Soc Psychiatry Psychiatr Epidemiol DOI 10.1007/s00127-014-0923-z
ORIGINAL PAPER
Stigma resistance in eating disorders Scott Griffiths • Jonathan M. Mond • Stuart B. Murray • Chris Thornton • Stephen Touyz
Received: 19 February 2014 / Accepted: 22 June 2014 Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose Stigma resistance, described as the capacity to counteract or remain unaffected by the stigma of mental illness, may play a crucial role in the fight against stigma. Little is known, however, about stigma resistance and its correlates in people with eating disorders. This study investigated stigma resistance in people currently diagnosed (n = 325) and recovered (n = 127) from anorexia nervosa, bulimia nervosa, and EDNOS. Methods Participants completed an Internet survey that included the Stigma Resistance subscale of the Internalized Stigma of Mental Illness Scale together with a battery of psychosocial and psychiatric measures. Results A minimal-to-low level of stigma resistance was exhibited by 26.5 % of currently diagnosed participants compared to just 5.5 % of recovered participants. Stigma resistance was significantly higher among the recovered than the currently diagnosed (Cohen’s d = 0.25) after controlling for differences in eating disorder and depression symptoms, attitudes about seeking psychological help,
S. Griffiths (&) S. Touyz School of Psychology, University of Sydney, Sydney, NSW 2006, Australia e-mail: [email protected]; [email protected] S. Griffiths S. B. Murray C. Thornton The Redleaf Practice, Wahroonga, NSW 2076, Australia J. M. Mond Research School of Psychology, The Australian National University, Canberra, ACT 0200, Australia J. M. Mond Department of Psychology, Macquarie University, Sydney, NSW 2109, Australia
self-esteem, years between symptom onset and diagnosis, and years since diagnosis. Greater stigma resistance among the currently diagnosed was associated with less marked eating disorder and depression symptoms, higher selfesteem, more positive attitudes about seeking psychological treatment, and lower internalized stigma. Conclusions Stigma resistance is a promising concept that warrants further study. Researchers should consider designing interventions that specifically cultivate stigma resistance in people with eating disorders as a complement to current interventions that target public perceptions of eating disorders. Clinicians may consider incorporating the concept into their practice to help patients rebuff the adverse effects of mental illness stigmatization. Keywords Stigma Stigma resistance Internalized stigma Eating disorders
Introduction Stigmatization is an unfortunate reality for people living with eating disorders. Research has shown that stigma toward eating disorders is perpetrated by adolescents, university students, members of the general population, medical professionals, and service-users themselves [1–5]. The most common forms of this stigma include the perceptions that people with eating disorders are attention seekers who are personally responsible for their condition and who have only themselves to blame for their predicament [1–5]. Preliminary research suggests that some of these stigmatizing attitudes can be changed, at least in the short-term [6, 7], offering hope for future long-term interventions that aim to improve eating disorders mental health literacy.
123
Soc Psychiatry Psychiatr Epidemiol
The aforementioned body of research has investigated a form of stigma called ‘‘enacted stigma,’’ which refers to discriminatory behaviours and stigmatizing attitudes that are perpetrated by others toward people with eating disorders [8]. Comparatively little research, however, has investigated the process by which enacted stigma becomes internalized in people with eating disorders. The internalization of stigma is transformative; a set of desired and valued self-identities (for example, as a parent, as a partner, or as an employee) is gradually subsumed by a stigmatized and devalued identity or set of identities [9]. The damage caused by internalized mental illness stigma is considerable. A recent review found that internalized mental illness stigma was associated with increased symptom severity, poorer treatment adherence, reduced self-esteem, reduced hope, and reduced empowerment [10]. Researchers have stressed the importance of disrupting the process of internalization, arguing that an overemphasis on investigating enacted stigma may have been counterproductive [11]. The result of these criticisms has been an influx of new research that aims to identify targets for interventions to reduce internalized stigma [12, 13]. To this end, researchers have identified stigma resistance as a potential line-of-defence against the internalization process. Rooted in the broader literature on resilience, stigma resistance describes the capacity to resist, counteract or otherwise remain unaffected by mental illness stigmatization [14]. In patients with schizophrenia, greater stigma resistance has been associated with reduced internalized stigma, reduced depression, greater selfesteem, greater empowerment, and improved quality of life [15, 16]. A similar pattern of results has been observed in patients with depression and bipolar disorder [17] and in patients with stigmatized biological diseases, including inflammatory bowel disease [18]. In addition, researchers have expressed optimism about the contribution of stigma resistance to recovery from mental illness [16, 17]. Tempering this enthusiasm, however, is the fact that no research has compared stigma resistance in people currently diagnosed with mental illness to those in recovery from mental illness. If people who had recovered from mental illness were shown to exhibit greater stigma resistance after controlling for expected differences in psychopathology, the evidence base suggesting that stigma resistance contributes to recovery from mental illness would be strengthened considerably. Stigma resistance might facilitate recovery from mental illness, including eating disorders, in several ways. According to the stress-coping model of stigma, an individual’s wellbeing is undermined when their perception of the harm due to stigma (the primary appraisal) exceeds their perceived ability to cope with the stigma (the secondary appraisal) [19, 20]. Stigma resistance may influence
123
the primary appraisal by leading to less harmful and/or less psychologically taxing appraisals of stigma. Alternatively, stigma resistance may act on the secondary appraisal by fortifying a person’s internal resources for coping with stigma, leading to more favourable perceptions of their ability to cope. Common to both explanations is that stigma resistance promotes and protects an individual’s psychological wellbeing, thus helping to facilitate recovery. Benefits to wellbeing may extend beyond eating disorder symptoms to symptoms of other mental illnesses, such as depression, and to self-esteem more generally. More positive appraisals related to stigma may also generalize to other forms of stigma, for example, the self-stigma associated with seeking psychological help, and may act as a buffer against the internalization of mental illness stigma into one’s self-identity. To our knowledge, no research has investigated the prevalence and correlates of stigma resistance as this applies to individuals with eating disorders or the potential role of stigma resistance in promoting recovery in this population. To address these issues, the present study investigated stigma resistance in people currently diagnosed with eating disorders and people recovered from eating disorders. Two major hypotheses were made: first, that stigma resistance would be associated with less marked eating disorder symptoms, less marked depression symptoms, more positive attitudes about seeking psychological treatment, higher self-esteem, and lower internalized stigma; and second, that stigma resistance would be greater in people that had recovered from eating disorders relative to people currently diagnosed with eating disorders, even after controlling for expected differences in eating disorder symptoms, depression symptoms, attitudes about seeking psychological treatment, and self-esteem.
Method Recruitment Participants were recruited via online and print advertisements circulated by various eating disorder organisations and community support groups, including major eating disorder charities, located in Australia, the United Kingdom, and the United States of America. The advertisements requested volunteers who were currently diagnosed with an eating disorder or in recovery from an eating disorder to participate in a brief online study about stigmatization. A link provided in the advertisement led participants to the survey, which began with the participant information statement and consent form, followed by a series of screening questions: ‘‘Have you ever been diagnosed with an eating disorder by a mental health professional, e.g., a
Soc Psychiatry Psychiatr Epidemiol
psychologist?’’, ‘‘Are you currently diagnosed with an eating disorder or are you recovered from an eating disorder?’’, ‘‘What eating disorder subtype are you currently diagnosed with?/What eating disorder subtype have you recovered from?’’ The survey then presented participants with the battery of psychiatric and psychosocial questionnaires. Once completed the survey could not be retaken. A total of 456 individuals completed the online survey and self-reported a current diagnosis of, or recovery from a diagnosis of, anorexia nervosa, bulimia nervosa, or an eating disorder not otherwise specified (EDNOS). All participants indicated that their current or former diagnoses were made by a mental health professional. Data integrity and exclusion All data were examined for duplicate responses, repeat Internet protocol addresses, and suspicious patterns of responding. In addition, three valid-responding checks were embedded in the study (e.g., For validity purposes, please select ‘‘Agree’’ as your answer to this question). Correct response rates to these questions have averaged 89 % in undergraduate samples [2, 21, 22]. The correct response rate in the present sample averaged 97.4 %, indicating that the vast majority of participants were responding meaningfully to the survey questions. Four participants who failed two or more validity checks had their data excluded. The final sample constituted 452 participants. Validity of the recovered group A purposeful decision was made not to provide participants with a definition of recovery because of widespread disagreement in the eating disorders field about what constitutes recovery from eating disorders [23, 24]. Instead, we devised four criteria that would provide support for the validity of the recovered group. These criteria were (a) significant and large differences (Cohen’s d [ 1.00) in eating disorder attitudes, as measured by the four subscales and the global score of the Eating Disorder ExaminationQuestionnaire (EDE-Q) [25], between the currently diagnosed and recovered groups, (b) significant differences in eating disorder behaviours, including purging, laxative use, diuretic use, objective binges, subjective binges, and hard exercise, as measured by the behavioural component of the EDE-Q, between the currently diagnosed and recovered groups, and (c) a median score of zero for each eating disorder behaviour for the recovered group. These three criteria were chosen because a marked reduction in eating disorder symptoms is common to all definitions of recovery from eating disorders [23, 24]. For anorexia nervosa we further specified that (d) the recovered group should report
a significantly higher body mass index (BMI) than currently diagnosed group. We did not specify an effect size for this criterion because BMI is a particularly contentious topic in discussions about recovery from anorexia nervosa [23]. Measures Stigma resistance and internalized stigma Stigma resistance and internalized stigma were measured using the 29-item, 5-subscale Internalized Stigma of Mental Illness (ISMI) scale [14]. The Stigma Resistance subscale consists of five reverse coded items: ‘‘In general, I am able to live life the way I want to,’’ ‘‘People with mental illness make important contributions to society,’’ ‘‘I can have a good fulfilling life, despite my mental illness,’’ ‘‘Living with mental illness has made me a tough survivor,’’ and ‘‘I feel comfortable being seen in public with an obviously mentally ill person.’’ Previous research has shown that the Stigma Resistance subscale is conceptually and psychometrically distinct from the four other subscales of the ISMI [16, 26, 27]. Consistent with previous research, the present study measured stigma resistance using the Stigma Resistance subscale and measured internalized stigma by summing the averages of the remaining four subscales of the ISMI [15–18]. These four subscales are Alienation, Stereotype Endorsement, Perceived Discrimination, and Social Withdrawal. Internalized stigma measured using the four subscales of the ISMI has demonstrated excellent internal consistency in previous research, with an average Cronbach’s a of 0.90 [26]. The Stigma Resistance subscale has been less internally consistent with an average Cronbach’s a of 0.64 [26], which is deemed acceptable by psychometric standards [28]. In the present study, Cronbach’s a for Internalized Stigma and Stigma Resistance was 0.90 and 0.65, respectively. Internalized stigma was measured in currently diagnosed participants only because the items that constitute this measure assume that the participant is currently suffering from a mental illness. Items on the Stigma Resistance subscale do not presuppose a current mental illness diagnosis and were administered to both the currently diagnosed and recovered group. Eating disorder symptoms The EDE-Q [25] measured disordered eating attitudes and behaviours. The attitudes component asks participants how frequently various eating attitudes have occurred during the past 28 days and contains four subscales: Dietary Restraint, Eating Concern, Shape Concern, and Weight
123
Soc Psychiatry Psychiatr Epidemiol
Concern. Responses are given using a 6-point scale (0 = no days, 6 = every day). In addition, mean scores on the four subscales were averaged together to calculate a global EDE-Q score. Cronbach’s a for the global EDE-Q was 0.96 in the present study. The behaviour component of the EDE-Q asks participants to rate how many times they have engaged in the following during the past 28 days: laxative use, diuretic use, purging, objective binges, subjective binges, and ‘‘hard’’ exercise. Cronbach’s a for eating disorder behaviours was 0.69 in the present study.
