University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content SOCIOCULTURAL INFLUENCES ON MENTAL HEALTH FUNCTIONING

OVER TIME: IMPLICATIONS FOR THE DESIGN OF COMMUNITY-BASED

SERVICES

by

Tam Quy Thi Dinh

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)

August 2008

Copyright 2008 Tam Quy Thi Dinh

ii

DEDICATION

This dissertation is dedicated to the memory of my loving father and
grandmother (ba ngoai). Through my father, I find the strength and perseverance to
achieve my dreams. It was through his hard work and sacrifices that I am able to
freely pursue my education and integrate my love for learning with my passion for
serving the community. Through my amazingly strong and intelligent grandmother,
definitely a woman beyond her time, I have gained new insights, compassion, and
respect for those with severe and persistent mental illness.
iii

ACKNOWLEDGEMENTS

Despite the oftentimes painfully long drawn out process of completing a
doctoral degree, the last five years have been the most fulfilling years of my life. I
would not have gotten to this point without the love, support, and guidance of certain
people in my life. First, I would like to thank my husband for understanding and
supporting my work and passion. Second, I am thankful for my son, Sebastian, who
through forced playtimes and naptimes has helped me to maintain a sense of balance
and perspective. Third, it is much easier to have a balanced life when you have a
mother like mine. I will never be able to thank or repay my mother for all her home-
cooked meals, care of my family, and nanny duty that allowed me to travel and work
long days and nights. Finally, although my sisters may never fully understand what I
do, their presence always makes my life more complete and meaningful.
I also must acknowledge the faculty at the School of Social Work. The
professors I have had the honor of studying or working with have helped to shape me
into the scholar I am today. I would especially like to thank Dr. John Brekke for
generously providing mentorship and access to data that enabled me to obtain a
NIMH training grant that supported completion of this dissertation.

iv

I am extremely thankful for his continual support and guidance throughout my
doctoral experience and beyond. Lastly, and probably the key reason for my
academic success thus far, is my mentor, dissertation chair, and friend, Dr. Ann
Marie Yamada. She has thoughtfully mentored and unconditionally supported me
through each phase of the doctoral program. She is an inspiration and I feel blessed
to have her as a role model.
v

TABLE OF CONTENTS

PAGE
Dedication ii
Acknowledgements iii
List of Tables viii
List of Figures ix
Abstract x

CHAPTER I INTRODUCTION 1
A. Study Rationale 1
B. Study Purpose 3
C. Study Overview 4
D. Study Aims 5

CHAPTER II LITERATURE REVIEW 6
A. Psychosocial Rehabilitation as a Conceptual 7
Framework
B. Functional Outcomes in PSR 11
1. Independent Living Functioning 12
2. Work Productivity Functioning 14
C. Sociocultural Factors in PSR 16
1. Minority Status as Sociocultural Factor 17
a. Ethnic minority mental health research 18
b. Rationale for minority status as proxy for 19
cultural influences
2. Multidimensional Sociocultural Support Systems 21
in PSR
a. Service support 22
b. Social support 24
c. Intrapsychic support 27
3. Significance 29
D. Preliminary Studies 31
1. Predicting PSR Service Outcomes 31
2 Exploratory Study of Culture, Coping, 32
and Schizophrenia

vi

CHAPTER III METHODS 34
A. Study Aims 35
B. Study Overview 35
C. Study Participants 36
1. Recruitment Sites 36
2. Participant Selection 36
3. Sample Size and Characteristics 37
D. Measurement Instruments and Variables 38
1. Demographics and Clinical Variables 38
2. Functional Outcomes 39
3. Minority Status 40
4. Support System Variables 40
a. Service support 41
b. Social support 41
c. Intrapsychic support 42
E. Data Analysis 43
1. Analytical Plan 43
2. Mplus Latent Growth Curve Modeling 45
3. Methods and Guidelines for Data Interpretation 50

CHAPTER IV RESULTS 51
A. Descriptive Statistics 51
B. Independent Living Functioning: Model Specifications and 57
Parameter Estimates
1. Growth Trajectory: Unconditional Model 58
2. Support Covariates: Conditional Model 62
3. Moderating Effects: Multiple Groups Model 64
4. Summary of Independent Living Functioning Results 68
C. Work Productivity Functioning: Model Specifications and 71
Parameter Estimates
1. Growth Trajectory: Unconditional Model 71
2. Support Covariates: Conditional Model 75
3. Moderating Effects: Multiple Groups Model 77
4. Summary of Work Productivity Functioning Results 80

vii

CHAPTER V DISCUSSION 82
A. Independent Living Functioning 83
1. Growth Trajectory of Independent Living Functioning 83
2. Support Covariates of Independent Living Functioning 84
a. Intrapsychic support 84
b. Social support 86
c. Service support 88
3. Moderating Effects of Minority Status 90
on Independent Living Functioning
a. Social support 91
B. Work Productivity Functioning 93
1. Growth Trajectory of Work Productivity Functioning 93
2. Support Covariates of Work Productivity Functioning 95
a. Intrapsychic support 95
b. Social support 96
c. Service support 97

3. Moderating Effects of Minority Status 98
on Work Productivity Functioning
a. Social support 98
b. Intrapsychic support 99
C. Within Group Comparisons 100
D. Study Strengths 102
E. Study Limitations 104
F. Future Research Direction 106
G. Implications for Community-based Services 110
H. Conclusion 112

REFERENCES 114

APPENDIX Table 15. Indices for Alternative Work Productivity Functioning 133
Models

viii

LIST OF TABLES
PAGE
Table 1 Fifteen U.S. Principles of Psychosocial Rehabilitation 8
Table 2 Fifteen Principles of Community-based PSR Programs 9

Table 3 Characteristics of Total Sample at Baseline, 6 Months, 54
and 12 Months

Table 4 Bivariate Comparisons Between Tri-ethnic Groups at Baseline 55

Table 5 Bivariate Comparsions By Minority Status at Baseline 56

Table 6 Correlation Table of Support Covariates and Functional Outcomes 57

Table 7 Fit Indices for Pertinent Independent Living Functioning Models 59

Table 8 Linear Unconditional Model Parameter Estimates 60
for Independent Living Functioning

Table 9 Conditional Model Parameter Estimates 64
for Independent Living Functioning

Table 10 Multiple Group Model Parameter Estimates 68
for Independent Living Functioning

Table 11 Fit indices for Work Productivity Functioning Models 70

Table 12 Nonlinear Unconditional Model Parameter Estimates 73
for Work Productivity Functioning

Table 13 Conditional Model Parameter Estimates 77
for Work Productivity Functioning

Table 14 Multiple Group Model Parameter Estimates 80
for Work Productivity Functioning

Table 15 Fit Indices for Alternative Work Productivity 133
Functioning Models

ix

LIST OF FIGURES

PAGE
Figure 1 Multi-systems PSR Conceptual Framework for Sociocultural 31
Influences on Functional Outcomes

Figure 2 A Conditional Model of Sociocultural Influences 47
on Functional Outcomes

Figure 3 A Multiple Group Model of Sociocultural Influences 49
on Functional Outcomes

Figure 4 Independent Living Functioning Score Over Time 61
for Sample

Figure 5 Random Independent Living Functioning Trajectories 62
of N=50 Participants

Figure 6 Independent Living Functioning Over Time 65
by Minority/Non-minority Groups

Figure 7 Work Productivity Functioning Score Over Time 74
for Sample

Figure 8 Random Work Productivity Functioning Trajectories 74
of N=50 Participants

Figure 9 Work Productivity Functioning Over Time 78
by Minority/Non-minority Groups

x

ABSTRACT

Despite the promising effects of psychosocial rehabilitation (PSR)
interventions on outcomes for individuals with severe and persistent mental illness
(SPMI), wide variation still exists in individual functional outcomes. This is
particularly true for ethnic minorities as empirical evidence has documented
persistent racial disparities in mental health outcomes. In this dissertation, the author
examines sociocultural factors—support systems and minority status —that
influence rehabilitative independent living and employment functional outcomes for
individuals with SPMI, using secondary data from a longitudinal study conducted at
four community-based psychosocial rehabilitation programs in Southern California.
Based on a psychosocial rehabilitation framework, available support systems
include support from PSR service system, social support from family and friends,
and personal intrapsychic support. Minority status will be treated as a proxy variable
for cultural influences to determine its moderating influence on support and
outcomes. The three specific aims are to (a) illustrate change patterns in treatment
outcomes over twelve months; (b) identify the strongest support determinants
(service support, social support, or intrapsychic support) for each functional
outcome; and (c) explore whether minority status exerts a significant moderating
effect on the relationship between the support variables and functional outcomes.
Muthen and Muthen’s (2007) Mplus Version 5 multivariate repeated model approach
xi

is used to run a series of latent growth curve models to study the functional outcome
growth structure and the predictors.
The study findings suggest that the functional change mechanism is a
complex issue. While independent living functioning exhibited continuous linear
growth from baseline to twelve months, work productivity experienced a slight
decrease after 6 months (Aim 1). Intrapsychic support explained most of the
variance for both functional outcomes at baseline (Aim 2). However it is social
support that has the most influence on the growth trajectory for both independent
living and work productivity (Aim 2). Minority status was found to moderate some
of the relationship between the available support systems and functional outcomes
(Aim 3). It is hoped that this study will contribute to the knowledge base needed to
facilitate the development of effective and culturally responsive psychosocial
rehabilitative interventions for ethnic minorities experiencing SPMI.
1

CHAPTER I
INTRODUCTION

A. Study Rationale
Psychosocial rehabilitation (PSR) interventions are receiving a great deal of
attention due to evidence of their positive effects on key outcomes for individuals
with severe and persistent mental illness (SPMI) (Brekke & Long, 2000; Bustillo,
Lauriello, Horan, & Keith, 2001; Penn & Mueser, 1996). Despite evidence of the
effectiveness of PSR on reducing relapse and re-hospitalization for individuals with
SPMI (Bustillo et al., 2001; Yamada, Korman, & Hughes, 2000), findings on the
impact of PSR services on functional outcomes such as independent living and
employment are diverse and conflicting (Brekke & Long, 2000; Bustillo et al., 2001;
Cook et al., 1996; Lauriello, Bustillo, & Keith, 1999; Phillips, Barrio, & Brekke,
2001). Even less is understood about functional outcomes for ethnic minorities.
Given the rapid growth of ethnic minority populations in the USA, the growing body
of evidence of ethnic/racial disparities in mental health services (Kupfer, Frank,
Grochocinski, Houck, & Brown, 2005; USDHHS, 2001), and resounding support for
evidence based research (Bellack, 2006; Ghaemi, 2008), there is a clear need for
research on cultural influences associated with ethnicity as they pertain to PRS
processes and outcomes (Lopez, 2002).
2

Understanding cultural influences on PSR outcomes requires identification of
contexts in which functioning occurs (Hitchcock et al., 2005; Hohmann & Shear,
2002; NIMH Workgroup, 2000). Those contexts include psychological, social,
ecological, and environmental mechanisms (and their interactions) that attenuate or
exacerbate outcomes (National Advisory Mental Health Council [NAMHC], 1999;
Steinberg & Avenevoli, 2000; Super & Harkness, 1999; USDHHS, 2001).
Sociocultural factors, consisting of combinations of social and cultural factors, are
believed to account for the wide variation in functional outcomes that currently
exists (Bhugra, 2006), especially across ethnic groups (Barrio, 2001; Snowden &
Yamada, 2005).
One example of a sociocultural factor that may affect PRS interventions is
the consumer support systems. In psychosocial rehabilitation, these support systems
are the available support that can be mobilized and sustained to improve functional
skills considered necessary for meaningful community life (Crosse, 2003; Spaniol &
Koehler, 1994). There is considerable theoretical and empirical evidence indicating
that support is vitally important to individuals functioning, especially those with
SPMI (Buchanan, 1995; Carpentier & White, 2002). Given that the principal focus of
PSR is on interactions between the individuals, their interpersonal networks, and the
social context (Davidson, O’Connell, Tondora, Lawless, & Evans, 2005; Spaniol &
Koehler, 1994), support in PSR interventions is acknowledged as being multi-
faceted. Support is conceptualized as coming from service providers, family and
3

friends, and intrapsychic characteristics. Although PSR treatment consists of a
combination of these support systems, most efforts to study PSR interventions have
either focused on comprehensive PSR programs, skills training, or cognitive
behavioral therapy rather than consumer support systems (Brekke, Long, Nesbitt, &
Sobel, 1997; Dilk & Bond, 1996; Penn, Waldheter, Perkins, Mueser, & Lieberman,
2005). In studies where support has been a focus, researchers have generally failed
to examine support systems from a psychosocial rehabilitation framework,
concentrating instead on single-system analyses of either PSR service support
(Barton, 1999; Brekke, Ansel, Long, Slade, & Weinstein, 1999; Brekke & Long,
2000) or social support from family and friends (Corrigan & Phelan, 2004; Evert,
Harvey, Trauer, & Herrman, 2003). These single-system analyses combined with
insufficient evidence-based research on available support systems hinder our
understanding of the effectiveness and range of the support systems in PSR
treatments.

B. Study Purpose
Increased emphasis on brief, cost-effective targeted interventions calls for
new efforts to isolate consumer support systems in order to identify the “effective
ingredients” that fit well with the multi-faceted nature of PSR (Penn et al., 2005). To
advance the current literature, this dissertation will test a more comprehensive model
of the multidimensional support systems in PSR interventions and examine how they
4

are moderated by minority status. The goal is to provide further insight to the change
mechanisms for functional outcomes and to create a foundation for future
examination of specific sociocultural constructs and contextual influences. This
study will also contribute to the evidence-based research literature on treatment
outcomes for individuals with SPMI. It is hoped that this research will assist in the
effort to reduce and eventually eliminate ethnic/racial disparities in mental health
service outcomes.

C. Study Overview
In response to recent NIMH directives (Division of Services and Intervention
Research, n.d. [DSIR]; Miranda, Nakamura, & Bernal, 2003) and inconsistent
findings regarding functional outcomes in PSR intervention research (Brekke &
Long, 2000; Bustillo et al., 2001; Cook et al., 1996; Lauriello et al., 1999; Phillips et
al., 2001), this study will apply a psychosocial rehabilitation framework to examine
sociocultural support systems, their relationships, and change trajectories for
rehabilitative functional outcomes for a sample of individuals with SPMI. Minority
status, used as a proxy variable for cultural influences (Phinney, 1996), will be
examined as a moderator on the relationship between systems of support and two
functional outcomes- independent living and work productivity. This dissertation
study is based on multiethnic data from a longitudinal research study conducted
between 1996 to 2001 at four community-based psychosocial rehabilitation programs
5

in Southern California, all serving individuals with SPMI. Overall, this study is
designed to clarify the influences of sociocultural factors (support systems and
minority status) on change mechanisms for functional outcomes in PSR.

D. Study Aims
The three specific aims for this study are to use latent growth curve modeling
as an analytical framework to

1. Illustrate change patterns in functional outcomes (independent living and
work productivity) over a twelve month period of PSR.

2. Identify the strongest support determinant or determinants (service support,
social support, or intrapsychic support) for each psychosocial functional
outcome for individuals with SPMI at baseline and 12 months.

3. Explore whether minority status exerts a significant moderating effect on
the relationship between support systems and functional outcomes for
individuals with SPMI.
6

CHAPTER II
LITERATURE REVIEW

The four topics addressed in this literature review are the psychosocial
rehabilitative framework, functional outcomes, sociocultural factors in psychosocial
rehabilitation, and preliminary studies. The purpose of this review is to present a
comprehensive overview of the PSR framework components and sociocultural
variables that are the focus of this dissertation study.
The objective of the first section is to provide a thorough conceptual
background from which the dissertation’s aims were generated. The conceptual
framework also allows for a clear understanding of the study variables’
multidimensionality and significance. The second and third sections summarize
current studies on support systems and disparities in PSR care for ethnic minorities.
The final section presents an overview of studies that established the direction for
this dissertation.

7

A. Psychosocial Rehabilitation as a Conceptual Framework
Psychosocial rehabilitation (PSR) has been defined as:
The process of facilitating an individual’s restoration to an optimal
level of independent functioning in the community…While the nature
of the process and the methods used differ in different setting,
psychosocial rehabilitation invariably encourages persons to
participate actively with others in the attainment of mental health and
social competence goals…The process emphasizes the wholeness and
wellness of the individual and seeks a comprehensive approach to the
provision of vocation, residential, social/recreation, education and
personal adjustment services. (Cnaan, Blankertz, Messinger, &
Gardner, 1988, p61).

Despite variation in process and methods, it has been argued that effective PSR
should incorporate or emphasize these ten essential elements: 1) individualization, 2)
hope, 3) consumer-directed, 4) focus on skills, 5) strengths and abilities, 6)
environomentally specific, 7) environmental support, 8) partnership approach, 9) on-
going support, and 10) skills and supports in balance (International Journal of
Psychosocial Rehabilitation website [IJPR], 2005). After an extensive review of the
PSR literature, Cnaan and his co-authors proposed thirteen principles of PSR (Cnaan
et al., 1988) which were ultimately built upon to become the fifteen guiding
principles of PSR services in the United States (Wimmerapdss website, March 3,
2008).

8

Table 1. Fifteen U.S. Principles of Psychosocial Rehabilitation
1 Utilization of full human capacity.
2 Equipping people with skills (social, vocational, educational, interpersonal and
others).
3 People have the right and responsibility for self-determination.
4 Services should be provided in as normalized environment as possible.
5 Differential needs and care.
6 Commitment from staff members.
7 Care is provided in an intimate environment without professional, authoritative
shield and barriers.
8 Early intervention.
9 Environmental approach.
10 Changing the environment.
11 No limits on participation.
12 Work-centered process.
13 There is an emphasis on a social rather than a medical model of care.
14 Emphasis is on the consumer’s strengths rather than on pathologies.
15 Emphasis is on the here and now rather than on problems from the past.

9

Community based organizations further transformed these principles to fit the unique
characteristics of the community managed PSR sector (Wimmerapdss website,
March 3, 2008).
Table 2. Fifteen Principles of Community-based PSR Programs
1 Flexibility of structure and service models.
2 Non-obligatory attendance.
3 Support for mobility and choice of service options.
4 Active participant involvement in services.
5 Support for participant decision-making.
6 Concentration on quality of relationships and interactions between participants
and staff.
7 Encouragement of peer support.
8 Responsiveness to participants’ needs.
9 Provision of most 'normal' environment.
10 Effective psychosocial rehabilitation.
11 Autonomous community accountability.
12 Utilization of a broad range of skills.
13 Active community education function.
14 Active advocacy function.
15 Cost-effectiveness: both operational and preventative.

Considered a mix of psychosocial and rehabilitative theory (Drake, Green,
Mueser, & Goldman, 2003), PSR is regarded as a paradigm shift from a medical
model focused on biological influences to a recovery-oriented model focused on the
impact of environmental factors (Bellack, 2006; USDHHS, 2001). Recovery for
individuals with SPMI is not a return to a state preceding the onset of mental illness
or symptom remission (Davidson et al., 2005), but a “process by which people with
psychiatric disabilities rebuild and further develop important personal, social,
environmental, and spiritual connections” (Spaniol & Koehler, 1994, p.1) while
10

living with the “effects and side effects of their mental illness” (Davidson et al.,
2005, p. 483).
The primary objective of PSR treatment is building and improving necessary
functional skills leading to meaningful lives in the community (Bachrach, 2000;
Penn et al., 2005). Consistent with psychosocial influences, PSR interventions are
designed to mobilize and sustain support systems (IJPR website, 2005; Spaniol &
Koehler, 1994). At the center of PSR is the individual consumer—in other words,
the intervention process is consumer-driven and tailored to individual needs.
Psychosocial rehabilitation providers recognize that individuals with SPMI are active
and reactive agents moving within complex, dynamic multidimensional care
systems. Transactional relationships between individuals and their environments are
considered the essence of psychosocial rehabilitation (Bachrach, 2000; Crosse,
2003).
Research on PSR interventions has produced promising findings in terms of
community reintegration for participants who receive services (USDHHS, 2001).
There is significant evidence indicating that PSR services reduce psychiatric
symptoms and increases cognitive and functional outcomes for individuals with
SPMI (Barton, 1999; Penn et al., 2005), but less research evidence exists indicating
that PSR programs generally result in positive outcomes for ethnic minorities as the
number of studies conducted on this topic is small (Baker, Stokes-Thompson, Davis,
Gonzo, & Hishinuma, 1999; Imada & Schiavo, 2005; Vega & Lopez, 2001).
11

The PSR characteristics and principles mentioned above provide a conceptual
framework to understand the importance of the relationship between the support
systems and functional outcomes of individuals with SPMI. The dynamic
relationship between individuals and the multiple facets of their environment guide
this study’s rationale that support available to individuals with SPMI exists on
multiple levels ranging from larger networks and institutions (service support), to
interpersonal (social support) to internal (intrapsychic support).

B. Functional Outcomes in Psychosocial Rehabilitation
Functional outcomes refer to the assessment of personal skills needed for
meaningful participation in the community (Bellack, 2006). Among potential PSR
treatment outcomes, functional outcomes represent a distinct domain (Brekke &
Long, 2000; Rosenblatt & Attkisson, 1993). The successful attainment of individual
functional outcome domains (e.g., independent living and work productivity) is
necessary for the stability and overall functioning of individuals with SPMI (Mueser
et al., 1997). In addition to perceiving functional outcomes as important to their well-
being (Hansson et al., 2002; Malik, Reeves, & Dellario, 1998), individuals with
SPMI have repeatedly indicated their preference for independent living (Tanzman,
1993) and employment (Mueser, Salyers, & Mueser, 2001). Some researchers argue
that functional outcomes are more accurate recovery criteria than clinical outcomes
for individuals with SPMI (Liberman, Kopelowicz, Ventura, & Gutkind, 2002) and
12

are more consistent with PSR goals to improve vocational status and independent
living (McKibbin, Brekke, Sires, Jeste, & Patterson, 2004; Rosenblatt & Attkisson,
1993).
Despite advancements in PSR interventions and research, wide variation
remains in terms of functional outcome predictors and the variability in the impact of
those predictors on functional outcomes (Brekke & Long, 2000; Bustillo et al., 2001;
Hoffman & Kupper, 1997). Outcome variability may reflect a core PSR framework
feature that integrates multiple sociocultural factors such as settings and service
modalities with consumer capacity to address multiple functional outcome goals
(Barton, 1999; Cook et al., 1996). Perhaps due to the complex interplay among these
variables, no systematic examination of how sociocultural factors influence the
rehabilitative process and subsequent functional outcomes has been published. Given
the importance of helping consumers achieve and maintain successful functioning in
their communities, the overall purpose of this study is to increase available
knowledge of how sociocultural factors such as support systems and minority status
impact the functional outcomes of individuals with SPMI.

