children with a high need having negative perceptions and knowledge of mental
health services was supported.

Figure 4. The Effect of Need Factor on Perception and Knowledge of Mental
Health Services
.02
Perception of MH
.72
contserv
e1
.85
.69
helpserv
e2
.83
.63
prvchara
e3
.80
ChiSquare=122.413
DF=43
RMSEA=.066
CFI=.952
.48
effcserv
e4
.69
.31
economic
e5
.56
.41
access
e6
.64
.23
need
e7
.48
.00
language
e8
.05
Need factor
.92
externlt
e12
.96
.54
internlt
e13
.74
.09
totcomt
e14
-.30
D1
.14

Hypothesis 2AII – Caregivers with a high predisposing factor will have negative
perceptions and knowledge of mental health services.

The second path tested was the path between the predisposing factor and
perceptions and knowledge of mental health services. The path coefficient
between the predisposing factor and perceptions and knowledge of mental health
services (Figure 5) was also significant ( β = -2.78, C.R. = -4.29, p < .001) when
the model fit the data well ( χ
2
= 122.509, df = 43, CFI = .947, RMSEA = .066).
Hypothesis 2AII of caregivers with a high predisposing factor having negative
perceptions and knowledge of mental health services was also supported. Since
72
the first criterion of the mediation effect of the independent variables predicting
the mediator was met, the second criterion was tested.
Figure 5. The Effect of Predisposing Factor on Perception and Knowledge of
Mental Health Services

.08
Perception of MH
.71
contserv
e1
.85
.69
helpserv
e2
.83
.63
prvchara
e3
.79
ChiSquare=122.509
DF=43
RMSEA=.066
CFI=.947
.49
effcserv
e4
.70
.32
economic
e5
.56
.42
access
e6
.64
.23
need
e7
.48
.00
language
e8
.05
Predisposing
.28
socsupp
e9
.53
.69
depress
e10
-.83
.18
stress
e11
-.43
D2
-.28

73
74
Hypothesis 2B – Caregivers with more positive perceptions and knowledge of
mental health services will be more likely to use mental health services for the
children under their care.

The next step in testing the mediation effect was in testing the second
criterion of the mediator predicting the outcome variable when the independent
variable is controlled. As with the first criterion, the need factor and the
predisposing factor needed to be tested as independent variables. First,
perceptions and knowledge of mental health services and mental health service
use controlling for the need factor was tested. As seen in Figure 6 the model fit
the data well ( χ
2
= 134.414, df = 52, CFI = .95, RMSEA = .061). However, the
path coefficient between perceptions and knowledge of mental health services and
mental health service use when the need factor was controlled was not statistically
significant ( β = .016, C.R. = .305, p = .76), failing to meet the second criterion.

Figure 6. The Effect of Perception and Knowledge of Mental Health Services
to Mental Health Service Use Controlling for the Need Factor

Perception of MH
.72
contserv
e1
.85
.69
helpserv
e2
.83
.63
prvchara
e3
.80
ChiSquare=134.414
DF=52
RMSEA=.061
CFI=.950
.48
effcserv
e4
.69
.31
economic
e5
.56
.41
access
e6
.64
.23
need
e7
.48
.00
language
e8
.05
Need factor
.94
externlt
e12
.97
.53
internlt
e13
.72
.09
totcomt
e14
-.30
.06
Mental Health
Service Use at 1 yr.
.24
.02
.14
D4

Next, perceptions and knowledge of mental health services and mental
health service use controlling for the predisposing factor was tested. As seen in
Figure 7, the model with the path between perceptions and knowledge of mental
health services and mental health service use controlling for the predisposing
factor fit well into the data ( χ
2
= 150.068, df = 52, CFI = .935, RMSEA = .066),
but the path coefficient from perceptions and knowledge of mental health services
and mental health service use was also not statistically significant ( β = -.081, C.R.
= -1.264, p = .206), failing to meet the second criterion of the mediation effect.
Since there were no statistically significant results, hypothesis 2B could not be
answered.
75
Figure 7. The Effect of Perception and Knowledge of Mental Health Services
to Mental Health Service Use Controlling for Predisposing Factor

Perception of MH
.71
contserv
e1
.85
.69
helpserv
e2
.83
.63
prvchara
e3
.79
ChiSquare=150.068
DF=52
RMSEA=.066
CFI=.935
.49
effcserv
e4
.70
.32
economic
e5
.56
.42
access
e6
.64
.23
need
e7
.48
.00
language
e8
.05
Predisposing
.28
socsupp
e9
.53
.66
depress
e10
-.81
.19
stress
e11
-.44
-.28
.01
Mental Health
Service Use at 1 yr.
.03
-.08
D3

After failing to meet the required criteria for the mediation effect, it was
concluded that perceptions and knowledge of mental health services do not
mediate the relationship between need and predisposing factors and mental health
service use by caregivers for the children under their care.

76
77
Research Question 3- Hypotheses 3AI, 3AII, 3BI, and 3BII
Preliminary Analyses
Since one of the main variables of interest was caregiver type as it was
used as a moderator, bivariate analyses were conducted to examine differences
between each caregiver type (i.e., birth parent, relative caregiver, and foster
parent) and need and predisposing factors and mental health service use. Results
provide a first glimpse as to the possible moderating effects by caregiver type.
For the outcome variable of mental health service use by the caregiver for the
children under their care, a χ
2
test showed that there were statistically significant
differences in mental health service use by caregivers for the children under their
care among the different caregiver groups χ
2
(2)

= 10.46, p <.05) with foster
parents (56.2%) having the highest service use and birth parents (37.1%) having
the lowest service use.
In examining need and predisposing factors, differences were found
among caregivers with regard to externalizing behaviors of the children under
their care F (2, 409) = 3.57, p = .029, level of stress of the caregiver F (2, 407) =
8.34, p = .000, level of depression of the caregiver F (2, 412) = 28.108, p = .000)
and level of social support of the caregiver F (2, 414) = 20.83, p = .000), with
higher levels of externalizing behaviors and higher levels of stress and depression
indicating a higher level of dysfunction for the children under their care and for
the caregiver. A higher level of social support indicated a greater support system
for the caregiver (see Table 5). The Tukey HSD (Honestly Significant
Difference) test was conducted in order to determine which caregiver group was
significantly different. The results of this test indicated that statistically
significant differences in externalizing behaviors for the children under their care
existed between birth parents and relative caregivers ( X
difference
= 4.03, p = .039)
with birth parents having children with a higher externalizing behavior score on
the CBCL or a higher dysfunction in their behavior than relative caregivers.
There were also statistically significant differences in the level of stress
experienced between birth parents and relative caregivers ( X
difference
= .265, p
= .004) with birth parents having a higher level of stress. There were also
statistically significant differences between birth parents and foster parents
( X
difference
= .277, p = .001) with birth parents again having a higher level of
stress than foster parents. Statistically significant differences were also found in
the level of depression experienced between birth parents and relative caregivers
( X
difference
= 6.47, p = .000) with birth parents having a higher level of depression.
Birth parents and foster parents ( X
difference
= 8.41, p = .000) also had statistically
significant differences in their level of depression with birth parents having the
highest level. Level of support was also statistically significant between birth
parents and relative caregivers ( X
difference
= - .181, p = .013) with relative
caregivers having a stronger support system, There were also statistically
significant differences between birth parents and foster parents ( X
difference
= -.384,
78
p = .000) in the level of support with foster parents having a higher level of
support. Relative caregivers and foster parents ( X
difference
= -.204, p = .006) also
had statistically significant differences in the level of support with foster parents
having a stronger support system (see Table 5).
Table 5. Need
1
and Predisposing
2
Factors and Mental Health Service Use
3
by
Caregiver Type

Variables of
Interest

Caregiver Type

Birth Relative Foster
Parent Caregiver Parent
(n=167) (n=110) (n=153)

Test

Total Competency

39.57 (7.85)

50.79
(13.04)

46.54(10.83)

F (2, 409) = 3.57**

Internalizing
Behavior

39.57 (8.23)

41.31 (8.91)

40.11 (9.00)

F (2, 409) = .66

Externalizing
Behavior

57.83 (13.46)

53.81
(12.99)

57.67(13.07)

F (2, 325) = 1.06

Stress

1.98 (.75)

1.71 (.62)

1.70 (.56)

F (2, 407) = 8.34**

Depression

15.57 (12.70)

9.10 (9.19)

7.16 (7.39)

F (2, 412) = 28.108**

Social Support

4.04 (.58)

4.22 (.52)

4.42 (.42)

F (2, 414) = 20.83**

Mental Health
Service Use

Yes
No

37.1
52.1

48.2
45.5

56.2
36.6

χ
2
(2)

= 10.46*
* p ≤ .05, **p ≤ .001
Note: Results are in means and standard deviations. Mental health service use is presented as
percentages.
1
The need factor which includes total competency, internalizing and externalizing behaviors is the
child’s symptomotology as measured by the caregiver.
2
Predisposing factors which include stress, depression and social support pertain to the caregiver.