Table 1 Categorical data for currently diagnosed and recovered participants, including self-reported diagnosis, treatment status, sex, current residence, employment status, highest level of education, and relationship status Variable
The 10-item Self-Stigma of Seeking Help scale (SSOSH) was used to measure the existence of negative attitudes about seeking psychological treatment [29]. The SSOSH has demonstrated predictive validity insofar as predicting which undergraduates sought help from a mental health professional over a 2-month period [29]. Internal consistency of the SSOSH among undergraduates is acceptable at 0.78 and test-reliability is acceptable at 0.82 over a 3-week period. Cronbach’s a for the SSOSH was 0.88 in the present study. Higher scores on this measure indicate more negative attitudes towards help-seeking.
Anorexia nervosa Bulimia nervosa
The 7-item Depression subscale of the 21-item Depression Anxiety Stress Scales (DASS-21) was used to measure depression symptoms [30]. The Depression subscale has demonstrated concurrent validity with other measures of depression and acceptable internal consistency (a = 0.88) in a general population sample of adults [30]. Cronbach’s a was 0.94 in the present study. Self-esteem The 10-item Self-Esteem Scale (SES) was used to measure global self-esteem [31]. Cronbach’s a was 0.89 in the present study. Statistical analyses
123
n
n
df
%
%
2 51.7 21.2
82 17
64.6 13.4
88
27.1
28
22.0
Currently in treatment for eating disorder
183
56.3
17
13.4
Not currently in treatment for eating disorder
133
40.9
108
85.0
9
2.8
2
1.6
13
4.0
2
1.6
309
95.1
123
96.9
3
0.9
2
1.6
98
30.2
29
22.8
110
33.8
40
31.5
Treatment status
Other
2
Sex Male Female Transgender Current residence United States of America United Kingdom
65
19.7
28
22.0
Other (11 countries total)
53
16.3
30
23.6
Paid employment
116
35.7
46
36.2
Student
157
48.3
67
52.8
52
16.0
14
11.0
Employment status
Other Highest level of education High school/ secondary school
81
24.9
20
15.7
University/college
216
66.5
94
74.0
28
8.6
13
10.2
Single
206
63.3
57
44.9
In a relationship
111
34.2
67
52.8
8
2.5
3
2.4
Other Relationship status
Other
All statistical analyses were conducted using SPSS version 21. To examine differences between the currently diagnosed and recovered group for categorical variables such as sex, a series of Chi-square tests of independence were conducted with the categorical variable entered as the first independent variable and diagnostic status (currently diagnosed or recovered) entered as the second independent variable. The results of these analyses are shown in Table 1. To examine differences between the currently
Statistical comparison
168 69
Australia
Depression symptoms
Recovered
Self-reported diagnosis
EDNOS
Attitudes about seeking psychological help
Currently diagnosed
v2 6.59
p 0.037
71.89 \0.001
2
2.00
0.368
3
4.80
0.187
2
1.92
0.383
2
4.46
0.107
2
13.40
0.001
diagnosed and recovered group for continuous variables such as age, a series of univariate ANOVAs were conducted with the continuous variable entered as the dependent variable and diagnostic status (currently diagnosed or recovered) entered as the independent variable. The results of these analyses are shown in Table 2.
Soc Psychiatry Psychiatr Epidemiol Table 2 Continuous data for currently diagnosed and recovered participants, including age, BMI, eating disorder symptomatology, depression symptomatology, self-esteem, attitudes about seeking help, internalized stigma, and years since symptom onset, since diagnosis, and since recovery
Variable
Age
Currently diagnosed
Recovered
Mean
Median
Mean
SD
Median
F (df) or U
SD
Statistical comparison p
24.61
7.14
23
25.30
8.02
23
U = 20,113.50
0.67
Anorexia nervosa
17.78
3.63
17.40
21.95
7.82
20.31
F(1, 237) = 46.83
\0.001
Bulimia nervosa and EDNOS
23.22
8.25
20.91
24.56
6.57
22.66
F(1, 184) = 0.80
Body mass index
0.373
Eating Disorders Examination Questionnaire Dietary restraint
4.27
1.42
4.60
1.89
1.62
1.60
F(1, \0.001 450) = 236.35
Eating concern
3.98
1.23
4.20
1.72
1.44
1.60
F(1, \0.001 450) = 279.09
Shape concern
5.11
1.09
5.50
3.21
1.76
3.25
F(1, \0.001 450) = 191.74
Weight concern
4.70
1.21
4.80
2.79
1.62
2.80
F(1, \0.001 450) = 184.44
Global score
4.51
1.06
4.80
2.41
1.45
2.31
F(1, \0.001 450) = 290.36
Laxative use
2.91
10.75
0
0.38
1.86
0
U = 11,409.50
\0.001
Diuretic use
0.93
5.28
0
0.07
0.38
0
U = 16,361.50
\0.001
Exercised hard Objective binge
7.70 7.51
9.68 16.46
3 0
3.35 1.55
6.69 4.82
0 0
U = 19,172.00 U = 14,358.50
0.024 \0.001
Subjective binge
11.56
14.77
10
4.434
6.83
0
U = 13,829.00
\0.001
Purging
11.53
22.84
1
0.81
3.79
0
U = 11,641.00
\0.001
13.08
5.57
14.00
7.20
5.68
7.00
F(1, 442) = 99.91
\0.001
Global self-esteem
9.16
5.07
9.00
15.04
5.80
15.00
Negative attitudes about seeking help
2.81
0.77
2.70
2.61
0.79
2.50
F(1, 450) = 6.19
0.013
Internalized stigma
2.51
0.48
2.52
Years between symptom onset and diagnosis
9.59
7.27
9.00
10.42
8.03
8
U = 16,565.00
0.005
Years since diagnosis
5.79
6.02
4.00
U = 16,361.50
0.001
Depression symptoms
Body mass index for participants with anorexia nervosa is given separately from participants with bulimia nervosa and EDNOS. Statistical comparisons were either ANOVA F tests or Mann– Whitney U tests
Years spent in recovery
To determine the relative amounts of stigma resistance exhibited by individuals currently diagnosed and recovered from eating disorders, each participant was assigned a level of stigma resistance according to their mean score on the Stigma Resistance subscale. The cut-offs for each level of stigma resistance were taken from previous research [15, 17]: Minimal (\2), Low (2—2.5), Moderate (2.5–3), High (3?). A Pearson correlational analysis was conducted to determine the relationships between stigma resistance, internalized stigma, eating disorder symptoms, depression symptoms, negative attitudes about seeking psychological help, and self-esteem. The role of stigma resistance in eating disorder recovery was examined by conducting a univariate ANCOVA with
7.44
7.01
6
2.63
3.33
1
F(1, \0.001 445) = 112.56
stigma resistance entered as the dependent variable, diagnostic status entered as the independent variable, and eating disorder symptoms, depression symptoms, attitudes about seeking psychological help, self-esteem, years between symptom onset and diagnosis, and years since diagnosis entered as covariates. The latter two variables were included as covariates because the results of the analyses shown in Table 2 indicated significant differences between the currently diagnosed and recovered group for these variables. Finally, the relationships between stigma resistance and the demographic and clinical variables included in the study were examined by conducting a series of univariate ANOVAs with diagnostic status (currently diagnosed versus recovered) entered as the independent variable. All
123
Soc Psychiatry Psychiatr Epidemiol
nervosa was significantly higher, for participants in the recovered group versus the currently diagnosed group. All four criteria that were devised to support the validity of the recovered group were satisfied by the data. First, scores on the four subscales and the global score of the EDE-Q were significantly lower for recovered participants compared to currently diagnosed participants and these differences were very large (Cohen’s d = 1.28–1.61). Global EDE-Q scores for the recovered group (M = 2.41) were higher than population norms for young adult women living in Australia and America (M = 1.52–1.74) [33, 34] but considerably lower than the cut-off ([4.0) for clinical significance [33, 34]. Second, the recovered group reported significantly fewer eating disorder behaviours across each of the seven behaviours assessed. Third, the median score in the recovered group for each eating disorder behaviour was zero. Fourth, participants that had recovered from anorexia nervosa had significantly higher BMIs than currently diagnosed participants and the size of this effect was large (Cohen’s d = 0.89). The ANCOVA conducted to examine the role of stigma resistance in eating disorder recovery revealed that stigma resistance was significantly greater in people who reported being recovered from an eating disorder (covariate-adjusted mean = 3.03, SD = 0.43) than in people who reported being currently diagnosed with an eating disorder (covariate-adjusted mean = 2.87, SD = 0.51), F(1, 425) = 4.26, p = 0.006. The size of the effect was small (Cohen’s d = 0.25). Table 3 shows the proportion of participants in the currently diagnosed and recovered group who fell into each level of stigma resistance. The subsequent ANOVAs conducted to explore the relationships between stigma resistance and the demographic and clinical variables shown in Table 1 revealed that stigma resistance did not significantly differ as a function of eating disorder subtype, treatment status, employment status, or highest level of education, but did differ significantly as a function of current residence, F(3, 447) = 4.78, p = 0.031, and relationship status,
significant between-groups differences indicated by the ANOVAs were followed-up by planned contrasts using the Holm–Bonferroni method to control the family-wise error rate at 0.05. Statistical assumptions Dependent variables that were slated for ANOVA had their data distributions tested for normality. Normality was established if (a) the absolute values of the skewness and kurtosis statistic were less than twice the standard error of the respective statistics [32], and if (b) visual inspection of the frequency histogram and normal Q–Q plot corroborated the skewness and kurtosis statistics. Variables deemed to be non-normal were instead analysed using non-parametric Mann–Whitney U tests. The variable of primary interest, stigma resistance, met both criteria for normality. Ethical standards All participants gave their informed consent prior to inclusion in the study. The study was approved by the Human Research and Ethics Committee of the University of Sydney.