1. Independent Living Functioning
In the PSR context, independent living functioning refers to the self-care
skills necessary for self-determined residence in the community (Goodman, Sewell,
Cooley, & Leavitt, 1993; Wright, 1980). Independent living is considered a first step
13

toward community integration for individuals with SPMI (Fellin, 1993; Flynn &
Aubry, 1999, Wong & Solomon, 2002). Self-care skills deficit is considered
characteristic of severe mental illness (Iyer, Rothman, Vogler, & Spaulding, 2005)
and individuals incapable of maintaining self-care activities are more likely to
become homeless or moved to institutionalized environments where service
providers assist with daily living activities (Nelson, Aubry, & Lafrance, 2007;
O’Hara, 2007). The diversity of existing housing options ranges from assisted living
at board and care, to supported housing in a group home, to independent living in
apartments in the community (Fakhoury, Murray, Shepard, & Priebe, 2002). For this
reason, functioning outcome assessments that only examine “living in the
community” provide little information on independent living functioning levels;
consequently, there is considerable variability in findings (Newman, Reschovsky,
Kaneda, & Hendrick, 1994).
A substantial amount of empirical evidence supports the many benefits of
independent community living (Fakhoury et al., 2002; Hansson et al., 2002; Wong &
Solomon, 2002). Hansson et al., (2002) report that individuals with SPMI in
independent living situations have subjectively better quality of life, shorter illness
duration, and better psychosocial functioning than those living in non-independent
situations. Other positive outcomes of living in supportive independent housing
include reduced homelessness, fewer service gaps, and lower hospitalization rates
(Wong & Solomon, 2002). From a policy perspective, community housing is often
14

considered a cost-effective alternative to hospitalization and in-patient treatment
(Harkness, Newman, & Salkever, 2004).
Variability in outcomes may be due to the wide range of research foci and
expansive definitions of “independent living”. Studies on living situation outcomes
for individuals with SPMI have mainly focused on assessment of one of the
following: length of time in community, desire for and satisfaction with independent
living; level of independence, and stability of living situation (Hansson et al., 2002;
Nelson, 2007; Tanzman, 1993; Yamada et al., 2000). In this study, independent
living functioning is conceptualized as having both the skills to take care of one self
and also the ability to utilize these skills in a living situation other than on the street
or in an inpatient treatment programs or hospitals. Current independent living
functioning is observed at three time points over a twelve month period (Aim 1).

2. Work Productivity Functioning
Work productivity functional outcome is conceptualized as possessing
necessary work skills in order to find and maintain employment (Goodman et al.,
1993; Iyer et al., 2005). Despite evidence-based practices designed to assist
individuals with SPMI obtain and maintain employment, the attainment of
employment is filled with obstacles and barriers (Cook, 2006; O’Hara, 2007).
Results from a recent longitudinal multi-site study of the US Schizophrenia Care and
Assessment program, with a sample of 2,300 individuals, indicated an employment
15

rate of only 17.2 percent (Salkever et al., 2007); of those with jobs, only 57.1%
worked 40 hours or more per week. Although the rate of employment is extremely
low for individuals with SPMI and the amount of work may vary, many individuals
express a desire to work (Auerback & Richardson, 2005; Mueser et al., 2001).
The extensive literature on work for individuals with SPMI reflects the
importance society place on work as a recovery outcome (Marrone & Golowka,
1999; Tsang, Lam, Ng, & Leung, 2000). While vocational programs may differ in
their level and type of support, most share the end goal of assisting individuals with
SPMI obtain and maintain employment (Tsang et al., 2000). There is strong
evidence that supported employment programs are the most successful in attaining
employment for individuals with SPMI (Burke-Miller et al., 2006; Drake et al., 2003,
2006; Mueser et al., 2004).
There are many personal and clinical benefits of sustaining employment
(Auerback & Richardson, 2005; Bell, Fiszdon, Greig, & Bryson, 2005). Auerback
and Richardson (2005) found that consumers gained a sense of normalcy and
working gave them a personal identity. Consumers in Provencher, Gregg, Mead, and
Mueser’s qualitative study (2002) also found that work increased their self esteem
and sense of personal competence. Benefits from employment for SPMI individuals
appear to extend to older individuals (Bell et al., 2005) and ethnic minorities (Baker
et al., 1999). Specifically, Bell et al. (2005) found that older individuals with SPMI
who participated in the six-month work therapy program performed at least as well
16

as their younger counterparts. In addition to clinical improvements, higher quality of
life was reported by both younger and older participants in this study by Bell and
colleagues.
Similar to independent living research, findings on work outcomes for
individuals with SPMI tend to be uneven due to the wide range of research foci,
expansive definitions of “work outcome” (Honey, 2000), and the lack of
standardized measures (Cook, 2006). Recent work outcome studies have focused on
length of employment, desire for employment, job satisfaction, level of
independence, and job stability (mostly measured by the number of jobs held over a
pre-defined period (Becker, Drake, Bond, Xie, Dain, & Harrison, 1998; Mueser et
al., 2001). In this study, work productivity functioning, conceptualized as possessing
necessary work skills in order to find and maintain employment, is observed at three
time points over a twelve month period (Aim 1).

C. Sociocultural Factors in Psychosocial Rehabilitation
Sociocultural factors, consisting of combinations of social and cultural
factors, are believed to account for the wide variation in functional outcomes
(USDHHS, 2001). With the shift from traditional medical models to recovery-
oriented models such as PSR, sociocultural factors are becoming more salient to
understanding treatment outcomes. Such factors are often operationalized as
anything within one’s environment, including interactions among factors at the
17

individual, family, sociocultural, service system, and organizational levels (NIMH
Workgroup, 2000). The contributory weight of specific sociocultural factors is
difficult to assess due to their multidimensional and interrelated nature (Barrio, 1998;
Nathan, Wylie, & Marsella, 2001; Snowden, 2003) One example of a sociocultural
factor that affects psychosocial rehabilitation functioning is consumer support
systems (Bengtsson-Tops & Hansson, 2001; Carpentier & White, 2002). The
support systems are in turn affected by other social and cultural influences associated
with consumer ethnic background (USDHHS, 2001). The next section focuses on
sociocultural domains that may relate to functional outcomes.

1. Minority Status as a Sociocultural Factor
A fundamental problem in cross-cultural research is operationalizing
sociocultural factors- that is, identifying constructs that represent culture (Barrio,
1998; Betancourt & Lopez, 1993). Complex sociocultural interactions surrounding
cultural experiences and ethnic/racial status are complex factors in the mental health
experiences of ethnic minorities (Barrio, 2000; Sue & Chu, 2003). In a Supplemental
Report on Mental Health, the Office of the US Surgeon General (USDHHS, 2001)
acknowledges the influence of culture on many aspect of mental illness, including
conceptualization, symptom manifestation, help-seeking decisions, treatment
experiences, and coping styles. According to that report, the lack of understanding of
cultural influences leads to ethnically based disparities in mental health care use and
18

outcomes (USDHHS, 2001; Vega & Lopez, 2001). These disparities have been
documented in terms of mental health service underutilization (Barrio et al., 2003;
Diala et al., 2000), unmet treatment needs (Wang et al., 2005), and both shorter
treatment participation periods, and higher dropout rates (Diala et al., 2000).
Combined, this evidence suggests that ethnic minorities are not getting optimal
treatment or services (Wang et al., 2005).
(a) Ethnic minority mental health research.
Researchers who have identified multiple barriers to PSR program access and
utilization report that ethnic-related characteristics play a significant role (Folsom et
al., 2007; USDHHS, 2001). Since PSR services often reflect mainstream Euro-
American cultural ideologies and characteristics, ethnic-specific needs may be
overlooked (Barrio & Yamada, 2005). Cultural differences between clinicians and
ethnic minority consumers can result in mistaken attributions of the nature and
causes of mental illness (Nathan et al., 2001). Biases and insufficient cultural
awareness among service providers have also been raised as important issues in the
literature on minority consumers (Snowden, 2003). The combination of
socioeconomic factors, cultural misunderstandings, and culturally inappropriate
treatment plans may result in lower service use by and poorer treatment outcomes for
ethnic minorities (Snowden, 2003; Telles et al., 1995; USDHHS, 2001)
Phillips et al., (2001) and Rogers, Anthony, Cohen, and Davies (1997) are
among those researchers who have documented ethnic disparities in outcomes and
19

reported that ethnic minorities with SPMI have poorer social functioning skills and
work outcomes than Euro-Americans. Over time, ethnic minorities may also exhibit
greater decline in their functioning outcomes than Euro-Americans. According to
Phillips and colleagues (2001) and Bae, Brekke, and Bola (2004), African Americans
showed the least improvement in social functioning, and Latinos had the greatest
decline in work functioning following PSR treatment. Compared to ethnic
minorities, Euro-Americans generally spend longer periods of time in the community
prior to re-hospitalization (Yamada et al., 2000). One aspect of poorer treatment
outcomes is that ethnic minorities with SPMI are more likely to experience relapses
leading to inpatient hospitalizations and psychiatric emergency room visits
(Snowden, 2003).
(b) Rationale for minority status as proxy for cultural influences.
Despite the rapidly growing number of ethnic minorities in the United States
(Hobbs & Stoops, 2002; Reeves & Bennett, 2004) and legislation requiring the
inclusion of ethnic minorities in federally funded research (Miranda et al., 2003),
ethnic minorities remain significantly under-represented in mental health research
(Anderson, Harada, Chiu, & Makinodan, 1995; Imada & Schiavo, 2005; Mak, Law,
Alvidrez, & Perez Stable, 2007; Vega & Lopez, 2001). There is evidence suggesting
that the under-representation of ethnic minorities in research makes it difficult to
understand sociocultural variation in the development of mental illness across
different ethnic groups (USDHHS, 2001). In addition, inadequate ethnic minority
20

sampling limits the ability to generalize findings (Patel, Doku, & Tennakoon, 2003),
resulting in the possibility of continued ineffective treatment for ethnically diverse
individuals and racial/ethnic disparities in treatment outcomes (Kupfer et al., 2005;
USDHHS, 2001).
Despite the concerted effort to increase the number of ethnic minority
participants in research studies, researchers continue to report challenges to
recruitment (Janson, Alioto, & Boushey, 2001). The recruitment literature discusses
an array of sociocultural barriers that influence willingness to participate in research
studies (Hinton, Guo, Hillygus, & Levkoff, 2000; Shavers, Lynch, & Burmeister,
2002; Tu, Chen, Chen, Lim, Ma, & Drescher, 2005). Mistrust of studies and
researchers is often cited as a major factor, especially among African Americans who
know of many examples of abuse, including the Tuskegee Study (Shavers et al.,
2001). Socioeconomic factors such as competing family responsibilities and lack of
transportation have also been given as reasons for not participating in research (Tu et
al., 2005). Sociocultural factors mentioned in the literature include linguistic
barriers, lack of knowledge about the research process, and stigmatization (Hinton et
al., 2000; Tu et al., 2005).
In the USA, African Americans and Latinos share some similar “experiences
associated with minority status, such as powerlessness, discrimination, and
prejudice” (Phinney, 1996, p.919). African Americans and Latinos also share more
collective/interdependent family values than Euro-Americans (Lefley, 1990). In
21

addition, they also share relatively similar outpatient service disparities and minority-
related mental health outcomes (Guarnaccia & Parra, 1996; Singelis, 1994).
Although the disparities noted for both African Americans and Latinos are not
directly attributable to ethnic heritage, it has been surmised to relate to being of
minority status (Phinney, 1996). Using minority status as a proxy variable for
cultural influence is an important initial step towards understanding ethnic/racial
disparities in functional outcomes.

2. Multidimensional Sociocultural Support Systems in PSR
Support is a crucial component of psychosocial rehabilitative interventions
(IJPR website, 2005; Spaniol & Koehler, 1994). Given the number of PSR
intervention modalities, support is conceptualized as multidimensional systems
within one’s self or in the environment that promote favorable illness outcomes
(Bachrach, 2000; Beels, 1981; Crosse, 2003). The different PSR support systems
mobilized and utilized include: service providers (intensity of service delivery;
friends and family; and the consumer’s internal drive (motivation and sense of
purpose). The support literature contains considerable evidence showing a positive
relationship between support quantity/quality and sustained mental health (Cohen &
Wills, 1985; Greenberg & Rosenheck, 2005), the ability to cope with stress and
increased quality of life (Bengtsson-Tops & Hansson, 2001; Bracke, Christiaens, &
Verhaeghe, 2008; Caron, Lecomte, Stip, & Renaud, 2005) and functional outcomes
22

(Brekke et al., 1997, 1999). Theoretically and empirically, support is vitally
important to an individual’s functioning, especially for an individual with SPMI
(Buchanan, 1995; Carpentier & White, 2002). However, most previously published
PSR research has focused on intervention effectiveness (Dilk & Bond, 1996) rather
than on examining the causal relationship of consumer support systems and
mechanisms of change. There has also been a propensity for single, rather than
multiple systems analyses that are more likely to reflect actual support resources
available to individuals with SPMI. Using psychosocial rehabilitation as a conceptual
framework of this study allows for a more complete examination of available support
resources. The three systems of available support discussed next are service support
from service providers, social support from family and friends, and internal
intrapsychic support from within consumers with SPMI.
(a) Service support.
Service support refers to support provided by PSR service providers and is
measured in terms of contact quantity (Brekke et al., 1997, 1999). Program
evaluations have demonstrated a strong need among stakeholders in mental health
systems to find cost-effective ways of providing treatment programs to their
consumers (Solomon, 1992; USDHHS, 2001). Typically the unit of analysis is the
individual treatment program (Solomon, 1992), with the goal being able to determine
program effectiveness in producing positive outcomes for individuals with SPMI.
Only recently have researchers started to look at treatment program characteristics to
23

determine relationships between specific components or conditions of programs and
positive outcomes (Brekke et al., 1997, 1997; Greenberg & Rosenheck, 2005).
Characteristics of psychosocial services include modality and focus, quantity,
and quality as perceived by PSR consumers (Brekke et al., 1997, 1999). Service
quantity is usually measured in terms of contact frequency, intensity, and length of
service time (Brekke et al., 1997, 1999; Dietzen & Bond, 1993). While a positive
relationship exists between frequency of office visits and consumer satisfaction level,
wide variation has been reported in the relationship between service quantity and
functional outcomes (Dietzen & Bond, 1993). Consumers participating in intensive
treatment programs have been shown to have better improvement on select
functional outcomes (Brekke et al., 1997; Greenberg & Rosenheck, 2005). Follow-
up studies indicate that these improvements were sustained at six and twelve months
(Brekke et al., 1999). Findings on the impact of the length of service time are more
mixed, with some evidence indicating a long-term positive impact on independent
living and social functioning, but not on employment status (Brekke et al., 1999).
Less effort has been made to determine if service quantity is equally obtained
across ethnically diverse consumers (NAMHC, 1999). Given the potential
importance of service quantity on functional outcomes, it is important to understand
whether consumer-provider relationships differ by ethnic background. Since the
bottom line of all service interactions is the connection between consumer and
service provider (Cooper, Brown, Vu, Ford, & Powe, 2001) and working alliance is
24

recognized as a strong predictor of functional outcomes (Hopkins & Ramsundar,
2006), stereotypes and lack of cultural awareness on the part of providers can result
in subtle service biases (Snowden, 2003). African American consumers may believe
that they experience lower quality service due to their race (Copeland, 2005). The
USDHHS (2001) also notes that stigma and cultural beliefs can influence consumer-
provider working relationships and consequently affect service quantity and quality.
Since positive consumer- provider relationships are crucial for PSR treatment
success (Bachrach, 2000; Crosse, 2003), the impact of minority status on service
quantity must be examined. In response to the wide variation in findings in this area,
this dissertation study will focus on determining the relationship between service
support (defined as “service intensity”) and functional outcomes (Aim 2) and will
examine the effect of ethnicity on this relationship (Aim 3).
(b) Social support.
Social support refers to instrumental and emotional support provided by
family and friends (Bengtsson-Tops & Hansson, 2001; Buchanan, 1995; Schwarzer,
Knoll, & Rieckmann, 2003; Walsh & Connelly, 1996). Social support has been
linked to mental health maintenance (Cohen & Wills, 1985), stress-coping capacity,
and quality of life (Bengtsson-Tops & Hansson, 2001; Caron et al., 2005). Empirical
social support studies have mostly focused on actual received support or perceived
available support (Buchanan, 1995; Schwarzer et al., 2003), with network type,
network density, and subjective social support experiences the most frequently used
25

indices (Albert, Becker, McCrone, & Thornicroft, 1998). The multiple dimensions of
social support make it difficult to precisely measure the effect of each individual
facet (Walsh & Connelly, 1996), therefore, researchers have reported a range of
findings on the importance of support.
Results from empirical studies on actual support for individuals with SPMI
have also been wide-ranging. For example, some researchers report that smaller
networks are associated with higher inpatient use and increased severity of clinical
symptoms (Albert et al., 1998) and larger networks are associated with positive self-
esteem and quality of life (Corrigan & Phelan, 2004; Goldberg, Rollins, & Lehman,
2003). However, little evidence has been found in support of a link between network
size and functional outcomes (i.e, occupational and social functioning) (Erickson,
Beiser, & Iacono, 1998). Findings on the significance of the type of network (e.g.
family vs. peer network, high density vs. low density) are also widespread. Family
members are most often the main source of emotional and financial support for
persons with SPMI (Bengtsson-Tops & Hansson, 2001; Walsh & Connelly, 1996). A
relationship has been reported between familial involvement and increased rates of
full-time employment (Evert et al., 2003), but neither family network density
(Erickson et al., 1998) nor family support (Yamada et al., 2000) have been verified
as good predictors of functional outcomes for community stability. In contrast,
Erickson et al. report that density and degree of supportiveness from non-kin
relationships are beneficial for consumer outcomes, even at five-year follow up.
26

Findings on perceived support are more consistent. Carpentier and White
(2002) found that perception of family cohesion is very important to initial help-
seeking behaviors and service continuation among consumers. Bengtsson-Tops and
Hansson (2001) are among several researchers who have consistently found
perceived support to be a good predictor of positive outcomes. Perceived support is
also viewed as a significant indicator of employment (Rogers, Anthony, & Lyass,
2004; Yanos, Rosenfield, & Horwitz, 2001), community adaptation (Clinton,
Lunney, Edwards, Weir, & Barr, 1998), and five-year global functional outcomes for
persons experiencing a first-episode of schizophrenia (Erickson et al., 1998).
The cultural influences embedded in ethnicity further compound the
complexity of the role of social support in consumer outcomes. Weisman, Rosales,
Kymalainen, and Armesto (2005) found that perceived family cohesion is associated
with higher emotional distress for Latinos and African Americans with SPMI but not
for their Euro-American counterparts. A collectivist orientation among many ethnic
minorities means that they tend to rely more heavily on their extended family
networks for support compared to Euro-Americans (Barrio, 2000). In many cultures,
family members fulfill multiple roles and provide support on many levels (Goldberg
et al., 2003; Jenkins & Schumacher, 1999). Since such support is constructed from
cultural values and experiences (Kim & McKenry, 1998), there is wide variation in
the contributory effects of social support factors across heterogeneous populations of
ethnically diverse individuals with SPMI. Thus, this study is meant to contribute to
27

the literature on the relationship between social support and treatment outcomes
(Aim 2) and how this relationship may differ in terms of minority status (Aim 3).
(c) Intrapsychic support: Motivational processes.
Intrapsychic support refers to the motivational drive of individuals with
SPMI (Barch, 2005). Intrapsychic factors such as diminished emotional range,
curbed interest, lack of sense of purpose and motivation, and diminished social drive
are generally conceptualized as deficit symptoms (Cohen & Docherty, 2004), and are
therefore mostly understood as an individual’s cognitive and motivational processes
(Galderisi et al., 2002).
Schizophrenia is often described and characterized by manifested symptoms
(negative or positive) associated with the illness (Andreasen, 1982; Bleuler, 1950;
Carpenter, 2007). Positive symptoms are overt bizarre behaviors, thought disorders,
and hallucinations, and negative symptoms include the absence of or diminished
state of emotional or behavioral processes (Velligan & Alphs, 2008). Within the
negative symptoms is a subgroup of deficit symptoms with distinct characteristics
(Buchanan, 2007; Cohen & Docherty, 2004). Unlike negative symptoms, deficit
symptoms are not derived from other neurological impairments, are usually present
for longer periods of time, and are not considered side effects of medications
(Bryson, Bell, Kaplan, Greig, & Lysaker, 1998; Cohen & Docherty, 2004). Despite
the difference that exists between the two, both negative and deficit symptoms are
strongly associated with poorer outcomes in all aspects for individuals with SPMI
28

(Kirkpatricks & Fisher, 2006; Kurtz, 2005; Mileve, Ho, Arndt, & Andreasen, 2005).
Especially for functional outcomes, a growing body of research findings suggests
that improvement in negative and deficit symptoms results in better functional
outcomes (Ho, Nopoulos, Flaum, Arndt, & Andreasen, 1998; Leeuwenkamp et al.,
2008).
Specifically within the deficit domain, Brekke, Levine, Wolkon, Sobel, &
Slade (1993) have found promising evidence of a relationship between intrapsychic
deficits and functional outcomes. Since the main tenet of psychosocial rehabilitation
is consumer-driven service, and given the fact that functional outcomes such as
independent living and work productivity require more purposeful action-oriented
behaviors, intrapsychic factors play an important role in an individual’s treatment
process. To find jobs or remain independent, individuals must not only make the
initial efforts but also be able to sustain them. Intuitively, the two most relevant
intrapsychic factors for initiating and sustaining action-oriented behaviors appear to
be motivation and sense of purpose.
There is an extensive body of research on motivational processes for the
general population in regards to education (Deci, Koestner, & Ryan, 1999; Deci,
Vallerand, Pelletier, & Ryan, 1991; Dembo, 2000; Dembo & Eaton, 2000; Glynn,
Aultman, & Owens, 2005) and the workplace (Latham & Pinder, 2005; Wright,
2001), but very little on motivational processes for individuals with SPMI. The
existing studies on the motivational processes of individuals with SPMI have mostly
29

focused on their association with neurocognitive impairments (Bell & Mishara,
2006; Cohen, Saperstein, Gold, Kirkpatrick, Carpenter, & Buchman, 2007).
Following Barch’s (2005) call for more research in the limited area of intrapsychic
factors for individuals with SPMI, this study will specifically target the motivational
processes of individuals with SPMI and combine indicators of motivation and sense
of purpose in an attempt to better understand the relationship between intrapsychic
support and functional outcomes (Aim 2) across ethnic groups (Aim 3).

3. Significance
In summary, the current psychosocial rehabilitation literature suggests that
sociocultural factors play an important role in identifying predictors of functional
outcomes for individuals with SPMI, but it is a role that can be difficult to determine
and measure. Since the end goal of PSR is community integration, functional
outcomes as indicators of positive treatment effects are much easier to measure. The
sociocultural factor that is the focus of this dissertation study is the available support
systems—a crucial component of PSR interventions. Unlike previous studies that
favored single-system analyses of specific PSR dimensions, the present study will
address the multidimensionality of the PSR framework by performing multi-system
analyses of the support dimension. The overall purpose is to understand the
relationship between available support systems and functional outcomes in an effort
to elucidate the functional change mechanisms. To address the gaps and variability
30

described in ethnic minority research, this study will also explore the moderating
effect of minority status on the relationship between support factors and functional
outcomes.
The conceptual model shown in Figure 1 will be utilized to clarify change
mechanisms in psychosocial rehabilitation functional outcomes. The support
dimension in PSR is conceptualized as being available through many systems:
support provided by service providers (service support), support provided by friends
and family (social support), and internal support from the individual’s motivational
processes (intrapsychic support). These support systems are hypothesized to affect
functional outcomes (e.g. independent living functioning and work productivity
functioning). Minority status is hypothesized to moderate the trajectory of change.

