Experienced Stigma in Severe Mental Illness

Published Date: 29 Mar 2018

Exploring experienced stigma in severe mental illness – contributing to validation of a psychometric instrument

Luís Pedro Santos de Mendonça

Table of contents (Jump to)

Acknowledgements 

Acronyms

1 Introduction

1.1 About stigma

• 1.1.1 Why to focus on stigma?
• 1.1.2 Evolution of the concept
• 1.1.3 Development of stigma
• 1.1.4 Different concepts of stigma
• 1.1.5 Correlates and consequences of stigma

1.2 Stigma research

• 1.2.1 Instruments to measure stigma - categories and criteria for psychometric properties
• 1.2.2 Perceived stigma
• 1.2.3 Self-Stigma
• 1.2.4 Experienced stigma
• 1.2.5 Stigma studies in Portugal

1.3 Consumer experiences of stigma questionnaire (CESQ)

• 1.3.1 Main research with psychometric data involving CESQ.

Acronyms

CASS - Clinician Assessment of Schizophrenic Syndromes

CAT - Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment

CESQ – Consumer Experiences of Stigma Questionnaire

CFA – Confirmatory Factor Analysis

CI – Confidence interval

CRPD - Convention on the Rights of People with Disabilities

DISC – Discrimination and Stigma Scale

DSSS – Depression Self-stigma Scale

EDS – Experiences of Discrimination Scale

EFA – Exploratory Factor Analysis

FBS - Frankfurter Befindlichkeits-Skala

GAF - Global Assessment of Functioning

GAS - Global Assessment Scale

HIV/AIDS – Human Immunodeficiency Virus / Acquired Immunodeficiency Syndrome

HSRS - Health Sickness Rating Scale

HSS – Stigmatisation Scale

ICCPR - International Covenant on Civil and Political Rights

ICD – International Classification of Diseases

ISE – The Inventory of Stigmatising Experiences

ISMI – Internalised Stigma of Mental Illness

KMO – Kaiser-Meyer Olkin statistic

M – Mean

MIDUS – MacArthur Foundation Midlife Development in the United States

MSA – Measures of sampling adequacy

MSS – Maristan Stigma Scale

NAMI - National Aliiance for Mentally Ill

PA – Parallel Analysis

PAF – Principal Axis Factoring

PANSS – Positive and Negative Syndrome Scale

PCM – Polychoric correlation matrix

PDD – Perceived devaluation and discrimination scale

PD-S - Paranoid-Depresivitäts-Skala

QOLI – Quality of Life Interview

RES – Rejection Experiences Scale

RMSEA - Root mean square error of approximation

SD – Standard deviation

SESQ – Self-esteem and Stigma Questionnaire

SFS – Social Functioning Scale

SLDS – Satisfaction with Life Domains Scale

SRER – Self Reported Experiences of Rejection

SS – Stigma Scale

SSMIS – Self-stigma of Mental Illness Scale

UDHR - Universal Declaration of Human Rights

WHO – World Health Organization

WLSMV - Means and Variance adjusted weighted least square

1 Introduction

1.1 About stigma

1.1.1 Why to focus on stigma?

Stigma is defined as a sign of disgrace or discredit. Authors agree it is a powerful negative attribute, having its impact on all social relations.

Stigma is present everywhere in our society. It affects different characteristics in people, ranging from sexual orientation to HIV/AIDS, several medical disorders, gender, race, unemployment or obesity. However, it is in mental health disorders that stigma has its most devastating impact, although not always obvious.

Discrimination, the enactment of stigma, appears closely associated to it. While stigma lies at the base of discrimination, discriminatory practices also promote and reinforce stigma. Discrimination is also about the conditions in which patients live, mental health budgets and the priority which we allow these services to achieve.¹ In other words, stigma and discrimination lead to social exclusion – a triad that is a key determinant of mental health.

Stigma and discrimination are violations of human rights. Intention and commitment to fight stigma are present in the spirit of legally binding treaties such as the Universal Declaration of Human Rights (UDHR)² , International Covenant on Civil and Political Rights (ICCPR)³, International Covenant on Economic, Social and Cultural Rights (ICESCR)⁴ and Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (CAT)⁵, and are explicitly mentioned on the Convention on the Rights of People with Disabilities (CRPD)⁶.

