Community Ment Health J (2012) 48:490–502

DOI 10.1007/s10597-011-9440-7

ORIGINAL PAPER

Merging the Fields of Mental Health and Social Enterprise:

Lessons from Abroad and Cumulative Findings
from Research with Homeless Youths
Kristin M. Ferguson

Received: 5 May 2010 / Accepted: 24 September 2011 / Published online: 2 October 2011
Ó Springer Science+Business Media, LLC 2011

Abstract Despite the growing integration of supported (Baron and Hartnagel 1997). Faced with difficulty in
employment within the mental health system in the United obtaining and maintaining employment, many homeless
States as well as the widespread use of social enterprises young people rely instead on informal income sources,
abroad, the fields of mental health and social enterprises both legal (e.g., selling recycled materials or self-made
remain largely separate in the USA. The mental health field items) and illegal (e.g., theft and selling drugs) (Gaetz and
currently lacks a response that strengthens homeless O’Grady 2002; Greene et al. 1999; Kipke et al. 1997;
youths’ existing human and social capital, provides them Whitbeck 2009). Compounding these behaviors, homeless
with marketable job skills and employment, and impacts youths also have histories of depression, trauma, substance
their mental health. To address this gap, this paper estab- abuse, low self-esteem, and physical and sexual abuse
lishes a case for using social enterprises with homeless (Cauce et al. 2000).
youths, drawing on both global precedents and findings Despite the myriad employment and mental health
from a mixed-methods study of a social enterprise inter- issues homeless youths confront, prior research indi-
vention with homeless youths. Recommendations are cates their limited service utilization (Kipke et al. 1997).
offered for how to integrate social enterprises with mental De Rosa et al. (1999) found very low engagement rates
health treatment as well as how to evaluate their impact on among this population in medical services (28%), sub-
mental health outcomes. stance abuse treatment (10%) and mental health services
(9%). Socially, homeless youths often have low levels of
Keywords Homeless youths
Mental health
Severe formal human capital (i.e., educational achievement)
mental illness
Social enterprise
Supported employment (Coleman 1990; Thompson et al. 2002) and social capital
(i.e., networks, connections and associations with groups
that provide both tangible and potential support and access
Introduction to resources) (Bourdieu 1993). Given their limited service
use, combined with high-risk behaviors, mental health
There are an estimated 3 million homeless youths and issues and low levels of formal human capital and social
young adults in the United States (Whitbeck 2009). capital, these youths are at risk for developing new and
Unemployment rates among this population range from 66 exacerbating existing mental health illnesses (Cauce et al.
to 71% (Ferguson and Xie 2008; Whitbeck 2009) and can 2000). In the absence of social resources necessary for the
be chronic, with many homeless young people averaging formation of human capital, homeless youths also experi-
more than 8 out of 12 months without work in a given year ence labor-market disadvantage and increased likelihood of
labor exclusion (Gaetz and O’Grady 2002).
The traditional service-delivery method with homeless
K. M. Ferguson (&) youths is the outreach model, which consists of mobile and
Silberman School of Social Work at Hunter College,
fixed-site services (Kipke et al. 1997). In each approach,
The City University of New York, 2180 3rd Avenue,
New York, NY 10035, USA providers offer health, mental health, vocational and social
e-mail: [email protected] services in the streets or in nonresidential drop-in centers.

123
Community Ment Health J (2012) 48:490–502 491

The role of the outreach worker in both approaches is to Thus, the purpose of this paper is to examine the current
connect youths with resources, provide counseling, and fields of mental health and social enterprise and to establish
foster an open and trusting relationship with them. a case for using social enterprises within the United States’
Although the outreach model aims to mitigate the health mental health system with homeless youths. Social enter-
and mental health problems that constitute barriers to prises operated within non-profit agencies offer an alter-
employment, vocational and mental health services remain native to existing vocational and clinical services for
largely separate (De Rosa et al. 1999; Kipke et al. 1997). homeless youths by providing them with employment
The outreach model also fails to recognize the youths’ training and placement that is integrated with mental health
entrepreneurial skills and to replace their street-survival treatment, all within a supportive and familiar setting
behaviors with other legal, income-generating activities. (Krupa 1998). The theoretical rationale for the relevance of
Instead, the outreach model focuses on providing basic social enterprises to mental health treatment draws on
services and can in effect sustain the youths’ livelihoods evidence-based supported employment data with adults,
and survival behaviors on the streets (Ferguson 2007). global precedents of social enterprises with at-risk youths,
Other traditional job-training programs, such as Job asset-based youth development theory, and preliminary
Corps, focus on job skills and placement yet do not have a findings from a mixed-methods study of a social enterprise
mental health component. These programs thus fail to intervention with homeless youths.
address the mental health issues that can hinder homeless
youths in obtaining employment. Likewise, transitional-
living programs are reported to positively influence Field of Mental Health
homeless youths’ mental health, wages, savings, housing
and employment (Rashid 2004). Yet often these programs Combining vocational rehabilitation and mental health
impose entry restrictions, such as requiring that clients do services in the USA for clients with severe mental illness
not have mental health problems that would preclude them dates back to the 1980s, when supported employment
from maintaining employment. As such, many of the models were introduced within the mental health system by
existing vocational programs are tailored to the most highly leaders in the psychiatric rehabilitation movement (Becker
functioning and engaged homeless youths with the least and Drake 2003). Supported employment was originally
severe mental health and substance abuse problems. used in vocational rehabilitation to rapidly place individ-
Two well-known employment models that integrate uals with mental illness in competitive employment and
vocational and clinical services for adults with mental ill- subsequently provide them with individualized vocational
ness are supported employment and social enterprises training and clinical support to assist them in maintaining
(Becker and Drake 2003; Warner and Mandiberg 2006). To employment (Wehman and Moon 1988). This ‘‘place-
date however, there has been limited research examining train’’ model came to replace the previously used stepwise
these employment models with homeless young people ‘‘train-place’’ approaches characteristic of pre-employment
with mental illness (Ferguson and Xie 2008; Nuechterlein training (e.g., skills training, sheltered workshops, trial
et al. 2008; Rinaldi et al. 2004). Not only do these young work programs) (Becker and Drake 2003).
people face illness-related barriers to employment common Supported employment research over the past few dec-
among housed youths with mental illness (e.g., stigma and ades indicates that integration of mental health and voca-
poor service coordination) (Carter and Wehby 2003), they tional rehabilitation produces better employment outcomes
also encounter challenges inherent to living on the streets for clients than when these services are offered in separate
(e.g., lack of housing, personal hygiene issues, food inse- settings (Becker et al. 2001; Becker and Drake 2003; Bond
curity and criminal records) (Kipke et al. 1997). Successful et al. 2001; 2007; Cook et al. 2005; Drake et al. 1999;
strategies to move homeless young people into formal Salyers et al. 2004). Clients who participate in vocational
employment therefore require individualized and long-term rehabilitation programs with integrated and coordinated
supports combining both clinical and vocational services, clinical services report improvements in relationships, self-
as their labor participation is often hindered by the dual esteem, hope and life satisfaction, in addition to gains in
challenges of mental illness and homelessness. Yet to date, income, work hours, work functioning and employability
in order to reach the population of homeless youths who (Bond et al. 2001; Cook et al. 2005; Drake et al. 1999;
are least connected with services, who have the most Salyers et al. 2004). Findings also reveal that clients who
pressing mental health needs, and who participate in the receive greater amounts of employment-specific vocational
most high-risk behaviors, the mental health field currently services and who remain for longer periods of time in
lacks a response that strengthens these youths’ human and vocational programs demonstrate better outcomes than
social capital, provides them with job skills and employ- those who receive fewer vocational services for shorter
ment, and impacts their mental health. periods (Cook 2006). Additionally, increased amounts of

