Content uploaded by Georg Schomerus
Author content
All content in this area was uploaded by Georg Schomerus on Oct 16, 2017
Content may be subject to copyright.
Please cite this article in press as: Schomerus, G., et al., Self-stigma in alcohol dependence: Consequences for drinking-refusal self-efficacy. Drug
Alcohol Depend. (2010), doi:10.1016/j.drugalcdep.2010.08.013
ARTICLE IN PRESS
G Model
DAD-3897; No. of Pages 6
Drug and Alcohol Dependence xxx (2010) xxx–xxx
Contents lists available at ScienceDirect
Drug and Alcohol Dependence
journal homepage: www.elsevier.com/locate/drugalcdep
Self-stigma in alcohol dependence: Consequences for drinking-refusal
self-efficacy
Georg Schomerusa,∗, Patrick W. Corriganb, Thomas Klauer a, Philipp Kuwerta,
Harald J. Freybergera, Michael Lucht a
aDepartment of Psychiatry, Ernst-Moritz-Arndt University Greifswald, Rostocker Chaussee 70, 18437 Stralsund, Germany
bIllinois Institute of Technology, Chicago, 3424 S. State Street, Chicago, IL 60616, USA
article info
Article history:
Received 28 June 2010
Received in revised form 30 August 2010
Accepted 30 August 2010
Available online xxx
Keywords:
Stigma
Discrimination
Substance abuse
Self-efficacy
Mental illness
abstract
Background: Public stigma and self-stigma are two facets of mental illness stigma. Self-stigma denotes
the internalization of negative public perceptions by persons with mental illness and has been shown
to decrease general self-efficacy. To date, self-stigma has not been examined in people suffering from
alcohol dependence, a particularly severely stigmatized mental disorder.
Methods: By adopting the Self-Stigma in Mental Illness Scale (SSMI), we developed the Self-Stigma in Alco-
hol Dependence Scale (SSAD). The scale is based on a focus-group derived list of 16 negative stereotypes
about alcohol dependent persons. It consists of four 16-item subscales measuring four hypothetical stages
of self-stigma, stereotype awareness (aware), stereotype agreement (agree), self-concurrence (apply),
and self-esteem decrement (harm). We employed the SSAD in a cross-sectional study of 153 patients
hospitalized for alcohol detoxification to examine its reliability and validity.
Results: The four stages of self-stigma could be reliably measured with the SSAD (Cronbach’s alpha,
0.86–0.93). Each step in the process of self-stigmatization was most closely associated with its preceding
step. Other significantly related independent variables in multiple regression analyses included desire
for social distance (associated with agree), duration of drinking problems (associated with apply) and
depressive symptoms (associated with apply and harm). Both apply and harm were significantly related
to reduced drinking-refusal self-efficacy in analyses controlling for depressive symptoms and variables
related to duration and severity of the drinking problem.
Discussion: The SSAD showed good validity and reliability measuring the stages of self-stigma in this
group. Self-stigma appears to be associated with lower drinking-refusal self-efficacy.
© 2010 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
Alcohol dependence is one of the most common and most severe
mental disorders. The World Health Organization estimates 76.3
million people worldwide suffer from alcohol use disorders (WHO,
2004), about 4% of all deaths and 5% of all disability adjusted
life years lost can be attributed to alcohol (Rehm et al., 2009).
Aggravating the physical, psychological and social harm of alcohol
dependence, those affected have to endure considerable stigmati-
zation (Fortney et al., 2004; Room, 2005). Both on a societal level
and within the field of mental disorders, alcohol dependent people
constitute one of the most severely stigmatized groups (Schomerus
et al., 2010).
Stigma burdens people with alcohol dependence in different
ways: Public stigma denotes stereotypes, prejudice and discrimi-
∗Corresponding author. Tel.: +49 3831 452109; fax: +49 3831 452105.
E-mail address: georg.schomerus@uni-greifswald.de (G. Schomerus).
nation affecting patients from the outside, from their social and
societal environment (Corrigan and Watson, 2002; Rüsch et al.,
2005). Compared to other mental disorders, rejection of alco-
hol dependent people is particularly strong (Angermeyer and
Matschinger, 1997; Link et al., 1999; Schomerus et al., 2006;
Baldwin et al., 2010) and negative stereotypes like being dan-
gerous or being unpredictable are readily associated with them
(Crisp et al., 2005). However, the process of stigmatization does
not stop outside those stigmatized. Self-stigma denotes a cog-
nitive and emotional process within the stigmatized subject,
taking place when a person internalizes prevalent negative views
about “someone like him/her” and applies these views to him-
self/herself. In people with severe mental illness, self-stigma has
been shown to be associated with depressive symptoms, low
self-esteem and low self-efficacy (Corrigan et al., 2006; Link et
al., 2001; Ritsher and Phelan, 2004), and to be associated with
increased need of inpatient treatment (Rüsch et al., 2009). Stud-
ies among college students and in the general population suggest
that self-stigma hinders professional help-seeking in case of mental
0376-8716/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.drugalcdep.2010.08.013
Please cite this article in press as: Schomerus, G., et al., Self-stigma in alcohol dependence: Consequences for drinking-refusal self-efficacy. Drug
Alcohol Depend. (2010), doi:10.1016/j.drugalcdep.2010.08.013
ARTICLE IN PRESS
G Model
DAD-3897; No. of Pages 6
2G. Schomerus et al. / Drug and Alcohol Dependence xxx (2010) xxx–xxx
illness (Schomerus, 2009; Schomerus et al., 2009; Vogel et al.,
2006).