3
Mental health service use is by the children under the caregiver’s care

79
80
The third research question aimed to examine the moderator variable of
caregiver type (i.e., birth parents, relative caregivers, and foster parents). Since
there was no mediation effect found in the second research question, the
conceptual model (Figure 2) was revised as the original third research question
aimed to investigate whether the mediation path from need and predisposing
factors to perceptions and knowledge of mental health services to mental health
service use by caregivers for the children and youth under their care differed
among caregiver type. The revised conceptual model aimed to investigate
whether the paths from need and predisposing factors to perceptions and
knowledge of mental health services and from need and predisposing factors to
mental health service use by caregivers for the children and youth under their care
differed among caregiver type without taking into account the path between
perceptions and knowledge of mental health services and mental health service
use to disregard the mediation effect previously hypothesized (see Figure 8).

Figure 8. Moderator Effect by Caregiver Type Revised Conceptual Model

To answer the third research question and evaluate the moderator effect of
caregiver type on the four paths, several steps were taken. Each path was tested
separately: 1) the need factor and perceptions and knowledge of mental health
services; 2) the need factor and mental health service use by caregivers for the
children under their care; 3) the predisposing factor and perceptions and
knowledge of mental health services; and 4) the predisposing factor and mental
health service use. Testing of the moderator effect was conducted using multiple
group analyses based on the framework of measurement invariance (Ployhart &
Oswald, 2004) and previously discussed in Chapter 3.

81
82
Hypothesis 3AI – Children with a high need factor will have birth parents and
relative caregivers with negative perceptions and knowledge of mental health
services.

The first model aimed to examine the moderator effect of caregiver type
on the association between the need factor and perceptions and knowledge of
mental health services (see Figure 4). A baseline model was a measurement
invariant model in which all factors loadings as well as the path from need factors
to perceptions and knowledge of mental health services were constrained to be
equal across the three caregiver types. This baseline model was compared with
the hypothesized model in which the path from the need factor to perceptions and
knowledge of mental health services was freed to be estimated across the three
caregiver types (i.e., a moderator model).
Results indicate that the baseline model fit the data well ( χ
2
= 122.413, df
= 43, CFI = .95, RMSEA = .066). Testing the hypothesized model also revealed
that it fit well into the data ( χ
2
= 325.154, df = 169, CFI = .90, RMSEA = .047).
These two models were then compared using the chi-square difference statistic.
Model comparison results are presented in Table 6. The hypothesized model was
statistically different from the baseline model in the chi-square difference tests,
resulting in the selection of the hypothesized model over the constrained model
( χ
2
difference
= 6.227, df
differences
= 2, p = .043), indicating that there were differences
among caregivers in the need factor and perceptions and knowledge of mental
health services. Table 7 shows estimated parameters in the hypothesized model.
83
The path coefficients of the need factor influencing the perceptions and
knowledge of mental health services of caregivers was statistically significant
among birth parents (B = .024, C.R. = 2.288, p = .022) and relative caregivers (B
= .036, C.R. = 2.732, p = .006), but not among foster parents (B = -.004, C.R. = -
.334, p = .738). Therefore, hypothesis 3AI of children with a high need factor
will have birth parents and relative caregivers with negative perceptions and
knowledge of mental health services was supported.
Table 6. Model Comparison Test Results for Need Factor and Perceptions and
Knowledge of Mental Health Services

Models χ
2
df CFI RMSEA
1. Hypothesized model 325.154 169 .90 .047
2. Baseline model 331.431 171 .95 .066

Model differences
between

Δ χ
2

Δdf

p

Model 1 & 2 6.277 2 .043

Note:
1. CFI = comparative fit index; RMSEA = root-mean-square error of
approximation.
2. Δ χ
2
= differences in chi-squares of models; Δdf = differences in degree of
freedom of models.

84
Table 7. Estimated Path Coefficients in the Moderator Model of the Need Factor
and Perceptions and Knowledge of Mental Health Services by Caregivers
Caregiver Type
B S.E. C.R.

Birth Parent

.024*

.011

2.288

Relative Caregiver

.036**

.013

2.732

Foster Parent

-.004

.011

- .334
** p < .001, *p<.05
Hypothesis 3AII – Children with a high need factor will have: a) foster parents
more likely to use mental health services for the children under their care and, b)
relative caregivers less likely to use mental health services for the children under
their care.

The next model examined the moderator effect of caregiver type on the
association between the need factor and mental health service use (see Figure 9).
The baseline model had a good fit with the data ( χ
2
= 36.287, df = 18, CFI = .944,
RMSEA = .049). The hypothesized model also fit the data well ( χ
2
= 29.276, df =
16, CFI = .96, RMSEA = .044). Model comparison results are presented in Table
9. The hypothesized model was statistically different from the baseline model in
the chi-square difference tests and, therefore, the hypothesized model was
selected over the constrained model ( χ
2
difference
= 7.011, df
differences
= 2, p = .030),
which indicated that there were differences among caregivers in the need factor
and mental health service use for the children under their care.

Figure 9. Moderator Effect of Need Factor and Mental Health Service Use by
Caregivers

0, 6.82
Need factor
40.13
totcomt
0, 71.18
e1
1.00
1
53.65
internlt
0, 79.24
e2
-3.29
1
56.77
externlt
0, 2.48
e3
-4.96
1
ChiSquare=36.287
DF=18
RMSEA=.049
CFI=.944
.60
Mental Health
Service Use at 1 yr.
0, .22
D1
1
-.05

Table 8. Model Comparison Test Results for Need Factor and Mental Health
Service Use

Models χ
2
df CFI RMSEA
1. Hypothesized model 29.276 16 .96 .033
2. Baseline model 36.287 18 .99 .044

Model differences
between

Δ χ
2

Δdf

p

Model 1 & 2 7.011 2 .030
Note:
1. CFI = comparative fit index; RMSEA = root-mean-square error of
approximation.
2. Δ χ
2
= differences in chi-squares of models; Δdf = differences in degree of
freedom of models.

85
86
Table 9 shows estimated parameters in the hypothesized model. The path
coefficient of the need factor influencing mental health service use by caregivers
for the children under their care was statistically significant among relative
caregivers (B = -.081, C.R. = -3.298, p = .001) and foster parents (B = -.058, C.R.
= -3.051, p = .002), but not significant among birth parents (B = -.015, C.R. = -
.928, p = .353). Hypothesis 3AII was supported for relative caregivers, but not
for foster parents. Foster parents did not have a positive association as
hypothesized. In fact, foster parents of children with a high need factor were less
likely to use mental health services. Relative caregivers of children with a high
need factor were also less likely to use mental health services for the children
under their care as hypothesized.

Table 9. Estimated path coefficients in the moderator model of the Need Factor
and Mental Health Service Use by caregivers
Caregiver Type
B S.E. C.R.

Birth Parent

-.015

.016

.353

Relative Caregiver

-.081**

.025

-3.298

Foster Parent

-.058**

.019

- 3.051
** p < .001

87
Hypothesis 3BI – Birth parents and relative caregivers’ with a high predisposing
factor will have negative perceptions and knowledge of mental health services.

The third model examined the moderator effect of caregiver type on the
association between the predisposing factor and perceptions and knowledge of
mental health services (see Figure 10). The baseline model had a poor fit with the
data ( χ
2
= 465.102, df = 171, CFI = .7974, RMSEA = .063). In addition, the
hypothesized model had negative variance which suggested that the measurement
invariance model of the predisposing factor across the three caregiver groups
could not be estimated. Therefore, the hypothesized model of the predisposing
factor and perceptions and knowledge of mental health services among caregivers
was not tested. Due to the problems encountered in the factor model of the
predisposing factor, hypothesis 3BI could not be answered.

Figure 10. Moderator Effect of Predisposing Factor and Perceptions and
Knowledge of Mental Health Services by Caregivers

0
Perception of MH
1.84
contserv
0, 1.35
e1
1.00
1
1.55
helpserv
0, 1.23
e2
.98
1
1.11
prvchara
0, .86
e3
.70
1
ChiSquare=465.102
DF=171
RMSEA=.063
CFI=.797
1.04
effcserv
0, 2.67
e4
.67
1
.73
economic
0, .98
e5
.39
1
1.79
access
0, 5.18
e6
.98
1
.91
Perceived
need
0, 1.16
e7
.32
1
.04
language
0, .11
e8
.02
1
0, .00
Predisposing
Factor
4.26
socsupp
0, .39
e9
1.00
9.26
depress
0, 119.55
e10
103904.93
1.76
stress
0, .39
e11
5294.32
0, 2.46
D1
8982.24
1
1 1 1

Hypothesis 3BII – The higher the predisposing factors the more likely that birth
parents and relative caregivers’ will use mental health services for the children
under their care.

