 

 

v
0‘ 4‘

.231! .6 .

f Ere... I...“ v.

. J"... «an ..

 

‘ 2
3 i 2.5
..; 1. ...: a. 4
1: .5.

11 V
3.1.56.5”
,iivce; . .
(sic. «X.
I»... v... . K
I

 

. .44
~ .. 1.1.5.! . :: ..¢)j-\.rul P7; ‘11.. ‘ ‘ . 1:. I
r (a {pizza 1111 14.. , a” Jeni ‘ u

i. ‘4. :1

 

N]

K»

m
Q3

LIBRARY
Michigan State
University

This is to certify that the
dissertation entitled

THE RELATIONSHIP BETWEEN SOCIAL NETWORK
SUPPORTS AND RECOVERY FROM MENTAL ILLNESS

presented by
FRANCESCA MARIA PERNICE-DUCA

has been accepted towards fulfillment
of the requirements for the

Doctoral degree in Family and Child Ecolggy

Major Professor’s Signﬂre
V’- 1 l3 / 2005
I

Date

MSU is an Afﬁrmative Action/Equal Opportunity Institution

 

a--uon-o-I-I-o-o-c-c-u-o-.-u—

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE
FEB 2 7 2003

 

 

{IQQQOQ

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 p:IClRC/Date0ue.indd—p.1

THE RELATIONSHIP BETWEEN SOCIAL NETWORK SUPPORTS AND
RECOVERY FROM MENTAL ILLNESS

By

Francesca Maria Pemice-Duca

A DISSERTATION
Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of
DOCTER OF PHILOSOPHY
Department of Family and Child Ecology

2005

ABSTRACT
THE RELATIONSHIP BETWEEN SOCIAL NETWORK SUPPORTS AND
RECOVERY FROM MENTAL ILLNESS

By

Francesca Maria Pemice-Duca

Emerging evidence indicates that social network supports may act as facilitators
to the recovery process from mental illness (Corrigan, Giffort, Rashid, Leary & Okeke,
1999; Corn'gan & Phelan, 2004). Although much has been studied in the area of social
support and mental health, the literature on the social networks of individuals and their
relationship to the recovery process requires further examination. Marriage and Family
Therapists (MFTs) can make a contribution to the understanding of social networks and
recovery through the application of Family Systems Theory being applied to work with
individuals living with chronic mental illness and their families. The Feminists’ re-
conceptualization (Knudson-Martin, 1996) of Bowen Family Systems Theory (Bowen,
1978) was used to understand the role of social and family networks to the recovery
process. A cross-section of a larger, longitudinal study was employed to analyze social
network proﬁles and a measure of recovery. A structured social network methodology
(Herman, 1997) was used to study the social networks of 221 individuals living with a
chronic psychiatric disability participating in psychosocial rehabilitation programs called
Clubhouses. A series of regression models were proposed to examine the relationship
between social and family networks supports and the process of recovery as measured by

the Recovery Assessment Scale (Corrigan, Giffort, Rashid, et al., 1999). A path analysis

was also used to explore whether clubhouse participation and cohesion was associated
with social support network dimensions and recovery as an outcome.

Five main themes emerged from the results. First, families dominate social
network support proﬁles among individuals living with chronic mental illness. Second,
the size of network supports is relatively small, consisting of an average 5 members.
Third, club members were least likely to be nominated as sources of support, while the
number of clubhouse staff supports was associated with greater recovery. Fourth,
engaging in reciprocally supportive relationships with social network supports was the
single most important predictor of recovery. And ﬁnally, being a regular participant at the
clubhouse is associated with greater afﬁliations and feelings of mutuality with club peers.
Club members who shared a greater sense of clubhouse community and cohesion with
club peers were also more likely to nominate more social network members, perceive
greater positive appraisals of social support, as well as perceive themselves as sources of
support to others. The path analysis failed to support the overall model predicting the
path between clubhouse participation, sense of community, social networks supports, and
recovery. However, this study did ﬁnd support for a prior empirical investigation
showing a signiﬁcant relationship between social support networks and recovery.
Furthermore, the measure of reciprocity with social support network members appears to

be signiﬁcantly related with the recovery process.

Copyright by
FRANCESCA M. PERNICE-DUCA
2005

DEDICATION

To my major professor, Dr. Esther Onaga for friendship, encouragement, and support.
Your work gives voice to those who are seldom heard, and inspiration to those who seek
hope. And to my best friend, and husband, Bradford Duca, and my two children, Dante

and Julien for being by my side through this ‘seemingly’ endless journey.

ACKNOWLEDGMENTS

I would ﬁrst like to acknowledge, my major professor, Dr. Esther Onaga, for the
support to complete this body of work. Without her guidance and vision, this work would
have not been possible. Dr. Onaga has been an overwhelmingly caring and responsive
mentor throughout my Michigan State graduate education, providing me with rich
educational and human experiences. I am blessed to have known and worked with such
an amazing woman; thank you for taking me as your student and sharing your insight,
wisdom, and knowledge.

I would secondly like to express my sincere appreciation to Dr. Sandra Herman,
the Flinn Project co-principle investigator and dissertation co-director. This study would
not have been possible without her partnership with Michigan State University. I would
like to thank her for dedicating the time to be an important part of the dissertation
committee and supporting me with her statistical and methodological insight. She has
been a wonderful mentor and friend, and I am honored to also be recognized as her
student.

Secondly, I would like to thank Dr. Marsha Carolan, the Marriage and Family
Therapy program clinical director and dissertation committee member, for her guidance
and support throughout my tenure in the clinical program. She exempliﬁes the values of
our MFT profession by upholding the professionalism and standards of our ﬁeld while
nurturing a more inclusive lens in which to understand relationships.

I would also like to express my gratitude to my committee members, Dr. Lilian

Phenice, Dr. Vincent Hoffman, and Dr. Ruben Parra Cardona. I was honored by their

vi

commitment and ﬂexibility in meeting my needs during this process. Dr. Cardona took
extra time to give me guidance and support toward my professional development.

I extend a special ‘thank you’ to the my Fee Family colleagues and research team
in West Fee Hall. I’d like to recognize the original Flinn research team (Sandy Herman,
Esther Onaga, Helen Irvine, Cathy Maddelana, SuMin Oh, Katie Weaver-Randall,and
Chandra O’Donnell) for all the work and time invested in collecting and entering data. I
would also like to acknowledge all the wonderful Clubhouses who participated in the
Flinn Project and that made this work possible. I would like to extend a special thank
you to those who have joined Fee Hall and have provided valuable assistance in projects
and tasks, especially Toko Oshio, who gave me a quick and fun lesson in SPSS AMOS.

I must express my deepest and warmest gratitude to all of those in my own
personal family and social network who have made this journey in my life less bumpy.
Thank you to my parents, Salvatore and Maria Pemice, and my mother and father in-law,
Frank and Denise Duca for loving and caring for my children when I needed it most. My
sisters, Mariangela and Liliana Pemice gave of their time to see me ﬁnish through the
end. I would like to thank my supportive sisters Sabrina Badalamenti and Fara Hitchcock
and their husbands, and my friend, Alissa Lincoln for the “count down”. I would like to
extend a special thanks to my good friends and MFT colleagues Dahlia Rosen and
Natasha Kendall, for reading, advising, editing, and dealing with me throughout this
process. They will forever be clear family friends. I would also like to extend recognition
to my generous friend, Fumi Dobrowitsky, for her kind ear and the endless supply of

goodies during the mad rush to ﬁnish this work.

vii

And ﬁnally, a very special recognition goes to my sons, Dante Salvatore (2 yrs.)
and Julien Bradford (7 mos.), who inspired me to continue to the end. Thank you for

making our lives happier and ﬁlling it with laughter.

viii

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... xi
LIST OF FIGURES ......................................................................................................... xii
CHAPTER 1
Introduction ....................................................................................................................... 1
Purpose .............................................................................................................................. 7
Theoretical Rationale ............................................................................................. 8
Ecological Rationale .................................................................................. 8
Clubhouse Program .................................................................................... 9
Family System Theory .............................................................................. l 1
Deﬁnition of Terms ................................................................................... l6
Assumptions & Limitations .................................................................................. 18
Overview ............................................................................................................... 21
CHAPTER 2
Review of the Literature .................................................................................................. 22
Mental Illness as a Problem ................................................................................. 22
Social Network Research ..................................................................................... 24
Deﬁnition .................................................................................................. 24
Measurement of Social Networks ............................................................ 24
Social Networks & Mental Illness ........................................................... 25
Family Networks ...................................................................................... 27
Recovery from Mental Illness .............................................................................. 29
Deﬁning Recovery ................................................................................... 30
Social Networks & Recovery ................................................................... 31
Social Networks, Recovery, & Clubhouses ............................................. 37
Summary .............................................................................................................. 39
CHAPTER 3
Method ............................................................................................................................. 42
Description of Flinn Clubhouse Project ............................................................... 42
Research Questions .............................................................................................. 43
Conceptual Models .............................................................................................. 45
Sample ................................................................................................................. 50
Procedures ............................................................................................................ 51
Measures .............................................................................................................. 53
Demographic Questions ........................................................................... 54
Clubhouse Participation ........................................................................... 57
Social Network Interview ........................................................................ 57
Recovery Assessment Scale ..................................................................... 58
Sense of Community/Cohesion Scale ...................................................... 60

ix

Data Analysis ....................................................................................................... 61

CHAPTER 4
Results .............................................................................................................................. 62
Sample Characteristics ......................................................................................... 62
Descriptive Statistics of Social Networks ............................................................ 67
Social Network Dimensions & Club Member Characteristic ............................... 71
Social Networks & Clubhouse Engagement ........................................................ 76
Recovery .............................................................................................................. 77
Intercorrelations among Path Study Variables ........................................ 89
Direct & Indirect Pathways to Recovery ............................................................. 91
CHAPTER 5
Discussion ........................................................................................................................ 95
Social Networks & Recovery .............................................................................. 96
Social Network Support Dimensions & Recovery .............................................. 98
Social Networks & Clubhouses ........................................................................... 99
Path Results .......................................................................................................... 99
Limitations ...................................................................................................................... 101
Clinical Implications ....................................................................................................... 104
Conclusion ...................................................................................................................... 108
Future Research .............................................................................................................. 109
APPENDICES ................................................................................................................ 1 10
Appendix A: Letter to Clubhouse ....................................................................... 112
Appendix B: Participant Consent ........................................................................ 113
Appendix C: Guardian Consent ........................................................................... 116
Appendix D: Human Subjects Approval Letter .................................................. 119
Appendix E: Demographic Questions .................................................................. 121
Appendix F: Release for Diagnostic Information ............................................... 132
Appendix G: Clubhouse Participation ................................................................ 134
Appendix H: Social Network Interview ............................................................. 136
Appendix 1: Recovery Assessment Scale ........................................................... 139
Appendix J: Clubhouse Sense of Community Scale .......................................... 143
REFERNCES .................................................................................................................. 147

LIST OF TABLES

1. Measures Included in the Current Study ........................................................... 54
2. Statewide demographic data on clubhouse members from the Michigan

Department of Community Funding Year 2000 compared to Flinn Project

Participants project year 2001 ........................................................................... 56
3. Participant Demographics ................................................................................. 65
4. Composition of Social Network ........................................................................ 69
5. Means & Standard Deviations for Social Network Dimensions ....................... 71
6. Descriptive Statistics for the Recovery Assessment Scale & Recovery

and Recovery Factors ......................................................................................... 77
7. Correlation of Members Characteristics, Clubhouse Engagement,

Social Network Supports by Recovery Dimensions ......................................... 79
8. Correlations among Social Network Support Clusters & Recovery

Dimensions ...................................................................................................... 8O
9. Summary of Social Network Models Predicting Recovery Assessment

Scale Total Score & Factors .............................................................................. 84
10. Relationship between Family Network Support Dimensions & Recovery

Domains ............................................................................................................. 87
11. Zero-Order Correlations of Demographic Variables, Predictor Variables

in Regression & Path Models ............................................................................ 90

xi

PPPT‘

LIST OF FIGURES

Model A: Total network support predicting recovery ....................................... 46
Model B: Network members ............................................................................. 47
Model C: Family network predicting recovery domains ................................... 48
Initial path-analytic model: Inﬂuence of clubhouse engagement on

social networks & changes in recovery ............................................................. 49
Path-analytic model: Inﬂuence of clubhouse engagement on social

networks to recovery .......................................................................................... 94

xii

CHAPTER 1

INTRODUCTION

Mental illness can have devastating effects on an individual’s family and social
relationships. Individuals with chronic or persistent mental illness can experience the
loss of support from friends, family or partners, resulting in small or restricted social
support resources. Small social support networks have been associated with mental health
concerns such as isolation (Brewer, Gadsden & Scimshaw, 1994), and increased
likelihood of depression (Lin, Ye & Ensel, 1999). Poor or inadequate social support
networks have also been associated with increased mortality rates among the general
population (Berkman, 1995; Berkman, Glass, Brissette & Seeman, 2000; House, Landis
& Umberson, 1988). One of the earliest research studies on social networks and mental
health began with Emile Durkheim’s (1897) empirical examination of the effects of the
lack of social network ties and community integration and the rate of suicide in
metropolitan areas. Between 1969 and 1985, the interest in social network and mental
health research proliferated with over 1,300 published research articles (Biegel,
McCardel & Mendelson, 1985).

Social support networks among people living with severe or chronic mental
illness such as schizophrenia, are typically small, and predominately consist of family
members or mental health professionals (Davidson, Hoge, Merrill, Rakfeldt & Grifﬁth,
1996; Goldberg, Rollins & Lehman, 2003; Hardiman & Segal, 2003; Perese, Getty &

Wooldridge, 2003). Research has shown that small or restricted social networks threaten

psychological and emotional well-being (Green, Hayes, Dickinson, Whittaker &
Gilheany, 2002; Pickens, 2003), quality of life (Tempier, Caron, Mercier & Leouffre,
1998), and increase the likelihood of psychiatric re-hospitalization (Goldberg, Rollins &
Lehman, 2003). Cut-off or estranged family relationships have also been correlated with
increased psychological distress and functional impairments (Doane, 1991; Fisk, Rowe,
Laub, Calvocoressi & DeMino, 2000 ; Froland, Brodsky, Olson & Steward, 2000).
Individuals living with chronic and persistent mental illness experience functional
impairments in daily living skills and social skills. These impairments can negatively
impact social opportunities. Traditional medical model approaches continue to view
these negative consequences of serious mental illness as inevitable, which can result in a
loss of hope, despair, and chronic grief. The notion of recovery from mental illness has
received increasing attention in the mental health ﬁeld in the last decade. Emerging
evidence indicates that social network supports play a signiﬁcant role in the experiences
of recovery from mental illness (Corrigan & Phelan, 2004). State and Federal Mental
health organizations are beginning to recommend recovery oriented practices in the
treatment of mental illness, emphasizing the importance of social ties as an integral part
of the recovery process (Hogan, 2003). Longitudinal studies spanned across the last 30
years have documented recovery from serious mental illness, such as schizophrenia
(DeSisto, Harding, McCormick, Ashikaga & Brooks, 1999; Harding, Brooks, Ashikaga,
Strauss, & Breier, 1987a; 1987b). These longitudinal studies challenge traditionally held
beliefs about chronic mental illness, and provide support for programs that increase social

and vocational opportunities.

Recovery has been studied as a subjective experience through qualitative studies
(Deegan, 1988; 2003), as well as an objective outcome measuring level of functioning
and the absence of symptoms (Harding, 1986). Subjective accounts have described
recovery from mental illness as “reawakening of hope from despair; breaking through
denial and achieving understanding and acceptance; moving from withdrawal to
engagement and becoming an active participant in life; it is active coping rather than
passive adjustment (Beale & Lambric, 1995, p. 5). Recovery oriented philosophy in
mental health has revolutionized service delivery options, including more peer support
programs and psychosocial psychiatric rehabilitation.

The advent of psychiatric deinstitutionalization in Michigan raised a number of
concerns among professionals and family members about transitioning former hospital
patients into community living arrangements. In 1960, nearly 20,000 Michigan residents
were living in psychiatric hospitals. From 1960 to 1998, the numbers drastically
declined, with 1,244 people living in state hospitals as of 1998 (Michigan Department of
Community Health; Division of Vital Records and Health Statistics, 2000). Individuals
who had resided in psychiatric hospitals were now living in a variety of community
residential settings ranging from lodge programs to adult foster care homes. Some have
noted that a major drawback of deinstitutionalization was the concerns for many
individuals to remain at risk for social isolation (Davidson, Hoge, Godleski, Rakfeldt &
Grifﬁth & Merrill, 1996). Psychiatric rehabilitation programs, such as Clubhouses,
became instrumental in transitioning individuals from the hospital to the community

setting (Mastboom, 1992). Successful community living has been associated with

frequent family contact, participation in social and recreational activities, and increased
social ties (Dickerson, Ringel & Parente, 1999).

Social ties and relationships translate into personal social networks of support
composed of family, friends, and others in the community. Social networks act as life
lines for individuals with mental illness, providing opportunities for socialization,
companionship, and support. Berkman, Glass, Brissete & Seeman (2000) theorize a
direct relationship between social networks and measures of psychological health, such as
depression, coping effectiveness, self-efﬁcacy, and overall well-being. In essence, social
networks “deﬁne and reinforce meaningful social roles... which in turn provide a sense
of value, belonging, and attachment” (Berkman et al., p. 849). The authors further
contend that

...measures of social integration or connectedness have been such powerful

predictors of mortality [because] these ties gives meaning to an individual’s life

by virtue of enabling him or her to participate in it fully, to be obligated (in fact

often to be a provider of support) and to feel attached to one’s community (p.

849).

The clubhouse program, which is based on psychosocial psychiatric rehabilitation
principles rather than a medical model of treatment, values social relationships and social
participation as an active agent of rehabilitation and recovery (Mastboom, 1992). It
promotes a non-pathological orientation in working with individuals, which is consistent
with Marriage and Family Therapy (MFI‘) values. Clubhouses are viewed as a place for

people with chronic and persistent psychiatric disabilities where they feel they can

belong, learn new skills, and socialize. Clubhouses have been designed to increase social
connections for individuals with little family or social network ties (Beard, 1992b).
Further, they have also been cited as catalysts to recovery in the narratives of clubhouse
members (Beard, 1992b; Ely, 1992; Deegan, 1988; Paul, 1992; Peckoff, 1992).

MFI theories such as Narrative, Structural, Experiential, Family
Psychoeducation, and Bowen Family Systems Theory, can be viewed from the
perspective of the recovery philosophy. Each of these approaches reaches beyond the
medical model to identify strengths and non-pathological solutions within an ecological
context to the challenges of mental illness. Narrative theory offers hope through re-
storying and reclaiming one’s identity as a person, not an illness (White & Epston, 1990).
Experiential approaches provide ‘here and now’ experiences challenging set patterns of
behaviors and attitudes (Whitaker & Keith, 1981). Salvador Minuchin introduced the
importance of social networks in family therapy to the treatment of chronic mental
illness, (Elizur & Minuchin, 1989) where an extended family member may become the
most important person in the social network, and psychosocial programs are deemed
necessary for recovery (Kaffman, 1989). The analysis of family subsystems can shed
light on the impact of the psychiatric disability across the family life span, family roles
and responsibilities, and shared experiences. Family Psychoeducation has proven crucial
in the treatment of major bi-polar diagnoses and schizophrenia (McFarlane, Dixon,
Lucksted, 2003). Tenets of Bowen Theory view healing from chronic mental illness as a
“self-regenerative phenomena [deﬁned as] not only being self responsible, but also self-

actualizing” (Bowen, 1967, cited in Friedman, 1991, p.159). The Feminist elaboration of

Bowen Family Systems Theory underscores the importance of development of self in the
context of social relationships (Knundson-Martin, 1996).

MFI‘ training, values, and therapeutic relational orientations can speciﬁcally
address systemic issues to facilitate the recovery process. This study bridges the
importance of social network supports to the process of recovery. For persons with
chronic mental illness, recovery may be characterized as a journey from alienation to
partnerships with family and peers. Recovery can be viewed as a proxy to differentiation
of self in the context of family and social relationships.

Mental illness is a signiﬁcant event in the life course of multigenerational
families. A salient relational feature of people living with chronic mental illness, such as
schizophrenia, is the dependency on family members for support over a life time, which
can pose a strain on family relations. Networks that are characterized as small, kin-
dominated, and overly dependent increases the likelihood of emotional reactivity (Leff,
1976), increased burden of care, and less satisfying relational contacts. High levels of
Expressed Emotion (EE) have been associated among people with such small, overly
dependent family networks (Leff, 1979; Vaugn & Leff, 1981). Approaches that
incorporate family network support have been found to reduce psychiatric relapse rates,
hospitalization readmissions, family burden of care, and overall cost of outpatient
treatment (Penn & Museser, 1996). The clinical implications of this study demonstrate
the need for MFI‘s and other mental health professionals to encourage a sense of hope of
recovery to individuals challenged with serious and chronic mental illness and their
families. The history of family systems approach, however, negatively portrayed the role

of families in the etiology of major mental illnesses. Treatment approaches for

schizophrenia, for example, blamed family members for the onset or relapse of psychosis.
Early Expressed Emotions studies (EB) (Brown, Birley & Wing, 1972) gave way for
family therapists to play a role in treatment, which was not embraced by family
organizations of the mentally ill (Nichols & McFarlane, 2001). More recent interventions
have also focused heavily on hospitalization or psychopharmacological approaches with
little regard for the role of the family (Walsh, 1996).

Community based mental health programs, such as Clubhouses can provide
individuals with opportunities to expand social network size beyond family members,
develop reciprocal and mutually supportive relationships with club peers which can lead
to more satisfying supportive relationships with family members.

Purpose

Early literature in the area of social networks suggests that social ties are
important to mental health (Biegel, McCardle & Mendelson, 1985). A key component of
the Clubhouse psychiatric rehabilitation program is to establish or maintain social
relationships. Clubhouses offer individuals opportunities to meet new friends to expand
personal networks, as well as to identify themselves as someone other than a person
living with mental illness (Macias & Rodican, 1997). While a number of studies have
found that people with severe mental illness have smaller networks, which mostly consist
of kin, the social networks of clubhouse members have not been sufﬁciently described.
Existing studies employ small sample sizes and are typically descriptive in nature (Beard,
1992; Perese, Getty & Wooldridge, 2003; Stein, Barry, Van Dien, Hollingsworth &
Sweeney, 1999; Stein, Rappaport & Seidman, 1995). Further, the recovery provides a

new lens from which the relationship between social network supports and mental health

can be investigated. It is unclear whether larger social support networks and the greater
level of network support are related to the recovery process. This study proposes to ﬁll
the existing gap in the literature by examining presence of family and social support
networks are related to the recovery process among a sample of clubhouse members with
chronic mental illness. In addition, this study examined the relationship between
clubhouse engagement, as measured by clubhouse participation and sense of
community/cohesion with club peers, the extent of social network support, and the
subjective experiences of recovery. It is contended that the social ecology of Clubhouses
does in fact foster a sense of community and provides opportunities to engage in
supportive reciprocal relationships (Herman, Onaga, Pemice-Duca, Oh, 2005).
Theoretical Rationale

Ecological Rationale

An important element to the ﬁeld of Marriage and Family Therapy and to the
discipline of Family and Child Ecology (FCE) is the relationship between individuals,
children, and families and their environments (Griffore & Phenice, 2001) The study of
social networks involves the reciprocal relations of individuals and their environments,
which is a core dimension of the human ecological model (Bronfenbrenner, 1989;
Bubloz, Eichert & Sontag, 1979). According to Hirsh (1981), a healthy social network
that consists of supportive and reciprocal relationships can provide enriching
opportunities, support development, and afﬁrm positive social identities. Therefore,
social networks can be viewed as ‘systems’ of inter-related individuals, information,
energy, resources, and communication in transactional processes within an ecological

context (Vaux, 1988). In this perspective, support is not a property of the person or the

environment, but rather the interaction between the two. This is consistent with
Bronfenbrenner’s (1979) theory of human development and viewing behavior as a joint
function of the person and environment.

Two frameworks guide research questions and hypotheses. First, the
underpinnings of clubhouses as they relate to social networks will be presented. Second,
the contributions of the Feminist Perspective on Bowen Family Systems Theory is
reviewed to discuss the concepts of recovery and development of self and differentiation.

Clubhouse Programs

“A clubhouse is a community organized to help people living with serious mental
illness as they manage their illness and rejoin the worlds of employment, education,
family, and friends” (International Center for Clubhouse Development, 1998). The ﬁrst
psychosocial clubhouse emerged nearly 52 years ago when 10 residential psychiatric
patients discharged from the New York State Psychiatric Hospital decided they needed a
place to meet and discuss ways to readjust into society. The meeting place became
known as “Fountain House.” Today, Fountain House represents the model rehabilitation
program found across the US. and the world. There are clubhouses modeled after
Fountain House located in over 20 countries.

Like New York’s Fountain House, four fundamental principles guide the
Michigan clubhouse programs: (a) the clubhouse belongs to its members, (b) daily
attendance is desired and makes a difference to other members, (c) members feel wanted
as contributors, and ((1) members feel needed (Beard, Propst, & Malamud, 1982).
Clubhouse programs offer a range of community supports such as housing assistance,

employment training and placement, and self-help resources.

The clubhouse model has an egalitarian social structure with members and staff
sharing in clubhouse work and decision-making. The central tenet of the clubhouse
model is what is known as the "work-ordered day." It mimics a normal workday in that
the day begins at 9:00 A.M. and essentially ends at 5:00 P.M., with social activities and
support groups occurring after hours. The work-ordered day is designed to provide
individuals with a workday structure that incorporates work ethics and social skills
needed to prepare one for community reintegration. Clubhouse members work side-by-
side along with clubhouse staff, interacting through the work-ordered day activities.
Clubhouse participants are referred to as “members”, and membership is voluntary.

The clubhouse was designed to address the needs of people living with chronic or
persistent mental illness who have encountered losses in social skills, friendships, family
connections, and employment (Mastboom, 1992). As a rehabilitation program, clubs
assist peOple in leading more productive, community oriented lives by encouraging skill
development within an environment that supports them to meet the demands of daily
living, socialization, and employment (Anthony, Cohen, Farkas & Gagne, 2002).

According to Beard, Propst & Malmund (1982), social interaction is an important
aspect of the program. These authors assert that members ‘feeling needed’ is one of the
three core elements of the clubhouse model. Therefore, it is contended that through
clubhouse participation, members gain a sense of connection with others and, thereby,
reduce isolation and increase social ties. Further, members also elicit support from their
social support networks and engage in mutually supportive reciprocal interactions with

network supports.

10

Family Systems Theory

Social relationships are a universal, fundamental human need almost as powerful
as the need for food, which if not satisﬁed, could thwart positive psychological
development and adjustment (Baumeister & Leary, 1995). Research has shown that
social isolation, lack of social ties and personal networks are predictive of death or
mortality in the general population (Berkman, 1995; Berkman, Glass, Brissette, Seeman
2000; Cohen, 1988; House, Landis & Umberson, 1988).

Social relationships provide a number of psychological and practical beneﬁts.
Social relationships can be examined through social networks, contact with others,
providing or giving support, or developing an afﬁliation with others. Social network
supports have been argued to “enhance psychological well-being because they fulﬁll
basic human needs”. . .for “autonomy, relatedness and competence” which foster
enhanced “self - regard, self regulation, vitality, and feelings of connectedness with
others” (Ryan & Solky, 1996, p. 250).