Results Descriptive data and the outcomes of the Chi-square tests of independence are shown in Table 1 and descriptive data and the outcomes of the ANOVA and Mann–Whitney tests are shown in Table 2. As shown in Table 1, there were no significant differences between the diagnosed and recovered groups in self-reported sex, current residence, employment status, and highest level of education, while significant differences did emerge between the groups for self-reported diagnosis, treatment status, and marital status. As shown in Table 2, eating disorder symptomatology was significantly lower, and BMI for participants with anorexia
Table 3 Number (n) and proportion (%) of participants with various levels of stigma resistance Stigma resistance
Anorexia nervosa
Bulimia nervosa
EDNOS
All subtypes
Currently diagnosed
Recovered
Currently diagnosed
Recovered
Currently diagnosed
Recovered
Currently diagnosed
Recovered
n
n
n
n
n
n
n
n
%
%
%
%
%
%
%
%
Minimal (\2)
6
3.6
0
0.0
2
2.9
0
0.0
3
3.4
0
0.0
11
3.4
0
0.0
Low (2–2.5)
35
20.8
4
4.9
16
23.2
0
0.0
13
14.8
3
10.7
64
23.1
7
5.5
Moderate (2.5–3) High (3?)
85 42
50.6 25.0
34 44
41.5 53.7
37 14
53.6 20.3
6 11
35.3 64.7
45 27
51.1 30.7
7 18
25.0 64.3
167 83
51.4 25.5
47 73
37.0 57.5
Data are given for each eating disorder subtype and for all eating disorder subtypes combined, and for currently diagnosed participants and recovered participants
123
Soc Psychiatry Psychiatr Epidemiol Table 4 Correlations between stigma resistance and eating disorder symptoms, depression symptoms, self-esteem, attitudes about seeking psychological treatment, and internalized stigma for currently diagnosed participants and recovered participants Variable
Correlation with stigma resistance Currently diagnosed
Recovered
N
r
N
r
Global EDE-Q score
325
-0.26 \0.001
127
-0.11
DASS-21 depression subscale score
317
-0.36 \0.001
127
-0.29 \0.001
Global self-esteem
320
0.36 \0.001
127
Attitudes about seeking psychological treatment
325
-0.29 \0.001
127
Internalized stigma
325
-0.40 \0.001
p
p 0.205
0.35 \0.001 -0.24
0.008
F(2, 448) = 5.34, p = 0.005. The follow-up contrast analyses suggested that stigma resistance was significantly higher among participants who were residing in the USA compared to the remaining countries, F(1, 447) = 12.69, p \ 0.001. Interestingly, stigma resistance was significantly higher in the minority of participants who reported their family status as ‘‘other’’ compared to participants who reported that they were single, F(1, 448) = 6.07, p = 0.014, and compared to participants who reported being in a relationship, F(1, 448) = 9.09, p = 0.003. The results of the correlational analysis are shown in Table 4. Stigma resistance was associated with less marked eating disorder symptoms, less marked depression symptoms, more positive attitudes about seeking psychological treatment, and higher self-esteem. Among participants who were currently diagnosed, stigma resistance was associated with lower internalized stigma.
Discussion The present study investigated stigma resistance, the capacity to resist or be unaffected by mental illness stigmatization, in people who self-reported a current diagnosis, or recovery from a diagnosis, of anorexia nervosa, bulimia nervosa, or EDNOS. The results revealed that 26.5 % of currently diagnosed participants had low-to-minimal stigma resistance. In comparison, a study of people with self-reported depression or bipolar disorder found that 32 % had low-to-minimal stigma resistance [17], whilst 50.7 % of people with self-reported schizophrenia exhibited low-to-minimal stigma resistance [15]. In the present study there were no significant differences in stigma
resistance between the eating disorder subtypes. However, low-to-minimal stigma resistance was exhibited by just 5.5 % of people in recovery from an eating disorder, implying that stigma resistance may play a role in the recovery process. Both major hypotheses of the present study were supported. Among both the currently diagnosed and the recovered, stigma resistance was significantly associated with less marked eating disorder symptoms, less marked depression symptoms, more positive attitudes about seeking psychological treatment and higher self-esteem. In addition, for currently diagnosed participants, greater stigma resistance was associated with lower internalized stigma. Moreover, stigma resistance discriminated participants currently diagnosed with an eating disorder from participants recovered from an eating disorder, even after controlling for differences in eating disorder and depression symptoms, attitudes about seeking psychological help, self-esteem, years between symptom onset and diagnosis, and years since diagnosis. Collectively, the results suggest that stigma resistance plays an important and positive role in people with eating disorders. The primary mechanism by which stigma resistance achieves positive outcomes in people with eating disorders may be through disruption of the link between enacted stigma and internalized stigma. Stigma resistance may lead to less harmful appraisals of stigma or it may bolster an individual’s ability to cope with stigma, leading to a more favourable appraisal of their ability to cope [19, 20]. In turn, these favourable appraisals may reduce the internalization of stigma, promoting wellbeing and facilitating recovery. These positive appraisals may also generalize to other forms of stigma, for example, the self-stigma associated with seeking psychological help, which may explain why greater stigma resistance was associated with more positive attitudes about mental health care in the present study. The notion that more positive appraisals protect and promote an individual’s psychological wellbeing was supported by the positive interrelationships we observed between stigma resistance, depression symptoms, eating disorder symptoms, and self-esteem. It is important, however, to acknowledge the possibility that the process of recovery itself is what facilitates the development of stigma resistance. Individuals who receive psychotherapy for an eating disorder may develop cognitive strategies that allow them to mount a superior resistance to stigma, implying that stigma resistance is a helpful by-product of the recovery process. Additional research exploring these mechanisms is needed. Stigma resistance demonstrates promise for future interventions into eating disorder stigmatization and for clinicians who treat eating disorders. Because an intervention to raise stigma resistance would be likely to target
123
Soc Psychiatry Psychiatr Epidemiol
the stigmatized individual, it can be viewed as complementary to recent efforts that target public perceptions of stigma resistance with the aim of reducing enacted stigma [7]. A two-sided approach that targets both enacted and internalized stigma may provide a more comprehensive and effective strategy to reduce stigmatization of eating disorders. An additional benefit of an intervention that targets stigma resistance in the individual rather than enacted stigma among the public may be the restoration of agency to the sufferer, with potential benefits to empowerment, hope, and self-esteem. Future research is needed to explore these possibilities. Clinicians may also benefit from cultivating stigma resistance in their practice, insofar as it equips their patients with the skills needed to combat the widespread prevalence of eating disorder stigmatization. Several strengths and limitations of the present study are noted. First, the internal consistency of the Stigma Resistance subscale in the present study was relatively low (a = 0.65). Although this figure is higher than in previous research [15–17] and is acceptable by published psychometric standards [28], there is scope to improve the psychometric properties of the subscale, or to develop an independent instrument [16]. Second, the sample may not be fully representative of the broader population of eating disorders patients. In particular, patients who are acutely ill, patients without Internet access, and patients unaffiliated with eating disorder organisations may have been underrepresented. Third, there was potential for inaccuracy in participants’ self-reported diagnoses and self-reported recovery status. For example, some participants who identified as recovered may have very recently completed a course of inpatient treatment, making their status as ‘‘recovered’’ a matter for debate. Fourth, although attitudes about help-seeking were controlled for in comparisons of currently diagnosed versus recovered patients, data bearing on differences in the type and length of treatment actually received were not collected. Fifth, the cross-sectional design of the study precludes any conclusions about causality. Particular strengths of the study include the recruitment of a large, geographically diverse sample of individuals with current or past eating disorders and the novel investigation of stigma resistance in this population. We offer several suggestions for future research. The literature on stigma resistance and eating disorders may benefit from the development of a stigma resistance instrument that is specific to eating disorders and which is populated by items that assess an individual’s ability to resist the specific types of stigma directed toward eating disorders. For example, one in three members of the general public stigmatize individuals with eating disorders as ‘‘having only themselves to blame’’ and being ‘‘able to pull themselves together’’, but less than one in ten members of the public stigmatize schizophrenia in this way [4].
