Associative stigma among mental health professionals: implications for professional and service user well-being.

Abstract

In contrast with growing attention given to the stigma experiences of mental health service users, the stigma literature has paid almost no attention to mental health professionals. This study focuses on experiences of associative stigma among these professionals. We investigate the link between associative stigma and three dimensions of burnout as well as job satisfaction among mental health professionals, and the link of associative stigma with self-stigma and client satisfaction among service users. Survey data from 543 professionals and 707 service users from diverse mental health services are analyzed using multilevel techniques. The results reveal that among mental health professionals associative stigma is related to more depersonalization, more emotional exhaustion, and less job satisfaction. In addition, in units in which professionals report more associative stigma, service users experience more self-stigma and less client satisfaction. The results reveal that associative stigma is related to more depersonalization, more emotional exhaustion, and less job satisfaction among mental health professionals.

Full text

Page 1

http://hsb.sagepub.com/
Social Behavior
Journal of Health and
http://hsb.sagepub.com/content/53/1/17
The online version of this article can be found at:

DOI: 10.1177/0022146512439453
2012 53: 17Journal of Health and Social Behavior
Mieke Verhaeghe and Piet Bracke
Professional and Service User Well-Being
Associative Stigma among Mental Health Professionals : Implications for


Published by:

http://www.sagepublications.com
On behalf of:




American Sociological Association
can be found at:
Journal of Health and Social Behavior

Additional services and information for





http://hsb.sagepub.com/cgi/alerts
Email Alerts:

http://hsb.sagepub.com/subscriptions
Subscriptions:



http://www.sagepub.com/journalsReprints.nav
Reprints:


http://www.sagepub.com/journalsPermissions.nav
Permissions:
What is This?

- Mar 1, 2012 Version of Record >>
at Universiteit Gent-Vakgroep on March 1, 2012hsb.sagepub.comDownloaded from

Page 2

Journal of Health and Social Behavior
53(1) 17 –32
© American Sociological Association 2012
DOI: 10.1177/0022146512439453
http://jhsb.sagepub.com
Stigma and Health
One-quarter to one-third of the general population
experiences a mental health problem at least once
(World Health Organization International Consor-
tium in Psychiatric Epidemiology 2000), yet mental
illness remains one of the most stigmatized human
conditions. Empirical studies based on theoretical
frameworks such as the labeling perspective (Scheff
1966), modified labeling perspective (Link 1987;
Link et al. 1989), and theory of self-stigma (Corri-
gan and Watson 2002) have revealed detrimental
consequences for job opportunities (Link 1982;
Glozier 1998), housing opportunities (Page 1977),
life satisfaction (Markowitz 1998), self-esteem,
self-efficacy (Link et al. 2001; Wright, Gronfein,
and Owens 2000), and obtaining professional care
(Vogel, Wade, and Hackler 2007). The literature on
the stigma of mental illness typically focuses on the
general public (Angermeyer et al. 2005; Phelan
et al. 2000) or mental health service users (Link
et al. 1997; Rosenfield 1997) and has paid almost no
attention to mental health professionals. To address
this deficit, this study focuses on experiences of
associative stigma among these professionals. We
define associative stigma as stigma that mental
439453 HSBXXX10.1177/0022146512439453Verha
2012
eghe and BrackeJournal of Health and Social Behavior
1Ghent University, Belgium
2Katho University College, Kortrijk, Belgium
Corresponding Author:
Mieke Verhaeghe, Ghent University, Department of
Sociology, Korte Meer 5, B-9000 Ghent, Belgium
E-mail: mieke.verhaeghe@katho.be
Associative Stigma
among Mental Health
Professionals: Implications for
Professional and Service User
Well-Being
Mieke Verhaeghe1,2 and Piet Bracke1
Abstract
In contrast with growing attention given to the stigma experiences of mental health service users, the
stigma literature has paid almost no attention to mental health professionals. This study focuses on
experiences of associative stigma among these professionals. We investigate the link between associative
stigma and three dimensions of burnout as well as job satisfaction among mental health professionals, and
the link of associative stigma with self-stigma and client satisfaction among service users. Survey data from
543 professionals and 707 service users from diverse mental health services are analyzed using multilevel
techniques. The results reveal that among mental health professionals associative stigma is related to
more depersonalization, more emotional exhaustion, and less job satisfaction. In addition, in units in which
professionals report more associative stigma, service users experience more self-stigma and less client
satisfaction. The results reveal that associative stigma is related to more depersonalization, more emotional
exhaustion, and less job satisfaction among mental health professionals.
Keywords
associative stigma, burnout, job satisfaction, mental health services, multilevel research, self-stigma, service
user satisfaction
at Universiteit Gent-Vakgroep on March 1, 2012hsb.sagepub.com Downloaded from

Page 3

18
Journal of Health and Social Behavior 53(1)
health professionals experience because they are
associated with persons who belong to a stigmatized
category in society, namely, people with mental
health problems. We investigate the association of
associative stigma with three dimensions of burnout
and job satisfaction among professionals and the
link of associative stigma with self-stigma and cli-
ent satisfaction among service users.
BACKGrOUnD
Stigma and Mental Health Professionals
Mental health professionals are generally thought
to experience stigma in two primary ways (Schulze
2007). First, professionals are often—implicitly or
explicitly—regarded as perpetrators of stigma.
Several recent empirical studies document the
stigmatizing attitudes of mental health profession-
als (Hugo 2001; Jorm et al. 1999; Nordt, Rossler,
and Lauber 2006) and reveal how mental health
professionals’ treatment of service users contrib-
utes to stigmatization for service users (Chaplin
2000; Sartorius 2002). Second, mental health pro-
fessionals may be considered victims of stigma
when the general public has negative attitudes
about them (Von Sydow and Reimer 1998). For
example, McGuire and Borowy (1979) found that
mental health professionals have lower profes-
sional prestige than do other health professionals.
This article draws heavily on Halter’s (2008)
concept of “associative stigma.” Associative
stigma, or courtesy stigma (Goffman 1963), can be
defined as stigma that persons experience not
because of their own (attributed) characteristics
but because they are associated with persons who
belong to a stigmatized category in society. This
concept has mainly been applied to family mem-
bers of stigmatized persons (Angermeyer, Schulze,
and Dietrich 2003; Phelan 2005; Phelan, Bromet,
and Link 1998).
However, the concept can also be applied to
mental health professionals, who may be the target
of negative attitudes or treatment because they are
associated with the target group they serve. As
Halter (2002:24) describes, “the public . . . may
tend to associate the nurses with the pathologies
they treat, resulting in the stigmatization of the
specialty itself.” Mental health specialties such as
psychiatric nursing appear to have a very low sta-
tus when compared with other nursing specialties.
A recent study found that nurses, in general, per-
ceive psychiatric nurses as unskilled, illogical,
idle, and disrespected (Halter 2008). Schulze
(2007:145) reports that mental health professionals
are portrayed in the media as “neurotic, unable to
maintain professional boundaries, drug or alcohol
addicted, rigid, controlling, ineffectual, mentally
ill themselves, comically inept, uncaring, self-
absorbed, having ulterior motives, easily tricked
and manipulative, foolish and idiotic.”
Both the original labeling perspective (Scheff
1966) and the modified labeling perspective (Link
et al. 1989) acknowledge that mental health ser-
vices are crucial locations for the labeling and
elicitation of stigma associated with mental health.
However, the precise mechanisms that operate
within mental health services and the role that
professionals and their interactions with service
users play in contributing to stigma remain unclear.
Clarifying the specific mechanisms of the stigma
process can enhance our understanding of how
mental health stigma operates.
From Goffman’s (1963) work, we know that
courtesy stigma is acquired through connections
with stigmatized people. However, much of our
knowledge about courtesy, or associative, stigma
processes in general is derived from studies on
family members of persons with mental health
problems. This perspective provides a limited
understanding of associative stigma that is particu-
lar to family relationships and does not necessarily
translate to other kinds of relationships. The
dynamics between mental health professionals and
their service users, which is a professional rela-
tionship, may be very different from those among
family members. Whereas professionals are con-
nected with service users only through a social
relationship, family members also have a biologi-
cal relationship (Phelan 2005). Moreover, the pro-
fessional nature of the social relationships between
mental health professionals and users may mani-
fest in several ways (Kitson 2003; Scanlon 2006).
Professional relationships are work related and
limited in time and place and are also less personal
because a team of professionals is typically respon-
sible for several service users, and often for one
aspect of care, in accordance with the specific
profession. In addition, the contractual and explicit
therapeutic nature of the care relationship is in
contrast to lay caring, which is based on motives
such as love, altruism, duty, and necessity. For
these reasons, we pay particularly close attention
to the work context and the specific dynamics
between mental health professionals and mental
health service in this study. However, we expect
some degree of similarity across contexts, such as
the consequences of stress for mental health or the
at Universiteit Gent-Vakgroep on March 1, 2012hsb.sagepub.comDownloaded from