123
492 Community Ment Health J (2012) 48:490–502

vocational services have been found to have a positive contact, social context, and social identity—all of which
impact on employment outcomes, whereas greater amounts affect their health and mental health (Harnois and Gabriel
of clinical services are associated with poorer employment 2000).
outcomes (Cook 2006). These findings suggest that One particular social entrepreneurship model that pro-
enhancing the amount of vocational services to clients with vides job opportunities and clinical services for people with
mental illness to complement or exceed their existing psychiatric and physical disabilities is the affirmative
levels of clinical services may benefit them in obtaining business, or social firm (Krupa et al. 2003; Warner and
and maintaining competitive employment. Mandiberg 2006). Affirmative businesses are a type of
social enterprise created with the dual mission of
employing individuals with disabilities and providing a
Field of Social Entrepreneurship product or service needed by society. They are distinct
from sheltered workshops in that affirmative businesses
Social entrepreneurship approaches that integrate business operate according to a defined set of principles. Namely,
ventures with a social mission also emerged in the USA and they hire over one-third of employees with disabilities or
Canada in the 1980s. These socially conscious, business- labor-market disadvantages; they pay a fair market wage to
minded ventures focus on community economic develop- employees; and they operate without subsidies (Krupa et al.
ment as well as workforce development and job creation for 2003; Warner and Mandiberg 2006).
disadvantaged populations (Cooney and Williams Shanks Similarly, work integration social enterprises (WISEs)
2010; Krupa et al. 2003; Warner and Mandiberg 2006). create employment opportunities integrated with social
Social enterprises can refer to a nonprofit organization, a supports and hire individuals with a history of unemploy-
socially minded business, or a revenue-generating venture ment, low educational and employment skills, and mental
established to create positive social impact in the context of a health challenges (Cooney and Williams Shanks 2010).
financial bottom line (Dees 1998). Common social enter- Collectively, both affirmative businesses and WISEs offer
prises used with vulnerable populations include vocational transitional and permanent competitive employment to
cooperatives, affirmative businesses/social firms, work people with disabilities in a supportive, empowering and
integration social enterprises (WISEs) and microenterprises community-based setting. Findings from several studies
(Cooney and Williams Shanks 2010; Krupa 1998; Midgley suggest that these social businesses facilitate access to
and Livermore 2004). employment for clients experiencing homelessness and/or
In contrast to supported employment efforts that teach mental illness, who benefit from teamwork with peers,
individuals with mental illness to ‘‘fit in’’ to the existing while acquiring the vocational skills and clinical services
labor market, social enterprises do not encourage confor- needed to obtain and maintain a job (Ho and Chan 2010;
mity to a particular job description or setting. Rather, social Krupa et al. 2003; Shaheen and Rio 2007; Warner and
enterprises utilize a community economic development Mandiberg 2006).
approach to neutralize labor-market conditions of individ-
ualism, competition and profit that can create employment
disadvantage for persons with mental illness (Krupa et al. Theoretical Similarities Between Approaches
2003). This is evidenced in social enterprises such as
Homeboy Industries in Los Angeles, a non-profit organi- The basic theory of psychiatric rehabilitation using sup-
zation that oversees multiple self-sustaining social busi- ported employment and social enterprises is that individ-
nesses operated by gang-involved young people (Homeboy uals’ functional adjustment can be improved by creating a
Industries 2008). These businesses focus on the strengths of supportive environment and enhancing their skills or abil-
gang-involved youths and reduce their labor disadvantage ities (Anthony et al. 1990; Krupa et al. 2003). Likewise, the
by developing new employment opportunities for them basic theory of recovery is that individuals can get better
within their communities. Adopting a community eco- from their illness and pursue meaningful life goals, such as
nomic development approach, these businesses also create employment (Deegan 1988). Rehabilitation and recovery
local sources of employment and provide products and are instrumental for individuals with mental illness and can
services for residents (Cooney and Williams Shanks 2010; be promoted by both mental health systems and local
Krupa et al. 2003). Additionally, social enterprises offer communities. For instance, mental health systems can
opportunities for empowerment that are consistent with stimulate rehabilitation and recovery by integrating ser-
principles of mental illness recovery (Warner and Mandi- vices into natural, community-based settings. Similarly,
berg 2006). The underlying philosophy is that economic supportive community environments can facilitate reha-
and personal well-being are interrelated. Through employ- bilitation and recovery by creating opportunities for
ment, individuals benefit from time structure, social employment, education, housing, recreation and social

123
Community Ment Health J (2012) 48:490–502 493

support. Locating a job that has a strong goodness-of-fit Unlike the social enterprise movement, supported
between the individual and the work setting/responsibilities employment also benefits from treatment manuals and
and having access to ongoing clinical support during job formal agency trainings, through which it continues to be
tenure can be enabling, normalizing and health-promoting disseminated within mental health systems (Swanson et al.
for persons with mental illness (Becker and Drake 2003). 2008). Presently, there is a lack of outcomes data on the
effectiveness of social enterprises as well as data compar-
ing outcomes between supported employment and social
Rationale for and Evidence of Separate Fields enterprises (Young 2010). Studies of this nature are needed
before conclusions can be drawn about the relative effec-
Despite growing integration of the supported employment tiveness of social enterprises with persons with severe
approach in the USA mental health system (Becker and mental illness.
Drake 2003) as well as the widespread use of social Second, federal mental health funding traditionally has
enterprises in Europe and Canada (Krupa et al. 2003; not supported social enterprises as a behavioral interven-
Warner and Mandiberg 2006), the fields of mental health tion for populations with mental illness. Review of study
and social enterprises largely remain separate in the USA. abstracts from the National Institutes of Health (NIH)
The former is dominated by researchers from psychiatry, Research Portfolio Online Reporting Tool Expenditures
psychology, medicine and social work, who use the med- and Results (RePORTER) from 1985 to 2009 using the
ical model, clinically defined treatment protocols, ran- search terms ‘‘social enterprise,’’ ‘‘social firm,’’ ‘‘affirma-
domized trials, and outcomes based on diagnostic criteria tive business,’’ ‘‘vocational cooperative,’’ and ‘‘small
of mental illness (Becker and Drake 2003; Drake et al. business intervention’’ reveals that to date, no social
1999). In contrast, the latter is populated by researchers enterprise intervention has been funded by NIH. One
from business, economics, international relations, public NIMH-funded abstract does appear under the search term
policy and social work, who use social and economic ‘‘social enterprise.’’ This R01 study explores prisoners’
development models, United Nations’ standards and best- views about research as a social enterprise. The principal
practice guidelines, international case studies, and out- investigator’s use of ‘‘social enterprise’’ refers to research
comes based on improvements in human, social and eco- as an activity that is endorsed by society as potentially
nomic development (Cooney and Williams Shanks 2010; contributing to the social good, and which necessarily
Krupa 1998; Midgley and Livermore 2004). involves cooperation between scientists (in their roles as
The limited integration of these fields is further evi- investigators) and individual citizens (in agreeing to be
denced by three observations (1) the lack of social enter- participants), with the focus on informed consent as a
prise interventions readily adopted within mental health necessary safeguard for protecting research participants
agencies, (2) the absence of social enterprise interventions (personal communication, N. Poythress, September 3,
funded by the National Institutes of Health, and (3) the 2009). Review of RePORTER abstracts clearly highlights
limited empirical evidence on the mental health benefits the dearth of federally funded social enterprise interven-
of social enterprises with vulnerable populations. Each tions with populations with severe mental illness.
observation is discussed below. Third, in the existing social enterprise research, studies
First, social enterprises are not commonly used by documenting participants’ mental health outcomes are
mental health agencies. In their global review of affirma- lacking, as most research has focused on business sus-
tive businesses, Warner and Mandiberg (2006) highlight tainability and social mission risk (Cooney and Williams
that this model is much more prevalent in Europe and Shanks 2010). In the first worldwide overview of the social
Canada than in the USA. One explanation for its limited firm movement (Warner and Mandiberg 2006), no data
use in the American mental health system is the presence of were reported on the specific mental health impacts of
alternative, evidence-based supported employment models. involvement in social firms. Within the extant research on
The feasibility, efficacy and effectiveness of supported outcomes associated with social enterprises, findings—
employment models have been demonstrated over decades although scant—suggest the positive impact of work on
through randomized controlled trials (RCTs) comparing mental health symptoms in adults. For instance, Krupa
these models with conventional vocational services et al. (2003) assessed the influence of participation in an
(Becker and Drake 2003; Bond et al. 2001, 2007; Cook affirmative business on the well-being of adults with psy-
et al. 2005; Drake et al. 1999; Salyers et al. 2004). Yet chiatric disability. Qualitative data suggest reductions in
since social enterprises have not been part of the mental depressive symptoms, substance use and physical illness.
health landscape in the USA, they were likely not com- Similar studies evaluating the effects of paid work on
prised as part of the control groups to which supported mental illness reveal that adults with schizophrenia dem-
employment models traditionally have been compared. onstrate improvements in mental health symptoms, in