123
Moreover, we stress the importance of defining and operationalising ‘‘recovery’’ in future research that uses a similar method to the present study. The lack of consensus surrounding definitions of recovery is not confined to eating disorders [35] and the ability of future research to make valid comparisons between studies will depend on researchers being explicit about their definitions and operationalisations of recovery. In conclusion, the present study found support for two major hypotheses about stigma resistance: first, that stigma resistance would be associated with positive psychosocial and psychiatric outcomes in people currently diagnosed and recovered from eating disorders, and second, that stigma resistance would discriminate between the currently diagnosed and the recovered after controlling for expected differences in psychopathology. Stigma resistance may bolster current intervention strategies that focus primarily on enacted stigma and which neglect internalized stigma. Clinicians may also benefit by fostering stigma resistance in their patient. Conflict of interest On behalf of all authors the corresponding author states that there is no conflict of interest.
References 1. Fleming J, Szmukler GI (1992) Attitudes of medical professionals towards patients with eating disorders. Aust N Z J Psychiatry 26:436–443. doi:10.3109/00048679209072067 2. Griffiths S, Mond JM, Murray SB, Touyz S (2013) Young peoples’ stigmatizing attitudes and beliefs about anorexia nervosa and muscle dysmorphia. Int J Eat Disord 47:189–195. doi:10. 1002/eat.22220 3. Mond J, Marks P (2007) Beliefs of adolescent girls concerning the severity and prevalence of bulimia nervosa. Aust J Psychol 59:87–93. doi:10.1080/00049530601148413 4. Crisp AH, Gelder MG, Riz S, Meltzer HI, Rowlands OJ (2000) Stigmatisation of people with mental illnesses. Br J Psychiatry 177:4–7. doi:10.1192/bjp.177.1.4 5. Robinson KJ, Mountford VA, Sperlinger D (2013) Being men with eating disorders: perspectives of male eating disorder service-users. J Health Psychol 18:176–186. doi:10.1177/1359105 312440298 6. Crisafulli MA, Von Holle A, Bulik C (2008) Attitudes towards anorexia nervosa: the impact of framing on blame and stigma. Int J Eat Disord 41:333–339. doi:10.1002/eat.20507 7. Zwickert K, Rieger E (2013) Stigmatizing attitudes towards individuals with anorexia nervosa: an investigation of attribution theory. J Eat Disord 1:5. doi:10.1186/2050-2974-1-5 8. Link BG, Yang LH, Phelan JC, Collins PY (2004) Measuring mental illness stigma. Schizophr Bull 30:511–541 9. Yanos PT, Roe D, Markus K, Lysaker PH (2008) Pathways between internalized stigma and outcomes related to recovery in schizophrenia spectrum disorders. Psychiatr Services 59: 1437–1442. doi:10.1176/appi.ps.59.12.1437 10. Livingston JD, Boyd JE (2010) Correlates and consequences of internalized stigma for people living with mental illness: a systematic review and meta-analysis. Soc Sci Med 71:2150–2161. doi:10.1016/j.socscimed.2010.09.030
Soc Psychiatry Psychiatr Epidemiol 11. Thornicroft G, Rose D, Kassam A, Sartorius N (2007) Stigma: ignorance, prejudice or discrimination? Br J Psychiatry 190: 192–193. doi:10.1192/bjp.bp.106.025791 12. Lasalvia A, Zoppei S, Bortel TV, Bonetto C, Cristofalo D, Wahlbeck K, Thornicroft G (2013) Global pattern of experienced and anticipated discrimination reported by people with major depressive disorder: a cross-sectional survey. Lancet 381:52–55. doi:10.1016/S0140-6736(12)61379-8 13. Thornicroft G, Brohan E, Rose D, Sartorius N, Leese M (2013) Global pattern of experienced and anticipated discrimination against people with schizophrenia: a cross-sectional survey. Lancet 373:408–415. doi:10.1016/S0140-6736(08)61817-6 14. Ritsher JB, Otilingam PG, Grajales M (2003) Internalized stigma of mental illness: psychometric properties of a new measure. Psychiatry Res 121:31–49. doi:10.1016/j.psychres.2003.08.008 15. Brohan E, Elgie R, Sartorius N, Thornicroft G (2010) Selfstigma, empowerment and perceived discrimination among people with schizophrenia in 14 European countries: the GAMIANEurope study. Schizophr Res 122:232–238. doi:10.1016/j.schres. 2010.02.1065 16. Sibitz I, Unger A, Woppmann A, Zidek T, Amering M (2009) Stigma resistance in patients with schizophrenia. Schizophr Bull 37:316–323. doi:10.1093/schbul/sbp048 17. Brohan E, Gauci D, Sartorius N, Thornicroft G (2011) Selfstigma, empowerment and perceived discrimination among people with bipolar disorder or depression in 13 European countries: the GAMIAN-Europe study. J Affect Disord 129:56–63. doi:10. 1016/j.jad.2010.09.001 18. Taft TH, Ballou S, Keefer L (2013) A preliminary evaluation of internalized stigma and stigma resistance in inflammatory bowel disease. J Health Psychol 18:451–460. doi:10.1177/135910531 2446768 19. Ru¨sch N, Corrigan PW, Heekeren K, Theodoridou A, Dvorsky D, Metzler S, Ro¨ssler W (2014) Well-being among persons at risk of psychosis: the role of self-labelling, shame, and stigma-stress. Psychiatr Services 65:483–489 20. Kaiser CR, Major B, Mccoy SK (2004) Expectations about the future and the emotional consequences of perceiving prejudice. Pers Soc Psychol Bull 30:173–184. doi:10.1177/014616720 3259927 21. Griffiths S, Angus D, Murray SB, Touyz S (2014) Unique associations between young adult men’s emotional functioning and their body dissatisfaction and disordered eating. Body Image 11:175–178. doi:10.1016/j.bodyim.2013.12.002
22. Griffiths S, Murray SB, Touyz S (2013) Drive for muscularity and muscularity-oriented disordered eating in men: the role of set shifting difficulties and weak central coherence. Body Image 10:636–639. doi:10.1016/j.bodyim.2013.04.002 23. Dawson L, Rhodes P, Touyz S (2014) Defining recovery from anorexia nervosa: a Delphi study to explore expert practitioners’ views (submitted) 24. Yu J, Agras SW, Bryson S (2013) Defining recovery in adult bulimia nervosa. Eat Disord 21:374–379. doi:10.1080/10640266. 2013.827536 25. Fairburn CG, Beglin SJ (1994) Assessment of eating disorders: interview or self-report questionnaire? Int J Eat Disord 16: 363–370. doi:10.1002/1098-108X(199412)16:4\363::AID-EAT 2260160405[3.0.CO;2 26. Boyd JE, Adler EP, Otilingam PG, Peters T (2014) Internalized Stigma of Mental Illness (ISMI) Scale: a multinational review. Compr Psychiatry 55:221–231. doi:10.1016/j.comppsych.2013. 06.005 27. Lysaker PH, Roe D, Yanos PT (2007) Toward understanding the insight paradox: internalized stigma moderates the association between insight and social functioning, hope, and self-esteem among people with schizophrenia spectrum disorders. Schizophr Bull 33:192–199. doi:10.1093/schbul/sbl016 28. Kline P (2000) The handbook of psychological testing, 2nd edn. Routledge, London 29. Vogel DL, Wade NG, Haake S (2006) Measuring the self-stigma associated with seeking psychological help. J Counsel Psychol 53(3):325–337. doi:10.1037/0022-0167.53.3.325 30. Henry J, Crawford J (2005) The short-form version of the Depression Anxiety Stress Scales (DASS-21): construct validity and normative data in a large non-clinical sample. Br J Clin Psychol 44:227–239 31. Rosenberg M (1965) Society and the adolescent self-image. Princeton University Press, Princeton 32. Sheskin DJ (2000) Handbook of parametric and nonparametric statistical procedures, 2nd edn. Chapman & Hall/CRC, New York 33. Luce KH, Crowther JH, Pole M (2008) Eating Disorder Examination Questionnaire (EDE-Q): norms for undergraduate women. Int J Eat Disord 41:273–276. doi:10.1002/eat.20504 34. Mond JM, Hay PJ, Rodgers B, Owen C (2006) Eating Disorder Examination Questionnaire (EDE-Q): norms for young adult women. Behav Res Therapy 44:53–62. doi:10.1016/j.brat.2004.12.003 35. Frank E (2005) Describing course of illness: does our language matter? Int J Eat Disord 38:7–8. doi:10.1002/eat.20145
123
ORIGINAL PAPER
Stigma resistance in eating disorders Scott Griffiths • Jonathan M. Mond • Stuart B. Murray • Chris Thornton • Stephen Touyz
Received: 19 February 2014 / Accepted: 22 June 2014 Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose Stigma resistance, described as the capacity to counteract or remain unaffected by the stigma of mental illness, may play a crucial role in the fight against stigma. Little is known, however, about stigma resistance and its correlates in people with eating disorders. This study investigated stigma resistance in people currently diagnosed (n = 325) and recovered (n = 127) from anorexia nervosa, bulimia nervosa, and EDNOS. Methods Participants completed an Internet survey that included the Stigma Resistance subscale of the Internalized Stigma of Mental Illness Scale together with a battery of psychosocial and psychiatric measures. Results A minimal-to-low level of stigma resistance was exhibited by 26.5 % of currently diagnosed participants compared to just 5.5 % of recovered participants. Stigma resistance was significantly higher among the recovered than the currently diagnosed (Cohen’s d = 0.25) after controlling for differences in eating disorder and depression symptoms, attitudes about seeking psychological help,
S. Griffiths (&) S. Touyz School of Psychology, University of Sydney, Sydney, NSW 2006, Australia e-mail: [email protected]; [email protected] S. Griffiths S. B. Murray C. Thornton The Redleaf Practice, Wahroonga, NSW 2076, Australia J. M. Mond Research School of Psychology, The Australian National University, Canberra, ACT 0200, Australia J. M. Mond Department of Psychology, Macquarie University, Sydney, NSW 2109, Australia
self-esteem, years between symptom onset and diagnosis, and years since diagnosis. Greater stigma resistance among the currently diagnosed was associated with less marked eating disorder and depression symptoms, higher selfesteem, more positive attitudes about seeking psychological treatment, and lower internalized stigma. Conclusions Stigma resistance is a promising concept that warrants further study. Researchers should consider designing interventions that specifically cultivate stigma resistance in people with eating disorders as a complement to current interventions that target public perceptions of eating disorders. Clinicians may consider incorporating the concept into their practice to help patients rebuff the adverse effects of mental illness stigmatization. Keywords Stigma Stigma resistance Internalized stigma Eating disorders
Introduction Stigmatization is an unfortunate reality for people living with eating disorders. Research has shown that stigma toward eating disorders is perpetrated by adolescents, university students, members of the general population, medical professionals, and service-users themselves [1–5]. The most common forms of this stigma include the perceptions that people with eating disorders are attention seekers who are personally responsible for their condition and who have only themselves to blame for their predicament [1–5]. Preliminary research suggests that some of these stigmatizing attitudes can be changed, at least in the short-term [6, 7], offering hope for future long-term interventions that aim to improve eating disorders mental health literacy.