Murray Bowen, whose early work focused on a clinical population of people with
schizophrenia and their families, viewed families as central to the development of the
competent, autonomous adult (Bowen, 1966). Families are a primary unit of social
relationships. Bowen viewed the family as an emotional and relational system. Bowen
(1966) contended that the symptoms of mental illness resulted from a loss of self within
the emotional system of the family. He believed one’s inability to function autonomously
in the context of the family was predictive of functional impairments. The theory also

postulates that family relations characterized by high levels of tension or physical and

11

emotional cut-offs exacerbate psychiatric symptoms. Bowen asserted that an inability to
differentiate from others lead to increased emotional reactivity or high levels of tension
and conﬂict when relating with signiﬁcant others such as family members.

“Cut-offs” are deﬁned as the loss or disruption of an important social or family
relationships. Cut-offs can be an emotional or physical separation from a social
relationship. The ability to maintain social ties rather than “cut-off” from family during
signiﬁcant life events, such as the experience of mental illness, can determine future
recovery and quality of life (Tittler, 1998). This suggests that for individuals with chronic
mental illness, social and family relations may run the risk of cut-offs because of strained
interactions or withdrawal that can be a result of persistent symptoms.

At the other end of the continuum, enmeshment or fusion of family members is
believed to thwart individual autonomy (Bowen, 1966). Fusion may be manifested
through over-dependency between family members. For instance, a relative with a mental
illness may be dependent on family members for ﬁnancial or instrumental support, while
simultaneously attempting to become more autonomous. This, in turn, creates
ambivalent emotional ties to family members, which results in a signiﬁcant source of
stress.

Emotional reactivity, cut-offs and relational fusions are the consequences of the
poorly differentiated person, according to Bowen (Kerr & Bowen, 1988; Bowen, 1978).
Bowen’s signature concept of differentiation of self is deﬁned as the ability to experience
closeness with others while maintaining independence from them. Differentiation is the
means by which individuals move toward delineation of the self in relation to other(s).

Less differentiated individuals are believed to become easily distressed and suffer more

12

psychiatric symptoms, while highly differentiated persons demonstrate greater
psychological adjustment.

While the benchmark of the healthy individual, according to original Bowen
Family Systems Theory, is characterized as the “autonomous”, “being for self”, “goal-
directed” person (Walsh & Scheinkman, 1989), these are typically qualities valued in
males in Western societies. In contrast, the inclusive Feminist model of Bowen Family
Systems Theory incorporates differentiation in the context of relationships. It values an
orientation toward others, as well as social ties and emotional identiﬁcation with others
(Walsh & Scheinkman, 1989). In this view, individuals develop in the context of
relationships, not separate from them. And like recovery, differentiation of self is
independent of symptoms and severity of diagnosis (Greene, Gilbert, Hamilton &
Rolling, 1986).

Social networks or relationships are hypothesized to foster the development of
self-esteem, personal competencies, and independence (Cobb, 1976), which are
consistent with the recovery from mental illness philosophy. The concept of
differentiation is described as life long process in the development of self in relation to
others. Similarly, the concept of recovery has also been described as continuous journey
that incorporates multiple relational and autonomous dimensions, including the
willingness to ask for help, reliance on others, and goal orientation and success (Corrigan
& Phelan, 2005). More highly differentiated individuals may remain in satisfying contact
with families, and establish more mutually based relationships with others.

The study incorporates an expanded version of Bowen’s Family Systems Theory

(Bowen, 1978) with an inclusive deﬁnition of differentiation. Although Bowen did not

13

directly talk about recovery from mental illness, Bowen theory does postulate that
positive family relationships that support a balance between connection and autonomy
are related to increased differentiation. Bowen Theory focused on how the “self”
functions in relation to others. In an expanded feminist version of Bowen’s Family
systems theory, individuality (i.e., capacity to function as an autonomous self) and
togetherness (i.e., capacity to function in relationship to others) are viewed as distinct
dimensions of emotional differentiation equally lead to development of self (Knudson-
Martin, 1994; 1996). In contrast to Bowen’s original conceptualization of differentiation,
individuals can possess varying capacities of autonomy and togetherness. We can
develop self either through separateness or connection with others. Persons who tend to
experience the self through their connections with others develop mutual relationships
and well integrated sense of self. This may be indicative of a recovery process in which
individuals begin to engage in more mutually supportive relationships with members of
their social networks as opposed to being solely dependent on them for emotional and
social support. If the development of self can be theorized as a proxy of the recovery,
then positive social network supports are viewed as a critical element to the recovery
process. Therefore social networks characterized as small and highly dependent on family
place family relations at increased risk for stress and emotional reactivity. Increased
stress and emotional reactivity is theorized to hinder differentiation, which may also
affect the psychological process of recovery

Based on this framework, social network relations with family members
characterized by satisfying contacts and reciprocity would be predictive of greater

subjective accounts of recovery. Further, greater family network supports would also be

14

related to speciﬁc recovery domains that reﬂect aspects of the development of self, such
as personal conﬁdence and being oriented toward personal goals and success.

Overall, this study examines the relationship between social network measures
and recovery through the inclusive lens of Bowen Family Systems’ Theory. If the
construct of recovery can serve as a proxy to Knundson- Martin’s reconceptualization of
development of self in relations to others, then social network supports are hypothesized
to be signiﬁcantly related to the process of recovery. Therefore, it is reasonable to
question which social and family network dimensions relate to recovery and whether
engagement in clubhouse programs can serve as a way for people to increase network
size and support, as well as engage in satisfying and mutual relationships with families

and others.

15

Definition of Terms

The following section describes the terms used in this study.

Serious/Chronic Mental Illness/Severe Psychiatric Disability: Federal deﬁnition
of severe mental illness includes a diagnosable mental, behavioral, or emotional disorder
of sufﬁcient duration that results in functional impairments which substantially interferes
with or limits one or more major life activities. Participants in this study have a primary
diagnosis of major depressive disorder, bi-polar disorder, or psychotic disorder, such as
schizophrenia.

Community Support Programs (CSPs): Community based psychosocial
rehabilitation program for individuals with chronic mental illness (National Institute of
Mental Health [NIMH]). Designed to decrease repeated hospitalization and increase
psychosocial functioning and community integration.

Psychosocial Rehabilitation: (Also known as psychiatric rehabilitation).
Psychosocial rehabilitation (PSR) programs is a universal term to refer to approaches that
incorporate social aspects of treatment and recovery, that involve daily living skills
rehabilitation, social skill development, and role functioning (Anthony, Cohen, Farkas &
Gagne, 2002). A core value of psychosocial rehabilitation is the person-centered
approach which is a processing involving the whole person: their strengths, interests,
hopes, and fears (Anthony, Cohen, Farkas & Gagne, 2002). It views the person as an
individual rather than an illness or a label, focusing on choice, growth potential, and
improving functioning rather than reducing symptoms. Programs are a deviation from

the traditional medical model of ‘day treatment programs’ in which mental health

16

consumers are recipients of services as opposed as directors of their recovery. Clubhouse
programs are a form of psychosocial rehabilitation programs.

Clubhouse Program: A voluntary PSR program delineated by membership.
Clubhouses are based on an egalitarian structure to promote cohesion, equality, and social
connectedness. Core elements of the clubhouse are social interaction, work, and
empowerment.

Clubhouse Member: Anyone with a psychiatric disability who attends the
clubhouse at least once at any time.

Clubhouse staff: A person employed by the clubhouse and auspice agency to
work side by side with clubhouse members. Clubhouse staff may or may not be
recovering from a psychiatric disability. Clubhouses maintain a low staff to member
ratio, typically, one staff for every 15-20 members.

Clubhouse peers: Anyone who participates in the clubhouse program, including
staff and members.

Social Network: A list of people nominated by the focal person who are
perceived to be individuals the focal person perceives to have a relationship with or is a
source of support (e. g., turns to in times of need, to have fun, or to talk to).

Social Network Support Dimensions: Measures pertaining to the number of
people on a network and the extent of perceived support, perceived reciprocity, level of
satisfaction, level of importance, and frequency of contact with network members.

Reciprocity in Social NetwoLk Relations: The extent to which one perceives the

relationship as a exchange of mutual support between both parties. Reciprocity is an

17

important element in relationships (Wellman, 1985). It provides equity in the relationship
and satisﬁes the need to give and meaningfulness in social relationships.

Clubhouse PJarticipation: Extent of participation in the clubhouse through the
‘work-ordered’ day, number of hours attending the club in a given day or week, and
length of clubhouse membership. Social activities with clubhouse peers during and after
clubhouse hours consist of social gatherings, outings, or community events.

Sense of Community: A community may be viewed as a group of people in a
shared environment with common characteristics or interests, or in a social relationship.
Sense of community refers to concepts of cohesion which can fellowship through shared
values (McMillan & Chavis, 1986). Buckner (1988) describes it as a “sense of
belongingness, fellowship, ‘we-ness’ experienced in the context of a functional group or
geographically based collective” (p. 773).

Clubhouse Engagement: A construct speciﬁed in this study that delineates the
extent to which clubhouse members are engaged in the clubhouse philosophy and
demonstrate a sense of afﬁliation with peers. It is the extent to which members attend the
club and have developed supportive and reciprocal relationships with club peers.

Recovery: The recovery literature deﬁnes recovery as both a subjective and
experience. For the purposes of this study, recovery is deﬁned as a psychological
construct deﬁned as an ongoing process involving the perception or experiences of a
person living a satisfying life given the constraints of mental illness.

Assumptions and Limitations
For the purposes of this study, it is assumed that utilizing self-report interviews

with pe0ple with mental illness is a valid and reliable method to gather information on

18

clubhouse member characteristics and perceptions of experiences. It also assumes that
gathering social network information is a valid way to understand who is considered part
of the member’s network and their relationship to that member.

In the strictest sense of the concept, Social Network Analysis (SNA) is a method
of mapping relationships and connections between people, groups, organizations, and
other entities. The term ‘node’ is used to denote people or entities. Lines depicting and
relationship or social exchanges are drawn between nodes. This type of mapping
provides a visual and mathematical representation of relationships. SNA frequently
involves several levels of analysis; with individuals embedded in multiple networks. This
is referred to as ‘nested design’ or hierarchical data. Examples of studies using nested
data may be families embedded in a larger network, much like Bronfenbrenner’s
Microsystems (1979). We may be nested in our work organizations, individuals nested in
families, families nested in neighborhoods, neighborhoods in communities, in societies,
and so forth (Hannaman, 1998). Traditional SNA was not used in this study because it
was outside the original scope of the larger clubhouse study.

There are a number of limitations presented by the current study. First, it is
questionable whether the perception of who is in one’s social network is correct. Studies
that employ true social network methodology typically utilize multiple sources to conﬁrm
or deny whether someone is part of an individual’s social network. These studies can be
expensive to conduct. But as Stein, Rappaport, and Seidman’s (1995) showed, network
members from 97 clubhouse members’ social network were corroborated by at least one
other person in the network. Results from matched interviews also demonstrated high

correlations between clubhouse member and the nominated social network member. In

19

sum, the study found moderate to high agreement between reporters on factual
information about the network structure (e. g., frequency of contact, type of contact) and
global aspects of the relationship (e.g., satisfaction of the relationship).

Second, the current study also is limited by its use of a cross-sectional design,
which does not allow for causal inferences, a test of directionality, or an examination of
changes over time. The purpose of the current study, however, is not to test the impact of
mental illness on social relationships, but rather to describe the social network supports of
clubhouse members and the relationship to recovery.

Third, this study also fails to distinguish between positive or negative social
network relationships. A number of studies have revealed that not all social network
nominations are individuals with whom the participant shares a positive relationship
(Brenner, Norvell & Limacher, 1989; Green, Hayes, Dickinson, Whittacker & Gilheany,
2002; Pickens, 2003; Page], Erdly, & Becker, 1987; Rosenﬁeld & Wenzel, 1997; Uchino,
Holt-Lunstad, Smith & Bloor, 2004). Although the social network interview does not ask
participants to report individuals with who they share a negative relationship, it does
inquire about the importance and satisfaction with each network member.

The results of the structured social network interview are dependent on the
methodology, such as the type of questions asked to elicit network information. The
interview does not ask participants about the speciﬁc type of support received. For
example, speciﬁc questions of who provides ﬁnancial, emotional, and instrumental
support can conjure rather different network nominations (Bass & Stein, 1997; Lovell,
Barrow & Hammer, 1984). Instead, general support questions were used to probe and

gather information about the people who provide emotional and ﬁnancial support, as well

20

as who they engage in leisure activities with. A unique aspect of the study, which is
absent from other social network support studies involves mental health participants, is
the extent to which individuals provide or reciprocate support to network members.
Overview
This chapter included an introduction, purpose of the study, and theoretical
rationale. The theoretical underpinnings of the clubhouse model and the contribution of a
Feminist Elaboration of Bowen Family Systems Theory were presented to support the

research questions. Deﬁnition of terms, and assumptions and limitations were delineated.

A review of the literature relevant to this study is presented in Chapter Two.
Chapter Three presents research questions, conceptual models, and provides a description
of the data set, participants, and measures for this study. Chapter Four describes the

results, followed by the discussion and conclusion found in Chapter Five.

21

CHAPTER TWO

REVIEW OF THE LITERATURE

This chapter consists of a literature review that addresses major areas of this
study. First, a broad overview of mental illness as a problem is presented, followed by a
review of the social network literature, and a brief review of the role of family networks.
A literature search on social networks and recovery from mental illness resulted in a
handful of relevant articles. Therefore, studies falling under the umbrella of social
networks, social network supports, and mental health were reviewed. Mental health
outcome studies examining the relationship between social relationships and
psychological well-being, wellness, and functioning were used as a proxies to the
construct of recovery. Social network research involving clubhouse programs and other
mental health consumer programs were included.

Mental Illness as a Problem

The label of a severe psychiatric disability from a community mental health
deﬁnition typically includes diagnoses of schizophrenia, major depression, and bi-polar
depression. In a given year, 5% to 7% of adults have serious mental illness (US.
Department of Health & Human Services, 1999). Individuals with severe psychiatric
disabilities have reduced life expectancy (Berren, Hill & Merikel, et al., 1994) and
increased risk of death (Kouzis, Eaton & Leaf, 1995). Mental illness is ranked as the 4‘h
most debilitating illnesses among non-communicative diseases in the world (WHO,
2001). Over 15% of the burden of disease in industrialized countries is accounted by

mental illness, which is greater than the burden caused by all cancers combined (Murray

22

& Lopez, 1996). In a given year, approximately 6.5 million Americans become disabled
by severe mental illness (Narrow, 1998). Data from the Global Burden of Disease study
found that among mental illness, major depression, bipolar, schizophrenia, and severe
obsessive compulsive disorder constitute the top four disorders resulting in the greatest
number of years lost of healthy life (Murray & Lopez, 1996). Schizophrenia is estimated
to affect 2.5% of the nation’s population. Major depression is the leading cause of
number of healthy years lost to a disability. It is projected that psychiatric and
neurological conditions will grow to reach the number one position in disease burden by
2020, greater than cardiovascular and cancer conditions (Murray & Lopez, 1996).
According to the Surgeon General’s Mental Health Report (US. Department of Health &
Human Services, 2001), the direct costs of mental illness are “exceedingly high” with the
US. spending nearly 70 billion dollars on services in a given year. Yet the indirect costs
have totaled more than 80 billion in lost productivity in work, school, and in the home
(Rice & Miller, 1996; 1998). It is estimated that only 16% of people experiencing or
suffering from psychiatric disability actually seek treatment (Narrow, 1998).

The social consequences associated with severe psychiatric disabilities may
arguably be more devastating than the illness itself. While the loss of daily living skills,
employment or housing permeates the demographic proﬁles of persons hospitalized for a
mental illness, stigma, social rejection (Estroff, 1981), and the loss of social and family
relationships (Boydell, Gladstone & Crawford, 2002; Davidson & Stayner, 1997;
Deegan, 1988), saturate the narratives of many of those affected by mental illness

(Macias & Rodican, 1997, Newton, 2001).

23

Social Network Research
Deﬁnition

Social support, social ties, and social networks are all distinctly deﬁned concepts
with roots to the attachment and sense of belonging literature (Vaux, 1988). Theories
posited on the fundamental need to belong or attach have been proposed by a number of
well known theorists (e.g., Bowlby, Ainsworth, Homey, Stack Sullivan, Freud, Alder,
and Bowen). Attachment theory has been used to describe the human propensity toward
social contact, bonding, and relating that is believed to be associated with positive
psychological adjustment, as well as human survival.

Although many theorists and researchers use the terms of social network and
social support interchangeably, they are very distinct concepts (Berkman, Glass, Brissette
& Seeman, 2000; Vaux, 1988). Social networks appear to be an umbrella term for a
number of concepts related to social relationships and social supports. Mitchell ( 1969)
originally deﬁned social networks from the general systems perspective as “the system of
relationships with other individuals” (p. 12 cited in Vaux, 1988). Personal social
networks are also deﬁned as a collection of individuals who know and interact with the
focal person (Milardo, 1988) and as “support that leads one to believe that he or she is
cared for, loved, valued, and belongs to a network with mutual obligations” (Cobb, 1976,
p. 7, cited in Vaux, 1988).

Measurement of Social Networks

Social networks can be examined in terms of size, structure, relationships,

composition (i.e., proportion of family, friends, etc.), homogeneity of network members

(Uchino, Holt-Lundstad, Smith & Bloor, 2004), frequency of contact, geographic

24

proximity, intimacy, degree of reciprocity, and type of support (Vaux, 1988). Other
methods to examine social networks include the density (i.e., extent to which members
know one another) and the multiplexity (i.e., extent to which members fulﬁll more than
one role or function) (Goldberg, Rollins & Lehman, 2003).

Different types of support can be elicited from a social network support interview.
Common forms of support include ﬁnancial (e.g., extent to which you can depend on
someone for money), instrumental (e.g., practical assistance, transportation, house
cleaning), material (e.g., clothing, supplies, food), and emotional (e.g., listening,
guidance, advice). Speciﬁc questions about these forms of support can evoke different
sources of support and network sizes (Bass & Stein, 1997; Lovell, Barrow & Hammer,
1984).

Some social network studies have demonstrated gender differences. Alexander’s
(2001) interviews with 18 men diagnosed with major depression found that having a
signiﬁcant partner did not increase the likelihood of nominating that person as a
conﬁdent on the men’s social network. Male participants reported refraining from
burdening others with problems, desiring to keep problems to themselves, or waiting for
others to notice a problem. In another small qualitative study, women identiﬁed more
reciprocal relationships among women (Pickens, 2003). Some have argued that women
are socialized to illicit help from their social networks (Rogers, Anthony & Lyass, 2004),
while men are socialized to be less dependent on others.

Social Networks and Mental Illness
Social networks among people with serious mental illness are small, ranging from

ﬁve to ten members, but always half the size reported in the general population (Bass &

25

Stein, 1997; Borge, Martinsen, Rudd, Watne & Friis, 1999; Froland, Brodsky, Olson &
Steward, 2000; Green, Hayes, Dickinson, Whittaker & Gilheany, 2002; Goldberg, Rollins
& Lehman, 2003; Hardiman & Segal, 2003; Perese, Getty & Wooldridge, 2003;
Rosenﬁeld & Wenzel, 1997; Seidman, Sololove, McElroy, Knapp & Sabin, 1987).
Networks consisting of less than ﬁve members have been associated with more severe
psychiatric diagnoses, such as schizophrenia and related psychotic disorders (Goldberg et
al., 2003). Social relationships among people with psychiatric disabilities are
characterized as less reciprocal, with a greater likelihood of being a recipient of support
than a provider of support (Wilson, Flanagan & Rynders, 1999). Studies have shown the
social networks of individuals with severe psychiatric disabilities are typically composed
of relatives, mental health services providers or professionals, while friendships are
limited or rare (Rosenﬁeld & Wenzel, 1997).

The concern that emerges from these studies is that small networks often have
more network members performing multiple functions of support, which may contribute
to burden of care for families (Fadden, Bebbington, Kuipers, 1987) and increase
emotional reactivity among family members (Leff, 1979). Larger networks are presumed
to be more accessible and diverse, thus placing less burden on any one network member,
and increasing the likelihood of greater forms of support.

Studies have found increases in social network size to reduce loneliness and
isolation, and improve psychosocial functioning and psychiatric symptoms (Biegal, Tracy
& Corvo, 1994; Davidson, Shahar, Stroyner, Chinman, Rakfedlt & Kraemer Tebes, 2004;
Froland, 1978; Froland, Brodsky, Olsm & Steward, 1979). Small social networks may be

the result of disrupted relationships from the presence of a serious mental illness

26

(Boydell, Gladstone & Crawford, 2002), repeated or prolonged hospitalizations (Albert,
Becker, McCrone & Thronicroft, 1998; Froland et al., 2000), psychotic symptoms
(Goldberg et., a1, 2003), social withdrawal (Rosenﬁeld & Wenzel 1997), stigma and
social rejection (Perlick, Rosenheck, Clarkin, Sirey, Salahi, Struening & Link, 2001).
Studies have also pointed to brain abnormalities such as ventricle size differences that
inhibit the development of social relationships among persons with schizophrenia
(Seidman, 1983; Seidman, Sokolove, McElory, Knapp & Sabin, 1987). Although many
of these studies are correlational employing cross-sectional data, thus making it difﬁcult
to assess directionality and causality, they have established a strong relationship between
the presence of mental illness, the lack of social network ties, and poor psychological
well-being.
Family Networks

Studies have shown that positive connections between family members and
friends who are accessible, express affection, visit and join for social activities, and are
available in a time of need, play a signiﬁcant role in fostering psychological well-being
and personal self efﬁcacy among people living with chronic mental illness. Among a
small sample of 30 mental health consumers, extended family has been found to provide
a signiﬁcant portion of the emotional support (81.4%), as compared to co-workers
(74.5%), friends (64.3%), and family of procreation (63%) (Connelly & Walsh, 1996).

Caring for a family member with mental illness may limit the number of positive
interactions and create feelings of emotional stress and caregiver burden (Hatﬁeld &
Leﬂey, 1987). Stressful or strained network relations risk emotional or physical cut-offs

from important relationships. Bowen Theory asserts that unsupportive family

27

relationships are predictive of psychological distress. In a study that included nearly 300
psychiatric interviews, Fisk, Rowe, Laub & DeMino, (2000) found individuals who had
experienced severed or estranged family relations reported greater psychological distress,
and continued to maintain a “powerful emotional connection” to family members despite
the cut-off relationship. Positive family relations were found to inﬂuence successful
transitions to independent community living, which has been corroborated by other
studies (Wood, Hurlburt, Hough & Hofstetter, 1998; Calsyn & Winter, 2003).

Persons with severe psychiatric disabilities frequently report less satisfaction with
the quality of their family relationships (Lehman, Ward & Linn, 1982), but signiﬁcantly
turn to family more often in times of need, (Froland, et al., 1979). Family relationships
have often characterized as dependent and non-reciprocal among individuals with
psychiatric disabilities, (Green, Hayes, Dickinson, Whittaker & Gilheany 2002). The role
of family networks from the consumer perspective was explored through a small
qualitative study by Green et al., (2002). Participants described having the comfort of
family “being there " even if relationships were characterized as emotionally
unsupportive. Mental health consumers primarily viewed their relationships with family
as ‘overly-dependent’, on ﬁnancial or material support, which resulted in high level of
emotional ambivalence toward kin. The onset of mental illness was blamed for
disrupting parent-child relationships, “friends drifting” away, and feelings of loneliness.
The desire for social contact for many was so strong, that some reported simply “going
shopping” to engage in social contact. Psychosocial mental health programs were

credited for providing social connections that participants would have not otherwise

formed.

28

Building social networks and social ties with others beyond families can reduce
the burden on families. Reform efforts in community mental health services raised
awareness of increasing interventions that: 1) build new network ties; 2) maintain and
strengthen existing social ties; and 3) enhance family ties (Biegel, Tracy & Corvo, 1994).

Cross-cultural studies among people living with schizophrenia revealed
differences in recovery rates due to familial connections. In a review of early WHO
cross-cultural research in recovery from schizophrenia, Calabrese & Corrigan (2005),
discussed how developing countries were 30% more likely to meet recovery criteria from
schizophrenia than more industrialized countries like Germany and the US. The authors
contended that cultures in developing countries place greater importance to maintaining
family and social relationship, and social roles (e. g., teacher, mother, worker), while
Western cultures tend to place greater emphasis on autonomy from the nuclear family
and de-emphasize the importance of extended family members.

Recovery from Mental Illness

Although recovery is uniquely deﬁned by the individual, it has also come to be
deﬁned as a measurable outcome in psychosocial programs, as well as a subjective
attitude independent of symptoms (Anthony, 1993; Corrigan & Ralph, 2005;
Deegan,1988; Liberrnan & Kopelowicz, 2005; Liberrnan, Kopelowicz, Ventura &
Gutkind, 2002; Schiff, 2004). Available and accessible social and family network support
are signiﬁcantly associated with recovery (Corrigan & Phelan, 2004; Liberrnan,
Kopelowicz, Ventura & Gutkind, 2002). Recovery has also been associated with

participation in family psychoeducation (Resink, Rosenheck & Lehman, 2003).

29

Deﬁning Recovery
The word “recovery” can evoke variety of meanings, each individually

construed to reﬂect personal experiences. The concept of recovery is multidimensional,
encompassing physical and mental health, as well as interpersonal well-being (DeMasi,
Markowitz, Videka-Sherman, Sofka, Knight & Carpinello, 1996). Anthony (1993)
summarizes recovery as a deeply personal and unique process of changing one’s
attitudes, values, feelings, goals, and skills in life given the constraints of mental illness.
Some studies have demonstrated two-thirds of people with schizophrenia achieving full
recovery (Davidson & McGlashan, 1997; Harding, Brooks, Ashikaga, Strauss & Breier,
1987). While recovery has been typically deﬁned from a medical model that involves
elimination of symptoms and returning to previous levels of functioning (Mueser,
Corrigan, Hilton, Tanzman, Schaub, Gingerich, Essock, Tarrier, Morey, Vogel-Schibilia
& Herz, 2002), others have deﬁned recovery as a process despite the presence of
psychiatric symptoms (Resnick, Rosenheck & Lehman, 2003).