Page 4

Verhaeghe and Bracke
19
processes of emotional contagion in relationships
(see below). Therefore, a better understanding of
the consequences of associative stigma processes
for mental health professionals might also enhance
our understanding of the social complexities of
mental health care more broadly.
This study focuses on two research questions
concerning the effects of associative stigma. First,
in line with research on the adverse effects of
stigma for service users (Link et al. 2001; Markow-
itz 1998) and the negative consequences of asso-
ciative stigma perceived by family members
(Östman and Kjellin 2002), we investigate the
effects of associative stigma on the work-related
well-being of mental health professionals. Second,
as the social interaction between service providers
and service users constitutes a key process in men-
tal health care, this study will also investigate the
relationship between associative stigma among
professionals’ and service users’ well-being.
Associative Stigma among Mental Health
Professionals and Their Work-Related Well-
Being
In our examination of the effects of associative
stigma on mental health professionals, we focus spe-
cifically on the effects on work-related well-being,
given that associative stigma among mental health
professionals is acquired through association with
their target groups in their workplaces. We treat asso-
ciative stigma as a job stressor, analogously to other
studies that consider stigma a general stressor (Rüsch
et al. 2009). We build on the social stress perspective
(Pearlin 1981, 1989) as well as perspectives that spe-
cifically emphasize the relationship between working
conditions and mental health (Karasek 1979) and
utilize burnout and job satisfaction as indicators of
work-related well-being.
Burnout is a key indicator of work-related
well-being, especially among persons performing
“people-work” or “emotion-work.” Work-related
well-being is conceived of as a response to chronic
emotional and interpersonal stressors on the job
(Maslach, Schaufeli, and Leiter 2001). We treat
associative stigma as one of these chronic emo-
tional and interpersonal stressors that might pro-
duce burnout and will investigate the relationship
of associative stigma with each of three dimen-
sions of burnout—cynicism or depersonalization,
emotional exhaustion, and perceived inefficacy
(Maslach et al. 2001). Job satisfaction is a more
general indicator of work-related quality of life.
We expect that the distress related to associative
stigma will eventually decrease job satisfaction.
Both indicators of work-related well-being are
interrelated, and their causal order is not always
clear-cut (Dolan 1987), but in line with other
research (Maslach et al. 2001; Wolpin, Burke, and
Greenglass 1991), we consider job satisfaction as
the final outcome indicator.
Other determinants of work-related well-being
should not be ignored, however. Autonomy and
support, which are assumed to impede burnout and
stimulate job satisfaction, represent two job char-
acteristics that are particularly important in human
service organizations in general and in (mental)
health care specifically (Ben-Zur and Michael
2007; Karasek 1979; Maslach et al. 2001; Ross,
Altmaier, and Russell 1989). Furthermore, we
include an indicator of the mental health status of
professionals to control for their affective status as
a source of common method variance (Podsakoff,
McKenzie, and Lee 2003). This leads to our first
two main hypotheses: (1) A larger number of asso-
ciative stigma experiences will be associated with
more burnout for mental health professionals and
(2) a larger number of associative stigma experi-
ences will be associated with less job satisfaction
for mental health professionals due to burnout (see
Figure 1). Our examination of both relationships
will control for the level of job autonomy and con-
trol, support from colleagues, and professionals’
mental health status.
Associative Stigma among Mental Health
Professionals and the Well-Being of Mental
Health Service Users
Our second goal is to investigate whether associa-
tive stigma among professionals is associated with
the well-being of service users. We anticipate an
association, given the crucial role that interper-
sonal contact plays in the provision of mental
Associa?ve
s?gma
(professionals)
Client
sa?sfac?on
(service users)
Job sa?sfac?on
(professionals)
Burnout
(professionals)
Self-s?gma
(service users)
Figure 1. Conceptual Model
at Universiteit Gent-Vakgroep on March 1, 2012hsb.sagepub.com Downloaded from