123
Soc Psychiatry Psychiatr Epidemiol
The aforementioned body of research has investigated a form of stigma called ‘‘enacted stigma,’’ which refers to discriminatory behaviours and stigmatizing attitudes that are perpetrated by others toward people with eating disorders [8]. Comparatively little research, however, has investigated the process by which enacted stigma becomes internalized in people with eating disorders. The internalization of stigma is transformative; a set of desired and valued self-identities (for example, as a parent, as a partner, or as an employee) is gradually subsumed by a stigmatized and devalued identity or set of identities [9]. The damage caused by internalized mental illness stigma is considerable. A recent review found that internalized mental illness stigma was associated with increased symptom severity, poorer treatment adherence, reduced self-esteem, reduced hope, and reduced empowerment [10]. Researchers have stressed the importance of disrupting the process of internalization, arguing that an overemphasis on investigating enacted stigma may have been counterproductive [11]. The result of these criticisms has been an influx of new research that aims to identify targets for interventions to reduce internalized stigma [12, 13]. To this end, researchers have identified stigma resistance as a potential line-of-defence against the internalization process. Rooted in the broader literature on resilience, stigma resistance describes the capacity to resist, counteract or otherwise remain unaffected by mental illness stigmatization [14]. In patients with schizophrenia, greater stigma resistance has been associated with reduced internalized stigma, reduced depression, greater selfesteem, greater empowerment, and improved quality of life [15, 16]. A similar pattern of results has been observed in patients with depression and bipolar disorder [17] and in patients with stigmatized biological diseases, including inflammatory bowel disease [18]. In addition, researchers have expressed optimism about the contribution of stigma resistance to recovery from mental illness [16, 17]. Tempering this enthusiasm, however, is the fact that no research has compared stigma resistance in people currently diagnosed with mental illness to those in recovery from mental illness. If people who had recovered from mental illness were shown to exhibit greater stigma resistance after controlling for expected differences in psychopathology, the evidence base suggesting that stigma resistance contributes to recovery from mental illness would be strengthened considerably. Stigma resistance might facilitate recovery from mental illness, including eating disorders, in several ways. According to the stress-coping model of stigma, an individual’s wellbeing is undermined when their perception of the harm due to stigma (the primary appraisal) exceeds their perceived ability to cope with the stigma (the secondary appraisal) [19, 20]. Stigma resistance may influence
123
the primary appraisal by leading to less harmful and/or less psychologically taxing appraisals of stigma. Alternatively, stigma resistance may act on the secondary appraisal by fortifying a person’s internal resources for coping with stigma, leading to more favourable perceptions of their ability to cope. Common to both explanations is that stigma resistance promotes and protects an individual’s psychological wellbeing, thus helping to facilitate recovery. Benefits to wellbeing may extend beyond eating disorder symptoms to symptoms of other mental illnesses, such as depression, and to self-esteem more generally. More positive appraisals related to stigma may also generalize to other forms of stigma, for example, the self-stigma associated with seeking psychological help, and may act as a buffer against the internalization of mental illness stigma into one’s self-identity. To our knowledge, no research has investigated the prevalence and correlates of stigma resistance as this applies to individuals with eating disorders or the potential role of stigma resistance in promoting recovery in this population. To address these issues, the present study investigated stigma resistance in people currently diagnosed with eating disorders and people recovered from eating disorders. Two major hypotheses were made: first, that stigma resistance would be associated with less marked eating disorder symptoms, less marked depression symptoms, more positive attitudes about seeking psychological treatment, higher self-esteem, and lower internalized stigma; and second, that stigma resistance would be greater in people that had recovered from eating disorders relative to people currently diagnosed with eating disorders, even after controlling for expected differences in eating disorder symptoms, depression symptoms, attitudes about seeking psychological treatment, and self-esteem.
Method Recruitment Participants were recruited via online and print advertisements circulated by various eating disorder organisations and community support groups, including major eating disorder charities, located in Australia, the United Kingdom, and the United States of America. The advertisements requested volunteers who were currently diagnosed with an eating disorder or in recovery from an eating disorder to participate in a brief online study about stigmatization. A link provided in the advertisement led participants to the survey, which began with the participant information statement and consent form, followed by a series of screening questions: ‘‘Have you ever been diagnosed with an eating disorder by a mental health professional, e.g., a
Soc Psychiatry Psychiatr Epidemiol
psychologist?’’, ‘‘Are you currently diagnosed with an eating disorder or are you recovered from an eating disorder?’’, ‘‘What eating disorder subtype are you currently diagnosed with?/What eating disorder subtype have you recovered from?’’ The survey then presented participants with the battery of psychiatric and psychosocial questionnaires. Once completed the survey could not be retaken. A total of 456 individuals completed the online survey and self-reported a current diagnosis of, or recovery from a diagnosis of, anorexia nervosa, bulimia nervosa, or an eating disorder not otherwise specified (EDNOS). All participants indicated that their current or former diagnoses were made by a mental health professional. Data integrity and exclusion All data were examined for duplicate responses, repeat Internet protocol addresses, and suspicious patterns of responding. In addition, three valid-responding checks were embedded in the study (e.g., For validity purposes, please select ‘‘Agree’’ as your answer to this question). Correct response rates to these questions have averaged 89 % in undergraduate samples [2, 21, 22]. The correct response rate in the present sample averaged 97.4 %, indicating that the vast majority of participants were responding meaningfully to the survey questions. Four participants who failed two or more validity checks had their data excluded. The final sample constituted 452 participants. Validity of the recovered group A purposeful decision was made not to provide participants with a definition of recovery because of widespread disagreement in the eating disorders field about what constitutes recovery from eating disorders [23, 24]. Instead, we devised four criteria that would provide support for the validity of the recovered group. These criteria were (a) significant and large differences (Cohen’s d [ 1.00) in eating disorder attitudes, as measured by the four subscales and the global score of the Eating Disorder ExaminationQuestionnaire (EDE-Q) [25], between the currently diagnosed and recovered groups, (b) significant differences in eating disorder behaviours, including purging, laxative use, diuretic use, objective binges, subjective binges, and hard exercise, as measured by the behavioural component of the EDE-Q, between the currently diagnosed and recovered groups, and (c) a median score of zero for each eating disorder behaviour for the recovered group. These three criteria were chosen because a marked reduction in eating disorder symptoms is common to all definitions of recovery from eating disorders [23, 24]. For anorexia nervosa we further specified that (d) the recovered group should report
a significantly higher body mass index (BMI) than currently diagnosed group. We did not specify an effect size for this criterion because BMI is a particularly contentious topic in discussions about recovery from anorexia nervosa [23]. Measures Stigma resistance and internalized stigma Stigma resistance and internalized stigma were measured using the 29-item, 5-subscale Internalized Stigma of Mental Illness (ISMI) scale [14]. The Stigma Resistance subscale consists of five reverse coded items: ‘‘In general, I am able to live life the way I want to,’’ ‘‘People with mental illness make important contributions to society,’’ ‘‘I can have a good fulfilling life, despite my mental illness,’’ ‘‘Living with mental illness has made me a tough survivor,’’ and ‘‘I feel comfortable being seen in public with an obviously mentally ill person.’’ Previous research has shown that the Stigma Resistance subscale is conceptually and psychometrically distinct from the four other subscales of the ISMI [16, 26, 27]. Consistent with previous research, the present study measured stigma resistance using the Stigma Resistance subscale and measured internalized stigma by summing the averages of the remaining four subscales of the ISMI [15–18]. These four subscales are Alienation, Stereotype Endorsement, Perceived Discrimination, and Social Withdrawal. Internalized stigma measured using the four subscales of the ISMI has demonstrated excellent internal consistency in previous research, with an average Cronbach’s a of 0.90 [26]. The Stigma Resistance subscale has been less internally consistent with an average Cronbach’s a of 0.64 [26], which is deemed acceptable by psychometric standards [28]. In the present study, Cronbach’s a for Internalized Stigma and Stigma Resistance was 0.90 and 0.65, respectively. Internalized stigma was measured in currently diagnosed participants only because the items that constitute this measure assume that the participant is currently suffering from a mental illness. Items on the Stigma Resistance subscale do not presuppose a current mental illness diagnosis and were administered to both the currently diagnosed and recovered group. Eating disorder symptoms The EDE-Q [25] measured disordered eating attitudes and behaviours. The attitudes component asks participants how frequently various eating attitudes have occurred during the past 28 days and contains four subscales: Dietary Restraint, Eating Concern, Shape Concern, and Weight
123
Soc Psychiatry Psychiatr Epidemiol
Concern. Responses are given using a 6-point scale (0 = no days, 6 = every day). In addition, mean scores on the four subscales were averaged together to calculate a global EDE-Q score. Cronbach’s a for the global EDE-Q was 0.96 in the present study. The behaviour component of the EDE-Q asks participants to rate how many times they have engaged in the following during the past 28 days: laxative use, diuretic use, purging, objective binges, subjective binges, and ‘‘hard’’ exercise. Cronbach’s a for eating disorder behaviours was 0.69 in the present study.