123
494 Community Ment Health J (2012) 48:490–502

particular with regards to positive, hostility and emotional experience and are often earlier in their trajectory of mental
discomfort symptoms. Reductions in both symptom inten- illness (Ferguson and Xie 2008; Nuechterlein et al. 2008;
sity and hospitalization rates were also noted (Bell et al. Rinaldi et al. 2004). In contrast to adults, homeless youths
1996). Lastly, qualitative findings from a study of 51 work are still being initiated into the role of ‘‘worker;’’ it thus
integration social enterprises (WISEs) in Hong Kong sug- may be particularly important for them to obtain a breadth
gest that participants from disadvantaged groups (e.g., new of experience and hands-on training. Also unlike adults
immigrants, persons with disabilities, ex-offenders and with severe mental illness, these youths are unlikely to
unemployed youths) experienced improvements in social receive governmental assistance, relying instead on non-
capital, social support, and job and life satisfaction (Ho and governmental drop-in centers (Kipke et al. 1997). Due to
Chan 2010). their added independence from families, homeless youths
need to achieve economic self-sufficiency in order to
survive.
Establishing a Case for Social Enterprises Given the unique situation of homeless youths, social
with Homeless Youths enterprises thus may be envisioned as one of several
potential employment options for this population. Consis-
While extant research suggests that social enterprise tent with the belief that there is no ‘‘wrong path’’ to
involvement is associated with positive changes in adults’ employment among homeless individuals (Shaheen and
mental health symptoms, there has been little research on Rio 2007), social businesses focus on work readiness and
the effect of social enterprises on the mental health of offer clients multiple pathways to employment, including
homeless youths. To date, there is considerable evidence job training, social-enterprise employment, and other
regarding the effectiveness of supported employment competitive employment (Krupa et al. 2003; Warner and
models (e.g., Individual Placement and Support [IPS]) with Mandiberg 2006). Although social enterprise interventions
adults with severe mental illness. A meta-analysis of 11 do not yet constitute an evidence-based treatment for
RCTs of vocational rehabilitation interventions for adults homeless youths, they draw upon concepts from evidence-
with severe mental illness reveals that supported employ- based supported employment models for adults with
ment programs—and IPS in particular—have demonstrated mental illness and from empirical precedents of social
better employment outcomes than conventional vocational enterprises with street youths around the world.
rehabilitation programs, both in rates of competitive
employment and employment of any type. However, only Social Enterprise Lessons from Abroad
61% of participants successfully obtained competitive
work at some point during the studies (Twamley et al. In the international literature, various examples exist of
2003). Job retention rates in IPS are also moderate at using social enterprise interventions with street youths to
various follow-up time points, including 60% at 6 months, develop niche markets for particular products or as portals
47% at 3 months, 41% at 15 months (Twamley et al. to competitive employment. Mexico currently uses a model
2003), and 35% at 24 months (Bond et al. 2007). While of triangular solidarity, which brings together different
extant evidence reveals that the IPS is the most effective sectors including non-governmental organizations, public
evidence-based employment intervention to date with institutions and corporations to address the street-youth
adults with severe mental illness, there remains great need phenomenon. In this model, non-profit organizations,
for improvement in outcomes. Despite its demonstrated which include Covenant House, assist youths in exiting the
effectiveness with adults, the IPS model has yet to be streets and receiving the necessary health, mental health,
examined with homeless youths, nor has it been combined and social services. The public sector, through the Ministry
with social enterprises when clients’ preferences are to start of Education, provides the youths with vocational training
their own business (Nuechterlein et al. 2008; Rinaldi et al. and certification programs to teach them skills and increase
2004). their employability. Finally, local and national corporations
When conceptualizing interventions with new popula- agree to hire the youths who have completed the training
tions with whom existing evidence-based approaches have program (Covenant House 2001).
not been developed, it is optimal to consider the population Children’s Development Banks (CDB) (Bal Vikas
being studied, rather than relying solely on theories or Banks), which are owned, managed and operated by street
methods that have not yet been applied to this population children themselves, currently exist in India, Bangladesh,
(Eap and Nagayama Hall 2008). The homeless youth Pakistan, Afghanistan and Nepal and serve hundreds of
population is distinct from adults with severe mental ill- street children (Children’s Development Bank 2004).
ness, on whom the IPS model has been developed and Operating as a cooperative banking system, CDBs enable
normed, since these youths have had less employment street children who are members to save and withdraw

