Psychiatry Research 239 (2016) 79–84

Contents lists available at ScienceDirect

Psychiatry Research

journal homepage: www.elsevier.com/locate/psychres

Negative cognitions about the self in patients with persecutory : An empirical study of self-compassion, self-stigma, schematic beliefs, self-esteem, fear of madness, and suicidal ideation

Nicola Collett a,n, Katherine Pugh b, Felicity Waite b, Daniel Freeman b a The Oxford Institute of Clinical Psychology Training, University of Oxford, Isis Education Centre, Roosevelt Drive, Warneford Hospital, Oxford, OX3 7JX, UK b Department of Psychiatry, University of Oxford, Warneford Hospital, Oxford, OX3 7JX, UK article info abstract

Article history: There has been growing awareness of the high prevalence of negative cognitions about the self in pa- Received 21 April 2015 tients with persecutory delusions, and it has been proposed that paranoid fears build upon these per- Received in revised form ceived vulnerabilities. This study aimed to investigate for the first time a wide range of different con- 25 January 2016 ceptualisations of the negative self, and to examine associations with suicidal ideation, in patients with Accepted 17 February 2016 persecutory delusions. Twenty-one patients with persecutory delusions and twenty-one non-clinical Available online 18 February 2016 individuals completed measures relating to negative self cognitions. The delusions group also completed Keywords: a measure of suicidal ideation. It was found that the patients with persecutory delusions had low self- compassion, low self-esteem, increased fears of being mad, beliefs of inferiority to others, negative self- schemas, and low positive self-schemas when compared to the non-clinical control group. The effect Suicidal ideation sizes (Cohen's d) were large, and the different conceptualisations of negative self cognitions were highly Negative self-concept Self-compassion associated with one another. Self-stigma did not differ between the two groups. Furthermore, suicidal Self-esteem ideation was highly associated with low self-compassion, low self-esteem, fears of madness, and negative self-schema but not self-stigma. This study shows marked negative self cognitions in patients with persecutory delusions. These are likely to prove targets of clinical interventions, with patient preference most likely determining the best conceptualisation of negative self cognitions for clinicians to use. & 2016 Elsevier Ireland Ltd. All rights reserved.

1. Introduction five concepts together in patients with persecutory delusions: 1). self-compassion, 2). schema, 3). self-stigma, 4). fears of madness Three recent literature reviews have all confirmed a connection and 5). self-esteem. Further, we examine their links to the clini- of to negative cognitions about the self (Garety and cally important problem of suicidal ideation in patients with Freeman, 2013; Kesting and Lincoln, 2013; Tiernan et al., 2014). It psychosis. has been hypothesised that paranoia builds upon the sense of vulnerability that low self-worth triggers (Freeman et al., 2005). 1.1. Five conceptualisations of negative self cognitions Consistent with this, a longitudinal study with 301 patients with fi psychosis indicated that negative self beliefs predict the later se- Self-compassion has been de ned by Neff (2003a) as com- verity of persecutory delusions (Fowler et al., 2012). Furthermore, prising three components: self-kindness versus self-judgement in two experimental studies by our group show that in individuals our responses to pain or failure, common humanity versus isola- tion when understanding one's own suffering, and being mindful with paranoid ideation the induction of negative self cognitions versus over-identification when paying attention to our own suf- leads to an increase in paranoia (Freeman et al., 2014a; Atherton fering (Neff, 2003a). Self-compassion involves an open, non-jud- et al., 2016). This has led to a clinical test of targeting negative self gemental stance to our own pain and suffering, which is viewed as beliefs in patients with persecutory delusions (Freeman et al., part of the human condition and promotes self-kindness (Neff, 2014b). Although there are multiple ways that negative self cog- 2003b). Two recent meta-analyses have found that self-compas- nitions can be conceptualised, studies have mainly focussed upon sion is an important explanatory variable in understanding psy- negative self beliefs and low self-esteem. In this paper we examine chopathology and wellbeing (MacBeth and Gumley, 2012; Zessin et al., 2015). n Corresponding author. Eicher et al. (2013) found in a study of 88 participants with E-mail address: [email protected] (N. Collett). either schizophrenia or that higher scores http://dx.doi.org/10.1016/j.psychres.2016.02.043 0165-1781/& 2016 Elsevier Ireland Ltd. All rights reserved. 80 N. Collett et al. / Psychiatry Research 239 (2016) 79–84

Table 1 Means, standard deviations, and ranges for the clinical group on the symptom measures.