CRPD actually demands that signatories ‘take all appropriate measures to eliminate discrimination on the basis of disability by any person, organisation or private enterprise’, and to ‘adopt immediate, effective and appropriate measures … to combat stereotypes, prejudices and harmful practices relating to persons with disabilities … in all areas of life’.⁶

From the part of the World Health Organization, tackling stigma, discrimination and social exclusion is a major concern of the General Assembly, with of the General Assembly, with reflection in the WHO Mental Health Action Plan 2013-2020⁷.

At regional level, in European Union, commitment to fighting stigma and discrimination is a consequence of signing treaties like European Convention on Human Rights, European Social Charter, European Convention on the Prevention of Torture and Inhuman or Degrading Treatment or Punishment and, specifically, Recommendation Rec(2004)10, of the Committee of Ministers to member states, concerning the protection of the human rights and dignity of persons with mental disorder.

Still at regional level, and in line with WHO Mental Health Action Plan, stigma and discrimination is one of the main action areas of European Mental Health Action Plan.⁸

At national level, fighting stigma, discrimination and social exclusion is a component of policies, plans and programs worldwide.

In a time when quality mechanisms tend to be implemented into healthcare systems, there is also a trend to develop parts of quality standards that have statements on fighting stigma at a local level. NICE quality standards are a good example⁹. To implement stigma into quality standards is, by itself, a strategy to fight it, by turning each service user in a potential advocate, as Byrne noted¹.

Therefore, there is the need to foster development of indicators that can be used regarding mental illness stigma.

1.1.2 Evolution of the concept

Stigma is a word that has its reminiscences in the Greek civilization. Stigma were body marks that were intentionally applied to individuals- the stigmatized - that carried unacceptable moral or individual traits, as compared to standards in that society. Christians absorbed the concept, adding two other meanings to those body marks – to indicate a holy grace or to indicate a sign of deformity/physical disease.

Anyway, even in early days of Christianity, stigma implied, from the social point of view, firstly, “imputing a meaning into something” even if it did not have that meaning, and, secondly, dealing with deviations to a social norm.

Goffmann¹⁰ was the first author to theorize stigma. To Goffmann, stigma is the result of a gap between perceived attributes and stereotypes. It is a matter of perspective, not reality. it is “in the eye of the beholder”. Stereotypes are selective perceptions that categorize people, and that exaggerate differences between groups (‘them and us’) in order to obscure differences within groups.¹¹

He defines three types of attributes:

• Body(physical) – e.g. visible deformities in the body, deformity caused by physical disease
• Character (personal) – e.g. mental illness, criminal conviction
• Tribal (Social) – e.g. stigma of one group against another.

Goffmann also distinguishes between “discredited” and discreditable”. Those concepts were further developed by Jones et al.¹² , who proposed six dimensions of stigma:

• Concealability indicates how obvious or detectable the characteristic is to others.
• Course indicates whether the stigmatizing condition is reversible over time. Irreversible conditions provoke more negative attitudes than others.
• Disruptiveness indicates the extent to which a mark blocks or diminishes interpersonal interactions.
• Aesthetics reflects what is attractive or pleasing to one's perceptions. When applied to stigma, it means whether a mark provokes a reaction of disgust.
• Origin refers to how the condition came into being. Perceived responsibility on the conditions will carry more negative attitude.
• Peril, refers to feelings of danger or threat induced in others. This can mean physical threat (as in “contamination”) or simply uneaseness.

According to Byrne, stigma is connoted with a few negative attributes. Shame is its first expression, resulting from perception as indulgence or as a weakness, despite centuries of knowledge, media campaigns and “the decade of the brain”. Blame is also an attitude that appears associated to shame.¹ Maintaining secrecy is the maladaptive way some people find to cope with shame, but it can lead to deleterious consequences.

1.1.3 Development of stigma

Negative attitudes towards people with mental illness, according to Byrne ¹, exist since playschool and extend into early adulthood. This is suggested by several studies: Weiss¹³ examined a cohort of children of elementary school age and confirmed the prejudices eight years after; Green¹⁴ compared attitudes between several studies using the same measures, that ranged over 22 years, and found consistent results indicating community had the same negative attitudes. This objects the common belief that with increased scientific knowledge about mental illness, stigma would tend to disappear.

1.1.4 Different concepts of stigma

Stigma concept has evolved in the last fifteen years.