Page 5

20
Journal of Health and Social Behavior 53(1)
health care. The job stress literature has demon-
strated the link between job stress and perfor-
mance. Professionals’ working conditions in
mental health services can affect their interactions
with service users (Holland et al. 1981; Weisman
and Nathanson 1985). Moreover, higher stress
among health professionals is associated with
worse interpersonal performance with service
users (Motowidlo, Packard, and Manning 1986;
Stewart and Barling 1996). In a review study,
Cohen (1980:95) suggests that stress is related to
“a decreased sensitivity to others. This includes a
decrease in helping, a decrease in recognition of
individual differences and an increase of aggres-
sion.” This decreased sensitivity to others reflects
one of the dimensions of burnout, as mentioned
above. Thus, professionals with more associative
stigma experiences could be less involved with the
service users, resulting in strained relationships
between professionals and service users. These
poor relationships may, in turn, enhance the stigma
experiences of service users (Goffman 1963).
Schulze (2007:138–39) showed that poor contact
with mental health professionals and poor quality
of mental health services are the second and third
most frequent stigma experiences mentioned by
service users and their families: “Patients felt stig-
matized by a lack of interest in their person . . .
patients did not get the personal attention they
needed, craving for personal contacts with some-
one else than their fellow-patients.” Another study
revealed that impersonal and standardized care can
contribute to self-stigma (Verhaeghe and Bracke
2008).
Stigma experiences of service providers might
also enhance feelings of stigma among service
users because of processes of emotional contagion
that can occur in emotional labor (Pugh 2001).
Displayed emotions of service providers can affect
the moods of service users and thus have an impact
on their attitudes toward the services (Barger and
Grandey 2006; Pugh 2001). Emotional reactions
are a key element in the stigma process (Link et al.
2004). Therefore, professionals with associative
stigma experiences could display their related
emotions, which in turn could affect the emotional
state of their service users.
This leads to our third and fourth research
hypotheses: Experiences of associative stigma will
have negative effects on burnout and job satisfac-
tion, which in turn (3) will be positively associated
with self-stigma among service users and (4)
will be negatively associated with service user
satisfaction.
Whereas previous empirical studies on the link
between work-related well-being of professionals
and satisfaction of service users have already con-
firmed the link between professionals’ burnout and
clients’ satisfaction (Halbesleben and Rathert
2008; Leiter, Harvie, and Frizzell 1998) and the
link between professionals’ job satisfaction and
clients’ satisfaction (Weisman and Nathanson
1985), this study will focus on the role of stigma
experiences of professionals and service users. The
third and fourth hypotheses will be analyzed from
the point of view of service users, with their self-
stigma experiences and satisfaction as dependent
variables (see Figure 1). We control for two other
dimensions of stigma experiences—social rejec-
tion and stigma expectations—that do not directly
refer to experiences within the current mental
health service organization but could affect self-
stigma experiences. Furthermore, we will take into
account service users’ mental health status, length
and intensity of their current treatment, length of
their total treatment history, age, gender, educa-
tion, income, and marital status.
DATA AnD METHODS
Data
We used survey data from a larger research project
on stigma established in 2005 in Flanders, the
Dutch-speaking region of Belgium. Belgium is
characterized by a very late deinstitutionalization
of mental health patients and still has one of the
largest numbers of psychiatric beds per 100,000
inhabitants in Europe. A two-stage sampling pro-
cedure was used: First, organizations were sam-
pled, and second, service users and providers
within these organizations were selected. Five
types of services were enrolled in the study.
Psychiatric hospitals offer only mental health care,
whereas psychiatric wards in general hospitals
provide specialized mental health care within the
context of general hospitals. Whereas both provide
predominantly full-time inpatient care, part-time
at Universiteit Gent-Vakgroep on March 1, 2012hsb.sagepub.comDownloaded from

Page 6

Verhaeghe and Bracke
21
care is increasing. In addition to medical care and
psychotherapy, a whole range of vocational or day-
structuring activities is offered in these settings.
Community mental health centers offer ambula-
tory mental health care for service users who come
once a week or once a month, most often for coun-
seling. Psychiatric rehabilitation centers and day
activity centers are settings that day patients typi-
cally attend during the week. The composition of
the service user population and the type of activi-
ties offered (e.g., counseling, vocational activities,
recreational activities) varies considerably. In
2005, Flanders counted 38 psychiatric hospitals,
34 psychiatric wards of general hospitals, 76 com-
munity mental health services, 9 psychiatric reha-
bilitation centers, and 47 day activity centers. To
obtain diversity with regard to type of institution,
we included a sample of 8 organizations from each
organization type described above. These organi-
zations were randomly selected; refusals (n = 10)
were replaced by other organizations from the
same type, except when they were announced too
late (n = 3) or when no other organization of that
type existed (n = 1). From the 46 organizations that
were contacted, the final sample size was 36 (78
percent): 8 psychiatric hospitals, 7 general hospi-
tals, 8 day activity centers, 7 psychiatric rehabilita-
tion centers, and 6 community mental health
centers.
Within these organizations, units specifically
for youth or the elderly were excluded, as were
units specializing in service users with cognitive
disorders or mental retardation. Furthermore,
among the selected centers and units, service users
were excluded if they had cognitive disorders or
mental retardation, if they were in a stage of too
acute illness to participate (determined by profes-
sionals), and if their knowledge of Dutch was
insufficient. The service users who met the criteria
and who were present on a date that had been
agreed on beforehand with the supervisor were
invited to participate. Informed consent was
obtained after an introduction by the researcher. Of
the 1,174 eligible service users, 846 (72 percent)
agreed to participate. Within the selected centers
and units, all mental health professionals were
invited to complete questionnaires; 597 of the 954
invited (63 percent) agreed to participate. Due to
ethical considerations and privacy legislation, we
were not able to gather information about the ser-
vice users and professionals who were unwilling to
participate in the study. Due to missing values on
some key variables, the final working sample for
the current analyses includes 707 service users and
543 professionals. Of the 707 service users, 389
receive services at psychiatric hospitals, 102 at
psychiatric wards of general hospitals, 66 at com-
munity mental health centers, 92 at day activity
centers, and 58 at psychiatric rehabilitation cent-
ers. Their mean age is 39.59 (SD = 11.76), and 54.9
percent are women. Of the 543 professionals, 349
provide services at psychiatric hospitals, 80 at
psychiatric wards of general hospitals, 36 at com-
munity mental health centers, 37 at day activity
centers, and 41 at psychiatric rehabilitation cen-
ters. The working sample consists of 272 psychiat-
ric nurses, 73 general nurses, 64 psychologists, 60
vocational trainers, 47 social workers, 21 peda-
gogues, 15 physiotherapists, 14 psychiatrists, 14
social nurses, and 96 service providers in a miscel-
laneous category consisting of a diversity of occu-
pations. The mean age of the sample is 37.6 (SD =
10.45); 75.1 percent are women.
Measures
Mental health professionals. Concerning associa-
tive stigma among mental health professionals, no
standardized measure exists. We used four items
with five answer categories ranging from 1 (never)
to 5 (often), to ask professionals (1) whether some
people react negatively when they hear the profes-
sionals work in mental health care organizations,
(2) whether some people make jokes about the pro-
fessionals’ working there, (3) whether the
professionals feel ashamed of working there, and
(4) whether they are sometimes reluctant to tell
other people where they work (see the appendix). A
majority of the professionals answered “never” to
the last two questions. Furthermore, the original
measure had an alpha reliability coefficient of .51,
which is moderate for a four-item scale. Therefore,
the instrument was not used as a Likert-type scale
but instead as an index. After having dichotomized
the items by distinguishing experiences that never
happened from those that at least seldom occurred,
affirmative answers to the four items were summed.
The index, which refers to the number of types of
at Universiteit Gent-Vakgroep on March 1, 2012 hsb.sagepub.comDownloaded from

