Abstract
Background:
Internalized stigma negatively affects lives and prognosis of individuals with psychosis.
Aim:
This study aimed to identify correlates of internalized stigma among individuals with psychosis in a sample of community care users in the Czech Republic.
Methods:
A cross-sectional study was conducted among 133 community service users with psychosis. A shortened version of the Internalized Stigma of Mental Illness (ISMI-10) scale was used alongside the 5-level EQ-5D version (EQ-5D-5L), assessing health-related quality of life. Descriptive and linear regression analyses were performed in order to determine levels of internalized stigma and its correlates.
Results:
High levels of internalized stigma were reported in 25% of participants. Lower internalized stigma levels were associated with better self-reported health status and being married, and higher internalized stigma with a longer period of time since initial contact with psychiatric care.
Conclusion:
Lower internalized stigma levels are associated with better self-reported health-related quality of life. In addition, clients having used psychiatric care for longer periods of time reported significantly higher internalized stigma levels. Therefore, authors suggest self-stigma reduction interventions based in a community setting with an emphasis on targeting clients with chronic psychosis.
Introduction
Stigma has a substantial impact on various facets of life for people with mental illness. A large body of literature focuses on public stigma and its reduction, but much has discounted internalized stigma and the role it plays in the lives of individuals with psychosis. Internalized stigma (also referred to as personal stigma or self-stigma) is defined as the endorsement and internalization of negative stereotypes and prejudices associated with mental illness in persons with mental health conditions (Corrigan & Watson, 2002).
Prevalence rates for internalized stigma for people with mental illness vary and are difficult to compare due to inconsistencies in metrics used and populations assessed (Link, Yang, Phelan, & Collins, 2004). International prevalence rates of internalized stigma for individuals with psychosis range from 18.8% to 44.2% (Drapalski et al., 2013; Lv, Wolf, & Wang, 2013; Mosanya, Adelufosi, Adebowale, Ogunwale, & Adebayo, 2014; Werner, Aviv, & Barak, 2008), and within Europe, 33% to 41.7% of patients with psychosis have moderate-to-high levels of internalized stigma (Brohan, Elgie, Sartorius, & Thornicroft, 2010; Krajewski, Burazeri, & Brand, 2013). Research indicates that prevalence rates of internalized stigma can be greater in patients living with psychosis than in patients with non-psychotic mental illness (Lien et al., 2015). While a recent study found no significant variance between internalized stigma levels and various demographic variables in patients with psychosis (Vrbová et al., 2015), one systematic review detected marginal associations among patient sociodemographic variables and levels of internalized stigma, reporting a significant association between literacy rates and internalized stigma (Gerlinger et al., 2013).
The quality of life and mental health of people with psychosis is significantly poorer for individuals with internalized stigma (Chan & Mak, 2014; Holubova et al., 2016; Mashiach-Eizenberg, Hasson-Ohayon, Yanos, Lysaker, & Roe, 2013; Park, Bennett, Couture, & Blanchard, 2013; Tang & Wu, 2012), and this trend persists regardless of patient demographics (Mosanya et al., 2014). Consequences of internalized stigma in patients with psychosis include decreased social functioning (Gerlinger et al., 2013; Hill & Startup, 2013; Lysaker, Davis, Warman, Strasburger, & Beattie, 2007; Yanos, Roe, Markus, & Lysaker, 2008), activity levels (Moriarty, Jolley, Callanan, & Garety, 2012), social power (Campellone, Caponigro, & Kring, 2014), vocational functioning (Gerlinger et al., 2013; Yanos, Lysaker, & Roe, 2010), self-esteem (Mashiach-Eizenberg et al., 2013; Staring, Van der Gaag, Van den Berge, Duivenvoorden, & Mulder, 2009; Yanos et al., 2008), self-efficacy (Hill & Startup, 2013), as well as increased depressive symptoms (Lien et al., 2015; Park et al., 2013; Schrank, Amering, Hay, Weber, & Sibitz, 2014; Valiente, Provencio, Espinosa, Duque, & Everts, 2015; Yanos et al., 2008) and symptom severity (Chan & Mak, 2014; Holubova et al., 2016; Livingston, Rossiter, & Verdun-Jones, 2011; Lysaker et al., 2007). Finally, internalized stigma undermines treatment adherence (Fung, Tsang, & Chan, 2010; Gerlinger et al., 2013; Kvrgic, Cavelti, Beck, Rüsch, & Vauth, 2013; Tsang, 2013; Tsang, Fung, & Chung, 2010; Uhlmann et al., 2014; Vrbová et al., 2014), acts as a barrier to help-seeking (Anderson, Fuhrer, & Malla, 2013) and contributes significantly to a greater psychosocial impact of stigma when receiving psychiatric treatment (Świtaj, Chrostek, Grygiel, Wciórka, & Anczewska, 2016).