The fourth model examined the moderator effect of caregiver type on the
association between the predisposing factor and mental health service use (see
88
Figure 11). Unfortunately, the baseline model did not fit the data ( χ
2
= 92.085, df
= 2, CFI = .89, RMSEA = .13) as well as the hypothesized model ( χ
2
= 81.829, df
= 16, CFI = .324, RMSEA = .103). In addition, the chi-square difference tests
indicated no statistical differences between models ( χ
2
difference
= 4.256, df
differences

= 2, p = .119). These results indicated that hypothesis 3BII of the higher the
predisposing factor the more likely that biological parents and relative caregivers’
will use mental health services for the children under their care could not be
answered.
Figure 11. Moderator Effect of Predisposing Factor and Perceptions and
Mental Health Service Use by Caregivers

0, .04
Predisposing
Factor
1.76
stress
0, .28
e1
1.00
1
9.27
depress
0, 18.81
e2
31.07
1
4.28
socsupp
0, .15
e3
-1.00
1
ChiSquare=92.085
DF=18
RMSEA=.098
CFI=.303
.62
Mental Health
Service Use at 1 yr.
0, .23
D1
.30
1

Overview of Major Findings

This study examined the influence that caregivers in the child welfare
system have on the utilization of mental health services by the children under
their care. The first research question examined whether the child’s need factor
89
90
(i.e., internalizing and externalizing behaviors, and total competency) and
predisposing factors (i.e., the caregiver’s social support, stress level, and level of
depression) were significant in the utilization of mental health services by
children controlling for the demographic characteristics of the caregiver such as
age, marital status, education, ethnicity, and income. The only demographic
characteristic statistically significant was age of the caregiver with older
caregivers more likely to use mental health services for their children. Only
externalizing behaviors of the child and stress level of the caregiver were
statistically significant in mental health service use by children with the higher the
externalizing behaviors of the child and level of stress of the caregiver the more
likely children were to use mental health services.
The second research question examined whether perceptions and
knowledge of mental health services mediated need and predisposing factors and
mental health service use by caregivers for the children under their care. No
mediation effect was found between perceptions and knowledge of mental health
services and need and predisposing factors and mental health service use by
caregivers for their children. There were, however, statistically significant
associations between need and predisposing factors and perceptions and
knowledge of mental health services with the higher the need factor the more
negative the perceptions and knowledge of mental health services and the higher
91
the predisposing factors the more negative the perceptions and knowledge of
mental health services were by the caregiver.
The last research question examined whether caregiver type (i.e., birth
parents, relative caregivers, and foster parents) was a moderator in the paths
between need and predisposing factors, perceptions and knowledge of mental
health services, and mental health service use by the children under their care. A
moderating effect was found with the higher the need factors the more negative
the perceptions and knowledge of mental health services among birth parents and
relative caregivers. Furthermore, the higher the need factor the less likely that
relative caregivers and foster parents were to use mental health services for the
children under their care. Significance and implications of these findings are
discussed in the next chapter.

92
CHAPTER V: DISCUSSION

Previous research studies have generally focused on the child’s
demographic and case characteristics to better understand the underutilization of
mental health services by children and youth in the child welfare system.
However, children and youth in the child welfare system are supervised by
multiple individuals and systems that are responsible for children receiving
needed services. One important and crucial individual in a child’s life while in
care is their caregiver, as children generally do not seek services for themselves.
Caregivers are the “gatekeepers” for children’s well-being while in the child
welfare system. To date, little is known as to the role that a child’s caregiver
plays in the utilization of mental health services for the children under their care.
This study addressed this gap by examining need and predisposing factors,
perceptions and knowledge of mental health services, and mental health service
use to obtain a better understanding as to the influence that caregivers have on the
utilization of mental health services for the children under their care while in the
child welfare system. In addition, since only kin caregivers have generally been
examined in the child welfare literature this study extends the knowledge base by
using caregiver type as a moderator to examine the role the three most common
types of caregivers (birth parents, relative caregivers, and foster parents) have on
the utilization of mental health services for children in the child welfare system.
The study produced a number of interesting and unexpected findings which have
93
implications for both practice/policy and the research fields of child welfare and
mental health. Each section will summarize the study’s major findings in the
context of current knowledge and discuss their implications.
Caregivers’ Characteristics and Mental Health Service Use
Results of this study indicate that demographic characteristics such as
ethnicity, income, education, and marital status of the caregiver are not
independently associated with children’s utilization of mental health services as
previously noted in research studies. Age was a strong predictor, however, with
older caregivers more likely to have children under their care utilized mental
health services when age was entered into the two regression models examining
need and predisposing factors and mental health service use. Even after
controlling for demographic characteristics, age continued to be a significant
predictor. This finding departs from previous research studies that have
documented that minority, low income, poorly educated, single, and older
caregivers are the least likely to utilize mental health services for the children
under their care (Ehrle & Geen, 2002; Zima et al., 2000).
There are several reasons that explain the departure from past findings.
Previous research studies have generally only examined kin (relative) caregivers
and utilization of mental health services with children in the child welfare system.
This study is one of the first to incorporate the three most common types of
caregivers in the child welfare system (birth parents, relative caregivers, and
94
foster parents), and therefore, provides a broader perspective when examining
caregivers. Findings suggest that examining the combined effect of the different
types of caregivers in the child welfare system provides a unique perspective in
which demographic characteristics, other than age, are not significant in the
utilization of mental health services for the children under their care. In addition,
results suggest that younger caregivers are less likely to utilize mental health
services for the children under their care which closely resemble findings from
community samples in which younger caregivers with children in mental health
services are more likely to drop out of treatment (Kazdin, Holland, & Crowley,
1997).
Need and Predisposing Factors and Mental Health Service Use
Utilization of mental health services has previously been associated with a
child’s psychological need. Research studies have consistently documented that
children displaying externalizing behaviors are more likely to be deemed by
caregivers in need of mental health services (Zima et al., 2000; Shore et al., 2002;
Alegria et al., 2004; Timmer, Sedlar & Urquiza, 2004). Studies also indicate that
the greater the impairment in social functioning and the greater the severity of the
child’s aggressive behavior the more likely that parents or foster parents are to
seek mental health services (McKay, McCadam, & Gonzales, 1996; Farmers et al.,
1999; Leslie et al., 2000; Brannan & Heflinger, 2005). This study’s finding is
consistent with previous research studies as it was found that children with high
95
externalizing behaviors were more likely to use mental health services.
Internalizing behaviors and total competency were not significant with this
sample.
The fact that internalizing behaviors are not significant is also consistent
with research studies in the child welfare literature with abused children, as
generally children are only referred to mental health services when externalizing
behaviors are displayed (Garland et al., 1996; Blumberg et al., 1996). This
finding suggests that caregivers of maltreated children only seek mental health
services when the child’s behavior is severe. Indicating that children who may
have a need for services, but are not displaying any problematic behaviors are not
referred for services. This is a concern given that the effects of maltreatment if
left untreated can have long lasting consequences into adulthood. Research
studies have long established an association between child abuse and the
increased likelihood of psychiatric disorders and suicidal behavior in adulthood
(Molnar, Berkam & Buka, 2001; Molnar, Buka & Kessler, 2001; MacMillan et al.,
2001; Afifi, Brownridge & Sareen, 2006; Enns et al., 2006). In addition to
psychiatric problems, studies have found that a high percentage of substance
abuse can be attributed to abuse experienced in childhood with 56% of lifetime
drug problems, 63% of illicit drug addiction, and 64% of intravenous drug use
(Dube et al., 2003). A high percentage of suicide attempts can also be attributed
96
to child abuse with 80% of suicide attempts among adolescents and 64% of adult
suicide attempts (Dube at al., 2006).
These findings have practice implications. Social workers and caregivers
need to be educated as to the mental health issues that children may face after an
incident of abuse. Children need to be assessed and closely monitored to
determine whether there is a need for mental health services regardless of whether
the child is exhibiting behaviors indicative of mental health intervention, as some
children may not display any symptoms/behaviors and may be left untreated.
When examining predisposing factors and mental health service use by
caregivers for the children under their care, results show that a caregiver’s level of
stress is significant in determining whether mental health services are used for
their children. Results from this study are consistent with previous research
findings that indicate that the higher the level of stress of the caregiver the higher
the likelihood of the utilization of mental health services for their children (Ezzell
et al., 1999; Kolko et al., 1999; Timmer, Sedlar & Urquiza, 2004). Although past
research has found level of depression and social support for caregivers’ to also
be significant in the utilization of mental health services for their children, these
two predisposing factors are not significant with this sample. One clear
explanation for this can be attributed to stress playing a detrimental role in the
lives of caregivers. For example, research studies have found that the more the
stress in disruptive family and social relationships, with neighbors and the
97
community, interrupted work and personal time, and restrictions on personal
activities the more likely that the child is to use mental health services (Angold et
al., 1998; Brannan & Heflinger, 2005).
Even though a caregiver’s level of stress in this study is a positive
outcome and contributes to a child’s use of mental health services, a cautionary
word must be made. Caregivers in the child welfare system are overburdened, not
only from the demands of the children under their care, but also from the various
systems (e.g., child welfare, court, and school) they interact with and often seek
services only in a crisis situation (Geen, 2003). It is, therefore, the responsibility
of the child welfare system to educate caregivers to understand their children’s
behavior early on so that their level of stress does not determine whether services
are sought. Children should receive needed services regardless of the caregiver’s
psychological well-being.
Perceptions and Knowledge of Mental Health Services