Deﬁnitions share similar themes to Bowen’s observations of healing from chronic
mental illness. Bowen (1967) viewed healing as a “self-regenerative
phenomena. ..[deﬁned as] not only being self-responsible, but also self-actualizing ...the
act of taking responsibility for one’s own emotional being and destiny is not only key to
survival, but that very attitude creates the self that is the necessary resource for that end”
(cited in Friedman, 1991, p. 159). Likewise, major proponents in the domain of recovery
also view healing as a self-actualizing journey traveled through self-efﬁcacy, positive
social identity, change in attitude, hope, and self—advocacy (Ralph & Muskie, 2000;

Beale & Lambric, 1995). Ralph’s (2000) review of the recovery literature prepared for

30

the Mental Health Report of the Surgeon General (US. Department of Health and Human
Services, 1999), described ‘extemal factors’ such as social relationships,
interconnectedness with others, family, friends, and professional support, and the
presence of people who encouraged and believed in the individuals’ ability to cope and
recovery from mental illness as indicative of the recovery process.
Social Networks and Recovery

Until recently, the concept of recovery from mental illness has not been a distinct
measurable construct in quantitative studies (Loveland, Weaver-Randall, & Corrigan,
2005; Ralph, Kidder & Phllips, 2000). The concept of recovery is closely linked to the
notion of wellness or well-being (Ralph, & Corrigan, 2005). Traditional mental health
outcome studies have used a constellation of measures assessing ‘wellness or well-being’
that are now considered to reﬂect various dimensions of the recovery process (Liberrnan
& Kopelowicz, 2005; Resnick, Rosenheck & Lehman, 2004). Studies have demonstrated
strong associations between social network support and indictors of wellness, including
health status (Berkman, 2000), mental health (Biegel, McCardle & Mendelson, 1985;
D’Augelli,1983) psychosocial and psychological functioning (Rogers, Anthony & Lyass,
2004) clinical symptoms and quality of life (Goldberg, Rollins & Lehman, 2003;

Markowitz, 2001).

A panel study with 4,000 subjects conducted by Calsyn and Winter (2002),
conﬁrmed a recursive relationship between network contact and psychiatric symptoms.
Increased contact with friends and family reduced psychotic symptoms, which also lead
to increased support from family and friends. Path models hypothesizing the relationship

between network support and presence of psychotic symptoms conﬁrmed that friends and

31

family reduced symptoms at two consecutive time points in the study. As symptoms
worsened, contact with friends and family increased.

Social networks composed of peers with a mental illness appear have positive
impacts on psychological well-being as compared to networks composed of ‘normals’
(i.e., those without a mental illness) (Boydell, Gladstone & Crawford, 2002). The Yale
Supported Socialization Partnership Project (Davidson, Haglund, Stayner, Rakfeldt,
Chinman & Kraemer Tebes, 2001), however, found that mental health consumers
signiﬁcantly expressed a greater desire to socialize with others who were unlike
themselves (i.e., shared a psychiatric disability). Participants who reported attending
social mental health clubs, such as clubhouse programs, reported a desire to develop
social connections based on mutual interests instead of the shared experience of a
disability. Rosenﬁeld & Wenzel (1997) found whether they are based within the mental
health system or not, any supportive ties increased wellness.

A classic social network study by Froland, Brosky, Olson & Stewart (1979)
examined the available network support and the contribution to social and psychological
adjustment among mental health consumers in inpatient, outpatient, and day treatment
settings. Social networks characterized as less reciprocal, less satisfying, and having
infrequent contact with family members were indicative of consumers who expressed
more distress, were less productive, and had repeated experiences with hospitalization
and treatment. Increased psychological distress was present among those who reported
absent or less supportive family networks. The importance of this early study raised

awareness of the relationship between absent, or poor family relations and the need to

32

strengthen family and social ties for those who placed less emphasis on these sources of
support.

Goldberg, Rollins & Lehman (2003) used secondary cross-sectional data to
examine demographic, clinical, and life satisfaction as social network correlates. Larger
social networks were signiﬁcantly related to higher level of education, greater life
satisfaction, greater satisfaction with social relations, and higher self-esteem ratings.
Network size, however, was unrelated to age, gender, or history of hospitalization.

In a controlled experimental study, Davidson et a1. (2004) found outcomes
measuring psychological well-being (e.g., life quality, self-esteem), depression,
psychiatric symptoms, and health were largely dependent on the frequency of contact
with others. Participants were randomly assigned into one of three conditions: (1) a
mental health consumer paired with another mental health consumer, (2) a mental health
consumer paired with a community volunteer, (3) and a stipend only group. Researchers
hypothesized that increases in well-being are a result of engaging in social activities with
their assigned partner. Results showed that participants paired with mental health
consumers who failed to maintain regular contact faired better than those paired with a
community volunteer. These mental health consumers who were paired with a non-
mental health consumer community volunteer and did not engage in regular social
activities together deteriorated over time. However, if participants meeting with a
community volunteer met regularly, they faired better than all other groups.

Brier and Strauss (1984) described the role of social relationships and recovery
with 20 hospitalized individuals experiencing psychosis. Data on the beneﬁts of social

ties was collected during the hospitalization period and one year following discharge. The

33

authors identiﬁed 12 themes characterizing ways in which social network relationships
are beneﬁcial to the recovery process. Themes relevant to aspects of the current study are
described. First, social relationships provided constancy in associating with people they
knew before the psychiatric hospitalization. This helped individuals to connect to a pre-
hospital identity. Social relationships also provided social approval and integration
which increased a sense of acceptance and belonging to a larger community. Modeling
socialization behavior of others assisted individuals in incorporating positive behavior.
Reciprocal relations were important in equalizing the relationship in the ability to share
with others as wells as be of assistance to others. Social relationships also provided the
motivation to achieve higher levels of functioning through encouragement.

The current study attempted to replicate some ﬁndings from a similar
investigation between social network supports and recovery. Corrigan & Phelan (2004)
examined secondary data collected from 1,824 individuals involved in Consumer
Operated Services Project (COSP) (Campbell, Johnsen, Lichtenstein, Noel, Yates,
McDorel Herr, et a1. 2003). The Recovery Assessment Scale factors (Corrigan, Giffort,
Rashid, Leary & Okeke, 1999) served as dependent variables while social network
dimensions assessed by the Social Network Scale (SN S) were employed as independent
variables (Stein, Rappaport & Seidman, 1995). The SNS measure requires participants to
generate a list of people in the focal person’s life who are perceived as supportive.
Network size is equated to ‘network support’ and participants rated the extent of
satisfaction and mutuality with each network member. The Recovery Assessment Scale is
composed of a ﬁve factors structure deﬁning recovery as (1) Personal Conﬁdence and

Hope; (2) Willingness to Ask for Help; (3) Goal and Success Orientation; (4) Reliance on

34

Others; and (5) Not Dominated by Symptoms (Corrigan, Salzer, Ralph, Sangster & Keck,
2004). The study demonstrated that satisfaction with network supports was signiﬁcantly
related to overall network size and family network size. Network support was also related
to the recovery dimensions of Personal Conﬁdence and Hope, Willingness to Ask for
Help, Goal and Success Orientation, and Reliance on Others. Family network support
was signiﬁcantly related to Reliance on Others, but failed to reach signiﬁcance with other
recovery dimensions. However, network support size of friends and professionals
correlated with all the recovery domains. The study also failed to ﬁnd a relationship
between network mutuality (extent of relationship reciprocity) and recovery domains.
This study is important for several reasons. First, it uses a structured social network
measure to assess the number and size of networks, as well as gather information about
the degree of reciprocity between the focal person and network members. Second, it
examines the correlations between speciﬁc network clusters and recovery factors. A
limitation of the study is that it equates ‘network size’ to ‘support’ suggesting that each
network member is perceived to provide the same level of support to the focal person.
Creating a culture of healing and recovery is the current trend for psychosocial
services by increasing best practices around the goals of recovery (Anthony, 2000; Frese,
Stanley, Kress, & Vogel-Scibilia, 2001; Torry, & Wyzik, 2000). At the core of the
recovery movement in mental health, there is a focus on social, functional, and growth-
potential aspects of the individual. In contrast, the medical model relies on pathology and
treating the symptoms instead of the whole person. Recovery from mental illness, does
not indicate a ‘cure’ from the illness; instead individuals report attitudinal changes, or a

return to social and cognitive functioning. Clubhouses can serve as a place for individuals

35

to lead more productive, community oriented lives. The clubhouse program was
designed to address the needs of people with chronic or persistent mental illness who
have encountered a sense of loss in skills, friendships, family connections, and
employment (Mastboom, 1992). Clubhouse program encourage participation in the
operations and daily maintenance of the clubhouse. Members are encouraged to work in
various ‘units’ of the clubhouse with other members, such as the kitchen, or clerical units.
Social activities typically occur after hours. Although a signiﬁcant element of the
clubhouse program is to facilitate skill development, social interaction and social
activities play an important role in clubhouse culture. In fact, the number one reason
clubhouse members come to the clubhouse is to socialize (Herman, Onaga, Ferguson,
Pemice-Duca, Oh & Weaver Randall, 2002). The clubhouse model encourages
engagement in philosophy of the clubhouse through participation in the work of the
clubhouse with club peers as well as promoting social interaction. The main domains of
clubhouse programs consist of: social (e.g., recreation, support groups, outreach, psycho-
education, social activities), vocational (e.g., supported, transitional, competitive, peer-
run employment programs), educational supports, residential (e.g., housing information,
supported housing options, clubhouse apartments), medical (e.g., medication
management with nurses or community based psychiatrists), and ﬁnancial services (e. g.,
member bank). It is unclear whether or not social interactions at the clubhouse increase a
club member’s social networks supports, and such reciprocal interaction generalizes to

relationships beyond the clubhouse doors.

36

Social Networks, Recovery, and Clubhouses

Social network support has a history of being incorporated into a variety of social
intervention programs ranging from health care promotion to adolescent parenting to
delinquency prevention (Whittaker & Garbarino, 1983). Peer support has been proposed
as a key ingredient in programs for people with severe mental illness (Campbell &
Leaver, 2003). Programs designed to increase social networks among people with severe
mental illness do in fact achieve success (Davidson, Haglund, Stayner, Rakfeldt,
Chinman, & Kraemer Tebes, 2001; Wilson, Flanagan, & Rynders, 1999; Mowbray &
Tan, 1992).

Mental health services in the era following de-institutionalization have strongly
followed a social support framework of intervention, attempting to increase social contact
and engagement by increasing social network resources. In the 1970’s, mental health
policy in the US. utilized informal social networks and support systems as resources for
mental health patients transitioning into the community following long-term
hospitalization. From a policy and services standpoint, less reliance on formal
professional support systems and services helped to contain costs associated with
providing a continuum of care. In 1977, the National Institute of Mental Health
developed one of the ﬁrst national initiatives to utilize the social network research and
psychosocial rehabilitation services began assisting persons with chronic mental illness
with housing, daily living skills, employment and socialization opportunities (Turner &
TenHoor, 1978).

Positive and supportive mental health environments foster a greater likelihood of

nominating peers in the social network of consumers of mental health services (Hardiman

37

& Segal, 2003). The FRIENDS program, which is based on the philosophy that social
networks evolve from building a strong caring intentional community, has been found to
increase and maintain social networks over time to impact overall functioning (Wilson,
Flanagan, & Rynders, 1999. The central values of the FRIENDS program are recovery
oriented and recognizing mental health well-being has a direct relationship to the
involvement with others.

Clubhouse programs are being recognized as a valuable and cost effective option
to assist in the recovery process from mental illness (Landis, 1999). Social interaction
and activities play an important role in clubhouse culture. The clubhouse has been found
to foster cohesion, friendships, and feelings of belonging, otherwise known as a ‘sense of
community’ (Herman, Onaga, Pemice-Duca, Oh, Ferguson, 2005). In an unpublished
report to the Department of Community Health (Herman, Onaga, Ferguson, Pemice-
Duca, Oh & Weaver-Randall, 2002). Clubhouse members reported socializing and
making friends as the number one reason for attending the program. The clubhouse has
also been deemed a place where these friendships have lead to personal narratives of
recovery (Macias & Rodican, 1997). Creating a culture of healing is the current trend for
psychosocial services by increasing best practices around the goals of recovery (Anthony,
2000; Frese, Stanley, Kress, & Vogel-Scibilia, 2001; Torry & Wyzik, 2000).

Social network studies with clubhouse members have been limited and mixed.
Stein et al., (1999) found that clubhouse members nominated an average of 16 network
members, with an average of 5 friends, 7 family members, and 4 professionals. In a
separate study by Stein, Rappaport and Seidman (1995), clubhouse members identiﬁed

an average of 6 network members as sources of help or assistance, those of which 2.4

38

were family network members, 1.8 friends, and 1.6 professionals. Clubhouse members
also identiﬁed at least one conﬂictual relationship in their network. The study also found
that when reports about the quality of relationship matched a nominated family members’
report, clubhouse members exhibited higher social adjustment scores and lower
symptoms and distress. Further, clubhouse members who reported greater dissatisfaction
with family network relations experienced more severe psychiatric symptoms, and poor
social functioning, especially when the quality of the relationship was incongruent
between reporters. The authors contend that the lack of congruency in reporting between
clubhouse members and nominated family network members may be in part due to high
emotional reactivity. This suggests lower satisfactions with contact and relationships may
be indicative of lower levels of differentiation.

Another study with 34 clubhouse members, Perese, Getty & Wooldridge (2003)
found that clubhouse members reported fewer friendships and social network support as
compared to the general population of non-mental health consumers. Clubhouse members
were more likely to report friendships are more positive sources of support than family
members. Interestingly, the study found that clubhouse members who reported greater
sources of social support were also more likely to participate in self-help support groups
at the clubhouse.

Summary
The studies reviewed reﬂect the social networks characteristics among people
with serious mental illness and the relationship between social relationships and mental
health, speciﬁcally psychological well-being and recovery. There were few studies that

speciﬁcally described the social network characteristics of clubhouse members (Perese et

39

al., 2003; Stein, Rappaort & Seidman, 1995; Stein, Barry, Van Dein, Hollingsworth &
Sweeny, 1999), and only one study to date speciﬁcally examines social network support,
reciprocity, and network satisfaction with the recovery domains set forth by the Recovery
Assessment Scale (Corrigan & Phelan, 2004). This measure of a subjective experience of
recovery is used in this study.

Most social network studies using sample of people living with serious mental
illness employed small sample sizes and mostly qualitative. Cross-sectional studies
demonstrate a strong correlation between network dimensions and recovery type
outcomes. Longitudinal and controlled experimental designs demonstrated a causal link
between the lack of social network supports and increased psychiatric symptoms, poor
psychological functioning, and lower quality of life.

Clubhouse studies and other non-traditional mental health programs provide
evidence that people tend to seek emotional and social support that provide socialization
opportunities. These opportunities are hypothesized to improve overall social adjustment
by way of increasing self worth or afﬁrrning one’s social identity as someone other than a
mental health patient. According to Hirsch (1981) personal networks are conceptualized
as ‘personal communities’ that work to support healthy social identities to in turn
inﬂuence psychological well-being. One pathway to deﬁning our social identities is to
achieve meaningful participation in one’s community. The clubhouse program is an
environment that is designed to create, enhance, and maintain personal networks
composed of family, friends, club peers, and others. Therefore personal networks are

theorized to afﬁrm our social identity through aspects of support and mutual reciprocity.

40

Overall, as Goldberg, Rollins & Lehman (2003) stated, “whether considered
outcomes or predictors, clinical and psychosocial functioning are clearly related to social
network structure” (p. 399). These studies highlight the signiﬁcance of this research in
supporting existing programs designed to increase or improve social network relations,

and expanding our understanding of the role of social and family network supports to the

recovery process.

41

CHAPTER THREE

METHOD

The current study is part of the Flinn Clubhouse Project, funded in part by the
Flinn Family Foundation. The larger study employed a longitudinal design with one
follow-up period with numerous measures. By contrast, this study utilizes a cross-
sectional dataset collected from the ﬁrst interviews with clubhouse members. This
chapter describes the methodology for the Flinn Clubhouse Project and the present study.
Research questions are also presented.

Description of the Flinn Clubhouse Project

The Flinn project was conducted by a joint partnership between the
Dr. Esther Onaga of Michigan State University and Dr. Sandra Herman, of the Michigan
Department of Community Health. The original study examined 40 clubhouses in
Michigan, and sampled 18 clubhouses for in-depth clubhouse member interviews. The
criteria for selecting the 18 clubhouses included: 1) clubhouses were Medicaid enrolled,
2) clubhouse managers completed a clubhouse program assessment developed by the
Flinn Project researchers, 3) Clubhouse staff and members participated in a clubhouse
‘values’ Delphi survey developed by the Flinn Project researchers. The Values survey
assessed the importance of club values from members and staff. The survey was used to
validate the study’s logic model by identifying which components of the model
clubhouse members and staff believed to be important in a ‘very good clubhouse.’ The
values assessed were recovery, treatment choice and control, community and social

integration, skills and abilities, and partnerships between staff and members. Responses

42

were received for 313 members, 136 staff, and 40 managers responded to the survey.
Their responses were then aggregated to obtain a clubhouse rating of importance for each
value.

Eighteen clubhouses were selected based on the results of the Staff and Member
Values Survey to determine which clubs the Flinn Research Team was to visit.
Clubhouses were ranked from lowest to highest on the way they scored the values survey.
A sub-sample of 18 clubhouses were selected to represent value scores from low, middle
and high ranges, in addition to daily member attendance, total members employed,
location, and percent of members with schizophrenia. These clubhouses represent the
range in size, location, and clubhouse environments. Three of these 18 clubhouses were
excluded because they served as pilot sites to test the interview protocol. Thus a sample
of 221 participants from 15 clubhouses participated in this study.

Research Questions

Based on the review of literature, there are limited studies examining the social
network support dimensions of clubhouse members. Further, the lack of empirical
quantitative studies investigating the relationship between social network supports and
recovery warrant further examination. Variables selected for the models and the
hypothesized paths were based on a theoretical rationale that integrates the clubhouse
philosophy, social network research, and Feminist informed Bowen Family Systems
Theory. The purpose of this study was three-fold: (1) Describe the role of social
networks among clubhouse members; (2) Examine the relationship between social

network supports and recovery, and; (3) Test a path model that incorporates the overall

43

theory of the relationship between clubhouse engagement, the social network supports,
and the recovery process.
Research Questions

Research questions are divided into three sections covering social network
characteristics, social network and recovery, and social networks and clubhouses.
Social Network Support Characteristics

1. What is the size of the social network?

2. What type of relationships compose the network?

3. Do social network support dimensions (support, reciprocity, satisfaction with
relationship, importance of network relationship, and frequency of contact) vary
with age, gender, level of education, ethnicity, living arrangements, family
composition, level of functioning, clubhouse participation, and clubhouse sense of
community/ cohesion?

Social Network Supports and Recovery

4. Are social network support dimensions, (support, reciprocity, satisfaction with
relationship, importance of network relationship, and frequency of contact)
related to recovery and recovery domains?

a. Which social network support dimensions are more predictive of the
recovery?
b. Which network supports (e.g., family, friends, clubhouse staff, clubhouse

members, etc.) are more predictive of recovery?

5. Is there a positive relationship between greater family network supports (i.e.,
support, reciprocity, satisfaction with contacts, importance of family members on
the network, and frequency of contact) and recovery?

a. Are greater family network supports predictive of the recovery domains of
Personal Conﬁdence & Hope, Reliance on Others, and Goal & Success
Orientation?
b. What type of activities characterizes the interactions between clubhouse
members and nominated family members?
Social Network Supports and Clubhouses

6. What is the relationship between clubhouse engagement, as measured by level of
clubhouse participation and extent of clubhouse sense of community/cohesion,
social network supports (size, support, reciprocity, satisfaction and contact), and
recovery?

An exploratory path analysis was used to examine whether coming to a psychosocial
clubhouse and developing mutual and reciprocal relationships with others is related to
greater social networks, more satisfying contacts with others, and greater support and
reciprocity with network members, which then is related to greater sense of recovery.
Conceptual Models
The relationship between social network support dimensions and recovery from
mental illness among clubhouse users was examined. The study incorporated a human
ecology framework to study social networks and integrated Feminist informed Bowen

Family Systems Theory. Figures 3.1 through 3.4 display the relationships among the

45

variable of interest in this study. Speciﬁcally, the relationships between predictor

variables and criterion variables are presented in graphic format.

 

Perceived
Network
Support

 

 

Perceived
Network
Reciprocity

 

 

 

 

Network Size Recovery

 

 

 

 

Network
Contact

 

 

Network
Satisfaction

 

 

 

Figure 3.1. Model A: Total Network Support Predicting Recovery

46

 

Family Network
Size

 

 

 

 

 

Friendship Recovery
Network Size

 

 

 

 

 

 

Clubhouse
Network Size

 

 

 

 

Clubhouse
Member
Network Size

 

 

Professional
Network Size

 

 

 

Figure 3.2. Model B: Network Members

A series of regression models examining the relationship between personal networks and
recovery dimensions were performed. Regression Model A (Figure 3.1) hypothesized that
increased network support, network reciprocity, network contact, network satisfaction,
and importance are predictive of increased recovery scores as measured by the Recovery
Assessment Scale (Corrigan, Giffort, Rashid, Leary & Okeke, 1999).

Regression Model B (Figure 3.2) examines which group of network member
contributes to overall recovery scores. Regression Model C (Figure 3.3) examines the

extent of family network variables and the relationship to recovery domains such as

47

Personal Conﬁdence and Hope, Goal and Success Orientation, and Reliance on Others.
These domains are suggested to be indicators of greater development of self or

differentiation among individuals living with a chronic mental illness.

 

 

 

 

 

 

 

 

 

 

 

 

 

Recovery
Domains
Family Network
Size '\
Perceived Family Personal
Network Support Conﬁdence &
Hope
Perceived Family
Network Reliance on
Reciprocity Others
Family Network
Contact Goal & Success
Orientation

 

 

 

 

 

 

 

 

Family Network
Satisfaction ___/

 

 

 

Figure 3.3 Model C: Family network predicting recovery domains

48

 

Perceived
Support

 

 

 

Perceived
Reciprocity

 

 

 

 

 

 

Clubhouse Sense of Recovery
Participation H Community

 

V

 

 

 

 

 

 

 

 

 

Network Size

 

 

 

 

Network
Satisfaction

 

 

Figure 3.4 Initial path-analytic model: Inﬂuence of clubhouse engagement on social
networks and changes in recovery.

The path model in Figure 3.4 hypothesizes that clubhouse participation leads to
increased sense of community and cohesion among clubhouse peers, which consists of
developing relationships based on mutual support. Club community and cohesion is
hypothesized to be related to opportunities to increase network size, support, as well as
engage in more satisfying and reciprocal relationships with others. The presence of larger
networks, greater levels of support and reciprocity, satisfying contact with network
members, and increased contact with network is then hypothesized to be related to greater

level of recovery.

49

Corrigan & Phelan (2004) conducted Pearson Product Moment correlations
between social network size and recovery domains also used in this study. Their analysis
with over 1,000 mental health consumers, did not measure the quality of network support.
It was implied by the number of network members nominated. This study utilized both
number of network supports nominated as well as the level of perceived support received
and provided to each network member. The path model incorporates elements of
clubhouse philosophy that are assumed to foster greater network size that and sources of
support. The initial path model hypothesized greater clubhouse participation is related to
greater sense of community and cohesion among clubhouse members, which is a measure
of mutual support, sense of belonging, and afﬁliation with peers. Having a sense of club
community and cohesion then is hypothesized to be related to greater overall social
network size, network satisfaction, perceived network support and reciprocity, which is
then related to greater recovery.

Sample

Participants for this study were clubhouse members from 15 clubhouses in
Michigan who participated in the Michigan Flinn Clubhouse Project. A total of 221
clubhouse members volunteered for an in-depth, face to face structured interview.

A power analysis (Cohen, 1992) with an alpha level at .05, power at .80, and medium
effect size (.30) and 8 predictors variables requires a minimum of 107 participants. The
project team contacted the clubhouses to arrange for an on-site visit to conduct interviews
with clubhouse members. Up to 15 interviews were allotted for each clubhouse, however,
the number of participants who volunteered from each clubhouse ranged from 10 to 17

members to yield an average of 14 volunteers from each site. Seven clubhouses had 15

50

volunteers, three clubs had 16 volunteers, and ﬁve clubhouses had 10 to 14 volunteers.
To be eligible for clubhouse membership, one is required to have a psychiatric diagnosis.
Interviews were excluded from consideration if the interviewee was severely cognitively
impaired and unable to complete or comprehend the interview protocol. Data for the ﬁrst
interview was collected from August, 2000 through February, 2001.
Procedures

This study utilized a cross-section of the data from the original longitudinal
design. Clubhouse participants were self-selected volunteers. In keeping with the tenets
of clubhouse values, permission to recruit clubhouse members was obtained from the
clubhouse membership through consensus at their general house meeting. A detailed
letter describing the project and purpose of the in-depth interview was mailed to each of
the 18 clubhouses. Letters described the nature of the interview, the potential for a
follow-up interview six months later, and a ﬁnancial compensation of $20. Letters were
addressed to clubhouse members, staff, and the clubhouse manager (Appendix A).
Clubhouses interested in participating notiﬁed the research team by fax or telephone to
indicate their voluntary participation. At that time, a date was also selected for the ﬁrst
interview visit. Clubhouses were provided ﬂyers advertising the project. Clubhouse
managers were provided a volunteer sign up sheet with 15 interview slots and 5 waiting
lists positions for the pre-scheduled interview date. Clubhouse managers distributed
"consent to contact" forms to potential clubhouse interviewees so that a researcher team
member could contact them for an interview at different time or date. Interviewees who
signed up for an interview and could not attend the pre-scheduled interview date were

asked to complete and submit a ‘consent to contact’ form to the research team to a

51

schedule a separate interview time. Club managers assisted in gathering information
about potential interviewees to increase likelihood of representation on diagnosis,
employment, and length of membership. A waiting list with alternates was used as in the
case that scheduled interviewees failed to show for an interview. Interviews were
conducted in a private area at the clubhouse or at nearby facility. Each interview was
approximately one hour in length.

The project guaranteed participants that all interview information would be kept
strictly conﬁdential and they would be paid for their participation. Consent from each
participant was obtained prior to the start of the interview (Appendix B) and interviewees
were compensated $20.00 for each completed interview. Participants with legal guardians
were required to obtain consent from their guardian prior to scheduling an interview with
the research team (Appendix C). Clubhouses were compensated $100.00 for hosting the
on-site interviews for the day and assisting in recruitment procedures. All study
procedures for the larger study and the current proposal were approved by human
subjects review committee at Michigan State University (Appendix D). Interviews were
assigned a site ID number followed by an individual ID number to replace names. A
master list of project ID names with corresponding names was kept by the principal MSU
investigator in a locked password protected computer ﬁle.

Interviewers were instructed to maintain notes pertaining to the quality of the
interview, questions that arose during the interview, and observations related to the
participant’s interview behavior (e.g., whether participants were having difficulty

completing particular questions or sections of the interview, difﬁculty with

52

comprehension, cognitive impairments, or psychotic symptoms interfering with
completion) . Incomplete or incoherent interviews were excluded in the ﬁnal database.
Measures

A one-hour structured interview included the following content areas: 1)
clubhouse participation, 2) relationships with staff and members, 3) employment, 4)
social support networks, 5) health & medications, 6) history of mental illness, 7) mental
health service use, 8) extent of daily functioning, 9) sense of recovery, 10) sense of
community, 11) staff relationships, and 12) demographic information. The interview
protocol combined established survey instruments with questionnaires developed by the
Flinn Project researchers to address speciﬁc aspects of the clubhouse. An overview of

measures and data sources relevant to this study is provided in Table 3.1.

53

Table 3.1

Measures Included in Current Study.

 

 

Construct Measure Variables

Demographics Demographic Questions Gender
Age
Ethnicity

Clubhouse Participation

Clubhouse Cohesion

Social Network

Clubhouse Participation Section

Sense of Community (Buckner,
1988)

Social Network Questionnaire
(Herman, 1997)

Level of education
Marital status

Family composition
Diagnosis

Level of Functioning
History of Hospitalization
Living arrangements

Clubhouse attendance (# of
days by # of hours per day)

Total Score

Network Size

Network Composition
Frequency of contact with
members

Perceived Support
Perceived Reciprocity
Relationship Satisfaction
Relationship Importance of
relationship

 

Recovery Recovery Assessment Scale Total score
(Corrigan, Giffort, Rashid, Leary & 5 Recovery Factor Scores
Okeke, 1999).