Measure Mean SD Range

Psychotic Symptoms Rating Scale (PSYRATS) 16.86 3.14 9–22 and Deservedness Scale (PaDS) 76.62 19.03 17.5–100 Positive and Negative Syndrome Scale- PANSS (positive scale) 20.38 3.78 14–29 Positive and Negative Syndrome Scale- PANSS (negative scale) 14.95 6.03 8–29 Positive and Negative Syndrome Scale- PANSS (Total Score) 75.29 13.46 52–100 Beck Depression Inventory (BDI) 22.81 11.43 0–44 Beck Scale for Suicidal Ideation 12.76 9.51 0–34

on the self-compassion scale were negatively associated with po- unable to tell the difference between reality and imagination, sitive symptoms of psychosis. Qualitative accounts from patients worrying that the mind is falling apart, and a fear of being locked with psychosis highlight the potential role of self-compassion in up forever. Fear of recurrence of psychosis is a predictor of the facilitating recovery in psychosis (Waite et al., 2015). Current re- development of psychosis-related post-traumatic stress disorder search studies have found positive outcomes for compassion fo- (White and Gumley, 2009). Gumley et al. (2015) found that fear of cussed therapy for psychosis (Mayhew and Gilbert, 2008; Braehler recurrence of psychosis was associated with increased risk of re- et al., 2013). However, these studies have not focussed on reducing lapse and increased emotional distress, suggesting that in- symptoms of psychosis, and they do not focus specifically on dividual’s idiosyncratic appraisals may influence relapse. persecutory delusions. Mills et al. (2007) found in a study of 131 Finally, self-esteem has been considered as a central factor in students that paranoia was significantly negatively correlated with the occurrence of persecutory delusions, with differing views on scores on the self-compassion sc (Neff, 2003a), indicating in par- whether paranoia reflects a defence against low self-esteem ticular the presence of self-judgement, isolation, and over-identi- (Bentall et al., 2001) or is a direct reflection of low self-esteem fication in those with higher paranoia scores. Hutton et al. (2013) (Freeman et al., 2002). Multiple studies, but not all, demonstrate found that a group of 15 patients with persecutory delusions were that low self-esteem is present in patients with persecutory de- significantly less likely to self-reassure compared to non-clinical lusions (e.g. Combs et al., 2009; Freeman et al., 1998). For instance, controls. These authors propose that a decreased capacity to self- in a study of 154 participants who ranged across the continuum of reassure may activate perceptions of threat. Lincoln et al. (2013) paranoia, paranoia was found to be associated with lower self- found in an experimental study that a brief compassion focused esteem, and fluctuations in self-esteem were predictive of para- manipulation decreased paranoid thoughts in a non-clinical po- noia (Thewissen et al., 2008). Findings from studies on self-esteem pulation. However, no studies have yet specifically explored levels interventions for people with psychosis have demonstrated im- of self-compassion in a clinical group of people with persecutory provements in positive symptoms (Lecomte et al., 1999; Hall and delusions, although specific elements have been examined in one Tarrier, 2003) which may also lend support for a connection be- study (Hutton et al., 2013). tween self-esteem and positive symptoms of psychosis. Negative self-schema, defined as stable negative self evalua- tions which influence the individual’s interpretation of situations 1.2. Persecutory delusions and suicide (Beck et al., 1979), is a concept well-established in depression re- search (Beck, 1967). In contrast to self-compassion, their potential It is estimated that approximately 5% of patients with psychosis role in paranoia has been tested in cross-sectional (e.g. Gracie commit suicide (Hor and Taylor, 2010). Paranoid ideation in the et al., 2007; Smith et al., 2006), longitudinal (e.g. Fowler et al., general population too is associated with suicidal ideation (Free- 2011; Oliver et al., 2012; Freeman et al., 2013b) and experimental man et al., 2011). To our knowledge there are no psychological studies (e.g. Freeman et al., 2008), using the Brief Core Schema studies of specific psychological constructs that are associated Scale (Fowler et al., 2006). In a recent pilot randomised controlled with suicide specifically in patients with persecutory delusions. trial with patients with persecutory delusions, there was evidence But an association has been established between suicidal ideation that targeting negative self cognitions may lead to reductions in and hopelessness as predictors for attempted and completed sui- paranoia (Freeman et al., 2014b). To date negative self beliefs have cide in psychosis (Klonsky et al., 2012; King et al., 2001). Turner been most strongly connected to paranoia, compared to, for ex- et al. (2012) found an association between internal and external ample, self-esteem (Fowler et al., 2006). shame associated with psychosis, and symptoms of trauma in fifty The experience of mental health problems often leads to fur- patients with psychosis, who were not acutely psychotic at the ther difficulties for how people view themselves. Self stigma is the time. Internal shame was also associated with depression (Turner internalisation of harmful stereotypes, which results in a reduction et al., 2012), suggesting that shame may be a key emotion in de- in self-esteem (Link and Phelan, 2001; Corrigan and Watson, 2002; pression in psychosis. Fialko et al. (2006) found that suicidal Corrigan et al., 2011). Corrigan et al. (2011) propose that inter- ideation in people with psychosis was associated with depressed nalising stereotypes about mental illness, for example concerning mood, low self-esteem, and negative beliefs about the self and dangerousness and poor recovery rates, decreases self-esteem in others, pointing towards negative self-cognitions as an important individuals with serious mental health difficulties. In a study of 49 link to suicidal ideation. Overall, these findings highlight the im- people with schizophrenia, self-stigma was associated with low portance of considering suicidal ideation in relation to negative self-esteem (Rodrigues et al., 2013). No studies to date have ex- self-cognitions in individuals with persecutory delusions. plored self-stigma and persecutory delusions specifically. The current study set-out to examine these five different con- A specific type of self-stigma in relation to mental health pro- ceptualisations of negative self cognitions specifically in patients blems is the fear that one is ‘mad’, and in a cross-sectional study with persecutory delusions. It was hypothesised that patients with higher levels of fears of madness have been associated with persecutory delusions, compared to non-clinical controls, would greater distress associated with persecutory delusions (Bassett demonstrate negative self-schema, self-stigma, fears of madness, et al., 2009). Typical fears of madness are concerns that one is low self-compassion and low self-esteem. Of these constructs, N. Collett et al. / Psychiatry Research 239 (2016) 79–84 81