The process of self-stigmatization has so far not been empiri-
cally studied in alcoholism, although it may have particularly grave
consequences. Recovering from alcohol dependence requires pro-
found behavioral changes, and here, a stigma-related decrease in
self-efficacy is likely to cause considerable harm. Indeed, the con-
cept of self-efficacy as introduced by Bandura (1977) has early been
adopted to alcoholism (Annis, 1986). Meanwhile, several studies
have demonstrated that higher alcohol-related self-efficacy pre-
dicts better outcome in alcohol dependent persons (Greenfield et
al., 2000; Ilgen et al., 2005; Sitharthan and Kavanagh, 1991; Zingg
et al., 2009). However, self-efficacy has so far not been considered
in the context of self-stigma in alcohol dependence.
In this study, our aim is to explore the process of self-
stigmatization in alcoholism. For this purpose, we use a progressive
model of self-stigma proposed by Corrigan and co-workers
(Corrigan et al., 2006; Corrigan et al., in preparation). It consists of
four stages that represent successive, interrelated steps. Its starting
point is being aware of negative stereotypes other people endorse
(stereotype awareness or aware). The next step is to personally
agree with these stereotypes (stereotype agreement, agree), then
to apply these stereotypes to oneself (self-concurrence, apply),
and finally to suffer from low self-esteem due to application of
these stereotypes (harm). Decreased self-esteem, which has also
been described as a consequence of self-stigma (Link et al., 2001;
Ritsher and Phelan, 2004), is thus conceptualized as the final stage
of self-stigma in this model. The four stages of self-stigmatization
have been measured in patients with various mental disorders (but
not substance related disorders) with an according scale, the Self-
Stigma of Mental Illness Scale (SSMI, Corrigan et al., 2006), which
has shown good reliability and validity (Corrigan et al., 2006; Rüsch
et al., 2006).
We report on the adoption of the SSMI to alcohol dependent peo-
ple, creating the Self-Stigma in Alcohol Dependence Scale (SSAD),
and on a study among alcohol dependent patients undergoing
detoxification to examine its reliability and validity. We exam-
ined a sample of hospitalized patients for this purpose because we
assumed that self-stigma would be particularly relevant in those
severely affected and in those with an established diagnosis of alco-
hol dependence. In particular, we were interested whether results
obtained with the SSAD are consistent with Corrigan’s progres-
sive model of self-stigma, how self-stigma relates to duration and
severity of the drinking problem, and whether self-stigma is related
to drinking-refusal self-efficacy in people hospitalized for alcohol
dependence.
2. Methods
2.1. Development of the Self-Stigma in Alcohol Dependence Scale
We adopted the Self-Stigma of Mental Illness Scale (SSMI) (Corrigan et al., 2006)
to be used in alcohol dependent persons. To elicit the four steps of self-stigma (aware,
agree,apply, and harm), the original scale uses a list of 10 negative stereotypes
that patients rate four times (in different order) following different introductory
clauses. Stereotype awareness is rated after the introductory clause “I think the
public believes most persons with mental illness are...”, stereotype agreement
following “I believe most persons with mental illness are...”, self-concurrence or
application after “because I have a mental illness, I am...”, and self-esteem decre-
ment rated following the phrase “I currently respect myself less because I am...”.
Answers are given on 9-point Likert scales with 9 =“I strongly agree”.
Since stereotypes for alcohol dependence potentially differ from those con-
nected to “mental illness”, we conducted a focus-group study with three focus
groups consisting of persons with alcohol dependence (n= 12), staff (n= 6) and lay
persons including family members of those affected (n= 8). Their responses gen-
erated lists of common stereotypes associated with alcohol dependent persons.
Altogether, a list of 34 stereotypes was compiled and, by eliminating redundancies,
subsequently reduced to 16 stereotypes: unreliable, emotionally unstable, violent,
living on other people’s expenses, self-pitying, lazy, resolving conflicts only with
alcohol, weak-willed, unable to ever get away from alcohol, unable to keep a regu-
lar job, to blame for their problems, not to be trusted, disgusting, dirty and unkempt,
below average intelligence, unpredictable. The last 7 stereotypes of this list are iden-
tical to the original SSMI. From these 16 stereotypes, we generated the Self-Stigma
of Alcohol Dependence (SSAD) scale comprising 4 ×16 items in a format similar to
the SSMI. Examples for items are: “I think the public believes most people with alco-
hol problems are violent” (aware), “I think most people with alcohol problems are
violent” (agree), “Because I have alcohol problems, I am violent” (apply), “I currently
respect myself less because I am violent” (harm). Items were rated on 5-point Lik-
ert scales with 5 =“I strongly agree”. Four scales were determined with sum scores
ranging from 16 to 80. We chose 5-point instead of 9-point Likert scales to reduce
the complexity of the items. We pilot tested the instrument in a convenience sample
of 14 persons diagnosed with alcohol dependence. As a result, we chose the expres-
sion “alcohol problems” for the questionnaire, because the medically correct term
“alcohol dependence” was not readily accepted by some participants.
