Introduction

People with mental illness have to cope with widespread stigma and discrimination [1]. Since having a mental illness often is a concealable stigmatized identity, individuals face the dilemma of whether or not to disclose their condition to others. Depending on the circumstances, disclosure may lead to rejection or to social support; non-disclosure or secrecy on the other hand may protect from discrimination, but is associated with negative long-term outcomes and the threat of discovery can be a constant stressor [2–4]. This has led to the development of interventions meant to support people with mental illness in their disclosure decisions, either as decision aids [5] or as manualised peer-led group interventions [6, 7].

Stigma is a complex phenomenon that includes past experiences of discrimination and the anticipation of future discrimination by others [8, 9]. Stigma also affects internal cognitive and emotional processes in terms of stress-coping reactions and self-concept. Stigma stress occurs if people with mental illness believe that stigma-related harm exceeds their coping resources [10, 11]; self-stigma implies that individuals agree with public stereotypes and internalize them, leading to low self-esteem and demoralization [12, 13].

It remains unclear which aspects of stigma and discrimination are related to disclosure-related distress versus to being comfortable with disclosure. To further improve the above-mentioned interventions, this study had the primary aim to identify the aspects of the stigma process that are most strongly associated with disclosure-related distress as targets for interventions. We expected higher levels of stigma variables to be associated with reduced comfort disclosing and wanted to examine four different indices of experiencing and reacting to stigma and discrimination (anticipated discrimination; experienced discrimination; stigma stress; self-stigma) as variables that are associated with the comfort disclosing one’s mental illness to friends and family. As a secondary aim, we tested the hypothesis that more comfort disclosing would be related to greater well-being, controlling for symptom levels.

Methods

Participants and recruitment

This research is part of the Mental Illness-Related Investigations on Discrimination study, a cross-sectional survey of individuals using secondary mental health services [14]. The study received ethical approval by the East of England/Essex 2 Research Ethics Committee (11/EE/0052) and has been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. All participants provided written informed consent after being fully informed about study procedures and prior to their inclusion in the study. Two hundred and two participants were recruited from community mental health teams in London, UK. Nearly half (n = 96, 48 %) had a diagnosis of a schizophrenia spectrum disorder, 20 % (n = 41) bipolar disorder and 32 % (n = 65) depression. Over half (n = 110, 55 %) were female; the mean age was 42 years; 55 % (n = 110) reported Black or mixed ethnicity; and 23 % (n = 46) were in employment. Details of recruitment and participant characteristics are described elsewhere [14].

Measures

The dependent variable for our analyses, the level of comfort participants felt with disclosing a mental illness to family and friends, was assessed by one item that has been used in previous studies [15, 16] (see footnote of Table 1 for wording and scoring; in our study: M = 3.7, SD = 2.3). Independent variables were examined as follows: The level of anticipated discrimination was assessed by the Questionnaire on Anticipated Discrimination [17], a 14-item self-report scale assessing the extent to which participants expect to be treated unfairly in different life domains. The scale yields a mean score between 0 and 3, with higher scores equalling greater anticipated discrimination (in our study M = 1.5, SD = 0.5, Cronbach’s alpha 0.88). The cognitive appraisal of mental illness stigma as a stressor was assessed by a previously validated 8-item Stigma Stress Scale [10, 18, 19], scored from 1 to 7 with higher scores equaling higher agreement. Four items assessed the primary appraisal of mental illness stigma as harmful (e.g. “Prejudice against people with mental illness will have harmful or bad consequences for me”; in our study M = 4.8, SD = 1.7; Cronbach’s alpha 0.90) and four items the secondary appraisal of perceived resources to cope with stigma (e.g. “I have the resources I need to handle problems posed by prejudice against people with mental illness”; in our study M = 5.0, SD = 1.5; Cronbach’s alpha 0.82). A single stress appraisal score was computed by subtracting perceived resources from perceived harmfulness. A higher difference score indicates the appraisal of stigma as more stressful, exceeding personal coping resources.