123
Community Ment Health J (2012) 48:490–502 495

money as well as to receive loans for economic develop- receive technical training and education concerning spe-
ment activities. Across South Asia, CDBs have been suc- cific vocational skills; (2) small-business skill acquisition is
cessful in teaching street children valuable business skills, a separate 4-month course that focuses on business-related
assisting them in saving for their futures and securing loans skills needed to start a social enterprise, such as account-
for new businesses, and providing these children with a ing, budgeting, marketing and management; (3) affirmative
path out of poverty and social exclusion. Despite consid- business formation and distribution is the 12-month phase
erable social entrepreneurship activity with homeless in which participating youths establish a social enterprise
youths in the international realm, a major limitation to date among their peers in a supportive, empowering and com-
with these initiatives is the lack of outcome evaluation munity-based setting; and (4) clinical services is the mental
data on program efficacy and effectiveness (Cooney and health component of the model, which is woven throughout
Williams Shanks 2010). all stages over 20 months.
The Social Enterprise Intervention (SEI) clinical social
worker tracks all participating youths through weekly
Social Enterprise Intervention (SEI) Model meetings to identify, assess, prioritize and treat their target
areas of need. The social worker tailors the intensity and
In light of the lack of global social enterprise outcome focus of the clinical services to the severity of the youths’
evaluation data, combined with few social enterprise presenting conditions. Consistent with research on sup-
interventions within the field of mental health in the United ported employment programs that integrate vocational and
States, the author collaborated with a community-based, clinical services, it is expected that the frequency of clin-
homeless youth agency to develop, pilot and evaluate the ical services will decrease over the duration of the inter-
Social Enterprise Intervention (SEI) model with homeless vention (Cook 2006).
youths. The SEI model consists of vocational and business Review of the literature on homeless youths suggests
training, small-business development, supportive mentor- that in addition to participating in high-risk behaviors, they
ship and clinical services using a harm-reduction approach are likely to experience depression, post-traumatic stress
to improve the mental health, prosocial behaviors, social disorder and substance abuse, among other mental illnesses
support, and service utilization of homeless youths. The (Cauce et al. 2000). For youths with depression, bipolar or
SEI seeks to train youths in vocational and business skills anxiety disorders, the clinical social worker uses cognitive
and to treat them for existing mental health conditions to behavioral therapy, coupled with referrals to collaborating
reduce the likelihood of them moving from situations of psychiatrists for medication when requested by the youths.
high-risk to manifest mental health conditions and chronic Histories or symptoms of trauma are addressed through
homelessness (Tyler and Johnson 2006). SEI treatment individual and group trauma-intervention services, such as
effectiveness is reflected by increases in mental health cognitive behavioral therapy, support groups, relaxation
status, social support and service utilization as well as techniques, and referrals for medication. Motivational
reductions in high-risk sexual behaviors and substance interviewing is also used to reduce the youths’ high-risk
abuse (see Fig. 1). behaviors, especially substance abuse (Miller and Rollnick
Designed as a 20-month intervention, the SEI combines 2002). In addition, various individual and group-based
vocational and clinical services to assist homeless youths in harm-reduction strategies including safe-sex practices,
developing job skills and motivation to change in order to health education, sexually transmitted disease prevention,
make more-informed life and employment choices. There HIV counseling and testing, and substance abuse referrals
are four stages in the SEI model (See Table 1): (1) voca- are used by the social worker to reduce the youths’ high-
tional skill acquisition is a 4-month course in which youths risk behaviors through small achievable steps (Denning

Fig. 1 Social enterprise

intervention model Mental Health Status
• Depression
• Self-esteem
• Satisfaction with life

Social Enterprise High-Risk Behaviors

Intervention (SEI) • Survival sex
• Drug/Alcohol use
Engagement Peer Mentoring Retention
Vocational Skills Social Support
Business Skills • Peer support
Social Enterprise • Family support
Clinical Services
Harm Reduction Service Utilization
• Number of agencies used

123
496 Community Ment Health J (2012) 48:490–502

Table 1 Stages of the social enterprise intervention

Stage Time Characteristics Key mechanisms Objectives

I. Vocational 4 Months Vocational skill classes are taught by a Vocational skills course (3 days/ Youths learn vocational
skill qualified instructor to a small group of week for 1.5 h each class over skills needed to form
acquisition homeless youths. Peer mentors assist the 4 months) business with peers
instructor with classes Peer mentoring (access to mentors up Youths and mentors
to 20 h/week) develop rapport and
trust
II. Small 4 Months Youths participate in a seminar on small- Business skills course (3 days/week Youths learn business
business skill business skills (e.g., accounting, for 1.5 h each class over 4 months) skills
acquisition budgeting, marketing, management). Peer mentoring (access to mentors up Youths develop business
MBA students assist youths in conducting to 20 h/week) plan
a feasibility analysis of the business and
Youths identify local
in exploring product marketability.
market for target
Mentors continue to assist instructor
product(s)
Youths develop trust with
mentors
III. Affirmative 12 Months Affirmative business established to develop Affirmative business (up to 40 h/ Youths have increased
business and market youths’ target product(s). week for 12 months) opportunities for
formation and Youths design website to showcase/sell Peer mentoring (access to mentors up applying skills
distribution product(s). Youths locate venues/clients to 16 h/week) Affirmative business
to sell product(s). Mentors assist youths in works towards self-
business activities. SEI instructor sufficiency with sales
provides business oversight from target product(s)
IV. Clinical 20 Months Youths meet with SEI clinician to identify, Clinical services (29/week in month Clinician/youths identify
services assess, prioritize and treat target areas of 1; 19/week in months 2–6; 29/ clinical needs and goals
(ongoing need. Clinician uses motivational month in months 7–12; as needed Clinician assesses/treats
throughout interviewing, cognitive behavioral in months 13–20 youths’ clinical issues
stages I–III) therapy and harm-reduction strategies, as Harm-reduction Clinician tracks youths’
determined from baseline assessment
Service connection progress

2001). Evidence suggests that continuous access to high- areas of interest and direction of the social enterprise.
quality, integrated substance abuse and mental health There is evidence that client involvement in planning ser-
treatment helps people recover (Drake et al. 2008). vices can increase their satisfaction with services, which
The Social Enterprise Intervention (SEI) model draws may have implications for both client retention and out-
on the theoretical framework of asset-based youth devel- comes (James and Meezan 2002).
opment and thriving (Benson 1999, 2003). Specifically, Further, through the SEI mental health component,
youths’ internal developmental assets are categorized in the social worker partners with the youths to develop
four areas: commitment to learning, social competencies, and exercise positive values, particularly responsibility and
positive values and positive identity. Collectively, the SEI restraint. The social worker helps the youths identify and
components aim to strengthen the youths’ internal assets to prioritize their areas of need as well as take personal
enhance positive outcomes and protect them against high- responsibility for their actions. Similarly, by learning to use
risk behavior. For instance, the vocational and business diverse harm-reduction strategies, the youths are able to
courses seek to promote the youths’ commitment to practice the positive value of restraint, in particular with
learning by offering them training that integrates their their high-risk behaviors. Finally, the SEI model is
entrepreneurial skills and strengths, is tailored to their designed to promote positive identity in the youths by
previous experience, and promotes active learning. SEI affirming their capacity for self-direction in their relation-
courses are also designed to promote social competencies ship with the social worker, thus strengthening the youths’
in the youths, particularly planning and decision making, personal power. Through participation in the social enter-
by incorporating them in the decision-making aspects of prise, the youths are able to develop and strengthen their
the courses. To help build their social competencies, the areas of vocational expertise, thus enhancing their sense of
SEI instructor partners with the youths to establish class purpose. Additional descriptions of the SEI theoretical
rules, consequences for rule-breaking and incentive struc- framework and logic model are published elsewhere
tures for attendance. The youths also help the instructor (Ferguson 2007; Ferguson and Islam 2008; Ferguson and
formalize the class syllabi, depending on their identified Xie 2008).