Page 7

22
Journal of Health and Social Behavior 53(1)
stigma experiences one ever had, has values from 0
(no stigma experiences) to 4 (four types of stigma
experiences) (M = 1.73, SD = 0.84).
Burnout was measured by a Flemish translation
of the Maslach Burnout Inventory (Vlerick 1995),
with 22 items with scores from 0 (never) to 6
(every day). Mean scores on the three conventional
subscales were computed. Depersonalization was
measured using five items (M = 0.72, SD = 0.66,
alpha = .59), emotional exhaustion was measured
using nine items (M = 1.30, SD = 0.89, alpha =
.83), and the experience of failing in personal
accomplishment was measured using eight items
(M = 1.09, SD = 0.80, alpha = .79).
Job satisfaction was operationalized by a Likert-
type scale composed of five items derived from Pfef-
fer and Davis-Blake (1990) and Martin and Roman
(1996), with scores ranging from 1 (absolutely disa-
gree) to 5 (absolutely agree). The scale refers to
general feelings of pleasure and motivation in the
current job, using items such as “I do my job with
pleasure” (M = 4.03, SD = 0.70, alpha = .83).
Job autonomy and control were measured by an
instrument (Haynes et al. 1999) consisting of six
items with scores from 1 (not at all) to 5 (very
much) (M = 3.20, SD = 0.76, alpha = .82). An
example item is, “To what extent do you determine
the methods and procedures you use in your
work?”
To operationalize relationships with colleagues,
a four-item indicator was used (Haynes et al.
1999), with answering scores ranging from 1 (not
at all) to 5 (totally) (M = 4.18, SD = 0.71, alpha =
.88). An example item is, “To what extent can you
count on your colleagues to help you with a diffi-
cult task at work?”
To measure mental health status, a 12-item ver-
sion of the General Health Questionnaire (Goldberg
and Hillier 1979) was used. Mean scores range from
1 to 4, with higher scores’ referring to better mental
health (M = 3.16, SD = 0.29, alpha = .81).
Finally, two main sociodemographic character-
istics were included as control variables: gender
(women = 1, men = 2) and age (in years).
Mental health service users. Self-stigma is con-
ceptualized as feelings of shame and inferiority
that are directly related to treatment in the current
mental health service organization. We used a mea-
sure derived from Fife and Wright (2000) consisting
of 5 items with scores from 1 (completely false) to
5 (completely true) (M = 2.76, SD = 1.29, alpha =
.92). An example item is, “Since I have come to
this center, I have come to feel inferior.” Social
rejection refers to negative interactions with people
outside of the center due to having received help at
the center. It was measured by a scale derived from
Fife and Wright, using 5 items with scores from 1
(completely false) to 5 (completely true) (M = 3.11,
SD = 1.22, alpha = .91). An example item is, “Since
I have come to this center, some people treat me
with less respect.” Stigma expectations refer to
beliefs about what the general population thinks
about persons with mental health problems, and it
was measured by the Devaluation Discrimination
scale developed by Link et al. (1989). We trans-
lated this scale and slightly adapted it by replacing
references to “mental hospital” and “ex-…” with
“persons who receive(d) psychological help,” as
suggested by Link et al. (2002). This instrument
consists of 12 items with answer categories from 1
(completely disagree) to 4 (completely agree) (M =
2.69, SD = 0.42, alpha = .83).
Client satisfaction was measured using a Dutch
translation (De Brey 1983) of the Client Satisfac-
tion Questionnaire (Nguyen, Attkisson, and Steg-
ner 1983). It consists of 8 items such as “How
would you rate the quality of the services you
received?” with scores from 1 to 4 (M = 3.07, SD
= 0.53, alpha = .89). Psychiatric symptoms were
operationalized by the 18-item version of the Brief
Symptom Inventory (Derogatis 2001), using a
Dutch translation of its parent instrument, the
SCL-90 (Arrindell and Ettema 1986). The items
are scored from 0 (not at all) to 4 (always) (M =
1.43, SD = 0.94, alpha = .94).
Professionals provided information about the
psychiatric diagnosis of service users. Three main
diagnostic categories were used as dichotomous
variables (1 = present, 0 = absent)—mood-related
disorders (29.1 percent), psychotic-related disor-
ders (19.3 percent), and substance-related disor-
ders (26.8 percent). Service users for whom
information was missing—nearly 9 percent of the
sample—were included as a separate category.
Length of current treatment was measured in
months (M = 10.69, SD = 8.68), whereas number
of years since first treatment (M = 17.20, SD =
31.8) was computed as the difference between
at Universiteit Gent-Vakgroep on March 1, 2012 hsb.sagepub.comDownloaded from

Page 8

Verhaeghe and Bracke
23
current age and the age at which the service user
received mental health care for the first time in his
or her life. Intensity of current treatment was meas-
ured as the number of hours a week the service
user spends in the current mental health center
(M = 79.81, SD = 68.39).
Finally, the following background variables
were taken into account: gender (women = 1, men
= 2), age (in years), and marital status (married or
cohabiting = 1, 24.7 percent; single, divorced, or
widowed = 0). Education was measured by means
of four categories (primary education = 1, college
or university degree = 4; M = 2.92, SD = 0.77).
Income was measured by a proxy variable indicat-
ing how easily one can get by with one’s income
(1 = very difficult, 6 = very easy; M = 3.31,
SD = 1.45).
Analyses
To respect the clustered data-sampling procedure,
multilevel analyses were performed using the pro-
gram HLM 6 (Raudenbush, Bryk, and Cheong
2004), which accounts for the nonindependence of
the observations. To address our first research
problem concerning the professionals, the follow-
ing model was specified for each of the four
dependent variables:
= γ+γ++
∑
p
0010
,
ijp pijjij
YXur
with γ00 as the intercept, γ1p as p indexed first-level
regression coefficients (i.e., fixed effects corre-
sponding to gender, age, mental health status,
autonomy, etc.), rij as the individual-level error,
and u0j as the organization-level error. The errors
are assumed to follow normal distributions with
means 0 and variances σ² and τ00, respectively.
Concerning the second research problem, data
on two levels were needed. We could not link the
data for each service user with characteristics of
his or her service provider in a direct way, as no
one-to-one relationship between service users and
service providers exists. Since a team of profes-
sionals share responsibility for a service user, data
were aggregated to the team level by taking the
average value for each variable for the team.
Therefore, a multilevel analysis with two levels
was performed for service users, with characteris-
tics of the service users at the lowest level and
aggregate scores of mental health professionals at
the highest level. The models were specified as
follows:
00100
,
= γ+γ+γ++
∑
p
∑
q
ijppijqqjjij
YXZur
with γ00 as the intercept, γ1p as p first-level fixed
effects corresponding to service user covariates,
γ0q as q second-level fixed effects corresponding to
service provider covariates, rij~N(0, σ²) as the indi-
vidual level error, and u0j~N(0, τ00) as the level-2
error.
The models specified in equations 1 and 2 were
estimated using the restricted maximum likelihood
method as the sample size for each unit was rela-
tively small. For all variables at the interval or ratio
level, grand mean centering was applied. No sam-
pling weights were used. Within scales, item cor-
relation substitution was used to deal with missing
values: A missing value was replaced by the value
on the item within the scale that has the highest
correlation with the item (Huisman 1999). Cases
with missing data information on other variables
were omitted from the analyses (listwise deletion).
Finally, standardized multilevel regression coeffi-
cients are reported in the next section. They are
based on the fixed effects reported by HLM 6, with
robust standard errors. These coefficients are com-
puted by multiplying the unstandardized coeffi-
cients with the standard deviation of the respective
variable and dividing by the standard deviation of
the dependent variable.
rESUlTS
Associative Stigma Experiences among
Mental Health Professionals
Generally speaking, experiences of associative
stigma can be considered rather low (see the
appendix). The mean score is 1.74, which means
that service providers experience, on average,
nearly two of the four types of associative stigma.
Very few professionals mentioned feeling ashamed
of working in the current mental health center or
hesitant to tell other people about it. In contrast, a
(1)
(2)
at Universiteit Gent-Vakgroep on March 1, 2012hsb.sagepub.comDownloaded from