31

Figure 1. Multi-systems Conceptual Framework for Sociocultural Influences on Functional
Outcomes.

D. Preliminary Studies
1. Predicting Psychosocial Rehabilitative Service Outcomes
Brekke et al. (1993, 1997, 1999) conducted a series of studies to examine
positive change determinants and mechanisms associated with PSR interventions.
Findings suggest that the intensity, continuity, specificity, and longitudinality of
services are all associated with more positive outcomes (Brekke et al., 1997, 1999);
intensity was found to have the strongest association. In a later study, Brekke, Long,
and Kay (2002) developed a social support model based on social competence,
relationship quality, satisfaction, relationship symmetry, and number of close friends.
The researchers provided initial evidence of construct validity, however, their
analysis was limited to social support from friends/family and did not include other
Service
Support
Minority Status

Social
Support
Functional Outcome
Trajectories

Independent Living
Work Productivity
Intrapsychic
Support
32

potential sources such as intrapsychic support. Brekke et al. (1993) also found
promising evidence of a relationship between intrapsychic deficits and functional
outcomes. The literature also contains contradictory findings regarding ethnic
differences in functional outcomes. Bae, Brekke, and Bola (2004) found few
significant differences, but Phillips, Barrio, and Brekke (2001) reported significant
differences in change trajectories for functional outcomes, especially for vocational
and social functioning.
These studies set the foundation for the current dissertation in a number of
ways. First, these findings regarding service support and social support substantiate
their credibility as potential predictors of functional outcomes. Second, the
variability in the functional outcome results suggests a need to consider other
explanations for the mechanism of change. Third, given the multi-dimensionality
nature of PRS, a multi-system analysis might better explain functional outcome
change than these previously conducted single system analyses.

2. An Exploratory Study of Culture, Coping, and Schizophrenia
As part of an NIMH-sponsored study, Yamada, Dinh, and Brekke (2006)
conducted semi-structured interviews with 20 ethnically diverse consumers
participating in three psychosocial community-based rehabilitation programs in
Southern California. A connection was observed between ethnicity and both the
relevance and nature of various types of support received by consumers. Consumers
33

expressed a clear emphasis on the importance of feeling comfortable with the care
provider, a scenario that requires an understanding and awareness of sociocultural
influences. Family support was also identified as being important to consumers, as
indicated by their stated desire for more contact with their families. Even when
interviews revealed limited family support, ethnic minorities were reluctant to
present their families in a negative light. Traces of sociocultural influences were also
observed in intrapsychic characteristics, with many consumers from collectivist-
oriented cultures expressing a strong desire to work in order to support their families,
while Euro-Americans frequently expressed a desire for personal financial
independence from their families. These findings from Yamada et al.’s pilot work
(see Yamada & Brekke, in press) encourage further exploration of the relationship
between ethnic minority status and types of support as they affect functional aspects
of recovery.

34

CHAPTER III
METHODS

The purpose of this dissertation is to examine the factors, relationships, and
change trajectories between sociocultural factors and rehabilitative functional
outcomes in individuals with SPMI over time. It is hoped that greater understanding
of these change mechanisms will facilitate the development of more effective and
culturally relevant interventions for individual with SPMI, and ultimately promote
the successful community integration of consumers from diverse ethnic
backgrounds. The conceptual model in Figure 1 illustrates the hypothesized
relationship between each system of support and functional outcome, as well as the
moderating effect that minority status may have on these change trajectories. As
Cook and Jonikas (1996) suggest, it is essential to have a multidimensional
functional outcome model to address multiple PSR intervention goals.

35

A. Study Aims

1. To illustrate change patterns in functional outcomes (independent living
and work productivity) over a twelve-month period of PSR.

2. To identify the strongest support determinant (service support, social
support, or intrapsychic support) for individuals with SPMI for each
psychosocial functional outcome for individuals at baseline and 12 months.

3. To explore whether minority status exerts a significant moderating effect
on the relationship between different systems of support and functional
outcomes for individuals with SPMI.

B. Study Overview
This study will use secondary data from a longitudinal study of psychosocial
rehabilitation for individuals with SPMI receiving services from four community-
based psychosocial rehabilitation programs located in urban Los Angeles, California.
Participants were recruited upon admittance into the programs and monitored for one
year. The psychosocial assessments, at baseline, 6 months, and 12 months, were
completed by trained research interviewers (Bae et al., 2004; Brekke, Nagagami,
Kee, & Green, 2005).
36

C. Study Participants
1. Recruitment Sites
The four community-based psychosocial rehabilitation treatment programs,
all associated with a county mandated mental health initiative, were designed to
provide integrated and comprehensive as well as continuous and long-term
rehabilitative services (Young, Sullivan, Murata, Sturm, & Koegel, 1998). Based on
a shared PSR philosophy, the four programs offer vocational, independent living,
and social rehabilitation, mental health treatment, substance abuse treatment, and 24-
hours crisis response (Brekke et al., 1999, 2000, 2005). Each program has a low staff
ratio (approximately 20:1). Consumers progress along vocational, independent, and
socialization rehabilitation continuums in an effort to achieve maximum community
adjustment and integration.

2. Participant Selection
The four eligibility criteria for inclusion in the original study (Brekke et al.,
2005) were (a) diagnosis of schizophrenia, schizoaffective, or schizophreniform
disorder; (b) residence in Los Angeles county for at least three months;(c) age 18
year old and older; and (d) no diagnosis of a neurological or substance abuse
disorder in the previous six months. Diagnoses were determined using clinical
records, a DSM-IV checklist, and collateral reports from the admitting clinician and
on-site psychiatrist. All subjects signed an informed consent under protocols
37

approved by the Institutional Review Board at the University of Southern California.
The original sample size of the parent study was 164 (82% completion rate at 12
months). There was no statistically significant differential attrition across the
program sites.

3. Sample Size and Characteristics
In this current study, there were two additional inclusion criteria. Given the
small numbers of individuals of Asian, Native American, or other ethnic
backgrounds, only those participants from the three most represented ethnic groups
were included: African Americans, Euro-Americans, and Latinos. Since service
support is a predictive covariate in this dissertation study, those who did not receive
any days of service were eliminated from the present sample, resulting in a final
sample of 136 individuals.
There were 64 Euro-Americans (47%), 56 African Americans (41%), and 16
Latinos (12%) in the current study. There were 94 males (69%) and 42 females
(31%). The average age was 38.2 years old (sd=9.18). On average, functioning
scores were low. The average independent living score for the three time points was
3.83 (sd=1.56), not yet reaching marginally self-sufficient. The average work
productivity functioning score for all time points of 2.36 (sd=1.52) indicates that
occasional attempts at productivity were unsuccessful and constant supervision may
38

be required. The mean BPRS total score of 37.7 (sd=10.0) over three time points
reveals a low symptomatic sample.
Study retention was high, with 81.6% of the sample still enrolled at 6 months
and 72.8% at 12 months. Euro-Americans had the lowest attrition rate: 87.5% were
still enrolled at 6 months and 75% at 12 months. The next lowest rate was for
Latinos (81.3% still enrolled at 6 months and 75% at 12 months), followed by
African Americans (75% at 6 months and 69.6% at 12 months). The attrition rates
were not significantly different between the three ethnic groups.

D. Measurement Instruments and Variables
1. Demographics and Clinical Variables
Five variables were used for descriptive and comparative purposes: age,
gender, marital status, education level, and psychiatric symptoms. Age and
education levels are measured and reported in years. Marital status is measured as
single, married, living together, divorced, and separated status. The marriage and
living together categories are combined and reported as living together. The
divorced and separated categories are combined and reported as separated.
Psychiatric symptoms were assessed with the Expanded Brief Psychiatric Rating
Scale (Lukoff, Nuechterlein, & Ventura, 1986).

39

2. Functional Outcomes
Independent living functioning within PSR refers to the self-care skills
necessary for self-determined residence in the community (Wright, 1980). Work
productivity functional outcome is conceptualized as possessing necessary work
skills in order to find and maintain employment (Iyer et al., 2005). The two
outcome variables (independent living and work productivity) are represented by two
items on the Role Functioning Scale (RFS) (Goodman et al., 1993). In addition to
providing anchored descriptions, the RFS independent living and work productivity
items capture both the quantity and quality of functioning. The independent living
item has a score range from 1 to 7, with 1 indicating no skills or ability to take care
of oneself in an independent living situation; individuals at this level often receive
multiple and lengthy hospital services. A score of 7 indicates an optimal care level;
these individuals can independently manage their own personal needs in independent
living situations. For work productivity, a score of 1 reflects a lack of skills or
attempts to work and a score of 7 represents optimal work performance and skills in
an independent working situation. The RFS has been found to be a reliable outcome
measure for individuals with SPMI (Brekke et al., 1997, 1999, 2001; Green &
Gracely, 1987).

40

3. Minority Status
Minority status was used as a proxy variable for sociocultural influences
within the treatment process. Although separate analyses for each of the three ethnic
groups in this study would provide greater insights regarding sociocultural
differences, the small Latino sample size would compromise the generalizability of
any findings. Since African Americans and Latinos share relatively similar minority-
related mental health outcomes and outpatient service disparities, and since they both
share more collective family orientations than Euro-Americans (Guarnaccia & Parra,
1996; Lefley, 1990; Singelis, 1994), they were combined into a single group for
analyses in this study. Minority status is thus used as a proxy for ethnic and cultural
variation. African Americans and Latinos are combined as the minority group
(n=72) and persons of non-Hispanic Euro-American heritage (n=64) are considered
as the non-minority group.

4. Support System Variables
The three types of support used in the analyses are service support from
service providers, social support from family members and friends, and intrapsychic
support within consumers. All support variables are continuous, with different scale
ranges as follows.

41

(a) Service support.
Service support refers to treatment support provided by PSR service
providers: measured by the quantity of contact (Brekke et al., 1997, 1999). Service
support was measured using a service intensity indicator successfully used in a
previous study of similar agency sites (Brekke et al., 1997). Intensity will be
measured as the number of days that an individual receives a service from the
admitting rehabilitation agency during the 365 days after being recruited for this
study. The source of this information is service contact data gathered by program
staff for billing and administrative purposes and review by county auditors; billing is
done in 15-minute increments. Additional data for the participants were retrieved
from throughout the Los Angeles County mental health system for the study period.
The data indicate that 93% of all outpatient service contacts from county and county-
contracted agencies during the study year were with the rehabilitation program to
which the participants were first admitted when they were recruited for the original
study.
(b) Social support.
Social support refers to instrumental and emotional support provided by
family and friends (Bengtsson-Tops & Hansson, 2001; Buchanan, 1995; Schwarzer
et al., 2003; Walsh & Connelly, 1996). The social support variable represents a
combination of family network and immediate social network relationships items
from Goodman et al.’s (1993) RFS. The family network relationship item provides
42

an anchored description and captures both the quantity and quality of family support;
the immediate social network relationships item does the same for friendships.
Possible scores for both items range from 1 to 7, with 1 representing no or severely
limited contact and 7 representing positive relationships in which the consumer
makes active contributions in a consistent and reciprocal manner. Cronbach’s alphas
for the four RFS items were .510 at baseline, .589 at 6 months, and .680 at 12
months—below the .80 criterion level for high reliability, indicating that they do not
measure the same constructs. This supports the decision to use two of the items
(independent living and work productivity) as outcome variables and the other two
(family network relationships and immediate social network relationships) as
predictive social support covariates.
(c) Intrapsychic support.
Intrapsychic support refers to the motivational drive of individuals with
SPMI (Barch, 2005). Intrapsychic support indicates the degree to which individuals
report a combined sense of purpose and internal motivation (Galderisi et al., 2002).
The two variables are parts of Heinrichs, Hanlon, and Carpenter’s (1984) Quality of
Life Scale (QLS). Sense of purpose is measured in terms of the degree to which a
person expresses realistic and integrated life goals. Responses to this item are given
on a Likert scale ranging from 0 (no sense of purpose) to 6 (realistic and integrated
plans). Motivation is measured in terms of the extent to which a person is capable of
43

initiating or sustaining goal-directed activity. The Likert response scale also ranges
from 0 (lack of motivation) to 6 (no evidence of a significant lack of motivation).

E. Data Analysis
1. Analytic Plan
The first part of the analytic plan was an examination of the descriptive
statistics (i.e., frequencies, measures of central tendency and dispersion) of the study
sample, each ethnic group, and the ethnic minority/non-minority groups. A one-way
analysis of variance (ANOVA) was used to test relationships between nominal level
and interval/ratio level variables, and a chi-square test was used to examine the
relationship between the two nominal variables.
The second part of the analysis made use of Muthen and Muthen’s (2007)
Mplus Version 5 to run a series of latent growth models to study the growth structure
and predictors of the functional outcomes. Based on Singer and Willlett’s (2003)
recommendations, the modeling was conducted in three stages: estimating the
unconditional model, approximating the preliminary conditional models, and
computing the full conditional model. Unlike previous regression analysis of means
comparisons at different time points, growth curve modeling is better suited for
longitudinal data, since it offers opportunities to examine variability and to provide
insights to change (Acock, 2005). This approach acknowledges that various support
levels do not exist in a vacuum, but in fact operate and interact at multiple levels.
44

These complex multilevel interactions may explain (at least in part) the considerable
variability observed in functional outcomes for individuals with SPMI.
Mplus Version 5 allows for analyses of individual growth trajectories across
time points and subsequent examinations of the effects of covariates to determine
systematic differences in initial status (intercept), rate of change (slope), or a
combination of the two (Muthen & Muthen, 2007). In the univariate approach to
hierarchical linear modeling, the Level 1 model represents each individual’s
trajectory of change as a function of person-specific parameters plus random errors,
and the Level-2 model describes variation in these growth parameters across a
population (Raudenbush & Bryk, 2002). In contrast, Mplus’s multivariate repeated-
measures approach requires (a) the specification of main effects and interactions that
describe the expected trajectory for different subgroups or kinds of individuals, and
(b) the specification of variation and covariation in the repeated measures over time
(Muthen & Muthen, 2007).
According to Muthen and Muthen (2007), Mplus Version 5 was especially
designed for smaller clinical samples. Unlike other statistical programs that suffer
greatly from missing data, Mplus is still capable of using all available data for model
estimation. Using the full information maximum likelihood (FIML) approach, the
missing data becomes part of the model parameter estimation, meaning that each
parameter can be estimated directly without data imputation for each individual.

45

2. Mplus Latent Growth Curve Modeling
The first step in latent growth curve modeling is fitting the basic growth
curve. Also known as the “unconditional model,” this represents the growth curve of
the functional outcome variables without any covariates or interaction effects (Aim
1). The two latent variables are the intercept and slope (Acock, 2005). Growth curves
provide empirical evidence for determining proper specification of the individual
growth equation and baseline statistics for evaluating subsequent models. It is
important to have a model that fits well with the observed data in order to generate
interpretable explanations (MacCallum, 1995).
Final model determination is based on selected statistical fit indices. Since
each index type provides information on different aspects of model fit, multiple
criteria are recommended for evaluating a growth curve model (Hu & Bentler, 1995).
In most cases, the X
2
test of model fit is the first indicator of how well the overall
model fits the data, but the sensitivity of X
2
to sample size (Marsh, Balla &
McDonald, 1988) dictates a need for additional fit indices (known as “incremental
indices”) to evaluate model fit. This study makes use of the comparative fit index
(CFI) (Bentler, 1990), Tucker-Lewis index (TLI) (Tucker & Lewis, 1973), root mean
square error approximation (RMSEA) (Steiger & Lind, 1980), and standard root
mean square residual (SRMR) (Joreskog & Sorbom, 1981). CFI and TLI compare
existing and null models; a value equal to or greater than .90 indicates good model
fit. While CFI adjusts for model complexity (Bentler, 1990), TLI penalizes for
46

model complexity (Tucker & Lewis, 1973). Similar to X
2
, the SRMR test is based on
predicted versus observed covariance matrixes. The smaller the SRMR, the better the
fit; an SRMR of 0 is considered a perfect fit. A value equal to or less than .05 is
generally considered good fit (Joreskog & Sorbom, 1981). While RMSEA also tests
predicted versus observed covariances and does not require a comparison to a null
model, it penalizes for lack of parsimony (Steiger & Lind, 1980). An RMSEA value
(also known as discrepancy per degree of freedom) of .08 or less is considered an
adequate fit, and a value of .05 or less is considered a good fit (Hu & Bentler, 1995).
To address Aim 2 (to determine whether different support systems impact
initial level and rate of change in independent living and work productivity), the
predictive covariates are entered into the unconditional model, and the intercepts and
slopes of the functional outcomes are regressed on the predictive covariates (i.e.,
service intensity, social support, and intrapsychic support) (Muthen & Muthen,
2007). The model can be built using one of two methods; the choice is usually driven
by the conceptual rationale of the study in question. The first method consists of
adding one covariate at a time to the unconditional model, then comparing outputs to
determine if the additional covariate contributes to the model. The second method is
to inject all covariates at the same time (Hoyle, 1995). Due to this study’s
psychosocial framework (in which different support dimensions interact with each
other), it is conceptually reasonable to input all three support covariates at the same
time when building the conditional model. Figure 2 below illustrates the statistical
47

model of the relationships among the support covariates and work productivity
functioning.

Figure 2. A Conditional Model of Sociocultural Influences on Functional Outcomes
Note: The model for independent living is the same with the levels for independent living at
baseline (IL1), 6 months (IL2), and 12 months (IL3) substituted for work productivity.

The final step (Aim 3) is determining whether the trajectories differ
according to minority status. For the multiple group analysis, models are
simultaneously estimated within each group, with the least restrictive model analysis
being conducted first. According to this model, all regression coefficients are
allowed to vary freely between minority and non-minority groups, then the fit

Work1

Work2

Work3
Social
Support

i
s
Service
Support
Intrapsychic
Support
48

between the model and a second multiple group analysis of a more restrictive model
are compared. In the more restrictive model, all regression coefficients are fixed to
be equal between the two groups (Byrne, 2001). A non-significant Chi-square
difference test between the two groups would indicate that their

parameters are equal
suggesting that minority status does not exert a moderating influence on the
relationship between the support covariates and functional outcomes (Muthen &
Muthen, 2007).
A statistically significant chi square value shows that constraining the
parameters decreases the model fit, indicating a difference in the model between the
two groups (Muthen & Muthen, 2007). This suggests a potential for a minority
moderating effects on the relationship between the support covariates and outcome
variables. However since the Chi-square test is sensitive to sample size (i.e., it may
not detect a moderating effect in a small sample such as that used in this research)
(Marsh et al., 1988), it is important to conduct additional multiple group analyses in
which parameters are freed one at a time in order to identify the potential moderating
effect (Byrne, 2001). Accordingly, the Chi-square

of this model with one freed
parameter can be compared to the Chi-square

of a model where all parameters are
freed. Once a moderating effect is found, an examination of regression coefficients
from the least restrictive model will provide a more detailed understanding of the
moderating effect in relation to all other covariates.
49

Figure 3 below, a statistical model of the relationships among the support
covariates and work productivity functioning moderated by minority status, is a
pictorial illustration of the statistical analyses discussed in the previous section.

Figure 3. A Multiple Group Model of Sociocultural Influences on Functional Outcomes
Note: The model for independent living is the same with the levels for independent living at
baseline (IL1), 6 months (IL2), and 12 months (IL3) are substituted for work productivity.

Work1

Work2

Work3
Social
Support

i
s
Service
Support
Intrapsychic
Support
Minority
Status
50

3. Methods and Guidelines for Data Interpretation
In this study, a p-value equal to or less than .05 is considered statistically
significant. Since a small sample size can limit detection of a statistically significant
relationship, a p-value between .05 and .10 is reported as indicating a statistical
trend. Effect size standards can differ depending on sample size (Garson, n.d.). In
general, an r value of .80 or larger represents a large effect; however, in small
clinical studies, .10 is said to represent a small effect, .30 a medium effect, and .50 a
large effect (Cohen, 1988). Z-scores and p-values are reported when statistical
significance is determined. Non-standardized coefficients (b) and non-standardized
standard errors are reported when discussing intercept, slope, or covariate values and
effect sizes. The standardized coefficient ( β) is used when discussing the variance
contribution of a specific covariate, since standardized coefficients are comparable
across variables within a model and reflect variance strength (Garson, n.d). Since
small sample size limits the ability to detect moderating effects in multiple group
analyses, an examination of where parameters differ for both groups in the non-
constrained model will be conducted to avoid committing Type II errors (Bryne,
2001). While it may not be possible to definitely conclude that differences exist, it is
possible to hypothesize their most likely location.

51

CHAPTER IV
RESULTS

The study results are presented in three sections: descriptive statistics,
independent living specifications and parameter estimates, and work productivity
specifications and parameter estimates. Organized by functional outcome—
independent living and work productivity—the second and third sections
respectively present a linear progression of analysis and understanding. Three
subsections are used to correspond to study Aims 1, 2 and 3 within each functional
outcome section. Figures depicting estimated growth trajectories for each functional
outcome are included, as are tables that summarize the results of specific analyses.