Table 2 Means, standard deviations, t values, confidence intervals, p values, and effect sizes for measures of negative self-cognition.

Measure Mean (SD) t Value (df) Confidence interval (95%) p Value Effect size (d)

Clinical Non-Clinical

Self Compassion Scale (SCS) 2.40 (0.53) 3.64 (0.92) 5.34 (40) 1.72 to0.77 o0.001 d¼1.60 Brief Core Schema Scale Positive self (BCSS-P) 6.60 (4.71) 14.95 (4.73) 5.65 (39) 11.34 to5.37 o0.001 d¼1.76 Social Comparison Scales (SCS) 608.37 (323.84) 1147.48 (337.40) 5.14 (38) 751.30 to326.91 o0.001 d¼1.78 Rosenberg Self Esteem Scale (RSES) 11.95 (5.63) 21.10 (4.49) 5.81 (40) 12.32 to5.96 o0.001 d¼1.79

Table 3 Mean ranks, mean scores, U scores, p values, and effect sizes for measures of negative self.

Measure Mean rank Mean scores (standard deviation) U score p Value Effect size (d)

Clinical Non-clinical Clinical Non- clinical

Brief Core Schema Scale Negative self (BCSS-N) 28.50 13.86 9.45 (5.02) 3.05 (4.98) 30.00 po0.001 1.28 (large) Self-Stigma of Mental Illness Scale (SSMIS) 23.58 17.71 31.21 (14.52) 25.19 (13.56) 141.00 p¼0.113 0.43 (medium) Mental Health Worries Questionnaire (MHWQ) 25.88 16.36 34.40 (16.41) 23.57 (6.17) 112.50 p¼0.010 0.88 (large)

Table 4 Correlations of suicide with negative self-constructs and depression.