Page 9

24
Journal of Health and Social Behavior 53(1)
larger number of professionals acknowledged
experiencing negative reactions to their jobs, and
even more of them reported that people make jokes
about their work.
We found no differences in associative stigma
between women (M = 1.76, SD = 0.79) and men
(M = 1.67, SD = 0.98) or according to age (p = .621).
However, professionals with better mental health
reported significantly less stigma (p = .016). Further-
more, we found differences between professional
groups: Nurses reported the most stigma experiences
(M = 1.90), followed by physiotherapists (M = 1.65),
social workers (M = 1.47), and psychologists (M =
1.38). Furthermore, differences between the types of
organizations appeared: psychiatric hospitals (M =
1.85), psychiatric wards of general hospitals (M =
1.86), community mental health centers (M = 1.64),
day activity centers (M = 1.41), and psychiatric reha-
bilitation centers (M = 0.88).
Associative Stigma among Mental Health
Professionals and Their Work-Related Well-
Being
Associative stigma is positively associated with
depersonalization and emotional exhaustion but
not with failure in personal accomplishment (see
Table 1). Furthermore, associative stigma is nega-
tively associated with job satisfaction. This effect
diminishes slightly when controlling for burnout,
revealing that the negative effect of stigma on job
satisfaction can be partially attributed to higher
levels of emotional exhaustion. For both deperson-
alization and emotional exhaustion, associative
stigma is an even more important determinant than
job autonomy. Furthermore, job autonomy is asso-
ciated with fewer experiences of failure of per-
sonal accomplishment. In addition, professionals
who feel more work autonomy are more satisfied
with their jobs, which can be partially attributed to
greater feelings of personal accomplishment.
Supportive relationships with colleagues seem
important for job satisfaction and for all dimen-
sions of burnout; the exception is depersonaliza-
tion (p = .052). More positive mental health status
is associated with less burnout and more job satis-
faction. No consistent relationships were found
regarding sociodemographic background vari-
ables. Men report more depersonalization com-
pared to women, whereas older professionals
report less depersonalization. Men are less satis-
fied with their jobs, which can be attributed to
greater feelings of depersonalization. Furthermore,
older professionals are more satisfied with their
Table 1. The Effect of Associative Stigma, Job Characteristics, and Control Variables on Three Dimensions
of Burnout and Job Satisfaction among Mental Health Professionals—results from a Multilevel Analysis
(Standardized Multilevel regression Coefficients)
Depersonalization
Emotional
Exhaustion
Failure of Personal
Accomplishment
Job
Satisfaction
Gender
Age
Mental health status
Support colleagues
Autonomy
Associative stigma
Depersonalization
Emotional exhaustion
Failure of personal
accomplishment
Variance components
τ00
σ²
.172***
−.133***
−.161**
−.086
−.045
.151***
.053
−.059
−.258***
−.165***
−.079
.136**
−.017
−.021
−.182***
−.167***
−.181***
.027
−.100*
.067*
.163***
.265***
.246***
−.161***
−.077
.043
.082*
.209***
.206***
−.126**
−.088
−.134**
−.168***

.004
.391
.061
.499
.009
.691
.02
.343
.011
.319
Note: Deviance statistics not reported since restricted maximum likelihood was used as method of estimation.
*p < .05. **p < .01. ***p < .001.
at Universiteit Gent-Vakgroep on March 1, 2012 hsb.sagepub.com Downloaded from

Page 10

Verhaeghe and Bracke
25
jobs, which is attributable to lower levels of
depersonalization. In sum, findings indicate that
associative stigma is associated with more deper-
sonalization and emotional exhaustion and with
less job satisfaction, which largely supports our
first and second hypotheses.
Associative Stigma among Mental Health
Professionals and the Well-Being of Mental
Health Service Users
Associative stigma is positively associated with
self-stigma among service users (see Table 2,
Model 1.1). This effect is not explained by any one
of the three dimensions of burnout generally or by
depersonalized treatment of service users specifi-
cally (see Model 1.2), showing that higher levels
of self-stigma among service users are not attribut-
able to more impersonal treatment by the profes-
sionals. Burnout has no significant effect on
self-stigma. Furthermore, enhanced feelings of
self-stigma among service users of units where
professionals have higher levels of associative
stigma cannot be attributed to the job satisfaction
of the professionals (see Model 1.3).
We also found that service users receiving care
in units where the professionals indicate more
associative stigma are less satisfied with the ser-
vices (see Model 2.1). Like the results for self-
stigma, this association cannot be explained by any
of the dimensions of burnout, and none of the latter
is associated with service user satisfaction (see
Model 2.2). However, lower levels of service user
satisfaction in units where professionals report
more associative stigma can be explained by ser-
vice users’ self-stigma (see Model 2.3.A). Associa-
tive stigma therefore appears to enhance self-stigma
among service users, which then leads to lower
service user satisfaction. However, another expla-
nation for the link between associative stigma
among professionals and service user satisfaction
is also supported by the data (see Model 2.3.B). In
units with higher levels of associative stigma, pro-
fessionals are less satisfied with their jobs, as
indicated earlier in Table 1. This lower job satis-
faction level seems to partially explain why higher
associative stigma levels are linked with lower
service user satisfaction. In units with higher job
satisfaction levels, service users are more satisfied.
Concerning the other service user–level varia-
bles, the strongest effect is found for symptoms:
The more symptoms are present, the more self-
stigma and the less service user satisfaction are
reported. The importance of self-stigma is revealed
by the fact that nearly half of the effect of symp-
toms on service user satisfaction is due to self-
stigma and that self-stigma is the most important
determinant of service user satisfaction. Further-
more, service users experiencing more social
rejection report more self-stigma and less service
user satisfaction. The latter effect is explained
when taking self-stigma into account, however,
which again confirms the importance of self-
stigma. In addition, service users with prolonged
service use over the course of their lives report less
self-stigma, whereas users with more intensive
current treatment report more self-stigma. Finally,
men, younger people, and those with psychotic
disorders are generally less satisfied with the ser-
vices received.
To summarize, hypotheses 3 and 4 were only
partially supported by the data. In accordance with
the third hypothesis, we found a positive relation-
ship between associative stigma among profes-
sionals and self-stigma among service users.
However, this link was not explained by burnout.
Hypothesis 4 was partially confirmed because we
found a negative link between associative stigma
and service user satisfaction, which was reduced
when taking self-stigma into account. However, an
alternative mechanism, via job satisfaction, was
also supported by the data.
DISCUSSIOn AnD COnClUSIOn
The point of departure for this article is that mental
health professionals may experience associative
stigma and that this may affect their work experi-
ences and the quality of the service they provide.
Two questions were addressed: First, what are the
effects of associative stigma on professionals’
work-related well-being? Second, what are the
consequences for the well-being of their service
users? We found that associative stigma is associ-
ated with depersonalization and emotional exhaus-
tion among mental health professionals, and the
latter contributes to a decrease in job satisfaction.
Associative stigma among professionals can
at Universiteit Gent-Vakgroep on March 1, 2012hsb.sagepub.comDownloaded from