Link and Phelan have added discrimination to Jones’ original dimensions.¹⁵ Still, in 2001 the same authors present two major challenges for the concept of stigma.

The first challenge is that researchers who research stigma do so from their own vantage point, giving priority to their scientific theories and research techniques rather than words and perceptions about people they study, which lead to misunderstanding of the experience of people being stigmatized and to perpetuation of assumptions that are unsubstantiated.

The second challenge is about individualization of stigma and the fact that in research it tends to be considered as an attribute or a mark of the individual rather than a designation or tag that others affix to a person.

Thus, Link and Phelan propose a definition of stigma based on a convergence a few components:

• Distinguishing and labelling human differences – oversimplification of salient differences between human beings occurs, with further labelling of individuals.
• Associating human differences to negative attributes – Labels previously mentioned are associated to negative sterotypes, as previously described by Goffmann. Categories and stereotypes are often “automatic” and facilitate “cognitive efficiency”.
• Separating “us from them” – Social labels connote a separation between the group that stigmatizes (“us”) and the group that is being stigmatized (“them”). For example, some people talk about people who have schizophrenia as being “schizophrenics”.
• Status loss and discrimination – stigma leads to loss of status in social hierarchy, and to discrimination, both at individual and at structural levels.

Link and Phelan also emphasize that stigma is a matter of power – certain groups in the society have the power to stigmatize. Stigma is also a matter of degree – there is a continuum between its existence and its absence.

Corrigan¹⁶, has an opposing view, focused on cognitive and behaviour features of mental illness. He proposed a model in which stigma was categorized either as public or self stigma.

Public stigma is defined as the reaction that the general population has to people with mental illness. Self stigma is the prejudice which people with mental illness turn against themselves.

In each of the categories, stigma is broken down into three elements: stereotypes (cognitive knowledge structures) prejudice (cognitive and emotional consequence of stereotypes) and discrimination (behavioural consequence of prejudice)¹⁷ .

Thornicroft et al.¹⁸, elaborate on this framework, stating that stigma is composed of problems at three levels: Knowledge, Attitudes and Behaviour.

Mental health knowledge is also known in the literature as mental health literacy. A study by Jorm et al. in Australia has shown better knowledge was correlated with better recognition of the features of depression, and better compliance with help seeking or medication and/or psychotherapy compliance.¹⁹ Nevertheless, by citing contradicting evidence, Thornicroft¹⁸ states that “an increase in knowledge about mental illness does not necessarily improve either attitudes or behaviour towards people with mental illness.”

Negative attitudes, also known as prejudice, is the most studied component. According to Thornicroft, it can predict more strongly actual discrimination than do stereotypes. Attitudes have been widely researched. There are studies regarding both public, healthcare practitioners (and medical students) and caregivers.

Thornicroft emphasizes the importance of studying actual behaviour, stressing that most of the studies have focused on attitudes towards hypothetical situations, rather than actual stigmatizing and discriminative behaviour. Thornicroft proposes a shift from research focused on stigma to research focused on discrimination.¹⁸

1.1.5 Correlates and consequences of stigma

Stigma can have profound impact both at individuals with mental illness and their relatives.

Rüsch et al.¹⁷ list four negative consequences of public stigma:

• Everyday life discriminations encountered in interpersonal relations and depictions in media
• Structural discrimination – inequity in the access to opportunities in private and public institutions.
• Self-stigma (versus empowerment)
• Fear of stigma as a barrier to use health services.

About self-stigma and empowerment, Rüsch et al. comment, firstly, that self-stigma and empowerment are on the same continuum of self-esteem. They also remark that people may have different reactions to public stigma – while some people react with low self-esteem (self-stigmatized), some people might react with anger or indifference. They point out a possible explanation for this resides both within group identification with public stigma and perceived legitimacy of it. They also point the issue of self-disclosure – a person who considers mental illness is a part of his/her identity will more likely reveal his/her condition to others.

Secondly, Rüsch et al. comment on the relationship between stigma and service use. People decrease usage of psychiatric services in order to overcome public stigma. This is supported by evidence showing associations of this lack of usage with negative reactions from family members and poorer social status.

Lack of usage of psychiatric services is intrinsically linked to decreased treatment compliance and, therefore, poorer prognosis.^20,21

Personal stigma has shown to be associated with variables at different domains, in a systematic review and meta-analysis conducted by Livingston and Boyd.