This is the first study to examine a caregiver’s perceptions and knowledge
of mental health services and mental health service use for the children under
their care in the child welfare literature. Results from this study are informative
and somewhat surprising.
Research studies in the children’s mental health literature have
consistently shown that children’s psychological needs are associated with a
caregiver’s perceptions and knowledge of mental health services (Flisher et al.,
98
1997; Pavuluri, Luk & McGee, 1996; Richardson, 2001; McKay et al., 1998;
Zima et al., 2000; Yeh et al., 2003). For example, Richardson (2001) found that
parents with children who had been identified with a mental health problem were
two times more likely to expect the mental health professionals to be
untrustworthy and to provide poor care. Flisher et al. (1997) also found that
parents of children with unmet need were more likely to believe that services
would take too much time or be inconvenient. Results from this study corroborate
previous findings in which caregivers of children with a high psychological need
are more likely to have negative perceptions and knowledge of mental health
services.
It is difficult to ascertain the causality of a child’s psychological need and
a caregiver’s negative perceptions and knowledge of mental health services
because of the cross sectional nature of this study. However, there is a salient
explanation as to this negative finding. First, it is important to note that a child’s
need or psychological symptomology was rated by the child’s caregiver using the
CBCL, and therefore, it is the caregiver who deemed the child’s behavior
problematic. One reason for the study’s current findings is the caregiver’s past
experiences with the mental health system. Because 45% of the children in this
sample also had an open case with the department of mental health, findings may
be attributed to the familiarity of mental health services. Yeh et al. (2003)
concluded in their research study that caregivers who are the most familiar with
99
mental health services may also enforce more barriers to service use because of
their knowledge of the obstacles in receiving services. Another reason other than
barrier enforcement is the caregiver’s exposure to the quality of the mental health
services provided. If past experiences included treatment not being effective or
lengthy, caregivers may perceive mental health services as ineffective, and
therefore, have negative perceptions and knowledge of mental health services.
These findings have implications for both the child welfare and mental
health systems. Findings suggest that focus needs to be placed on the caregiver to
provide them the opportunity to be incorporated in their children’s treatment plans
and goals, as studies of caregivers in the children’s mental health literature have
shown that inclusion of caregivers as partners in their children’s mental health
treatment leads to service utilization and effectiveness (see Richards et al., 2008).
Inclusion of caregivers in treatment may serve to eliminate the negative
perceptions and knowledge of mental health services and increase utilization of
these services.
While a handful of research studies have examined a caregiver’s
predisposing factors and perceptions and knowledge of mental health services,
research findings from this study corroborate previous research findings. In one
of the few studies conducted, results show that caregivers with high stressors and
identified obstacles tend to believe that therapy is not relevant and too demanding
(Nock, Phil, & Kazdin, 2001). This study’s findings are consistent with previous
100
studies in that caregivers with a high predisposing factor have negative
perceptions and knowledge of mental health services. This suggests that
caregivers are more likely to believe that mental health services are not a solution
or are an ineffective method when their own psychological well-being is affected.
Again, this is an important finding for the child welfare system as it suggests that
social workers need to focus on caregivers as well as children when providing
resources. However, because few studies have examined a caregiver’s
perceptions and knowledge of mental health services, additional research needs to
be conducted to determine the generalizability of this finding to caregivers in the
child welfare system.
As previously noted, this study is one of the first to examine perceptions
and knowledge of mental health services, need and predisposing factors, and
mental health service use by the children under the caregiver’s care in the child
welfare literature. One of the central aims of this study was to examine whether
perceptions and knowledge of mental health services served as a mediator in
service use. Findings from this study depart from previous research studies that
have consistently documented that a caregiver’s perceptions and knowledge of
mental health services are significant in the utilization of mental health services
for the children in their care (e.g., Flisher et al., 1997; Pavuluri, Luk & McGee,
1996; Richardson, 2001; McKay et al., 1998; Zima et al., 2000; Yeh et al., 2003).
Findings from this study indicate that a caregiver’s perceptions and knowledge of
101
mental health services do not mediate need and predisposing factors and mental
health service use for their children.
Methodological issues could explain the departure from past findings. For
example, data for perceptions and knowledge of mental health services were
collected using the Measurement of Parental Endorsement of Barriers developed
by Yeh and colleagues. Only one study to date has used this measure to examine
barriers to service use by caregivers, and the subsample (i.e., the child welfare
sector) used in the present study derives from this study. In the study conducted
by Yeh et al. (2003), researchers found that caregivers of children with a high
psychological need were more likely to report fewer barriers or positive
perceptions and knowledge of mental health services, but were the least likely to
use mental health services for their children. This suggest that this instrument
needs further testing to obtain an in depth understanding of the findings. Further
testing is needed to understand whether the barriers being enforced actually
provide an accurate description of a caregiver’s perceptions and knowledge of
mental health services and whether the instrument is a reliable measure when used
with caregivers in the child welfare system.
A caregiver’s characteristics and contextual factors could also explain the
departure from past findings. For example, studies have shown that when a
child’s psychological need is added to the model a caregiver’s stress and history
of psychopathology continues to increase the likelihood of service use beyond
102
need (Farmer et al., 1997; Angold et al., 1998), suggesting that a caregiver’s
perceptions and knowledge of mental health services may not be a significant
predictor and that a caregiver’s psychological well-being is a stronger predictor.
In addition, because caregivers in the child welfare system are expected to abide
by child welfare agency and court mandates, these mandates may be a stronger
influence in whether caregivers seek and utilize mental health services for their
children than their own perceptions and knowledge of these services. Qualitative
research may be able to provide a better understanding as to a caregiver’s
perceptions and knowledge of mental health service use with children in the child
welfare system.
Caregiver Type

Kin caregivers and foster parents have been the focus in the child welfare
literature when examining utilization of mental health services by children
(Timmer, Sedlar, & Urquiza, 2004; McMillan et al., 2004; James et al., 2004;
Leslie et al., 2000; Blumberg et al., 1996). Of these studies, however, none have
examined the three most common types of caregivers in the child welfare system
(i.e., birth parents, relative caregivers, and foster parents) and the role that need
and predisposing factors and a caregiver’s perceptions and knowledge of mental
health services play in the utilization of children’s mental health services. This
study extends the examination of traditional caregivers in the child welfare
literature by adding birth parents as caregivers.
103
Findings from previous research studies indicate that parents of children
with a high psychological need have negative perceptions and knowledge of
mental health services (Richard, 2001; Fletcher et al., 1997). Prior research
studies have not included relative caregivers, but because of the nature of the
relationship with the child, a birth parent’s perceptions and knowledge of mental
health services may be similar to that of a relative caregiver. Findings from this
study corroborate previous research studies in the children’s mental health
literature. This negative relationship among birth parents and relative caregivers
suggests that caregivers of children with a familial connection may view mental
health services as ineffective even when the children under their care are in need
of services. This is an important finding to the field of child welfare as previous
studies have found that children placed with relatives generally receive less
support and monitoring by social workers (Berrick & Barth, 1994; Dubowitz et al.,
1993). Findings from this study provide evidence as to the importance of
educating and closely monitoring cases where a child remains or is returned to a
birth parent or is placed with a relative.
The child welfare literature has primarily focused on relative caregivers
and studies indicate that children placed with relative caregivers are the least
likely to utilize mental health services (Timmer, Sedlar, & Urquiza, 2004;
McMillan et al., 2004; James et al., 2004; Leslie et al., 2000; Blumberg et al.,
1996). This study’s findings are consistent with previous research studies in that
104
relative caregivers of children with a high psychological need were less likely to
use mental health services for the children under their care. One explanation for
this could be that children placed with relatives are likely to be closely bonded
with family and relatives may have the same stigma that birth parents have when
dealing with mental health service use. For example, Richardson (2001) found
that stigma was a concern with parents of children with unmet need as parents did
not want relatives to find out the child was receiving services. Another
explanation can be attributed to relative caregivers often seeking services only in
times of crisis (Geen, 2003), which suggests that relative caregivers do not want
any involvement with the child welfare system. A third explanation is that
relative caregivers are more tolerant of their children’s behavior. In a study
conducted by Owens et al (2002), researchers found that parents were more likely
to enforce barriers that included the belief that the problem (i.e., psychiatric
problem) can be handled without treatment. Lastly, relative caregivers are less
knowledgeable of the workings of the child welfare and mental health systems
(Geen, 2003), and therefore, are unable to access services. Again, this is an
important finding for both the child welfare and mental health systems as
caregivers need to be targeted and provided educational trainings as to the
importance of mental health treatment for children with a high psychological need.
Targeting caregivers may lead to a higher utilization of mental health services by
children.
105
One surprising finding in this study that contradicts previous research is
that foster parents of children with a high psychological need were less likely to
utilize mental health services for the children under their care. Previous research
studies have consistently documented that children placed in foster care are more
likely to receive mental health services (Leslie et al., 2004; McMillen et al., 2004;
James et al., 2004; Garland et al., 2000; Leslie et al., 2000; Garland et. al., 1996).
The negative association between psychological need and children’s mental
health service use among foster parents can be attributed to a foster parent’s
negative past experiences with the mental health system or with past experiences
with children with high psychological need and placement instability, as research
studies have documented that placement change is generally associated with
behavior problems (James et al., 2004). Using Timmer, Sedlar, and Urquiza’s
(2004) Social Exchange Theory where relationships are maintained when the
benefits of the relationship to the individual outweigh the costs, findings from this
study suggest that foster parents may not be invested in the relationship with some
foster children, and therefore, may choose not to seek services. Another reason is
that foster parents do not always follow through on recommendations or advocate
for their children’s mental health needs (Garland et al., 2000).
Close monitoring of children placed in foster homes is warranted as
historically the child welfare literature has shown that children placed with foster
parents are more likely to use mental health services and social workers may take
106
for granted that children are receiving needed services. Further research is also
needed to determine the causality of this surprising negative association.
One methodological problem that was encountered and prevented the
examination of predisposing factors, perceptions and knowledge of mental health
services, and mental health service use using caregiver type as a moderator was
the fact that the latent variable of predisposing factor when caregiver type was
used did not fit the data. In other words, the latent variable predisposing factor
did not appear to be consistent among the three caregiver groups. This is not
surprising given that the scales used to measure social support, level of depression,
and stress level (i.e., Social Support Provision Scale, Center for Epidemiological
Studies – Depression Scale, and the Caregiver Strain Questionnaire) have
previously not been used with a population of caregivers in the child welfare
system. These scales have generally been used with only birth parents. When
examining the three different types of caregivers (birth parents, relative caregivers,
and foster parents) the scales do not appear to be appropriate. Further researcher
is needed to examine whether scales used in previous studies in the children’s
mental health literature are reliable measures to use with the different types of
caregivers in the child welfare system.
Limitations