A. Descriptive Statistics
Demographic and clinical variables for the entire sample at baseline, six
months, and twelve months are given in Table 3. The baseline sample of N=136
participants (94 male, 42 female) consisted of 64 Euro-Americans (47%), 56 African
Americans (41%) and 16 Latinos (12%) diagnosed with schizophrenia or
schizoaffective disorder. Most of the sample was either single (N=98) or
divorced/separated (N=37)—only one study participant was married. Just under one-
third (N=42, or 32%) had high school educations; the average number of years of
formal education for the sample was 11.97 (sd=1.75). The average age was 38.2
52

years (sd=9.18). The average number of days receiving service after joining the
study was 85.52 days (sd=54.87). The sample had limited interpersonal with family
and friends (x ¯ =3.34, sd=1.39) and lacked motivation (x ¯ =2.82, sd=1.53). The
average baseline score for independent living functioning of 3.26 (sd=1.70) was
higher than the baseline score for work productivity functioning 1.96 (sd=1.49).
Average Brief Psychiatric Rating Scale (BPRS) score for the entire sample was
39.16 (sd=10.43) at baseline.
Data from bivariate comparisons involving the tri-ethnic group at baseline
indicate significant relationships with gender and age (Table 4). Results from
Pearson chi square tests indicate significantly higher numbers of African American
males than females (45 vs. 8) and Euro-American males than females (40 vs. 24).
Results from a one-way ANOVA indicate that the Latino participants were younger
(x ¯ =31.69, sd=8.36) and the African Americans older (x ¯ =40.34, sd=8.58) at
statistically significant levels (F[2,133]=5.980, p<.05). Separate ANOVA data show
that at baseline, the Euro-Americans had higher intrapsychic support levels (x ¯ =3.09,
sd=1.55) than the African American (x ¯ =2.65, sd=1.53) and Latino (x ¯ =2.31,
sd=1.28) participants, but only at the level of a trend toward significance.
Results from a comparison (independent t-test) of demographic and clinical
variables between Euro-Americans and minority groups are presented in Table 5.
They indicate a statistically significant difference in intrapsychic support at baseline
between the non-minority (x ¯ =3.09, sd=1.55) and minority groups (x ¯ =2.31, sd=1.28)
53

(t[129]=1.99, p<.05). Although not statistically significant, the data also show that
minority group individuals had more days of service support, less social support, less
work productivity functioning, and less independent living functioning than non-
minority individuals at baseline. Pearson correlation data for the support covariates
and functional outcome variables are shown in Table 6. Statistically significant
correlations were found between social support and both intrapyschic support (r=.31)
and independent living (r=.31). The data also indicate significant correlations
between intrapsychic support and both independent living functioning (r=.32) and
work productivity functioning (r=.45), as well as between independent living
functioning and work productivity functioning (r=.43). No statistically significant
correlations were found between service support and any of the other predictive and
outcome variables. Service support was found to have inverse relationships with
intrapsychic support, independent living functioning, and work productivity
functioning.
54

Table 3. Characteristics of Total Sample at Baseline, 6 Months, and 12 Months
Baseline
(N=136)
6 months
(N=111)
12 months
(N=99)
Gender
Female
Male

42 (31%)
94 (69%)

Ethnicity
African American
Euro-American
Latino

56 (41%)
64 (47%)
16 (12%)

42 (38%)
56 (50%)
13 (12%)

39 (39%)
48 (49%)
12 (12%)
Marital Status
Single
Living Together
Separated

98 (72%)
1 (1%)
37 (27%)

85 (77%)
1 (1%)
24 (22%)

75 (75%)
4 (4%)
21 (11%)
Age
Mean
SD
Range

38.21
9.18
18-62 years

Education
Mean
SD
Range

11.97
1.75
8-18 years

Retention 81.6% 72.8%

Mean (SD)

Mean (SD)

Mean (SD)
Service Support
1
85.52 (54.87)
Social Support
2,a
3.34 (1.39) 3.61 (1.52) 3.83 (1.62)
Intrapsychic Support
3,b
2.82 (1.53) 3.11 (1.58) 3.24 (1.67)
Working Productivity
4,a
1.96 (1.49) 2.61 (1.86) 2.52 (1.94)
Independent Living
5,a
3.26 (1.70) 4.00 (1.74) 4.16 (1.88)
BPRS 39.16 (10.43) 38.06 (9.47) 36.54 (10.06)
Notes: 1. Number of days of treatment in program after admission to study
2. Combined Family Network item and Immediate Social Network item from the Role
Functioning Scale 3. Combined of Sense of Purpose item and Degree of Motivation item
from the Quality of Life Scale 4. Work Productivity item from the Role Functioning Scale
5. Independent Living item from the Role Functioning Scale
a. The scale range from 1 to 7 b. The scale range from 0 to 6.

55

Table 4. Bivariate Comparisons Between Tri-ethnic Groups at Baseline

Euro
American
(N=64)
African
American
(N=56)

Latino
(N=16)

Bivariate
X
2
and F (df)
Gender
Female
Male

24 (18%)
40 (29 %)

11 (8 %)
45 (33 %)

7 (5%)
9 (7 %)

X
2
(2)=5.87**
Marital Status
Single
Living Together
Separated

45 (33%)
1 (1%)
18 (13%)

40 (29%)
0
16 (12%)

13 (10%)
0
3 (2%)

X
2
(6)=4.63
Age
Mean
SD
Range

37.97
9.21
18-62 years

40.34
8.58
24-61 years

31.69
8.36
21-55 years

F(2,133)=5.98**
Education
Mean
SD
Range

11.90
1.70
8-16 years

12.17
1.84
8-18 years

11.53
1.64
8-15 years

F(2,128)=0.86

Mean (SD)

Mean (SD)

Mean (SD)

Service Support
1

83.64 (52.38) 80.29 (58.52) 111.38 (46.73) F(2,133)=2.10
Social Support
2,a

3.50 (1.42) 3.09 (1.35) 3.60 (1.49) F(2,132)=1.60
Intrapsychic Support
3,b
3.09 (1.55) 2.65 (1.53) 2.31 (1.28) F(2,133)=2.29*
Working Productivity
4,a
2.00 (1.54) 1.91 (1.43) 1.94 (1.53) F(2,133)=0.06
Independent Living
5,a

3.41 (1.60) 3.16 (1.89) 3.06 (1.39) F(2,133)=0.44
BPRS Total Score 39.73 (9.72) 39.20 (11.46) 36.75 (9.64) F(2,133)=0.52
Notes: 1. Number of days of treatment in program after admission to study 2. Combined
Family Network item and Immediate Social Network item from the Role Functioning Scale
3. Combined of Sense of Purpose item and Degree of Motivation item from the Quality of
Life Scale 4. Work Productivity item from the Role Functioning Scale 5. Independent
Living item from the Role Functioning Scale a. The scale range from 1 to 7 b. The scale
range from 0 to 6.
*p<.10, **p<.05.

56

Table 5. Bivariate Comparisons By Minority Status at Baseline

Notes: 1. Number of days of treatment in program after admission to study 2. Combined
Family Network item and Immediate Social Network item from the Role Functioning Scale
3. Combined of Sense of Purpose item and Degree of Motivation item from the Quality of
Life Scale 4. Work Productivity item from the Role Functioning Scale 5. Independent
Living item from the Role Functioning Scale a. The scale range from 1 to 7 b. The scale
range from 0 to 6.
*p<.05

Non-minority
(N=64)
Minority
(N=72)
Bivariate
X
2
or t (df)

Gender
Female
Male

24 (18%)
40 (29 %)

18 (13%)
54 (40%)

X
2
(1)=2.48
Marital Status
Single
Living Together
Separated

45 (33%)
1 (1%)
18 (13%)

53 (39%)
0
19 (14%)

X
2
(3)=3.79
Age
Mean
SD
Range

37.97
9.21
18-62 years

38.42
9.21
21-61 years

t(134)=0.28
Education
Mean
SD
Range

11.90
1.70
8-16 years

12.03
1.80
8-18 years

t(129)=0.41

Mean (SD)

Mean (SD)

Service Support
1
83.64 (52.38) 87.19 (57.30) t(134)=-0.38
Social Support
2,a
3.50 (1.42) 3.20 (1.36) t(133)=1.26
Intrapsychic Support
3,b
3.09 (1.55) 2.58 (1.48) t(134)=1.99*
Working Productivity
4,a
2.00 (1.54) 1.92 (1.44) t(134)=0.33
Independent Living
5,a
3.41 (1.60) 3.14 (1.78) t(134)=0.92
BPRS 38.58 (9.56) 38.65 (11.06) t(134)=0.60
57

Table 6. Correlation Table of Support Covariates and Functional Outcomes

1 2 3 4 5 6 7 8 9
Service
Support 1 .01 -.12 -.16

-.22*

-.19 -.13

.02

-.01
Social
Support 1 .31* .31*

.15

.15 .03

.15

.31*
Intrapsychic
Support 1 .32*

.35*

.29* .45*

.49*

.44*
Independent
Living1 1

.60*

.59* .43*

.43*

.37*
Independent
Living2

1

.75* .37*

.47*

.37*
Independent
Living3
1
.39*

.36*

.43*
Work
Productivity1

1

.47*

.44*
Work
Productivity2

1

.70*
Work
Productivity3

1
Notes: 1=Service Support, 2=Social Support, 3=Intrapsychic Support, 4=Independent Living
Functioning at baseline, 5=Independent Living at 6 months, 6=Independent Living at 12
months, 7=Work Productivity at baseline, 8=Work Productivity at 6 months, 9=Work
Productivity at12 months
* p< .05 (two tailed)

B. Independent Living Functioning: Model Specifications and Parameter Estimates
Results from model fit tests and independent living functioning parameter
estimates are presented in this section. Subsection 1 corresponds to study Aim 1
(describing the growth trajectory for independent living functioning for the entire
sample over a 12 month period), Subsection 2 to Aim 2 (describing the full
conditional model with the inclusion of all support covariates), and Subsection 3 to
Aim 3 (addressing the moderating effects of minority status on the relationship
between the support covariates and independent living functioning).
58

Fit indices for all independent living functioning models discussed in this section are
presented in Table 7.

1. Growth Trajectory: Unconditional Model to Test Aim 1
Due to the limited number of observations (n=3 time points), the
unconditional model was initially restricted to a linear growth model to determine
whether a linear model was valid for measuring independent living functioning.
Overall, fit index data indicated a good fit for the linear unconditional model
(X
2
(1)=3.79, p=0.05; CFI=0.98; TLI=0.94; SRMR=0.04). The one exception was the
RMSEA test result of 0.14 (90% confidence interval, 0.00-0.31), which did not meet
the accepted criterion of a good fit. To improve model fit, the last time observation
was freed up. After the model modification was made, the nonlinear model fit did
not exceed that of the linear model (X
2
(0)=0.00, p=0.00), therefore the linear model
was used as the unconditional model for independent living functioning.
59

Table 7. Fit Indices for Pertinent Independent Living Functioning Models

X
2
df p CFI TLI SRMR
RMSEA
(Confidence
Interval)

ΔX
2
( Δdf)
Unconditional
Model
1
3.79 1 0.050.980.940.04
0.14
(0.00-0.31)

Conditional Model
2
6.39 4 0.17 0.99 0.96 0.03
0.07
(0.00-0.16)

Multiple Group
Least restrictive
3
18.87 130.130.960.93 0.06
0.08
(0.00-0.16)

I-Service Support
4
19.16 14 0.16 0.97 0.95 0.06
0.07
(0.00-0.15)
0.29
(1)
S-Service Support
5
19.90 14 0.13 0.96 0.94 0.06
0.08
(0.00-0.15)
1.04
(1)
I-Social Support
6
21.28 140.100.960.92 0.07
0.09
(0.00-0.16)
2.41*
(1)
S-Social Support
7
24.06 14 0.05 0.94 0.89 0.07
0.10
(0.02-0.17)
5.19**
(1)
I-Intrapsychic
Support
8
18.91 140.170.970.950.06
0.07
(0.00-0.15)
0.04
(1)
S-Intrapsychic
Support
9
18.94 140.170.970.950.06
0.07
(0.00-0.15)
0.07
(1)
Intercept
10
19.14 140.160.970.950.07
0.07
(0.00-0.15)
0.45
(1)
Slope
11
19.21 140.160.970.950.06
0.07
(0.00-0.15)
0.34
(1)
Multiple Group
Most restrictive
12
28.66 21 0.12 0.95 0.95 0.07
0.07
(0.00-0.13)
9.79
(8)
Notes. 1.Unconditional linear model, 2.Conditional linear time invariant model, 3.Least
restrictive multiple group linear time invariant model (all parameters are freed), 4.Multiple
group model with the Intercept of Service Support freed, 5.Multiple group model with the
Slope of Service Support freed, 6.Multiple group model with the Intercept of Social Support
freed, 7.Multiple group model with the Slope of Social Support freed, 8.Multiple group
model with the Intercept of Intrapsychic Support freed, 9.Multiple group model with the
Slope of Intrapsychic Support freed, 10.Multiple group model with the Intercept freed,
11.Multiple group model with the Slope freed, 12.Most restrictive multiple group linear time
invariant model (all parameters are fixed)
* trend toward significance, ** significant at p=.05
60

Parameter estimates for the unconditional linear model for independent living
functioning reveal a statistically significant mean baseline level (z=23.61, p=.00)
(Table 8)—in other words, the average independent living level of 3.34 at baseline
was significantly different from 0. Furthermore, the resulting slope (z=5.464,
p=.000) indicates that the study participants showed statistically significant
improvement over the 12 month study period (Figure 4).

Table 8. Linear Unconditional Model Parameter Estimates for Independent Living

Functioning

b S.E. z-score p-value
Mean
Intercept 3.34 0.14 23.61 0.00
Slope 0.43 0.08 5.46 0.00

Variance
Intercept 1.63 0.41 3.96 0.00
Slope 0.21 0.21 1.02 0.31

Covariance
Intercept-Slope 0.13 0.22 0.58 0.56

61

Figure 4. Independent Living Functioning Score Over Time For Sample

According to the unstandardized coefficient of 0.427, the entire sample gains 0.427
unit of independent living functioning within the passage of 6 months. The variance
for the intercept (z=3.957, p=.000) indicates significant variability in the baseline
independent living values for all individuals in the sample. Despite this variability,
the non-significant slope variance (z=1.016, p=.309) indicates that the rate of change
among study participants was relatively homogeneous. This relationship is captured
in Figure 5. —that is, the figure shows uneven starting levels but relatively uniform
change rates. The non-significant covariance between the intercept and slope
indicates an absence of any relationship between baseline score and rate of
improvement for independent living.

62

Figure 5. Random Independent Living Functioning Trajectories of N=50 Participants

2. Support Covariates: Conditional Model to Test Aim 2
The full conditional model with all predictive covariates was created to
determine the impacts that individual covariates have on change rates for
independent living functioning (Aim 2). The conditional model was based on the
linear equality constraints growth model. The fit indices for the full conditional
model indicate a good model fit (X
2
(4)=6.39, p=0.17; CFI=0.99; TLI=0.96;
SRMR=0.03; RMSEA=0.07, 90% confidence interval, 0.00-0.16) (Table 5).
Results from parameter estimates of relationships between the covariates and
initial status (intercept) and rate of change (slope) for independent living functioning
are presented in Table 9. Results from the conditional independent living functioning
model indicate statistical significance for both the intercept of the intercept (z=4.46,
p=0.00) and the intercept of the slope (z=2.980, p=0.000). The intercept of the
63

intercept is the predicted level of independent living (b=1.98, SE=0.44) when time
and all covariates equal zero; the intercept of the slope is the slope for time (b=0.88,
SE=0.30) when all support covariates are kept constant at zero. Statistical
significance was only found for the residual variance of the intercept (z=1.152,
p=0.001), indicating the possibility that a covariate not discussed in this study might
explain the variability observed in the baseline levels. The residual variance for the
slope (z=0.815, p=0.415) was not statistically significant, suggesting a lack of further
variance requiring explanation. Accordingly, the three predictive covariates in this
study were sufficient for explaining most of the independent living growth trajectory.
Data for the regressed intercept on service, social, and intrapsychic support
indicate statistically significant impacts of both social support (z=2.33, p=0.02) and
intrapsychic support (z=3.41, p=0.00) on initial level of independent living.
Specifically, greater social support and intrapsychic support were associated with
higher independent living functioning at baseline. Intrapsychic support ( β=0.36,
SE=0.11) was found to have the strongest effect on initial level of independent
living, followed by social support ( β=0.25, SE=0.10). The results also suggest a
trend toward significance for system support (z=-1.71, p=.09). However, despite its
significant impact on initial independent living level, intrapsychic support did not
significantly impact the change rate in independent living status. A trend toward
significance was found for social support (z=-1.835, p=.067).

64

Table 9. Conditional Model Parameter Estimates for Independent Living Functioning

b S.E z-score p-value
Intercept regressed on
Service Support -0.00 0.00 -1.71 0.09
Social Support 0.22 0.10 2.33 0.02
Intrapsychic Support 0.30 0.09 3.41 0.00

Slope regressed on
Service Support -0.00 0.00 -0.40 0.69
Social Support -0.14 0.06 -1.84 0.07
Intrapsychic Support 0.00 0.05 0.01 0.99

Intercepts
Intercept 1.98 0.44 4.46 0.00
Slope 0.88 0.29 2.98 0.00

Residual Variances
Intercept 1.15 0.36 3.18 0.00
Slope 0.17 0.20 0.82 0.42

3. Moderating Effects: Multiple Groups Model to Test Aim 3
A multiple group analysis was conducted in which all parameters were freely
estimated (least constrained model) to determine the moderating effects of either
minority status on relationships between different systems of support and
independent living functioning among individuals with SPMI. Three of the five fit
indices met the criteria for a good model fit. Results from an X
2
test of model fit
(X
2
(13)=18.87, p=0.13) were non-significant, and both CFI and TLI values exceeded
0.90 (CFI=0.96, TLI=0.93). The RMSEA value of 0.08 (90% confidence interval,
0.00-0.16) indicated an adequate fit, and the SRMR value of .06 was higher than the
preferred maximum .05 level. Independent living functioning growth trajectories for
minority/non-minority individuals are illustrated in Figure 6.
65

Figure 6. Independent Living Functioning Over Time by Minority/Non-minority Groups

Parameter estimates for the least constrained multiple group model suggest
that even though the non-minority group started out with higher independent living
functioning levels (b=2.47, SE=0.71), they were not significantly different from the
baseline levels for individuals in the minority group (b=1.46, SE=0.57) (see Table
10). Results for the independent living growth rate when all covariates were held at
zero show that minority individuals (b=1.29, SE=0.40) had much faster growth
increases than non-minorities (b=0.68, SE=0.46). Notable differences were also
found between the minority and non-minority groups regarding relationships
between the support covariates and independent living functioning outcome. For the
non-minority group, intrapsychic support (z=1.99, p=0.05) was the only covariate
66

exerting a statistically significant impact at the baseline level of independent living
functioning. For the minority group, both intrapsychic support (z=2.49, p=0.01) and
social support (z=2.77, p=0.01) had statistically significant influences on the baseline
level of independent living functioning. Although intrapsychic support was found to
have a significantly positive effect on both groups at baseline, it did not significantly
influence the growth rate of either group. In fact, none of the support covariates were
found to have significant influences on change rates for the non-minority group.
Social support (z=-2.93, p=0.00) was the only covariate that significantly impacted
the growth rate of independent living for the minority group.
Results for the multiple group analysis in which all parameters were
constrained in terms of minority or non-minority group membership indicate a
poorer model fit for some indices. Although the differences were slight, data from
three of the five indices suggest a poorer model fit for the more restrictive model
(X
2
(21)=28.66, p=0.12; CFI=0.95; SRMR=0.07) than the model in which all
parameters were allowed to vary freely. The X
2
difference between the fully
restrictive and less restrictive models (X
2
(8)=9.79) was not significant at the p=.05
level, suggesting that the models were non-invariant between the minority and non-
minority groups.
However, since small sample sizes can mask the potential influences of
moderating effects, additional multiple group analyses in which parameters were
constrained one at a time were conducted. Results from a chi square test on the social
67

support slope revealed a significant difference ( ΔX
2
(1)=5.196), suggesting that
minority status altered the relationship between social support and independent
living growth rate. Social support apparently had an inverse relationship with the
independent living functioning slope for minority individuals (b=-0.26, SE=0.09)
compared to individuals from the non-minority group (b=.020, SE=0.081). The
results from a chi square test comparing social support between the two groups at
baseline ( ΔX
2
(1)=5.196) revealed a trend toward significance. Parameter estimates
for the multiple group analysis showed a more significant positive impact of social
support on initial status for the minority group (b=0.38, SE=0.14) compared to the
non-minority group (b=0.08, SE=0.14) (Table 8). While results from the least
constrained model suggested a moderating effect on the intercept of the slope
between the minority (z=3.24, p=.00) and non-minority (z=1.52, p=0.13), data from
a chi square test comparing the two models did not indicate statistical significance
( ΔX
2
=0.45, Δdf=1) (Table 7).

68

Table 10. Multiple Group Model Parameter Estimates for Independent Living

Functioning

Non-minority Minority
b S.E. z-score
p-
value b S.E. z-score
p-
value
Intercept
Service Support -0.00 0.00 -0.54 0.59 -0.01 0.00 -1.56 0.12
Social Support 0.08 0.14 0.58 0.57 0.38 0.14 2.77 0.01
Intrapsychic Support 0.26 0.13 1.99 0.05 0.30 0.12 2.49 0.01

Slope
Service Support -0.00 0.00 -1.17 0.24 0.00 0.00 0.21 0.84
Social Support 0.02 0.08 0.25 0.81 -0.26 0.09 -2.93 0.00
Intrapsychic Support 0.01 0.08 0.10 0.92 -0.02 0.07 -0.28 0.78

Intercepts
Intercept 2.47 0.71 3.49 0.00 1.46 0.57 2.56 0.01
Slope 0.68 0.45 1.52 0.13 1.29 0.40 3.24 0.00

Residual Variances
Intercept 1.47 0.42 3.52 0.00 1.20 0.35 3.38 0.00
Slope 0.15 0.14 1.06 0.29 0.06 0.12 0.47 0.64
Notes: Non-minority=Euro-Americans, Minority=combined African Americans and Latinos

4. Summary of Independent Living Functioning Results
The results indicate continuous linear growth in independent living
functioning over a twelve month period. For the complete sample, intrapsychic
support had the greatest impact on the independent living baseline level, followed by
social support; a trend toward significance was found for system support. No
significant predictors emerged from the independent living functioning growth
trajectory, though social support showed a trend toward significance. An inverse
linear relationship was found between social support and independent living—that is,
69

an association was found between higher levels of social support and lower levels of
independent living.
A more complex picture emerges from a comparison between minority and
non-minority group members. Intrapsychic support was identified as the most
important baseline factor for individuals in the non-minority group, and social
support as the most important for members of the minority group. Regarding the
growth trajectory, social support was the only significant predictor for minority
individuals; however, the inverse linear relationship indicates an association between
lower social support and a higher independent living growth trajectory. None of the
covariates were found to be significant for individuals in the non-minority group.

70

Table 11. Fit Indices for Pertinent Work Productivity Functioning Models

X
2
df p CFI TLI SRMR
RMSEA
(Confidence
Interval)

Δ X
2
( Δdf)
Unconditional
Model
1
0.74 2 0.69 1.00 1.02 0.02
0.00
(0.00-0.13)

Conditional Model
2
4.61 5 0.47 1.00 1.01 0.02
0.00
(0.00-0.11)

Multiple Group
Least restrictive
3
7.33 11 0.77 1.00 1.05 0.04
0.00
(0.00-0.09)

I-Service Support
4
7.33 12 0.84 1.00 1.06 0.04
0.00
(0.00-0.07)
0.00
(1)
S-Service Support
5
7.33 12 0.84 1.00 1.06 0.04
0.00
(0.00-0.07)
0.01
(1)
I-Social Support
6
8.08 12 0.78 1.00 1.05 0.04
0.00
(0.00-0.08)
0.76
(1)
S-Social Support
7
8.27 12 0.76 1.00 1.05 0.04
0.00
(0.00-0.09)
0.95
(1)
I-Intrapsychic
Support
8
7.68120.811.001.060.04
0.00
(0.00-0.08)
0.35
(1)
S-Intrapsychic
Support
9
9.51120.661.001.030.05
0.00
(0.00-0.10)
2.18*
(1)
Intercept
10
8.50120.751.001.050.04
0.00
(0.00-0.09)
1.18
(1)
Slope
11
7.48120.821.001.060.04
0.00
(0.00-0.08)
0.16
(1)
Multiple Group
Most restrictive
12
13.29 19 0.82 1.00 1.05 0.06
0.00
(0.00-0.07)
5.97
(8)
Notes. 1. Unconditional nonlinear equality constraints model 2. Conditional nonlinear
equality constraints time invariant model 3. Least restrictive multiple group nonlinear
equality constraints time invariant model (all parameters are freed) 5.Multiple group model
with the Intercept of Service Support freed, 6.Multiple group model with the Slope of
Service Support freed, 7.Multiple group model with the Intercept of Social Support freed,
8.Multiple group model with the Slope of Social Support freed, 9.Multiple group model with
the Intercept of Intrapsychic Support freed, 10.Multiple group model with the Slope of
Intrapsychic Support freed, 11.Multiple group model with the Intercept freed, 12.Multiple
group model with the Slope freed, 12. Most restrictive multiple group nonlinear equality
constraints time invariant model (all parameters are fixed). * trend toward significance
71

C. Work Productivity Functioning: Model Specifications and Parameter Estimates
All fit indices for growth curve models of work productivity functioning are
shown in Table 11. In this section, model specifications and parameter estimates for
work productivity functioning as they correspond to the three aims are provided.
Subsection 1 illustrates the growth trajectory for the work productivity functioning of
the entire sample over a 12 month period (Aim 1). Subsection 2 discusses the full
conditional model, including all support covariates and their relationships with work
productivity functioning (Aim 2). The multiple group analyses in Subsection 3
address the moderating effects of minority status on individual relationships between
the support covariates and work productivity functioning (Aim 3).