Studies clarifying the relationships between sociodemographic characteristics and stages of illness for patients with internalized stigma lack, yet they are fundamental to the successful design and implementation of effective stigma reduction interventions (Gerlinger et al., 2013). We aimed to identify correlates of internalized stigma among people with psychosis in a sample of community care users in the Czech Republic in order to identify risk factors which should be taken into account when designing self-stigma interventions.
Methods
Participants and study design
This cross-sectional study was nested into a year-long longitudinal study (Kondrátová, König, Mladá, & Winkler, 2019; Winkler et al., 2018) measuring the cost and outcomes of psychosis. Data were collected during the second measurement, out of four conducted within the study.
Data were collected in fall of 2015 within the eight community mental health care centres throughout seven regions of the Czech Republic. Community mental health care is provided as a part of the social service scheme in the Czech Republic. Usually, community mental health care centres are non-governmental organizations providing one or more registered social services (e.g. social rehabilitation, social counselling and sheltered workshops). Eligible clients for this study were users with a psychotic disorder having used community care services for at least 3 months at the beginning of the study. Participants were randomly selected from all clients meeting the selection criteria in each centre. A total of 635 community service users met the inclusion criteria and were considered eligible for the study and were subsequently randomized. From the total number of respondents who were asked to participate in the study (N = 277), 139 refused to participate for various reasons: they were not in a good health status (N = 29), hospitalized (N = 26), doubted their confidentiality as participants in the study (N = 22) and others (such as length of study or not being interested in participating). There were no significant differences among the respondents and the eligible clients who refused to participate in the study in terms of age, gender, education, family status or years of service use with a particular provider. A total of 133 respondents participated in the second measurement as two were hospitalized, one moved to different city, one stopped using community care services and one decided not to take part in the study. Respondents were given a small amount of money for participating in each wave. Further information about the study may be found elsewhere (Winkler et al., 2018).
Measures
Internalized stigma was measured by a shortened version of the Internalized Stigma of Mental Illness (ISMI-10) scale (Boyd, Otilingam, & DeForge, 2014). Respondents rated eight negatively oriented and two positively oriented items according to a 4-point scale (1 = strongly disagree, 2 = disagree, 3 = agree, 4 = strongly agree). Mean scores were computed afterwards with a higher score indicating higher internalized stigma (if one item had missing value, a score was computed from nine items; in the case of additional missing values, a score was not computed). Scores above 2.5 are considered as high internalized stigma (Ritsher & Phelan, 2004). Cronbach’s alpha tested at .71, indicating good reliability of the scale in this survey. The psychometric properties of the Czech version are satisfactory (Ociskova, Prasko, Kamaradova, Grambal, & Sigmundova, 2015; Vrbová et al., 2016). The ISMI scale is self-reported.
Health-related quality-of-life score was assessed by a standardized measurement tool, 5-level EQ-5D version (EQ-5D-5L), consisting of five dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. These domains are evaluated on five levels: no problems, slight problems, moderate problems, severe problems and extreme problems (Herdman et al., 2011).