In the psychosocial domain, stigma has been negatively associated with hope, self-esteem, empowerment/mastery, self-efficacy, quality of life and social support/integration, both at group and individual levels.^22–24

In the psychiatric domain, stigma has been positively associated with symptom severity and negatively with treatment adherence²². There are mixed results regarding association of stigma to diagnosis, illness duration, hospitalizations, insight, treatment setting, functioning and medication side effects, with most of the studies failing to show any statistically significant association.

Regarding socio-demographic variables, both gender, age, education, employment, marital status, income and ethnicity have failed to show any consistent results.²² We should note, however, that some studies have shown significant associations, both positive and negative, regarding each of the variables, with stigma.

1.2 Stigma research

Wahl et al., in 1999, mention four types of stigma research:

• Research that involves self-reports from general public.
• Research using vignettes or profiles of individuals and study participants’ ratings of people described.
• Analogue behaviour studies,(“experimental studies”) in which people are led to believe they are dealing with a person with mental illness.
• They note, however, there was, at the time, few research focused on mental health consumer, and his personal experiences of mental stigma.²⁵

The paradigm changed and nowadays there is a relatively large number of instruments to measure personal experiences of mental stigma.

1.2.1 Instruments to measure stigma - categories and criteria for psychometric properties

In 2010, Brohan et al. ²⁶, reviewed systematically 75 studies with instruments to measure personal experiences of mental stigma. Quality criteria for health status questionnaires have been thoroughly reviewed by Terwee et al,²⁷ and are briefly described in Table 1.

Table 1 - Criteria for quality of psychometric instruments^26,27

Content validity	Clear description is provided of the measurement aim, target population, concepts that the questionnaire is intended to measure, and investigators or experts involved in item selection
Internal consistency	Factor analysis performed on adequate sample size (minimum of 100 subjects, 4 to 10 subjects per variable); Cronbach alpha between 0,70 and 0,95
Construct validity	Specific hypotheses should be assessed (e.g. expected correlations between measures or expected differences in scores between “known” groups); at least 75% of the results are in correspondence with those hypotheses in a subgroup of at least 50 patients
Test-retest reliability	Intraclass correlation coefficients (ICC) or cohen’s Kappa ≥ 0,70 in a sample of at least 50 patients
Floor-ceiling effects	Are considered to be absent if less than 15% of respondents achieved the lowest or the highest possible score.

Brohan et al.²⁶ considered instruments to measure personal experiences of stigma in three categories:

• Perceived stigma
• Self-Stigma
• Experienced stigma

The found fourteen measures, used in the studies, which are listed in Table 2, and that were, thus, grouped in each of those categories. Instruments used were also assessed as to their psychometric properties, according to criteria by Terwee et al.²⁷

Table 2 - Scales assessing stigma experienced by people with experience of mental illness (Adapted from Brohan et al.²⁶)

Scale	Measures Perceived stigma	Measures experienced stigma	Measures self-stigma	Measures other
PDD – Perceived devaluation and discrimination scale28	Perceived discrimination (6 items) Perceived devaluation (6 items)	No	No	No
ISMI – Internalised Stigma of Mental Illness29	No	Discrimination experience (5 items)	Alienation (6 items) Stereotype endorsement (7 items) Social withdrawal (6 items)	Stigma resistance (5 items)
SSMIS – Self-stigma of Mental Illness Scale30	Stereotype awareness (10 items)	No	Stereotype agreement (10 items) Stereotype selfconcurrence (10 items) Self-esteem decrement (10 items)	No
CESQ – Consumer Experiences of Stigma Questionnaire25,31	No	Experiences of stigma (9 items) Experiences of discrimination (12 items)	No	No
RES – Rejection Experiences Scale32	No	Rejection experiences (11 items)	No	No
DSSS – Depression Self-stigma Scale33	Public stigma (4 items)	Stigmatizing experiences (6 items)	General selfstigma (9 items) Secrecy (9 items)	Treatment stigma (4 items)
SRER – Self Reported Experiences of Rejection34	No	Rejection experiences (12 items)	No	No
SS – Stigma Scale35	No	Discrimination (12 items)	Disclosure (11 items)	Positive aspects (5 items
ISE – The Inventory of Stigmatising Experiences36	Perceived stigma 2 items	Experienced stigma 2 items	Social withdrawal 1 item	Impact of stigma (5 item)
SESQ – Self-esteem and Stigma Questionnaire37	Feelings of stigmatisation (8 items)	No	No	Self-esteem (6 items)
HSS – Stigmatisation Scale38,39	Perceived stigma (15 items)	No	No	No
MIDUS – MacArthur Foundation Midlife Development in the United States40	No	Major discrimination (11 items) Day to day discrimination (11 items)	No	No
DISC – Discrimination and Stigma Scale41,42	Anticipated discrimination (4-items)	Experienced discrimination (32 items)	No	No
EDS – Experiences of Discrimination Scale43	No	Has discrimination occurred (1 item) Specific settings of discrimination (8 items)	No	Stressfulness of discrimination in specific settings (8 items)
Maristan stigma scale (MSS)44,45	Health professionals (4 items)	No	Self-Stigma (4-items)	Informal Networks (11 items) Socio-institutional (12 items)