Table 1 Multiple linear regression examining socio-demographic, clinical and stigma variables as predictors of comfort disclosing one’s mental illness (R 2 = 0.291)
Full size table

The level of experienced discrimination in the last 12 months was assessed using the 21-item unfair treatment subscale (http://www.sapphire.iop.kcl.ac.uk/SAPPHIRE%20Resources.html#DISC) of the Discrimination and Stigma Scale [20] yielding a count of life domains in which individuals experienced discrimination (in our study M = 4.6, SD = 3.4, range 0–15). Self-stigma was assessed by the 29-item Internalized Stigma of Mental Illness Inventory [21] yielding a mean score between 1 and 4 with higher scores indicating more self-stigma (in our study M = 2.3, SD = 0.5; alpha 0.88). Mental well-being was measured using the 14-item Warwick-Edinburgh Mental Well-being Scale (recommended by users of secondary mental health services: http://www.mhrn.info/pages/mhrn-news-for-researchers.html) [22] with higher sum scores, ranged from 14 to 70, indicating more well-being (in our study M = 41.2, SD = 11.6; alpha 0.94). Psychiatric symptoms were assessed by the 18-item version of the Brief Psychiatric Rating Scale [23] with items rated from 1 to 7 and a sum score between 18 and 126, with higher scores equaling more symptoms (M = 35.6, SD = 11.2). Clinical diagnoses were taken from electronic patient records.

Analyses

We ran a multiple linear regression model to examine variables associated with comfort disclosing one’s mental illness to friends and family as dependent variable. Anticipated discrimination, stigma stress, experienced discrimination and self-stigma were entered as independent variables. In the regression, we controlled for socio-demographic and clinical variables (Table 1). Because the use of one stigma stress difference score as an independent variable could impose undue equality constraints [24, 25], we repeated the regression with the two appraisal scores as independent variables instead of the one difference score. If R 2 in the latter regression does not significantly increase, it suggests there are no undue equality constraints and the difference score can be used as independent variable. The association of comfort disclosing with well-being was examined by bivariate and partial correlations, the latter controlling for psychiatric symptoms. All analyses were run in SPSS 20, and results were considered significant at p < 0.05.

Results

Greater comfort disclosing was associated with lower levels of anticipated discrimination and with less stigma-related stress (Table 1). Psychiatric inpatient treatment in the last year was associated with reduced comfort disclosing. Comfort with disclosure was not significantly related to past experiences of discrimination, self-stigma, socio-demographic or other clinical variables. Multi-collinearity in the regression model was acceptable with all variance inflation factors below 1.9 and all tolerance values above 0.5. The regression explained more than a quarter of the variance in disclosure comfort (R 2 = 0.291). With both appraisal scores as two independent variables in the equation, instead of one stigma stress difference score, both scores were significantly associated with disclosure comfort, and the amount of variance explained by the model remained unchanged (R 2 = 0.291).

With respect to our second aim, we found a significant positive correlation between comfort disclosing and mental well-being (r = 0.27, p < 0.001). In a partial correlation, this association remained significant after controlling for levels of psychiatric symptoms as assessed by the Brief Psychiatric Rating Scale (r = 0.16, p = 0.026).

Discussion

A combination of both anticipated external threats (discrimination) and internal processes (stigma stress appraisals) render individuals more vulnerable to disclosure-related distress which in turn is associated with reduced well-being. Disclosure appears to be harder for individuals with recent psychiatric inpatient treatment, perhaps because this is an indicator of severe and hence more stigmatising illness. Anticipated future discrimination may have a greater impact on disclosure than discrimination experienced in the past [8]. Since disclosure is often risky and thus potentially stressful, stigma stress appraisals could be more closely related to disclosure (dis)comfort than to self-stigma [6]. The link between more comfort disclosing and greater well-being is consistent with recent findings on perceived benefits of disclosure and better quality of life [26].

Limitations of our study need to be considered. Our data are cross-sectional and conclusions on causality are limited. The findings are based on a large sample from an urban area in the UK but cannot be generalised to other settings. Participants were current service users and disclosure concerns may be higher among people with mental illness not in contact with services. The well-being scale has not yet been validated among people with mental illness. Finally, comfort with disclosure was measured as a general trait and does not necessarily reflect actual disclosure decisions in specific situations.

In conclusion, we found that anticipated discrimination by others as an external threat as well as the perception that one’s resources to cope with discrimination are insufficient may lead to reduced comfort disclosing one’s mental illness to friends and family. This is consistent with recent models of disclosure [27] and has implications for intervention research. We need initiatives that reduce public stigma [28, 29] as well as support for people in their disclosure decisions and in their efforts to cope with stigma as a stressor [5, 6].