123
Community Ment Health J (2012) 48:490–502 497

Preliminary Findings from SEI Pilot Work Survival sex was a single-item variable that measured
the number of times in the past year in which the youths
SEI Pilot Study Sampling Procedure received money, food, drugs, clothing, shelter or other
items in exchange for sex. Substance use was assessed
Quantitative Sampling and Measures using two variables on the Adult Self Report (ASR), each
measured over the 6 months prior to the interview
Convenience sampling was used to select the SEI and (Achenbach 1997). Alcohol use measured the number of
control-group youths (ages 18–24) from the homeless days the youths had been drunk. Drug use measured the
youth drop-in center that hosted the study. Two screening number of days the youths had used drugs for non-medical
criteria guided recruitment: (1) the youths had to have purposes. For men, borderline clinical ranges comprised
attended the agency two or more times a week for the raw scores between 27 and 47 (10 and 19 for women) for
month prior to the study; and (2) the youths had to commit alcohol, and 21 and 118 (4 and 19 for women) for drug use.
to attending the program for both the 4-month vocational Clinical ranges comprised raw scores above 47 (19 for
and business trainings. Of the 100 youths approached women) for alcohol, and 118 (19 for women) for drug use.
during the 1-month engagement stage, 20 signed up to Peer support was a composite-score variable of the sum
participate. Of these 20, 16 participated in the intervention. of four items on the Friends Subscale of the ASR. The
Although our response rate for SEI participation was low, it response categories for each item range from none to 5 or
approaches the typical response rate of 20 to 25% among more for questions including: ‘‘About how many close
the general population in intervention studies (Leonard friends do you have?’’ The range of scores is from 0 to 12,
et al. 2003). Difficulties in client recruitment are even more with higher scores indicating greater peer support. Family
pronounced with populations who have limited service support was a single-item variable that asked the youths to
engagement, such as homeless street youths (De Rosa et al. indicate the frequency with which they see, write or talk to
1999; Kipke et al. 1997). A separate control group of 12 their immediate family. Responses range from less than
youths was also formed in the agency. These youths opted 1–2 times per year to everyday. The range of possible
not to participate in the SEI, but consented to the baseline scores is from 1 to 5, with higher scores reflecting
and follow-up interviews. An attempt was made to match increased frequency of family contact. Finally, service
these youths with the SEI group on age, gender and eth- utilization was a single-item variable measuring the num-
nicity. The control group received usual-care agency ber of other agencies—other than the host agency—at
services. which the youths were receiving services.
Researchers conducted a 60- to 90-min structured
interview with the youths, which was designed to assess Qualitative Sampling and Questionnaire Development
mental health status (depression, self esteem, satisfaction
with life), high-risk behaviors (survival sex and substance For the qualitative focus groups, purposeful sampling was
use), social support (peer and family) and service utiliza- used to select information-rich cases for in-depth study of
tion. Human subjects’ approval was granted from the uni- the processes and outcomes of the SEI Program. A
versity. Depression was assessed using the Reynolds homogeneous sampling procedure was adopted in which
Depression Screening Inventory (RDSI) (Reynolds and the young adults who remained in the SEI Program after
Kobak 1998). The range of RDSI raw scores is 0–63. having participated for 4 months were selected for the
Scores of 10 or less indicate no depression; 11–15 indicate interviews (Patton 1990).
mild clinical severity; 16–24 indicate moderate clinical To create the focus-group questions, the PI had devel-
severity, and scores of 25 or more indicate severe clinical oped hunches from prior research and practice with
severity. The Cronbach’s alpha for the 19 items on the homeless youths, which were subsequently discussed with
RDSI was a = 0.84. Self-esteem was measured using the the agency staff for their input. The PI’s existing practice
Rosenberg Self-Esteem Scale (RSE) (Rosenberg 1979). and research experience with this population facilitated the
Each item is rated using a 4-point Likert-type scale. The development of several preconceived ideas—or suspi-
range of RSE raw scores is 0–30. Scores between 15 and 25 cions—about key concepts and how they may be related
are within normal range, whereas scores below 15 suggest (Rodwell 1998). The PI functioned as a human instrument
low self-esteem. The Cronbach’s alpha for the 10 RSE to shape the questions used for further inquiry with the
items was a = 0.87. Satisfaction with life was assessed youths. Some of the focus group questions included: In
using the Satisfaction with Life Scale (SWLS) (Diener what ways has the SEI workshop affected your relation-
et al. 1985). Each item is scored from 1 to 7 on a Likert- ships with friends or family?, How do you feel about
type scale. Scores range from 5 to 35. The Cronbach’s yourself now after the SEI program versus when you
alpha for the 5 SWLS items was a = 0.74. began?, and In what ways has the SEI workshop affected

123
498 Community Ment Health J (2012) 48:490–502

your life? (for additional information on the focus groups, in the control group (P = 0.10). Although these two dif-
see Ferguson and Islam 2008). ferences only reflect trends in the data, there were statis-
tically significant within-groups differences for the SEI
SEI Pilot Study Findings youths between baseline and follow-up for both peer sup-
port (P = .02) and depression (P = .005). Overall, the SEI
Quantitative Findings showed some success in improving the youths’ satisfaction
with life, family support, peer support and depression
Twenty-eight homeless youths (ages 18–24) participated in (Ferguson and Xie 2008).
the SEI pilot (16 in the SEI and 12 in the control group).
Participants were on average 21 years old. Twenty youths Qualitative Findings
were male and 8 were female. Eleven youths identified as
African American, 6 as Hispanic, 6 as Caucasian, 4 as mixed Qualitative participatory evaluation methods were also
or other ethnicity, and 1 as Asian (Ferguson and Xie 2008). used in the pilot to explore SEI processes and outcomes
There were no significant differences between the SEI with the youths. The PI and trained research assistant
and control groups on baseline variables, nor on the rate of facilitated three focus groups during the pilot. The constant
study attrition. Eleven of the 16 SEI youths (69%) and 8 of comparative method was used to analyze participants’
the 12 control-group youths (67%) were available at follow verbatim transcripts from the summative focus group
up. Retention among both groups was respectable, given (Glaser and Strauss 1967). From the data, a model of cli-
that client-retention rates among the general population in ent-identified changes from the SEI was developed to
intervention studies range from 25 to 85%, yet most are include mental health, employment, service-related,
between 45 and 65% (Leonard et al. 2003). To prevent behavioral and societal outcomes (Ferguson and Islam
attrition among SEI youths, staff provided monthly 2008) (See Table 3). Qualitative findings suggest that the
incentives for attendance, including bus tokens and art SEI influenced participants’ self-esteem, motivation,
materials. Staff also used social networking technologies to employability, relationships with peers and family, and
communicate with the youths between class sessions. pro-social behaviors.
Table 2 compares changes in key outcome measures
between the SEI and control groups. Results from inde- Table 3 Constant comparative method of qualitative data analysis
pendent sample t-tests demonstrate that the SEI Program
had a 6.45-unit increase in total life satisfaction from Primary code Code sub-categories
baseline to the end of follow-up, compared with a 2.25-unit Mental health Family respect, self-esteem, motivation,
decrease in the control group (P = .02). The SEI Program outcomes goal-orientation
also had a 0.50-unit increase in family support over the Employment Job skills, employment search, labor networks,
study period, as compared to a 1.20-unit decrease observed outcomes future employment plans
in the control group (P = .03). Regarding additional bor- Service-related Relationships with staff, service engagement,
derline-significant differences, the SEI had a 3.00-unit outcomes social networks
increase in peer support over the study period, whereas the Behavioral Respite from street life, destructive behavior
outcomes
control group experienced a 0.13-unit increase (P = 0.06),
Societal outcomes Positive aspects of homeless youths, image
Additionally, the SEI group had a 5.45-unit decrease in with authorities
depressive symptoms, compared with no change observed

Table 2 Baseline, follow-up and change scores of outcome measures for SEI and control groups
Variable Baseline mean (SD)a Follow-up mean (SD) Change score mean (SD)b P (1-tailed) Cohen
effect sizec
SEI Ctrl SEI Ctrl SEI Ctrl

Satisfaction with 18.81 (7.16) 20.58 (8.69) 22.27 (6.36) 17.13 (10.11) 6.45 (5.89) -2.25 (11.61) 0.02 0.954
life (SWLS)
Family support 2.91 (0.83) 3.50 (1.51) 3.22 (1.48) 3.17 (1.47) .50 (1.05) -1.20 (1.64) 0.03 1.155
Peer support (ASR) 6.50 (3.92) 7.83 (4.04) 8.64 (2.20) 7.50 (3.82) 3.00 (4.20) .13 (3.14) 0.06 0.722
Depression (RDSI) 12.13 (8.63) 14.00 (9.53) 9.27 (7.71) 12.25 (12.65) -5.45 (5.56) .00 (12.02) 0.10 0.590
a
SEI: n = 16; Ctrl: n = 12
b
Change scores available only for subjects who had valid observations for both baseline and follow-up interviews
c
Standardized effect size corrected for small sample size