Table 1 Categorical data for currently diagnosed and recovered participants, including self-reported diagnosis, treatment status, sex, current residence, employment status, highest level of education, and relationship status Variable
The 10-item Self-Stigma of Seeking Help scale (SSOSH) was used to measure the existence of negative attitudes about seeking psychological treatment [29]. The SSOSH has demonstrated predictive validity insofar as predicting which undergraduates sought help from a mental health professional over a 2-month period [29]. Internal consistency of the SSOSH among undergraduates is acceptable at 0.78 and test-reliability is acceptable at 0.82 over a 3-week period. Cronbach’s a for the SSOSH was 0.88 in the present study. Higher scores on this measure indicate more negative attitudes towards help-seeking.
Anorexia nervosa Bulimia nervosa
The 7-item Depression subscale of the 21-item Depression Anxiety Stress Scales (DASS-21) was used to measure depression symptoms [30]. The Depression subscale has demonstrated concurrent validity with other measures of depression and acceptable internal consistency (a = 0.88) in a general population sample of adults [30]. Cronbach’s a was 0.94 in the present study. Self-esteem The 10-item Self-Esteem Scale (SES) was used to measure global self-esteem [31]. Cronbach’s a was 0.89 in the present study. Statistical analyses
123
n
n
df
%
%
2 51.7 21.2
82 17
64.6 13.4
88
27.1
28
22.0
Currently in treatment for eating disorder
183
56.3
17
13.4
Not currently in treatment for eating disorder
133
40.9
108
85.0
9
2.8
2
1.6
13
4.0
2
1.6
309
95.1
123
96.9
3
0.9
2
1.6
98
30.2
29
22.8
110
33.8
40
31.5
Treatment status
Other
2
Sex Male Female Transgender Current residence United States of America United Kingdom
65
19.7
28
22.0
Other (11 countries total)
53
16.3
30
23.6
Paid employment
116
35.7
46
36.2
Student
157
48.3
67
52.8
52
16.0
14
11.0
Employment status
Other Highest level of education High school/ secondary school
81
24.9
20
15.7
University/college
216
66.5
94
74.0
28
8.6
13
10.2
Single
206
63.3
57
44.9
In a relationship
111
34.2
67
52.8
8
2.5
3
2.4
Other Relationship status
Other
All statistical analyses were conducted using SPSS version 21. To examine differences between the currently diagnosed and recovered group for categorical variables such as sex, a series of Chi-square tests of independence were conducted with the categorical variable entered as the first independent variable and diagnostic status (currently diagnosed or recovered) entered as the second independent variable. The results of these analyses are shown in Table 1. To examine differences between the currently
Statistical comparison
168 69
Australia
Depression symptoms
Recovered
Self-reported diagnosis
EDNOS
Attitudes about seeking psychological help
Currently diagnosed
v2 6.59
p 0.037
71.89 \0.001
2
2.00
0.368
3
4.80
0.187
2
1.92
0.383
2
4.46
0.107
2
13.40
0.001
diagnosed and recovered group for continuous variables such as age, a series of univariate ANOVAs were conducted with the continuous variable entered as the dependent variable and diagnostic status (currently diagnosed or recovered) entered as the independent variable. The results of these analyses are shown in Table 2.
Soc Psychiatry Psychiatr Epidemiol Table 2 Continuous data for currently diagnosed and recovered participants, including age, BMI, eating disorder symptomatology, depression symptomatology, self-esteem, attitudes about seeking help, internalized stigma, and years since symptom onset, since diagnosis, and since recovery
Variable
Age
Currently diagnosed
Recovered
Mean
Median
Mean
SD
Median
F (df) or U
SD
Statistical comparison p
24.61
7.14
23
25.30
8.02
23
U = 20,113.50
0.67
Anorexia nervosa
17.78
3.63
17.40
21.95
7.82
20.31
F(1, 237) = 46.83
\0.001
Bulimia nervosa and EDNOS
23.22
8.25
20.91
24.56
6.57
22.66
F(1, 184) = 0.80
Body mass index
0.373
Eating Disorders Examination Questionnaire Dietary restraint
4.27
1.42
4.60
1.89
1.62
1.60
F(1, \0.001 450) = 236.35
Eating concern
3.98
1.23
4.20
1.72
1.44
1.60
F(1, \0.001 450) = 279.09
Shape concern
5.11
1.09
5.50
3.21
1.76
3.25
F(1, \0.001 450) = 191.74
Weight concern
4.70
1.21
4.80
2.79
1.62
2.80
F(1, \0.001 450) = 184.44
Global score
4.51
1.06
4.80
2.41
1.45
2.31
F(1, \0.001 450) = 290.36
Laxative use
2.91
10.75
0
0.38
1.86
0
U = 11,409.50
\0.001
Diuretic use
0.93
5.28
0
0.07
0.38
0
U = 16,361.50
\0.001
Exercised hard Objective binge
7.70 7.51
9.68 16.46
3 0
3.35 1.55
6.69 4.82
0 0
U = 19,172.00 U = 14,358.50
0.024 \0.001
Subjective binge
11.56
14.77
10
4.434
6.83
0
U = 13,829.00
\0.001
Purging
11.53
22.84
1
0.81
3.79
0
U = 11,641.00
\0.001
13.08
5.57
14.00
7.20
5.68
7.00
F(1, 442) = 99.91
\0.001
Global self-esteem
9.16
5.07
9.00
15.04
5.80
15.00
Negative attitudes about seeking help
2.81
0.77
2.70
2.61
0.79
2.50
F(1, 450) = 6.19
0.013
Internalized stigma
2.51
0.48
2.52
Years between symptom onset and diagnosis
9.59
7.27
9.00
10.42
8.03
8
U = 16,565.00
0.005
Years since diagnosis
5.79
6.02
4.00
U = 16,361.50
0.001
Depression symptoms
Body mass index for participants with anorexia nervosa is given separately from participants with bulimia nervosa and EDNOS. Statistical comparisons were either ANOVA F tests or Mann– Whitney U tests
Years spent in recovery
To determine the relative amounts of stigma resistance exhibited by individuals currently diagnosed and recovered from eating disorders, each participant was assigned a level of stigma resistance according to their mean score on the Stigma Resistance subscale. The cut-offs for each level of stigma resistance were taken from previous research [15, 17]: Minimal (\2), Low (2—2.5), Moderate (2.5–3), High (3?). A Pearson correlational analysis was conducted to determine the relationships between stigma resistance, internalized stigma, eating disorder symptoms, depression symptoms, negative attitudes about seeking psychological help, and self-esteem. The role of stigma resistance in eating disorder recovery was examined by conducting a univariate ANCOVA with
7.44
7.01
6
2.63
3.33
1
F(1, \0.001 445) = 112.56
stigma resistance entered as the dependent variable, diagnostic status entered as the independent variable, and eating disorder symptoms, depression symptoms, attitudes about seeking psychological help, self-esteem, years between symptom onset and diagnosis, and years since diagnosis entered as covariates. The latter two variables were included as covariates because the results of the analyses shown in Table 2 indicated significant differences between the currently diagnosed and recovered group for these variables. Finally, the relationships between stigma resistance and the demographic and clinical variables included in the study were examined by conducting a series of univariate ANOVAs with diagnostic status (currently diagnosed versus recovered) entered as the independent variable. All
123
Soc Psychiatry Psychiatr Epidemiol
nervosa was significantly higher, for participants in the recovered group versus the currently diagnosed group. All four criteria that were devised to support the validity of the recovered group were satisfied by the data. First, scores on the four subscales and the global score of the EDE-Q were significantly lower for recovered participants compared to currently diagnosed participants and these differences were very large (Cohen’s d = 1.28–1.61). Global EDE-Q scores for the recovered group (M = 2.41) were higher than population norms for young adult women living in Australia and America (M = 1.52–1.74) [33, 34] but considerably lower than the cut-off ([4.0) for clinical significance [33, 34]. Second, the recovered group reported significantly fewer eating disorder behaviours across each of the seven behaviours assessed. Third, the median score in the recovered group for each eating disorder behaviour was zero. Fourth, participants that had recovered from anorexia nervosa had significantly higher BMIs than currently diagnosed participants and the size of this effect was large (Cohen’s d = 0.89). The ANCOVA conducted to examine the role of stigma resistance in eating disorder recovery revealed that stigma resistance was significantly greater in people who reported being recovered from an eating disorder (covariate-adjusted mean = 3.03, SD = 0.43) than in people who reported being currently diagnosed with an eating disorder (covariate-adjusted mean = 2.87, SD = 0.51), F(1, 425) = 4.26, p = 0.006. The size of the effect was small (Cohen’s d = 0.25). Table 3 shows the proportion of participants in the currently diagnosed and recovered group who fell into each level of stigma resistance. The subsequent ANOVAs conducted to explore the relationships between stigma resistance and the demographic and clinical variables shown in Table 1 revealed that stigma resistance did not significantly differ as a function of eating disorder subtype, treatment status, employment status, or highest level of education, but did differ significantly as a function of current residence, F(3, 447) = 4.78, p = 0.031, and relationship status,
significant between-groups differences indicated by the ANOVAs were followed-up by planned contrasts using the Holm–Bonferroni method to control the family-wise error rate at 0.05. Statistical assumptions Dependent variables that were slated for ANOVA had their data distributions tested for normality. Normality was established if (a) the absolute values of the skewness and kurtosis statistic were less than twice the standard error of the respective statistics [32], and if (b) visual inspection of the frequency histogram and normal Q–Q plot corroborated the skewness and kurtosis statistics. Variables deemed to be non-normal were instead analysed using non-parametric Mann–Whitney U tests. The variable of primary interest, stigma resistance, met both criteria for normality. Ethical standards All participants gave their informed consent prior to inclusion in the study. The study was approved by the Human Research and Ethics Committee of the University of Sydney.