2.2. Validation study
2.2.1. Sample. 153 persons with a diagnosis of alcohol dependence according to
ICD-10 were recruited between September 2009 and March 2010 in a psychiatric
alcohol detoxification unit in Mecklenburg – Western Pomerania, Germany. The unit
is the single referral centre for alcohol detoxification for a rural region of 110,000
inhabitants. Treatment is scheduled to last 2 weeks and includes medication, group
therapy, physical activity, introduction to self-help groups and individual therapy
if indicated. We invited all alcohol dependent patients having completed the acute
phase of detoxification and being able to give informed consent to participate in the
study, except those not cognitively or physically capable to answer a questionnaire
and persons with insufficient language or reading skills. We thus approached 223
persons, 153 of which (69%) agreed to participate by giving written informed con-
sent and filling out the study-questionnaire. The study was approved by the local
ethics committee of the Ernst-Moritz-Arndt University Greifswald. We compared
participants and non-participants by gender (chi-square test, p= 0.69), age (t-test,
p= 0.79) and number of previous hospital detoxifications (t-test, p=0.47) and found
no significant differences. Our final sample included 121 (79%) male participants.
Mean age was 46.8 years (SD 8.8), 54 participants (35.3%) lived in a stable rela-
tionship, 35 (22.9%) were currently employed (including supported employment),
136 (88.9%) had completed at least 8 years of schooling. The mean self-reported
duration of the drinking problem was 13.1 years (SD 8.8). 36 participants (11.5%)
underwent their first detoxification treatment, 81 (52.9%) had a history of 2–10
hospital detoxifications, 36 (23.5%) of more than 10 detoxifications.
2.2.2. Instruments. In addition to the newly developed SSAD, we employed mea-
sures to elicit desire for social distance, present symptoms of depression and anxiety,
drinking-refusal self-efficacy, and variables related to duration and severity of the
drinking problem.
2.2.3. Social distance. Social distance is an established measure of individual dis-
criminating attitudes towards members of a minority group (Jorm and Oh, 2009).
We adopted a scale used in previous research to measure social distance towards
people with various mental disorders (Angermeyer and Matschinger, 1997; Link et
al., 1987). The scale asks whether respondents are willing to engage in several forms
of every day contact with “someone having an alcohol problem”, for example: “If
you had a room to rent, would you accept someone having an alcohol problem
as a tenant?” Answers were again given on 5-point Likert scales anchored with
1 = “definitely” and 5 = “definitely not”, yielding a sum score from 7 to 35 points,
high values indicating high desire for social distance. We expected social distance
to be associated with agreement to negative stereotypes.
2.2.4. Depressive symptoms and anxiety. To measure depressive and anxiety symp-
toms, which we considered potentially relevant to the process of self-stigmatization
in patients just having completed detoxification, we used the 18-item version of the
Brief Symptom Inventory (BSI-18) (Derogatis, 1993, 2001; German version: Franke,
2000). The scale provides brief descriptions of 18 psychological symptoms/problems
that are rated on a 5-point Likert-scale (“not at all” to “extremely”) in terms of how
much each has distressed or bothered the respondent during the past 7 days. Scor-
ing the BSI-18 involves summing ratings across subsets of six items each to create
measures of the symptoms depression and anxiety (we did not employ the third
subscale, somatization, in the present study). The BSI-18 has been used previously
in alcohol dependent persons (Buri et al., 2007) and the validity of its scales has
been proven across various patient populations (e.g. Wang et al., 2010). In our sam-
ple, Cronbach’s alphas of the entire scale (0.94) and scales for depression (0.87) and
anxiety (0.89) were high, and the two subscales correlated significantly and highly
with each other (Pearson’s correlation coefficient 0.75).
2.2.5. Drinking-refusal self-efficacy. The “Kurzfragebogen zur Abstinenzzuversicht”
(“Short Questionnaire on Abstinence Confidence”, KAZ-35) is a validated German
35-item instrument to measure drinking-refusal self-efficacy (Körkel and Schindler,
1996). The KAZ-35 is an adoption from the Situational Confidence Questionnaire
(SCQ-100) (Annis, 1986) and parallels one of its short forms, the SCQ-35 (Sandahl et
al., 1990). Participants are asked to rate their confidence to reject a drink in various
Please cite this article in press as: Schomerus, G., et al., Self-stigma in alcohol dependence: Consequences for drinking-refusal self-efficacy. Drug
Alcohol Depend. (2010), doi:10.1016/j.drugalcdep.2010.08.013
ARTICLE IN PRESS
G Model
DAD-3897; No. of Pages 6
G. Schomerus et al. / Drug and Alcohol Dependence xxx (2010) xxx–xxx 3
Table 1
Mean, standard deviation (SD) and reliability coefficients of the scales employed.
Mean SD Cronbach’s alpha
SSAD aware 46.5 15.0 0.93
SSAD agree 40.1 13.3 0.93
SSAD apply 30.6 9.6 0.86
SSAD harm 28.9 9.7 0.87
Social distance 23.7 6.6 0.87
Drinking-refusal self-efficacy (KAZ-35) 70.1 24.5 0.98
Severity of alcohol dependence (SESA) 50.1 21.7 0.95
BSI-18 depression 7.7 6.1 0.87
BSI-18 anxiety 6.5 5.7 0.89
SSAD: Self-Stigma in Alcohol Dependence Scale.
situations from 0 (not confident at all) to 100 (totally confident). To assess drinking-
refusal self-efficacy, we calculated a mean score across all items ranging from 0 to
100.
2.2.6. Severity of alcohol dependence. We used the severity scale of alcohol depen-
dence (SESA, John et al., 2003) to measure severity of dependence. The SESA is a
valid self-report instrument consisting of 28 items and seven scales covering the
diagnostic criteria of the alcohol dependence syndrome. Four scales (narrowing
of drinking repertoire, somatic withdrawal symptoms, drinking to avoid with-
drawal, and psychological withdrawal symptoms) use a 5-point Likert-scale from
0 = never to 4 = daily, three scales (increase of tolerance, extreme increase of toler-
ance, decrease of tolerance) use a dichotomous response scale (1 = yes, 0 = no). A
weighted sum score can be calculated from the seven scales, and norms for persons
hospitalized for alcohol detoxification exist (John et al., 2001). According to these
norms, 14% of our sample suffered from mild, 56% from moderate, and 30% from
severe alcohol dependence. We used the SESA sum score for our study as a measure
of global severity of alcohol dependence (range 0–100).