1. Growth Trajectory: Unconditional Model to Test Aim 1
Due to the limited number of observations (3 time points), the unconditional
model was initially restricted to a linear growth model. Results from a X
2
test of the
model fit for the linear unconditional growth model for the work outcome at
baseline, 6 months and 12 months indicate statistical significance (X
2
(1)=5.54,
p=0.02), suggesting a poor fit between the linear growth model and the data. With
the exception of the comparative fit index (CFI=.95), all fit indices were below
acceptable criteria for an adequate model fit (TLI=.85; SRMR=.06; RMSEA=.18,
90% confidence interval, 0.06-0.34).
72

Two primary modifications were made to increase the fit of the linear growth
model: freeing the last time point observation of work productivity to turn the model
into a nonlinear unconditional model and imposing the equality of constraints. Note
that four observations are generally required for quadratic models (Muthen &
Muthen, 2007) and only three were available for this study. Fit indices for the
nonlinear model could not be computed due to the lack of a positively defined
residual covariance matrix. Solving this problem required the imposition of equality
constraints on work productivity functioning at each time point. Fit indices for all
alternative latent growth curve models for work productivity are shown in Table 15
in the appendix. As shown, the best fit was observed for the nonlinear growth curve
model with equality constraints (X
2
(2)=0.74, p=0.69). The other fit indices used in
this research provided additional support for this model (CFI=1.00; TLI=1.02;
SRMR=.02; RMSEA=.00, 90% confidence interval, 0.00-0.13).
The parameter estimates data for the unconditional work productivity
functioning model indicate a significant mean initial level (z=15.39, p=0.00) and a
positive mean rate of change in work productivity functioning (z=3.61, p=.00) from
baseline to 6 months (Table 12).

73

Table 12. Nonlinear Unconditional Model Parameter Estimates for Work

Productivity Functioning

The average work productivity level of 1.96 at baseline is significantly different from
0. According to the significant mean growth rate across the sample, individual
participants gained an average of 0.57 units of work productivity from baseline to 6
months (Fig. 7). Variances for both the intercept (z=3.76, p=.00) and slope (z=2.77,
p=.01) were found to be statistically significant, indicating significant variability in
both initial work productivity functioning level and growth rate for the entire sample.
Data on initial status variability and growth is depicted in Figure 8. The non-
significant covariance between the intercept and slope indicates a lack of any
relationship between an individual’s baseline position and growth rate in work
productivity functioning.

b S. E. z-score p-value
Mean
Intercept 1.96 0.13 15.39 0.00
Slope 0.57 0.16 3.61 0.00
Variance
Intercept 1.16 0.31 3.76 0.00
Slope 1.17 0.42 2.77 0.01
Covariance
Intercept-Slope -0.01 0.27 -0.02 0.98
74

Figure 7. Work Productivity Functioning Score Over Time for Sample

Figure 8. Random Work Productivity Trajectories of N=50 Participants

75

2. Support Covarites: Conditional Model to Test Aim 2
The next step was to construct a conditional model based on the nonlinear
unconditional equality constraint model. Support covariates were added to the
nonlinear unconditional model to determine their effects on the initial status and rate
of change of work productivity functioning. The results reveal a good model fit
(X
2
(5)=4.61, p=0.47; CFI=1.00; TLI=1.01; SRMR=0.02; RMSEA=0.00, 90%
confidence interval, 0.00-0.114) for the conditional model. As shown in Table 13,
both the intercept of the intercept (z=3.366, p=0.000) and the intercept of the slope
(z=-2.404, p=0.016) were statistically significant. The intercept of the intercept
represents the predicted value of work productivity (b=1.307, SE=0.388) when time
and all covariates equal zero. The intercept of the slope represents the slope for time
(b=-1.23, SE=0.15) when all support covariates are held at zero. Both residual
variances for the intercept (z=2.71, p=0.01) and slope (z=2.32, p=0.02) were found
to be statistically significant, suggesting the possibility of other covariates explaining
the leftover work productivity functioning variance.
When the intercept was regressed on system, social, and intrapsychic support,
only intrapsychic support (z=5.96, p=0.00) was found to exert a significant impact
on the initial level of work productivity functioning. Intrapsychic support ( β=0.65,
SE=0.10) explained 65% of the variance work productivity functioning. A trend
toward significance was found for social support at baseline (z=-1.81, p=.07).
Despite its significant impact on initial level of work productivity functioning,
76

intrapsychic support did not exert a significant impact on the growth rate in work
productivity functioning (z=1.13, p=0.26). However, social support and system
support did contribute to the work productivity functioning growth trajectory. Social
support had the most significant impact on the growth rate of work productivity
functioning (z=2.59, p=0.01) and system support made a smaller but still significant
contribution (z=1.94, p=0.05).

77

Table 13. Conditional Model Parameter Estimates for Work Productivity

Functioning

b S.E. z-score p-value
Intercept
Service Support -0.00 0.00 -1.08 0.28
Social Support -0.15 0.09 -1.81 0.07
Intrapsychic Support 0.46 0.08 5.96 0.00

Slope
Service Support 0.01 0.00 1.94 0.05
Social Support 0.27 0.11 2.59 0.01
Intrapsychic Support 0.11 0.09 1.13 0.25

Intercepts
Intercept 1.31 0.39 3.37 0.00
Slope -1.23 0.51 -2.40 0.02

Residual Variance
Intercept 0.69 0.26 2.71 0.01
Slope 0.85 0.37 2.32 0.02

3. Moderating Effects: Multiple Groups Model to Test Aim 3
Performing a multiple group analysis for the least restrictive model was the
first step toward determining whether minority status moderated relationships
between the individual support covariates and work productivity outcome. As shown
in Table 11, all of the fit indices for the least restrictive model with no constraints on
any of the parameters for either group indicate a good model fit (X
2
(11)=7.33,
p=0.77; CFI=1.00; TLI=1.05; SRMR=0.04; RMSEA=0.00, 90% confidence interval,
0.00-0.09).
78

The parameter estimate results shown in Table 14 indicate some notable
differences between the individuals in the minority and non-minority groups, with
the former showing greater growth trajectory homogeneity. According to the growth
trajectory for work productivity functioning when all covariates are held at zero,
individuals in the non-minority group (b=-1.60, SE=0.73) experienced slower growth
trajectory increases than minority individuals (b=-1.15, SE=-0.75) (Figure 9).

Figure 9. Work Productivity Functioning Over Time by Minority/Non-minority Groups

Intrapsychic support captured the most variance in baseline work productivity
levels for both groups, with a trend toward significance in the work productivity
growth trajectory for non-minority group members (z=1.82, p=0.07) but not for
minority group members (z=-0.30, p=0.77). Among minority individuals, social
support (z=2.41, p=0.02) was found to have greater influence on work productivity
79

growth trajectory compared to individuals in the non-minority group (z=1.49,
p=0.14).
A second multiple group analysis for the most restrictive model (i.e., all
parameter constraints are equal for both groups) indicated a comparable model fit
(X
2
(19)=13.29, p=0.0.82; CFI=1.00; TLI=1.05; SRMR=0.06; RMSEA=0.00, 90%
confidence interval, 0.00-0.07). Results from a chi square test ΔX
2
(8)=5.97 were not
significant at a p=.05, suggesting that the two models were similar for both the
minority and non-minority groups. Since chi-square tests are sensitive to sample size,
additional multiple group analyses with parameters used one at a time as constraints
were conducted. In this situation, the test results for the relationship between
intrapsychic support and the slope ( ΔX
2
(1)=2.18) indicated a trend toward
significance, suggesting that minority status holds a potential for moderating the
impact of intrapsychic support on the growth rate for work productivity functioning.
As shown in Table 14, intrapsychic support exerted a significant and positive impact
on the work productivity functioning slope for the non-minority group (b=.24,
SE=0.13), but not for the minority group (b=-0.04, SE=0.14).

80

Table 14. Multiple Group Model Parameter Estimates for Work Productivity

Functioning

Non-minority Minority
b S.E.
z-
score
p-
value b S.E.
z-
score
p-
value
Intercept
Service Support -0.00 0.00 -0.60 0.55 -0.00 0.00 -0.92 0.36
Social Support -0.24 0.12 -1.90 0.06 -0.09 0.12 -0.74 0.46
Intrapsychic Support 0.41 0.12 3.42 0.00 0.51 0.10 4.93 0.00

Slope
Service Support 0.01 0.00 1.52 0.13 0.01 0.00 1.48 0.14
Social Support 0.20 0.13 1.49 0.14 0.41 0.17 2.41 0.02
Intrapsychic Support 0.24 0.13 1.82 0.07 -0.04 0.14 -0.30 0.77

Intercepts
Intercept 1.67 0.65 2.58 0.01 1.00 0.49 2.05 0.04
Slope -1.64 0.73 -2.24 0.03 -1.15 0.75 -1.53 0.13

Residual Variances
Intercept 0.95 0.36 2.63 0.01 0.51 0.29 1.76 0.08
Slope 0.93 0.47 2.00 0.05 0.69 0.47 1.49 0.14
Notes: Non-minority=Euro-Americans, Minority=combined African Americans and Latinos

4. Summary of Work Productivity Functioning Results
The nonlinear unconditional model offered a better description of the
observed work productivity data than the linear growth of independent living. As
shown in Figure 8, most growth occurred from baseline to 6 months, after which it
started to decline. For the total sample, intrapsychic support had the greatest impact
on the baseline level of work productivity, explaining most of the variance of the
initial status. Social support, which showed a trend toward significance at baseline,
was the most important factor in the change trajectory over the three time points,
81

followed by system support. Higher levels of both social and system support are
associated with greater work productivity over time. According to a comparison of
minority and non-minority groups, intrapsychic support had the greatest impact at
baseline for both groups. Furthermore, intrapsychic support was identified as the
most important growth trajectory factor for individuals in the non-minority group,
especially in terms of work productivity. For minority individuals, social support was
the strongest growth trajectory factor.
82

CHAPTER V
DISCUSSION

To the author’s knowledge, this is the first examination of the relationship
between available support systems for individuals with SPMI and their functional
outcomes over a 12 month period. The current psychosocial rehabilitation literature
suggests that sociocultural factors play an important but hard-to-detect role in
identifying functional outcome predictors for individuals with SPMI (Barrio, 2001;
Iyer et al., 2005; Snowden & Yamada, 2005). The sociocultural factors examined in
this study (service, social, and intrapsychic support) all have strong potential to assist
in PSR intervention.
Unlike previous studies whose primary focuses have been on single-system
analyses of specific PSR dimensions, this study reflects the multidimensional nature
of the PSR framework. Multi-systems analyses of support dimensions were
performed to distinguish among the different levels and facets of support systems
that can predict functional outcomes and elucidate functional change mechanisms.
The study findings suggest that the functional change mechanism is a complex issue.
The discussion of the results is organized by functional outcomes (independent living
and work productivity). These two sections respectively present discussion and
reflection on the possible interpretations of the results from Aims 1-3. The support
covariates are discussed in the order of the strength of their association with each
83

functional outcome for each study aim. Both the significant and trend toward
significance moderating effects are reported and discussed. Finally, the strengths
and limitations of the study, the future research directions, and the possible
implications for direct community based practice are discussed in this chapter.

A. Independent Living Functioning
Given the shift to a recovery model for persons with SPMI (Davidson et al.,
2005; Jacobson & Greenly, 2001), understanding change mechanisms for functional
outcomes has increasingly become the primary goal in outcome research (Brekke &
Long, 2000; Mckibbins et al., 2004). Independent living functioning is of particular
importance, since it is considered one of the first steps toward community integration
(Fellin, 1993; Flynn & Aubry, 1999, Wong & Solomon, 2002).

1. Growth Trajectory of Independent Living Functioning
While significant variability was noted in the initial level of independent
living functioning within the sample, the sample-wide growth trajectory indicates a
homogeneous and continuous upward movement from baseline to 12 months (Aim
1). A difference of a single point on the RFS is considered clinically significant; the
b=.43 x2 (b=.86) rate of change over a one-year period found in this study is
therefore considered close to clinical significance (Goodman et al., 1993). The
improvement observed in the sample is consistent with existing studies on
84

independent living outcomes. In other studies, individuals with SPMI have shown
improvement in independent living outcomes over 3 years (Phillips et al., 2001), 5
years (Lipton, Siegel, Hannigan, Samuels, & Baker, 2000) and up to 10 years (Drake
et al., 2006). Specifically, Lipton et al. (2000) reported that 75% of their sample
maintained stable housing after 1 year, 64% after 2 years, and 50% after 5 years.
In summary, this study’s two primary findings regarding growth curve
trajectory could be interpreted as (a) individuals in the sample were more likely to
improve in their independent living functioning, and (b) even those whose
independent living functioning was unstable at baseline showed potential to improve
over a one year period.

2. Support Covariates of Independent Living Functioning
(a) Intrapsychic support.
Regarding Aim 2 (determining the most predictive support covariate of
independent living functioning), the results show that intrapsychic support explained
more of the baseline variance in independent living than the other two types of
support. Intrapsychic support for individuals with SPMI was found to be positively
related to their level of independent living functioning, suggesting that higher levels
of internal support are associated with higher levels of independent living
functioning. Despite the positive association at baseline, intrapsychic support did not
exert a significant impact on the independent living outcome growth trajectory.
85

Given that intrapsychic deficit is frequently conceptualized as a core aspect
of schizophrenia that other functional deficits emerge from (Heinrichs et al., 1984),
the finding that the level of intrapsychic support is associated with independent
living functioning at baseline is as could be expected. The finding is consistent with
previous literature; the negative effects of deficit symptoms, such as intrapsychic
drive, on functional outcomes such as independent living are well established in
schizophrenia research (Ho et al., 1998; Kirkpatricks & Fisher, 2006; Kurtz, 2005;
Mileve et al., 2005), with at least one current research team suggesting that decreases
in deficit symptoms lead to increases in independent living functioning
(Leeuwenkamp et al., 2007). In light of the consumer-driven principle of PSR, it
makes intuitive sense that intrapsychic support might be essential to an individual’s
functional status, but it is surprising that this relationship is only found at baseline. It
is not clear why the relationship is not observed at 6 month or 12 month follow up
assessments as little previous literature has examined longitudinal patterns of persons
with SPMI. Combined, these results invite the interpretation that intrapsychic
support could be initially important to the independent functioning of adults with
SPMI who are not fully engaged in PRS. However, as individuals with SPMI begin
to receive and participate in PRS, they may learn skills, acquire additional external
support, and gain access to tangible resources (e.g., benefits, supported housing) that
do not rely on internal processes to obtain or sustain independent living skills.
86

It has been shown that consumers in PRS can obtain greater independence in their
living situation than those without such services (Bustillo et al., 2001; Fakhoury et
al., 2002).
(b) Social support.
In this study, even though it did not explain as much variance as intrapsychic
support, social support did have a significantly positive relationship with the initial
status of independent living functioning. Individuals in the sample who had higher
levels of social support tended to have higher initial levels of independent living
functioning. However, the effects of social support change over time. Social support
has a significantly inverse relationship with the growth trajectory of independent
living functioning, suggesting that increases in social support correspond to
decreases in the rate of improvement.
The positive relationship between social support and independent living
functioning at baseline is consistent with findings from other studies indicating that
higher levels of social support may increase functional outcomes (Clinton et al.,
1998; Erickson et al., 1998; Evert et al., 2002). Corrigan and Phelan (2004) offer an
explanation in that they found that larger network size and greater satisfaction with
social relationships have positive associations with stronger focus on goals and
success. Although more research is needed to determine how social networks help
result in more goal oriented successes, one possible reason is that social network
members may actively encourage individuals with SPMI to achieve functional
87

outcomes such as independent living skills development. The finding is also
consistent with other studies that describe such support as a good predictor of initial
help-seeking and community integration (Bengtsson-Tops & Hansson, 2001;
Carpentier & White, 2002; Clinton et al., 1998). Furthermore, since baseline
directionality cannot be established, this result may also be interpreted as meaning
that higher levels of independent living result in higher levels of social support. As
suggested in Hansson et al. (2002), individuals with more self-care skills are also
capable of engaging in more satisfying social interactions.
Despite the positive association at baseline, over time, social support appears
to negatively influence independent living functioning. The finding suggest that
over time, the more social support individuals have, the less likely they are to
increase their independent living functioning. This negative relationship between
social support and independent living functioning initially appears to contradict
previous findings (Calsyn & Winter, 2002; Corrigan & Phelan, 2004; Hansson et al.,
2002). A further review of the support literature, however, reveals much more
complexity in the relationship of social support with functional outcomes (Erickson
et al., 1998; Yamada et al., 2000). Therefore, it is possible that this finding may be
interpreted several ways. As previously mentioned, social support can have multiple
functions (Rogers et al., 2004)—for instance, strong positive social networks of
individuals with SPMI can assist with their daily living or housing needs (Callsyn &
Winter, 2002; Snowden, 2007). Although we do not currently understand the end
88

results of the assistance provided by the social networks, one possible consequence
may be that this reliance on the social network may impede movement toward
housing independence. Another possible explanation for the negative relationship
between social support and independent living over time may be that the greater the
capability to live independently in an apartment, the fewer the opportunities for
human social contact (Siegel et al., 2006). It may be that living at home with family
members or in a supported living environment like a board and care with other
residents allows for more regular social interactions. Finally, as consumers obtain
more independent living situations they may begin to receive less support from
supportive peers and staff within the PSR environment (Barrio et al., 2003); better
functioning often is viewed as indicative of less need for support in a system that is
focused on serving those most in need. Over time, outside of the supportive PSR
setting, consumers may experience stigma and other stressors that interfere with their
independent living skills and as they decompensate or struggle they may again begin
to receive increasingly greater amounts of social support. Qualitative research is
needed to provide further clarity into the dual nature of social support reported in this
study.
(c) Service support.
Unlike intrapsychic and social support, system support has an inverse
relationship with independent living outcome at baseline—in other words, the more
contact the individuals in the study sample had with service providers, the lower the
89

level of their independent living functioning. This is consistent with findings by
Barrio et al. (2003) regarding a study of 4, 249 multi-ethnic individuals with SPMI
who were receiving county wide mental health services. Barrio and her colleagues
found that individuals living in institutional residence were more likely to use more
case management services than those in independent living situations. Again, this is
in line with standard social service treatment processes since service providers tend
to spend more time with lower functioning individuals because they need more
assistance. Those with higher levels of independent living skills might be less likely
to need or seek services from their providers.
System support did not significantly improve independent living functioning
over time. This is inconsistent with previous findings connecting more provider
contact with improvement in functional outcomes (Brekke et al., 1997, 1999). There
are several possible explanations for this inconsistency. In Brekke et al.’s studies,
service intensity was measured in terms of minutes of staff-consumer contact,
whereas in this study it is measured by days of staff-consumer contact. The more
precise operationalization of the variable may account for the disparate results. In
addition, Brekke et al. combined independent living, social support, and work
functioning into a single global functional outcome measure, whereas in this study
independent living and work productivity were conceptualized as being distinct from
each other, and separate analyses were conducted for each outcome. Other types of
90

functioning may have altered the observed relationship with intensity of service
support reported by Brekke and colleagues.
3. Moderating Effects of Minority Status on Independent Living Functioning
A small number of compelling studies on ethnic minority mental health
services offer evidence that ethnic minorities experience outcome disparities (Bae et
al., 2004; Phillips et al., 2001; Rogers et al., 1997). Although the practice of
tailoring interventions to make them culturally relevant has been gaining momentum
in recent years (DHHS, 2001), the limited number of ethnic minorities actually
included in research (Imada & Schiavo; 2005, Mak et al., 2007; Vega & Lopez,
2001) has made it difficult to understand how sociocultural factors influence
treatment and outcomes. Ideally, separate analyses for the three ethnic groups that
were the focus of this research would have provided a richer understanding of the
effects of subtle cultural differences on change mechanisms. Still, findings from
multiple group analyses between minority and non-minority groups revealed
potential group differences that would otherwise be masked if a simple analysis of
the entire sample were conducted. In this section, statistically significant
moderating effects are reported as “moderating effects”. In light of the limited
number of studies including sufficient numbers of ethnic minorities for subgroup
analyses of outcome data (Mak et al., 2007), and the limited power to detect true, but
small differences in this study, non-significant group differences of interest are
reported as “group differences” here. Even though it is not possible to conclude that
91

a minority status difference exists, it is important to provide leads for potential
relationship warranting future study.
(a) Social support.
Group differences were observed in the relationship between social support
and the baseline level of independent living functioning. Although social support
demonstrated a positive relationship with the initial level of independent living
functioning for both groups, social support was a significant predictor of independent
living functioning only for minority group consumers. Although not testable in this
study, this finding is consistent with the current literature on the collective
orientation of ethnic minorities, through which behaviors are generally influenced by
group or social network values (Hofstede & McCrae, 2004; Mak, 2005; Snowden,
2007). Thus, while it appears that internal motivation is important to functional
outcomes, for ethnic minorities the influences of a social network seems to supersede
internal drive or desire. Other studies have found several positive benefits to
consumers who maintain a strong social network. A recent World Health
Organization report (Hopper, 2007) study found that individuals with SPMI in less
developed countries tend to have better daily living functioning than those in
developed countries. One interpretation is that close social networks were
responsible for the better functioning observed in countries where there is less access
or use of PSR or even medication. Along the same line, Hopper and Wanderling
(2000) found that a better integrated family structure and opportunities for mutual
92

responsibilities of consumers and their families are associated with increases in
independent living functioning.
In this study, minority status has a statistically significant moderating effect
on the relationship between social support and the independent living growth
trajectory. Specifically, social support had a highly significant inverse relationship
with independent living growth trajectory for ethnic minorities. In contrast, social
support was not a significant predictor of improvement for non-minority individuals.
In other words, minority status appears to alter the growth trajectory- minority
consumers with greater social support have a tendency to decrease in independent
living functioning over the one year period of measurement. While this finding may
initially be perceived as counterintuitive, it is actually consistent with the collective
orientation of many ethnic minorities. While not measured in this study, African
Americans and Latinos may have a greater collective self identity than do Euro-
Americans (Triandis, 1994). A number of studies have shown that minority families
provide more care to family members with SPMI (Mak, 2005; Snowden, 2007) and
that minority individuals with SPMI tend to live with family members (Barrio et al,
2003). Snowden specifically found that African Americans and Latinos were more
likely to live with their families and receive family support than Euro-Americans.
Although at this time we do not fully understand how a collective oriented
identity impacts functional outcomes, we could speculate that individuals with
sufficient support from family members and friends may have relatively intact self-
93

care skills, but over the long run become less inclined to live independently as it is
incongruent with their family values that encourage remaining at home with family
(Barrio, 2000). Another possible explanation may be that due to their collectivist
orientation, the guilt and sadness of leaving close-knit family and friendship units
may make it difficult to move toward independent housing situations. Conversely,
those receiving limited social support may not have the same sense of family
obligation and are naturally forced toward independent living. Despite the common
impression that all ethnic minorities are marked by strong family support, lack of
family support is also very much a reality in non-Western cultures (Yang, Law,
Chow, Andermann, Steinberg, & Sadavoy, 2005). Independent living may also be an
indirect reflection of that lack—for instance, individuals with severe and persistent
mental illness frequently “burn their bridges” with families and friends and lose their
once close-ties. There is a need for additional research to shed light on these possible
interpretations.