123
Community Ment Health J (2012) 48:490–502 499

Opportunities and Tensions of Merging Mental Health to hire additional staff (Ferguson and Xie 2008). Further,
and Social Enterprise Fields agency staff benefit in their professional development by
receiving training in designing, implementing and evaluat-
Overall, the SEI pilot produced both opportunities and ing evidence-based practices, and in administering stan-
tensions for the host agency youths and staff. Initial SEI dardized assessment tools with their clients (Becker and
feasibility work revealed improvements in the youths’ Drake 2003; Drake et al. 1999).
depressive symptoms, satisfaction with life, self-esteem, With respect to local community gains, established
goal-orientation, family and peer support, and delinquent social enterprises create both job opportunities for local
behavior. Aside from these outcomes, there are clear ben- residents and needed products and services for the com-
efits to clients, agencies and communities from the use of munity (Cooney and Williams Shanks 2010). Microenter-
social enterprise interventions with homeless youths. prises including the Grameen Bank cooperatives and their
First, through social-enterprise involvement, homeless global replications have been used throughout the world to
youths learn vocational and business skills, gain access to a increase household income and enhance community eco-
social enterprise, and receive continuous mentoring from nomic development for impoverished individuals and
peers and supportive staff. Regarding mental health bene- communities (Cooney and Williams Shanks 2010; Yunus
fits, the youths receive ongoing mental health services 2003, 2008). Community residents may also benefit from
tailored to their individual conditions and treatment goals, purchasing products created by local social enterprises. A
as well as referrals to health, mental health, and social recent study of over 200 Israeli social-business organiza-
services. For a highly service-disengaged population, social tions found that the major products offered to host com-
enterprises have shown some initial success in engaging munities were restaurant and food services (e.g., catering),
and retaining homeless youths in employment and clinical clothing services (e.g., tailoring), flower sales and tele-
services (Ferguson and Xie 2008). Although the SEI pilot phone services (Gidron and Yogev 2010). Social enter-
study’s focus on graphic design and the required time prises in the USA offer an array of services to local
commitment may have precluded some youths from par- residents, including restaurant and food services, street
ticipating, 69% of the youths remained active at 9 months. cleaning, manufacturing, maintenance, pest control, retail
In future replications, the SEI can be adapted to teach a and furniture upholstery (Cooney and Williams Shanks
variety of skills (e.g., woodworking, food preparation, 2010). Lastly, urban centers that are home to street youths
landscaping, silk-screening, etc.), using diverse training who engage in illicit behaviors in public spaces may also
styles and time periods (Gidron and Yogev 2010). Pre- benefit from transitioning these young people from crimi-
liminary focus groups with youths, staff and community nal activity to employment (Baron and Hartnagel 1997).
residents as well as a local market analysis will facilitate Despite the opportunities that may derive from social
selecting the most appropriate vocational/business skills, enterprise interventions, blending the fields of mental
length of training needed for skill mastery, and nature and health and social enterprise has not been without tensions.
products of the business. Solicitation of youths’ and staff’s First, as with any new mental health program, social
input in the intervention design and in the direction of the enterprise interventions require seed capital or start-up
social enterprise may in turn positively influence both cli- funding (Warner and Mandiberg 2006; Young 2010). One
ent retention and outcomes (James and Meezan 2002). way to address this is to request in-kind technical assis-
Regarding agency benefits, social enterprises provide an tance from faculty and students at local business schools or
on-site and integrated setting in which clients receive from community members with expertise in social enter-
vocational training and clinical services, and at the same prises. Donations of needed equipment such as machinery,
time, have the space to practice and hone their skills (Krupa course materials and computer hard and software can also
1998; Krupa et al. 2003). In an era of increasing unem- be solicited from local companies and community groups
ployment and limited federal and private funding for non- that would benefit from establishing the social enterprise
profit agencies, businesses that are run within non-profits and training the youths in specialized skills. For instance,
create sources of employment for clients along with alternate in the SEI pilot, online social-networking technologies
funding streams for host agencies (Cooney and Williams were used to advertise the SEI Program, to hire instructors
Shanks 2010). This income derived from the sale of services with expertise in graphic design, and to solicit technical
is vital for non-profits, as it now constitutes the largest and assistance, donations, and guest lectures by social entre-
fastest growing revenue source for non-profits in the USA preneurs. Similarly, public or private funding agencies
(Young 1998). Community-based agencies that host RCTs could provide start-up grants for innovative program
of social enterprise interventions are also eligible for multi- implementation for this highly vulnerable youth popula-
year intervention funding from the National Institutes of tion. Researchers can in turn use pilot studies of social
Health to develop and test evidence-based interventions and enterprise interventions to gather feasibility data to solicit

123
500 Community Ment Health J (2012) 48:490–502

larger grants to test their effectiveness in RCT studies. The federal guidelines for intervention research present a
author is currently funded by the National Institute of challenge to community-based agencies that may not issue
Mental Health (NIMH) to conduct an effectiveness study of mental health diagnoses or use diagnostic labels in their
the SEI with homeless youths with mental illness. SEI pilot service provision. One way to handle this discrepancy is to
data were crucial in preparing a successful proposal. train agency clinicians as part of the research study to
Second, social enterprise interventions are still in their administer diagnostic assessment tools (e.g., the Diagnostic
infancy within the mental health field (Warner and Mandi- Interview Schedule for Children [DISC] or the Composite
berg 2006). As a result, this approach has few successful International Diagnostic Interview [CIDI]) to determine
precedents to emulate and limited evidence-based research client diagnoses in their own practice (Shaffer et al. 2000;
to inform agency staff in the design of such programs. Lack Wittchen 1994). It is also vital for researchers to raise
of empirical data places researchers at a disadvantage in awareness among agency staff regarding the usefulness of
publishing pilot findings and in soliciting competitive federal client diagnoses in providing appropriate mental health
grants as well. Standard treatment manuals for developing treatment and in evaluating client symptom changes over
and replicating social enterprise interventions also remain time. Practitioners may also object to researchers screening
outstanding. In the absence of such documentation, research- clients for mental illness as an inclusion criterion, as this
ers and practitioners are forced to re-create study protocols practice can exclude other symptomatic and vulnerable
and treatment regimens with each new project. It is vital to clients who could benefit from the intervention (Drake
establish the feasibility of social enterprises within the et al. 2001). In this case, researchers can train agency staff
mental health system and to develop treatment manuals in as part of the research study to offer the intervention or
order to train mental health personnel and replicate studies. another evidence-based intervention to clients who do
Additionally, when using a community-based participa- not qualify. Wait-lists also help equalize opportunities for
tory research approach to intervention development, con- clients who do not initially qualify for the study.
siderable time is required to gain credibility for novel
interventions from the broader scientific community,
researchers and funders (Christopher et al. 2008; Minkler Conclusions
et al. 2003). With respect to the SEI, the author invested
1 year in developing university-agency partnerships in Los There is both theoretical and practical utility in blending
Angeles at which to host the SEI pilot. While initially labor- the fields of mental health and social enterprise. With
intensive, these partnerships proved essential for gaining respect to theory-building, extant data suggest that vul-
access to staff and clients with which the intervention could nerable populations involved in social enterprise interven-
be developed and tested. Over the following 2 years, the SEI tions experience positive changes in their employment and
intervention was designed and tested in a pilot study. Con- mental health outcomes. These findings support the inte-
comitantly, the author developed the conceptual and theo- gration of vocational and clinical services through blending
retical frameworks needed to communicate social enterprise the respective fields of mental health and social enterprise.
ideas to the broader scientific community through publica- Social enterprise interventions also blend the basic theories
tions and presentations, and to funding agencies through of psychiatric rehabilitation and of recovery by offering
proposals. Novel interventions that draw from the interna- persons with mental illness both mental health treatment
tional realm and from inter-disciplinary fields can be unfa- and economic opportunity within a supportive community
miliar to federal funders, which in turn can delay funding for environment. On a practical level, social enterprises can
ongoing research (Hokenstad and Midgley 2004). The author complement the existing array of vocational options for
initiated the NIMH SEI proposal in 2006 and received homeless populations as well as expand the supported
funding in 2009. Supplementary bridge funding can be employment approach by providing options for clients who
essential during federal proposal review cycles to sustain an seek self-employment and social-enterprise employment.
intervention’s momentum at a host agency, to keep inter- To accomplish this goal of greater integration, the fields
vention staff employed, and to nurture the ongoing of mental health and social enterprise have much to learn
researcher-agency relationship (Pietroburgo 2006). from the other. Yet to advance knowledge and progress
Finally, discrepancies may exist between federal fund- between fields, service integration—not segregation—is
ing requirements and community-based agency ethos key. Given their benefits to consumers and society, as well
regarding diagnosing clients with mental illness. Federal as their demonstrated success, social enterprise interven-
funding for mental health interventions commonly require tions are well poised to function as a bridge between the
mental health diagnoses as inclusion criteria in RCTs as fields of mental health and business. To the extent that they
well as for tracking symptom improvement (Becker and are created to assist in rehabilitation and recovery for
Drake 2003; Drake et al. 1999). These gold-standard persons with mental illness, they have an identity in the