The questionnaire also consisted of sociodemographic variables (gender, age, education, family status, economic activity and income) and health-related variables (age of initial contact with psychiatric care, number of months since last hospitalization and attitudes towards medication).
Statistical analysis
Descriptive and linear regression analyses were performed in IBM SPSS Statistics 23. A mean score was measured for each ISMI item as well as the ISMI scale. A multiple linear regression model was created in order to identify correlates of internalized stigma as measured by the ISMI scale. For assessing multicollinearity, we checked the tolerance and variance inflation factor (VIF). If otherwise not specified, in all analyses, p-values less than .05 were considered statistically significant. Analysed variables include sociodemographic characteristics, economic activity, health status and psychiatric history of patients.
Results
Participants
The sample consisted of 133 community care users with experience of psychosis. Their sociodemographic and health characteristics are shown in Table 1.
Characteristics of respondents (N = 133).
One participant did not fill in the answer.
Outcomes
Table 2 shows the mean ISMI score of internalized stigma among participants (2.27; confidence interval (CI): 2.20–2.36). According to the two-category interpretation of the ISMI score of Ritsher and Phelan (2004), there are not overall high rates of internalized stigma in this population. However, the prevalence of high internalized stigma scores among participants was 25.0%.
Characteristics of ISMI score.
If we look closer at the items which constitute the ISMI score (Table 3), we can see that the highest mean answer and the highest percentage of more stigmatizing answers include the following items: ‘Having a mental illness has spoiled my life’ (M = 2.76; % of more stigmatizing answers = 60.2) and ‘People without mental illness could not possibly understand me’ (M = 2.55; % of more stigmatizing answers = 50.4). All other items have a mean answer lower than 2.5, with the lowest mean answer 1.83 for the item ‘Mentally ill people tend to be violent’ (% of more stigmatizing answers = 18.0).
Mean answers and percentages of answer to ISMI items.
Not all respondents answered all items.
The linear regression model created for the identification of internalized stigma predictors is presented in Table 4. The ISMI score was set as a dependent variable. Independent variables were gender, age, education, family status, economic activity and first contact with psychiatric care and health status. Results show that lower internalized stigma levels were associated with better health status (unstandardized beta coefficient (β) = –0.010, p = < .001) and being married (unstandardized beta coefficient (β) = –0.272, p = .032). Higher internalized stigma levels were associated with a longer period of time since initial contact with psychiatric care (unstandardized beta coefficient (β) = 0.010, p = .019). The model accounted for 23.1% of sample variation (adjusted R2 = .231).
Linear regression model for ISMI score.
Note: Bold p-values indicate results significant on the level of 0.05.
Discussion
Our results indicate that overall levels of internalized stigma are not exceptionally high in community service users with psychosis in the Czech Republic. Still, 25.0% reached ISMI score higher than 2.5, indicating moderate-to-high levels of internalized stigma. Similar studies conducted in community-based service users with psychosis report more prevalent rates of moderate-to-high ISMI scores, such as in Hong Kong (38.3%) and Guangzhou, China (49.5%) (Young & Ng, 2016), as well as in psychiatric out-patients in Ethiopia (46.7%) (Assefa, Shibre, Asher, & Fekadu, 2012). In a cross-sectional study of persons with mental illness across six countries in Europe, the distribution of moderate-to-high levels of ISMI scores was 33%, ranging from 15% in Sweden to 57% in Croatia (Krajewski et al., 2013). In the United States, 36.1% of people with a severe mental illness showed higher levels of self-stigma (West, Yanos, Smith, Roe, & Lysaker, 2011). As many studies examining internalized stigma included more broad populations of people with mental illness, our results are relatively low when considering the findings from past research show that patients with psychosis can have higher levels of self-stigma in comparison with patients with other mental illnesses (Lien et al., 2015). Past studies in the Czech Republic report exponentially higher rates of self-stigma for in-patients (63.2%; Grambal et al., 2016) and for out-patients with schizophrenia (64.0%; Holubova et al., 2016).