3. Results
Table 1 shows mean scores, standard deviations, and reliabil-
ity coefficients of the scales used in this study. All scales showed
good internal consistency. The subscales of the newly developed
SSAD did also show excellent reliability in the small sub-sample of
women (n= 32) with Cronbach’s alpha values of 0.95 (aware), 0.96
(agree), 0.92 (apply) and 0.91 (harm).
Corrigan and co-workers have brought forward the hypoth-
esis that self-stigma forms progressively, i.e. that each stage
is precondition for the following stage. Accordingly, scores are
expected to be highest for aware, to diminish progressively for
each stage thereafter, and to be lowest for harm (Corrigan et
al., in preparation). In order to judge whether SSAD scores fol-
low this hypothesized pattern we examined the mean values of
the subscale scores (Table 1). Decreasing from aware to harm,
scores were consistent with a progressive model of self-stigma.
Repeated measurement ANOVA proved that differences between
subscale scores were statistically significant (F= 123.7, p< 0.001).
Subsequent t-tests for paired samples confirmed the significance
of the observed differences between proximate subscales (aware-
agree,t= 7.04, p< 0.001; agree-apply,t= 9.76, p< 0.001; apply-harm,
t= 3.27, p< 0.01).
Next we examined the correlation matrix of the four scales.
Assuming a progressive model of self-stigma, correlations of prox-
imate scales were expected to be higher than correlations of more
distal scales. The top half of Table 2 shows that each scale corre-
lates most highly with its immediate predecessor scale; correlation
coefficients are lower between more distant scales, being lowest
between the most distant scales aware and harm. The correlation
pattern of the subscales was thus consistent with their hypothe-
sized relation. Overall, effect sizes were high for the correlations
aware-agree and apply-harm (p< 0.001), moderate for agree-apply
and agree-harm (p< 0.001), and small for aware-apply (p< 0.001)
and aware-harm (p< 0.01).
The bottom half of Table 2 shows correlations of the four self-
stigma scales with conceptually different, but potentially related
constructs. Desire for social distance is an established measure to
elicit personal rejection of members of a minority group (Jorm and
Oh, 2009). In terms of convergent validity, desire for social distance
towards someone with alcohol problems was expected to be associ-
ated with agreeing with negative stereotypes about such a person.
Table 2 shows that social distance did indeed correlate strongest
and most significantly with stereotype agreement (moderate effect
size, p< 0.001), while correlations with other subscales were small
and less significant.
Studies using the original SSMI found apply and harm inversely
related to general self-efficacy (Corrigan et al., 2006). For our
alcohol-specific measure, we were interested in the relation of
the four subscales to drinking-related self-efficacy. As a potential
negative consequence of self-concurrence of negative stereotypes
(apply) and subsequent self-esteem decrease (harm), we expected
drinking-refusal self-efficacy to correlate inversely with these
two subscales. Table 2 shows that drinking-refusal self-efficacy
is indeed significantly related to the inverse of apply and (even
stronger) of harm (moderate effect size, p< 0.001), indicating con-
current validity of the SSAD.
Finally, Table 2 shows the relation of depressive symptoms and
anxiety to the four subscales. Along the lines of studies using the
SSMI, we expected depressive symptoms to be associated with
apply and harm.Table 2 shows that both symptom scales are signif-
icantly correlated to apply and harm, with overall moderate effect
sizes and largest coefficients for depressive symptoms, followed by
anxiety.
Beyond this bivariate correlational analysis, we explored two
sets of multiple regression models (Table 3): first, we looked at
potentially related explanatory variables for the different stages of
self-stigmatization in order to examine to what extent each step in
the process of self-stigmatization is related to its preceding step,
and to what extent other potentially relevant variables contribute
to the formation of self-stigma. Second, we examined the relation
of the final stages of self-stigma, apply and harm, to dinking-refusal
self-efficacy when being controlled for depressive symptoms and
variables depicting severity and duration of the drinking problem.
To establish whether the distribution of the data was adequate for
multiple regression analyses, we conducted Kolmogorov–Smirnov
tests for normal distribution of residuals for every dependent vari-
able in our regression models, which were all non-significant.
The first set of regression models is summarized in the left half
of Table 3. Sum scores of the four subscales of the SSAD (aware,
agree,apply and harm) serve as dependent variables. Except for
aware, which is at the beginning of the hypothetical process of
self-stigma, in each model the preceding stage of self-stigma is
entered as independent variable. Additional independent variables
include social distance, depressive symptoms, and variables related
to the drinking problem: severity of alcohol dependence, years of
Table 2
Pearson’s product moment correlations for the four self-stigma scales and related
constructs.
SSAD aware SSAD agree SSAD apply SSAD harm
SSAD aware 1.00 0.69*** 0.32*** 0.26**
SSAD agree 1.00 0.51*** 0.40***
SSAD apply 1.00 0.83***
SSAD harm 1.00
Social distance 0.27** 0.47*** 0.24** 0.19*
Drinking-refusal
self-efficacy (KAZ-35)
−0.15 −0.11 −0.34*** −0.42***
BSI-18 depression 0.04 0.05 0.33*** 0.38***
BSI-18 anxiety 0.00 0.08 0.28*** 0.30***
SSAD: Self-Stigma in Alcohol Dependence Scale.