123456789

1. Beck Suicide Scale (BSS) – 2. Self Compassion Scale (SCS) 0.64 – p¼0.002 3. Fear of Madness (MHWQ) 0.48 0.64 – p¼0.033 po0.001 4. Rosenberg Self-Esteem Scale 0.77 0.80 0.67 – (RSES) po0.001 po0.001 po0.001 5. Self-Stigma of Mental Illness 0.19 0.27 0.41 0.22 – Scale (SSMIS) p¼0.420 p¼0.095 p¼0.009 p¼0.165 6. Social Comparison Scale (SCS) 0.67 0.85 0.52 0.79 0.12 – p¼0.002 po0.001 po0.001 po0.001 p¼0.456 7. Brief Core Schema Scale (negative) 0.56 0.82 0.73 0.75 0.39 0.76 – p¼0.009 po0.001 po0.001 po0.001 p¼0.014 po0.001 8. Beck Depression Inventory (BDI) 0.54 0.73 0.72 0.79 0.46 0.66 0.66 – p¼0.012 po0.001 po0.001 po0.001 p¼0.002 po0.001 po0.001

associations with self-esteem were expected to be the smallest, Twenty one control participants were recruited from a participant since it has been argued that some studies have not shown asso- pool. The participant pool was developed from respondents to a ciations between clinical paranoia and low self-esteem (Bentall previous advert for participation in studies (Freeman et al., 2013b). et al., 2001). We did expect the different constructs to be highly The non-clinical control participants had no self-reported history correlated. Finally, we hypothesised that negative self cognitions of a mental health problem. They completed measures online in participants with persecutory delusions will be associated with using Qualtrics (Smith et al., 2011). suicidal ideation. 2.2. Measures

2. Method 2.2.1. Measures of psychotic experiences 2.2.1.1. Psychotic Symptom Rating Scale – Delusions (PSYRATS) A cross-sectional, between-groups comparison was conducted. (Haddock et al., 1999). The PSYRATS is a 6 item assessment of preoccupation, conviction, distress and disruption associated with 2.1. Participants the over the previous fortnight. Higher scores indicate greater delusional severity. The PSYRATS has good reliability and Twenty-one participants with persecutory delusions were re- validity for the assessment of delusions in participants with psy- cruited from clinical teams in Oxford Health NHS Foundation chosis (e.g. Drake et al., 2007). Trust. Inclusion criteria were: experience of a current persecutory delusion as defined by Freeman and Garety (2000); a clinical di- 2.2.1.2. The Positive and Negative Syndrome Scales (PANSS) (Kay agnosis of non-affective psychosis; and aged eighteen years or et al., 1987). The PANSS is a 30 item interviewer-rated tool to as- over. Participants were excluded if English was not their first sess psychiatric symptom severity present in the previous 72 h. language, or if they were unable to complete questionnaire mea- Ratings are delivered using a 7 point scale (1-7). Higher scores are sures. The clinical group completed measures in an assessment indicative of increased symptom severity. Kay (1991) found good session with the first author. reliability and validity for the PANSS in a cohort of 101 participants The non-clinical control group was matched by age and gender. with schizophrenia. 82 N. Collett et al. / Psychiatry Research 239 (2016) 79–84