Several limitations should be considered in evaluating the results of this
study. First, power analysis indicated that the sample size (N=430) used in this
107
study was an adequate size for purposes of this study. However, a larger sample
size could have provided the opportunity to examine other group factors not
examined. For example, racial/ethnic status, court mandates, and prior mental
health service use could also have been used as a moderator to provide a better
understanding of some of the current findings.
Second, the cross sectional nature of the study made it difficult to examine
causality with some of the variables of interest. Since all variables with the
exception of mental health service use were collected at baseline it was difficult to
determine what effect a child’s psychological need had on perceptions and
knowledge of mental health services or with the caregiver’s predisposing factors.
Examining the data longitudinally could provide a better understanding of the
study’s findings.
Third, the outcome variable used in this study was a dichotomous variable
(i.e., used of any specialty mental health services or no use of any specialty
mental health services in the previous 12 months). This limited the ability to
examine type of treatment used, frequency, and duration. Being able to examine
mental health services more specifically could provide a better understanding of
the association between mental health service use, need and predisposing factors,
and perceptions and knowledge of mental health services.
Lastly, there was one methodological problem encountered. This problem
could be attributed to the statistical technique used. Structural equation modeling
108
was used because of its ability to examine various regression models
simultaneously and because of the ability to use caregiver type as a moderator.
However, latent variables have previously not been used in the children’s mental
health and child welfare literature. There were problems encountered using the
latent variable predisposing factor. The latent variable was constructed using
scale scores from the Social Support Provision Scale, Center for Epidemiological
Studies – Depression Scale, and the Caregiver Strain Questionnaire. Although it
was tested using a measurement model and results indicated that the model fit
well into the data there were problems when attempting to used the latent variable
with the moderator of caregiver type. Results indicated that the variable was not
consistent across the three caregiver types. This made it impossible to test any
hypotheses that contained the latent variable of predisposing factor and the
moderator.
In spite of these limitations, this research study is important because it
begins to identify factors that influence the utilization of mental health services by
children in the child welfare system other than what has previously been
examined. It also provides a better understanding as to the role that the caregiver
plays in the utilization of mental health services by the children in their care. In
addition, this is the first study to date that has examined the three most common
types of caregivers (birth parents, relative caregivers, and foster parents) in the
child welfare system. Findings from this study provide both the child welfare and
109
mental health systems evidence that caregivers are a crucial and central figure in a
child’s receipt of mental health services and need to be targeted to begin to
address the underutilization of services for children in the child welfare system.

110
References

Achenbach, T.M. (1991a). Manual for the Child Behavior Checklist/4-18 and
1991 profile. Burlington, VT: University of Vermont, Department of
Psychiatry.

Achenbach, T.M. (1991b). Manual for the Youth Self-Report and 1991 profile.
Burlington: University of Vermont, Department of Psychiatry.

Afifi, T.O., Brownridge, D.A., Cox, B.J., Sareen, J. (2006). Physical Punishment,
Childhood Abuse, and Psychiatric Disorders. Child Abuse and Neglect, 30,
1093-1103.

Alegria, M., Canino, G., Lai, S., Ramirez, R.R., Chavez, L., Rusch, D., & Shrout,
P.E. (2004). Understanding Caregivers’ Help-Seeking for Latino
Children’s Mental Health Care Use. Medical Care, 42 (5), 447-455.

Alvidrez, J. (1999). Ethnic Variations in Mental Health Attitudes and Service Use
Among Low-Income African American, Latina, and European American
Young Women. Community Mental Health Journal, 35(6), 515-530.

Anderson, J. C., & Gerbing, D. W. (1988). Structural Equation Modeling in
Practice: A Review and Recommended Two-Step Approach.
Psychological Bulletin May, 103(3), 411-423.

Angold, A., Messer, S.C., Stangl, D., Farmer, E.M., Costello, E.J., & Burns, B.J.
(1998). Perceived Parental Burden and Service Use for Child and
Adolescent Psychiatric Disorders. American Journal of Public Health,
88(1), 75-80.

Aron, A. & Aron, E. N. (1999). Statistics for Psychology (2nd ed.). NJ: Prentice-
Hall.

Baron, R.M., & Kenny, D.A. (1986). The Moderator-Mediator Variable
Distinction in Social Psychological Research. Journal of Personality and
Social Psychology, 51 (6), 1173-1182.

Becker, M., Jordan, N., & Larsen, R. (2006). Behavioral Health Service Use and
Costs among Children in Foster Care. Child Welfare, 85(3), 633-647.

Bentler, P., & Chou C.P. (1987). Practical Issues in Structural Modeling.
Sociological Methods & Research, 16, 78-117.
111
Berrick, J.D. (1998). When Children Cannot Remain Home: Foster Family Care
and Kinship Care. The Future of Children, 8(1), 72-87.

Blumberg, E., Landsverk, J., Ellis-MacLeod, E., Ganger, W., & Culver, S. (1998).
Use of the Public Mental Health System by Children in Foster Care: Client
Characteristics and Service Patterns. Journal of Mental Health
Administration, 23 (4), 398-405.

Brannan, A.M., Heflinger, C.A., & Foster, E.M. (2003). The Role of Caregiver
Strain and Other Family Variables in Determining Children’s Use of
Mental Health Services. Journal of Emotional and Behavioral Disorders,
11 (2), 77-91.

Brannan, A.M, & Heflinger, C.A. (2005). Child Behavioral Service Use and
Caregiver Strain: Comparison of Managed Care and Fee for Service
Medicaid Systems. Mental Health Services Research, 7(4), 197-211.

Browne, M.W., & Cudeck, R. (1993). Alternative Ways of Assessing Model Fit.
In K.A. Bollen & J. S. Long (Eds), Testing Structural Equation Models
(pp. 136-162). Newbury Park, CA: Sage.

Burnette, D., (1999). Custodial Grandparents in Latino Families: Patterns of
Service Use and Predictors of Unmet Needs. Social Work, 44 (1), 22-34.

Burns, B.J., Phillips, S.D., Wagner, R., Barth, R.P., Kolkom, D.J., Campell, Y., &
Landsverk, J. (2004). Mental Health Needs and Access to Mental Health
Services by Youth Involved with Child Welfare: A National Survey.
Journal of the American Academy of Child and Adolescent Psychiatry, 43,
960-978.

Byrne, B. M. (1994). Structural Equation Modeling with EQS and EQS/Windows:
Basic Concepts, Applications, and Programming. Thousand Oaks, CA:
Sage.

Cepeda-Benito, A., & Short, P. (1998). Self-Concealment, Avoidance of
Psychological Services, and Perceived Likelihood of Seeking Professional
Help. Journal of Counseling Psychology, 45, 58-64.