1.2.2 Perceived stigma

Perceived or felt stigma, according to Scambler et al⁴⁶ original definition, refers principally to the fear of enacted stigma, but also encompasses a feeling of shame associated with the illness. Van Brakel et al⁴⁷, however, remove the feeling of shame from that definition, considering research about perceived stigma as research in which “people with a (potentially) stigmatized health condition are interviewed about stigma and discrimination they fear or perceive to be present in the community or society”.

Perceived stigma can refer both to what an individual thinks most people would believe towards a certain group of the society or what that individual thinks about him personally as a member of a stigmatized group.⁴⁸ Components of perceived stigma reported in the literature as measurable variables include stereotype awareness (perception by the individual of how individuals with mental illness are viewed by “most other people” in the society)¹⁶ and personal expectations or fears of encountering stigma.

Perceived stigma is addressed in the vast majority (79%) of the studies reported by Brohan et al. Seven measures were used in the literature to measure it: PDD, SSMIS, ISE, HSS, SESQ, DSSS and DISC.

PDD^26,28 is the most commonly used scale. It totals 12 items - its two subscales measure perceived discrimination and perceived devaluation – a way of measuring stereotype awareness. Perceived stigma is also measured in 10 item stereotype awareness subscale in SSMIS³⁰. HSS investigates perceptions of how the person feels they have been personally viewed or treated by the society. In 2 of its items, DISC addresses the expectation of being stigmatized in various aspects of life – a concept called anticipated discrimination. Although in a specific setting and about a specific group, MSS^44,45 “health professionals” subscale measures in our opinion perceived stigma regarding healthcare professionals, so it would fit in perceived stigma category.

Regarding psychometric properties, all of the measures above mentioned reported on content validity. PDD, SESQ and DSS did not report whether target population was involved in selecting items in the scale. DSSS and SESQ reported results on internal consistency. However, PDD; SSMIS, ISE and HSS, although have calculations for Cronbach’s alpha, do not have factor analysis. SSMIS and SESQ have measured test retest reliability. MSS has been multiculturally tested, and its content validity was assessed. Cronbach alpha, internal consistency and test-retest reliability have been reported and meet criterion level.

1.2.3 Self-Stigma

Self-stigma is considered, by Corrigan, the internalization of the public stigma. For Corrigan et al, there are three components in self stigma: negative belief about the self (e.g., character weakness, incompetence) – cognitive response, agreement with beliefs expressed by the public or the society and negative emotional reaction (e.g., low self-esteem, low self-efficacy) – affective response and behaviour response to prejudice (e.g., failing to pursue work and housing opportunities)^16,49

Self-stigma is assessed by ISMI, SSMIS, DSSS, SS and ISE.

Alienation, stereotype endorsement and social withdrawal subscales in ISMI, measure self-stigma, which correspond to its affective, cognitive and behavioural dimensions⁵⁰. SSMIS measures self-stigma through three sub-scales: stereotype agreement; stereotype self-concurrence and self-esteem decrement^26,30. SS has a “disclosure” subscale, which focus on the three dimensions already mentioned^26,35. ISE contains one item on social withdrwal³⁶. DSSS addresses self-stigma through two subscales – general self-stigma and secrecy: general self-stigma measures personal stereotype awareness. Secrecy subscale can be comparable to social withdrawal subscale in ISMI and disclosure scale in SS³³. MSS^44,45 has a 4 item subscale on self-stigma.

According to Brohan, all the measures reported on content validity. DSSS did not report on target population involvement in item selection. SSMIS and ISE reported on partial criteria for internal consistency, reporting Cronbach’s alpha calculation but not factor analysis. ISMI; DSSS and SS have full internal consistency analysis.

ISMI, SSMIS and SS have been reported to have measured test-retest reliability.

1.2.4 Experienced stigma

According to Brohan and van Brakel, experienced stigma is the “experience of actual discrimination and/or participation restrictions on the part of the person affected”^26,47.

For the purpose of this definition, measuring experienced stigma can refer to measuring experiencing stigma in general or a report of experiences of stigma in specific situations or areas of life.²⁶

By measuring experienced stigma, one can, thus, assess direct effects of public stigma on the stigmatized individual.

Measures of experienced stigma include ISMI, CESQ, SRES, DSSS, SRE, SS, ISE, MIDUS, DISC and EDS.

CESQ will be addressed in a separate chapter. We should note it is the most used scale that addresses only experienced stigma.

ISMI’s discrimination experience subscale has 5 items. It addresses both perceived discrimination and general experiences of discrimination.⁵⁰

RES also measures self-stigma through both its subscales. However, it is developed only in Swedish.

SS discrimination subscale has 12 items on general stigma experiences and specific experiences.³⁵ ISE has two questions on general stigma experiences.³⁶ DSSS contains 6 items about feelings of stigmatization after experiencing or disclosing depression.³³

DISC has 32 items addressing personal experiences of stigma in several areas of life.⁴¹

MIDUS⁴⁰ and EDS⁴³ examine, besides experienced stigma, reasons for stigma.

MSS^44,45 health professionals subscale measures patients’ experienced stigma when they contact healthcare professionals, so we can include that subscale in experienced stigma category.

From the psychometric point of view, we should note that RSE, DSSS, SRE and EDS do not report target population involvement in selecting the items; EDS do not report on internal consistency as ISMI, DSSS, SS and MIDUS met full criteria for internal consistency. DISC⁴² RES, SRE and IDE have been reported to have Cronbach’s alpha calculated, but have no factor analysis; only SS and ISMI measure test-retest reliability.

Other subscales

Other subscales identified in Brohan’s review measure additional elements of stigma, not covered by the perceived, experienced and self-stigma categories. Those include ISMI’s “stigma resistance”, SS’s “positive aspects”, “self-esteem” in SESQ, “treatment stigma” in DSSS and “stressfulness of stigma events” in EDS.

Stigma resistance, positive aspects and self-esteem seem to be, according to Brohan et al, more related with self-estigma. Treatment stigma does not measure mental health stigma, but a related construct. “Stressfulness” is more related to discrimination, as it measures magnitude of experienced discrimination.

MSS’s MSS’s informal networks subscale measures stigma perceived and experienced by the individual on the part of their informal network of care. It would be related with perceived and experienced stigma. Socio-institutional subscale measures aspects related to perception about mental health services organization.

Table 3 summarizes psychometric properties of stigma measures, excluding CESQ, according to Brohan et al.²⁶

CESQ psychometric properties will be summarized in Section 1.3.2.

Table 3 - Assessment of psychometric properties of stigma measures (adapted from Brohan et al.)²⁶

+ = positive rating of property; ? = indeterminate rating of property, -= negative rating of property, 0 – no information available. Rating of properties is according to Terwee criteria.²⁷

Scale	Content Validity	Internal Consistency	Construct validity	Test-retest reliability	Floor/ Ceiling effects
PDD28	?	?	+	0	0
ISMI50	+	+	+	+	0
SSMIS30	+	?	+	+	0
RES32	?	?	+	0	-
DSSS33	?	+	+	0	0
SRE34	?	?	+	0	-
SS35	+	+	+	+	0
ISE36	+	?	0	0	0
SESQ37	?	+	+	?	0
HSS38,39	+	?	+	0	0
MiDUS40	?	+	+	0	-
DISC42,51	+	0	0	0	0
EDS43	?	0	+	0	0
MSS44	+	+	0	+	0

1.2.5 Stigma studies in Portugal

In Portugal, there have been few studies about stigma in mental illness. Most of them are about public stigma, by Palha et al.⁵²

There is also noticeable work from M. Gonçalves Pereira et al., regarding stigma in caregivers.⁵³

Personal has been increasingly studied over the last few years. There is record of an authorized translation of ISMI⁵⁴, but there are no psychometric studies about it.

We should also note a study by Sousa et al.⁵⁵, correlating self stigma and recovery potential in Schizophrenia.

Nevertheless, there are no published studies on experienced stigma.

CESQ, the most studied scale that measures experienced stigma, has an authorized translation and retroversion in Portugal, but had not its psychometric properties documented until now.

1.3 Consumer experiences of stigma questionnaire (CESQ)

CESQ was developed in 1999 by Wahl ²⁵, with the collaboration of National Alliance for Mentally Ill (NAMI).

Its development was based on the fact that most of the research done regarding stigma until then seldom focused on real life experience of mental health consumer.

The questionnaire is available as a self-reporting tool, but has also been administered in face to face interviews.

The questionnaire has two sections. The “stigma” section includes, in its original version, nine questions. ”Discrimination” section has 12 items, in its original version.

It has been reported to be translated in chinese⁵⁶, and polish^57–59. There is also a Portuguese authorized translation of CESQ, with its corresponding backtranslation⁶⁰ , which remains unpublished until date. Chinese translation does not have psychometric data.

1.3.1 Main research with psychometric data involving CESQ.

In Wahl’s original study, a postal survey which had 1301 respondents, selected within a sample obtained through NAMI mailing list, respondents had been diagnosed mainly with bipolar disorder (25%) but also with schizophrenia (19%) and major depression (15%).

Most commonly reported stigma experiences included witnessing stigmatizing comments about mental illness (50% often or very often), encountering hurtful or offensive media portrayals of it (47% often or very often), or being treated as less competent by others once their illness had been disclosed (36% often or very often). 26% of them often had the experience of sometimes being shunned or avoided. 16% of the respondents reported receiving support from friends seldom or never. More than half (55%) of the consumers said they worried that others would view them as service users.

Discrimination experiences, such as being turned down for a job, were less reported: One in three respondents reported that they had been turned down a job, and 28% found work environment unfriendly.

Wahl’s study has a few limitations: First, it is based in a sample of patients affiliated with an advocacy organization, and who choose to respond to a stigma questionnaire. Wahl himself questions if it represents the typical mental health service user. Secondly, it is based in a heterogeneous sample, which includes patients both with common and severe mental illness. Thirdly, it does not establish a correlation between stigma and other variables such as patients socio-demographic characteristics, illness severity, gender, mood-related variables, or illness insight³¹.

Dickerson et al.³¹, in 2002, applied CESQ to 74 outpatients with Schizophrenia, modifying it by administering it verbally in a face-to-face interview, and by replacing the term “consumer” with “person who has mental illness”, person who has a psychiatric disorder” or “Person who uses psychiatric services”. They assessed symptom severity quality of life and social functioning, socioeconomic status, gender, mental illness attribution and Depression.

Item to scale correlations were calculated, regarding stigma and discrimination section of CESQS. All of the items in the stigma section correlated with total score for that section. Only 6 of 12 items in Discrimination section correlated with total score for discrimination.

In their sample, Dickerson et al. found that ranking of responses was similar to NAMI-affiliated sample from Wahl. However, average rating of CESQ items was higher in Wahl’s sample, meaning patients affiliated with NAMI would report more frequently stigma experiences.

Surprisingly, they did not find correlations of stigma with psychiatric symptoms, depression, subjective satisfaction with living situation or subjective satisfaction with Safety.

There was inverse correlation between several subjective variables from quality of life scale and Stigma total score: Subjective satisfaction with daily activities, Subjective satisfaction with family contact, subjective satisfaction with finances, and Subjective satisfaction with health. The variable measuring Perceived Adequacy of Finances to Meet Daily Needs from Quality of Life Interview was inversely correlated with total stigma score. In a multiple regression analysis model on the stigma total score, where the former variables were entered, only Perceived Adequacy of Finances remain significant in the equation.

Socioeconomic status based on parental occupation was significantly correlated with discrimination total, and remained significant in univariate comparisons.

Bagley and King have done an exploratory analysis on three stigma scales, including CESQ. However, only 13 items were included. Criteria for inclusion was that each item correlated 0,4 or greater with the scale total, and the resulting scale had an alpha value of at least 0,80.³⁸ Only 7 items have “survived the alpha analysis”. However, the study does not address the fact that there are two subscales in CESQ, and does not treat them differently nor does a factor analysis.

Switaj et al.⁵⁷, in 2009, applied stigma section of CESQ to 153 inpatient and outpatient patients with schizophrenia, diagnosed according to ICD-10 criteria, aged between 18 and 65 and that were not dependent on drugs of alcohol, also applying Satisfaction with life domains (SLDS) scale in order to assess subjective quality of life, Global Assessment of Functioning (GAF) for global functioning, Frankfurter Befindlichkeits-Skala (FBS) and Paranoid-Depresivitäts-Skala (PD-S) to assess patient rated severity of symptoms and Clinician Assessment of Schizophrenic Syndromes (CASS) in order to measure severity of psychiatric symptoms.

Regarding CESQ, they found out that up to 86% of the patients avoid to disclose the fact that they are receiving psychiatric treatment to people outside the family. 40 to 70% of the patients reported at least sometimes being avoided by others, treated as less competent, advised to lower one’s expectations in life, having worried about being viewed unfavourably by others, hearing negative comments about the mentally ill or encountering offensive media depictions of mental illness. Questionnaire has shown good internal consistency. (Cronbach’s alpha = 0,79)

Regarding relationship between clinical and social-demographic characteristics and stigma, they found a negative association with subjective quality of life, which matches previously described data.

They also found, through stepwise regression multiple analysis of socio-demographic and clinical characteristics and stigma that patients who became ill at a younger age felt more stigmatized. (beta=-0.33 ; p<0.001)

There was a positive correlation between stigma and patient rated symptoms, in both Paranoid and Depressive subscales of PD-S.

However, there were no significant associations, between the level of stigmatization experienced by the participants and: socio-demographic characteristics such as age, sex, education, marital status, housing status, current employment or number of years in employment; and with type of psychiatric setting, duration of illness, number of involuntary hospitalizations, medication side effects, social functioning and overall severity of clinician rating symptoms. This goes in line with the data reported by Dickerson et al.

In 2012, Switaj and colleagues⁵⁸ also applied CESQ to 442 patients, treated in various psychiatric care facilities in Warsaw. 63% of them were diagnosed with psychotic disorders, and 14% of them with affective disorders. In accordance with the previous study, a majority of patients reported having often or very often avoided telling others about their mental health problems. More than a third indicated having often or very often heard others make unfavourable or offensive statements about people with mental illness, worried about being viewed unfavourably by others, and been treated as less competent. Nevertheless, it should be noted that only a small fraction of subjects believed that people who knew about their illness seldom or never treated them fairly, and that their friends were not understanding and supporting.

Regarding discrimination experiences, they were less reported, in parallel with what happened in previous studies by Wahl and Dickerson et al.: Just above half of those interviewed said that in their written applications for jobs, housing, school, or licenses of all types they often or very often avoided mentioning that they are in psychiatric treatment. Approximately 1 in 4 respondents said that their colleagues or superiors at work who knew about their mental health problems seldom or never were supportive and accommodating, and about 1 in 5 thought that representatives of the justice system had not treated them with consideration and kindness. Slightly less than 10% reported that they were often or very often turned down for a job. No further correlations were sought.

In 2013, Switaj led a group that made a secondary analysis of the previous described samples, performing a factor analysis in order to assess construct validity of stigma section of the scale⁵⁹.

They decided not to use the discrimination section of CESQ. This decision is due to the fact that in their sample, the discrimination section was proved to have unsatisfactory internal consistency (Cronbach’s α = 0,63), which is corroborated by work by Solomon et al⁶¹ and to the fact that as many as 7 items in discrimination section were endorsed by a marginal proportion of respondents.

They selected only cases which completed the full stigma section –373 in their 2012 study (sample one) and 136 sample of the 2009 study (sample two.

Regarding data analysis, they started to randomly split sample one in two demographically and clinically equivalent subsamples.

Internal consistency of the 9-item subscale was not acceptable regarding inter-item correlation (0,302 in sample one and 0,282 on sample two), although Cronbach alpha has acceptable in both samples. (0,799 and 0,788 respectively).

Two of the items in the subscale were found to be psychometrically poor in Exploratory factor analysis and were, thus, discarded. The remaining 7 items model parameters of fit were in the acceptable ranges in confirmatory factor analysis.

Concurrent validity was assessed using SLDS (quality of life), PD-S, FBS (patient reported symptoms) and GAF (global functioning) and CASS (clinician-rated symptoms). Pearson