Demographics

Demographic Questions. Information on demographics consisted of the following: age,

gender, ethnicity, and level of education, living arrangements, diagnosis, level of

disability, and history of hospitalization (see Appendix E). Diagnostic information was

54

obtained through the clubhouse manager’s client ﬁles. A release of information was
signed prior to the interview by each participant to release information pertaining to
diagnosis contained in their membership ﬁle (Appendix F). The level of one’s disability,
was assessed using the Social Functioning Scale (Birchwood, Smith, Cochrane &
Wetton,l990). This measure has 13 questions pertaining to an individual’s perception of
their ability to function independently. The scale was initially designed to measure the
efﬁcacy of family interventions with people with schizophrenia. Results from a validity
and reliability study indicated the measurement of skill and behavior was relevant to the
population and is a reliable and valid instrument. Questions include: “How well are you
able to... use public transportation, budget money, cook for yourself, take care of
personal hygiene?.” The measure was adapted to the Flinn Project. Additional response
categories were created to delineate the extent to which clubhouse members were able to
function independently or with assistance. The measure was scored using the following
categorical response scale: 1) able; 2) able with club help; 3) able with other help; and 4)
need help not currently receiving.

Clubhouse members who participated in this study were similar to the general
population of clubhouse members in Michigan for study year of 2000 (Michigan

Department of Community Health 2000; see Table 3.2).

55

Table 3.2

Statewide Demographic Data on Clubhouse Members from the Michigan Department of
Community Health Funding Year 2000 compared to Flinn Project Participants Project

 

 

Year 2001
State Data on Flinn Project
Clubhouse Clubhouse
Members Members
Demographic Variables Categories (N = 3,613) ( N = 221)
Age in Years Range 17 to 89 21 to 66
Mean 44.08 43.30
Standard Deviation 1 1.35 9.92
Gender Male 52.7% 46.6%
Female 47.3% 53.4%
Ethnicity Native American 0.7% 1.8%
Asian Paciﬁc Islander 0.5% 0.0%
African American/Black 13.0% 10.4%
White 83.5% 81.9%
Latino 1.7% 0.5%
Multi-racial 0.5% 4.5%
Arab American 0.1% 0.5%
Living arrangement Homeless 3.3% 0.0%
Living with family — dependent 23.9% 1 1.7%
Living alone, with spouse, or non- 37.0% 49%
relative
Foster family home 4.9% 0.0%
Specialized residential home 9.1% 1.0%
General group home 14.9% 17.2%
Prison, jail, juvenile detention 0.4% 0.0%
Nursing care facility 0.6% 0.0%
Institutional setting 1.3% 0.0%
Support independence program (SIP) 2.3% 11.5%
Employment status
Unemployed 27.2% 61.5%
Educational status Less than high school 24.2% 17.2%
High school GED 70.0% 39.8%
In school K-12 4.3% NA
In training program 0.7% 4.1%
Special education 0.9% NA
Income <$5,000 14.0% 9.2%
$5,000 to $9,999 68.9% 68.6%
$10,000 to $14,999 13.7% 15.9%
$15,000 to $19,999 2.6% 2.3%
>=$20,000 3.8% 2.3%
Axis 1 Diagnosis Schizophrenia & other psychosis 59.6% 53%
Mood disorders 28.6% 32.9%
All other 11.7% 14.2%

 

56

Clubhouse Participation. Two questions pertaining to clubhouse participation
comprised the ‘participation’ variable (Appendix G) . Participants were asked to estimate
the number of days they attended the clubhouse per week. They were also asked to report
how long they stayed at the clubhouse during their visit. A numeric value was derived
by multiplying number of days attended the clubhouse by number hours per day.

Social Network Interview (adapted from the Substance Abuse and Mental Illness
Project; Herman, 1997). Measures of clubhouse members’ social networks was based on
the social network analysis approach of McCallister and Fischer (1983). This measure
has been used as a valid measure in studies employing psychiatric populations (Ribisl,
1995). The Social Network Interview methodology uses probe questions to facilitate the
nominations of network members, which has been a recommended method for research
with people with schizophrenia and other special populations (Phillips, 1981, cited in
Ribisl, 1995).

Four probe questions were used to elicit personal network members. The ﬁrst
question was “When you are concerned about a personal matter- - for example,
something you are worried about or you are concerned about someone you are close to - -
who do you talk with?” Three additional questions followed: “Who do you spend your
time with, that is- who do you hang out with?”; “who would you ask if you needed to
borrow some money?” and “is there anyone else important in your life who you have not
mentioned?”

The social network method begins by asking each respondent the ﬁrst probe

question and then writing down the ﬁrst name and the ﬁrst initial of the last name on the

57

network list. Responses were recorded for this question until the respondent was finished
and was prompted with “is there anyone else. . .?” The same procedure was used for
questions 2 to 4. After enumerating the network list, respondents were asked to complete
the demographic information for each nominated network member (i.e., sex,
relationship), type of interaction (i.e., “what types of things do you & __ do
together”), frequency of contact, how important the person is to the respondent, how
satisﬁed the participant is with their relational contact, how much support the network
member provides, and how much support the participant provides to the network member
(see Appendix H for network measure).

Social network characteristics compiled from this network measure include:
network size (number of distinct people enumerated on the list), composition (clusters or
groups of type of people, e.g., family, friends, professionals), type of interaction (e.g.,
what do you do with this person?) , frequency of contact (1: yearly or less, 2 = few
times a year, 3 = monthly, 4 = weekly, and 5 = daily), satisfaction with relationship
contacts (1 = not at all to 5 = extremely satisﬁed), relationship, importance (1 = not all
important to 5 = extremely important), extent of perceived support ( 1 = none to 5 =
extremely), and extent of support provided (i.e., reciprocity, 1: none to 5 = extremely) .

Social network size was represented as the total number of social network
members nominated, up to 12. If a participant reported they had no network members to
offer support, they were retained in the sample to represent lack of a support network.

Recovery. The Recovery Assessment Scale (RAS) (Corrigan, Giffort, Rashid,
Leary & Okeke, 1999). The Recovery Scale was used to measure the extent to which

clubhouse members perceived a sense of recovery from mental illness (Appendix I). The

58

Recovery Scale is a 41-item questionnaire with a 5 point Likert scale assessing extent of
agreement (1 = strongly disagree to 5: strongly agree). Items reﬂecting recovery as a
psychological construct include: “ I can identify what triggers the symptoms of my
mental illness, ” “Fear doesn’t stop me from living the way I want to,” “I can handle it if
I get sick again.”

The development of the RAS was based on the analysis of narratives among
people with severe mental illness and their recovery process. Overall, the Recovery scale
yielded test —retest reliability at r = .88 and Cronbach’s alpha = .93 for internal
consistency (Corrigan, Giffort, Rashid, Leary & Okeke, 1999). The RAS has been found
to be positively associated to indicators of well-being, such as quality of life, social
support, empowerment, and self-esteem, but inversely related to psychiatric symptoms
and age (Corrigan et al., 1999). Corrigan, Salzer, Ralph, Sangster & Keck, (2004)
examined the factor structure of the RAS through a conﬁrmatory factor analysis with a
sample of 1,824 participants from the national Consumer Operated Services Project. The
study revealed that recovery is a multidimensional construct composed of ﬁve factors,
which correlate with other measures of psychological functioning and symptoms.
Cronbach’s alphas ranged between .74 and .87. The factors closely match those found in
an earlier unpublished report by the Flinn Project (Herman, Onaga, Ferguson, Pemice-
Duca, Oh & Weaver-Randall, 2002). RAS factors are based on 24 of the 41 scale items.
RAS factors are: Personal Conﬁdence and Hope (or = 0.87), Willingness to ask for help

(or = 0.84), Goal and success orientation (or = .82), Reliance on others (or = .74), No

domination by symptoms (or = .74) (e. g., “I understand how to control the symptoms of

mental illness”).

59

To determine the overall recovery, a total score was calculated for each
participant. The greater the score, the greater the subjective attitude of recovery.

Sense of Community/Cohesion Scale (Buckner, 1988). Sense of community,
which is one concept of cohesion, has been found to be a core element of the clubhouse
model (Herman, Onaga, Pemice-Duca, Oh, Ferguson, 2005). Sense of community is
assumed to facilitate the development of social networks within the clubhouse (Appendix
J). To measure the extent to which individuals felt as though they belonged to the
clubhouse, attained mutual relationships, were accepted, and perceived themselves as
part of a clubhouse culture was measured with the Sense of Community/Cohesion Scale
developed by Buckner (1988). The original instrument was developed to measure
neighborhood cohesion (Neighborhood Cohesion Instrument, NCI Buckner, 1988) and
consisted of three scales measuring attraction to the neighborhood, neighboring
characteristics, and psychological sense of community. The original version was
consisted of 39 items with test retest reliability of r = .80 and internal consistency of .97.

Flinn project researchers utilized the Sense of Community Scale to understand
clubhouse community from the perspective of the clubhouse member. Eleven items from
the original Buckner scale and 5 items were developed by the Flinn project team from a
previous clubhouse concept mapping study (Herman, Onaga, Pemice-Duca, Oh,
Ferguson, 2005) were included to reﬂect speciﬁc items to the clubhouse community and
cohesion. Statements included: “I feel like I belong to this clubhouse”, “Being part of the
clubhouse helps me deal with mental illness”, “Being a member of this clubhouse helps
me have hope for the future.” A principle components analysis revealed a two structure

instrument, comprised of Beneﬁts of Membership and Recovery, and Sense of

60

Community and Fellowship. The internal consistency of the measure is .91. A total score
is the sum of all item responses; the greater the score, the greater sense of community.

The resulting clubhouse sense of community scale contains 15 statements,
measured on a 5 point Likert scale. Respondents were asked to indicate the extent to
which they agreed or disagreed with each statement.

Data Analyses

The analyses involved four main components: (1) descriptive statistics of
clubhouse members, (2) descriptive statistics of the social networks of clubhouse
members (3) two separate sets of multiple regression analyses using aggregated social
network variables and recovery (4) three separate multiple regression analysis employing
family network support variables and three different recovery factor dependent variables,
(5) one path analytic model hypothesizing relationships between clubhouse participation,
cohesion, social networks, and recovery. Path analysis is an extension of multiple
regression, except it is used to predict to more than one dependent variable
simultaneously. The path coefﬁcients for the model were determined using SPSS AMOS
statistical program. The program generates indices for model ﬁt. Demographic variables
that signiﬁcantly correlate with the dependent variables were entered ﬁrst as covariates in

the model.

61

CHAPTER FOUR
RESULTS

Several analyses were conducted to explore the relationship between social
networks and recovery. A series of multiple regression analyses were performed between
social network predictors and recovery. A path model exploring the relationship between
clubhouse engagement, social network support dimensions, and recover was analyzed
using SPSS AMOS 5.0. First, descriptive statistics on clubhouse member demographic
variables and social network variables were computed. Second, the results of multiple
regression analyses are presented. Finally, an initial hypothesized path model was
explored. An alpha level of .05 was used for all statistical tests. An alpha level of .01 was
used for any demographic variables that co-vary with the criterion variable recovery and
recovery domains.
Sample Characteristics

A total of 221 participants from 15 clubhouses participated in an interview
between August 2000 and January 2001. Sample characteristics are similar to clubhouse
demographic statistics collected on 3,613 clubhouse members by the Michigan
Department of Community Health during the 2000 funding year (see Table 3.2 in Chapter
3). Table 3.2 shows that the sample used in this study is representative of clubhouse
members in the state. Table 4.1 summarizes clubhouse member demographics.

The sample was comprised of slightly greater number of women than men, 53.4%
vs. 46.6% respectively. Participants ranged in age from 21 to 66 years, with an average

of 43 years. About a third of the clubhouse members were between the ages of 36 and 45

62

years (34.1%). The majority of the participants were Caucasian (81.9%). African
Americans were the second largest group (10.4%). Statewide data on clubhouse
demographics demonstrate a slightly greater percentage of African American clubhouse
participation (13%) than the current sample. Asian Americans were not represented in the
current sample, and comprise of .05% of clubhouse membership statewide. Level of
income for over 50% of the sample ranged between $6,000 and $9,000 per year, while
26% reported income between $9,000 and $25,000. A portion of the participants were
living on less than $6000 per year (17.8%). The main source of income reported was
from Social Security (93.2%).

The majority of clubhouse members never married (80%), while 11.8% indicated
having a spouse or a live-in partner. Divorced, separated, or widowed clubhouse
members constituted 8.2% of the sample. A third of clubhouse members reported having
children (33.5%), with 16.4% of the sample as parents of children under the age of 18.
Family composition and marital characteristics are similar to other studies involving
clubhouse members (Perese, Getty, & Wooldridge, 2003). Nearly a fourth of the sample
reported living with family members, while 75% reported living alone or a non-relative
roommate. Overall, 69.7% of clubhouse members were living in independent living
arrangements. Clubhouse members were typically unemployed (61.5%). A majority of
clubhouse members completed high school or experienced some college (66.9%).

The sample consisted mostly of clubhouse members diagnosed with
schizophrenia and other psychotic disorders (52.5%), followed by mood disorders
(32.6%). A total of 180 (81%) participants reported hospitalization data. Participants

recalled the year in which they were last hospitalized. On average, clubhouse members

63

reported 5.6 years since their last psychiatric hospital admission. Number of years since

last hospitalization was calculated from the year the interview was conducted.

Table 4.1

Participant Demographics (N = 221)

 

Variable Categories % of Number of Cases
sample
Gender Male 46.6% 103
Female 53.4% 118
Ethnicity Caucasian 81 .9% 1 81
African American 10.4% 23
Latino .5% 1
Arab American .5% 1
Multi-racial 4.5% 10
Native American 1.8% 4
Other .5% 1
Age (M = 43.3)
21 - 35 21.3% 47
36 — 45 34.1% 82
46 - 56 28.7% 69
57 - 66 10.6% 23
Level of Education Less than High School 17.2% 38
Diploma, GED 39.8% 88
Some College, less than degree 27.1% 60
Associates Degree or Certiﬁcate 9.0% 20
Program
Bachelor’s or Master’s Degree 6.8% 15
Religious Afﬁliation Christian 83.1% 182
Jewish .9% 2
Muslim .5% 1
Agnostic .5% 1
Other 5.4% 12
None 9.5% 21
Employment Status Employed 38.5% 85
Unemployed 61.5% 136
Residency Status Lives with Family 25.3% 56
Lives with Roommate(s) 29.4% 65
Lives Alone 45.2% 100
Living Arrangement Private Residence 70.7% 156
Group Home 17.2% 38
Residential Treatment Setting 1.0% 2
Supervised Home or Apartment 11.5% 25
Who Participant Lives With Lives with Family of Procreation 10.4% 23
Lives with Family of Origin 14.5% 32
Lives Alone 46.2% 102
Lives with Roommate 29.4% 65

 

65

Table 4.1 cont,

 

Variable Categories % of Number of Cases
sample

Marital Status Never Married 80.1% 177
Spouse/Live in Partner 11.8% 26
Divorced/Separated 7.7% 1 8
Widowed .5% 1

Primary Psychiatric Mood Disorder 32.9% 72

Diagnosis
Schizophrenia & Related 53.0% 1 16
Other Axis I & II Diagnoses 14.2% 31
Declined to disclose .9% 2

 

Level of the functioning measured by the Social Functioning Scale (Birchwood,
Smith, Cochrane, & Wetton, 1990) assessed ‘how able’ clubhouse members performed
daily living tasks. This variable was used as a possible covariate to recovery. The lowest
possible score is 13 with a maximum of 39 on the scale. Results of the scores ranged
from 15 to 39 with a mean of 33.5, a mode of 38 and a standard deviation of 4.6. A
frequency distribution revealed that most clubhouse members have sufﬁcient independent
living skills and appeared not to be hindered by their disability. Because of a lack of
sufﬁcient range, this variable was removed as possible covariate in the subsequent
analyses.

Clubhouse members attended the clubhouse an average of 18 hours a week. The
median number of years of clubhouse membership was 3 years, ranging between less

than 1 year and 28 years.

66

Descriptive Characteristics of Social Networks
Research Question I & 2: What is the size of the social network? What type of
relationships composes the network?

To address the ﬁrst set of research questions of the study, descriptive statistics
were computed to calculate the percentage of family, friends, and others nominated on
the social network interview. The relationship of the nominated network member to the
focal person was originally coded in the following 10 categories on the Social Network
Questionnaire: ( 1) Family member, (2) friend, (3) neighbor, (4) professional, (5) club
staff, (6) club member, (7) co-worker, (8) group home member, (9) guardian, (10) school
friends, and (11) other. Research team members were required to specify ‘other’ types of
the relationship if it did not fall within one of these 10 categories. Additional relational
categories that emerged included payees, and pastors or clergy. Frequency distributions
revealed that clubhouse members rarely nominated guardians (.4%), payees (.4%) group
home peers (1.6%), and co-workers (2.7%), neighbors (3.1%) pastors, and clergy (4.6%).
School friends were non-existent on the network. The low frequency distributions
allowed for type of relationship to be collapsed into ﬁve main categories: (1) Family, (2)
Friends, (3) Professionals, (4) Clubhouse staff, (5) Clubhouse members. Relationships
coded as ‘Professionals’ consisted of counselors, therapists, medical professionals (e.g.,
nurses, doctors, psychiatrists), church clergy, as well as adult foster care staff. Friends
consisted of roommates, neighbors, co-workers, and general friendships and
acquaintances in the community. Family consisted of family of origin and procreation,

ﬁancés, signiﬁcant others, and legal guardians.

67

Descriptive statistics on the composition of the social network were performed.
Means and percentages of network relationships are presented in Table 4.2. The
calculation of means and standard deviations for the number of network members
nominated in each cluster does not include the value of ‘0’ for ‘no’ network members.
These calculations were based on the number of members nominated on the network.
The percentage of family, friends, clubhouse peers, and professionals are also
summarized. The social networks of clubhouse members were fairly small, with an
average of ﬁve nominations. Only two participants were identiﬁed as isolates, with no
social network nominations reported. In contrast, 22 (9%) clubhouse members reported
having networks composed of 10 or more people. Over 97% of the participants identiﬁed
at least one or more members on their social networks. Family members were the largest
group to compose social networks (76%). The second largest group was friends with
49% of network composition. Clubhouse staff ranked as the third largest group (42%)
followed by professionals (34%), and ﬁnally clubhouse members (28%). Unexpectedly,

clubhouse members made up the smallest composition of the network.

68

Table 4.2

Composition of Social Network*

 

 

 

 

Variable Value Frequency % Mean i SD/
Range
Family 76% 2.6: 1.7 (1— 8)
# of Family 0 52 23.5
Members
1 54 24.4
2 45 20.4
3 32 14.5
4 15 6.8
5 8 3.6
6 or more 15 6.8
Friends 49% 2.1: 1.5 (1- 9)
# of Friends 0 1 1 1 50.2
1 47 21.3
2 33 14.9
3 13 5.9
4 7 3.2
5 4 1.8
6 or more 6 2.8
Club Staff 42% 2.2 i 1.5 (1 - 9)
# of Club staff 0 128 57.9
1 41 18.6
2 28 12.7
3 5 2.3
4 12 5.4
5 or more 7 3.3

 

*Calculations of means and standard deviations do not include the value of ‘0’

69

Table 4.2 cont.,

 

 

 

 

Variable Value Frequency % Mean _-_1-_ SD/
Range
Professionals 34% 1.5 i .80 (1-4)
# of Professionals 0 144 65.2
1 48 21.7
2 18 8.1
3 9 4.1
4 2 .9
Club Members 28% 1.8 i 1.1 (1 — 5)
# of Club Members 0 158 71.5
1 37 16.7
2 14 6.3
3 6 2.7
4 2 .9
5 4 1.8
Total # of Network
Nominations 0 2 99% 5.0 i: 3.0 (1-12)
1 19 8.6
2 34 15.4
3 28 12.7
4 31 14.0
5 21 9.5
6 18 8.1
7 21 9.5
8 or more 47 21.3

 

*Calculations of means and standard deviations do not include the value of ‘0’

70

Social Network Dimensions & Club Member Characteristics
Research Question 3: Do social network support dimensions vary with age, gender,
level of education, ethnicity, living arrangements, family composition, level of
functioning, clubhouse engagement (level of participation and clubhouse
communi ty/cohesi on ) ?

Statistics were computed for respondents who reported the presence of at least
one member from the following network clusters: family, friends, professionals and
clubhouse staff, and clubhouse members. Statistics were also computed for total network
supports. One way ANOVAs were performed on social network support dimensions and
club member characteristic. AN OVAs were performed for all social network support
relationships (i.e., Family network, friend network, professionals, etc). Descriptive

statistics for total social network support dimensions are presented by network cluster in

Table 4.3.

Table 4.3

Means and Standard Deviations for Social Network Dimensions

 

 

Total Clubhouse Clubhouse
Network Family Friends Professionals Staff Members
(N: 221) (n: 169) (n=111) (n=77) (n=92) (n=63)
M/SD M/SD M/SD M/SD M/SD M/SD
Variables
Perceived Support 3.1/.82 3.1/.74 3.1/.75 3.2/.64 3.2/.63 3.2/.68
Perceived Reciprocity 2.7/1.06 2.8/.99 2.8/.96 2.6/1.0 2.8/1.0 3.0/.80
Importance 3.4/.73 3.4/.64 3.4/.59 3.4/.61 3.4/.65 3.3/.61
Contact 3.9/.82 3.8/.71 3.8/.72 4.0/62 4.1/.68 4.1/.54
Satisfaction 3.0/.92 3.1/.78 2.9/80 3.1/.76 3.2/.72 3.1/.64

 

71

Studies have shown that individuals diagnosed with schizophrenia and have
smaller social networks. Clubhouse members with schizophrenia are the largest
diagnostic group in the overall study, which is representative of the diagnostic
representation of clubhouse users across the state. Therefore to differentiate between
diagnostic groups, diagnostic categories were collapsed to form two distinct categories
(i.e., Schizophrenia and Related Disorders vs. Others) and dummy coded (1 =
schizophrenia, 0 = non-schizophrenia). There were 103 clubhouse members diagnosed
with schizophrenia and related disorders, and 116 members with mood disorders and
other type I and H diagnoses. Effect sizes for signiﬁcant independent variables are
reported. Independent variables include: diagnosis, gender (1: female, 0 = male), living
arrangements (1 = dependent, 0 = independent), ethnic group (1: minority, 0 = non-
minority), level of education (1 = less than diploma, 2 = diploma or equivalent, 3 =
greater than high school ), marital status (1: married, live-in partner, widowed,
separated/divorced, 0 = never married), and family composition (0 = no children, 1 =
have children). Pearson Product Moment Correlations were performed between
continuous demographic variables and social network variables. These variables include:
(1) age, (2) number of years since last hospitalization, (3) clubhouse participation (4)
clubhouse membership, (5) and level of functioning. Signiﬁcant results are summarized
below. An average of 1 to 3 cases had missing data on one or more independent or
dependent variable. Pairwise deletion of cases was performed for statistics.

No signiﬁcant differences were found between total network size and diagnosis, F
(I, 218) = 3.04, p =.08. Clubhouse members with schizophrenia rated their networks as

more important, F ( 1,218) = 4.15, p = .0, d = .27 small eﬁ’ect than club members with

72

other diagnoses. Club members with schiz0phrenia reported greater contact with overall
network members, F(1,218) = 6.96, p = .00, d = .36, small effect (Schizophrenia = 103,
M = 4.6/2. 7; Other = 5.4/3.3); contact with friends, F(1,219): 7.22, p = .00, d =.53,
medium effect(Schizophrenia = 5, M = 4.1/.51; Other = 60, M = 3. 7/.83 ),' family, F ( 1,
219) = 4.87, p = .02, d =.36 small effect, (Schizophrenia =70, M = 4.0/58; Other =98,
M = 3. 7/. 78); and clubhouse staff, F(1,219) = 5.30, p = .02, d = .48, (Schizophrenia =
45, M =4.3/.53; Other = 47, M = 4.0/. 78). However, these members had signiﬁcantly
smaller family networks, F(1, 219) = 7.22, p = .00, d = .36, small effect (Schizophrenia
= 103, M = 2.1/1.6; Other = 116, M = 2.6/1.3).

Greater satisfaction with overall network relationships were found among men
than women, F(1,215) = 5.85, p = .01, d = .3, small effect (Males = 103, M = 3.2/.23;
Females = 118, M = 2.9/1.9). Males also rated satisfaction with family contact higher
than females, F(1, 167) = 7.30, p =.00, d = .53, moderate effect (Males = 86, M =
3.2/.69; Females = 83, M = 2.8/. 83). However, the importance of friends on the network
was greater among females, F ( 1, 110) = 6.14, p = .01, d = .52, moderate effect (Males =
66, M = 3.2/.67; Females = 45, M = 3.5/.51).

The sample included a small number of ethnic minorities, which resulted in
unequal sample sizes for the analysis of variance. Given the problem of unequal sample
sizes, the Brown-Forsythe and Welch statistics were used to assess difference in social
network variables by ethnic group. These statistics test for the equality of group means
and is preferable to the F statistic when the assumption of equal variances does not hold.
The Brown-Forsythe and Welch statistics were all signiﬁcant. A total of 39 clubhouse

members who reported family on their networks were identiﬁed as ethnic minorities as

73

compared to 180 non-minorities. Ethnic minorities reported more frequent contact with
network members, F ( 1, 218), = 4.53, p = .03, d = .37. small eﬁ‘ect; professionals, F ( 1,
76) = 13.95, p = .00, d = 1.71, large eﬂect(Minority = 14, M = 4.5/.36; Other = 63, M =
3.8/.61),° and clubhouse staff, F ( I , 92) = 4.89, p = .02, d =.61, moderate effect(Minority
= 17, M = 4.5/.41; Other = 76, M = 4.1/. 71); greater family reciprocity, F ( I, 168) =
6.24, p=.01, d = .54, moderate effect(Minority = 27, M = 3.2/. 73; Other = 142, M =
2.7/1.0).

Level of education was divided into three groups: (1) less than high school
diploma, (2) high school diploma or equivalent, and (3) greater than high school.
Clubhouse members with a high school education or greater reported less reciprocity with
network members, F(2, 217) = 3.20, p = .04, [(1) =38, M = 3.1/.92; (2) = 87, M =
2.8/.99; (3): 93, M = 2.6/1.0]; less satisfaction with network contacts, F (2, 216) = 3.12,
p = .04[(1) = 38, M = 3.35/. 77; (2): 87, M = 3.10/81; (3) = 92, M = 2.95/86; and
nominated less clubhouse staff, F(2, 220) = 3.08, p =.04[(1) = 38, M = 1.44/21; (2 )=
88, M = .88/1.4; (3 ) =95, M = .74/1.2] as compared to club members with less than a
high school education.