123
Community Ment Health J (2012) 48:490–502 501

field of mental health among existing vocational rehabili- Christopher, S., Watts, V., McCormick, A. K. H. G., & Young, S.
tation and mental health treatment options. Likewise, to the (2008). Building and maintaining trust in a community-based
participatory research partnership. American Journal of Public
extent that they exist to facilitate rehabilitation and Health, 98(8), 1398–1406.
recovery for vulnerable populations within the context of a Coleman, J. (1990). The foundations of social theory. Cambridge:
financial bottom line, they are also grounded in the field of Harvard University Press.
business among private, for-profit options. In either case, Cook, J. A. (2006). Employment barriers for persons with psychiatric
disabilities: Update of a report for the president’s commission.
social enterprise interventions are consistent with the Psychiatric Services, 57(10), 1391–1405.
principles of rehabilitation and recovery, and can facilitate Cook, J. A., Lehman, A. F., Drake, R. E., McFarlane, W. R., Gold, P.
the integration of disparate fields of study for the benefit of B., Leff, H. S., et al. (2005). Integration of psychiatric and
persons with severe mental illness. vocational services: A multisite randomized, controlled trial of
supported employment. American Journal of Psychiatry, 62(10),
1948–1956.
Acknowledgments The pilot study described in this article was Cooney, K., & Williams Shanks, T. R. (2010). New approaches to old
supported by the Larson Endowment for Innovative Research at the problems: Market-based strategies for poverty alleviation. Social
University of Southern California School of Social Work. Service Review, 84(1), 29–55.
Covenant House. (2001). World economic forum in Mexico to include
a voice for street children. (http://www.casa-alianza.org).
Accessed 1 Dec 2010.
References De Rosa, C. J., Montgomery, S. B., Kipke, M. D., Iverson, E., Ma, J.
L., & Unger, J. B. (1999). Service utilization among homeless
Achenbach, T. M. (1997). Manual for the young adult behavior and runaway youth in Los Angeles, California: Rates and
checklist and young adult self report. Burlington, VT: Depart- reasons. Journal of Adolescent Health, 24(6), 449–458.
ment of Psychiatry, University of Vermont. Deegan, P. E. (1988). Recovery: The loved experience of rehabili-
Anthony, W., Cohen, M., & Farkas, M. (1990). Psychiatric rehabil- tation. Psychosocial Rehabilitation Journal, 11(4), 11–19.
itation. Boston: Center for Psychiatric Rehabilitation. Dees, J. G. (1998). Enterprising nonprofits. Harvard Business Review,
Baron, S. W., & Hartnagel, T. F. (1997). Attributions, affect, and Jan–Feb, 55–67.
crime: Street youths’ reactions to unemployment. Criminology, Denning, P. (2001). Strategies for implementation of harm reduction in
35(3), 409–434. treatment settings. Journal of Psychoactive Drugs, 33(1), 23–26.
Becker, D. R., Bond, G. R., McCarthy, D., Thompson, D., Xie, H., Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The
McHugo, G. J., et al. (2001). Converting day treatment centers to satisfaction with life scale. Journal of Personality Assessment,
supported employment programs in Rhode Island. Psychiatric 49(1), 71–75.
Services, 52(3), 351–357. Drake, R. E., Essock, S. M., Shaner, A., Carey, K. B., Minkoff, K.,
Becker, D. R., & Drake, R. E. (2003). A working life for people with Kola, L., et al. (2001). Implementing dual diagnosis services for
severe mental illness. New York: Oxford University Press. clients with severe mental illness. Psychiatric Services, 52(4),
Bell, M. D., Lysaker, P. H., & Milstein, R. M. (1996). Clinical 469–476.
benefits of paid work activity in schizophrenia. Schizophrenia Drake, R. E., McHugo, G. J., Bebout, R. R., Becker, D. R., Harris, M.,
Bulletin, 22(1), 51–67. Bond, G. R., et al. (1999). A randomized clinical trial of
Benson, P. L. (1999). A fragile foundation: The state of developmental supported employment for inner-city patients with severe mental
assets among American youth. Minneapolis, MN: Search Institute. disorders. Archives of General Psychiatry, 56(7), 627–633.
Benson, P. L. (2003). Developmental assets and asset-building Drake, R. E., O’Neal, E. L., & Wallach, M. A. (2008). A systematic
community: Conceptual and empirical foundations. In R. M. Lern- review of psychosocial research on psychosocial interventions
er & P. L. Benson (Eds.), Developmental assets and asset-building for people with co-occurring severe mental and substance use
communities: Implications for research, policy and practice (pp. disorders. Journal of Substance Abuse Treatment, 34, 123–138.
19–43). New York: Kluwer Academic/Plenum Publishers. Eap, S., & Nagayama Hall, G. C. (2008). Relevance of RCTs to
Bond, G. R., Becker, D. R., Drake, R. E., Rapp, C. A., Meisler, N., diverse groups. In A. M. Nezu & C. M. Nezu (Eds.), Evidence-
Lehman, A. F., et al. (2001). Implementing supported employ- based outcome research (pp. 425–443). New York: Oxford
ment as an evidence-based practice. Psychiatric Services, 52(3), University Press.
313–322. Ferguson, K. (2007). Implementing a social enterprise intervention
Bond, G. R., Salyers, M. P., Dincin, J., Drake, R. E., Becker, D. R., with homeless, street-living youth in Los Angeles. Social Work,
Fraser, V. V., et al. (2007). A randomized controlled trial 52(2), 103–112.
comparing two vocational models for persons with severe mental Ferguson, K., & Islam, N. (2008). Conceptualizing outcomes with
illness. Journal of Consulting and Clinical Psychology, 75(6), street-living young adults: Grounded theory approach to evalu-
968–982. ating the Social Enterprise Intervention. Qualitative Social
Bourdieu, P. (1993). Sociology in question. London: Sage. Work: Research and Practice, 7(2), 217–237.
Carter, E. W., & Wehby, J. H. (2003). Job performance of transition- Ferguson, K., & Xie, B. (2008). Feasibility study of the Social
age youth with emotional and behavioral disorders. Exceptional Enterprise Intervention with homeless youth. Research on Social
Children, 69(4), 449–465. Work Practice, 18(1), 5–19.
Cauce, A. M., Paradise, M., Ginzler, J. A., Embry, L., Morgan, C. J., Gaetz, S., & O’Grady, B. (2002). Making money: Exploring the
Lohr, Y., et al. (2000). The characteristics and mental health of economy of young homeless workers. Work, Employment &
homeless adolescents: Age and gender differences. Journal of Society, 16(3), 33–456.
Emotional and Behavioral Disorders, 8(4), 230–239. Gidron, B., & Yogev, D. (2010). Social business organizations in
Children’s Development Bank (CDB). (2004). All about CDB. (http:// Israel: Preliminary data on a new phenomenon. Beersheba: Ben
www.childrensdevelopmentbank.org). Accessed 1 Dec 2010. Gurion University, Israeli Center for Third Sector Research.