B. Work Productivity Functioning
1. Growth Trajectory of Work Productivity Functioning
In this study, the trajectory for work productivity functioning differed sharply
from the linear trajectory for independent living functioning. Individuals in the study
sample experienced rapid improvement in work productivity from baseline to 6
months (Aim 1), which is similar to the growth trajectory described by Larson, Barr,
94

Kuwabara, Boyle and Glenn (2007). However, work productivity functioning began
to decline at 6 months, a pattern consistent with that reported by Gates et al. (2005),
who found that job loss tended to occur at two time points: during the first month,
and close to the sixth month of employment.
Although not amenable to analysis in this study, a decrease in service
intensity is one possible explanation for declining work productivity functioning that
was observed six months after the baseline assessment was completed. In previous
studies, using different samples from within the same PSR agencies, Brekke et al.
(1992, 1999) found that service intensity was nearly three times higher during the
first six months than during the second six months of the study period. Thus it may
be that the work functioning of individuals with SPMI in this sample increased in the
first 6 months due to the likely higher service intensity and also declined as service
intensity decreased. Another possible explanation is the gradual occurrence of job-
related problems, such as poor onsite relationships with co-workers and bosses
(Auerback & Richardson, 2005)—that is, individuals with SPMI may have the
necessary job skills to perform work-related tasks, but do not have the necessary
social skills to handle interpersonal workplace relationships. Becker et al. (1998)
found that more than one-half of their study sample left their jobs without being fired
or in anticipation of new jobs, and that pre-employment skills training did not reduce
the risk of early job termination – suggesting that there may be other on the job
factors that may reduce work productivity.
95

2. Support Covariates of Work Productivity Functioning
(a) Intrapsychic support.
Intrapsychic support explained the most variance of the baseline level of
work productivity functioning (Aim 2). The statistically significant positive
relationship indicates that higher levels of intrapsychic support is related to higher
level of work productivity functioning. Given that PSR intervention is consumer-
driven, it makes intuitive sense that individual motivational processes could be
essential in initiation of the treatment process (Auerback & Richardson, 2005;
Mueser et al., 2001). Mueser et al. found that those individuals with SPMI who
wanted to work were three times more likely to find jobs than those who were not so
motivated. In Auerbach and Richardson’s qualitative study, work was a “contributor
to the person’s identity”, a “conduit to personal growth” and provided “various types
of personal gains” (2005, p.268). These personal benefits of work cited by
individuals with SPMI could be reasons for the strong relationship found here
between intrapsychic support and work productivity.
Yet over time, intrapsychic support did not significantly influence the work
productivity functioning growth rate. The results from the present study take on new
meaning in light of Rosenheck et al.’s (2005) findings that although consumers who
scored higher on the intrapsychic functioning scale were more motivated to work, the
expression of a strong desire to work did not translate into more work days compared
to individuals who expressed a weak desire to work. Accordingly, the present
96

findings suggest that although individual motivation may be necessary to get the
employment process started, there may be other factors at play and that there may be
other barriers that cannot be overcome by willpower alone (Auerback & Richardson,
2005; Cook, 2006).
(b) Social support.
The inverse relationship between social support and work productivity at
baseline is inconsistent with previous findings that high social support facilitates
functional outcomes such as work productivity (Evert et al., 2003; Rogers et al.,
2004). The present study results indicate that having a more positive relationship
with more family members and friends may lower work productivity. Although it
can not be discerned from this study, a possible explanation may be that more
positive relationships with family members and friends translate into greater
financial assistance for individuals with SPMI. More family financial support (not
uncommon in ethnic families) means that individuals with SPMI have less reason to
work (Bengtsson-Tops & Hansson, 2001; Walsh & Connelly, 1996). According to
Goldberg et al. (2003), social networks in ethnic cultures tend to be more
“multiplex” compared to Euro-American social networks. Multiplexity in social
networks means that other functions (e.g., providing financial assistance and places
to stay) are provided in addition to emotional support (Goldberg et al., 2003).
In this study, a significant positive association was found between social
support and the growth trajectory in work productivity functioning—in other words,
97

support from family members and friends apparently helps individuals maintain and
increase their work productivity (Everet et al., 2003; Rogers et al., 2004; Yanos et
al., 2001). Although it is beyond the scope of the current study, there may be on-the-
job stressors that hinder work productivity, and social support could help individuals
with SPMI deal with these stressors and maintain employment (Haertl, 2005).
Corrigan and Phelan (2004) have also found that individuals expressing greater
satisfaction with support levels are more goal- and success-oriented, both of which
obviously help in terms of maintaining or increasing work productivity functioning.
An alternative explanation could be that greater social support helps individuals
develop better social skills that they can use at work. There are numerous studies
stating that the ability to get along with others is one of the more important factors in
vocational performance (Anthony & Jansen, 1984; Kee, Green, Mintz & Brekke,
2003).
(c) Service support.
While no significant impact was found for service support on the baseline
level of work productivity functioning, it was found to be an important factor for
work productivity functioning over time. The former finding is in line with the
treatment philosophy of PSR, which states that all consumers should be given equal
opportunities for change regardless of their level of functioning (Brekke, Hoe, Long,
& Green, 2007). The latter finding is consistent with Brekke et al. (1992, 1997), data
that indicate that more contact with service providers is helpful in terms of
98

improving work productivity functioning. This suggests that ongoing support from
service providers is crucial for maintaining and increasing the work productivity of
individuals with SPMI.

3. Moderating Effects of Minority Status on Work Productivity Functioning
(a) Social support.
In this study, work productivity functioning at baseline and over time differed
between minority and non-minority consumers (Aim 3). A trend toward significance
was found for social support among individuals in the non-minority group, but no
impact was found for baseline levels among minority consumers. As previously
mentioned, the inverse association of social support and work productivity noted for
non-minority consumers may imply that those with less social support are more
likely to need to work to supplement their incomes because they cannot depend on
their social support networks for financial assistance. This speculation needs to be
further explored in future studies.
Similar to independent living, a group difference by ethnic minority status
was also found in the relationship of social support and the work productivity growth
trajectory—that is, the association was statistically significant for ethnic minorities
but not for the non-minority group. This appears to be consistent with existing
studies asserting the strong influence of collective orientation among ethnic
minorities (Oyserman et al., 2002; Singelis, 1994). Employment is highly valued in
99

many ethnic minority cultures because it shows that an individual is capable of
contributing to the collective good (Falicov, 2001). Although we cannot determine
the factors responsible for the results in this study, it may be that close relationships
with social groups encourage individuals with SPMI to pursue employment. It could
also be that ethnic minority individuals feel obliged to contribute to their groups’
well-being by earning an income. In the sample used in the present study, it appears
that the social group could be acting as an external motivating factor (as opposed to
intrapsychic support) when impacting growth in work productivity functioning for
ethnic minorities.
(b) Intrapsychic support.
Intrapsychic support was the most important determinant of growth in work
productivity functioning for Euro-Americans. The significant positive correlation
indicates that for non-minority individuals, internal motivation was key to
improvement in work productivity. This finding appears to align with the literature
on individualism and collectivism that describes non-minority individuals as
individualistic (Oyserman et al., 2002; Singelis, 1994), and therefore more reliant on
self-motivation processes than on support from friends, family members, or service
providers. Since minority individuals tend to be more collectivist in their orientation
and more strongly affected by their social systems, intrapsychic support may not be a
significant predictor of their work productivity growth.

100

C. Within Group Comparisons
None of the support covariates significantly explained the rate of improvement
of independent living functioning. In all other cases, intrapsychic support had the
most impact on independent living functioning and work productivity functioning for
Euro-Americans. In most situations, intrapsychic support increases the baseline or
growth score for both functional outcomes. Since PSR is consumer-driven, it makes
intuitive sense that individual motivational processes could drive the treatment
process (Auerback & Richardson, 2005; Mueser et al., 2001). This appears to be
especially true for individualistic Euro-Americans who may be more driven by their
internal motivational drive (Oyserman et al., 2002; Singelis, 1994) than their social
support.
Within group comparisons are a little bit more complicated for ethnic
minorities. For both functional outcomes, at baseline, ethnic minorities appear to be
similar to Euro-Americans as in the importance of internal motivational drives. This
again underscores the main tenet of PSR-to provide consumer-driven services. Yet,
despite the baseline relationship between intrapsychic support and functional
outcomes for ethnic minorities, social support was the overall paramount influence.
Given the potential of a more collectivistic nature of ethnic minorities, it stands to
reason that these consumers could be more impacted by their social networks than
their own internal drive (Barrio, 2000).
101

However, it is important to note that social support does not always bring
about positive change for ethnic minorities. Over time, social support may assist in
the improvement of work productivity functioning but actually aid in the decline of
independent living functioning. From these results, it can be speculated that social
support may not always have positive effects on individuals with SPMI. In fact
Weisman et al. (2005) found that perceived family cohesion is related to higher
emotional distress for ethnic minorities but not for Euro-Americans. More social
support may mean more pressure to do what is right for the group even if the
individual does not like or agrees with it (Singelis, 1994). Although it can not be
validated in this study, these results may also be explained by the different cultural
values that ethnic minorities place on independent living and work productivity. It is
normal and often valued for multiple generations to live together under one house in
collectivistic cultures (Singelis, 1994). Given the importance of living together in
collectivist cultures, it could be difficult for the individuals with SPMI to move
toward independent living (Goldberg et al., 2003; Jenkins & Schumacher, 1999); the
social network would most likely encourage or pressure the individuals to stay at
home. Thus, independent living, a highly valued PSR outcome is not as valued in
many ethnic minority communities (Barrio, 2000). Work, on the other hand, is
highly valued within collectivistic communities because the financial benefits from
work could ultimately benefit the whole group (Falicov, 2001).
102

Thus it may be reasonable for an ethnic minority social network to encourage and
support their members to work.

D. Study Strengths
This study has a number of strengths worth noting. These strengths have
significantly contributed to advancing an understanding of the change mechanisms in
PSR functional outcomes. The most notable strength of this study is the exploration
of the relationship of three types of available support systems for individuals with
SPMI and their functional outcomes over a 12 month period. Most studies on
support systems for individuals with schizophrenia have been limited to examining
single systems (service intensity, family support, peer support, etc). By taking into
consideration multidimensional support systems, it was possible to statistically
determine their individual effects while controlling for the other support systems.
This is not only consistent with the multidimensional characteristics of PSR
frameworks, but provides a greater understanding of the effectiveness and range of
the support systems operating for consumers engaged in PSR.
Another study strength is the longitudinal data with three time points. A
cross-sectional design would have only shown correlations between two variables at
one time point without providing information on directionality. Having three
outcome time points allows for a better understanding of how support covariates
impact functional outcomes over time. The longitudinal data made it possible to see
103

how the impact of support systems may change over time. As the findings indicate,
a support covariate may have a positive impact on the functional outcome at baseline
but have a different impact over time.
Given the limited number of clinical schizophrenia studies involving ethnic
minorities, the inclusion of African Americans and Latinos is another strong point of
this study, as were the multiple group analyses, which assisted in the search for
potential cross-cultural differences. Although this study’s comparisons between one
minority and one non-minority group provided less culturally specific information
compared to a tri-ethnic comparison, the results still exceeded those that would have
been produced had the entire sample been combined into a single group. The
analyses by ethnic minority status offer an important preliminary glimpse of
sociocultural differences in the arena of PSR.
Additional strengths are noted in the conceptualization and operational
definition of the functional outcome constructs. Most functional outcome researchers
use a single factor to represents a sum of all or some type of functioning. In this
study, independent living functioning and work productivity functioning were
analyzed separately, thus allowing for a better understanding of the individualized
growth trajectories and facilitating the development of clearer implications for
community-based service providers.

104

E. Study Limitations
Several study limitations are important to acknowledge. One limitation was
having three instead of four time points, thereby inhibiting the construction of a
growth curve model (Muthen & Muthen, 2007). For the work productivity outcome,
additional time points would have made it possible to determine declines or increases
after 12 months; the nonlinear model precluded such an analysis.
The small sample size for the minority and non-minority groups comparative
analysis is another limitation. The small sample size in this study inhibited the
researcher’s ability to detect statistical significance in relationships, especially small
and medium effects. Results from chi square and fit indices for the multiple group
analysis indicate that the unconstrained model had a good fit with the data. While
this suggests some differences may exist between the minority and non-minority
groups, the statistical power was insufficient for determining specific parameters.
This increases the potential for Type II errors—that is, concluding no differences
when one or more might exist.
Another limitation due to combining two ethnic groups into one is that in
doing so, we may overlook multiple differences in culture, world view, and
socioeconomic, political, and service disparities. Furthermore, there is great
heterogeneity within each ethnic group—for example, the Latino designation did not
acknowledge distinctions among Mexicans, Latin Americans, or other groups with
Hispanic backgrounds. Additionally, level of acculturation of Latino consumers was
105

not taken into consideration. Finally, the Latino sample size was significantly
smaller than the African American sample size. Thus interpretation of the minority
results may not truly be representative of Latinos.
Despite the attempt to clearly define the study constructs, limitations still
exist in providing a definite operationalized definition of independent living and
work productivity that would make findings comparison with other studies more
conceptually defensible. The relationships between the support covariates and
functional outcomes offered insight into change mechanisms, however, the social
support construct could have been better defined. In this study, social support was
defined as a combination of family and social relationships. Some researchers have
suggested that certain aspects of social support may be more important to different
ethnic groups—for instance, African Americans and Latinos may place more
importance on family and extended family networks (Ajrouch et al., 2001), while
Euro-Americans place more importance on friends and coworkers (Holschuh, 2003;
Kim & McKendry, 1998).
Another limitation is that this study only focused on three covariates (service
support, social support, and intrapsychic support) and one moderating covariate
(minority status). There may also be other covariates such as neurocognitive abilities
or quality of provider/consumers relationship that could equally or better detail the
functional outcome mechanisms.

106

F. Future Research Direction
To the author’s knowledge, this is the first examination of the relationship
between available support systems for individuals with SPMI and their functional
outcomes over a 12 month period. The study findings have two important
implications for the field of PSR research: 1) that the role of the support systems
within the functional outcome mechanism is complicated and complex and 2) this
study provides potential leads for future studies, especially for further examination of
minority/non-minority differences.
The steady improvement in independent living over a 12 month period and
the increase in work productivity from baseline to 6 months suggest that the
individuals in the sample generally improved in their functional outcomes. This
suggests promising results for PSR. Thus it would be beneficial for future studies to
look at whether external barriers such as policies to increase access to safe and
affordable housing, and jobs, in combination with the PSR consumer support system
would increase the chance that individuals with SPMI attain the goals of independent
living and work productivity.
Similar to other studies in which functional outcomes were observed as
following different pathways (Brekke, Raine, Ansel, Lencz, & Bird, 1997), the
unconditional model analysis results in this study depict different growth trajectories
for independent living and work productivity functioning. Consistent with the work
of Anthony and Liberman (1992) and Strauss and Carpenter (1972), these functional
107

domains appear to be “open linked systems” (Waghorn, Chant & King, 2007). Two
reasons for this assertion are (a) the significant Pearson correlation between the two
functional outcomes suggests that they are not completely independent of each other,
and (b) the different baseline levels and growth trajectories of independent living
functioning and work productivity functioning indicate a limited relationship
between the two. This finding has important implications for future research on
PSR service delivery. Research is needed to determine whether PSR treatment plans
tailored to the specific functional outcomes produce better outcomes than those who
provide more standardized or general services and how to best tailor services to be
most effective for diverse consumers. It would be also interesting to explore whether
PSR treatment plans are adjusted according to challenges and conflicts that arise as
consumers work towards their functional outcome goals.
Although the conceptualized model used in this study was found to have a
good fit with the data, other covariates may fit equally well. Therefore, researchers
should look for alternative models to use to test other possible covariates. The results
indicate that intrapsychic support and social support were important in the change
process for the study sampled. In particular, the findings show how the impact of
different support systems may differ according to time of measurement (at baseline,
6 months, 12 months) as well as according to minority status.
Future studies would benefit from more in-depth and qualitative exploration
of how intrapsychic characteristics and qualities such as the motivational processes
108

of individuals with SPMI affect treatment progress and change over the course of
treatment. Given that PSR is consumer-driven, addressing the motivational
processes of individuals with SPMI appears warranted (Mueser et al., 2001). Even
though service providers may intuitively understand the importance of an
individual’s motivational processes on functional outcomes, no concerted effort has
been made to specifically address or incorporate them into the treatment process
(Barch, 2005). The findings from this study suggest that the intrapsychic processes
of individuals with SPMI may be central treatment components that deserve to be
given greater weight during assessment and used more effectively for treatment
(Haertl, 2005).
Findings on the effects of social support in this study suggest a complex dual
nature. For the entire sample, social support had a positive association with
independent living functioning at baseline but a negative association over the long
run. Regarding the impact of social support on work productivity, social support
was related to decreased work productivity at baseline and increases at 6 months and
12 months. Three related research directions stemming from these findings are (a) a
need to further explore the dual nature of social support—that is, to provide more
evidence as to why social support can both help and hinder functional outcomes; (b)
a need to further understand the relationship between social support and each
specific functional domains; and (c) a need to learn how these influences may change
during the treatment process.
109

Results from multiple group analyses revealed cross-ethnic differences that
may not have been detected had no distinction been made between African
Americans/Latinos and Euro-Americans. A number of minority/non-minority group
differences were found; the most notable being the impact of social support on the
change trajectories for both independent living and work productivity functioning for
ethnic minorities. Social support did not influence the functional outcome change
trajectories for Euro-Americans, but did influence those trajectories for ethnic
minorities. The only significant predictor of work productivity change for Euro-
Americans was intrapsychic support.
These findings have several research implications. Cross-group differences
suggest that sociocultural factors could impact the types of support being mobilized.
First it is essential that this dissertation study be replicated, with the addition of a
larger sample size, with greater ethnic diversity so as to allow a minimum of a tri-
ethnic comparison. The findings in this study demonstrate the potential importance
for researchers to actively recruit a sufficient number of ethnic minority participants
in order to conduct complete statistical analyses that provide more detailed
information regarding sociocultural variations (Mak et al., 2007). In addition, it
would also be important for future studies to explore how the sociocultural
background of individuals with SPMI influences the assessment and intervention
phases of PSR. Barrio (2000) has discussed potential problems when the inherent
strengths of collectivist and family-centered societies are not taken into consideration
110

when designing and providing community-based treatment plans for individuals with
SPMI. Currently many PSR interventions are driven by Euro-American standards
and norms (Lefley, 1990). It may also be beneficial for future studies to explore
how PSR interventions could be adapted to include cultural strengths in order to
produce the best possible functional outcomes. Adjusting treatment strategies to fit
the cultural norms and realities of individuals with SPMI can reduce the number of
ineffective intervention strategies that are put into effect (Telles et al.,1995), thereby
potentially reducing disparities in functional outcomes for ethnic minorities (Phillips
et al., 2001).

G. Implications for Community-based Services
The study findings have two important implications for the field of PSR
research: 1) that the role of the support systems within the functional outcome
mechanism is complicated and complex and 2) this study findings bring about
potential leads for future studies especially for minority/non-minority differences.
The study findings raise as many complex questions for the field as answers.
Below are highlights of how the finding could be applied to community-based
services in general and according to two functional domains.

111

General Implications
• The effectiveness of PSR community-based interventions in
increasing independent living and work productivity functioning
outcomes is promising.
• The significance of intrapsychic support suggests that there may be a
need to address or incorporate intrapsychic support into the treatment
process, especially during assessment.
• The different growth trajectories suggest that treatment plans may
need to be tailored to specific functional outcome domains instead of
simply following a general treatment plan protocol.
• The dual nature of social support suggests a need for extra care to be
taken when service providers mobilize the social support networks of
individuals with SPMI, especially when determining how and where
such support assists or undermines functional goals.
• Group differences between the minority and nonminority groups
suggest a need to address the cultural background of consumers with
SPMI and potentially incorporate cultural strengths into the treatment
plan.

112

Work Outcomes
• The decline in work functioning at 6 months suggests a need for
service providers to probe more deeply about working conditions and
experiences to assist their clients in facing difficult jobsite situations
and to provide continuous employment support.

Living Situation
• The potential for isolation as consumers gain more independent living
functioning suggests that even though individuals with SPMI may
achieve their independent living functioning outcome goals, service
providers must continue to provide ongoing case management and to
focus on helping their customers maintain and build social networks.

H. Conclusion
This dissertation study represents an attempt to provide a more thorough
understanding of how functional outcomes change over a twelve month period for
individuals with SPMI, how multidimensional support systems influence these
functional outcomes, and how minority status might moderate these relationships.
The results of this study, provide further insight into change mechanisms for
functional outcomes, and can be used to establish a foundation for future
examinations of the role of specific sociocultural constructs that could be at play.
113

Most importantly, these findings support the need to further develop the study of
support systems in multiethnic samples to maximize the practical clinical
implications for community-based service providers. It is hoped that this research
will contribute to the efforts to reduce and eventually eliminate ethnic disparities in
mental health service outcomes.

114

REFERENCES

Acock, A. C. (2005). Growth curves and extensions using Mplus. Presented at
Academica Sinica. Retrieved October 25, 2007, from
http://oregonstate.edu/~acock/growth-curves/Growth Curve Presentation.doc.

Albert, M., Becker, T., Mccrone, P., & Thornicroft, G. (1998). Social networks and
mental health service utilization – A literature review. International Journal of
Social Psychiatry, 44, 248-266.