Greater frequency of family contact was found among club members who were
ever married, F(1, 168) = 7.39, p = .00, d = .58, medium effect (N0 = 133, M = 3. 7/. 75;
Yes = 36, M =133, M = 4.1/.45). No social network differences were among club
members with children and those without children. Clubhouse members living
independently reported greater family contact, F( 1, 168) = 4.81, p = .03, d = .29, small
eﬂectﬂndependent = 115, M = 3.95/66; Dependent = 54, M =3. 70/. 79); and more

frequent contact with friends, F(1, 110) = 8.28, p = .00, d = .61, medium eﬁ’ect

74

(Independent = 79, M = 4.0/65; Dependent = 32, M = 3.59/80). Although club
members who were living independently reported greater contact with family, they were
more likely to report receiving less family support than club member members living in
dependent arrangements such as adult foster care home, residential centers, or with
family of origin, F(1, 168) = 5.22, p = .02, d = .37, small effect (Independent = 115, M
= 3.08/ 78; Dependent =54 = 3.35/61). Clubhouse member living independently were
also more likely to report less total network support, F ( I, 217) = 5.54, p = .01, d = .35,
small eﬂect (Independent = 152, M = 3.11/76; Dependent.=66, M = 3.36/63).
Correlations were performed among all social network variables and identiﬁed
continuous variables (i.e., age, clubhouse participation, clubhouse membership,
hospitalization, and level of functioning). Younger club members perceived greater
family support, r(169) = -.16, p = .03, and less likely to report professionals on their
networks, r(221) = .15, p = .02. Clubhouse members participated in the clubhouse more
often were more likely to report greater satisfaction with their network supports, r(221) =
.14, p = .03; report greater contact with staff, r( 92 ) = .35, p = 00; and greater contact
with professionals, r(77)= .35, p = .00. However, the length of one’s clubhouse
membership was not related to any social network variables. Clubhouse members who
were more recently hospitalized were more likely to perceive greater reciprocal relations
between themselves and clubhouse staff, r( 77) = -.26, p = .02. This ﬁnding may be
suggest that staff members attempted to reach out and include more recently hospitalized

club members in mutually supportive activities of the clubhouse.

75

Social Networks and Clubhouse Engagement

To examine whether greater clubhouse engagement was related to nominating
more clubhouse peers and more overall network members, an additional set of Pearson
Product Moment Correlations were performed with clubhouse participation, length of
clubhouse membership, and sense of clubhouse community/ cohesion variables with
total network size, and number of clubhouse staff and clubhouse members nominated on
the network. No associations were found between level of clubhouse participation, the
length of one’s membership and the overall size of the social network, r( 220) = .00, and
likelihood of nominating more clubhouse peers, r(220): .07. Further, the level of club
participation and length of membership were not associated with the number of
clubhouse staff or members nominated on the network. However, clubhouse sense of
community/cohesion was positively related to level of participation, r(221) = .19, p =
.00; and the number of clubhouse staff nominated on the network, r(221) = .24, p = .00.
Sense of club community/cohesion showed a moderate relationship with perceived
network support, r(221) = .3 7, p = .00; perceived reciprocity, r(221) = .44, p = .00;
importance of network members, r(221) = .26, p= .00; and satisfaction with network
relations, r(221) =.36, p = .00. Network contact was not related to sense of club
community/cohesion.

The relationship between club cohesion and family network support was also
explored among participants who reported family on their networks. The purpose was to
explore whether afﬁliations with the clubhouse was related to more satisfying contacts
with family network members, greater family reciprocity, and reliance on less support.

Club sense of community/cohesion demonstrated a modest positive correlation with

76

perceived family reciprocity, r(169)=.42, p =.00; perceived family support, r(169)=.36,
p = .00; and weak association with satisfaction with family contacts, r(169)=.32, p =
.00.
Recovery

Research Question 4: Are social network support dimensions, (support, reciprocity,
satisfaction with relationship, importance of network relationship, and frequency of
contact) related to recovery and recovery domains?

Means and standard deviations for the Recovery Assessment Scale are presented
in Table 4.4. Scores were greater in the recovery domains of Reliance on Others, Goal &

Success Orientation, and Willingness to Ask for Help.

Table 4.4

Descriptive Statistics for the Recovery Assessment Scale and Recovery Factors

 

Recovery Assessment Scale Dimensions

 

N = 220 Mean i SD
Personal Conﬁdence & Hope 3.87/78
Willingness to Ask for Help 4.10/82
Goal & Success Orientation 4.22/67
Reliance on Others 4.24/61
No Domination by Symptoms 3.63/95
Total Score 165.27/23.28

 

77

Correlations between recovery scores and selected demographic variables, and aggregate
social network support variables are presented in Table 4.5. The strength of these
relationships is similar to the correlations found in Corrigan & Phelan’s (2004) study.
Total network size demonstrated a weak relationship with Personal conﬁdence and Hope,
Goal and Success Orientation, Reliance on Others, and the total recovery score.
Perceived support and reciprocity demonstrated a modest relationship with the total
recovery score, and across all ﬁve recovery domains. Frequency of contact with network
members appeared to be least related to recovery domains, while greater satisfaction with
network contact was related to greater recovery. Social network support measures
appeared to have the least relationship with recovery domain of No Domination by
Symptoms. Family network size was not signiﬁcantly related to the total recovery score.
Weak relationships emerged between number of family supports and Personal
Conﬁdence & Hope, and Reliance on Others. However, the number of clubhouse staff
reported as supports was signiﬁcantly related to recovery, with modest relationships in all
recovery domains except Willingness to Ask for Help. A weak relationship emerged
between number of clubhouse members nominated and Reliance on Other, and total
recovery score. The presence of professionals nominated on the network did not
correlate with recovery or recovery domains. The number of friends nominated as
supports demonstrated a weak relationship with Personal Conﬁdence & Hope, and Goal

& Success Orientation, and overall recovery.

78

Table 4.5

Correlation of Member Characteristics, Clubhouse Engagement, Social Network

Supports by Recovery Dimensions, N = 220

 

 

Personal Willingness Goal & Reliance on No Total
Conﬁdence & to Success Others Domination Recovery
Hope Ask for Orientation by Score
Help Symptoms
Gender -.09 -.06 -.02 -.01 .05 -.06
Ethnicity .08 .04 -.03 -.05 .08 .03
Age .00 .04 -.08 -. 10 14"I -.01
Education -.20** -.08 -. 16* .02 -.03 -.1 1
Living Arrangements -.11 -. 16* -.07 -.15* -.1 l -.24**
D‘ignos‘s °.f -.18** -25" -.12 -.13* -. 10 22*
Schizophrenia
Clubhouse Participation .08 .03 .01 .03 .07 .05
Sense Of an: slur: scar: an: an: an:
Community/Cohesion .42 .28 .45 .47 .26 .50
Total Network Size .24** .12 .l8** .29** .12 .25**
Total Perceived Support .33* .28" .43" .37** .21** .41**
Tm“ PC'FC’VC" 42** 35** 43** 37M 18" 46**
Reciprocrty ' ' ' ' ' '
Total Network .14* .16* .23** .24** .02 .19**
Importance
Total Network Contact .08 .04 .10 .18** . 10 .13*
T0931 Neiwork .34** .26** .34** .3o** .16* .35**
Satisfaction
Family Network Size .13* .11 .04 .16* .06 .12
Friend Network Size .13* .12 .16* .10 -.02 .13*
SPFOfessmna' New“ -.03 -.06 -.05 .02 -.06 -.03
rze
Club Staff Network Size .26* .18 .28** .39** .33" .36**
Club Member Network .09 .01 .12 . 13* .1 1 .13*
Size

 

79

Table 4.6

Correlations among Social Network Support Clusters and Recovery Dimensions

 

Personal Willing- Goal& Reliance No Dom- Total

 

Conﬁ- ness to Success on Others ination by Recovery

f _ dence & Ask for Orientation Symptoms Score
3; a; Social Network Hope Help
8 2 Variable
Z O

Perceived Support .29** .22** .42** .33** .23** .39**

Perceived Reciprocity .39" .31** .43" .33** .17* .44**
3: % Satisfaction .31 ** .19* .29** .22” .20** .29**
g ‘E‘ Importance .10 .09 .18* .19* .01 .14
I“ = Contact -.00 -.2 .01 .06 .09 .06

Perceived Support .38“ .23** .44** .31" .13 .39“
m c: Perceived Reciprocity .38“ .25* .52** .24* .25“ .45"
E T Satisfaction .30* .16 .35* .20* .12 .26**
Li: :- Importance .09 .08 .24* .14 .04 .15

Contact -.08 -. 10 .03 .07 .01 -.01
“g N Perceived Support .26* .18 .28** .39" .33” .36“
t2 7" Perceived Reciprocity .39" .38" .41“ .38" .22** .45"
,2 I: Satisfaction .19 .22* .18* .36M .29” .27**
U Importance .23* .12 .27** .23* 17 .27**

Contact .02 -.01 .03 .10 . 15 .1 1
7: Perceived Support .23* .24* .21 .26* .31** .36**
E I; Perceived Reciprocity .44" .38** .42" .39" .30" .50**
g g Satisfaction .35M .22 25* .29** .17 .34M
5 Importance .12 .18 .14 .14 .15 .18

Contact .21 .04 .21 .10 .17 .23*
g Perceived Support .09 .19 .24 .09 .14 .20
.E, 8 Perceived Reciprocity .28* .43** .35* .03 .11 .32"
E II Satisfaction .24 .13 .19 .09 .16 .14
3 7 Importance .08 .15 .16 .08 .10 .10
0 Contact .00 -.01 .16 .18 .14 .17

 

80

Correlations between recovery scores and social network supports by cluster (e.g.,
family network, friends, professionals) are presented in Tables 4.6. A breakdown of the
quality of support by each network cluster provides a more complete view of which
aspect of the supportive relationship was related to greater recovery. Perceived support,
reciprocity and satisfaction were signiﬁcantly related to recovery among club members
who reported family, friends, club staff, and professionals in their networks. Clubhouse
staff emerged as important relationship in the network composition, and greater contact
with professionals was weakly related to recovery. Reciprocity emerged as the only
variable related to recovery and recovery domains among club members who nominated

other club members on the social network.

Research Question 4a: Which social network support dimensions are important to
recovery?

Standard multiple regression analyses were performed to test a series of models
between social network measures and recovery total score. Table 4.7 displays the
unstandardized regression coefﬁcients (B) and standard error of (B), the standardized
regression coefﬁcient ( [3 ), the R2, and adjusted R2 for Models A through (3. Social
network support variables served as the independent variables and recovery served as the
criterion. SPSS was used for all analyses and missing cases were deleted pairwise for all
regression analyses. Assumptions for regression analysis were examined and skewness,
outliers, and normality, linearity, and homoscedasticity of residuals were within normal

ran ge.

81

For the ﬁrst model (i.e., Network Support), total network size, perceived network
support, perceived network reciprocity, network satisfaction, and frequency of contact
with network members were entered as predictor variables to predict recovery total score.
R for regression was signiﬁcantly different from zero, F (6, 219) = 19.29, p <.001.
Three of the six predictors contributed signiﬁcantly to the model. Total network
reciprocity was slightly more important to predicting recovery total score. Altogether,
29% of the variability in recovery scores was predicted by knowing the scores of these

three variables.

Research Question 4b: Which network supports (e.g., family, friends, clubhouse staff,
clubhouse members, etc.) are more predictive of recovery?

A regression analysis was performed with predictors in Model B and recovery
total score. The total number of family members, friends, professionals, clubhouse peers
was entered into the model as independent variables. This analysis included the entire
sample of participants who reported a presence or lack of support from any of these
network clusters. R for regression was signiﬁcantly different was zero, F (5, 219) = 8.43,
p < .01. The number of clubhouse staff support nominated on the social network was
more important to recovery than other relationships. The number of supportive
friendships emerged as the second most level of support. Although the model is
signiﬁcant, the size of network support by cluster only explains less than 10% of the

variance in recovery.

82

Research Question 5: Is there a positive relationship between greater family network
supports (i.e., support, reciprocity, satisfaction with contacts, importance of family
members on the network, and frequency of contact) and recovery?

To test the notion that greater family network supports (i.e., Model C)
is indicative of the recovery process, a standard multiple regression was performed.
Family network size, perceived family support & reciprocity, satisfaction with family
contacts, and family network contact were entered as independent variables. R for
regression was signiﬁcantly different was zero, F(6, 219) = 9.07, p < .001. Reciprocal
support with family members was more indicative to recovery, followed by greater levels
of perceived support. Although the model is signiﬁcant, only 20% of the variability in
recovery was predicted by knowing the scores of these ﬁve variables. Frequency of

family contact and satisfaction with family contacts were not signiﬁcant in the model.

83

Table 4.7

Summary of Social Network Models Predicting Recovery Assessment Scale Total Score

 

 

and Factors

Model Variables 3 SE B [3 t

Model A Network Support, Dependent Variable: RAS total score
Total Network Size 1.37 .44 .18 3.12**
Total Network Support 4.04 2.18 .14 1.85T
Total Network Reciprocity 6.60 1.60 .30 4.1 1*"‘
Total Network Contact -0.97 1.77 -.03 -.54
Total Network Satisfaction 3.67 1.80 .14 204*
R2=.29 .Adjusted R2 = .29, R = .54**
F(5,219) = 17.75, p = .00“

Model B Network Member, Dependent Variable: RAS total score
No. of Family 1.43 .824 .1 1 1.74
No. of Friends 2.29 1.00 .15 2.29*
No. of Professionals -l.l6 1.76 -.04 -.66
No. of Clubhouse Staff 3.09 1.05 .19 2.94**
No. of Clubhouse Members 2.88 1.51 .12 1.901'
R2=.O8 Adjusted R2 = .06, R = .29**
F(5, 219) = 4.00, p = .00

Model C Family Network, Dependent Variable: RAS total score
No. of Family .858 .928 .06 .92
Family Support 5.59 2.60 .18 2. 14*
Family Reciprocity 7.26 2.01 .31 3.60**
Family Satisfaction 2.27 2.31 .08 .98
Family Contact -.877 2.33 -.02 -.37
R’=.24 .Adjusted R2 = .22. R = .49**
F(5, 166) = 10.51, p = .00“

 

*p<.05, **p<.01, ’p=.06

84

Research Question 5a: Are greater family network supports predictive of the recovery
domains of Personal Conﬁdence & Hope, Reliance on Others, and Goal & Success
Orientation ?

To examine the notion that indicators of recovery occur in the context of positive
family supportive relationships, 3 separate series of standard multiple regression analyses
were performed. The domains of Personal Conﬁdence & Hope, Reliance on Others, and
Goal & Success Orientation were selected to reﬂect aspects of the ‘development of self in
relation to others’ concept of differentiation as conceptualized by Knudson -Martin’s
Feminist perspective of Bowen Family Systems Theory. It was expected that family
relationships characterized as more supportive, more satisfying, and more reciprocal
would be indicative of participants who are more conﬁdent, hopeful, and goal oriented,
and able to utilize social supports. It is argued that family network supports inﬂuence
recovery through means of relating that supports autonomy, which is consist with the
Feminist informed Bowen’s Family Systems Theory of development of self and
differentiation. Table 4.9 displays the unstandardized regression coefﬁcients (B ) and
standard error of ( B ), the standardized regression coefﬁcient ( B ), the R2, and adjusted
R2. Sample sizes used in this set of analyses are based on participants who nominated at
least one family member on the network. Family network size, perceived family network
support, perceived family network reciprocity, family network satisfaction, family
network importance, and family contact were entered as predictor variables and the
recovery dimensions of Goal and Success Orientation, Personal Conﬁdence & Hope,

Reliance on Others (Model D, E, & F).

85

The regression analyses demonstrated mixed results between family network
support variables and indicators of recovery. Results of Model D demonstrate greater
perceived reciprocity with family network members and satisfaction with contacts were
positively related to increases in a sense of personal conﬁdence and hope. The results
from Model B indicate perceived support and reciprocity as a signiﬁcant predictor to the
recovery domain of Reliance on Others. In the last Model, the recovery domain of Goal
and Success Orientation was also dependent on greater perceived support and reciprocity

with family.

86

Table 4.8

Relationship between Family Network Support Dimensions and Recovery Domains

 

Model Variables 3 SE B [3 t

 

Model D: Family Network Support and Personal Conﬁdence & Hope

No. of Family 0.09 .029 .024 0.33
Family Support 0.10 .086 .l 1 1.25
Family Network Reciprocity 0.22 .063 .32 3.5 8**
Family Network Importance -0. 16 .089 -.15 -1.84
Family Network Satisfaction 0.17 .075 .20 238*
Family Network Contact 01 1 0.07 -.1 l -1.54

R2=.22 .Adjusted R2 = .19, R = .46**
F(6, 166) = 7.34, p = .00

Model B : Family Network Support and Reliance on Others

No. of Family .002 .025 .08 1.06
Family Support .158 .072 .20 2. 17*
Family Network Reciprocity .1 12 .053 .19 210*
Family Network Importance -.01 l .075 -.01 -0.15
Family Network Satisfaction .039 .063 .05 0.62
Family Network Contact .003 .061 .00 0.05

R2=.15.Adjusted R2 = .12 , R = .39**
F(6, 167) = 4.99, p = .00

Model F: Family Network Support & Goal and Success

Orientation
No. of Family -.007 .027 -.02 -.28
Family Support .238 .080 .26 2.95**
Family Network Reciprocity .196 .059 .29 3.31 **
Family Network Importance -.098 .083 -.096 -1.17
Family Network Satisfaction .089 .070 .10 1.24
Family Network Contact -.093 .068 -. 10 -1.37

R’=.25.Adjusted R’ = .22 , R = .50**
F(6, 167) = 9.15, p :00“-

 

* p <.05, **p < .01, "p = .05

87

Research Question 5b. What type of activities characterizes the interactions between
clubhouse members and nominated family members?

A total of 169 clubhouse members nominated at least one family member. A
qualitative analysis of the open-ended responses was performed on a Single open ended
question. Participants were asked what they did with each person nominated on the social
network (i.e., “What type of things do you do with Sally?”). Open-ended responses were
examined for participants who nominated family members on the social network.
Responses were coded as follows: (1) Social/Leisure interactions, (2) Financial/Material
support interactions, (3) Instrumental support interactions, (4) Emotional support
interactions (5) Reciprocal support interactions (e.g., participant indicated provided some
type of support to network members). Responses may have included various types of
interactions. Each type of interaction was coded separately, thus yielding non-exclusive
categories. Each category was ranked by frequency of occurrence. Overall, clubhouse
members described positive interactions and activities with family members.
Social/Leisure activity was the most frequently identiﬁed interaction (f = 272/123%).
Social and leisure activities with family members were described as “visiting”, “talking

9’ ‘6

on the phone, going out to eat,” “hanging out,” or “taking walks.” Instrumental
Support Interaction ranked second, as many clubhouse members described family
network members assisting them with shopping, transportation, assistance with

community resources, or doing laundry. Emotional Support Interactions ranked third

(e. g., “talk about problems,” “family gives me advice,” “talk about personal matters”).

88

Research Question 6: What is the relationship between clubhouse engagement, as
measured by level of clubhouse participation and extent of clubhouse sense of
community/cohesion, and greater social network supports (size, support, reciprocity,
satisfaction, and contact), and recovery?
Intercorrelations among Path Study Variables

Zero-order correlations among all 21 study variables are featured in Table 4.9.
The intercorrelations among the demographic variables are displayed in the ﬁrst triangle
(Variables 1 — 6) of the diagonal. The next segment is composed of social network
predictor variables (7- 12), followed by the recovery domains and total score (13 — 18).
Clubhouse engagement variables are located on the last two rows of the matrix. The
sample size for this matrix was 221, and 220 for recovery factors. Correlations were

calculated using pairwise deletion of cases.

89

Eco-Enemiom u _ .mLmocwEQ «EDEQSEom-eoz u o die—ESE“ 3:028.er u_ Amer»: 32309..»- Ewto ..o 3.8885059“. u c Jena—omega $.33 52052 n _ .btoEE-coz
n 8 peace on: u _ .oaeou .t. o 856 2 2an> 82:222.. one... c6. 880 . .8. v .2 us a 282:»: at. a c... e5 8V a o... a .5828: pa ...2. A ...c 88.8 n 2 H82

 

VN ow. cm. to. 9». mm. NV. N_. on. on. 3.. pm. 2. 8.. ac.- Sr 3. mo. co. mags—Eouuo 028m .om

 

- mo. 5. mo. 5. no. mo. oo. 2. no: no. 2. co. 8.- cm: 2.- mo. 3. 3 .- :o_.m&o_§wm £30 .2
- on. 3. am. we. co. m _. mm. 2. 9o. 3.. mm. mm. _ _.- 2.- 5.- mo. 8.. _So.r bo>ooom .m—
mega—cam

. 3. am. an. 2.. 2. 2. 8. 2. a. 2. 2.- 2.. 8.- 2. 8. 8. 38.28.862.622

- On. 0*. mn- w_. em. on. hm. mm. mm. 3.- w... No. 2.- no: 5.- 3050 co pong—0&6—
- on. on. 3. 3”. mm. 9.. mv. 2. 2.- Nb.- 2: we: mo: .8.- mmouoswumV—aoc.m_

 

 

 

 

- 8. 8. cm. 2. mm. 8. 2. 8.- 2.- 8.- 3. 8. 8.- .8 6.3 Swenﬁui
. . . 86:8
- 8 em. 2. 2.. 8. 4m 2- :.- 8.- 8. 8. 8.- 888.50 3828.2
- 8. mm. 8. R. 8.- 2. 8. :.- 8.- 8. 8. cos—Bu .2
- a. 2.. mm. 2. 8. 8.- 2.- 8.- 8. 2.- cocoa-2.8.:
- 2.. .e. 5.- 2. 8. 8.- 2.. Sr 8. 8528528
- on. 2. 2. 2.- 2.- :.- 2. 8.- E8563. .8
- 2. S. 2.- 2.- :.- 8. 5.- 8&5 .w
- :.- 2.- 8. 5. 8. 8. 8895 .6 a .e
- 2. 8. 8. s.- a. £52888 .e
- 3. 5. 8.- mo. EoEowcmE wag-— .m
- 2. 2.- 2. 8:822 .e
- 8.- :. ow< .m
- s. bees-m .N
- Mona—DO .—

 

 

0. w. E f n. 3 v.— N_ 2 o_ o w h o n v m N _

2on6: 56% VS» towmaoxmom 5 326.23» 26.22.20 we SEEM-i 5636.25» oEQEmoEono 36.23339 .5956ng
0.? 05$.

9O

Direct and Indirect Pathways to Recovery

Path analysis was used to explore the relationship clubhouse engagement
variables, social networks, and recovery. Path analysis is a series of hierarchical
regressions predicting to multiple dependent variables. Standardized coefﬁcients are used
to generate the direct and indirect effects of predictor variables to dependent variables
(Klem, 2000). A path analysis using SPSS AMOS 5.0 statistical program was used to test
the model (see Figure 4.1). Clubhouse participation served as an exogenous variable.
Sense of clubhouse community/cohesion served as a mediating variable between level of
clubhouse participation and social network variables. The Sense of Community/Cohesion
scale was used as a measure of clubhouse engagement in this study to assess the extent to
which clubhouse developed as sense of members engaged with peers, experienced a
sense of bonding through meaningful activities, and expressed loyalty to club friendships.
It was expected that greater club cohesion would generalize beyond the clubhouse walls
and be more predictive of greater network supports; increase network size, greater
satisfaction with contacts, more reciprocal relationships with others.

Demographic covariates related to recovery were not entered into the model. A
cut-off of r = .25 or 6% variance explained for possible demographic covariates, or an
alpha level below .01 was used to include covariates. This is to reduce the likelihood of a
Type Ierror. A diagnosis of schizophrenia has a weak association with recovery with a
coefﬁcient of .22 at the .01 level. This variable was excluded from the path analysis. The
rationale for excluding this variable from the analysis is based on results of the regression

model A (Network Support) in predicting recovery. Such a weak association added very

91

little to the overall model to explain social relationships and recovery. In fact, entering
diagnosis into the model resulted in a .05 R2 change.

Assessing ‘goodness of ﬁt’ in path models involves testing individual path
coefﬁcients and an overall test of the model with all its paths. Structural equation
modeling goodness of ﬁt measures were used as indicators of model ﬁt. A model that ﬁts
the data will reveal the sum of path values from social network support variables to
recovery to equal the regression coefﬁcient for recovery predicted by network variables.

The relationship of the social network support variables to recovery is modest,
with poor model ﬁt. Regression coefﬁcients are presents in the path model located in
Figure 4.1. A signiﬁcant chi-square value indicates the speciﬁed full social network
support model was different from the underlying data covariance structure x2 (11) = 124.
.34, p = .00. Three additional indices were used to assess model ﬁt; the Root Mean
Square Error of Approximation (RMSEA), which indicates a good model if values are
close to zero; Akaike Information Criterion (AIC) close to zero reflects good ﬁt; and
Expected cross-validation index, (ECVI) reﬂects the discrepancy between model-implied
and observed covariance matrices. Goodness of ﬁt tests were also consistent in
indicating that the social network support model did not ﬁt the data, RMSEA = 0.217;
AIC = 172.34; ECVI = .78 . Good of ﬁt statistics determine if the model tested should be
accepted or rejected. The purpose of this analysis was to explore a model hypothesizing
relationships among variables, given the assumptions of the study.

Regression coefﬁcients remain uninterpretable as a result of poor model ﬁt.
Although regression coefﬁcients are signiﬁcant, the paths hypothesized may have

unidentiﬁed variables not included in the model.

92

The correlations implied in the model between perceived network support, reciprocity,
satisfaction, and network size to recovery are small and not equal to the observed
correlations (see table 4.9). However, model ﬁtness does not imply the model is correct
(Klem, 1995). The position of the relationship in the model could be arranged in a variety
of ways and one may still achieve a good measure of ﬁt. But the results of the path model
indicate the speciﬁcation of the model may be incorrect; that is, there are other
moderating or intervening variables not identiﬁed in the model better able to explain the
variability in recovery.

The model was trimmed to include two network support variables: total network
support and total network reciprocity. The model ﬁt statistics were slightly better, but the

underlying covariance structure of the data did not ﬁt the model x2 (4) = 68.09, p = .00.

93

.bo>88 8 3850: 38m :0 EoEowmwco 08.2520 ..o 8:25.: ”—088 QEESN-Emm N .v .2:me

 

 

 

 

 

 

 

94

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5:88wa
x83qu
we.
3% x3382
* n _ .
582550 i cow—«Euthan—
bo>ooom 3mm. .8 Scum 88:22.0
.. s. 808908 ...mN.
338qu—
S. toga—6

 

3385a—

 

 

CHAPTER 5

DISCUSSION

The purpose of this study was to examine the relationship between social network
supports and a self—reported measure of recovery from mental illness. Clubhouse
members, on average nominated ﬁve network members, mostly consisting of kin. This
ﬁnding is similar to other social network studies using psychiatric samples, however
lower than what was expected among clubhouse members. Based on the assumptions of
the clubhouse model, it was expected that clubhouse members would dominate network
proﬁles, but clubhouse members were least likely to be nominated. Family emerged as
the largest network support cluster. Friends outside the clubhouse were the second largest
group to comprise the network. Clubhouse staff was the third largest network support
cluster. This ﬁnding was surprising, since clubhouses typically have a small staff to
member ratio. Further, greater sense of community was related to greater nominations of
clubhouse staff, not members. Contrary to the clubhouse principles that promote member
and staff collegiality, these results suggests the presence of a less egalitarian structure
between members and staff than expected. Perhaps the simple fact that staff are paid
employees while members are not, already creates a hierarchical division. Very early
studies of groups of people living with psychiatric disabilities have found the powerful
influence professional staff play in social and communication dynamics between staff
and people with psychiatric disabilities (Learner & Fairweather, 1963). The presence of

staff encourages less dependency on other clubhouse members for support,

95

Club members may have also perceived other club members as part of a larger
collective, therefore making it difﬁcult to identify individual members. During data
collection, some participants attempted to nominate the clubhouse, as opposed to
isolating one or two members.