Page 11

26
Table 2. The Effect of Associative Stigma and Burnout among Mental Health Professionals on Experiences of Self-Stigma and Client Satisfaction among Mental
Health Service Users. Controlled for Service Users’ Sociodemographic Characteristics and Features of Mental Health (Service Use)—results from Multilevel
Analyses (Standardized Multilevel regression Coefficients)
Self-Stigma
Client Satisfaction
Model 1.1
Model 1.2
Model 1.3
Model 2.1
Model 2.2
Model 2.3.A
Model 2.3.B
Model 2.4
Service provider variables
Associative stigma
.134***
.147***
.146***
−.119*
−.120*
−.082
.063
–.043
Depersonalization
−.018
−.019
−.032
−.040
.063
–.041
Faillure of personal accomplishment
.075
.074
−.017
−.001
.078
.101
Emotional exhaustion
−.012
−.012
.097
.087
.065
.117
Job satisfaction
−.001
.080*
.166*
Service user variables

Gender
−.036
−.037
−.037
−.090*
−.090*
−.102**
−.092*
–.104**
Age
.070
.065
.064
.098°
.100*
.113*
.096*
.108*
Education
.021
.024
.024
−.052
−.053
−.049
−.054
–.050
Marital status
.008
.012
.012
.028
.027
.030
.026
.029
Income
.029
.030
.030
.049
.050
.058
.050
.058
Symptoms
.320***
.321***
.321***
−.183***
−.186***
−.103**
−.183***
–.100*
number of years since first treatment
−.090*
−.085*
−.084*
.031
.033
.013
.038
.018
length current treatment
.021
.018
.018
−.017
−.013
−.009
−.009
–.005
Intensity current treatment
.110**
.103**
.103**
−.047
−.030
−.004
−.035
–.010
Mood disorder
−.042
−.040
−.040
−.032
−.033
−.044
−.091*
–.014
Psychotic disorder
.047
.042
.041
−.082*
−.084*
−.080
.031
.028
Substance disorder
−.068
−.069
−.069
−.044
−.044
−.058
−.036
–.048
Missing
−.023
−.018
−.019
−.044
−.061
−.069
−.087
–.083
Social rejection
.317***
.319***
.319***
−.097*
−.091*
−.015
−.041
–.055
Stigma expectations
−.012
−.010
−.010
.034
.033
.029
−.072
–.080
Self-stigma
−.252***
–.253***
Variance components

τ00
.041
0.037
.041
.036
.038
.032
.034
.028
σ²
1.104
1.106
1.105
.222
.222
.212
.222
.212
Note: Deviance statistics are omitted; see comment in Table 1.
°p < 0.10. *p < .05. **p < .01. ***p < .001.
at Universiteit Gent-Vakgroep on March 1, 2012 hsb.sagepub.comDownloaded from

Page 12

Verhaeghe and Bracke
27
clearly be considered a job stressor, in line with
other studies considering stigma a stressor for
those stigmatized (Rüsch et al. 2009) and for those
associated with the stigmatized such as family
members (Baxter 1989; Östman and Kjellin 2002).
Although professionals differ from family mem-
bers in their relationships with service users in
several ways, similar processes with regard to the
link between stigma and well-being appear. Our
finding of these similarities is in accordance with
the findings of other scholars who stress that dif-
ferences between lay caring and professional car-
ing should not be exaggerated (Hem and Heggen
2003; Kitson 2003): considering, for instance, that
emotions cannot be eliminated from professional
care.
Findings further demonstrate that the detrimen-
tal effects of associative stigma among service
providers spill over to mental health service users,
as associative stigma is also associated with self-
stigma and dissatisfaction among service users.
Regarding the link between associative stigma and
service user satisfaction, two processes emerged
from the data. First, associative stigma is related to
job dissatisfaction, which is directly linked with
service user satisfaction. A similar association
between provider satisfaction and user satisfaction
has been found in other empirical studies (Weis-
man and Nathanson 1985). The fact that the rela-
tionship between professionals’ associative stigma
experiences and service user satisfaction can be
explained by job satisfaction is in accordance with
the job stress literature, which states that job satis-
faction is related to better interpersonal perfor-
mance (Cohen 1980). The quality of relationships
between professionals and service users is crucial
to the provision of quality mental health services.
Not surprisingly, service users are more satisfied in
units with higher levels of job satisfaction among
professionals. Emotional contagion effects could
be involved, too, when displayed emotions related
to satisfaction among professionals affect the
mood of service users, resulting in higher levels of
reported service user satisfaction (Barger and
Grandey 2006; Pugh 2001).
We also found that associative stigma is directly
related to self-stigma among service users, which
seems to contribute to dissatisfaction with mental
health services. This process is also in accordance
with arguments that associative stigma worsens
interpersonal relations between service providers
and users. This finding supports other studies
showing how negative interpersonal relationships
can lead to (self-)stigma among service users
(Schulze 2007; Verhaeghe and Bracke 2007),
which in turn enhances feelings of dissatisfaction.
Emotional contagion processes, which link stigma
experiences of service providers and users, could
also be at play, but that seems less plausible as
only a few providers report feelings of shame or
inferiority.
Several limitations of this study should be
noted. First, in the absence of a well-defined stand-
ardized measure of associative stigma among ser-
vice providers, we designed an instrument for this
study. This inventory needs to be developed fur-
ther, taking into account that negative reactions
from others are more apparent than negative reac-
tions or emotions from service providers them-
selves. Second, the cross-sectional design of the
study limits our confidence concerning the causal
nature of the relationships. Alternative causal path-
ways cannot be excluded. Service providers with
low job satisfaction levels might, for instance,
report more associative stigma because they see all
work-related issues in a negative way. However,
given that the association between associative
stigma and job satisfaction remains after control-
ling for mental health status, this alternative path-
way seems less plausible. Furthermore, associative
stigma experiences generally involve how others
react to the professionals. It might be less plausible
that job dissatisfaction—which is not necessarily
reported to others—could elicit stigmatizing reac-
tions in the environment of professionals. The
reasoning applies even more for service users. An
alternative causal pathway would be that high lev-
els of self-stigma in service users would lead to an
environment in which professionals react in a more
negative way. As we do not assume a direct rela-
tionship between the environment of professionals
and the service users, this pathway seems less
plausible.
Third, the results are limited to Belgium, which
is characterized by a late deinstitutionalization of
mental health patients. The issue of stigma might
be especially relevant in such a context. There is
some evidence that stigma might be lower in more
at Universiteit Gent-Vakgroep on March 1, 2012hsb.sagepub.comDownloaded from