Results Descriptive data and the outcomes of the Chi-square tests of independence are shown in Table 1 and descriptive data and the outcomes of the ANOVA and Mann–Whitney tests are shown in Table 2. As shown in Table 1, there were no significant differences between the diagnosed and recovered groups in self-reported sex, current residence, employment status, and highest level of education, while significant differences did emerge between the groups for self-reported diagnosis, treatment status, and marital status. As shown in Table 2, eating disorder symptomatology was significantly lower, and BMI for participants with anorexia
Table 3 Number (n) and proportion (%) of participants with various levels of stigma resistance Stigma resistance
Anorexia nervosa
Bulimia nervosa
EDNOS
All subtypes
Currently diagnosed
Recovered
Currently diagnosed
Recovered
Currently diagnosed
Recovered
Currently diagnosed
Recovered
n
n
n
n
n
n
n
n
%
%
%
%
%
%
%
%
Minimal (\2)
6
3.6
0
0.0
2
2.9
0
0.0
3
3.4
0
0.0
11
3.4
0
0.0
Low (2–2.5)
35
20.8
4
4.9
16
23.2
0
0.0
13
14.8
3
10.7
64
23.1
7
5.5
Moderate (2.5–3) High (3?)
85 42
50.6 25.0
34 44
41.5 53.7
37 14
53.6 20.3
6 11
35.3 64.7
45 27
51.1 30.7
7 18
25.0 64.3
167 83
51.4 25.5
47 73
37.0 57.5
Data are given for each eating disorder subtype and for all eating disorder subtypes combined, and for currently diagnosed participants and recovered participants
123
Soc Psychiatry Psychiatr Epidemiol Table 4 Correlations between stigma resistance and eating disorder symptoms, depression symptoms, self-esteem, attitudes about seeking psychological treatment, and internalized stigma for currently diagnosed participants and recovered participants Variable
Correlation with stigma resistance Currently diagnosed
Recovered
N
r
N
r
Global EDE-Q score
325
-0.26 \0.001
127
-0.11
DASS-21 depression subscale score
317
-0.36 \0.001
127
-0.29 \0.001
Global self-esteem
320
0.36 \0.001
127
Attitudes about seeking psychological treatment
325
-0.29 \0.001
127
Internalized stigma
325
-0.40 \0.001
p
p 0.205
0.35 \0.001 -0.24
0.008
F(2, 448) = 5.34, p = 0.005. The follow-up contrast analyses suggested that stigma resistance was significantly higher among participants who were residing in the USA compared to the remaining countries, F(1, 447) = 12.69, p \ 0.001. Interestingly, stigma resistance was significantly higher in the minority of participants who reported their family status as ‘‘other’’ compared to participants who reported that they were single, F(1, 448) = 6.07, p = 0.014, and compared to participants who reported being in a relationship, F(1, 448) = 9.09, p = 0.003. The results of the correlational analysis are shown in Table 4. Stigma resistance was associated with less marked eating disorder symptoms, less marked depression symptoms, more positive attitudes about seeking psychological treatment, and higher self-esteem. Among participants who were currently diagnosed, stigma resistance was associated with lower internalized stigma.
Discussion The present study investigated stigma resistance, the capacity to resist or be unaffected by mental illness stigmatization, in people who self-reported a current diagnosis, or recovery from a diagnosis, of anorexia nervosa, bulimia nervosa, or EDNOS. The results revealed that 26.5 % of currently diagnosed participants had low-to-minimal stigma resistance. In comparison, a study of people with self-reported depression or bipolar disorder found that 32 % had low-to-minimal stigma resistance [17], whilst 50.7 % of people with self-reported schizophrenia exhibited low-to-minimal stigma resistance [15]. In the present study there were no significant differences in stigma
resistance between the eating disorder subtypes. However, low-to-minimal stigma resistance was exhibited by just 5.5 % of people in recovery from an eating disorder, implying that stigma resistance may play a role in the recovery process. Both major hypotheses of the present study were supported. Among both the currently diagnosed and the recovered, stigma resistance was significantly associated with less marked eating disorder symptoms, less marked depression symptoms, more positive attitudes about seeking psychological treatment and higher self-esteem. In addition, for currently diagnosed participants, greater stigma resistance was associated with lower internalized stigma. Moreover, stigma resistance discriminated participants currently diagnosed with an eating disorder from participants recovered from an eating disorder, even after controlling for differences in eating disorder and depression symptoms, attitudes about seeking psychological help, self-esteem, years between symptom onset and diagnosis, and years since diagnosis. Collectively, the results suggest that stigma resistance plays an important and positive role in people with eating disorders. The primary mechanism by which stigma resistance achieves positive outcomes in people with eating disorders may be through disruption of the link between enacted stigma and internalized stigma. Stigma resistance may lead to less harmful appraisals of stigma or it may bolster an individual’s ability to cope with stigma, leading to a more favourable appraisal of their ability to cope [19, 20]. In turn, these favourable appraisals may reduce the internalization of stigma, promoting wellbeing and facilitating recovery. These positive appraisals may also generalize to other forms of stigma, for example, the self-stigma associated with seeking psychological help, which may explain why greater stigma resistance was associated with more positive attitudes about mental health care in the present study. The notion that more positive appraisals protect and promote an individual’s psychological wellbeing was supported by the positive interrelationships we observed between stigma resistance, depression symptoms, eating disorder symptoms, and self-esteem. It is important, however, to acknowledge the possibility that the process of recovery itself is what facilitates the development of stigma resistance. Individuals who receive psychotherapy for an eating disorder may develop cognitive strategies that allow them to mount a superior resistance to stigma, implying that stigma resistance is a helpful by-product of the recovery process. Additional research exploring these mechanisms is needed. Stigma resistance demonstrates promise for future interventions into eating disorder stigmatization and for clinicians who treat eating disorders. Because an intervention to raise stigma resistance would be likely to target
123
Soc Psychiatry Psychiatr Epidemiol
the stigmatized individual, it can be viewed as complementary to recent efforts that target public perceptions of stigma resistance with the aim of reducing enacted stigma [7]. A two-sided approach that targets both enacted and internalized stigma may provide a more comprehensive and effective strategy to reduce stigmatization of eating disorders. An additional benefit of an intervention that targets stigma resistance in the individual rather than enacted stigma among the public may be the restoration of agency to the sufferer, with potential benefits to empowerment, hope, and self-esteem. Future research is needed to explore these possibilities. Clinicians may also benefit from cultivating stigma resistance in their practice, insofar as it equips their patients with the skills needed to combat the widespread prevalence of eating disorder stigmatization. Several strengths and limitations of the present study are noted. First, the internal consistency of the Stigma Resistance subscale in the present study was relatively low (a = 0.65). Although this figure is higher than in previous research [15–17] and is acceptable by published psychometric standards [28], there is scope to improve the psychometric properties of the subscale, or to develop an independent instrument [16]. Second, the sample may not be fully representative of the broader population of eating disorders patients. In particular, patients who are acutely ill, patients without Internet access, and patients unaffiliated with eating disorder organisations may have been underrepresented. Third, there was potential for inaccuracy in participants’ self-reported diagnoses and self-reported recovery status. For example, some participants who identified as recovered may have very recently completed a course of inpatient treatment, making their status as ‘‘recovered’’ a matter for debate. Fourth, although attitudes about help-seeking were controlled for in comparisons of currently diagnosed versus recovered patients, data bearing on differences in the type and length of treatment actually received were not collected. Fifth, the cross-sectional design of the study precludes any conclusions about causality. Particular strengths of the study include the recruitment of a large, geographically diverse sample of individuals with current or past eating disorders and the novel investigation of stigma resistance in this population. We offer several suggestions for future research. The literature on stigma resistance and eating disorders may benefit from the development of a stigma resistance instrument that is specific to eating disorders and which is populated by items that assess an individual’s ability to resist the specific types of stigma directed toward eating disorders. For example, one in three members of the general public stigmatize individuals with eating disorders as ‘‘having only themselves to blame’’ and being ‘‘able to pull themselves together’’, but less than one in ten members of the public stigmatize schizophrenia in this way [4].