Despite research showing smaller social networks among individuals with more
severe diagnoses, such as schizophrenia, this study demonstrated no difference in
network size among club members with and without schizophrenia. This suggests
clubhouse members are equally likely to have small to larger networks, independent of
diagnoses. However severity of the diagnoses was associated with smaller family
networks. These members reported greater contact with family, as well as friends. This
suggests that clubhouse members with schizophrenia have access to other forms of
support beyond the family. Most studies have found that over-dependency on families for
support was indicative of increased family stress.

Social Networks & Recovery

The results of the study showed social network supports dimensions to be
positively related to subjective experiences of the recovery process. However
associations were weak to modest. According to the analysis, the single most important
predictor to recovery was reciprocity. Clubhouse members who perceived themselves in
more reciprocal, supportive relationships with their support network were more to be
farther along in their recovery process.

The network support size of club staff and friendships outside the clubhouse were
more predictive of overall recovery than the size of family, professionals, and clubhouse

members combined. Although family was the largest network cluster, it did not emerge

96

as a signiﬁcant predictor of recovery. A study from the Yale Supported Socialization
Partnership Project may provide an explanation. Mental health consumers in the
Supported Socialization Project (Davidson, Haglund, Stayner, Rakfeldt, Chinman &
Kraemer Tebes, 2001) who also attended clubhouses expressed a greater desire to be
around others ‘who were unlike themselves’ (i.e., have a psychiatric disability) and to
develop social connections based on mutual interests instead of the shared experience of
a disability. Findings from the unpublished Flinn Report (Herman, Onaga, Ferguson,
Pemice-Duca, Oh & Weaver-Randall, 2002) revealed that only 5% of clubhouse
members reported “receiving support from staff and other members” as main reason to
attend the club. Therefore, it may not be surprising that a low number of clubhouse
members made the list.

A large number of participants reported social and leisure activities with families,
followed by ﬁnancial and material support interactions. The study attempted to capture a
‘glimpse’ of the type of things club members do with family. A limitation of this
question, “what types of things do you do with _?”, was that it did not gather
information about the speciﬁc type of support each person provided. Responses that
expressed some type of support were spontaneous, since respondents were more inclined
to answer the question with a description of an activity. However, the analysis of the
open-ended item suggests that clubhouse members generally reported positive
interactions, consisting of fun and enjoyable activities with family (e. g., “going bowling”;
“going ﬁshing”; “talking on the phone”). Some club members also described the type of
support they provided their family, such as cutting the lawn for an elderly parent or

providing child care for a sibling.

97

Social Network Support Dimensions & Recovery

The strongest correlate to the construct of recovery found in this study was the
dimension of perceived reciprocity with network supports. Clubhouse members who
viewed themselves as ‘providers of support’ or giving back in a relationship were more
likely to have greater recovery scores. Having hope, conﬁdence, and strategies in
managing symptoms were all indicators of recovery in this study. Reciprocity emerged as
the single most important correlate because individuals feel part of a community they are
able to contribute to. For individuals living with chronic mental illness, this becomes
especially signiﬁcant in creating a view of themselves as active members of their support
networks, and not as passive recipients of support.

In terms of the inﬂuence of family network supports, the results indicated that
being able to ‘give back’ to family members who provide support was associated with
greater personal conﬁdence and hope, goal oriented activity. The study also found that
more satisfying and supportive contact with family was associated with greater aspects of
recovery as being more conﬁdent and hopeful. A qualitative exploration of the
interactions between family members and participants revealed a large percentage of
participants reporting social and recreational activities with family members. A small
number of participants also described things they did for family, such as cutting the lawn
for an older parent or babysitting for a sister. The notion that recovery can occur in a
context of supportive relationship provides some preliminary support for the Feminist
informed Bowen Family Systems Theory concept of differentiation (Knudson -Martin,
1996). More satisfying relationships with family contacts together with positive

appraisals of family support was characteristic of more conﬁdent and hopeful club

98

members, which can be argued to be indicative of greater level of differentiation. Greater
support and reciprocity with family members was also associated with being more goal
and success oriented, and more able to elicit and rely on social support. These ﬁndings
provide support for the notion that we develop and recover in the context of supportive
relationships. It recognizes the interdependency of social relationships and recovery.
Social Networks & Clubhouses

Clubhouse tenure was not associated with the number of club staff or members
nominated on the network. Both club tenure and weekly participation were also not
related to larger networks. This ﬁnding does not conﬁrm the assumption that merely
attending and being a clubhouse member is associated with larger social networks.
However, greater club participation (i.e., more days and hours per week) was associated
with increased sense of cohesion with club peers. Coming to the clubhouse more often
created more opportunities to build mutual relationships and develop an afﬁliation with
people with similar experiences. Greater clubhouse afﬁliation translated into
opportunities to increase social network size and foster greater reciprocity with network
members. Sense of clubhouse community also emerged as signiﬁcant correlate to
recovery. This ﬁnding is supported by another study in which clubhouse members
identiﬁed social connections and recovery as two major concepts deﬁning the clubhouse
experience (Herman, Onaga, Pemice-Duca, Oh, & Ferguson, 2005).
Path Results

One advantage of using path analysis is that multiple dependent variables can be
examined simultaneously unlike in regression. It was proposed that access to the

clubhouse through participation and gaining a sense of belonging with peers at the

99

clubhouse would increase network size, support, opportunities for mutual relationships
and overall satisfaction, in relation to recovery. The model did not ﬁt the data well, but
some important conclusions can still be drawn. First, clubhouse participation was
signiﬁcantly related to clubhouse cohesion; feelings of belongingness and the extent to
which members were engaged in mutually obligated relationships. Clubhouse members
who perceived themselves to be part of a greater community, were also more likely to
have greater numbers of network members, share in reciprocally supportive relationships.
These ﬁndings support claims of the approach of clubhouse rehabilitation. As Berkman
et al., (2000), from the Harvard School of Public Health stated in her analysis of
empirical evidence on the relationship between social networks and health:

...measures of social integration or connectedness have been such powerful

predictors of mortality [because] these ties gives meaning to an individual’s

life by virtue of enabling him or her to participate in [the network] fully, to

be obligated (in fact often to be a provider or support) and to feel attached to

one’s community (p. 849).

People tend to seek emotional and social support that provide socialization opportunities.
These opportunities may improve overall social adjustment by way of increasing self
conﬁdence or afﬁrming one’s social identity as someone other than a mental health
patient. Social networks were presumed to provide meaningful roles and opportunities for
mutual exchanges of support, which is a hypothesized to be related to recovery.

A major disadvantage using a path analysis method is that multiple models may
be correct or incorrect in explaining the data. Model speciﬁcations for this study were

incomplete. Since speciﬁcation is usually based on previous knowledge or theory,

100

potential variables related to recovery were excluded from the model and this study.
Recovery is a very complex construct that has multiple dimensions.
Social relationships, alone, doe not explain the variability in recovery. However, the
purpose of the analysis was to explore a pathway to recovery through clubhouse
engagement and social network supports. The standardized betas generated by the path
analysis and correlations between variables showed that clubhouse participation, sense of
community and the relationship between social network dimensions ﬁt the data well. The
expected relationship between total network support, reciprocity, network satisfaction
and network size and recovery contributed to poor model ﬁt, despite moderate
correlations. It is clear that recovery goes beyond the inﬂuence of social relationships.
The literature in recovery suggests a constellation of variables that contribute to recovery,
which were beyond the scope the current study.
Limitations

It is important emphasize that the sample used in this study, albeit on the severe
end of the continuum of mental illness, are more stable than a general population of
people living with chronic mental illness who are uninvolved with mental health services.
Further, the cause of one’s illness remains unknown; that is, whether the illness was
attributed by biological influences, cognitive - affective basis of behavior, or social
inﬂuences. The etiology of the illness was not a focus on the study, and participants were
selected on the basis of being a clubhouse member, not the illness.

A major limitation of this study was the use of cross-sectional data to test
predictive relationships between network support dimensions and recovery. It is equally

plausible the relationship between network supports and recovery is bi-directional. This

101

study does not imply causal relationships or emphasize direct, linear relationships
between variables. It is unclear from this study what affects recovery, since recovery may
have occurred before a clubhouse membership. However, studies in psychosocial
rehabilitation and recovery have found evidence for increased subjective and objective
reports of recovery. Further, randomized, controlled studies have demonstrated support
for the direction of greater network supports, social contact, and increased psychological
functioning, adjustment, and well-bein g. Overall, the strongest relationship to emerge
from this study was between perceived reciprocity and recovery. Whether recovery is an
outcome or a predictor, perceiving oneself as an equal player in a meaningful relationship
where there is some level of give and take does relate to increase in a sense of purpose,
hope, and conﬁdence. Reciprocity may work to reinforce positive social identities.

The study failed to include a more diverse representation of clubhouse members
with cultural and ethnic backgrounds. Although low, this is consistent with demographic
proﬁles for clubhouses across the state of Michigan. Early research shows no signiﬁcant
differences in network size and characteristics between Caucasian and African American
persons with schizophrenia, however, older African Americans with schizophrenia tend
to have smaller, kin dependent networks without outside formal contacts (Cohen &
Kochanowicz, 1989).

Another limitation of the study included the poor distinction of social network
relations during data collection. Family of origin, procreation, and other kin were simply
grouped into the category of ‘family’ without differentiating the individual’s role within
the family. Nearly all social network studies reviewed failed to differentiate who in the

family was included in the network. Understanding who in the family is most often part

102

of the network may clarify the quality of the relationships and who in the family of
people living with chronic mental illness are sources of support. A great majority of
clubhouse members fell within the age range of 36-45, which indicate active activity with
family of origin and procreation.

Survey measures such as the Corrigan Recovery Scale may not adequately
capture the dynamic process of recovery from mental illness. Yet, direct clinical
observations or measures may assume that the absence of symptoms is indicative of
recovery. The recovery construct is considered an attitudinal scale and based on
reﬂections of personal experiences and feelings. There has been minimal research to test
the validity and reliability of the Recovery Assessment scale used in this study
(Loveland, Weaver Randall & Corrigan, 2005). Further, the item response design ranged
from 1 (strongly disagree) to 5 (strongly agree), which included a middle response of 3
(not sure). Including a middle alternative can affect responses and the conclusions that
can be drawn from the data. Inclusion of a middle alternative in attitudinal scales may
attract a signiﬁcant number of respondents who may be unsure of their opinion. It may be
argued that being unsure of agreeing or disagreeing with a statement may suggest that the
respondent is ﬁnding themselves in the cross-road between low recovery experiences and
high indicators of recovery experiences.

The research and literature in recovery is ill deﬁned and continually emerging.
The results of this study provide some evidence of the relationship between social
network support and recovery, as were found by Corrigan & Phelan (2004). In sum, ﬁve
main ﬁndings emerged from this study. First, families continue to dominate social

network support proﬁles among consumers of mental health; second, network support

103

sizes are still relatively small; third, club members were least likely to be nominated as
sources of support; fourth, engaging in reciprocal supportive relationships with members
of one’s social network was the single most important concept related to recovery; and
ﬁfth, being a regular participant at the clubhouses was related to having a sense of
clubhouse community. Overall, sense of community was signiﬁcantly related increased
perceptions of reciprocal relationships with others, more satisfying contact with network
supports, and modestly related to an increase in social network size. Taken together,
however, these network support dimensions failed to be largely associated with recovery
as an overall outcome as revealed by the path analysis. This leads to further exploration
of additional correlates to the recovery process not explained by social relationships
alone. Loveland, Weaver Randall & Corrigan suggest that social indicators comprised of
variables such as employment status, history of hospitalization, and housing may be more
related to the construct of recovery, which are considered standards of independent
functioning.
Clinical Implications

The literature review found a lack of reciprocity as a salient feature of the social
relationship between people living with chronic schizophrenia and network members.
This characteristic leaves one highly dependent on others, which can signiﬁcantly result
in strained or stressed interpersonal conﬂict, and increase emotional reactivity among
family members. Although clubhouse members signiﬁcantly reported greater perceived
support from the network, the perception that members viewed themselves as being a

provider of support was important in the recovery process.

104

A recovery orientation offers hope, while normalizing the experiences that can
accompany the illness. These approaches are consistent with the values of MFI’s, where
there is less emphasis on pathology and more attention to individual strengths, personal
experiences, afﬁrmation, and acceptance of self. MFTs work collaboratively, using a
non-pathological framework to cognitively reframe one’s internalized stigma of being
labeled with a mental illness. MFl‘s clinically focus on personal growth and acceptance,
management of symptoms in contrast to elimination of the illness, and utilizing extended
family supports, which are all parallel to the dimensions of recovery.

According to the Feminist Elaboration of Bowen Family Systems Theory
(Knundson-Martin, 1996), individuals develop in the context of relationships, moving
toward differentiation in relation to others, not independent from them. The tension
between the needs for individuality and the needs for togetherness are not viewed as
competing needs, but as compatible. Development of self occurs through connection with
others which fosters increased capacity to orient oneself toward relationships, experience
more satisfying social contacts, while increasingly developing a distinct sense of self.
Overall, this inclusive view of differentiation assists in examining how others are
involved with the focal person, by creating a picture of how this person interacts with
others.

The concept of recovery in this study is reﬂective of a personal ideology that
incorporates autonomy, social support, and personal conﬁdence. The capacity to function
autonomously, while staying connected and engaged with family members was a key
ﬁnding of this study. Reciprocity with family members emerged as the single most

important predictor to recovery among participants who identiﬁed family support in their

105

social network proﬁle. This suggests that recovery is dependent on mutual exchanges of
support, which is indicative of one who is more integrated and connected on Knudson-
Martin’s (1996) dimensions of differentiation. The notion that one can give back in a
supportive relationship has signiﬁcant implications for interventions with family
members. MFTs can carry forth that message of “hope and recovery” and identify
strategies to increase the likelihood of subjective experiences of recovery, such as
conﬁdence, hope, autonomy, and reliance on positive social supports. A timely issue of
the Family Therapy Magazine published by the American Association of Marriage and
Family Therapists highlighted the recovery movement and described clubhouses, and
other psychosocial or peer support programs (AAMFI‘, May/June, 2005). This study
gives attention to an important area in mental health. It gives the message that people
living with chronic mental illness can experience meaningful supportive relationships
with others, especially families.

Social Network Therapy (SNT) (Wasylenki, et al., 1992) is a speciﬁc way the
results of this study can be translated or generalized into interventions with individuals
and their families. Clinicians can assist individuals and their family members to move
from small, kin—dominated networks to include people in the community. Case studies
and outcome evaluations using Social Network Therapy have demonstrated greater
satisfying family contact between clients and members of their family, less reliance on
family support, and greater reciprocity in family relationships. SNT has been described
as an intervention to increase overall network size by members beyond the immediate
family. Eco-maps provide a concrete baseline assessment on one’s social network

connections. Clinical strategies include ‘constructing’ a network by adding new

106

members. This can be achieved by including peer support programs, clubhouses, and
other groups. Clinicians also hold consultation meetings with network members to effect
change, as well as take on the role as the network ‘coach’. This aspect attempts to
educate members about the illness, reduce emotional reactivity among members, and
increase more positive interactions. This last strategy is very similar to Bowen Family
System Theory of ‘coaching’ (Bowen, 1978).

The study also demonstrated that contact with family was not an overall predictor
of recovery domains. This can suggest that as networks expand to include others beyond
kin, there is less reliance on family. This can also help in making for more satisfying
contact with family members, instead of contact characterized by high level of strain,
burden, or negativity. More satisfying contact with family network members was related
to greater personal conﬁdence aspect of recovery, which is consistent with greater level
of differentiation.

Overall, the clinical implication of the current study is relevant to professionals in
the ﬁeld of Marital and Family for a number of reasons. First, the current study
highlights the parallels between MFI‘s training and the construct of recovery, and the
need for MFTs to return to the roots of the ﬁeld with a more expansive perspective of
mental illness and families. Marital and Family Therapists are in a unique position to
respond to the need for training professionals to work with people with psychiatric
disabilities and their families (Leroy, Zipple, Marsh, Finely, 2000; Zipple, Spaniol &
Rogers, 1990), since the modern family therapy movement was an outgrowth of
psychiatry (Broderick & Schrader, 1991). Given the infamous history between family

therapy and mental illness, it is important that MFTs shift toward using effective models

107

in working with individuals and their families. For example, family psychoeducation
models have demonstrated efﬁcacious in reducing relapse and hospitalization.
Incorporating an orientation toward recovery compliments these any of these treatment

approaches while maintaining consistency with MFI‘s values.

Conclusion

Mental illness affects not only individuals, but the families who care about them
as well. This research showed that social networks members signiﬁcantly contribute to
their recovery process. More importantly, the opportunity to ‘provide support or give
back’ to family members, or others is a crucial aspect of the recovery process. Family
therapists who work with individuals with chronic psychiatric disabilities are encouraged
to assist them in ways to expand social network resources. Assisting them to cognitively
reframe their identity as a potential resource to others and focusing less on the self as
‘disabled’ would be helpful.

Caution must be taken in interpreting and generalizing the results of this study,
however. This study used a cross-sectional design, which does not allow us to assert that
increased social network supports led to increased experiences in recovery. However, this
study did establish a relationship between network supports and recovery, as well as
clubhouse engagement, social network supports, and recovery. Relationships were
modest, suggesting social relationships alone do not account for the variability in
recovery. However, the results of this study support ﬁndings found in one other social
network study using the same recovery measure (Corrigan & Phelan, 2004; Corrigan,

Giffort, Rashid, Leary, & Okeke, 1999).

108

Future Research

Although a number of articles are written about recovery, there is little research
examining social network supports and the recovery process. Organizations, programs,
and mental health agencies are moving toward creating a culture of recovery. Future
research can continue to examine the role social network supports and recovery. Studies
must include baseline assessments before interventions and track network growth over
time. A scale assessing differentiation along with a subjective measure of recovery can
service as a method of assessing how recovery and differentiation of self change in
relation to others. It is important to emphasize that differentiation does not equate to
autonomy. However, as the interpersonal environment of individuals living with chronic
mental illness expands to include members beyond kin, it increases the likelihood of
more balanced, reciprocal, and satisfying relationships with family that may not be
dominated by high levels of emotional reactivity which is characteristic of small, overly
dependent family networks (Leff, 1976). Clubhouses can serve as an environment to
create new networks, decrease dependency on family, and develop social skills through
socialization and interactions in mutual and social reciprocal relationships. Reciprocation
of support with members of one’s social support network is powerful in changing views

of the self from a passive recipient of support to an active resource to others.

109

APPENDICES

110

APPENDIX A

Letter to Clubhouse

111

Appendix A
Letter to Clubhouse
Dear Clubhouse Members, Managers, 6: Staff:

In the past several years, very little formal research on the benefits of
psychosocial programs and clubhouses has been conducted in the human services field.
However, over the past two years, Michigan State University and the Michigan
Department of Community Health have partnered on a project called ‘The Flinn
Clubhouse Pro ject" which has been designed to help us understand how clubhouse
programs benefit their members.

During this time, The Flinn Clubhouse Project has been working with several
Michigan clubhouses in developing a greater understanding of clubhouse programs for
people with mental illness. To date, several clubhouses have participated in various
phases of the project such as competing mail in surveys, implementing a member -
driven computer database system, and participating in site visits.

These and several other project activities have invaluably contributed to the
clubhouse knowledge base in Michigan. One of the intended outcomes of the project
will be to compile a handbook of unique clubhouse practices and positive psychosocial
outcomes in Michigan clubhouses.

As we move into our third year, The Flinn Clubhouse Project would like to
interview members about their experiences as a clubhouse member and their
experience with mental illness. Our hope is that an in-depth interview will provide
richer, more meaningful information that incorporates the multifaceted aspects of
each individual's experiences and the impact of their participation in psychosocial
programs.

On the following page is an outline of how we will attempt to facilitate
participation of those who are interested in the Member Interview. Please review the
process and feel free to call Sandy Herman (517/335-0130) or Esther Onaga
(517/355-0166) for clarification or further information.

We hope that your clubhouse will seriously consider this opportunity to be part
of a unique knowledge base that will help increase understanding of clubhouse
operations and people with mental illness.

Sincerely,

 

Katie Weaver-Randall

 

Sandra Herman, PhD

 

Chandra Donnell

 

Esther Onaga, PhD

 

SuMin Oh

 

Francesca Pemice-Duca

112

APPENDIX B

Participant Consent

113

Appendix B

Participant Consent

Flinn Clubhouse Project
Psychosocial Rehabilitation in Michigan

Participant Consent Form

Bypass

The purpose of the interview(s) is to learn about your experiences as a clubhouse
member and your experiences with mental illness. The interview(s) will cover a
variety of t0pics that are related to your clubhouse membership and who you are.

We will ask you demographic information about where you live, your
clubhouse participation and how you feel about the clubhouse, and information
about areas in your life that may change with participation. These areas are things
like your feelings of empowerment and recovery, your achievements in daily life,
your work and social activities, and your physical and mental health.

Interview Procedures
Participating in the interview(s) will involve the following:

Contacting You: We will be interviewing you over the next year. The ﬁrst interview
will take place in the late summer/early fall of 2000. The second interview may take
place 6 months later, in the spring/summer of 2001. You will be asked to give us
permission to contact people or agencies who will be able to assist us un contacting
you in case we are unable to locate you for a possible 2"d interview. We will only ask
them how we can contact you, and no other questions.

Interviews: The interview(s) will have several sections and questions. The
interview(s) will be approximately 1 hour long. You can withdraw from the
interview(s) at any time. Your answers will be strictly conﬁdential during and after
the interview(s). Your privacy will protected to the maximum extent allowable by
law.

Beneﬁts to You

In the past, many people have found participating in this type of study an
interesting and educational experience. For your participation in the interview(s),
you will receive $20 in cash for the ﬁrst interview and we will pay you in cash if we
are able to conduct a second interview.

Risks

We anticipate no risks to you from participation in these interviews. Some
questions may be about difﬁcult or emotional subjects. If feel uneasy about any of
these questions, you can choose not to answer the questions or end the interview.

Voluntary Participation

114

Your participation in these interview(s) is completely voluntary. Whether or
not you agree to participate will have no effect on the services you receive from the
clubhouse or mental health services. You are free to withdraw from participating at
any time. You do not have to respond to any question you do not want to answer.

Conﬁdentiality

All information during the interview(s) will be kept strictly conﬁdential.
Your privacy will be protected to the maximum extent allowable by law. We will not
use your name on the interview(s). Instead a number will be used to code your
answers. The only people who will have access to your answers will be the Flinn
Clubhouse Project Staff. We will be interview about 300 people and all the answers
will be groups together and not on an individual basis.

Questions or Concerns

If you have any questions or concerns regarding this project, please call the
people who are in charge of this project, Dr. Esther Onaga at 517/355-0166 or Dr.
Sandra Herman 517/335-0130.

If you have any questions and concerns about your rights as a participant of a
study, please call Dr.David Wright, Chair, Michigan State University, University
Committee on Research Involving Human Subjects (UCRIHS) 517/355-2180

 

Consent Statement

You are being asked to participate in a study that may involve two separate
interviews 6 months apart. You indicate your voluntary agreement t participate in the
interview(s) under the conditions listed above by signing this consent form.

I have read and been explained the procedures and nature of the interview(s). I have
had an opportunity to raise questions and have them answered. I voluntarily agree to
participate.

 

participant name (please print)

 

 

participant signature date

 

 

interview signature date

115

APPENDIX C

Guardian Consent

116

Appendix C

Guardian Consent

Guardian Consent for clubhouse member to be interviewed
Flinn Clubhouse Project
Member Name

Puppose

The purpose of this interview is to ask the member named above a variety of topics-
demographics information on who he or she is and where he or she lives,
information about clubhouse participation and how he or she feels about the club,
and information about areas in his or her life that may change with club
participation. These areas are things like feelings of empowerment and recovery,
achievements in daily life, work and social activities, and physical and mental health
status.

 

Amount of contact

We will be interviewing clubhouse members over the next year, once in late the
summer and fall of 2000 and possibly at another time in the spring and summer of
2001. Each interview will take about an hour.

Beneﬁts

The clubhouse member will be paid $20 for the ﬁrst completed interview. There will
not necessarily be any other direct beneﬁts to him or her.

Risks

No risk to the member is anticipated form participating these interviews. If the
member feels uneasy about any of the questions, he or she can choose not to answer
them or end the interview.

Voluntary participation

The member’s participation is completely voluntary. Whether or not he or she agrees
to participate will have no effect on the services received. The member is free to stop
the interview at any time. The member does not have to answer any questions he or
she does not want to answer.

Conﬁdentiallv

All information given to us will be kept strictly conﬁdential. Individual responses
and information will not be shared with others. The only people who will have
access to member’s answers will be the Flinn Clubhouse Project Staff. All data will
be complied together and presented together, not on an individual basis. The privacy
of the member will be protected to the maximum extent allowable by law.

If you have any questions or concerns regarding this project, please call the people
who are in charge of this project, Dr. Esther Onaga at 517/355-0166 or Dr. Sandra
Herman 517/335-0130.

117

If you have any questions and concerns about your rights as a participant of a
study, please call Dr. David Wright, Chair, Michigan State University, University
Committee on Research Involving Human Subjects (UCRIHS) 517/355-2180.

Consent Statement

By signing this consent form, I am being asked to agree to permit the clubhouse
member named above, for whom I am legal guardian, to participate in this study
under the conditions listed above. A copy of this form will be provided to me.

 

guardian signature

 

date

118

APPENDIX D

Human Subjects Approval Letter

119

 

‘ OFFICE or
RESEARCH
ETHICS AND
STANDARDS

'erslty Committee on
Research Involving
Human Subjects

ichigan State University
202 Olds Hall

East Lansing, MI

48824

517/355-2180
FAX: 517/432-4503

v.msu.edu/user/ucrihs
nail: ucrihs@msu.edu

1;; an afﬁrmative—action,
Opportunity institution.

Initial IRB
Application
Approval

MICHIGAN STATE

U N l V E R S I T Y
Human Subjects Approval Letter

Appendix D

November 9, 2004

T01 Esther Onaga
27 Kellogg
Msu

Category: EXEMPT 1—4
November 5, 2004
November 4, 2005

Re: IRB ii 04-874
Approval Date:
Expiration Date:

Title: MICHIGAN PSYCHOSOCIAL CLUBHOUSE PROJECT DATA

The University Committee on Research involving Human Subjects (UCRIHS) has completed their review of ,
your project. I am pleased to advise you that your project has been approved.

The committee has found that your research project is appropriate in design. protects the rights and welfare of
human subjects. and meets the requirements of MSU‘s Federal Wide Assurance and the Federal Guidelines

(45 CF R 46 and 21 OF R Part 50). The protection of human subjects in research is a partnership between the
IRB and the investigators. We look forward to working with you as we both fulﬁll our responsibilities.