123
502 Community Ment Health J (2012) 48:490–502

Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded Reynolds, W. M., & Kobak, K. A. (1998). Reynolds depression
theory: Strategies for qualitative research. Hawthorne, NY: screening inventory. Professional manual. Lutz, FL: Psycholog-
Aldine Publishing Company. ical Assessment Resources, Inc.
Greene, J. M., Ennett, S. T., & Ringwalt, C. L. (1999). Prevalence and Rinaldi, M., McNeil, K., Firn, M., Koletsi, M., Perkins, R., & Singh,
correlates of survival sex among runaway and homeless youth. S. P. (2004). What are the benefits of evidence-based supported
American Journal of Public Health, 89(9), 1406–1409. employment for patients with first-episode psychosis? Psychiat-
Harnois, G., & Gabriel, P. (2000). Mental health and work: Impact, ric Bulletin, 28, 281–284.
issues and good practices. Geneva: World Health Organization Rodwell, M. K. (1998). Social work constructivist research. New
and International Labor Organization. York: Garland Publishing, Inc.
Ho, A. P., & Chan, K. T. (2010). The social impact of work- Rosenberg, M. (1979). Conceiving the self. New York: Basic Books.
integration social enterprise in Hong Kong. International Social Salyers, M. P., Becker, D. R., Drake, R. E., Torrey, W. C., & Wyzik,
Work, 53(1), 33–45. P. F. (2004). A ten-year follow-up of a supported employment
Hokenstad, M. C., & Midgley, J. (2004). Lessons from other program. Psychiatric Services, 55(3), 302–308.
countries: Current benefits and future opportunities. In M. Shaffer, D., Fisher, P. W., Lucas, C. P., Dulcan, M. K., & Schwab-
C. Hokenstad & J. Midgley (Eds.), Lessons from abroad: Stone, M. E. (2000). NIMH diagnostic interview schedule for
Adapting international social welfare innovations (pp. 1–12). children version IV (NIMH DISC-IV): Description, differences
Washington, DC: NASW Press. from previous versions, and reliability of some common
Homeboy Industries. (2008). Welcome to Homeboy Industries. (http:// diagnoses. Journal of the American Academy of Child and
www.homeboy-industries.org/index.php). Accessed 5 Dec 2010. Adolescent Psychiatry, 39(1), 28–38.
James, S., & Meezan, W. (2002). Refining the evaluation of treatment Shaheen, G., & Rio, J. (2007). Recognizing work as a priority in
foster care. Family and Society, 83(3), 233–244. preventing or ending homelessness. Journal of Primary Preven-
Kipke, M. D., Unger, J. B., O’Connor, S., Palmer, R. F., & LaFrance, tion, 28(3–4), 341–358.
S. (1997). Street youth, their peer group affiliation and differ- Swanson, S. J., Becker, D. R., Drake, R. E., & Merrens, M. R. (2008).
ences according to residential status, subsistence patterns, and Supported employment: A practical guide for practitioners and
use of services. Adolescence, 32(127), 655–669. supervisors. Kearney, NE: Morris Publishing.
Krupa, T. (1998). The consumer-run business: People with psychiatric Thompson, S. J., Pollio, D. E., Constantine, J., Reid, D., & Nebbitt, V.
disabilities as entrepreneurs. Work, 11, 3–10. (2002). Short-term outcomes for youth receiving runaway and
Krupa, T., LaGarde, M., & Carmichael, K. (2003). Transforming homeless shelter services. Research on Social Work Practice,
sheltered workshops into affirmative businesses: An outcome 12(5), 589–603.
evaluation. Psychiatric Rehabilitation Journal, 26(4), 359–367. Twamley, E. W., Jeste, D. V., & Lehman, A. F. (2003). Vocational
Leonard, N. R., Lester, P., Rotheram-Borus, M. J., Mattes, K., Gwadz, rehabilitation in schizophrenia and other psychotic disorders.
M., & Ferns, B. (2003). Successful recruitment and retention of The Journal of Nervous and Mental Disease, 191(8), 515–523.
participants in longitudinal behavioral research. AIDS Education Tyler, K. A., & Johnson, K. A. (2006). Pathways in and out of
and Prevention, 15(3), 269–281. substance use among homeless-emerging adults. Journal of
Midgley, J., & Livermore, M. (2004). Social development: Lessons Adolescent Research, 21(2), 133–157.
from the global south. In M. C. Hokenstad & J. Midgley (Eds.), Warner, R., & Mandiberg, J. (2006). An update on affirmative
Lessons from abroad: Adapting international social welfare businesses or social firms for people with mental illness.
innovations (pp. 117–135). Washington, DC: NASW Press. Psychiatric Services, 57(10), 1488–1492.
Miller, W. R., & Rollnick, S. (2002). Motivational interviewing: Wehman, P., & Moon, M. S. (1988). Vocational rehabilitation and
Preparing people to change addictive behavior (2nd ed.). New supported employment. Baltimore, MD: Paul Brookes.
York: Guilford Press. Whitbeck, L. B. (2009). Mental health and emerging adulthood
Minkler, M., Blackwell, A. G., Thompson, M., & Tamir, H. (2003). among homeless young people. New York: Psychology Press.
Community-based participatory research: Implications for public Wittchen, H. U. (1994). Reliability and validity studies of the WHO-
health funding. American Journal of Public Health, 93(8), composite international diagnostic interview: A critical review.
1210–1213. Journal of Psychiatric Research, 28(1), 57–84.
Nuechterlein, K. H., Subotnik, K. L., Turner, L. R., Ventura, J., Young, D. R. (1998). Commercialism in nonprofit social service
Becker, D. R., & Drake, R. E. (2008). Individual placement and associations: Its character, significance, and rationale. Journal of
support for individuals with a recent-onset of schizophrenia: Policy Analysis and Management, 17(2), 278–297.
Integrating supported education and supported employment. Young, D. R. (2010). The state of theory and research on social
Psychiatric Rehabilitation Journal, 31(4), 340–349. enterprises. Paper presented at the conference on exploring
Patton, M. Q. (1990). Qualitative research and evaluation methods. social enterprises, UCLA School of Public Affairs, October 29,
Newbury Park, CA: Sage. 2010.
Pietroburgo, J. (2006). Charity at the deathbed: Impacts of public Yunus, M. (2003). Banker to the poor: Micro-lending and the battle
funding changes on hospice care. American Journal of Hospice against world poverty. New York: Public Affairs, Perseus Books
and Palliative Medicine, 23(3), 217–223. Group.
Rashid, S. (2004). Evaluating a transitional living program for Yunus, M. (2008). A world without poverty: Social business and the
homeless, former foster care youth. Research on Social Work future of capitalism. New York: Public Affairs, Perseus Books
Practice, 14(4), 240–248. Group.

123