Page 13

28
Journal of Health and Social Behavior 53(1)
deinstitutionalized care (Angermeyer, Link, and
Majcherangermeyer 1987; Verhaeghe, Bracke, and
Bruynooghe 2007). Furthermore, cross-national
differences in levels of stigma have been reported
(Pescosolido et al. 2008). Replication of this study
in other contexts is needed for greater certainty
about its generalizability.
Fourth, due to ethical considerations and privacy
legislation, information about people refusing to par-
ticipate could not be gathered. Also, selection effects
cannot be excluded. We believe that most of these
potential selection effects might contribute to an
underestimation of the levels of stigma and overesti-
mation of well-being as reported by both service
providers and users. Hence, we have reason to believe
that the detrimental outcomes of the processes
revealed in our analyses are underestimated. First,
stigma acts as a major barrier, respectively, to seeking
professional mental health care (Vogel et al. 2007)
and to choosing to work in the mental health profes-
sions (Cutler et al. 2009). In addition, both service
users and professionals with more severe stigma
experiences or lower satisfaction probably drop out
of the settings more readily. Furthermore, among
those providers and users who were eligible, those
with more stigma experiences and those with lower
satisfaction levels were probably less inclined to
participate in this study.
Fifth, the processes revealed could differ between
mental health services, as stigma experiences can be
affected by service characteristics (Angermeyer et al.
1987; Verhaeghe and Bracke 2007). Our descriptive
analysis revealed differences in associative stigma
between types of services and professionals, and
other studies have also revealed differences in atti-
tudes toward several professions (Von Sydow and
Reimer 1998). However, it would go too far to dif-
ferentiate between services and professionals in our
analyses.
Despite these shortcomings, this study is impor-
tant because it is the first to pay particular attention to
the empirical link between associative stigma experi-
ences among service providers and its consequences
for both service providers and service users, using a
multilevel research design and analysis techniques on
a relatively large data set.
What are the implications of an association
between associative stigma and service users’
well-being? Schulze (2007) pointed out that some
scholars consider professionals “victims” of stigma
because they are stigmatized too, while others see
professionals as “offenders” when they stigmatize
service users by the way they treat them. Our results
suggest that both aspects are present, as mental
health professionals who are victims become
offenders when the detrimental effects of associa-
tive stigma experiences spill over from providers to
service users due to specific features of the relation-
ship between professionals and service users in
human service organizations. An unexpected find-
ing was that burnout and particularly depersonalized
treatment play no role in the relationship between
associative stigma among service providers and
self-stigma among service users. Future research
could help to clarify which characteristics of the
relationship between professionals and service users
play a key mediating role. It is notable that, in gen-
eral, we found low levels of associative stigma and
almost no feelings of self-stigma or internalized
stigma among the service providers. Selection
effects could play a role here, leading service pro-
viders who are more vulnerable or sensitive to
associative stigma to leave their jobs. However,
more substantial reasons could be present, too.
Future research will need to investigate why profes-
sionals do not seem to internalize stigma.
This study has broader theoretical implications
for several research domains. First, in the stigma
domain, two processes related to stigma experi-
ences among mental health service users have been
stressed: (1) direct discrimination and devaluation,
which refer to negative behaviors from the general
public toward service users (Angermeyer and
Matschinger 2005; Scheff 1966), and (2) more
subtle social-psychological processes such as
expectations of devaluation and discrimination
(Link et al. 1989) and internalized stigma (Corri-
gan and Watson 2002) by service users. Our results
point to a third type of stigma process in which
professionals play a more active role: Their asso-
ciative stigma experiences seem to reproduce
stigma during service encounters.
Second, these findings have important implica-
tions for the work/health experiences of mental health
professionals. These professionals are known to
be particularly vulnerable to burnout because of
the nature of their work, such as the very intense
interaction with service users and the confrontation
at Universiteit Gent-Vakgroep on March 1, 2012hsb.sagepub.com Downloaded from

Page 14

Verhaeghe and Bracke
29
with challenging behaviors (Jenkins and Elliott 2004;
Moore and Cooper 1996; Sullivan 1993). This study
suggests that associative stigma should be added to
this list of demanding work features, as associative
stigma can be considered a specific job stressor for
mental health service providers, in analogy with
stigma, which can be considered a stressor for service
users (Rüsch et al. 2009).
Based on the results and their implications
summarized above, we conclude that associative
stigma deserves more research attention. Future
studies should start with the development of a
more sophisticated instrument for measuring asso-
ciative stigma. Also, a more diverse range of out-
come measures needs to be included when studying
the consequences for service providers and users.
In addition, researchers should pay attention to the
intermediate processes that link associative stigma
with service users’ well-being. Another research
domain that remains relatively unexplored is that
of the determinants of associative stigma. As not
all professionals report the same level of associative
stigma, it is important to determine what facilitates
or impedes these experiences.
The topic of associative stigma has important
policy implications. As far as we know, actions to
improve this type of work-related well-being have
not previously paid explicit attention to the issue of
stigma, with the exception of educational pro-
grams aimed at destigmatizing the profession to
make it more attractive for medical students (Cut-
ler et al. 2009). However, this study suggests that
stigma ought to be considered a real job stressor
for this professional group and that its importance
should not be ignored. Another policy implication
regards the finding that associative stigma among
professionals appears to be a stigma-enhancing
mechanism among service users, which has been
largely ignored; therefore, the role of professionals
might have been underestimated. Initiatives aimed
at reducing stigma experiences among service
users should pay more attention to mental health
professionals as both targets of the stigma associ-
ated with mental health care and contributors to the
stigma experiences of mental health service users.
FUnDInG
The authors disclosed receipt of the following financial
support for the research, authorship, and/or publication of
this article: This study is supported by a grant from the
Special Research Fund of Ghent University and by the
Flemish Research Council.
rEFErEnCES
Angermeyer, Matthias C., P. Breier, S. Dietrich, D. Ken-
zine, and H. Matschinger. 2005. “Public Attitudes
toward Psychiatric Treatment: An International Com-
parison.” Social Psychiatry and Psychiatric Epidemi-
ology 40:855–64.
Angermeyer, M. C., B. G. Link, and A. Majcheranger-
meyer. 1987. “Stigma Perceived by Patients Attend-
ing Modern Treatment Settings—Some Unanticipated
Effects of Community Psychiatry Reforms.” Journal
of Nervous and Mental Disease 175:4–11.
Angermeyer, Matthias C. and H. Matschinger. 2005.
“Labeling—Stereotype—Discrimination: An Investi-
gation of the Stigma Process.” Social Psychiatry and
Psychiatric Epidemiology 40:391–95.
Angermeyer, Matthias C., Beate Schulze, and Sandra Diet-
rich. 2003. “Courtesy Stigma: A Focus Group Study of
Relatives of Schizophrenia Patients.” Social Psychiatry
and Psychiatric Epidemiology 38:593–602.
Arrindell, W. A. and J. H. M. Ettema. 1986. SCL-90:
Handleiding bij een Multidimensionele Psychopa-
thologie-Indicator (SCL-90: Manual of a Multidi-
mensional Indicator of Psychopathology). Lisse, The
Netherlands: Swets & Zeitlinger.
Barger, P. B. and A. A. Grandey. 2006. “Service with a
Smile and Encounter Satisfaction: Emotional Conta-
gion and Appraisal Mechanisms.” Academy of Man-
agement Journal 49:1229–38.
Baxter, C. 1989. “Investigating Stigma as Stress in Social
Interactions of Parents.” Journal of Intellectual Dis-
ability Research 33:455–66.
Ben-Zur, H. and K. Michael. 2007. “Burnout, Social
Support, and Coping at Work among Social Workers,
Psychologists, and Nurses: The Role of Challenge/
Control Appraisals.” Social Work in Health Care
45:63–82.
Chaplin, R. 2000. “Psychiatrists Can Cause Stigma Too.”
British Journal of Psychiatry 177:467.
Cohen, S. 1980. “Aftereffects of Stress on Human
Performance and Social Behavior—A Review of
Research and Theory.” Psychological Bulletin
88:82–108.
Corrigan, P. W. and A. C. Watson. 2002. “The Paradox of
Self-Stigma and Mental Illness.” Clinical Psychology—
Science and Practice 9:35–53.
at Universiteit Gent-Vakgroep on March 1, 2012hsb.sagepub.com Downloaded from