Renewals: UCRIHS approval 15 valid until the expiration date listed above. If you are continuing your project,
you must submit an Application for Renewal application at least one month before expiration. lithe project is

completed. please submit an Application for Permanent Closure.

Revisions: UCRIHS must review any changes in the project, prior to initiation of the change. Please submit an
Application for Revision to have your changes reviewed. it changes are made at the time of renewal, please
include an Application for Revision with the renewal application. ,

Problems. it issues should arise during the conduct of the researCh, such as unanticipated problems, adverSe
events, or any problem that may increase the risk to the human subjects, notify UCRIHS promptly. Forms are

available to report these issues.

Please use the IRB number listed above on any forms submitted which relate to this project, or on any
correspondence with UCRIHS.

Good luck in your research. if we can be of further assistance, please contact us at 517-355—2180 or via email
at UCRIHS@msu.§d_u. Thank you for your cooperation.

Sincerely,

WM

Peter Vasiienko, PhD.
UCRIHS Chair

CI Francesca Pemice-Duca
8125 West Fee Hall

120

APPENDIX E

Demographic Questions

121

Appendix E

Demographic Questions

 

 

 

Clubhouse ID# (CLUBID):

Name of Participant: Project ID
(PRJID):

Time (TIME): 1 2

(circle one)

Date of Interview (DATE): / / Place of interview:

 

Interview Started : Interview Finished :

If the interview was completed in more than one session, then use the space
below to indicate when the interview was restarted and finally completed.

 

 

Interview Restarted : Interview Finally Finished :
Date / / Date / /
Completion Code:

B 1. (success) Interview successfully completed
El 2. (refused)Member Refused

 

Reason:
D 3. (unable) Member Unable to Respond
El 4. (terrnin) Interview Terminated Early (page number ) Reason

 

If this interview is not fully complete, then note plans for completion in the
"NOTES" section.

Name of Interviewer: Interviewer Code
(intvwr):

IF THERE Is ANYTHING THAT FUTURE INTERVIEWERS SHOULD KNOW ABOUT THIS
PERSON (LE. TOUCHY SUBJECT AREAS, SAFETY CONCERNS, ETC), PLEASE NOTE BELOW.

NOTES:

122

 

Read the following to the respondent:

 

This interview should take about one hour to complete. During this time I will be
asking you several questions about different parts of your life. Many of the
questions are easy to answer, others may require a little more time and thought. If
any of the questions make you feel uncomfortable please tell me.

The topics that will be covered in this interview include questions about
employment, your living situation, your use of the clubhouse program, your physical
and emotional health as well as several other topics. I want to assure you that all of
your responses to these questions are completely conﬁdential. Your responses will
not be shared with anyone at the clubhouse or any other agency.

DO you have any questions before we begin?
(If you are providing food, then ask) Do you have any special dietary restrictions
(e. g. low-salt, low-sugar) due to any health problems such as diabetes, hypertension,

or weight-restrictions? (If so, please describe in "Notes" section on previous page &
give appropriate food as a snack)

123

Living Arrangement (LA)

Instructions : First, have the participant describe their living situation.
Record any relevant information in the box provided below. Then, based on
the participant ’s description answer questions LA] and LAZ- Probe until
enough information is gathered in order to accurately answer LA] and LAZ.

Could you tell me a little bit about your living situation? For
example, where do you live, who do you live with?

 

 

 

LAl. Where does participant currently live? Check (/)one that best
describes participant’ s situation. J

 

Homeless (1)

 

Living in private residence with family members (2)

 

Group home (3)

 

Nursing care facility (4)

 

Residential Treatment setting (5)

 

House or apartment (6)

 

Supervised home/apartment (7)

Other
Specify (8)

 

 

 

 

 

 

124

 

LA2. Who currently lives with participant? [Check all that apply]

 

Item Category Yes

(1) No (0)

 

LA2a. alone

LA2b. parents

LA2e. siblings

LA2d. spouse or partner

LA2e. friends or roommates who are consumers or
clubhouse members
LA2f. friends or roommates who are not consumers or

clubhouse members
LA2g. own children under age 18
LA2h. own children over age 18
LA2i. grandparents
LA2j. aunts, uncles or other relatives
LA2k. others specify

El EIEICID El El EIEIEIEI
D DUDE! I'.'l El DUDE!

 

 

125

Family conﬁguration (CF)

 

Instructions. If answer LA2, live with spouse or partner, clarify
status and check appropriate box. Otherwise, ask
question CF 1.

CFl. Are you married or do you have a live-in partner?

D Yes, married (1)
D Yes, live-in partner (2)
D No (0)

CF2. Do you have children? (If already know participant has children from
LA2 check yes and ask CF2a and CF2b)

D No (0)
D Yes (1) If yes, How many children do you have:

CF2a Under age 18?

 

CF2b Over age 18?

 

126

Income (INC)

INCl. Which category best describes your income during the last 12
months (last year)? Check the one which best describes the
participant’s situation.

 

/ Income range / Income range

 

Less than $500 (1) $10,000 to $10,999 (12)

 

$500 to $999 (2) $1 1,000 to $14,999 (13)

 

$1,000 to $1,999 (3) $15,000 to $19,999 (14)

 

$2,000 to $2,999 (4) $20,000 to $24,999 (15)

 

$3,000 to $3,999 (5) $25,000 or more (16)

 

$4,000 to $4,999 (6) Unknown (99)

 

 

$5,000 to $5,999 (7)
$6,000 to $6,999 (8)
$7,000 to $7,999 (9)

 

 

 

$8,000 to $8,999 (10)

 

 

 

 

 

$9,000 to $9,999 (11)

 

INC2. What was the main (principal) source of your income? Check
all that apply.

 

Item Category Yes (1) No (0)
INC2a. Employment wages (l)
INC2b. Retirement Income (2)
INC2c. Alimony, child support
INC2d. SDA, SSI, SSDI

INC2c. Other public assistance

 

El
DD

INC2f. Unemployment compensation
INC2g. Spouse, Partner, or friend
INC2b. NO income source

INC2i.
Other(specify):

EIEIEIDEIEID
DDDDI’JD

 

 

127

Living Independence (LI)
(Birchwood, 1990 Social Functioning Scale)

 

 

Instructions: Read each question to the person and based on his or her response
circle the number that best describes the person’s ability to do or use each of the
following.

 

Introduction: Now I would like to tell me how well you are able to do each of the
following activities.
[Give Participant Card #3]

Abl Abl Need ,

withc wit: Help(not DO" ‘ No‘

Question Club other currently KPOW/ applicable
Able Help help receiving) Unsure (Specify)

How well are you
able to use public 4 3 2 1 9 8
transportation?

-LIl.

 

How well are you

able to handle 4
money (e. g.,

making change)?

How well are you

able to budget

money? 4 3 2 l 9 8

L12

L13.

 

How well are you
able to cook for 4
yourself?

L14.

 

How well are you
able to do the 4
weekly shopping? 3 2 1 9

L15.

 

How well are you
able to look for a
job? (or wanted to
look for a job)

L16
4;
t»
N
\O
00

 

How well are you
able to wash your 4 3 2 1 9 8
own clothes. .

L17.

 

How well are you
able to take care of 4 3 2 1 9 8
personal hygiene.

L18

 

How well are you
able to clean, tidy 4
your living space?

119

 

 

128

 

 

 

 

 

 

Able Need ,
with Helm"? 12fo Not
- Club Able with current Y a licable
Question Able Help other help receiving) Unsure pp
How well are you
:5 able to purchase 4 3 2 1 9 8
:: items from
"'1 shops?
How well are you
:3 able to leave the 4 3 2 1 9 8
:3 house alone?
. How well are you
2 able to choose 4 3 2 l 9 8
-l and buy clothes?
How well are you
on. able to care for 4 3 2 l 9 8
5 your personal

 

appearance?

 

129

DEMOGRAPHIC INFORMATION (DI)

I would like to find out a little bit about your background. You may skip any
questions that you do not want to answer.

 

 

D11.What is your date of birth?

 

 

 

 

 

Month Day Year
D12.I just want to conﬁrm your gender:
What is your gender? D Female (1)
B Male (2)

D13. How do you describe your race, ethnicity or cultural background?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DI3a. Race/ethnicity D13b. If Hispanic is selected, Mark all
that
apply

Mark one category No (0) Yes (1) Category
Native American (1) Mexican
Asian or Paciﬁc Islander (2) Mexican

American
African American or Black (3) Chicano/Chicana
White (4) Cuban
Hispanic (5) Puerto Rican
Multi-racial (6) Other Spanish
Arab American (7) None
Other (8)
Specify

 

 

 

130

Educational status
D16. How far did you go in school? (Highest level of school completed.) Check
one.

 

/
LEVEL

 

Never went to school (0)

 

1St - 8'h grade (1)

 

9th - 12th grade, but did not graduate (2)

 

High School Diploma (3)

 

GED (4)

 

Some college, less than degree (5)

 

Completed certiﬁcate or license program (such as chef,
plumber, electrician, etc.) (6)

 

2 year college diploma (7)

 

4 year college diploma (8)

 

Master degree (9)

 

 

 

 

Doctoral degree or professional degree (10)

 

131

APPENDIX F

Release for Diagnostic Information

132

Appendix F

Release for Diagnostic Information

Information on Diagnosis

You are participating in the Flinn Clubhouse Project. As part of the Flinn Project, the
team would like to obtain information about your diagnoses to assist in understanding the

effects of clubhouse participation on clubhouse members.

I give the staff of the Flinn Clubhouse Project permission to contact the clubhouse staff to

get information about my diagnoses.

If you have any questions or concerns regarding this project, please call the people who
are in charge of this project, Dr. Esther Onaga at 517/355-0166 or Dr. Sandra Herman
517/335-0130.

If you have any questions and concerns about your rights as a participant of a study,
please call Dr. David Wright, Chair, Michigan State University, University Committee on
Research Involving Human Subjects (UCRIHS) 517/355-2180

 

Participant Name (Print)

 

 

Participant Signature date

133

APPENDD( G

Clubhouse Participation

l 34

Appendix G
Clubhouse Participation

CPl. How long have you been coming to the clubhouse?

months years

 

CP2. On average, how often do you come to the clubhouse in a week?
(How many days a week?)

days a week

CP2a. If less than once a week ask, how often do you come to the club
each month?

times per month

 

CP3. When you come to the clubhouse, how long do you usually stay?

hours

135

APPENDIX H

Social Network Interview*

136

Appendix H
Social Network Interview*

*Scale was formatted to meet graduate school formatting requirements)
Directions:

1. Begin by asking probe question #1 at the bottom of the page. For each name that the
respondent mentions, turn to the next page and write down their ﬁrst name and ﬁrst letter
of their last name. Continue recording responses for this question until the respondent is
ﬁnished and prompt the respondent by asking, “Is there anyone else...?”

2. Continue asking questions 2-4. When you have recorded all the names of the
individuals for the remaining questions and the probe number fro each one, then proceed
to ﬁll in the remaining columns (e.g., sex, know from club, etc.). For each name go
across the row asking each of the questions for that one individual. Do this until the chart
is completely ﬁlled in for each person.

For this next section, I will be asking you questions about your friends and people who
you are close to. I will be asking you to list some of the names of your friends; however, I
will not ask you their full name, just their ﬁrst name and the ﬁrst letter of their last name.
This way you can protect their identity. Do you have any questions?

Probe questions:

1. when you are concerned about a personal matter—for example, something you
are worried about or you are concerned about someone you are close to—who do
you talk with?

<PROMPT for each question>
2. who do you spend your time with, that is — who do you hang out with?
3. who would you ask if you needed to borrow some money?

4. is there anyone else important in your life who you have not mentioned? <Probe>

137

 

N

 

_-
u—I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

o_
a
m
N.
o
w
v
m
N
_
a
slam wicm Clcm alum uncm olcm slam I5.
965 _oozomue ._|_. q M z
H.582: W w w. w
050: macawuo 17V .0,. m.
:m683wnw F W ..u..
boEobxoum 3%me coo—53-8”» H W
boEobxonm boEobxoum E. a oascnv .3qu5me bxooanv 89:2: piano m. m
:3 a Bivnv :3 a BEvHv EBEuvoEnm .5 a ozzvuw bﬁcoﬁnm ta: naonm w w
bamboo—cum bBﬂoooEnm .5 2:: «MN boaﬁouoEum a?» _a=o_mm£oauv m w
.5 2:: «MN .5 2:: «MN =~ 3 Sci .5 2:: «MN a 8:5 3...an Hoe—Eocum i no
o:o=n_ oco:n_ ml 55 =a a ~o=u_ 32 co banal income. “EoEHN m;
wig 03» ~50» 02ml: 88:8 So» 28» S ...l :23 88:8 on I a. omsoamtnzﬁﬂu— .5
so» ow toga moon 2253 55 so» Ba ﬂ Ear—SE 022 so» so» on meS 2523
5:5 >63 .32: 302 355mm Bom Ban on Etc 301 do on». 23>) £5 9 coax—om

 

 

 

 

 

 

138

APPENDIX I

Recovery Assessment Scale

139

Appendix I

Recovery Assessment Scale

RECOVERY ASSESSSMENT SCALE (REC)
[Give Participant Card #4]

Introduction: I am going to read you a list of statements that describe how people sometimes feel about
themselves and their lives. For each statement that I read, I want you to tell me which option on this
card describes the extent to which you agree or disagree with each statement.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Item Question Strongly Disagree Not Agree Strongly Don ’t
Disagree Sure Agree Know
REC]. I have a desnre to l 2 3 4 5 9
succeed.
REC2. I have my own plan for
how to stay or become 1 2 3 4 5 9
well.
REC3. I have goals in life that I l 2 3 4 5 9
want to reach.
REC4. I believe I can meet my 1 2 3 4 5 9
current personal goals.
RECS. 1 have a purpose in life. 1 2 3 4 5 9
REC6. Even when I don’t care
about myself, other I 2 3 4 5 9
people do.
REC7. I understand how to
control the symptoms of l 2 3 4 5 9
my mental illness.
REC8. I4can handle it if 1 get 1 2 3 4 5 9
Sle again.
REC9. I can identify what
triggers the symptoms of 1 2 3 4 5 9
my mental illness.
I can help myself
REC lO become better. 1 2 3 4 5 9
REC] 1. Fear doesn’t stop me
from living the way I l 2 3 4 5 9
want to.
REC 12. I know that there are
mental health services 1 2 3 4 5 9
that do help me.
REC 1 3. There are things that I
can do that help me deal 1 2 3 4 5 9
With unwanted
symptoms.
REC14. I can handle what
happens in my life. I 2 3 4 5 9
REC15. 1 like myself. 1 2 3 4 5 9
REC16. If people really knew
me, they would like me. 1 2 3 4 5 9

 

 

 

 

 

 

 

 

 

 

140

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Item Question Strongly Disagree Not Agree Strongly Don ’t
Disagree Sure Agree Know

REC17. I am a better person than 1 2 3 4 5 9
before my experience
with mental illness.

REC 18. Although my symptoms
may get worse, I know I l 2 3 4 5 9
can handle it.

REC19. IfI keep trying, I Will 1 2 3 4 5 9
continue to get better.

REC20. I have an idea of who 1 l 2 3 4 5 9
want to become.

REC21. Things happen for a 1 2 3 4 5 9
reason.

REC22. Something good Will 1 2 3 4 5 9
eventually happen.

REC23. I am the person most
responsible for my own 1 2 3 4 5 9
improvement.

REC24. I m hopeful about my 1 2 3 4 5 9
future.

REC25. Icontmue to have new 1 2 3 4 5 9
interests.

REC26. 21:15 important to have 1 2 3 4 5 9

REC27. Coping with my mental
illness is no longer the l 2 3 4 5 9
main focus of my life.

REC28. My symptoms interfere
less and less with my 1 2 3 4 5 9
life.

REC29. My symptoms seem to
be a problem for shorter l 2 3 4 5 9
periods of time each time
they occur.

REC30. I know when to ask for 1 2 3 4 5 9
help.

REC31. I am Willlng to ask for 1 2 3 4 5 9
help.

REC32. I ask for help, when I l 2 3 4 5 9
need it.

REC33. Eemg able to work IS 1 2 3 4 5 9
important to me.

REC34. I know what helps me
get better. 1 2 3 4 5 9

REC35. 1 can learn from my
mistakes. I 2 3 4 5 9

REC36. I can handle stress. 1 2 3 4 5 9

REC37. I have people I can count 1 2 3 4 5 9
on.

R . ' '

EC38 I can identify the early 1 2 3 4 5 9

 

warning signs of

becoming sick.

 

 

 

 

 

 

 

141

 

 

 

 

 

 

Item Question Strongly Disagree Not Agree Strongly Don ’t
Disagree Sure Agree Know
REC39. Even when 1 don’t
believe in myself, other 1 2 3 4 5 9
people do.
REC40. It is important to have a 1 2 3 4 5 9
variety of friends.
REC4l. It is important to have
healthy habits. 1 2 3 4 5 9

 

 

 

 

 

 

 

 

142

 

APPENDIX J

Clubhouse Sense of Community

143

Appendix J

Clubhouse Sense of Community

Sense of Community in Clubhouses (SOC)
excerpts from J.C. Buckner, 1988 — SOC Scale (items l-l3) & Clubhouse concept mapping results (items
14-20)

[Give Participant Card #4]

[moducm' n: Now I would like to ask some questions about the clubhouse. For each item that I read,
please tell me if you strongly disagree, disagree, agree, or strongly agree.

 

Question Strongly Disagree Not Sure Agree Strongly Don’t
Disagree Agree Know

Item

 

I feel like I belong to
this clubhouse. 1 2 3 4 5 9

SOCl

 

The friendships and
associations I have
with other people in l 2 3 4 5 9
my clubhouse mean a
lot to me.

SOC2

 

If the people in my
clubhouse were
planning something,
I'd think of it as l 2 3 4 5 9
something “we” were
doing rather than
“they” were doing.

SOC3

 

If I needed advice
about something. I
could go to someone I 2 3 4 5 9
in the clubhouse.

SOC4

 

I think I agree with
most people in my
clubhouse about what 1 2 3 4 5 9
is important in life.

SOCS

 

I feel loyal to the
members in my
clubhouse. 1 2 3

SOC6

 

 

I feel loyal to the staff
in my clubhouse. 1 2 3

 

 

 

 

 

 

SOC7

 

 

 

144

 

Item

Question

Strongly
Disagree

Disagree

Not Sure

Agree

Strongly
Agree

Don’t
Know

 

SOC8

I would be willing to
work together with
others on something
to improve my
clubhouse.

 

SOC9

I plan to remain a
member of the
clubhouse for a
number of years.

 

SOC10

I like to think of
myself as similar to
the people who are
part of this clubhouse.

 

SOCll

A feeling of
fellowship runs deep
between me and staff
in this clubhouse.

 

SOC12

A feeling of
fellowship runs deep
between me and
members in this
clubhouse.

 

SOC13

Being part of this
clubhouse gives me a
sense of community.

 

SOC14

Being part of this
clubhouse helps me to
deal with my mental
illness.

 

SOCIS

Belonging to this
clubhouse helps me
have hope for the
future.

 

SOC16

Being a member of
this clubhouse helps
reduce stigma that I
feel in the greater
community.

 

 

SOC17

 

Being a member of
this clubhouse gives
me a place to go.

 

 

 

 

 

 

 

145

 

 

 

 

 

 

Question Strongly Disagree Not Sure Agree Strongly Don’t
Disagree Agree Know

E
8

Being a member

helps me learn new
§ skills. 1 2 3 4 5 9
VJ

Being a member

helps me get a chance
§ to find paid work. 1 2 3 4 5 9
m

Being a member

gives me something

meaningful to do. 1 2 3 4 5 9

SOC20

 

 

 

 

 

 

 

 

146

 

REFERENCES

Albert, M., Becker, T., McCrone, W. & Thomicorft, G. (1998). Social networks and
mental health service utilization: a literature review. International Journal of
Social Psychiatry, 44, 248-266.

Alpass, F. M., & Neville, S. (2003). Loneliness, health, and depression in older males.
Aging and Mental Health, 7(3), 212-216.

Anthony, W. A. (1993). Recovery from mental illness: the guiding vision of the mental
health service system in the 19905. Psychosocial Rehabilitation Journal, 16(4),
11-23.

Anthony, WC. (2000). A recovery oriented service system: Setting some system level
standards. Psychiatric Rehabilitation Journal, 24(2), 149-158.

Anthony, W., Cohen, M., Farkas, M. & Gagne, C. (2002). Psychiatric rehabilitation, (2nd
edition). Boston, MA: Center for Psychiatric Rehabilitation, Boston University.

Antonucci, T. C., & Israel, B. A. (1986). Veridicality of social support: A comparison of
principal and network members responses. Journal of Consulting and Clinical
Psychology, 54(4), 432-437.

Alexander, J. (2001). Depressed men: An exploratory study of close relationships.
Journal of Psychiatric and Mental health Nursing, 8, 67-75.

Bass, L.A., & Stein, C. (1997). Comparing the structure and stability of network ties
using the social support questionnaire and the social network list. Journal of
Personal Relationships, 14(1), 123-132.

Baumeister, R. F., & Leary, M. R. (1995). The need to belong: desire for interpersonal
attachments as a fundamental human motivation. Psychological Bulletin, 117
(3), 497-529.

Beal, V., & Lambric, T. (1995). The recovery concept: Implementation in the mental
health system. A report by the Community Support Program Advisory
Committee, Ohio Department of Mental Health, Ofﬁce of Consumer Services.

Beard, J.H., Propst, R.N., & Malamud, TJ. (1982). The fountain house model of
psychiatric rehabilitation. Psychosocial Rehabilitation Journal 5(1), 47-53.

Beard, M. L. (1992). Social networks. Psychosocial Rehabilitation Journal, 16,(2), l 1-
116.

Berkman, L. F.(1995). The role of social relations in health promotion. Psychosomatic
Medicine, 57(3), 245-254.

147

Berkman, L.F.(2000). Social support, social network, social cohesion and health. Social
Work in Health Care, 31(2), 3-14.

Berkman, L. F., Glass, T., Brissette, I., & Seeman, T. E. (2000). From social integration
to health: Durkheim in the new millennium. Social Science & Medicine, 51(6),
843-857.

Berren, M. R., Hill, K. R., & Merikel, E. (1994). Serious mental illness and mortality
rates. Hospital and Community Psychiatry, 45(6), 604-605.

Biegel, D. E., McCardle, F... & Mendelson, S. (1985). Social Networks and Mental
Health: An Annotated Bibliography. CA: Sage, Inc.

Biegel, D. E., Tracy, E. M., & Corvo, K. N. (1994). Strengthening social networks:

intervention strategies for mental health case managers. Health and Social Work,
19(3), 206-216.

Birchwood, M., Smith, J ., Cochrane, R., & Wetton, S. (1990). The social functioning
scale: The development and validation of a new scale of social adjustment for

use in family intervention programs with schizophrenic patients. British Journal
of Psychology, 157, 853-859.

Blanch, A., Fisher, D., Tucker, W., Walsh, D., & Chassman, J ., (1993). Consumer-
practitioners and psychiatrists share insights about recovery and coping. Disability
Studies Quarterly, 13(2), 17-20.

Borge,L., Martinsen, E.W., Rudd, T., Watne, O. & Friis, S. (1999). Quality of life,
loneliness, and social contact among long-term psychiatric patients. Psychiatric

Services, 50(1), 81-84.

Bowen, M. (1959). Family relationships in schizophrenia. In Aurerback (Ed.),
Schizophrenia: An Integrated Approach. New York: Ronald Press.

Bowen, M. (1960). A family concept of schizophrenia. In D. Jackson (Ed.), The
Etiology of Schizophrenia. New York: Basic Books.

Bowen, M. (1965). Family psychotherapy with schizophrenia in the hospital and in
private practice. In I. Boszormngyi-Nagy & J.L. Framo (Eds), Intensive Family
Therapy. New York: Harper & Row.

Bowen, M. (1966). The use of family theory in clinical practice. Comprehensive
Psychiatry, 7(5), 345-374.

Bowen, M. (1978). Family Therapy in Clinical Practice. New York: Jason Aronson.

148

Boydell, K. M., Gladstone, B. M., & Crawford, E. S. (2002). The dialectic of friendship
for people with psychiatric disabilities. Psychiatric Rehabilitation Journal 26(2),
123-131.

Brenner, G.F., Norvell, N.K., & Limacher, M. (1989). Supportive and problematic social
interactions: A social network analysis. American Journal of Community
Psychology, 17(6), 831-836.

Brewer, P., Gadsden, K., & Scrimshaw, K. (1994). The community group network in
mental health: a model for social support and community integration. British
Journal of Occupational Therapy. 5 7( 12), 467-470.

Brier, A., & Strauss, J. S. (1984). The role of social relationships in the recovery from
psychotic disorders. American Journal of Psychiatry, 141(8), 949-955.

Broderick, C. B., & Schrader, S. S. (1991). The history of professional marriage and
family therapy. In. A.S. Gurman, & D. P.Kniskem, (Eds.), Handbook of family
therapy volume 11 (pp. 3-40). New York: Brunner/Mazel Publishers.

Bronfenbrenner, U. (1989). Ecological systems theory. Annals of Child Development, 6,
187-249.

Brown, G.W., Birley, J .L., & Wing, J .K. (1972). Inﬂuence of family life on the course of
schizophrenic disorders: A replication. British Journal of Psychiatry, 121, 241-
258.

Bubolz, M.M., & Sontag, S.M. (1993). Human ecology theory. In P.G. Boss, W.J.
Doherty, R. LaRossa, W.R. Schumm, & S.K. Steinmetz, (Eds.), Sourcebook of
family theories and methods (pp.419-447), New York: Plenum Press.

Buckner, J .C. (1988). The development of an instrument to measure neighborhood
cohesion. American Journal of Community Psychology, 16(6), 771-791.

Calsyn, R.J., & Winter, J .P. Social support, psychiatric symptoms, and housing: A causal
analysis. Journal of Community Psychology, 30(3), 247-259.

Campbell, J .,& Schraiber, R. (1989). In Pursuit of wellness: The well being project.
Sacramento, CA: California Department of Mental Health.

Campbell, J., & Leaver, J. (2003). Emerging New Practices in Organized Peer
Support (National Association of State Mental Health Program Directors and the
National Technical Assistance Center for State Mental Health, March 17-18,
2003, Alexandria, VA.

149

Campbell, I. (2005). Consumer-Operated Services Program: Multi-site Research
Initiative(Michigan Association of Community Mental Health Boards).
Baltimore, MD: National Research Institute.

Cohen, J. (1992). A power primer. Psychological Bulletin, 112(1), 155-159.

Cobb, S. (1976). Social support as a moderator of life stress. Psychosomatic Medicine,
38(5), 300-314.

Cohen, C. I., & Adler, A. (1986). Assessing the role of social network interventions with
an inner-city population. American Journal of Orthopsychiatry, 56(2), 278-
288.

Cook. J. A., & Hoffschmidt, S. J. (1993). Comprehensive models of psychosocial
rehabilitation. In R.W. Flexer and PL. Solomon (Eds.), Psychiatric Rehabilitation
in Practice (pp. 81-97). Boston: Andover Medical Publishers.

Cook, J. A., Lefley, H. P., Pickett, S. A., & Cohler, B. J. (1994). Age and family burden
among parents of offspring with severe mental illness. American Journal of
Orthopsychiatry, 64(3), 435-447.