123
Moreover, we stress the importance of defining and operationalising ‘‘recovery’’ in future research that uses a similar method to the present study. The lack of consensus surrounding definitions of recovery is not confined to eating disorders [35] and the ability of future research to make valid comparisons between studies will depend on researchers being explicit about their definitions and operationalisations of recovery. In conclusion, the present study found support for two major hypotheses about stigma resistance: first, that stigma resistance would be associated with positive psychosocial and psychiatric outcomes in people currently diagnosed and recovered from eating disorders, and second, that stigma resistance would discriminate between the currently diagnosed and the recovered after controlling for expected differences in psychopathology. Stigma resistance may bolster current intervention strategies that focus primarily on enacted stigma and which neglect internalized stigma. Clinicians may also benefit by fostering stigma resistance in their patient. Conflict of interest On behalf of all authors the corresponding author states that there is no conflict of interest.
References 1. Fleming J, Szmukler GI (1992) Attitudes of medical professionals towards patients with eating disorders. Aust N Z J Psychiatry 26:436–443. doi:10.3109/00048679209072067 2. Griffiths S, Mond JM, Murray SB, Touyz S (2013) Young peoples’ stigmatizing attitudes and beliefs about anorexia nervosa and muscle dysmorphia. Int J Eat Disord 47:189–195. doi:10. 1002/eat.22220 3. Mond J, Marks P (2007) Beliefs of adolescent girls concerning the severity and prevalence of bulimia nervosa. Aust J Psychol 59:87–93. doi:10.1080/00049530601148413 4. Crisp AH, Gelder MG, Riz S, Meltzer HI, Rowlands OJ (2000) Stigmatisation of people with mental illnesses. Br J Psychiatry 177:4–7. doi:10.1192/bjp.177.1.4 5. Robinson KJ, Mountford VA, Sperlinger D (2013) Being men with eating disorders: perspectives of male eating disorder service-users. J Health Psychol 18:176–186. doi:10.1177/1359105 312440298 6. Crisafulli MA, Von Holle A, Bulik C (2008) Attitudes towards anorexia nervosa: the impact of framing on blame and stigma. Int J Eat Disord 41:333–339. doi:10.1002/eat.20507 7. Zwickert K, Rieger E (2013) Stigmatizing attitudes towards individuals with anorexia nervosa: an investigation of attribution theory. J Eat Disord 1:5. doi:10.1186/2050-2974-1-5 8. Link BG, Yang LH, Phelan JC, Collins PY (2004) Measuring mental illness stigma. Schizophr Bull 30:511–541 9. Yanos PT, Roe D, Markus K, Lysaker PH (2008) Pathways between internalized stigma and outcomes related to recovery in schizophrenia spectrum disorders. Psychiatr Services 59: 1437–1442. doi:10.1176/appi.ps.59.12.1437 10. Livingston JD, Boyd JE (2010) Correlates and consequences of internalized stigma for people living with mental illness: a systematic review and meta-analysis. Soc Sci Med 71:2150–2161. doi:10.1016/j.socscimed.2010.09.030
Soc Psychiatry Psychiatr Epidemiol 11. Thornicroft G, Rose D, Kassam A, Sartorius N (2007) Stigma: ignorance, prejudice or discrimination? Br J Psychiatry 190: 192–193. doi:10.1192/bjp.bp.106.025791 12. Lasalvia A, Zoppei S, Bortel TV, Bonetto C, Cristofalo D, Wahlbeck K, Thornicroft G (2013) Global pattern of experienced and anticipated discrimination reported by people with major depressive disorder: a cross-sectional survey. Lancet 381:52–55. doi:10.1016/S0140-6736(12)61379-8 13. Thornicroft G, Brohan E, Rose D, Sartorius N, Leese M (2013) Global pattern of experienced and anticipated discrimination against people with schizophrenia: a cross-sectional survey. Lancet 373:408–415. doi:10.1016/S0140-6736(08)61817-6 14. Ritsher JB, Otilingam PG, Grajales M (2003) Internalized stigma of mental illness: psychometric properties of a new measure. Psychiatry Res 121:31–49. doi:10.1016/j.psychres.2003.08.008 15. Brohan E, Elgie R, Sartorius N, Thornicroft G (2010) Selfstigma, empowerment and perceived discrimination among people with schizophrenia in 14 European countries: the GAMIANEurope study. Schizophr Res 122:232–238. doi:10.1016/j.schres. 2010.02.1065 16. Sibitz I, Unger A, Woppmann A, Zidek T, Amering M (2009) Stigma resistance in patients with schizophrenia. Schizophr Bull 37:316–323. doi:10.1093/schbul/sbp048 17. Brohan E, Gauci D, Sartorius N, Thornicroft G (2011) Selfstigma, empowerment and perceived discrimination among people with bipolar disorder or depression in 13 European countries: the GAMIAN-Europe study. J Affect Disord 129:56–63. doi:10. 1016/j.jad.2010.09.001 18. Taft TH, Ballou S, Keefer L (2013) A preliminary evaluation of internalized stigma and stigma resistance in inflammatory bowel disease. J Health Psychol 18:451–460. doi:10.1177/135910531 2446768 19. Ru¨sch N, Corrigan PW, Heekeren K, Theodoridou A, Dvorsky D, Metzler S, Ro¨ssler W (2014) Well-being among persons at risk of psychosis: the role of self-labelling, shame, and stigma-stress. Psychiatr Services 65:483–489 20. Kaiser CR, Major B, Mccoy SK (2004) Expectations about the future and the emotional consequences of perceiving prejudice. Pers Soc Psychol Bull 30:173–184. doi:10.1177/014616720 3259927 21. Griffiths S, Angus D, Murray SB, Touyz S (2014) Unique associations between young adult men’s emotional functioning and their body dissatisfaction and disordered eating. Body Image 11:175–178. doi:10.1016/j.bodyim.2013.12.002
22. Griffiths S, Murray SB, Touyz S (2013) Drive for muscularity and muscularity-oriented disordered eating in men: the role of set shifting difficulties and weak central coherence. Body Image 10:636–639. doi:10.1016/j.bodyim.2013.04.002 23. Dawson L, Rhodes P, Touyz S (2014) Defining recovery from anorexia nervosa: a Delphi study to explore expert practitioners’ views (submitted) 24. Yu J, Agras SW, Bryson S (2013) Defining recovery in adult bulimia nervosa. Eat Disord 21:374–379. doi:10.1080/10640266. 2013.827536 25. Fairburn CG, Beglin SJ (1994) Assessment of eating disorders: interview or self-report questionnaire? Int J Eat Disord 16: 363–370. doi:10.1002/1098-108X(199412)16:4\363::AID-EAT 2260160405[3.0.CO;2 26. Boyd JE, Adler EP, Otilingam PG, Peters T (2014) Internalized Stigma of Mental Illness (ISMI) Scale: a multinational review. Compr Psychiatry 55:221–231. doi:10.1016/j.comppsych.2013. 06.005 27. Lysaker PH, Roe D, Yanos PT (2007) Toward understanding the insight paradox: internalized stigma moderates the association between insight and social functioning, hope, and self-esteem among people with schizophrenia spectrum disorders. Schizophr Bull 33:192–199. doi:10.1093/schbul/sbl016 28. Kline P (2000) The handbook of psychological testing, 2nd edn. Routledge, London 29. Vogel DL, Wade NG, Haake S (2006) Measuring the self-stigma associated with seeking psychological help. J Counsel Psychol 53(3):325–337. doi:10.1037/0022-0167.53.3.325 30. Henry J, Crawford J (2005) The short-form version of the Depression Anxiety Stress Scales (DASS-21): construct validity and normative data in a large non-clinical sample. Br J Clin Psychol 44:227–239 31. Rosenberg M (1965) Society and the adolescent self-image. Princeton University Press, Princeton 32. Sheskin DJ (2000) Handbook of parametric and nonparametric statistical procedures, 2nd edn. Chapman & Hall/CRC, New York 33. Luce KH, Crowther JH, Pole M (2008) Eating Disorder Examination Questionnaire (EDE-Q): norms for undergraduate women. Int J Eat Disord 41:273–276. doi:10.1002/eat.20504 34. Mond JM, Hay PJ, Rodgers B, Owen C (2006) Eating Disorder Examination Questionnaire (EDE-Q): norms for young adult women. Behav Res Therapy 44:53–62. doi:10.1016/j.brat.2004.12.003 35. Frank E (2005) Describing course of illness: does our language matter? Int J Eat Disord 38:7–8. doi:10.1002/eat.20145
123