Chalk, R., Gibbons, A., & Scarupa, H.J. (2002). The Multiple Dimensions of
Child Abuse and Neglect: New Insights into an Old Problem. Washington,
DC: Child Trends. Retrieved April 27, 2006, from
www.childtrends.org/Files/ChildAbuseRB.pdf.
112
Chou, C.P., & Bentler, P. (1990). Model Modification in Covariance Structural
Modeling: A Comparison among Likelihood Ratio, Lagrange Multiplier,
and Wald Tests. Multivariate Behavioral Research, 25(1), 115-136.

Clausen, J. M., Landsverk, J., Ganger, W., Chadwick, D., & Litrownik, A. (1998).
Mental Health Problems of Children in Foster Care. Journal of Child and
Family Studies, 7(3), 283.

Constantine, M.G. (2002). Predictors of Satisfaction with Counseling: Racial and
Ethnic Minority Clients’ Attitudes toward Counseling and Ratings of
Their Counselors’ General and Multicultural Counseling Competence.
Journal of Counseling Psychology, 49 (2), 255-263.

Courtney, M.E., Piliavin, I., Grogan-Kaylor, A., & Nesmith, A. (2001). Foster
Youth Transitions to Adulthood: A Longitudinal View of Youth Leaving
Care. Child Welfare, 80, 685-717.

Cutrona, C.E., & Russell, D. (1987). The Provisions of Social Relationships and
Adaptations to Stress. In W.H. Jones & D. Perlman (Eds), Advances in
Personal Relationships (Vol.1, pp.37-68) Greenwich, CT: JAI Press.

Dallam, S.J. (2001). The Long-Term Medical Consequences of Childhood
Maltreatment. In K. Franey, R. Geffner, & R. Falconer (Eds.). The Cost of
Child Maltreatment: Who pays? We all do. San Diego, CA: Family
Violence & Sexual Assault Institute.

Deane, F.P., & Todd, D.M. (1996). Attitudes and Perceived Likelihood of
Seeking Professional Help for Personal Problems or Suicidal Thinking.
Journal of College Students Psychotherapy, 10, 45-59.

Delphin, M.E., & Rollock, D. (1995). University Alienation and African
American Ethnic Identity as Predictors of Attitudes toward, Knowledge
about, and Likely Use of Psychological Services. Journal of College
Student Development, 36 (4), 337-346.

Diala, C., Muntaner, C., Walrath, C., Nickerson, K.J., LaVeist, T.A., & Leaf, P.J.
(2000). Racial Differences in Attitudes toward Professional Mental Health
Care and in the Use of Services. American Journal of Orthopsychiatry, 70
(4), 455-464.

DiLauro, M. (2004). Psychosocial Factors Associated with Type of Child
Maltreatment. Child Welfare, 83(1), 69-99.
113
dosReis, S., Zito, J.M., Safer, D.I., & Soeken, K.L. (2001). Mental Health
Services for Youths in Foster Care and Disabled Youth. American Journal
of Public Health, 91, 1094-1099.

Dube, S.R., Anda, R.F., Felitti, V.J., Chapman, D.P., & Giles, W.H. (2001).
Childhood Abuse, Household Dysfunction, and the Risk of Attempted
Suicide throughout the Life Span: Findings From the Adverse Childhood
Experience Study. Journal of the American Medical Association, 86,
3089-3096.

Dube, S.R., Felitti, V.J., Dong, M., Chapman, D.P., Giles, W.H., & Anda, R.F.
(2003). Childhood Abuse, Neglect, and Household Dysfunction and the
Risk of Illicit Drug Use: The Adverse Childhood Experiences Study.
Pediatrics, 111, 564-572.

Enns, M.V., Cox, B.J., Afifi, T.O., de Graaf, R., ten Have, M., & Sareen, J.
(2006). Childhood Adversities and Risk for Suicidal Ideation and
Attempts: A Longitudinal Population Based Study. Psychological
Medicine, 31, 1769-1778.

Ehrle, J., & Geen, R. (2002). Children Cared for by Relatives: What Services Do
They Need? The Urban Institute New Federalism National Survey of
American’s Families, B-47, 1-7.

Ezzell, C. E., Swenson, C. C., & Faldowski, R. A. (1999). Child, Family and Case
Characteristics: Links with Service Utilization in Physically Abused
Children. Journal of Child & Family Studies, 8(3), 271-284.

Flisher, A.J., Kramer, R.A., Grosser, R.C., Alegria, M., Bird, H.R., Bourdon,
K.H., Goodman, H., Greenwald, S., Horwitz, S.M., Moore, R.E., Narrow,
W.E., & Hoven, C.W. (1997). Correlates of Unmet Need for Mental
Health Services by Children and Adolescents. Psychological Medicine, 27,
1145-1154.

Farmer, E.M., Burns, B.J., Angold, A., & Costello, E.J. (1997). Impact of
Children’s Mental Health Problems on Families: Relationship with
Service Use. Journal of Emotional and Behavioral Disorders, 5, 230-238.

Farmer, E.M., Stangl, D.K., Burns, B.J., Costello, E.J., & Angold, A. (1999). Use,
Persistence, and Intensity: Patterns of Care for Children’s Mental Health
across One Year. Community Mental Health Journal, 35(1), 31-46.

114
Fritz, M.S., & MacKinnon, D.P. (In press). Required Sample Size to Detect the
Mediated Effect. Psychological Science.

Garland, A.F. & Besinger, B.A. (1997). Racial/Ethnic Differences in Court
Referred Pathways to Mental Health Services for Children in Foster Care.
Children & Youth Services Review, 19 (8), 651-666.

Garland, A. F., Hough, R. L., Landsverk, J. A., McCabe, K. M., Yeh, M., &
Ganger, W. C. (2000). Racial and Ethnic Variations in Mental Health Care
Utilization among Children in Foster Care. Children's Services: Social
Policy, Research, &Practice, 3(3), 133-146.

Garland, A. F., Landsverk, J. L., Hough, R. L., & Ellis-Macleod, E. (1996). Type
of Maltreatment as a Predictor of Mental Health Service Use for Children
in Foster Care. Child Abuse & Neglect, 20(8), 675-688.

Geen, R. (2003). “Foster Children Placed with Relatives Often Receive Less
Government Help”. Washington, D.C.: The Urban Institute.

Grogan-Kaylor, A. (2000). Who Goes into Kinship Care? The Relationship of
Child and Family Characteristics to Placement into Kinship Foster Care.
Social Work Research, 24(3), 132.

Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1992). Multivariate
Data Analysis. Prentice Hall, Englewood Cliffs, NJ.

Halfon, N., Berkowitz, G., & Klee, L. (1992). Mental Health Service Utilization
by Children in Foster Care in California. Pediatrics, 89, 1238-1244.

Halfon, N., Mendoca, A., & Berkowitz, G. (1995). Health Status of Children in
Foster Care: The Experience of the Center for the Vulnerable Child.
Archives of Pediatric and Adolescent Medicine, 149, 386-392.

Harman, J.S., Childs, G.E., & Kelleher, K.J. (2000). Mental Health Care
Utilization and Expenditures by Children in Foster Care. Archives of
Pediatric and Adolescent Medicine, 154, 1114-1117.

Horowitz, S.M., Hoagwood, K., Stiffman, A.R., Summerfeld, T., Weisz, J.R.,
Costello, E.J., et al. (2001). Reliability of the Services Assessment for
Children and Adolescents. Psychiatric Services, 52, 1088-1094.

115
Hurlburt, M.S., Leslie, L.K., Landsverk, J., Barth, R.P., Burns, B.J., Gibbons,
R.D., Slymen, D.J., Zhang, J. (2004). Contextual Predictors of Mental
Health Service Use among Children Open to Child Welfare. Archives of
General Psychiatry, 1217-1224.

Hu, L., & Bentler, P.M. (1999). Cutoff Criteria for Fit Indexes in Covariance
Structure Analysis: Conventional Criteria versus New Alternatives.
Structural Equation Modeling, 6, 1-55.

James, S., Landsverk, J., Slymen, D.J., & Leslie, L.K. (2004). Predictors of
Outpatient Mental Health Service Use – The Role of Foster Care
Placement Change. Mental Health Services Research, 6 (3), 127-141.

Kazdin, A.E., Holland, L., & Crowley, M. (1997). Family Experience of Barriers
to Treatment and Premature Termination from Child Therapy. Journal of
Consulting and Clinical Psychology, 65(3), 453-463.

Kelly, A.E., & Achter, J.A. (1995). Self-Concealment and Attitudes toward
Counseling in University Students. Journal of Counseling Psychology, 37,
178-184.

Kelly, B.T., Thornberry, T.P., & Smith, C.A. (1997). In the Wake of Childhood
Maltreatment. Washington, DC: National Institute of Justice. Retrieved
April 27, 2006, from http://www.ncjrs.gov/165257.pdf.