Apart from methodological reasons, the differences between in-patients and individuals using community services could be explained by the type of care they receive. It could be hypothesized that people using community care centres are in more stable health conditions, feel less excluded from society and have more social contacts than those hospitalized in in-patient care. However, half of the respondents agreed with the item ‘People without mental illness could not possibly understand me’ which could indicate that the socialization in community services is mostly among people with experience of mental illness and those people lack contact with the ‘healthy’ population. A comparison of samples from out-patient and community services and eventually those who do not use any service would be needed to explain the role of types of mental health care on self-stigma of people with mental illness.
A longer time since the initial contact with psychiatric care is associated with higher internalized stigma. Studies have identified several factors that can lead to change in self-stigma from a long-term perspective. Lysaker, Roe, Ringer, Gilmore, and Yanos (2012) studied whether participation in rehabilitation is linked to declines in self-stigma, finding that decreases in self-stigma may be correlated with increased self-esteem, while higher levels of emotional distress may be a barrier to self-stigma reduction. Another study focused on people with severe mental illness found out that a change in self-stigma was significantly negatively related to change in social functioning (Yanos et al., 2012). It would be beneficial to study how psychiatric care (various types of care) contributes to factors such as self-esteem and emotional distress to be able to explain this outcome of our study.
Our analysis showed better health status, being married and a shorter period of time since initial contact with psychiatric care as predictors of internalized stigma levels. The evidence surrounding these correlates is mixed. A study conducted in the Czech Republic found no statistically significant differences in mean self-stigma levels between out-patient groups according to marital status, but however found strong statistically significant differences in self-stigma levels between patients with a partner, and without a partner (Grambal et al., 2016), an association that the findings from the current study does not support. Another study conducted in Ethiopia found that single marital status was significantly associated with higher ISMI scores in psychiatric out-patients (Assefa et al., 2012). The conflicting and limited understanding of internalized stigma and its association with marital status as well as partnership status will require further investigation to understand nuances surrounding the effects of patients’ personal relationships and levels of internalized stigma. There are also variables indicated as correlates of self-stigma identified through past research, which were not found significant in our population. For example, in European countries, significant predictors of higher ISMI scores included age group (50–59 years old), economic activity, lower number of social contacts, and lower self-efficacy/self-esteem and sense of power/powerlessness (Krajewski et al., 2013).
However, there are some limitations to this study. First, our results are not generalizable to all people who experiences episode(s) of psychosis since our sample consists only of community care users having spent a long time in psychiatric services (17.7 years on average). There is a possibility that prevalence and correlates of internalized stigma will be different within institutionalized patients or out-patients. Second, some relevant variables were not collected due to the fact that this study did not primarily focus on internalized stigma. Such variables are influenced by internalized stigma, including self-esteem, as well as current symptoms of psychosis which have been indicated as important predictors in past research (Aakre, Klingaman, & Docherty, 2015). Nevertheless, the robust sample size and randomized sampling of the current study contribute important findings in the context of Eastern Europe where studies with such methodology are scarce and provides interesting results relevant to self-stigma within community care settings.
In conclusion, approximately 25% of community care users reported high levels of self-stigma. With the aim of increasing the overall positive impact of community care settings in the Czech Republic, screening could be used to detect and target this sub-population for self-stigma reduction interventions in order to increase social and vocational functioning, activity levels, social power, self-esteem, self-efficacy, treatment adherence and help-seeking, while decreasing depressive symptoms and symptom severity.
Footnotes
Acknowledgements
The authors are grateful to all respondents as well as to those who collected the data.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study is a result of the research funded by the project no. LO1611 with financial support from the MEYS under the NPU-I programme.