2.2.1.3. Persecution and Deservedness Scale (PaDS) (Melo et al., disagree”) is used for scoring. Items are totalled into an overall 2009). This is a 10 item self-report measure of the severity of scale which has been shown to have good internal consistency in paranoid thinking and the level of perceived deservedness for the individuals with serious mental health problems (Corrigan et al., persecution. Only the section on persecution was employed in this 2011; Corrigan et al., 2006). study. Higher scores represent greater levels of persecutory idea- tion. The PaDS demonstrates high validity and reliability and is 2.2.2.6. Social Comparison Scale (SCS) (Allan and Gilbert, validated for use in clinical samples (Melo et al., 2009). The scoring 1995). Participants’ beliefs about how they compare to others for this measure uses a percentage as described by Melo and were measured using an adapted version of the Social Comparison Bentall (2010). Scale. This 19 item self-report scale asks participants to rate how they have felt in relation to others during the previous week, on 2.2.2. Measures of negative self cognitions items including inferior/superior and incompetent/competent. 2.2.2.1. The Self-Compassion Scale (SCS) (Neff, 2003a). The SCS is a Eleven items were from the original scale (e.g. inferior versus su- 26 item self-report scale comprising statements (e.g. “I try to be perior) and eight items (e.g. failure versus success) were newly loving towards myself when I am feeling emotional pain”) for added. which responses can range from 1 (Almost Never) to 5 (Almost Always). An overall self-compassion score is provided, as well as 2.2.3. Measures of depression and suicidal ideation scores for sub-scales including self-judgement, isolation, over- 2.2.3.1. Beck Depression Inventory (BDI) (Beck, 1967). The BDI is a identification, self-kindness, common humanity, and mindfulness. 21-item self-report measure of symptoms of depression over the Scores for total self compassion were calculated by reverse scoring past fortnight. Each item involves selecting one of four statements, negative subscale items (self-judgement, isolation and over iden- with scores for each varying between 0 and 3. Higher total scores tification) before calculating subscale means for all six subscales. indicate higher levels of depression. Items tap cognitive distor- An overall mean of all six subscale means was then computed, as tions, hopelessness, suicidal ideation and physiological symptoms suggested by Neff (2003a). Higher total self-compassion scores including fatigue. The BDI demonstrates good construct validity indicate greater self-compassion in times of distress. The SCS de- and reliability (Beck et al., 1988). monstrates good construct validity. 2.2.3.2. Beck Scale for Suicidal Ideation (BSS) (Beck and Steer, 2.2.2.2. The Brief Core Schema Scale (BCSS) (Fowler et al., 1991). This is a 21-item self-report measure of thoughts, attitudes, 2006). The BCSS is a 24 item self-report measure which assesses and intentions about suicide. Items are rated on a three point scale beliefs about self and others. Items are rated using a 0–4 scale, and (0 to 2), with a total possible score of 38. Higher total scores in- comprises four factors (negative self, positive self, negative others dicate higher levels of suicidal thinking. The measure has good and positive others) with a score given for each factor. Only the internal reliability (0.96) and one study showed that the BSI had a items related to self were included in this study. For the six items test re-test reliability of 0.88 (po0.001) (Pinninti et al., 2002). In on negative self (e.g. “I am worthless”), the greater the total on this addition, Pinninti et al. (2002) found that the BSI was positively subscale, the greater the negative beliefs about the self. For the correlated with past suicide attempts (r¼0.46, po0.001) and re- positive subscale (e.g. “I am respected”), the larger the total, the commended it for use with participants with psychosis. more positive beliefs a person holds. Addington and Tran (2009) concluded that the psychometric properties of the BCSS make it 2.3. Analysis appropriate for use in this population. Data were analysed using the Statistics Package for Social Sci- 2.2.2.3. Self-Stigma of Mental Illness Scale (SSMIS) (Corrigan et al., ences (SPSS version 20) (IBM, 2011). For the first hypothesis, t-tests 2006). This 40 item measure comprises four subscales. Only the and Mann Whitney U tests were employed to test group differ- agreement subscale, which denotes the level to which people ences on the self-concept measures, depending on whether as- agree with stereotypes of mental illness, was employed in this sumptions for the use of parametric tests were met. A Cohen’sd study (e.g. “I think most persons with mental illness are to blame calculation was used to determine effect sizes, in order to establish for their problems”). This was to investigate stigma about mental which negative self-concepts showed the largest group differ- illness amongst people with and without persecutory delusions. ences. For the second main hypothesis, associations with suicidal Good internal consistency and concurrent validity for the measure ideation for the self-concept measures within the clinical group has been established (Corrigan et al., 2006). Responses are given were tested using Pearson’s r. Correlation coefficients were simi- on a 9 point scale (9¼strongly agree). Higher total scores indicate larly used to examine associations between the individual self- greater endorsement of self-stigma (Corrigan et al., 2011). concept measures. Power calculations to establish the required sample size were based on the Mental Health Worries Ques- 2.2.2.4. Mental Health Worries Questionnaire (MHWC)(Bassett tionnaire (MHWQ) data from Bassett et al. (2009), which was the et al., 2009). This is a 20 item self-report measure of fears of most similar study. It showed effect sizes (d) of 0.80. 