Kerker, B.D., & Dore, M.M. (2006). Mental Health Needs and Treatment of
Foster Youth: Barriers and Opportunities. American Journal of
Orthopsychiatry, 76(1), 138-147.

Kline, T. J. B. (1998). Principles and Practice of Structural Equation Modeling.
Methodology in the Social Sciences. New York, NY: Guilford Press.

Kline, R.B. (2005). Principles and Practice of Structural Equation Modeling 2nd
Ed. The Guildford Press. New York, New York.

Kolko, D.J., Selelyo, J., Brown, E. J. (1999). The Treatment Histories and
Services Involvement of Physically and Sexually Abusive Families:
Description, Correspondence, and Clinical Correlates. Child Abuse &
Neglect, 23, (5), 459-466.

116
Landsverk, J., & Garland, A.F. (1999). Foster Care and Pathways to Mental
Health Services. In P. A. Curtis and G. Dale, Jr. Lincoln (Eds). The Foster
Care Crisis: Translating Research into Policy and Practice. Child, Youth,
and Family Services. NE, University of Nebraska Press.

Landsverk, J., Garland, A.F. & Leslie, L.K. (2002). Mental Health Services for
Children Reported to Child Protective Services. In J.E.B. Myers, L.
Berliner, J. Briere, C.T. Hendrix, C. Jenny & T.A. Reid (Eds.), The
APSAC Handbook on Child Maltreatment (2
nd
ed.) (pp.487-507).
Thousand Oaks: Sage.

Leslie, L. K., Landsverk, J., Ezzet-Lofstrom, R., Tschann, J. M., Slymen, D.J., &
Garland, A.F. (2000). Children in Foster Care: Factors Influencing
Outpatient Mental Health Service Use. Child Abuse & Neglect, 24(4),
465-476.

Leslie, L.K., Hurlburt, M.S., James, S., Landsverk, J., Slymen, D.J., & Zhang, J.
(2004). Relationship between Entry into Child Welfare and Mental Health
Service Use. Psychiatric Services, 56 (8), 981-987.

Leslie, L.K., Hurlburt, M.S., Landsverk, J., Barth, R., & Slymen, D.J. (2004).
Outpatient Mental Health Services for Children in Foster Care: A National
Perspective. Child Abuse & Neglect, 28, 697-712.

MacCallum, R. C., Browne, M. W., & Sugawara, H. M. (1996). Power Analysis
and Determination of Sample Size for Covariance Structure Modeling.
Psychological Methods, 1, 130-149.

MacKinnon, D.P., & Dwyer, J.H. (1993). Estimating Mediated Effects in
Prevention Studies. Evaluation Review, 17 (2), 144-158.

MacKinnon, D.P., & Krull, J.L., & Lockwood, C.M. (2000). Equivalence of the
Mediation, Confounding and Suppression Effect. Prevention Science, 1 (4),
173-181.

MacMillen, H.L., Fleming, J.E., Streiner, D.L., et al. (2001). Childhood Abuse
and Lifetime Psychopathology in a Community Sample. American
Journal of Psychiatry, 158, 1878-1883.

117
McCabe, K.M., Yeh, M., Lau, A., Garland, A., & Hough, R. (2003).
Racial/Ethnic Differences in Caregiver Strain and Perceived Social
Support among Parents of Youth with Emotional and Behavioral Problems.
Mental Health Services Research, 5(3), 137-147.

McKay, M.M., McCadam, K., & Gonzales, J.J. (1996). Addressing the Barriers to
Mental Health Services for Inner City Children and Their Caretakers.
Community Mental Health Journal, 30 (4), 353-361.

McKay, M.M., Stoewe, J., McCadam, K., & Gonzales, J. (1998). Increasing
Access to Child Mental Health Services for Urban Children and Their
Caregivers. Health and Social Work, 23 (1), 9-15.

McMillen, J.C., Scott, L.D., Zima, B.T., Ollie, M.T., Munson, M.R., & Spitznagel,
E. (2004). Use of Mental Health Services among Older Youths in Foster
Care. Psychiatric Services, 55 (7), 811-817.

McMiller, W.P. & Weisz, J.R. (1996). Help-Seeking Preceding Mental Health
Clinic Intake among African-American, Latino, and Caucasian Youths.
Journal of the American Academy of Child and Adolescent Psychiatry, 35
(8), 1086-1094.

Molnar, B.E., Burkman, L.F., & Buke, S.L. (2001). Psychopathology, Childhood
Sexual Abuse and Other Childhood Adversities: Relative Links to
Subsequent Suicidal Behavior in the U.S. Psychological Medicine, 31,
965-977.

Molnar, B.E., Buka, S.L., & Kessler, R.C. (2001). Child Sexual Abuse and
Subsequent Psychopathology: Results from the National Comorbidity
Survey. American Journal of Public Health, 91, 753-760.

Mullen, P.E., Martin, J.L., Anderson, J.C., Romans, S.E., & Herbison, G.P.
(1996). The long-term impact of the physical, emotional, and sexual abuse
of children: a community study. Child Abuse & Neglect, 20 (1), 7-21.

Needell, B., Webster, D., Armijo, M., Lee, S., Cuccaro-Alamin, S., Shaw, T.,
Dawson, W., Piccus, W., Magruder, J., Exel, M., Smith, J., Dunn, A.,
Frerer, K., Putnam Hornstein, E., & Ataie, Y. (2006). Child Welfare
Services Report for California. Retrieved December 19, 2007, from
University of California at Berkeley Center for Social Services Research
website. URL: http://cssr.berkeley.edu/CWSCMSreports/

118
Newton, R., Litrownik, A., & Landsverk, J. (2000). Children and Youth in Foster
Care: Disentangling the Relationship between Problem Behavior and
Number of Placements. Child Abuse & Neglect, 24 (10), 1363-1374.

Nickerson, K.J., Helms, J.E., & Terrell, F. (1994). Cultural Mistrust, Opinions
about Mental Illness, and Black Students’ Attitudes toward Seeking
Psychological Help from White Counselors. Journal of Counseling
Psychology, 41(3), 378-385.

Nock, M.K., Phil, M., & Kazdin, A.E. (2001). Parent Expectancies for Child
Therapy: Assessment and Relation to Participation in Treatment. Journal
of Child and Family Studies, 10(2), 155-180.

Nock, M.K. & Photo, V. (2006). Parent Motivation to Participate in Treatment:
Assessment and Prediction of Subsequent Participation. Journal of Child
and Family Studies, 15(3), 345-358.

Owens, P.L., Hoagwood, K., Horwitz, S.M., Leaf, P.J., Poduska, J.M., Kellam,
S.G., & Ialongo, N.S., (2002). Barriers to Children’s Mental Health
Services. Journal of the American Academy of Child and Adolescent
Psychiatry, 41 (6), 731-738.

Pasztor, E.M., Hollinger, D.S., Inkelas, M., & Halfon, N. (2006). Health and
Mental Health Services for Children in Foster Care: The Central Role of
Foster Parents. Child Welfare Journal, 85 (1), 33-57.

Pavuluri, M.N., Luk, S., & McGee, R. (1996). Help-Seeking for Behavior
Problem by Parents of Preschool Children: A Community Study. Journal
of the American Academy of Child and Adolescent Psychiatry, 35, 215-
222.

Peifer, K.L., Hu, T.W., & Vega, W.A. (2000). Help Seeking by Persons of
Mexican Origin with Functional Impairments. Psychiatric Services, 51,
1293-1298.

Perry, B.D. (2001). The Neurodevelopmental Impact of Violence in Childhood. In
D. Schetky & E. Benedek (Eds). Textbook of Child and Adolescent
Forensic Psychiatry (pp. 221-238). Washington, DC: American
Psychiatric Press. Retrieved April 27, 2006, from
www.childtrauma.or/CTAMATERIALS/Vio_child.asp.

119
Pilowsky, D. (1995). Psychopathology Among Children Placed in Family Foster
Care. Psychiatric Services, 46, 906-910.

Ployhart, R. E., & Oswald, F. L. (2004). Applications of Mean and Covariance
Structure Analysis: Integrating Correlational and Experimental
Approaches. Organizational Research Methods, 7(1), 27.

Radloff, L.S. (1977). The CES-D Scale: A Self-Report Depression Scale for
Research in the General Population. Applied Psychological Measurement,
1, 385-401.

Radloff, L.S. & Teri, L. (1981). Use of the Center for Epidemiological Studies –
Depression Scale with Older Adults. Clinical Gerontologist, 5 (1/2), 119-
136.

Raghavan, R., Leibowitz, A.A., Andersen, R.M., Zima, B. T., Schuster, M.A., &
Landsverk, J. (2006). Effects of Medicaid Managed Care Policies on
Mental Health Service Use among a National Probability Sample of
Children in the Child Welfare System. Children and Youth Services
Review, 28, 1482-1496.

Raghavan, R., Inkelas, M., Franke, T., & Halfon, N. (2007). Administrative
Barriers to the Adoption of High Quality Mental Health Services for
Children in Foster Care: A National Study. Administration and Policy
Mental Health and Mental Health Services Research, 34, 191-201.