Page 15

30
Journal of Health and Social Behavior 53(1)
Cutler, J. L., K. J. Harding, S. A. Mozian, L. L. Wright,
G. G. Pica, S. L. Masters, and M. J. Graham. 2009.
“Discrediting the Notion ‘Working With “Crazies”
Will Make You “Crazy”’: Addressing Stigma and
Enhancing Empathy in Medical Student Education.”
Advances in Health Sciences Education 14:487–502.
De Brey, H. 1983. “A Cross-National Validation of the
Client Satisfaction Questionnaire: The Dutch Experi-
ence.” Evaluation and Program Planning 6:395–400.
Derogatis, L. R. 2001. BSI-18: Administration, Scoring
and Procedures Manual. Minneapolis, MN: National
Computer Systems.
Dolan, N. 1987. “The Relationship between Burnout and
Job-Satisfaction in Nurses.” Journal of Advanced
Nursing 12:3–12.
Fife, B. L. and E. R. Wright. 2000. “The Dimensionality
of Stigma: A Comparison of Its Impact on the Self
of Persons with HIV/AIDS and Cancer.” Journal of
Health and Social Behavior 41:50–67.
Glozier, N. 1998. “Workplace Effects of the Stigmatiza-
tion of Depression.” Journal of Occupational and
Environmental Medicine 40:793–800.
Goffman, E. 1963. Stigma: Notes on the Management of
Spoiled Identity. Englewood Cliffs, NJ: Prentice Hall.
Goldberg, D. P. and V. F. Hillier. 1979. “Scaled Version
of the General Health Questionnaire.” Psychological
Medicine 9:139–45.
Halbesleben, J. R. B. and C. Rathert. 2008. “Linking Phy-
sician Burnout and Patient Outcomes: Exploring the
Dyadic Relationship between Physicians and Patients.”
Health Care Management Review 33:29–39.
Halter, M. J. 2002. “Stigma in Psychiatric Nursing.” Per-
spectives in Psychiatric Care 38:23–29.
Halter, M. J. 2008. “Perceived Characteristics of Psy-
chiatric Nurses: Stigma by Association.” Archives of
Psychiatric Nursing 22:20–26.
Haynes, C. E., T. D. Wall, R. I. Bolden, and C. Stride.
1999. “Measures of Perceived Work Characteristics
for Health Services Research: Test of a Measure-
ment Model and Normative Data.” British Journal of
Health Psychology 4:257–75.
Hem, M. H. and K. Heggen. 2003. “Being Professional
and Being Human: One Nurse’s Relationship with a
Psychiatric Patient.” Journal of Advanced Nursing
43:101–108.
Holland, T. P., A. Konick, W. Buffum, M. K. Smith, and
M. Petchers. 1981. “Institutional Structure and Resi-
dent Outcomes.” Journal of Health and Social Behav-
ior 22:433–44.
Hugo, M. 2001. “Mental Health Professionals’ Attitudes
towards People Who Have Experienced a Mental
Health Disorder.” Journal of Psychiatric and Mental
Health Nursing 8:419–25.
Huisman, M. 1999. “Imputation of Missing Item
Responses: Some Simple Techniques.” Pp. 91–119 in
Item Nonresponse: Occurrence, Causes, and Imputa-
tion of Missing Answers to Test Items, edited by M.
Huisman. Leiden, The Netherlands: DSWO Press,
Leiden University.
Jenkins, R. and P. Elliot. 2004. “Stressors, Burnout, and
Social Support: Nurses in Acute Mental Health Set-
tings.” Journal of Advanced Nursing 48:622–31.
Jorm, A. F., A. E. Korten, P. A. Jacomb, H. Christensen,
and S. Henderson. 1999. “Attitudes towards People
with a Mental Disorder: A Survey of the Australian
Public and Health Professionals.” Australian and
New Zealand Journal of Psychiatry 33:77–83.
Karasek, R. A. 1979. “Job Demands, Job Decision Lati-
tude, and Mental Strain—Implications for Job Rede-
sign.” Administrative Science Quarterly 24:285–308.
Kitson, Alison L. (2003). A comparative analysis of lay-
caring and professional (nursing) caring relationships.
International Journal of Nursing Studies 40:503–510.
Leiter, M. P., P. Harvie, and C. Frizzell. 1998. “The Cor-
respondence of Patient Satisfaction and Nurse Burn-
out.” Social Science & Medicine 47:1611–17.
Link, B. G. 1982. “Mental Patient Status, Work, and
Income—An Examination of the Effects of a Psychiat-
ric Label.” American Sociological Review 47:202–15.
Link, B. G. 1987. “Understanding Labeling Effects in
the Area of Mental Disorders: An Assessment of
the Effects of Expectations of Rejection.” American
Sociological Review 52:96–112.
Link, B. G., E. Struening, F. T. Cullen, P. E. Shrout, and
B. P. Dohrenwend. 1989. “A Modified Labeling
Theory Approach to Mental Disorders—An Empiri-
cal Assessment.” American Sociological Review
54:400–23.
Link, B. G., E. Struening, S. Neese-Todd, S. Asmussen,
and J. C. Phelan. 2001. “The Consequences of Stigma
for the Self-Esteem of People with Mental Illnesses.”
Psychiatric Services 52:1621–26.
Link, B. G., E. Struening, S. Neese-Todd, S. Asmussen,
and J. C. Phelan. 2002. “On Describing and Seeking
to Change the Experience of Stigma.” Psychiatric
Rehabilitation Skills 6:201–31.
Link, B. G., E. L. Struening, M. Rahav, J. C. Phelan, and
L. Nuttbrock. 1997. “On Stigma and Its Consequences:
at Universiteit Gent-Vakgroep on March 1, 2012hsb.sagepub.comDownloaded from