Anderson, R., Harada, N., Chiu, V., & Makinodan, T. (1995). Application of the
behavioral model to health studies of Asian and Pacific Islander Americans.
Asian American and Pacific Islander Journal of Health, 3 (2), 128-140.

Andreasen, N.C. (1982). Negative symptoms of schizophrenia: Definition and
reliability. Archive of General Psychiatry, 43, 136-144.

Anthony, W. A., & Jansen, M. A. (1984). Predicting the vocational capacity of the
chronically mentally ill: Research and implications. American Psychologist, 39,
537-544.

Anthony, W. A., & Liberman, R. P. (1992). Principles and practice of psychiatric
rehabilitation. In R. P. Liberman (ed.). Handbook of psychiatric rehabilitation.
Boston: Allyn and Bacon.

Auerback, E. S., & Richardson, P. (2005). The long-term work experiences of
persons with sever and persistent mental illness. Psychiatric Rehabilitation
Journal, 28, 267-273.

Ayalon, L., & Young, M. A. (2005). Racial group differences in help-seeking
behaviors. The Journal of Social Psychology, 145, 391-403.

Bae, S. W., Brekke, J. S., & Bola, J. R. (2004). Ethnicity and treatment outcome
variation in schizophrenia. The Journal of Nervous and Mental Disease, 192,
623-628.

Bachrach, L. (2000). Psychosocial rehabilitation and psychiatry in the treatment of
schizophrenia-what are the boundaries? Acta Psychiatrica Scandinavica, 10, 6-
10.

Baker, F. M., Stokes-Thompson, J., Davis, O. A., Gonzo, R., & Hishinuma, E., S.
(1999). Two-year outcomes of psychosocial rehabilitation of Black patients
with chronic mental illness. Psychiatric Services, 50, 535-539.

115

Barch, D. M. (2005). The relationships among cognition, motivation, and emotion
in schizophrenia: How much and how little we know. Schizophrenia Bulletin,
31, 875-881.

Barrio, C. (1998). A longitudinal cross-ethnic study of sociocentricity in
schizophrenia. Unpublished doctoral dissertation. University of Southern
California, CA.

Barrio, C. (2000). The cultural relevance of community support programs.
Psychiatric Service, 51, 879-884.

Barrio, C. (2001). Cultural and schizophrenia: A cross-ethnic growth curve analysis.
Journal of Nervous and Mental Disease, 189, 676-684.

Barrio, C., & Yamada, A. M. (2005). Serious mental illness among Mexican
immigrant families: Implications for culturally relevant practice. Journal of
Immigrant & Refugee Service, 3, 87-106.

Barrio, C., Yamada, A. M., Hough, R. L., Hawthorne, W., Garcia, P., & Jeste, D. V.
(2003). Ethnic disparities in use of public mental health case management
services among patients with schizophrenia. Psychiatric Services, 54, 1264-
1271.

Barton, R. (1999). Psychosocial rehabilitation service in community support
systems: A review of outcomes and policy recommendations. Psychiatric
Services, 50, 525-534.

Becker, D.R., Drake, R.E., Bond, G.R., Xie, H., Dain, B.J., & Harrison, K. (1998)
Job terminations among persons with severe mental illness participating in
supported employment. Community Mental Health Journal, 34, 71-82.

Beels, C.C. (1981). Social support and schizophrenia. Schizophrenia Bulletin, 7, 58-
72.

Bell, M. D., Fiszdon, J. M., Greig, T. C., & Bryson, G. J. (2005). Can older people
with schizophrenia benefit from work rehabilitation? Journal of Nervous and
Mental Disease, 193, 293-301.

Bell, M. D., & Mishara, A.L. (2006). Does negative symptoms change relate to
neurocognitive change in schizophrenia? Implications for targeted treatments.
Schizophrenia Research, 81, 17-27.

Bellack, A. S. (2006). Scientific and individual models of recovery in
schizophrenia: Concordance, contrasts, and implications. Schizophrenia
Bulletin, 32, 432-442.

116

Bengtsson-Tops, A., & Hansson, L. (2001). Quantitative and qualitative aspects of
the social network in schizophrenic patients living in the community.
Relationship to sociodemographic characteristics and clinical factors and
subjective quality of life. The International Journal of Social Psychiatry, 47,
67-77.

Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin,
107, 238-246.

Bernal, G., & Saez-Santiago, E. (2006). Culturally centered psychosocial
interviews. Journal of Community Psychology, 34, 121-132.

Betancourt, H., & Lopez, S. R. (1993). The study of culture, ethnicity, and race in
American psychology. American Psychologist, 48, 629-637.

Bhugra, D. (2006). Severe mental illness across cultures. Acta Psychiatr Scand,
113, 17-23.

Bleuler, E. (1950). Dementia praecox or the group of schizophrenia. New York,
NY: International Universities Press.

Bracke, P., Christiaens, W., Verhaeghe, M. (2008). Self-esteem, self-efficacy, and
the balance of peer support among persons with chronic mental health problems.
Journal of Applied Social Psychology, 38, 436-459.

Brekke, J. S. (1987). The model-guided method of monitoring program
implementation. Evaluation Review, 11, 281-299.

Brekke, J. S., Ansel, M., Long, J., Slade, E., & Weinsten, M. (1999). Intensity and
continuity of services and functional outcomes in the rehabilitation of persons
with schizophrenia. Psychiatric Services, 50, 248-256.

Brekke, J. S., & Barrio, C. (1997). Cross-ethnic symptoms differences in
schizophrenia: The influence of culture and minority status. Schizophrenia
Bulletin, 23, 305-316.

Brekke J. S., Hoe M., Long J., & Green, M.F. (2007). How neurocognition and
social cognition influence functional change during community-based
psychosocial rehabilitation for individuals with schizophrenia. Schizophrenia
Bulletin.

Brekke, J. S., Kay, D. D., Lee, K. L., & Green, M. F. (2005). Biosocial pathways to
functional outcome in schizophrenia. Schizophrenia Research, 80, 213-225.

Brekke, J. S., Kohrt, B., & Green, M. F. (2001). Neuropsychological functioning as
a moderator of the relationship between psychosocial functioning and the
subjective experience of self and life in schizophrenia. Schizophrenia Bulletin,
27, 697-708.
117

Brekke, J. S., Levine, S., Wolkon, G., Sobel, G., & Slade, B. (1993). Psychosocial
functioning and subjective experience in schizophrenia. Schizophrenia Bulletin,
19, 599-608.

Brekke, J. S., & Long, J. D. (2000). Community-based psychosocial rehabilitation
and prospective change in functional, clinical, and subjective experience
variables in schizophrenia. Schizophrenia Bulletin, 26, 667-680.

Brekke, J. S., Long, J. D., & Kay, D. D. (2002). The structure and invariance of a
model of social functioning in schizophrenia. The Journal of Nervous and
Mental Disease, 190, 63-72.

Brekke, J. S., Long, J. D., Nesbitt, N., & Sobel, E. (1997). The impact of service
characteristics on functional outcome from community support programs for
persons with schizophrenia: A growth curve analysis. Journal of Consulting
and Clinical Psychology, 65, 464-475.

Brekke, J. S., Nakagami, E., Kee, K. S., & Green, M. F. (2005). Cross-ethnic
differences in perception of emotion in schizophrenia. Schizophrenia Research,
77, 289-298.

Brekke, J. S., Raine, A., Ansel, M., Lencz, T., & Bird, L. (1997).
Neuropsychological and psychophysiological correlates of psychosocial
functioning in schizophrenia. Schizophrenia Bulletin, 23, 19-28.

Brekke, J.S., & Test, M.A. (1992). A model for measuring the implementation of
community support programs: Results from three sites. Community Mental
Health Journal, 28, 227-247.

Bryson, G., Bell, M., Kaplan, E., Greig, T., & Lysaker, P. (1998). Affect recognition
in deficit syndrome schizophrenia. Psychiatry Research, 77, 113-120.

Buchanan, J. (1995). Social support and schizophrenia: A review of the literature.
Archives of Psychiatric Nursing, 2, 68-76.

Buchanan, R.W. (2007). Persistent negative symptoms in schizophrenia: An
overview. Schizophrenia Bulletin, 33, 1013-1022.

Burke-Miller, J. K., Cook, J. A., Grey, D. D., Razzano, L. A., Blyler, C. R., Leff, H.
S., Gold, P. B., Goldberg, R. W., Mueser, K. T., Cook, W. L., Hoppe, S. K.
Stewart, M., Blankertz, L., Dudek, K., Taylor, A. L., Carey, M. A. (2006).
Demographic characteristics and employment among people with severe mental
illness in a multisite study. Community Mental Health Journal, 42, 143-159.

118

Bustillo, J. R., Lauriello, J., Horan, W. P., & Keith, S. J., (2001). The psychosocial
treatment of schizophrenia: An update. The American Journal of Psychiatry,
158, 163-175.

Byrne, B. M. (2001). Structural equation modeling with AMOS: Basic concepts,
applications, and programming. Mahwah, NJ: Lawrence Erlbaum Associates,
Inc.

Calsyn, R.J., & Winter, J.P. (2002). Social support, psychiatric symptoms, and
housing: A causal analysis. Journal of Community Psychology, 30, 247-259.

Carpenter, W. T. (1996). The treatment of negative symptoms: Pharmacological and
methodological issues. The British Journal of Psychiatry, 168, 17-22.

Carpenter, W. T. Jr. (2007). Schizophrenia: Disease, syndrome, or dimensions?
Family Process, 46, 199-206.

Carpentier, N. & White, D. (2002). Cohesion of the primary social network and
sustained service use before the first psychiatric hospitalization. The Journal of
Behavioral Health Services & Research, 29, 404-418.

Caron, J., Lecomte, Y., Stip, E., & Renaud, S. (2005). Predictors of quality of life in
schizophrenia. Community Mental Health Journal, 41, 399-417.

Church, T. A. (2000). Culture and personality: Toward an integrated cultural trait
psychology. Journal of Personality, 66, 651-703.

Clinton, M., Lunney, P., Edwards, H., Weir, D., & Barr, J. (1998). Perceived social
support and community adaptation in schizophrenia. Journal of Advanced
Nursing, 27, 955-965.

Cnann, R. A., Blankertz, L., Messinger, K. W., & Gardner, J. R. (1998).
Psychosocial rehabilitation: Toward a definition. Psychosocial Rehabilitation
Journal, 11, 61-77.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences. (2
nd
ed.).
Hillsdale, NJ: Erlbaum.

Cohen, A. S., & Docherty, N. M. (2004). Deficit versus negative syndrome in
schizophrenia: Prediction of attentional impairment. Schizophrenia Bulletin, 30,
827-835.

Cohen, A. S., Saperstein, A.M., Gold, J.M., Kirkpatrick, B., Carpenter Jr., W.T.,
Buchanan, R.W. (2007). Neuropsychology of the deficit syndrome: New data
and meta-analysis of findings to date. Schizophrenia Bulletin, 33, 1201-1212.

Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering
hypothesis. Psychological Bulletin, 98, 310-357.
119

Cook, J. A. (2006). Employment barriers for persons with psychiatric disabilities:
Update of a report for the President’s Commission. Psychiatric Services, 57,
1391-1405.

Cook, J.A., & Jonikas, J.A. (1996). Outcomes of psychiatric rehabilitation service
delivery. In D.M. Steinwachs, L.M. Flynn, G.S. Norquist & E.A. Skinner
(Eds.). New directions for mental health services (pp. 33-47). San Francisco,
CA: Jossey-Bass Publishers.

Cook, J.A., Pickett, S.A., Razzano, L., Fitzgibbon, G., Jonikas, J. A., & Cohler, J. J.
(1996). Rehabilitation services for person with schizophrenia. Psychiatric
Annals, 26, 97-104.

Cooper, L. A., Brown, C., Vu, H. T., Ford, D. E., & Powe, N. R. (2001). How
important is intrinsic spiritually in depression care? A comparison of White and
African-American primary care patients. Journal of General Internal Medicine,
16, 634-638,

Copeland, V. C. (2005). African Americans: Disparities in health care access and
utilization. Health & Social Work, 30, 265-270.

Corrigan, P. W., & Phelan, S. M. (2004). Social support and recovery in people with
serious mental illnesses. Community Mental Health Journal, 40, 513-523.

Crosse, C. (2003). A meaningful day: Integrating psychosocial rehabilitation into
community treatment of schizophrenia. The Medical Journal of Australia, 178,
76-78.

Davidson, L., O’Connell, M. J., Tondora, J., Lawless, M., & Evans, A. (2005).
Recovery in serious mental illness: A new wine or just a new bottle?
Professional Psychology: Research and Practice, 36, 480-487.

Deci, E.L., Koestner, R., & Ryan, R.M. (1999). A metal-analytic review of
experiments examining the effects of extrinsic rewards on intrinsic motivation.
Psychological Bulletin, 125, 627-668.

Deci, E. L., Vallerand, R.J., Pelletier, L.G., & Ryan, R.M. (1991). Motivation and
education: The self-determining perspective. Educational Psychologist, 26325-
346.

Dembo, M.H. (2000). Motivation and learning strategies for college success: A
self-management approach. Mahwah, NJ: Erlbaum.

120

Dembo, M.H., & Eaton, M.T. (2000). Self-regulation of academic learning in
middle-level schools. The Elementary School Journal, 100, 473-490.

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, 455-464.

Dietzen, L. L., & Bond, G. R. (1993). Relationship between case manager contact
and outcome for frequently hospitalized psychiatric consumers. Hospital and
Community Psychiatry, 44, 839-843.

Dilk, M. N., & Bond, G. R. (1996). Meta-analytic evaluation of skills training
research for individuals with severe mental illness. The International Journal of
Psychosocial Rehabilitation, 64, 1337-1346.

Division of Services and Intervention Research (DSIR) (n.d.). Retrieved December
1, 2005 from National Institute of Mental Health website:
http://www.nimh.nih.gov/dsir/dsir.cfm &
http://www.nimh.nih.gov/dsir/dsir.pdf.

Drake, R. E., Green, A. I., Mueser, K. T., & Goldman, H. H. (2003). The history of
community mental health treatment and rehabilitation for persons with severe
mental illness. Community Mental Health Journal, 39, 427-440.

Drake, R. E., McHugo, G. J., Xie, H., Fox, M., Packard, J., & Helmstetter. (2006).
Ten-year recovery outcomes for consumers with co-occurring schizophrenia and
substance use disorders. Schizophrenia Bulletin, 32, 464-473.

Erickson, D. H., Beiser, M., & Iacono, W. G. (1998). Social support predict 5-year
outcome in 1
st
-episode schizophrenia. Journal of Abnormal Psychology, 107,
681-685.

Evert, H., Harvey, C., Trauer, T., & Herrman, H. (2003). The relationship between
social networks and occupational and self-care functioning in people with
psychosis. Soc Psychiatry Psychiatric Epidemiology, 38, 180-188.

Fakhoury, W.K.H., Murray, A., Shepard, G., & Priebe, S. (2002). Research in
supported housing. Social Psychiatry Psychiatric Epidemiology, 37, 301-315.

Falicov, C. J. (2001). The cultural meanings of money. American Behaviorist
Scientist, 45, 313-328.

Fellin, P. (1993). Reformulation of the context of community based care. Journal of
Sociology and Social Welfare, 20, 57–67.

Fitzmaurice, G. M., Laird, N.M., & Ware, J.H. (2004). Applied longitudinal
analysis. Hoboken, NJ: John Wiley & Sons, Inc.
121

Flynn, R., & Aubry, T. D. (1999). Integration of persons with developmental or
psychiatric disabilities: Conceptualization and measurement. In R. J. Flynn & R.
Lewey (Eds.), A quarter century of normalization and social role valorization
(pp. 271–304). Ottawa, Canada: University of Ottawa Press.

Folsom, D. P., Gilmer, T., Barrio, C., Moore, D. J., Bucardo, J., Lindamer, L. A.,
Garcia, Piedad, Hawthorne, W., Hough, R., Patterson, T., & Veste, D. V.
(2007). A longitudinal study of the use of mental health services by persons with
serious mental illness: Do Spanish-speaking Latinos differ from English-
speaking Latinos and Caucasians? American Journal of Psychiatry, 167, 1173-
1180.

Galderisi, S., Maj, M., Mucci, A., Cassano, G. B., Invernizzi, G., Vita, A.,
Dell’Ossom K. M., Daneluzzo, E., & Pini, S. (2002). Historical,
psychopathological,neurological, and neuropsychological aspects of deficit
schizophrenia: A multicenter study. The American Journal of Psychiatry, 159,
983-990.

Garson, G. David (n.d.). "Title of Topic", from Statnotes: Topics in Multivariate
Analysis. Retrieved February 12, 2008 from
http://www2.chass.ncsu.edu/garson/pa765/statnote.htm.

Gates, L. B., Klein, S. W., Akabas, S. H., Myers, R., Schawager, M., & Kaelin-Kee,
J. (2005). Outcomes-based funding for vocational services and employment of
people with mental health conditions. Psychiatric Services, 56, 1429-1435.
Ghaemi, S. N. (2008). Why evidence-based medicine can, and must, be applied to
psychiatry. Psychiatric Times, 25, 1-6.

Glynn, S. M., Aultmann, L. P., & Owens, A. M. (2005). Motivation to learn in
general education programs. The Journal of General Education, 54, 150-170.

Goldberg, R. W., Rollins, A. L., & Lehman, A. F. (2003). Social network correlates
among people with psychiatric disabilities. Psychiatric Rehabilitation Journal,
26, 247-258.

Goodman, S. H., Sewell, D. R., Cooley, E. L., et al. (1993). Assessing levels of
adaptive functioning: the Role Functioning Scale. Community Mental Heallth
Journal, 29, 199-131.

Green, R. S., & Gracely, E. J. (1987). Selecting a scale for evaluating services to
the chronically mentally ill. Community Mental Health Journal, 23, 91-102.

Greenberg, G. A., & Rosenheck, R. A. (2005). Continuity of care and clinical
outcomes in a national health system. Psychiatric Services, 56, 427-433.

122

Guarnaccia, P. J., & Parra, P. (1996). Ethnicity, social status, and the families’
experience of caring for a seriously mentally ill family member. Community
Mental Health Journal, 32, 243-260.

Haertl, K. (2005). Factors influencing success in a Fairweather model mental health
program. Psychiatric Rehabilitation Journal, 28, 370-377.

Hansson, L., Middelboe. T., Sorgaard, K. W., Bengtsson-Tops, A., Bjarnason, O.,
Merinder, L., Nilsson, L., Sandlund, M., & Vinding, H. R (2002). Living
situation, subjective quality of life and social network among individuals with
schizophrenia living in community setting. Acta Psychiatr Scand, 106, 343-350.

Harkness, J., Newman, S.J., & Salkever, D. (2004). The cost-effectiveness of
independent housing for the chronically mentally ill: do housing and
neighborhood features matter? Health Services Research, 39, 1341-1360.

Heine, S. (2001). Self as cultural product: An examination of East Asian and North
American Selves. Journal of Personality, 69, 881-906.

Heinrichs, D. W., Hanlon, T. E., & Carpenter, W. T. Jr. (1984). The quality of life
scale: An instrument for rating schizophrenic deficit syndrome. Schizophrenia
Bulletin, 10, 388-398.

Hinton, L., Guo, Z., Hillygus, J., & Levkoff, S. (2000). Working with culture: A
qualitative analysis of barriers to the recruitment of Chinese-American family
caregivers for dementia research. Journal of Cross-Cultural Gerontology, 15,
119-137.

Hitchcock, J. H., Nastasi, B. K., Dai, D. Y., Newman, J., Jayasena, A., Bernstein-
Moore, R., et al. (2005). Illustrating a mix-method approach for validating
culturally specific constructs. Journal of School of Psychology, 43, 259-278.
Ho, B. C., Nopoulos, P., Flaum, M., Arndt, S., & Andreasen, N. C. (1998). Two-
year outcome in first-episode schizophrenia: Predictive value of symptoms for
quality of life. American Journal of Psychiatry, 155, 1196-1201.

Hobbs, F., & Stoops, N. (2002). Demographic trends in the 20
th
century. Census
2000 Special Reports. Retrieved November 17, 2005 from
http://www.census.gov/prod/2002pubs/censr-4.pdf.

Hoffmann, H., & Kupper, Z. (1997). Relationships between social competence,
psychopathology and work performance and their predictive value for
vocational rehabilitation of schizophrenia outpatients. Schizophrenia Research,
23, 69-79.

123

Hofstede, G., & McCrae, R.R. (2004). Personality and culture revisited: Linking
traits and dimensions of culture. Cross-Culture Research, 38, 52-88.

Hohmann, A. A., & Shear, M. K. (2002). Community-based intervention research:
Coping with the “Noise” of real life in study design. American Journal of
Psychiatry, 159, 201-207.

Honey, A. (2000). Psychiatric vocational rehabilitation: Where are the customers’
views? Psychiatric Rehabilitation Journal, 23, 270-279.

Hopkins, M., & Ramsundar, N. (2006). Which factors predict case management
services and how do these services relate to consumers outcomes? Psychiatric
Rehabilitation Journal, 29, 219-222.

Hopper, K. (2007). Recovery from schizophrenia an international perspective: A
report from the WHO collaborative project, the International Study of
Schizophrenia. Oxford: Oxford University Press.

Hopper, K., & Wanderling, J. (2000). Revisiting the developed versus developing
country distinction in course and outcome in schizophrenia: Results from ISoS,
the WHO collaboration follow-up project. Schizophrenia Bulletin, 26, 835-846.

Hoyle, R. H. (1995). The structural equation modeling approach: Basic concepts and
fundamental issues. In R.H. Hoyle (Ed.). Structural equation modeling (pp.1-
15). Thousand Oaks, CA: Sage Publication, Inc.

Hu, L-T., & Bentler, P. M. (1995). Evaluating model fit. In R.H. Hoyle (Ed.).
Structural equation modeling (pp.76-99). Thousand Oaks, CA: Sage
Publication, Inc.

Imada, T., & Schiavo, R. S. (2005). The use of ethnic minority populations in
published psychological research. The Journal of Psychology, 139, 389-400.

International Journal of Psychosocial Rehabilitation website. (2005). Psychosocial
rehabilitation-Assertive community treatment. Retrieved on November 16, 2007
at http://www.psychosocial.com/descript.html.

Iyer, S. N., Rothman, T. L., Vogler, J. E., & Spaulding, W. D. (2005). Evaluating
outcomes of rehabilitation for severe mental illness. Rehabilitation Psychology,
50, 43-55.

Jacobson, N., & Greenley, D. (2001). What is recovery? A conceptual model and
explication. Psychiatric Services, 52, 482-485.

124

Janson, S. L., Alioto, M. E., & Boushey, H.A. (2001). Attrition and retention of
ethnically diverse subjects in a multicenter randomized controlled research trial.
Controlled Clinical Trials, 22, 236S-243S.

Jenkins, J. H., & Schumacher, J. G. (1999). Family burden of schizophrenia and
depressive illness: Specifying the effects of ethnicity, gender, and social
ecology. British Journal of Psychiatry, 174, 31-38.

Joreskog, K.G., & Sorbom, D. (1981). LISREL V: Analysis of linear structural
relationships by maximum likelihood and least squares methods. Chicago:
National Educational Resources.

Kee, K., Green, M. F., Mintz, J., & Brekke, J. S. (2003). Is emotional processing a
predictor of functional outcome in schizophrenia? Schizophrenia Bulletin, 29,
487-498.

Kim, H. K., & McKenry, P. C. (1998). Social networks and support: A comparison
of African Americans, Asians Americans, Caucasians, and Hispanics. Journal
of Comparative Family Studies, 29, 313-334.

Kirkpatrick B, Fischer B. (2006). Subdomains within the negative symptoms of
schizophrenia: commentary. Schizophrenia Bulletin, 32, 246-249.

Kupfer, D. J., Frank, E. J., Grochoncinski, V. J., Houck, P. R., & Brown, C. (2005).
African-American participants in a bipolar disorder registry: Clinical and
treatment characteristics. Bipolar Disorders, 7, 82-88.

Kurtz MM. (2005). Symptoms versus neurocognitive skills as correlates of everyday
functioning in severe mental illness. Expert Review of Neurotherapeutics, 6, 47-
56.

Larson, J. E., Barr, L. K., Kuwabara, S. A., Boyle, M.G., & Glenn, T. L. (2007).
Process and outcome analysis of a supported employment program for people
with psychiatric disabilities. American Journal of Psychiatric Rehabilitation,
10, 339-353.

Latham, G.P., & Pinder, C.C. (2005). Work motivation theory and research at the
dawn of the twenty-first century. Annual Review Psychology, 56, 485-516.

Lauriello, J., Bustillo, J., & Keith, S. J. (1999). A critical review of research on
psychosocial treatment of schizophrenia. Society of Biological Psychiatry, 46,
1409-1417.
Leeuwenkamp O, Velligan DI, Wang M, et al. (2007) Association between changes
on the negative symptom assessment scale and measures of functional outcome
in schizophrenia. Schizophrenia Bulletin, 33, 594-595.

Lefley, H.P. (1990). Culture and chronic mental illness. Hospital and Community
Psychiatry, 41, 277-286.
125

Lewis-Fernandez, R., & Diaz, N. (2002). The cultural formulation: A method for
assessing cultural factors affecting the clinical encounter. Psychiatric
Quarterly, 73, 271-295.

Liberman, R. P., Kopelowicz, A., Ventura, J., & Gutkind, D. (2002). Operational
criteria and factors related to recovery from schizophrenia. International
Review of Psychiatry, 14, 256-272.

Lipton, F. R., Siegel, C., Hannigan, A., Samuels, J., & Baker, S. (2000). Tenure in
supportive housing for homeless persons with severe mental illness. Psychiatric
Services, 51, 479-486.

Lopez, S. R. (2002). Mental health care for Latinos: A research agenda to improve
the accessibility and quality of mental health care for Latino. Psychiatric
Services, 53, 1569-1573.

Lucca A. M., & Allen, G. J. (2001). A statewide assessment of psychosocial
rehabilitation programs: General characteristics and services. Psychiatric
Rehabilitation Journal, 24, 205-213.

Lukoff, D., Nuechterlein, K. H., & Ventura, J. (1986) Manual for expanded Brief
Psychiatric Rating Scale (BPRS). Schizophrenia Bulletin, 12, 594 602.

Lysaker, P. H., & Davis, L. W. (2004). Social function in schizophrenia and
schizaffective disorder: Associations with personality, symptoms and
neurocognition. Health and Quality of Life Outcomes, 2, 1-6.

MacCallum, R. C. (1995). Model specification: Procedures, strategies, and related
issues. In R.H. Hoyle (Ed.). Structural equation modeling (pp.16-36).
Thousand Oaks, CA: Sage Publication, Inc.

Malik, K., Reeves, R. J., & Dellario, D. J. (1998). Barriers to community
integration for people with severe and persistent psychiatric disabilities.
Psychiatric Rehabilitation Journal, 22, 175-180.

Mak, W. W. S. (2005). Integrated model of caregiving: How macro and micro
factors affect caregivers of adults with severe and persistent mental illness.
American Journal of Orthopsychiatry, 75, 40-53.

Mak W.W. S., Law, R. W., Alvidrez, J., & Perez Stable, E. J. (2007) Gender and
ethnic diversity in NIMH funded clinical trials: Review of a decade of published
research. Administration and Policy in Mental Health, 34, 497 503.

126

Marks, L. I. (1998). Deconstructing locus of control: Implications for practitioners.
Journal of Counseling and Development, 76, 251-260.

Marrone, J., & Golowka, E. (1999). If work makes people with mental illness sick,
what do unemployment, poverty and social isolation cause? Psychiatric
Rehabilitation Journal, 23, 187-193.

Marsella, A. J., & Yamada, A. M. (2000). Culture and mental health: An
introduction and overview of foundations, concepts, and issues. In I. Cuellar &
F. A. Paniagua (Eds.). Handbook of multicultural mental health (pp.3-24). San
Diego, CA: Academic Press, Inc.

Marsh, H.W., Balla, J., & McDonald, R.P. (1988). Goodness of fit in confirmatory
factor analysis: The effect of sample size. Psychological Bulletin, 103, 391-410.

McKibbin, C. L., Brekke, J. S., Sires, D., Jeste, D. V., & Patterson, T. L. (2004).
Direct assessment of functional abilities: relevance to persons with
schizophrenia. Schizophrenia Research, 72, 53-67.

Milev P, Ho B.C., Arndt, S., Andreasen, N.C. (2005). Predictive values of
neurocognition and negative symptoms on functional outcome in schizophrenia:
a longitudinal first-episode study with 7-year follow-up. American Journal of
Psychiatry, 162, 495-506.

Miranda, J., Nakamura, R., & Bernal, G. (2003). Including ethnic minorities in
mental health intervention research: A practial approach to a long-standing
problem. Culture, Medicine, and Psychiatry, 27, 467-486.

Mueser, K. T., Becker, D. R., Torrey, W. C., Xie, H., Bond, G. R., Drake, R. E., &
Dain, B. J. (1997). Work and nonvocational domains of functioning in persons
with severe mental illness: A longitudinal analysis. The Journal of Nervous and
Mental Disease, 185, 419-426.

Mueser, K. T., Clark, R. E., Haines, M., Bond, G., R., Essock, S. M., Becker, D., R.,
Wolfe R., & Swain, K. (2004). The Hartford study of supported employment
for persons with severe mental illness. Journal of Consulting and Clinical
Psychology, 72, 479-490.

Mueser, K. T., & McGurk, S. R. (2004). Schizophrenia. Lancet, 363, 2063-2072.

Mueser, K. T., Salyers, M. P., & Mueser, P. R. (2001). A prospective analysis of
work in schizophrenia. Schizophrenia Bulletin, 27, 281-296.

127

Muthen, L.K., & Muthen, B.O. (1998-2007). Mplus user’s guide. Fifth Edition.
Los Angeles, CA: Muthen & Muthen.

Nathan, J. H., Wylie, A. M., Marsella, A. J. (2001). Attribution and serious mental
illness: Understanding multiple perspectives and ethnocultural factors.
American Journal of Orthopsychiatry, 71, 350-357.

National Advisory Mental Health Council. (1999). Bridging Science and Service: A
Report by the National Advisory Mental Health Council's Clinical Treatment
and Services Research Workgroup. NIH Publication No. 99-4353, 1999.
Retrieved December 1, 2005 from National Institute of Mental Health:
http://www.nimh.nih.gov/publicat/nimhbridge.pdf.

Nelson, G., Aubry, T., & Lafrance, A. (2007). A review of the literature on the
effectiveness of housing and support, assertive community treatment, and
intensive case management interventions for persons with mental illness who
have been homeless. American Journal of Orthopsychiatry, 77, 3, 350-361.

Newman, S. J., Reschovsky, J. D., Kaneda, K., & Hendrick, A. M. (1994). The
effects of independent living on persons with chronic mental illness: An
assessment of the section 8 certificate program. The Milbank Quarterly, 72,
171-198.

NIMH Workgroup (2000). Translating Behavioral Science Into Action: Report of the
National Advisory Mental Health Council Behavioral Science Group (NIH
Publication No. 00-4699). Retrieved on December 1, 2005 from Bethesda,
MD: National Institutes of Health/National Institute of Mental Health
website: http://www.nimh.nih.gov/publicat/nimhtranslating.pdf

O’Hara, A. (2007). Housing for people with mental illness: Update of a Report to
the President’s New Freedom Commission. Psychiatric Services, 58, 907-913.

Oyserman, D., Coon, H. M, & Kemmelmeier, M. (2002). Rethinking individualism
and collectivism: Evaluation of theoretical assumptions and meta-analysis.
Psychological Bulletin, 128, 3-72.

Patel, M.X., Doku, V., Tennakoon, L. (2003). Challenges in recruitment of research
participants. Advances in Psychiatric Treatment, 9, 229-238.

Penn, D. L., & Mueser, K. T., (1996). Research update on the psychosocial
treatment of schizophrenia. The American Journal of Psychiatry, 153, 607-617.

Penn, D. L., Waldheter, E. J., Perkins, D. O., Mueser, K. T., & Lieberman, J. A.
(2005). Psychosocial treatment for first-episode psychosis: A research update.
The American Journal of Psychiatry, 162, 2220-2232.

128

Phillips, E. S, Barrio, C., & Brekke, J. S. (2001). The impact of ethnicity on
prospective functional outcomes from community based psychosocial
rehabilitation for persons with schizophrenia. Journal of Community
Psychology, 29, 657-673.

Phinney, J.S. (1996). When we talk about American ethnic groups, what do we
mean? American Psychologist, 51, 918-927.

Provencher, H.L., Gregg, R., Mead, S. & Mueser, K.T. (2002). The role of work in
the recovery of persons with psychiatric disabilities. Psychiatric Rehabilitation
Journal, 26, 132-144.

Raudenbush, S.W., & Bryk, A. S. (2002). Hierarchical linear models: Applications
and data analysis methods, 2
nd
Ed. Thousand Oaks, CA: Sage Publications, Inc.

Raudenbush, S. Bryk, A., Cheong, Y. F., Congdon, R., & duToit, M. (2004). HLM6:
Hierarchical linear and nonlinear modeling. Lincolnwood, IL: Scientific
Software International, Inc.

Reeves, T. J., & Bennett, C. E. (2004). The Asian and Pacific Islander Population
in the United States: March 2002, Current Population Reports, P20-540, U.S.
Census Bureau, Washington, DC.

Resnick, B. (1998). Functional performance of older adults in a long-term care
setting. Clinical Nursing Research, 7, 230-246.

Rollnick S., & Miller, W.R. (1995). What is motivational interviewing?
Behavioural and Cognitive Psychotherapy, 23, 325-334.

Rogers, E. S., Anthony, W. A., Cohen, M., & Davies, R. R. (1997). Prediction of
vocational outcome based on clinical and demographic indicators among
vocationally ready consumers. Community Mental Health Journal, 33, 99-112.

Rogers, E. S., Anthony, W. A., & Lyass, A. (2004). The nature and dimensions of
social support among individuals with severe mental illnesses. Community
Mental Health Journal, 40, 437-450.

Rosenblatt, A., & Attkisson, C. C. (1993). Assessing outcomes for sufferers of
severe mental disorder: A conceptual framework and review. Evaluation and
Program Planning, 16, 347-363.

Rosenheck, R., Stroup, S., Keefe, R. S. E., McEvoy, J., Swartz, M., Perkins, D.,
Hsiao, J., Shumway, M., & Lieberman, J. (2005). Measuring outcome
priorities and preferences in people with schizoprehenia. British Journal of
Psychiatry, 187, 529-536.

129

Salkever, D. S., Karakus, M. C., Slade, E. P., Harding, C. M., Hough, R. L.,
Rosenheck, R. A., Swartz, M. S., Barrio, C., Yamada, A. M. (2007). Measures
and predictors of community-based employment and earnings of persons with
schizophrenia in a multisite study. Psychiatric Services, 58, 315-324.

Schwarzer, R., Knoll, N., & Rieckmann, N. (2003). Social support. In A. Kaptein,
& J. Weinman (Eds.). Introduction to health psychology. Oxford, England:
Blackwell.

Shavers, V. L., Lynch, C. F., & Burmeister, L.F. (2002). Racial differences in
factors that influence the willingness to participate in medical research studies.
Annual Epidemiology, 12, 248-256.

Sherbourne, C. D., & Stewart, A. L. (1991). The MOS social support survey. Social
Science and Medicine, 32, 705-714.

Siegel, C. E., Samuels, J., Tang, D-I, Berg, I., Jones, K., & Hopper, K. (2006).
Tenant outcomes in supported housing and community residences in New York
City. Psychiatric Services, 57, 982-991.

Singelis, T. M. (1994). The measurement of independent and interdependent self-
construals. Personality and Social Psychology Bulletin, 20, 580-591.

Singer, J.D., & Willett, J.B. (2003). Applied longitudinal data analysis: Modeling
change and event occurrence. New York, NY: Oxford University Press, Inc.

Silverstein, S. M., & Wilkniss, S. M. (2004). At issue: The future of cognitive
rehabilitation of schizophrenia. Schizophrenia Bulletin, 30, 679-692.

Snowden, L.R. (2001). Barriers to effective mental health services for African
Americans. Mental Health Services Research, 3, 181-187.

Snowden, L. R. (2003). Bias in mental health assessment and intervention: Theory
and evidence. American Journal of Public Health, 93, 239-243.

Snowden, L. R. (2007). Explaining mental health treatment disparities: ethnic and
cultural differences in family involvement. Cultural, Medicine, and Psychiatry,
31, 389-402.

Snowden, L. R., & Yamada, A. M. (2005). Cultural differences in access to care.
Annual Review of Clinical Psychology, 1, 143-166.

Solomon, P. (1992). The efficacy of case management services for severely
mentally disabled consumers. Community Mental Health Journal, 28, 163-180.

130

Spaniol, L., & Koehler, M. (1994). The experience of recovery. Boston: Center for
Psychiatric Rehabilitation.

Steiger, J.H., & Lind, J.C. (1980). Statistically-based tests for the number of
common factors. Paper presented at the annual Spring Meeting of the
Psychometric Society in Iowa City. May 30, 1980.

Steinberg, L., & Avenevoli, S. (2000). The role of context in the development of
psychopathology: A conceptual framework and some speculative propositions.
Child Development, 71, 66-74.

Strauss, J. S., & Carpenter, W. T. (1972). The prediction of outcome in
schizophrenia: I. Characteristics of outcome. Archives of General Psychiatry,
27, 739-746.

Sue, S., & Chu, J. Y. (2003). The mental health of ethnic minority groups:
challenges posed by the supplement to the Surgeon General’s report on mental
health. Culture, Medicine and Psychiatry, 27, 447-465.

Super, C. M., & Harkness, S. (1999). The environment as culture in development
research. In S. L. Friedman, & T. D. Wachs (Eds.), Measuring environment
across life span: Emerging methods and concepts (pp. 279-326). Washington,
DC: American Psychological Association.

Tanzman, B. (1993). An overview of surveys of mental health consumers’
preferences for housing and support services. Hospital & Community
Psychiatry, 44, 450-455.

Takeuchi, D.T., & Kim, K.F. (2000). Enhancing mental health services delivery for
diverse populations. Contemporary Sociology, 29, 74-83.

Telles, C., Karno, M., Mintz, J., Paz, G., et al., (1995). Immigrant families coping
with schizophrenia: Behavioral family intervention v. case management with a
low-income Spanish-speaking population. British Journal of Psychiatry, 167,
473-479.

Triandis, H.C. (1994). Culture and social behavior. New York: McGraw-Hill.

Tsang, H., Lam, P., Ng, B., & Leung, 0. (2000). Predictors of employment
outcomes for people with psychiatric disabilities: A review of the literature
since the mid ‘80s. Journal of Rehabilitation, 66, 19-25.

131

Tu, S.P., Chen, H.F, Chen, A., Lim, J., May, S., & Drescher, C. (2005). Clinical
Trials: Understanding and perceptions of female Chinese-American cancer
patients. Cancer Supplement, 104, 2999-3005.

Tucker, L.R., & Lewis, C. (1973). A reliability coefficient for maximum likelihood
factor analysis. Psychometrika, 38, 1-10.

U.S. Department of Health and Human Services. (2001). Mental Health: Race,
Culture, and Ethnicity: A Supplement to Mental Health: A Report of the
Surgeon General. Rockville, MD: U.S. Department of Health and Human
Services, Substance Abuse and Mental Health Services Administration, Center
for Mental Health Services, National Institutes of Health, National Institute of
Mental Health.

Vega, W.A., & Lopez, S.R. (2001). Priority issues in Latino mental health services
research. Mental Health Services Research, 3,189-200

Velligan, D. I., & Alphs, L. D. (2008). Negative symptoms in schizophrenia: The
importance of identification and treatment. Psychiatric Times, 125, 1-7.

Waghorn, G., Chant, D., & King, R. (2007). Classifying socially-valued role
functioning among community residents with psychiatric disorders. American
Journal of Psychiatric Rehabilitation, 10, 185-221.

Walsh, J., & Connelly, P. R. (1996). Supportive behaviors in natural support
networks of people with serious mental illness. Health and Social Work, 21,
296-303.

Wang, P.S., Lane, M., Olfson, M., Pincus, H.A., Wells, K.B., & Kessler, R.C.
(2005). Twelve-month use of mental health services in the United States:
Results from the National Comorbidity Survey Replication. Archive of General
Psychiatry, 62, 629-640.

Weisman, A., Duarte, E., Koneru, V., & Wasserman, S. (2006). The development
of a culturally informed, family focused treatment of schizophrenia. Family
Process, 45, 171-186.

Weisman, A., Rosales, G., Kymalainen, J., & Armesto, J. (2005). Ethnicity, family
cohesion, religiosity, and general emotional distress in patients with
schizophrenia and their relatives. Journal of Nervous and Mental Disease, 193,
359-368.

Wimmera PDSS website. Retrieved March 3, 2008, from
http://www.wimmerapdss.org.au/psr.htm.

132

Wong, Y.L.I, & Solomon, P. L. (2002). Community integration of persons with
psychiatric disabilities in supportive independent housing: A conceptual model
and methodological considerations. Mental Health Services Research, 4, 13-28.

Wright, B.E. (2001). Public-sector work motivation: A review of the current
literature and a revised conceptual model. Journal of Public Administration
Research and Theory, 11, 559-586.

Wright, G. N. (1980). Total rehabilitation (pp.733-789). Boston, MA: Little Brown
and Company. .

Yamada, A-M & Brekke, J. S. (in press). Addressing cultural disparities through
clinical competence rather than cultural competence: The need for assessment of
sociocultural issues in the delivery of psychosocial rehabilitation services.
Clinical Psychology Review.

Yamada, A.M., Dinh, T.Q., & Brekke, J.S. (2006). Impact of Sociocultural Factors
on Ethnically Diverse Consumers within Community-based Psychosocial
Rehabilitation Services. Unpublished manuscript, University of Southern
California, Los Angeles, CA.

Yamada, M. M., Korman, M., & Hughes, C. W. (2000). Predicting rehospitalization
of persons with severe mental illness. Journal of Rehabilitation, 66, 32-39.

Yang, J., Law, S., Chow, W., Andermann, L., Steinberg, R., & Sadavoy, J. (2005).
Assertive community treatment for persons with severe and persistent mental
illness in ethnic minority groups. Psychiatric Services, 56, 1053-1055.

Yanos, P. T., Rosenfield, S., & Horwitz, A. (2001). Negative and supportive social
interactions and quality of life among people diagnosed with severe mental
illness. Community Mental Health Journal, 37, 405-419.

Young, A.S., Sullivan, G., Murata, D., Sturm, D., & Koegel, P. (1998).
Implementing publicly funded risk contracts with community mental health
organizations. Psychiatric Services, 49, 1579-1584.
133

APPENDIX

Table 15. Indices for Alternative Work Productivity Functioning Models

Models X2 df p CFI TLI SRMR
RMSEA
(Confidence
Interval)
Linear Model
1
5.54 1 0.02 0.95 0.85 0.06
0.18
(0.06-0.34)
Nonlinear Model
2
0.00 0 0.00 1.00 1.00 0.00
0.00
(0.00-0.00)
Nonlinear with
equality constraint 0.74 2 0.69 1.00 1.02 0.02
0.00
(0.00-0.13)
Notes: 1. The residual covariance matrix (theta) is not positive definite. Problem involving
variable Work3. 2. The residual covariance matrix (theta) is not positive definite.
Problem involving variable Work

Abstract (if available)

Abstract Despite the promising effects of psychosocial rehabilitation (PSR) interventions on outcomes for individuals with severe and persistent mental illness (SPMI), wide variation still exists in individual functional outcomes. This is particularly true for ethnic minorities as empirical evidence has documented persistent racial disparities in mental health outcomes. In this dissertation, the author examines sociocultural factors -- support systems and minority status -- that influence rehabilitative independent living and employment functional outcomes for individuals with SPMI, using secondary data from a longitudinal study conducted at four community-based psychosocial rehabilitation programs in Southern California.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Longitudinal relationships of cognitive deficits, symptoms, and social functioning outcomes in community-based psychosocial rehabilitation programs: mechanisms of longitudinal change

PDF

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

PDF

Community integration of individuals with serious mental illness: a network perspective from India and United States

PDF

Mindfulness-based self-regulation for psychotic disorders: a feasibility study

PDF

Working alliance, hope, and functional outcome for individuals with schizophrenia: mechanisms of influence

PDF

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

PDF

The impact of family factors on the functioning of African-American consumers living with schizophrenia

PDF

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

PDF

Closing the health gap

PDF

Asian Americans served at community mental health agencies for a diagnosis of schizophrenia

PDF

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

PDF

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

PDF

The role of protective factors on outcomes for Latinos with schizophrenia

PDF

The Arukah Project: collaborating with the Church to improve African American mental health

PDF

Acculturation team-based clinical program: pilot program to address acculturative stress and mental health in the Latino community

PDF

The power of social media narratives in raising mental health awareness for anti-stigma campaigns

PDF

The formation and influence of online health social networks on social support, self-tracking behavior and weight loss outcomes

PDF

Building a trauma-informed community to address adverse childhood experiences

PDF

Maternal mental health matters: Creating FQHC+ to meet the complex needs of perinatal women of color

PDF

The effects of time and space on health status of older adults in China

Asset Metadata

Creator Dinh, Tam Quy Thi (author)

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

School School of Social Work

Degree Doctor of Philosophy

Degree Program Social Work

Publication Date 07/31/2008

Defense Date 05/29/2008

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

Tag community-based services,OAI-PMH Harvest,psychosocial rehabilitation,severe mental health,sociocultural factors

Place Name California (states)

Language English

Advisor Yamada, Ann Marie (committee chair), Brekke, John (committee member), Dembo, Myron H. (committee member)

Creator Email tamdinh@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1476

Unique identifier UC1441476

Identifier etd-Dinh-20080731 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-98104 (legacy record id),usctheses-m1476 (legacy record id)

Legacy Identifier etd-Dinh-20080731.pdf

Dmrecord 98104

Document Type Dissertation

Rights Dinh, Tam Quy Thi

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