Corrigan, P. W. (2005). Mental illness stigma as social injustice: yet another dream
to be achieved. In P. W. Corrigan (Ed.), 0n the stigma of mental illness:
Practical strategies for research and social change (pp. 315-320). Washington,
DC: American Psychological Association.

Corrigan, P. W., Giffort, D., Rashid, F., Leary, M., & Okeke, I. (1999). Recovery as a
psychological construct. Community Mental Health Journal, 35(3), 231-239.

Corrigan, P. W., & Phelan, S. M. (2004). Social support and recovery in people with
serious mental illness. Community Mental Health Journal, 40(6), 513-523.

Corrigan, P. W., & Ralph, R. O. (2005). Introduction: recovery as consumer vision and
research Paradigm. In R.O. Ralph & P. W Corrigan, (Eds) Recovery in mental
illness: Broadening our understanding of wellness, (pp. 3-17). Washington, DC:
American Psychological Association.

Corrigan, P. W., Salzer, M., Ralph, R. 0., Sangster, Y., & Keck, L. (2004). Examining
the factor structure of the recovery assessment scale. Schizophrenia Bulletin 30
(4), 1035-1041.

D’Augelli, A. (1983). Social support networks in mental health. In J. Whittaker, and J.
Garbarino, Social Support Networks: Informal Helping in the Human Services,
(pp.73-106). New York: Aldine De Gruyter.

150

Davidson, L., Hoge, M. A., Godleski, L., Rakfeldt, J ., Grifﬁth, E. E., & Merrill, M. E.
(1996). Hospital or community living? Examining consumer perspectives on
deinstitutionalization. Psychiatric Rehabilitation Journal, 19(3), 49-58.

Davidson, L., & McGlashan, T. H. (1997). The varied outcomes of schizophrenia.
Canadian Journal of Psychiatry, 42(1), 34-43.

Davidson, L., Sells, D., Sangster, 8., O’Connell, M (2005). Qualitative studies of
recovery: What can we learn from the person? In R. 0. Ralph, P. W. Corrigan,
(Eds.), Recovery in mental illness: Broadening our understanding of wellness
(pp. 147-170). Washington, DC: American Psychological Association.

Davidson, L., Shahar, G., Strayner, D. A., Chinman, M. J ., Rakfeldt, J ., & Kraemer
Tebes, J. (2004). Supported socialization for people with psychiatric disabilities:

Lessons from a randomized controlled trial. Journal of Community Psychology,
32(4), 453-477.

Davidson, L., & Stayner, D. (1997). Loss, loneliness, and desire for love: Perspectives
on the social lives of people with schizophrenia. Psychiatric Rehabilitation
Journal, 20(3) 3-12.

Deegan, G. (2003). Discovering recovery. Psychiatric Rehabilitation Journal, 26(4)
368-376.

Deegan, P. E. (1988). Recovery: The lived experience of rehabilitation. Psychosocial
Rehabilitation Journal, 11(4), 11-19.

DeMasi, M. E., Markowitz, F. E., Videka-Sherrnan, L., Soﬂta, C., Knight, E. L., &
Carpinello, S. E. (1996). Specifying dimensions of recovery. Proceedings:6"'
Annual national Conference on State Mental Health Agency Services Research

and Program Evaluation. Alexandria, VA: National Association of State Mental
Health Program Directors (NASMHPD) Research Institute.

DeSisto, M. Harding, C. M., McCormick, R.V., Ashikaga, T., & Brooks, G.W. (1999).
The Maine and Vermont three-decade studies of serious mental illness:
Longitudinal course comparisons. British Journal of Psychiatry, 167, 331-341.

Dickerson, F. B., Ringel, N., & Parente, F. (1999). Predictors of residential independence
among outpatients with schizophrenia. Psychiatric Services, 50(4), 515-519.

Doane, J. (1991). Emotion and attachment in families of schizophrenia: The struggle for

recovery. In E. F. Walker (Ed.), Schizophrenia: A Life Course Developmental
Perspective (pp. 213-241). San Diego: Academic Press, Inc.

151

Elizur, J ., & Minuchin, S. (1989). Institutionalizing madness: Families,
therapy, and society. BasicBooks.

Ely, I. (1992). Closeness in clubhouse relationships. Psychosocial Rehabilitation Journal,
16(2), 61-62.

Estroff, SE. (1981). Making it crazy: An ethnography of psychiatric clients in an
American community. Berkeley: University of California Press.

Fadden, G., Bebbington, P., & Kuipers, L. (1987). The burden of care: The impact of
functional psychiatric illness on the patient’s family. British Journal of
Psychiatry, 150, 285-292.

Falloon, I. R., Royd, J. L., et a1. (1985). Family management in the prevention of
morbidity of schizophrenia: clinical outcome of a two year longitudinal study.
Archives of General Psychiatry, 42(9), 887-896.

Frese, F.J., Stanley, J ., Kress, K., & Vogel-Schibilia, S. (2001). Integrating evidence-
based practices and the recovery model. Psychiatric Services, 52(11), 1462-1468.

Fisk, D., Rowe, M., Laub, D., Calvocoressi, L., & DeMino, K. (2000). Homeless persons
with mental illness and their families: Emerging issues from clinical work.
Families in Society, 81(4), 351-359.

Friedman, EH. (1991). Bowen theory and therapy. In A.S. Gurman & D.P. Kniskem,
(Eds.), Handbook of family therapy, Volume II (pp.134-l70). New York, NY:
Brunner/Mazel.

Froland, C. (1978). Improving the social adjustment of mental health clients: the case for
social support networks. Doctoral dissertation, school of public health, University
of California, Berkeley.

Froland, C., Brodsky, 6., Olson, M., & Steward, L. (2000). Social support and
adjustment: Implications for mental health professionals. Reprinted from 1979.
Community Mental Health Journal, 36(1) 61-75.

Glickman, M. (1992). The voluntary nature of the clubhouse. Psychosocial Rehabilitation
Journal, 16(2), 39—40.

Goldberg, R. W., Rollins, A. L., & Lehman, A. F. (2003). Social network correlates
among people with psychiatric disabilities. Psychiatric Rehabilitation Journal,
26(4), 393-402.

Green, G., Hayes, C., Dickinson, D., Whittaker, A., & Gilheany, B. (2002). The role and

and impact of social relationships upon well being reported by mental health
users: a qualitative study. Journal of Mental Health, 11(5), 565-5 79.

152

Greene, G. J ., Hamilton, N., & Rolling, M. (1986). Differentiation of self and psychiatric
diagnosis: an empirical study. Family Therapy, 13(2), 187-194.

Griffore, R. J., & Phenice, L. A. (2001). The language of human ecology: A general
systems perspective. Dubuque, IA: Kendall/Hunt Publishing Company.

Harding, C. M. (1986) Speculations on the measurement of recovery from severe
psychiatric disorder and the human condition. Psychiatric Journal of the
University of Ottawa, 11(4), 199-204.

Harding, C. M., Brooks, G. W., Ashikaga, T., Strauss, J. S., & Breier, A. (1987). The
Vermont longitudinal study of persons with severe mental illness, I: methodology,

study sample, and overall status 32 years later. American Journal of Psychiatry,
144,(6), 718-726.

Harding, C. M., Brooks, G. W., Ashikaga, T., Strauss, J. S., (1987a). The Vermont
longitudinal study of persons with severe mental illness: 11. Long-term outcome of

subjects who retrospectively met DSM-III criteria for schizophrenia. American
Journal of Psychiatry, 144(6), 727-735.

Harding, C. M., Brooks, G. W., Ashikaga, T., Strauss, J. S., (1987b). The Vermont
longitudinal study of persons with severe mental illness: 1. Methodology, study

sample, and overall status 32 years later. American Journal of Psychiatry, 144(6),
718-726.

Harding, C. M., & Zahniser, J. H. (1994). Empirical correction of seven myths about
schizophrenia with implications for treatment. Acta Psychiatrica Scandinavica
Supplementum, 90, 140-146.

Hardiman, E. R., & Segal, S. P. (2003). Community membership and social networks in
mental health self-help agencies. Psychiatric Rehabilitation Journal, 27(1), 25-
33.

Hatﬁeld, A. B., & Lefley, HP. (1987). Families of the mentally ill: Coping and
adaptation. New York: Guilford.

Hogan, F. (2003). The President's New Freedom Commission: Recommendations to
transform mental health care in America. Psychiatric Services, 54(11), 1467-
1474.

Herrnan, S. (1997). Final Progress Report to the National Institute of Mental Health on
Enhancing State Research Capacity: Effectiveness of Specialized Inpatient
Treatment for Mental Illness and Substance Abuse: Grant # R01MH46307,
Lansing: Michigan Department of Community Health.

153

Herman, S., Onaga, E., Pemice-Duca, F., Oh, S., & Ferguson, C. (2005). Sense of
community in clubhouse programs: Member and staff concepts. American
Journal of Community Psychology, in press.

Herman, S., Onaga, E., Ferguson, C., Pemice-Duca, F., Oh, 8., & Weaver Randall, K.
(2002). The Flinn clubhouse project: An evaluation report. Lansing, MI:
Michigan Department of Community Health.

Hirsch, B.J. (1981). Social networks and the coping process: Creating personal
communities. In B.H. Gottlieb (Ed.), Social networks and social support (pp. 149-
170). Beverly Hills: Sage Publications.

Holmes-Eber, P., & Riger, S. (1990). Hospitalization and the composition of mental
patients’ social networks. Schizophrenia Bulletin, 16(1), 157-164.

House, J .S., Landis, K.R., Umberson, D. (1988). Social relationships and health.
Science, 241 (4865), 540-545.

International Center for Clubhouse Development. (1998). Standards for Clubhouse
Programs. New York, NY.

Jacobson, N ., & Greenley, D. (2001). What is recovery? A conceptual model and
explication. Psychiatric Services, 52(4), 482-485.

Kaffman, M. (1989). Community management of mental illness: an alternative to
hospitalization. In J. Elizur & S. Minuchin, (Eds.), Institutionalizing madness
(pp. 195-242). BasicBooks.

Kerr, M.E., & Bowen, M. (1988). Family evaluation: An approach based on Bowen
theory. New York, NY: Norton & Co.

Klem, L. (2000). Path Analysis. In L.G. Grimm & P. R. Yarnold (Eds.) Reading &
Understanding multivariate statistics (pp. 65-98). Washington, DC: American
Psychological Association.

Knudson-Martin, C. (1994). The female voice: Applications to Bowen’s family systems
theory. Journal of Marital and Family Therapy, 20(1), 35-46.

Knudson-Martin, C. (1996). Differentiation and self-development in the relationship
context. The Family Journal, 4, 188-198.

Kouzis, A., Eaton, W. W., & Leaf, P. J. (1995). Psychopathology and mortality in the

general population. Social Psychiatry and Psychiatric Epidemiology, 30(4), 165-
170.

154

Kruger, A. (2000). Schizophrenia: Recovery and hope. Psychiatric Rehabilitation
Journal 24(1), 29-37.

Landis, J. B. (1999). Estimating net effects and costs of service options for persons with
serious mental illness. Psychiatric Services, 50(6), 735-738.

Learner, M. & Fairweather, G.W. (1963). Social behavior of chronic schizophrenics in
supervised and unsupervised work groups. Journal of abnormal & Social
Psychology. 67(3), 219-225.

Leff, J .P. (1976). Schizophrenia and sensitivity to the family environment. Schizophrenia
Bulletin, 2(4), 566-574.

Lehman, A.F., Ward, N .C., & Linn, LL (1982). Chronic mental patients: The quality of
life issue. American Journal of Psychiatry, 139(10), 1271-1276.

Liberrnan, R. P., Kopelowicz, A., Ventura, J ., & Gutkind, D (2002). Operational criteria
and factors related to recovery from schizophrenia. International Review of
Psychiatry, 14(4), 256-272.

Liberrnan, R. P., & Kopelowicz, A. (2005). Recovery from schizophrenia: A criterion
based deﬁnition. In R. 0. Ralph & P. W. Corrigan (Eds.), Recovery in mental
illness: Broadening our understanding of wellness (pp. 101- 129). Washington,
DC: American Psychological Association.

Loveland, D., Weaver Randall, K., & Corrigan, P.W. (2005). Research methods for
exploring and assessing recovery. In R.O. Ralph & P.W. Corrigan, (Eds.),
Recovery in mental illness: Broadening our understanding of wellness (pp. 19-
60). Washington, DC: American Psychological Association.

Lovell, A. M., Barrow, W., & Hammer, M. (1984). Social support and social network
interview. New York: New York State Psychiatric Hospital.

Lin, N., Xiaolan, Y., & Ensel, W.M. (1999). Social support and depressed mood: a
structural analysis. Journal of Health and Social Behavior, 40(4), 344-359.

McMillian, D.W., & Chavis, D.M. (1986). Sense of community: A deﬁnition and theory.
Journal of Community Psychology, 14, 6-23.

Macdonald, E., Sauer, K., Howie, L., & Albiston, D. (2205). What happens to social
relationships in early psychosis? A phenomenological study of young people’s
experiences. Journal of Mental Health, 14(2) 129-143.

Macias, C., Jackson, R., Schroeder, C., & Wang, Q. (1999). What is a clubhouse? Report

on the ICCD 1996 survey of USA clubhouses. Community Mental Health
Journal, 35(2), 181-190.

155

Macias, C., Barreira, P., Alden, M., & Boyd, J. (2001). The ICCD benchmarks for
clubhouses: A practical approach to quality improvement in psychiatric
rehabilitation. Psychiatric Services, 52(2), 207-213.

Macias, C., & Rodican, C. (1997). Coping with recurrent loss in mental illness: Unique

aspects of the clubhouse communities. Journal of Personal & Interpersonal Loss
2, 205-221.

Markowitz, F. E. (2001). Modeling processes in recovery from mental illness:
Relationships between symptoms, life satisfaction, and self-concept. Journal of
Health and Social Behavior, 42(1), 64-79.

Marsh, D. T., Dickens, R. M., Yachovich, N. S., Wilson, J. M., Leichliter, J. S., &
McQuillis, V. A. (1993). Troubled journey: siblings and children of people with
mental illness. Innovations and Research, 2(2), 17-28.

Mastboom, J. (1992). Forty clubhouses: model and practices. Psychosocial
Rehabilitation Journal, 16(2), 9-23.

McCallister, L., & Fischer, C. S. (1983). A procedure for surveying personal networks. In
R.S. Burt & M.J. Minor (Eds.), Applied Network Analysis: A Methodological
Introduction, 75-88. Beverly Hills, CA: Sage.

McFarlane, W.R., Dixon, L., Lukens, E., & Lucksted, A. (2003). Family
psychoeducation and schizophrenia: a review of the literature. Journal of Marital
and Family Therapy, 29(2) 223-245.

Michigan Department of Community Health (2000). Division of Vital Records and
Health Statistics. Lansing, MI.

Milardo, RM. (1988). Families and social networks: An overview of theory and
methodology. In. R.M. Milardo (Ed.), Families and social networks (pp.13-47).
Newbury Park, CA: Sage Publications.

Mirowsky, J., & Ross, C. (1986). Social patterns of distress. Annual Review of
Sociology, 12, 23-45.

Mowbray, C., & Tan, C. (1992). Evaluation of an innovative consumer-run service
model; The drop in center. Innovations and Research, 1(2), 33-42.

Mueser, K. T., Corrigan, P. W., Hilton, D., Tanzman, B., Schaub, A., Gingerich, S.,
Essock, S. M., Tarrier, N ., Morey, B., Vogel-Schibilia, & Herz, M. I. (2002).
Illness management and recovery: a review of the research. Psychiatric
Services 53(10), 1272-1284.

156

Murray C. L., & Lopez, A. D. (1996). The global burden of disease: a comprehensive
assessment of mortality and disability from diseases, injuries, and risk factors in
19990 and projected to 2020. Cambridge, MA: Harvard School of Public Health
on behalf of the World Health Organization and the World Bank, Harvard
University Press.

Narrow, W. E. (1998). One year prevalence of depressive disorders among adults 18 and
over in the US: National Institute of Mental Health Encatchment Study.
Unpublished.

Narrow, W. E., Rae, D. S., & Regier, D. A., (1998). National Institute of Mental Health
epidemiological study: One-year prevalence best estimates calculated from
Encatchment Area Study. Unpublished.

National Institute of Mental Health. (1980). Announcement of community support system
strategy development and implementation grants. Rockville, MD.

Newton, L. (2001). Self and illness: Changing relationships in response to life in the
community following prolonged institutionalization. The Australian Journal of
Anthropology, 12(2) 166-181.

Nichols, W.C., & MacFarlane, M.M. (2001). Family therapy and mental health:
historical overview and current perspectives. In M.M MacFarlane (Ed.), Family
therapy and mental health: innovations in theory and practice (pp. 3-20).New
York, NY: Hawthorn Clinical Practice Press.

Pagel, M.D., Erdly, W.W., & Becker, J. (1987). Social networks: We get by with (and in
spite of) a little help from our friends. Journal of Personality and Social
Psychology, 53(4), 793-804.

Paul, D. (1992). From crises to career: My story. Rehabilitation Journal, 16(2), 87-90.
Peckhoff, J. (1992). Patienthood to personhood. Rehabilitation Journal, 16(2), 5-8.

Penn, D.L., & Mueser, K.T. (1996). Research update on the psychosocial treatment of
schizophrenia. American Journal of Psychiatry, 153(5), 607-617.

Perese, E. F., Getty, C., & Wooldridge, P. (2003). Psychosocial club members’
characteristics and their readiness to participate in a support group. Issues in
Mental Health Nursing, 24, 153-174.

Perlick, D. A., Rosenheck, R. A., Clarkin, J. F., Salahi, J ., Strueining, E. L., & Link, B.
G. (2001). Adverse effects of perceived stigma on social adaptation of persons
diagnosed with bipolar affective disorder. Psychiatric Services 52(12), 162 7-
1632.

157

Pescosolido, B. A. (2001). The role of social networks in the lives of people with
disabilities. In G.L. Albrecht, K.D. Seelman & M.Bury (Eds.), Handbook of
Disability Studies, 468- 489.

Pettie, D., & Triolo, A.M. (1999). Illness as evolution: the search for identity and

meaning in the recovery process. Psychiatric Rehabilitation Journal, 22(3), 255-
262.

Phillips, S. L. (1981). Network characteristics related to the well-being of normals: A
comparative base. Schizophrenia Bulletin, 7(1), 117-123.

Pickens, J, M. (2003). Formal and informal social networks of women with serious
mental illness. Issues in Mental Health Nursing, 24(2), 109-127.

Propst, R. (1992). The standards for clubhouse programs: why and how they were
developed. Psychosocial Rehabilitation Journal, 16(2), 25-30.

Ralph, R. 0., Kidder, K., & Phillips, D. (2000). Can we measure recovery? A
compendium of recovery and recovery related instruments. The evaluation center
at the Human Services Research Institute, Cambridge, MA.

Ralph, R. 0., & Muskie, E. (2000). Review of recovery literature: A synthesis of a
sample of recovery literature 2000. Report prepared for National Technical
Assistance Center for State Mental Health Planning (NTAC), and National
Association for State Mental Health Program Directors (NASMHPD).

Ralph, R. O, Corrigan, P. W. (Eds.) (2005). Recovery in mental illness: Broadening our
understanding of wellness. Washington, DC: US American Psychological
Association.

Rands, M. (1988). Changes in social networks following marital separation and divorce.
In R.Milardo, (Ed.). Families and social networks (pp. 107-126), Newbury Park,
CA: Sage Publications.

Resnick, S. G., Rosenheck, R. A., & Lehman, A. F. (2004). An exploratory analysis of
correlates of recovery. Psychiatric Services, 55(5), 540-547.

Ribisl, K.M. (1995). A longitudinal study of the personal networks of individuals
receiving treatment for mental illness and substance abuse problems.
Unpublished doctoral dissertation, Michigan State University, East Lansing,
Michigan.

Rice, D.P., & Miller, LS. (1996). The economic burden of schizophrenia: Conceptual
and methodological issues, and cost estimates. In M. Moscarelli, & A. Rupp
(Eds.), Handbook of mental health economics and health policy, (Vol.1:
Schizophrenia). 321-334. Oxford, England.

158

Rice, D.P., & Miller, LS. (1998). Health economics and cost implications of anxiety and
other mental disorders in the United States. British Journal of Psychiatry,
1 73(34), 4-9.

Ridgeway, P. A., (1999). Re-storying psychiatric disability: Learning from ﬁrst person
narrative accounts of recovery. Lawrence, KS: University of Kansas School of
Social Welfare.

Rogers, S. B, Anthony, W., & Lyass, A. (2004). The nature and dimensions of social
support among individuals with severe mental illnesses. Community Mental
Health Journal, 40(5), 437-450.

Rosenﬁeld, S., & Wenzel, S. (1997). Social networks and chronic mental illness: A test
of four perspectives. Social Problems, 44(2), 200-216.

Ryan, R. M., & Solky, J. A. (1996). What is supportive about social support? In G. R.
Pierce, B. R. Sarason, & LG. Sarason, (Eds.), Handbook of Social Support and the
Family, (pp.249-267), New York: Plenum Press.

Seidman, L]. (1983). Schizophrenia and brain dysfunction: An integration of recent
neurodiagnostic ﬁndings. Psychological Bulletin, 94(2), 195-238.

Seidman, L.J., Sokolove, R.L., McElroy, C., Knapp, P.H., & Sabin, T. (1987). Lateral
ventricular size and social network differentiation in young, non-chronic
schizophrenic patients. American Journal of Psychiatry, 144(4), 512-514.

Skowron, E. A. & Friedlander, M. L. (1998). The differentiation of self
inventory: Development and initial validation. Journal of Consulting and Clinical
Psychology, 45(3), 235-246.

Spaniol, L., Zipple, A. M., Marsh, D. T., & Finely, L. Y (Eds.) (2000). The role of the
family in psychiatric rehabilitation. Boston, MA: Center for Psychiatric
Rehabilitation, Boston College.

Stein, C. H., Rappaport, J., & Seidman, E. (1995). Assessing the social networks of
pe0ple with psychiatric disability from multiple perspectives. Community Mental
Health Journal, 31(4), 351-367.

Stein, L. 1., Barry, K. L., Van Dien, G., Hollingsworth, E. J ., & Sweeney, J. K. (1999).
Work and social support: A comparison of consumers who have achieved stability

in ACT and clubhouse programs. Community Mental Health Journal, 35(2),
193-204.

Tabachnick, B., & Fidell, L. (1996). Using multivariate statistics. New York:
HaperCollins.

159

Tempier, R., Caron, J., Mercier, C., & Leouffre, P. (1998). Quality of life of severely
mentally ill individuals: A comparative study. Community Mental Health
Journal, 34(5), 477-485.

Thalheimer, W. & Cook, S. (2002, August). How to calculate effect sizes from published
research: A simpliﬁed methodology. Retrieved March 22, 2005. http://work-
learningcom/effect-sizes.htm.

Tittler, B. I. (1998). Family systems treatment of depression. In P.Titleman, (Ed.)

Clinical Applications of Bowen Family Systems Theory. Haworth Press, New
York, NY.

Torrey, W.C., & Wyzik, P. (2000). The recovery vision as a service improvement guide
for community mental health center providers. Community Mental Health
Journal, 36(2), 209-216.

Townsend, W., & Glasser, N. (2003). Recovery: the heart and soul of treatment.
Psychiatric Rehabilitation Journal, 27(1), 83-26.

Turkat, D., (1980). Social networks: theory and practice. Journal of Community
Psychology, 8(2), 99-109.

Turner, J. C., & TenHoor, W. J. (1978). The NIMH community support program: Pilot
approach needed to a social reform. Schizophrenia Bulletin, 4(3), 319-349.

Uchino, B. N., Holt-Lunstad, J ., Smith, T. W., & Bloor, L. (2004). Heterogeneity in
social networks: a comparison of different models linking relationships to

psychological outcomes. Journal of Social and Clinical Psychology, 23(2), 123-
139.

US. Department of Health and Human Services. (1999). Mental health: A report of the
Surgeon General. Rockville, MD: US. Department of Health and Human
Services, Substance Abuse and Mental Health Services Administration, Center for
Mental Health Services, National Institutes of Health, National Institutes of
Mental Health. Satcher, D.

US. Department of Health and Human Services (2001). Report of a Surgeon General ’s
working meeting on the integration of mental health services and primary health
care. Surgeon General's Working Meeting on the Integration of Mental Health
Services and Primary Health Care (30 November - 1 December 2000, Atlanta,
Georgia) Rockville, MD, United States. Public Health Service. Ofﬁce of the
Surgeon General

Vaughn, C.E., & Leff, J .P.( 1981). Patterns of emotional response in relatives of

160

schizophrenic patients. Schizophrenia Bulletin, 7(1), 43-44.

Vaux, A. (1988). Social support: Theory, research, and intervention. New York: Praeger
Publishers.

Walsh, F. (1996). Mental illness: What have we learned? In B. Abosh & A. Collins
(Eds.), Mental illness in the family (pp. 26-37). Toronto: University of Toronto
Press.

Walsh, J., & Connelly, P. R. (1996). Supportive behaviors in natural support networks of
people with serious mental illness. Health & Social Work, 21(4), 296-303.

Walsh, F., & Scheinkman, M. (1989). (Fe)male: The hidden gender dimension in models
of family therapy. In M. McGoldrick, C.M. Anderson, & F. Walsh, (Eds.),

Women in families: A framework for family therapy (pp. 16-41). New York:
W.W. Norton & Co.

Warner, R., Huxley, P., & Berg, T. (1999). An evaluation of the impact of clubhouse
membership on quality of life and treatment utilization. The International Journal
of Social Psychiatry, 45(4), 310-320.

Wasow, M.(1995). The skipping stone: ripple effects of mental illness on the family.
Palto Alto, CA.: Science and Behavior Book, Inc.

Wasylenki, D., James, S., Clark, G, Lewis, J ., Goering, P., & Gillies, L. (1992). Clinical
issues in social network therapy for clients with schizophrenia. Community
Mental Health Journal, 28 (5), 427-440.

Whitaker, C.A., & Keith, D.V. (1981). Symbolic-experiential family therapy. In A.S.
Gurman, & D.P. Kniskem (Eds.), Handbook of family therapy (pp. 187-224).
New York: Brunner/Mazel.

White, M., & Epston, D. (1990). Narrative means to therapeutic ends. New York: WW.
Norton & Company.

Whittaker, J .K., & Garbarino, J. (Eds.) (1983). Social support networks: Informal
helping in the human services .New York: Aldine De Gruyter.

WHO (2001). Mental health: new understanding, new hope. The World Health Report.
Geneva, World Health Organization.

Wilson, M. E., Flanagan, S., & Rynders, C.(1999). The friends program: a peer support
group model for individuals with a psychiatric disability. Psychiatric
Rehabilitation Journal, 22(3), 239-247.

161

Wood, P. A., Hurlburt, M. S., Hough, R. L., & Hofstetter, C. R. (1998). Longitudinal
assessment of family support among homelessness mentally ill participants in
support housing program. Journal of Community Psychology, 26(4), 327-344.

Zipple, A. M., Spaniol, L., & Rogers, S. (1990). Training mental health professionals to

assist families of persons who are mentally ill. Rehabilitation Psychology, 35(2),
121-129.

162

rlllllllllill