*p< 0.05.
** p< 0.01.
*** p< 0.001.
Please cite this article in press as: Schomerus, G., et al., Self-stigma in alcohol dependence: Consequences for drinking-refusal self-efficacy. Drug
Alcohol Depend. (2010), doi:10.1016/j.drugalcdep.2010.08.013
ARTICLE IN PRESS
G Model
DAD-3897; No. of Pages 6
4G. Schomerus et al. / Drug and Alcohol Dependence xxx (2010) xxx–xxx
Table 3
Linear regression models with SSAD subscales and drinking-refusal self-efficacy as dependent variables. Models are controlled for age and gender. Standardized regression
coefficients (beta), listwise deletion of cases with missing values, n= 125–132.
SSAD subscales Drinking-refusal self-efficacy (KAZ-35)
SSAD aware SSAD agree SSAD apply SSAD harm Model 1 Model 2 Model 3
Beta pBeta pBeta pBeta pBeta pBeta pBeta p
SSAD harm −0.340 0.000
SSAD apply 0.796 0.000 −0.240 0.009
SSAD agree 0.421 0.000
SSAD aware 0.600 0.000
Social distance 0.132 n.s. 0.273 0.000 0.036 n.s. 0.004 n.s.
Depressive symptoms (BSI-18) 0.021 n.s. 0.097 n.s. 0.265 0.001 0.156 0.007 −0.250 0.005 −0.177 n.s. −0.111 n.s.
Years of drinking problem 0.155 n.s. 0.003 n.s. 0.206 0.011 −0.039 n.s. −0.306 0.001 −0.240 0.009 −0.262 0.003
Severity of alcohol dependence (SESA) −0.120 n.s. 0.043 n.s. 0.137 n.s. −0.076 n.s. 0.091 n.s. 0.132 n.s. 0.105 n.s.
Number of hospital detoxifications 0.035 n.s. −0.143 n.s. −0.090 n.s. 0.052 n.s. −0.093 n.s. −0.134 n.s. −0.116 n.s.
Adj. R2n.s. 0.48 0.34 0.69 0.15 0.18 0.24
SSAD: Self-Stigma in Alcohol Dependence Scale.
drinking problem, and number of hospital detoxifications. Models
are controlled for age and gender. While neither social distance,
nor depression, nor drinking-related variables were significantly
associated with awareness of negative public stereotypes about
alcohol dependent persons, agreeing to these stereotypes was asso-
ciated with personal desire for social distance from persons with
alcohol problems, and applying these stereotypes to oneself was
associated with more years of drinking problems and depressive
symptoms. Consistent with the bivariate correlations reported in
Table 2, suffering from depressive symptoms was also related to
increased harm through self-esteem decrement.1With regard to
the proposed stages of self-stigma, our models were consistent with
the hypothesized immediate relationship between these stages.
Being aware of stereotypes was significantly associated with agree-
ing to these stereotypes, which in turn was associated with applying
them to oneself, which was strongly associated with stereotype-
related low self-esteem (harm). The model for aware remained
insignificant, while the other models explained 48% of the variance
of agree, 34% of apply, and 69% of harm.
In a second set of linear regression analyses (right half of
Table 3), we examine the potential influence of the two final stages
of self-stigma, apply and harm, on abstinence self-efficacy. Our
measure of drinking-refusal self-efficacy, the KAZ-35 mean score,
serves as dependent variable. In the first model, only depressive
symptoms, drinking-related variables, age and gender are entered
as independent variables. This model explains 15% of the vari-
ance of drinking-refusal self-efficacy, depression and duration of
drinking problem being significantly related to drinking-refusal
self-efficacy. In model 2 apply is added, which increases the amount
of explained variance by 3%. Both years of drinking problem and
apply contribute equally to drinking-refusal self-efficacy, while
depression is no longer significantly related. In model 3, instead
of apply,harm is entered as independent variable. Because of col-
inearity between both variables (Pearson’s correlation coefficient,
>0.8), we did not enter apply and harm simultaneously into the
same model. This would have resulted in a considerable loss of
power and demanded a larger sample size to detect independent
effects of both variables. The model including harm explains 24%
of the variance overall, harm thus accounting for an increase in
explained variance by 9% compared to model 1. Across all mod-
els, harm proved to be most strongly related to drinking-refusal
self-efficacy, followed by years of drinking problem.
1Similar models using symptoms of anxiety as independent variables instead of
depression resulted in similar, although smaller and less significant effects.
4. Discussion
Findings of this study are consistent with a stepwise process of
self-stigmatization in persons with alcohol dependence similar to
people with other severe mental illness, with stereotype aware-
ness (aware) at the beginning, followed by stereotype agreement
(agree) and self-concurrence (apply), and resulting in self-esteem
decrement (harm). Each step in the process of self-stigmatization
was most strongly related to its immediately preceding step. Other
variables independently related to self-stigmatization included
personal discriminatory attitudes towards alcohol dependent
patients (associated with agree), years of drinking problem (asso-
ciated with apply) and depressive symptoms (associated with
apply and harm). The final steps of self-stigmatization, apply and
harm, were significantly associated with lower drinking-refusal
self-efficacy, even when controlling for depressive symptoms and
variables related to the severity and duration of the drinking prob-
lem.