Richardson, L.A. (2001). Seeking and Obtaining Mental Health Services: What
Do Parents Expect? Archives of Psychiatric Nursing, 15 (5), 223-231.

Richardson, M.M., Bowers, M.J., Luzicki, T., Krall, D., & Jacobs, A.K. (2008).
Caregivers Involvement in the Intensive Mental Health Program:
Influence on Changes in Child Functioning. Journal of Child and Family
Studies, 17, 241-252.

Rittner, B. (1995). Children on the Move: Placement Patterns of Children under
Children’s Protective Services. Families in Society, 76, 469–477.

Rosenthal, J.A., & Curiel, H.F. (2006). Modeling Behavioral Problems of
Children in the Child Welfare System: Caregiver, Youth, and Teacher
Perspective. Children and Youth Service Review, 28, 1391-1408.

120
Silverman, A.B., Reinherz, H.Z., & Giaconia, R.M. (1996). The Long-Term
Sequelae of Child and Adolescent Abuse: A Longitudinal Community
Study. Child Abuse and Neglect, 20(8), 709-723.

Simms, M.D., Dubowitz, H., & Szilagyi, M.A. (2000). Health Care Needs of
Children in the Foster Care System. Pediatrics, 106(Suppl.), 909-918.

Shore, N, Sim, K.E., LeProhn, N.S., & Keller, T.E. (2002). Foster Parent and
Teacher Assessment of Youth in Kinship and Non-Kinship Foster Care
Placements: Are Behaviors Perceived Differently Across Settings?
Children and Youth Services Review, 24(1/2), 109-134.

Sobel, M.E. (1982). Asymptomatic Confidence Intervals for Indirect Effects in
Structural Equation Models. In S. Leinhart (Ed.). Sociological
Methodology (pp 290-312). San Francisco: Jossey-Bass.

Staudt, M. M. (2003). Mental Health Services Utilization by Maltreated Children:
Research Findings and Recommendations. Child Maltreatment, 8 (3), 195-
203.

Stiffman, A.R., Pescosolido, B., & Cabassa, L.J. (2004). Building a Model to
Understand Youth Service Access: The Gateway Provider Model. Mental
Health Services Research, 6 (4), 189-198.

Swenson, C.C., Brown, E.J., & Sheidow, A.J. (2003). Medical, Legal, and Mental
Health Service Utilization by Physically Abused Children and Their
Caregivers. Child Maltreatment, 8(2), 138-144.

Teicher, M.D. (2000). Wounds that time won’t heal: The neurobiology of child
abuse. Cerebrum: The Dana Forum on brain science, 2(4), 50-67.

Timmer, S.G., Sedlar, G., & Urquiza, A.J. (2004). Challenging Children in Kin
versus Nonkin Foster Care: Perceived Costs and Benefits to Caregivers.
Child Maltreatment, 9 (3), 251-262.

Ullman, J.B. (2001). Structural Equation Modeling. In B.G. Tabacknick & L.S.
Fidell (Eds.). Using Multivariate Statistics (4th ed, pp. 653-771). New
York, NY: Harper Collins.

121
U.S. Department of Health and Human Services (2008). The AFCARS Report:
Preliminary FY 2006 Estimates as of January 2008. Washington, DC:
Government Printing Office. Retrieved February 18, 2008, from
http://www.acf.dhhs.gov/programs/cb/stats_research/afcars/tar/report14.ht
m.

U.S. Department of Health and Human Services (2008). Child Maltreatment 2006.
Washington, DC: Government Printing Office. Retrieved January 18,
2008, from http://www.acf.dhhs.gov/programs/cb/pubs/cm06/index.htm.

U.S. Department of Health and Human Services (2007). Trends in Foster Care
and Adoption FY 2000 – FY 2005. Washington, DC: Government Printing
Office. Retrieved December 26, 2007, from
http://www.acf.hhs.gov/programs/cb/stats_research/index.htm.

Vega, W.A., Kolody, B., & Aguilar-Gaxiola, S. (2001). Help Seeking for Mental
Health Problems Among Mexican Americans. Journal of Immigrant
Health, 3(3), 133-140.

Vega, W.A., Kolody, B., Aguilar-Gaxiola, S., & Catalano, R. (1999). Gaps in
Services Utilization by Mexican Americans with Mental Health Problems.
American Journal of Psychiatry, 156, 928-934.

Yeh, M., Hough, R.L., Fakhry, F., McCabe, K., Lau, A., & Garland, A. (2005).
Why Bother With Beliefs? Examining Relationships between
Race/Ethnicity, Parental Beliefs about Causes of Child Problems, and
Mental Health Service Use. Journal of Counseling and Clinical
Psychology, 73 (5), 800-807.

Yeh, M., Hough, R.L., Garland, A., McCabe, K.M., & Lau, A. (2004). Parental
Beliefs about the Causes of Child Problems: Exploring Racial/Ethnic
Patterns. Journal of the American Academy of Child and Adolescent
Psychiatry, 43, 605-612.

Yeh, M., McCabe, K., Hough, R.L., Dupuis, D., & Hazen, A. (2003).
Racial/Ethnic Differences in Parental Endorsement of Barriers to Mental
Health Services for Youth. Mental Health Services Research, 5, (2), 65-77.

Zima, B. T., Bussing, R., Yang, X., & Belin, T. R. (2000). Help-Seeking Steps
and Service Use for Children in Foster Care. The Journal of Behavioral
Health Services & Research, 27(3), 271.

Abstract (if available)

Abstract Studies have documented that children in foster care are in need of mental health intervention, but not all foster care children receive the mental health services they need. One area that lacks attention and may help explain the underutilization of mental health services is the influence that the caregiver has on a child's utilization of services. Caregivers serve as gatekeepers for children while in the child welfare system, but few studies have focused on the factors that influence a caregiver's use of mental health services for the children under their care. This study represents a step toward addressing this gap in the literature by examining the influence that caregivers have on the utilization of mental health services for children and youth in the child welfare system. This study examined whether the caregivers' perceptions and knowledge of mental health services mediate need and predisposing factors and the utilization of mental health services for children under their care. This study also examined caregiver type as a moderator to understand whether type of caregiver was significant.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Immigrants at a loss: the need for services that promote child well-being among Latino families with child welfare contact

PDF

Implementation of peer providers in integrated health care settings

PDF

The effects of familism and cultural justification on the mental and physical health of family caregivers

PDF

From “soul calling” to calling a therapist: meeting the mental health needs of Hmong youth through the integration of spiritual healing, culturally responsive practice and technology

PDF

Interagency collaboration: cultivating resources and strengthening substance use service delivery in child welfare

PDF

One-two punch: when foster care fails, victims of child abuse and neglect are raised by the criminal justice system

PDF

Sociocultural influences on mental health functioning: implications for the design of community-based services

PDF

Impacts of caregiving on wellbeing among older adults and their spousal caregivers in the United States

PDF

The impact of psychosocial rehabilitation services on recovery outcomes for racial/ethnic minorities with severe mental illness

PDF

Mental Health First

PDF

Empowering rural youth: a comprehensive approach to mental health services in rural America

PDF

The roles of social isolation and secondary caregiver support in the stress process among primary caregivers of older adults

PDF

Enhancing capacity to provide cultural relatability in the in-home mental health service experience for urbanized and minoritized youths through workforce development strategy: a promising practi...

PDF

Strength-Based Reporting: a trauma-informed practice for mandated reporters, to address behavioral health concerns in children at risk of child welfare involvement

PDF

Delivering better care for children with special health care needs: analyses of patient-centered medical home and types of insurance

PDF

Rural minds initiative: navigating mental health wellness together at Gorham Middle/High School

PDF

Mental health advocacy and navigation partnerships: the case for a community collaborative approach

PDF

Post-investigation experiences of families with complex needs at risk of maltreatment: an examination of need, matched services, and changes in need

PDF

Multilevel influences on organizational adoption of innovative practices in children's mental health services

PDF

Neighborhood context and adolescent mental health: development and mechanisms

Asset Metadata

Creator Villagrana, Margarita (author)

Core Title Mental health service use by children and youth in the child welfare system: a focus on need and predisposing factors and caregiver type

School School of Social Work

Degree Doctor of Philosophy

Degree Program Social Work

Publication Date 11/17/2008

Defense Date 10/06/2008

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag caregivers,Child welfare,children and youth,foster care,mental health services,OAI-PMH Harvest

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Palinkas, Lawrence A. (committee chair), Chung, Ruth (committee member), Schneiderman, Janet U. (committee member)

Creator Email maggie1065@hotmail.com,villagrm@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1800

Unique identifier UC1104581

Identifier etd-Villagrana-2472 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-129018 (legacy record id),usctheses-m1800 (legacy record id)

Legacy Identifier etd-Villagrana-2472.pdf

Dmrecord 129018

Document Type Dissertation

Rights Villagrana, Margarita

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu