Development of Self-Stigma Inventory for Families of the patients with schizophrenia (SSI-F): validity and reliability study

ABSTRACT OBJECTIVE: It is known that relatives of the patients with schizophrenia tend to hide the illness from other people, are ashamed of their patients, and feel excluded from society. This situation is referred as self-stigmatization of families, and it may negatively affect the family functioning and therapeutic alliance. Assessing and evaluating the self-stigma of families are essential concerning family therapies and treatment of their patients. The purpose of this study is to develop a culturally sensitive inventory for the assessment of self-stigmatization for families of patients with schizophrenia in Turkey. METHODS: After examining the studies in the related field and conducting a focus group interview with the families of the patients with schizophrenia, a 19-item inventory was formed. One hundred and six relatives of the patients with schizophrenia and schizoaffective disorder were given a sociodemographic form, Self-Stigma Inventory for Families (SSI-F), Beck Depression Inventory (BDI), Beck Hopelessness Scale (BHS), Rosenberg Self-Esteem Scale (RSES), and Zarit Caregiver Burden Scale (ZCBS). Explanatory factor analysis and convergent validity were assessed as validity analysis, and internal consistency coefficient, item–total correlation, and test–retest reliability were calculated for reliability analysis. RESULTS: The sample consisted of 106 relatives whose 52% were female, 77% were married, mean age was 51 years, and level of education was 9 years. In explanatory factor analysis, three factors (social withdrawal, concealment of the illness, and perceived devaluation) with 14 items were detected, and the factors could explain 66.8% of the total variance. SSI-F was significantly correlated with Beck Depression Inventory (r = 0.48, P < 0.01), Beck Hopelessness Scale (r = 0.27, P < 0.01), Zarit Caregiver Burden Scale (r = 0.54, P < 0.01), and Rosenberg Self-Esteem Scale (r = −0.35, P < 0.01). Cronbach’s alpha coefficient for SSI-F total score was calculated as 0.88, and test–retest reliability coefficient of SSI-F was 0.93. CONCLUSIONS: This study shows that the SSI-F is a valid and reliable instrument for assessing self-stigmatization in the families of patients with schizophrenia. It can be considered as a valuable instrument to use for research and therapeutic purposes.


Introduction
Stigmatization is defined as making a disrespectful attribution to an individual or ascribing disgraceful and discreditable characteristics, without reflecting the reality, by the society since the person is considered as outside of the criteria that the community accepted as "normal" [1,2]. Patients with schizophrenia are generally regarded as "dangerous" and "unpredictable" individuals or as "outcasts," hence they face stigma frequently [3,4]. Stigmatization takes place in cognitive, emotional, and behavioural processes. First, stereotypes about mental illnesses emerge (e.g. patients are dangerous, they cannot look after themselves, a patient is an unknown quantity), then these negative stereotypes become stronger and turn into prejudiced behaviours, and consequently negative emotions (e.g. fear, disgust) appear. As a result, discrimination occurs through casting out of society [5,6,7]. Stigmatization constitutes a critical and secondary problem in addition to the disease especially for the individuals with severe mental illnesses such as schizophrenia. Experiencing stigma lowers the individuals' self-confidence, hinders their abilities to reach their targets, causes to miss social opportunities, and reduces their quality of live [5].
Mental illness and stigmatization are such blended terms that not just mentally ill patients are affected by the stigmatization, but the family members of the patients are also affected [8][9][10][11]. Beliefs about that inadequate parenting skills and detrimental environmental conditions provoke mental illnesses and the knowledge about the effects of heredity [12][13][14] might cause family members of the patients to hold themselves responsible for the illness and the poor prognosis. On the one hand, family members feel responsible for their relatives' illnesses, on the other hand, they exposed to negativistic attitudes from their environment and start to blame themselves. Difficulties of being a relative of a patient with schizophrenia and the psychosocial troubles are revealed through qualitative and quantitative studies [15][16][17][18][19]. In fact, some family members feel as if they have committed a disgraceful act and steer away from their social environments.
Family members are aware of that not only their patients are stigmatized, but also they are subjected to social stigma and devaluation as well [8,9,11,20]. While this awareness causes some family members to strive with and grow stronger, it creates some others to internalize the stigma and leads self-stigmatization. Then, by internalizing the social stigma, the individual embraces other people's stereotypes, and consequently, social withdrawal is observed with negative emotions like worthlessness and shame [21,22]. Stereotypical thoughts, prejudices, negative feelings, and exclusionary attitudes toward the patients may be internalized by their family members as well [11,23].
Experiencing self-stigma can lead family members to have lower self-esteem, hopelessness, despair, depression, hiding the illness and social withdrawal, impairments in individual and social/familial functioning, increase their burden, and decrease therapeutic alliance [9,11,17,24]. Self-stigma of the family members generally manifests itself as devaluation, feelings of insufficiency, social withdrawal, and concealment of the disease [9,11,23,25]. Stigma experience in family members is found to be significantly correlated with depression, suicidal thoughts, and higher caregiver burden [26][27][28].
Studies conducted in Turkey revealed that family members of the patients experience self-stigma [27,29], and pointed out that self-stigma is significantly correlated with depression and caregiver burden [27]. Researchers in studies mentioned early used the selfstigma scales that were developed for the patients [30,31] by adapting them into relatives of the patients. In Turkey, there is no culturally sensitive scale evaluates the self-stigma of the relatives of the patients. The purpose of this study is to develop a scale to assess the self-stigma of family members of the patients and to evaluate its psychometric properties.

Participants
One hundred and six relatives (mother, father, sibling, the spouse who lives together with) of the patients who are diagnosed with schizophrenia or schizoaffective disorder according to DSM-5 criteria [32] and outpatients of Kocaeli University School of Medicine Psychiatry Polyclinic between September 2016 and July 2017 were included to the study. Ethical approval for the study was taken from Kocaeli University Ethical Committee of Non-invasive Clinical Research (KÜ GOKAEK 2016/61). Family members of the patients were informed about the purpose and procedure of the study, and informed consent was taken from those who agreed to participate.

Inclusion criteria
Family members of the patients were taken to the study who agreed to participate, are 18-65 years old, do not have mental retardation, any current psychiatric disorder or neurological disease which may affect their judgment, and at least graduated from primary school.

Sociodemographic form for families
The sociodemographic information form includes the relative's age, sex, marital status, education level, employment status, economic conditions, and relationship with the patient.
Self-Stigma Inventory for Families (SSI-F) Self-stigma scales developed for the patients with mental illnesses [33,34], Internalized Stigma of Mental Illness Scale [30], and the scales developed for the family members or caregivers of the patients such as self-stigma scale [11], internalized stigma [23], and devaluation scale [9] were examined, and 25-item self-stigma scale for the relatives of the patients was formed. A focus group interview was conducted with 18 family members of the patients with schizophrenia, and the items of the scale were reevaluated. Then, a 19item scale was formed that was comprised of selfstigma statements that the family members emphasized and brought up themselves. The answer to each item was arranged as 5-point Likert type scale as "1 = do not agree, 2 = slightly agree, 3 = moderately agree, 4 = generally agree, 5 = totally agree." A pilot study was conducted with 19-item inventory with 18 family members, and the scale was finalized by reexamining the incomprehensible items.
Zarit Caregiver Burden Scale (ZCBS) Zarit Caregiver Burden Scale was developed by Zarit et al. [35] for assessing the burden of the caregivers of patients with Alzheimer's disease. In Turkey, Özlü et al. [36] conducted its reliability and validity study with relatives of the patients with schizophrenia. The Turkish version of the scale consisted of 19 items. Internal consistency of the scale was 0.83. Higher scores indicate greater burden.

Beck Depression Inventory (BDI)
The scale was developed by Beck et al. [37] to assess physical, emotional, and cognitive symptoms observed in depression and the study of its Turkish adaptation was conducted by Hisli [38]. It is a 21-item self-assessment scale. Cronbach's alpha coefficient was found as 0.90. Higher scores indicate the greater level of depression.

Beck Hopelessness Scale (BHS)
It is a 20-item scale developed by Beck, Lester, and Trexler [39], and adapted to Turkish culture by Durak [40]. Internal consistency of the scale was found as 0.86. Higher scores from the scale indicate the higher level of hopelessness.

Rosenberg Self-Esteem Scale (RSES)
In this study, 10-item Rosenberg Self-Esteem Scale [41] was used. Turkish reliability and validity study of the scale was conducted by Çuhadaroğlu [42]. Internal consistency of the scale was 0.71. Higher scores indicate higher self-esteem.

Statistical analysis
Statistical analyses were carried out with the SPSS 22 (IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp.). For examination of structure validity, the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy and Bartlett's tests of sphericity were utilized to check preliminary assumptions of factor analysis. The factor analysis was done by using Principal Component Analysis and Direct Oblimin Rotation. Corrected item-total correlations, Cronbach's alpha, and Cronbach's alpha if item deleted were calculated to assess the internal consistency of the scale. Pearson or Spearman correlations depending on the distributional features of the variables were used for the assessment of concurrent validity and test-retest reliability. Level of significance was set at a P-value of 0.05.

Results
Sociodemographic characteristics of the participants were given in Table 1.
The validity of the SSI-F

Construct validity
Explanatory factor analysis was conducted to examine the construct validity. The KMO test indicated excellent sampling adequacy (KMO= 0.804), and Bartlett's test of sphericity has suggested that a factor analysis may be useful for the data (χ 2 = 883.19, df= 91, P < 0.001). In the explanatory factor analysis of the 19item scale, five factors were found which can explain 72.6% of the total variance and has eigenvalue greater than 1. The scree plot showed that the slope started to change and drop dramatically after the third factor. Thus factor analysis was repeated by Direct Oblimin with three factors and the items with factor value less than 0.45 (item numbers 1, 3,11,12,19) were removed. Thereby, the scale had 3 factors and 14 items. It was revealed that first factor explains 44.6% of the total variance, the second factor explains 12.3% of the total variance, and the last factor explains 9.9% of the total variance, and together they explain 66.8% of the total variance. In terms of the items constitute the factors: first factor (9,8,16,10,7,15) was labelled as social withdrawal, second factor (13, 14, 5) described as concealment of the illness, and third factor (2,4,18,17,6) entitled as perceived devaluation (Table 2).

Content validity
The correlation between the total score of the scale and its subscales was calculated to assess the content validity (Table 3). A high correlation was found between the SSI-F total score and social withdrawal (r = 0.86), concealment of the illness (r = 0.74), and perceived devaluation subscale (r = 0.90).

Concurrent validity
Concurrent validity of the SSI-F was assessed with its correlation with BDI, BHS, RSES, and ZCBS. The results were given in Table 4. SSI-F was significantly correlated with all the scales. The scale was positively correlated with BDI, BHS, and ZCBS, and negatively correlated with RSES as expected.

The reliability of the SSI-F
Internal consistency reliability Item-total correlation and Cronbach's alpha if item deleted were analysed. In the internal consistency analysis, Cronbach's alpha coefficient of the scale was calculated as 0.88 for SSI-F final version and 19-item first version as well. Cronbach's alpha coefficient for the social withdrawal, concealment of the illness, and perceived devaluation factors were 0.84, 0.82 and 0.84 respectively. Data concerning the item-total correlations and Cronbach's alpha coefficients calculated for each item through an if item deleted technique can be found in Table 5. Item-total score correlation coefficients were between 0.49 and 0.75, and all were statistically significant (P < 0.001).

Test-retest reliability
In the test-retest reliability analysis, the data related to 30 individuals were analysed through the Spearman correlation test in a 1-month interval. The test-retest reliability coefficient of the scale was 0.93 (P < 0.001).

Discussion
In the current study, the validity and reliability of the newly developed scale (SSI-F) which was prepared to evaluate the self-stigma experience of family members of the patients with schizophrenia were examined. The fact that this scale was formed by reconsidering existing studies in the related field and working with the family members themselves enabled the scale to be a culturally sensitive, easy to comprehend, and time-saving instrument. Regarding psychometric properties, it was found to be reliable and valid 14-item scale with its three-factor structure. Items in the first factor were regarded as social withdrawal since it includes the evaluations of the people around the family members. These items reflect that the family members have thoughts about people were being afraid of them as if they can lose their control anytime, stay away from people in case they may make comments or jokes that would presumably hurt them, think that others do not care them, have thoughts about being a burden to others, believe that others are assuming that family members cannot take proper decisions, and stay away from others because of thinking they cannot be understood. The items in the second factor were entirely about the concealment of the illness. It showed that family members were unwilling to talk about the illness of their patients with their close circles, neighbours, or friends. In the third factor, lower self-confidence, thoughts on failing,   senses of uselessness, thoughts on feeling unable to be happy, and thoughts about their inability to fulfil their responsibilities were considered as the perceived devaluation of the family members. Social withdrawal, concealment of the illness, and perceived devaluation are the dimensions that are generally revealed in selfstigma scales [9,11,23]. Endorsement of stereotypes factor that was found in other scales was placed in both first and third factors in SSI-F. For example, the stereotypical judgments such as "patients cannot take proper decisions," "they are useless," "they cannot fulfil their responsibilities as other people do," "they may show unpredictable behaviours," "they are unimportant beings," and "they cannot be happy or successful" were being internalized by the relatives of the patients in their self-stigmatized thought processes. This scale, which was formed through the interviews conducted with family members of the patients, was found to reveal the self-stigmatization dimensions conceptually which were experienced by the relatives. Selfstigma feeling comprised of perceived devaluation, concealment of the illness, and thoughts and behaviours of social withdrawal. The correlations between those three factors were found to be significantly high. However, the fact that there was a moderate correlation between social withdrawal and concealment of the illness needs some explanation. We think that this situation might be a reflection of culture-specific properties. The family members hiding the name and diagnosis of their patients' illness from their relatives, friends, or neighbours do not necessarily indicate their withdrawal from the society. Studies conducted in Turkey [43,44] showed that people in society fear from the patient with psychotic illness and the patients are generally perceived negatively. This perception may have an influence on the family members of the patients for their concealment of the illness.
The SSI-F total score was significantly correlated with the BDI, BHS, RSES, and ZCBS. Similar results were shown in other studies investigating the selfstigma of the relatives of the patients with mental illnesses. Most of the studies presented a significant correlation between self-stigma and low self-esteem [20,31], hopelessness and depression [26][27][28]31], and higher caregiver burden [20,26,27]. In our study, especially the relation between self-stigma and family burden (r = 0.54) was explicit. Mak and Cheung [11] similarly found a relationship between self-stigma and caregiver burden, and they underlined that the family members tend to socially withdraw and conceal the illness because of the feelings of shame and insufficiency arising from living with their mentally ill relatives. Some studies also indicated that self-stigma has a mediator role on caregiver burden [24,25].
In Turkey, studies conducted with the family members of the patients revealed that the relatives also have stereotypical and stigmatized emotions, thoughts, and behaviours [29,43,44]. Since the relatives of the patients both stigmatize the patients, being stigmatized, and experience the self-stigma, these experiences might be a significant predicament that increases their distress and burden. It is obvious that throughout the therapy process of the patients, interventions and educations given to the relatives of the patients should take into consideration their feelings about self-stigmatization. We believe that this scale which is developed to evaluate the feelings of self-stigmatization is an important tool for research purposes and therapeutic processes.

Limitations of the study
This study has some limitations that should be considered. First, the study had been carried out in a single centre using a convenience sampling, whose patients are under treatment; thus the results may not be representative of all relatives of the patients some of whom may not currently be under treatment. Second, the study has a relatively small sample size that might have affected the statistical power. Third, we did not have the opportunity to compare this scale with another reliable and valid scale which directly assesses the self-stigma of the relatives, which unfortunately is not available in Turkish. Convergent validity analysis remained limited to other equivalent scales with which reliability and validity studies were conducted in Turkey. Fourth, the measurements used in this study were self-report scales that might not reflect the real feelings of the relatives. Notwithstanding these limitations, the present study is the first attempt to develop a tool that captures the cultural foundations of the concept of self-stigmatization of the people who have a relative with schizophrenia living in Turkey. Although more research is needed to replicate and further validate the measure, the findings of this study provided preliminary support of its validity. Future studies should also be done to validate in the relatives of other severe illnesses, such as dementia, alcohol and substance use disorder, and AIDS, who have been experiencing a tremendous amount of burden and suffering.