When examining self-stigma, some basic considerations about
causes and consequences of stigma are necessary. Since the pro-
cess of self-stigmatization unfolds internally, those affected could
wrongly be held responsible for it. It could be argued, for example,
that the personal attitudes of the affected persons, their depres-
sive symptoms, and the alcohol dependence itself could contribute
to stronger awareness and internalization of negative stereotypes.
However, the results of our regression analyses (Table 3) show
that personal characteristics are of minor importance to the for-
mation of self-stigma, and that the first step in the process of
self-stigmatization, stereotype awareness, is entirely unrelated to
personal attitudes (desire for social distance), depressive symp-
toms, and drinking-related variables. On the other hand, our results
are consistent with the hypothesis that self-stigma is a result of
public stigma: Although not examined in this study, it is well estab-
lished that negative public stereotypes about alcohol dependent
persons are very common, more than 70% of the general public
for example consider them dangerous or unpredictable (Link et
al., 1999; Crisp et al., 2005). The most likely cause for the first
step of self-stigma, stereotype awareness, is thus simply the high
prevalence of these stereotypes. It is a particular merit of Corrigan’s
progressive model of self-stigma that it captures both the origin of
self-stigma, the perception of negative public stereotypes, and the
process of their internalization. Our results corroborate this step-
wise conceptualization of self-stigma. In our analyses, each stage of
self-stigma was strongly related to its succeeding stage, and other
variables related to illness severity or mental distress were, if sig-
nificant, of considerably smaller influence. Our results thus are in
accordance with the view that self-stigma is primarily a conse-
quence of public stigma and suggest that it has to be regarded as a
Please cite this article in press as: Schomerus, G., et al., Self-stigma in alcohol dependence: Consequences for drinking-refusal self-efficacy. Drug
Alcohol Depend. (2010), doi:10.1016/j.drugalcdep.2010.08.013
ARTICLE IN PRESS
G Model
DAD-3897; No. of Pages 6
G. Schomerus et al. / Drug and Alcohol Dependence xxx (2010) xxx–xxx 5
particularly subtle, but relevant form of discrimination of alcohol
dependent persons.
Drinking-refusal self-efficacy has long been recognized as
important variable affecting the ability to change drinking behavior
(Annis, 1986; Greenfield et al., 2000). So far, very few studies
have examined personal predictors of self-efficacy in alcoholism,
finding higher drinking-refusal self-efficacy to be associated with
higher education, less severe substance related problems, female
gender (Ilgen et al., 2007; McKellar et al., 2008) as well as with
greater engagement in treatment or self-help-groups (Connors et
al., 2001; Ilgen et al., 2007). Our study brings another determi-
nant of drinking-related self-efficacy into focus. The concept of
self-stigmatization may usefully add to the understanding of the
formation of low self-efficacy in alcohol dependent persons. Our
results raise the question whether, to a certain extent, the stigma
of alcoholism could be a self-fulfilling prophecy: Belief in negative
stereotypes like being weak-willed or unable to ever quit drinking
could contribute to these very outcomes.
The stigma of alcoholism has recently been discussed from a
perspective of functionality: stigmatizing a deviant behavior like
heavy drinking could serve as a means to enforce social norms,
to clarify the boundaries of accepted behavior, and thus ultimately
help keeping those in danger within these boundaries (Phelan et al.,
2008; Schomerus et al., 2010). Seen from this perspective, stigma-
tization of alcoholism could be a rational, successful strategy to
improve public health – as has been controversially discussed for
other health-related behaviors like smoking (Bayer, 2008; Burris,
2008). Our findings, however, endorse the case against such a moral
strategy to combat alcohol addiction at least for our sample of
severely affected, hospitalized persons. They suggest that stigma is
a rather dysfunctional way to promote abstinence, since, by reduc-
ing abstinence self-efficacy, stigma may hinder people to achieve
sobriety.
The results of this study need to be discussed in the context
of its limitations. Although we took care to invite all eligible per-
sons to our study and did not find significant differences between
participants and non-participants with regard to age, gender and
number of previous detoxifications, our sample from a detoxifi-
cation unit cannot be regarded as representative for all persons
with alcohol dependence. On the one hand, persons being hospi-
talized for detoxification likely represent rather severe cases, on the
other hand, people agreeing to detoxification have usually achieved
some problem recognition and motivation to change. Most per-
sons suffering from alcohol dependence are not treated (Wittchen
and Jacobi, 2001), and those seeking help do so after considerable
delay (Wang et al., 2007). The influence of self-stigma (or avoid-
ance of self-stigma through denial of the problem) in the population
of untreated alcohol dependent persons or in those receiving out-
patient treatment is certainly of great interest, but was not covered
by this study. A second limitation is the cross-sectional nature of our
investigation. Clearly, a prospective examination of the prognostic
influence of self-stigma on future drinking behavior is necessary to
further determine the clinical relevance of this construct. Third, we
did not characterize our sample by means of a standardized diag-
nostic interview; hence the influence of psychiatric co-morbidities
on alcohol-related self-stigma could not be determined. We did,
however, control our analyses for present depressive and anxiety
symptoms and could show that self-stigma is related to, but only
partially explained by these. Finally, we did relate self-stigma to
alcohol-related self-efficacy, but did not examine other constructs
indicative of the negative consequences of self-stigma like shame
(Luoma et al., 2007) – this also remains a desideratum for future
studies.
In summary, self-stigma appears to be a relevant explanatory
variable for drinking-refusal self-efficacy and may thus probably
have negative effects on outcome in alcohol dependence ther-
apy. The SSAD showed good validity and reliability measuring the
stages of self-stigma in alcohol dependent patients. Prospective
studies examining the influence of self-stigma on drinking behav-
ior are warranted and should also include untreated persons with
alcohol dependence and persons receiving out-patient treatment.
Although used for research purposes in this study, the SSAD might
also prove valuable when used in individuals to examine severity
of self-stigma, or to establish the concept of self-stigma and reflect
on personal attitudes in psychotherapy. While means and standard
deviations reported in this paper might serve as an orientation to
judge individual severity of self-stigma, larger studies with more
diverse samples are needed to provide norms for this instrument.
Role of funding source
Nothing declared.
Contributors
Authors Schomerus, Corrigan, Lucht and Freyberger designed
the study. Authors Klauer and Schomerus conducted the statistical
analyses, authors Schomerus, Kuwert and Lucht conducted focus
groups for instrument development. Author Schomerus wrote
the first draft of the manuscript, and author Corrigan supervised
manuscript preparation. All authors contributed to interpreting the
results and have approved the final version of the manuscript.
Conflict of interest
None declared.
References
Angermeyer, M.C., Matschinger, H., 1997. Social distance towards the mentally ill:
results of representative surveys in the Federal Republic of Germany. Psychol.
Med. 27, 131–141.
Annis, H.M., 1986. A Relapse Prevention Model for the Treatment of Alcoholics.
Pergamon Press, New York.
Baldwin, M.L., Marcus, S.C., De Simone, J., 2010. Job loss discrimination and former
substance use disorders. Drug Alcohol Depend. 110, 1–7.
Bandura, A., 1977. Self-efficacy—toward a unifying theory of behavioral change.
Psychol. Rev. 84, 191–215.
Bayer, R., 2008. Stigma and the ethics of public health: not can we but should we.
Soc. Sci. Med. 67, 463–472.
Buri, C., Moggi, F., Giovanoli, A., Strik, W., 2007. Prescription procedures in medica-
tion for relapse prevention after inpatient treatment for alcohol use disorders
in Switzerland. Alcohol Alcohol. 42, 333–339.
Burris, S., 2008. Stigma, ethics and policy: a commentary on Bayer’s “Stigma and the
ethics of public health: Not can we but should we”. Soc. Sci. Med. 67, 473–475.
Connors, G.J., Tonigan, J.S., Miller, W.R., 2001. A longitudinal model of intake symp-
tomatology, AA participation and outcome: retrospective study of the project
MATCH outpatient and aftercare samples. J. Stud. Alcohol 62, 817–825.
Corrigan, P.W., Rafacz, J., Wilkniss, S., Rüsch, N., in preparation. Examining a pro-
gressive model of self-stigma and its impact on people with serious mental
illness.
Corrigan, P.W., Watson, A.C., 2002. Understanding the impact of stigma on people
with mental illness. World Psychiatry 1, 16–20.
Corrigan, P.W., Watson, A.C., Barr, L., 2006. The self-stigma of mental illness: impli-
cations for self-esteem and self-efficacy. J. Soc. Clin. Psychol. 25, 875–884.
Crisp, A.H., Gelder, M.G., Goddard, E., Meltzer, H.I., 2005. Stigmatization of people
with mental illnesses: a follow-up study within the Changing Minds campaign
of the Royal College of Psychiatrists. World Psychiatry 4, 106–113.
Derogatis, L.R., 1993. Brief Symptom Inventory. National Computer Systems, Min-
neapolis, MN.
Derogatis, L.R., 2001. Brief Symptom Inventory 18: Administration, Scoring, and
Procedures Manual. NCS Pearson Inc, Minneapolis, MN.
Fortney, J., Mukherjee, S., Curran, G., Fortney, S., Han, X.T., Booth, B.M., 2004. Factors
associated with perceived stigma for alcohol use and treatment among at-risk
drinkers. J. Behav. Health Serv. Res. 31, 418–429.
Franke, G.H., 2000. BSI – Brief Symptom Inventory von L. R. Derogatis – Deutsches
Manual. Beltz, Göttingen.
Greenfield, S., Hufford, M., Vagge, L., Muenz, L., Costello, M., Weiss, R., 2000. The
relationship of self-efficacy expectancies to relapse among alcohol dependent
men and women: a prospective study. J. Stud. Alcohol 61, 345–351.
Ilgen, M., McKellar, J., Moos, R., 2007. Personal and treatment-related predictors of
abstinence self-efficacy. J. Stud. Alcohol Drugs 68, 126–132.
Please cite this article in press as: Schomerus, G., et al., Self-stigma in alcohol dependence: Consequences for drinking-refusal self-efficacy. Drug
Alcohol Depend. (2010), doi:10.1016/j.drugalcdep.2010.08.013
ARTICLE IN PRESS
G Model
DAD-3897; No. of Pages 6
6G. Schomerus et al. / Drug and Alcohol Dependence xxx (2010) xxx–xxx
Ilgen, M., McKellar, J., Tiet, Q., 2005. Abstinence self-efficacy and abstinence 1 year
after substance use disorder treatment. J. Consult. Clin. Psychol. 73, 1175–1180.
John, U., Hapke, U., Rumpf, H.J., 2001. SESA Skala zur Erfassung der Schwere der
Alkoholabhängigkeit. Manual. Hogrefe, Göttingen.
John, U., Hapke, U., Rumpf, H.J., 2003. A new measure of the alcohol dependence
syndrome: the severity scale of alcohol dependence. Eur. Addict. Res. 9, 87–93.
Jorm, A.F., Oh, E., 2009. Desire for social distance from people with mental disorders.
Aust. N. Z. J. Psychiatry 43, 183–200.
Körkel, J., Schindler, C., 1996. Der Kurzfragebogen zur Abstinenzzuversicht
(KAZ-35)—Ein Instrument zur Erfassung der abstinenzorientienen Kompetenz-
zuversicht Alkoholabhaengiger. Sucht 42, 156–166.
Link, B.G., Cullen, F.T., Frank, J., Wozniak, J.F., 1987. The social rejection of former
mental patients—understanding why labels matter. Am. J. Sociol. 92, 1461–1500.
Link, B.G., Phelan, J.C., Bresnahan, M., Stueve, A., Pescosolido, B.A., 1999. Public con-
ceptions of mental illness: labels, causes, dangerousness, and social distance.
Am. J. Public Health 89, 1328–1333.
Link, B.G., Struening, E.L., Neese-Todd, S., Asmussen, S., Phelan, J.C., 2001. The conse-
quences of stigma for the self-esteem of people with mental illnesses. Psychiatr.
Serv. 52, 1621–1626.
Luoma, J.B., Twohig, M.P., Waltz, T., Hayes, S.C., Roget, N., Padilla, M., Fisher, G.,
2007. An investigation of stigma in individuals receiving treatment for substance
abuse. Addict. Behav. 32, 1331–1346.
McKellar, J., Ilgen, M., Moos, B.S., Moos, R., 2008. Predictors of changes in alcohol-
related self-efficacy over 16 years. J. Subst. Abuse Treat. 35, 148–155.
Phelan, J.C., Link, B.G., Dovidio, J.F., 2008. Stigma and prejudice: one animal or two?
Soc. Sci. Med. 67, 358–367.
Rehm, J., Mathers, C., Popova, S., Thavorncharoensap, M., Teerawattananon, Y., Patra,
J., 2009. Global burden of disease and injury and economic cost attributable to
alcohol use and alcohol-use disorders. Lancet 373, 2223–2233.
Ritsher, J.B., Phelan, J.C., 2004. Internalized stigma predicts erosion of morale among
psychiatric outpatients. Psychiatry Res. 129, 257–265.
Room, R., 2005. Stigma, social inequality and alcohol and drug use. Drug Alcohol
Rev. 24, 143–155.
Rüsch, N., Angermeyer, M.C., Corrigan, P.W., 2005. Mental illness stigma: concepts,
consequences, and initiatives to reduce stigma. Eur. Psychiatry 20, 529–539.
Rüsch, N., Corrigan, P.W., Wassel, A., Michaels, P., Larson, J.E., Olschewski, M.,
Wilkniss, S., Batia, K., 2009. Self-stigma, group identification, perceived legit-
imacy of discrimination and mental health service use. Br. J. Psychiatry 195,
551–552.
Rüsch, N., Lieb, K., Bohus, M., Corrigan, P.W., 2006. Brief reports: self-stigma, empow-
erment, and perceived legitimacy of discrimination among women with mental
illness. Psychiatr. Serv. 57, 399.
Sandahl, C., Lindberg, S., Rönneberg, S., 1990. Efficacy expectations among
alcohol-dependent patients—a Swedish version of the situational confidence
questionnaire. Alcohol Alcohol. 25, 67.
Schomerus, G., 2009. Obstacles in the way—stigma and help-seeking. Psychiatr. Prax.
36, 53–54.
Schomerus, G., Holzinger, A., Matschinger, H., Lucht, M., Angermeyer, M.C.,
2010. Public attitudes towards alcohol dependence. Psychiatr. Prax. 37, 111–
118.
Schomerus, G., Matschinger, H., Angermeyer, M.C., 2006. Alcoholism: illness beliefs
and resource allocation preferences of the public. Drug Alcohol Depend. 82,
204–210.
Schomerus, G., Matschinger, H., Angermeyer, M.C., 2009. The stigma of psychiatric
treatment and help-seeking intentions for depression. Eur. Arch. Psychiatry Clin.
Neurosci. 259, 298–306.
Sitharthan, T., Kavanagh, D.J., 1991. Role of self-efficacy in predicting outcomes from
a programme for controlled drinking. Drug Alcohol Depend. 27, 87–94.
Vogel, D.L., Wade, N.G., Haake, S., 2006. Measuring the self-stigma associated with
seeking psychological help. J. Counsel. Psychol. 53, 325–337.
Wang, J., Kelly, B., Booth, B., Falck, R., Leukefeld, C., Carlson, R., 2010. Examining
factorial structure and measurement invariance of the Brief Symptom Inventory
(BSI)-18 among drug users. Addict. Behav. 35, 23–29.
Wang, P.S., Angermeyer, M., Borges, G., Bruffaerts, R., Chiu, W.T., de Girolamo, G.,
Fayyad, J., Gureje, O., Haro, J.M., Huang, Y., 2007. Delay and failure in treatment
seeking after first onset of mental disorders in the World Health Organization’s
World Mental Health Survey Initiative. World Psychiatry 6, 177.
WHO, 2004. Global Status Report on Alcohol 2004. World Health Organization,
Geneva.
Wittchen, H.U., Jacobi, F., 2001. Die Versorgungssituation psychischer Störungen in
Deutschland. Bundesgesundheitsblatt - Gesundheitsforschung - Gesundheitss-
chutz 44, 993–1000.
Zingg, C., Schmidt, P., Kufner, H., Kolb, W., Zemlin, U., Soyka, M., 2009. The rela-
tionship between self-efficacy and abstinence from alcohol after outpatient and
inpatient treatment—a 2-year follow-up. Psychother. Psychosom. Med. Psychol.
59, 307–313.