80% power to madness and associated distress in the previous week (e.g. “I detect such an effect size at the conventional significance level of worry my mind is falling apart”). Responses relate to levels of 0.05 requires a sample size of 26 (G Power; Faul et al., 2007). All distress associated with fears ranging from 0 (not at all) to 5 hypothesis testing was two-tailed. (extremely distressing) for each statement item. Initial findings have shown good reliability and validity for the scale (Bassett et al., 2009). Higher total scores represent greater levels of dis- 3. Results tressing fears about madness. 3.1. Participant characteristics 2.2.2.5. Rosenberg Self-esteem Scale (RSES) (Rosenberg, 1965). The RSES is the most used self-esteem measure and comprises ten For the clinical group (n¼21) the mean age was 45.6 years items (e.g. “I feel that I’m a person of worth, at least on an equal (SD¼12.1; range: 21–66), compared to a mean age of 41.9 years plane with others.”). The higher the RSES score the lower self- (SD¼12.2; range: 22–61) in the control group (n¼21). The gender esteem. A four point Likert scale of agreement (4- “strongly balance was the same in both groups (males n¼10; 48% and N. Collett et al. / Psychiatry Research 239 (2016) 79–84 83 females n¼11; 52%). More participants in the clinical group were delusions were high, even compared to other studies of people unemployed (n¼18; 86%), compared with the control group with psychosis (e.g. Beck and Steer, 1991). Suicidal ideation was (n¼5; 24%). No significant difference was found between partici- highly associated with low self-esteem, negative self-schema, fears pants with persecutory delusions and control participants for age of madness, and lower levels of self-compassion. The evidence of (t¼10.88, df¼30.06, p¼10.38) or gender ( x2 ¼0.00, df¼1, links between suicidal ideation and a range of negative self views p¼0.74). However, there was a significant difference between the is entirely plausible, and consistent with previous research that groups for employment status, with significantly fewer employed has utilised fewer self-concept assessments (e.g. Fialko et al., participants in the delusion group (x2 ¼14.40, df¼1, p o0.001). 2006). Table 1 summarises the clinical characteristics of the persecu- Several important limitations must be considered in this study. tory delusions group. For the PSYRATS, the clinical group scored The cross-sectional design means that it is not possible to de- comparably to other studies with patients selected for current termine cause and effect, and it may be argued by some that the persecutory delusions (e.g. Freeman et al., 2014b, 2014c). The findings related to negative self-cognitions in this study are simply scores on the PaDS for levels of persecutory ideation in this group a result of the high levels of depression and/or paranoia in the were similar to those in a recent study by Melo and Bentall (2013), persecutory delusions group (i.e. the cognitions themselves are which found that the mean persecution score in a similar group secondary to affect). One consideration was whether to conduct an was 70.23% (SD¼17.37). The Beck Depression Inventory (BDI) was analysis controlling for level of depression, but it has been argued completed by both the group with persecutory delusions that this would be highly likely to remove genuine variance of (mean¼22.81; SD¼11.43) and the control group (mean¼6.43; interest (see Miller and Chapman, 2001). The sample size also was SD¼6.96) and a significant group difference was clearly apparent, insufficient to detect anything other than large differences be- U¼47.00, po0.001. tween the measures. Further, although the groups were well matched for age and gender, the groups were clearly biased in 3.2. The self-concept in patients with persecutory delusions terms of employment status, which would affect the scores. Future research would benefit from longitudinal studies in larger sample Tables 2 and 3 summarise the group scores on the measures of sizes. Although we believe that the pressing issue now is how best self-concept. It can be seen that there were significant and clear to treat negative self-cognitions in patients with psychosis (Free- group differences for all measures, apart from self-stigma. The man et al., 2014b, 2014c). A number of small studies have in- persecutory delusion group had many more negative self-cogni- vestigated treatment of negative self-cognitions, including self- tions and fewer positive self-cognitions. The effect size differences compassion (Mayhew and Gilbert, 2008; Braehler et al., 2013) and were all large. self-esteem (Hall and Tarrier, 2003). However, these studies have not tested improvements in persecutory delusions specifically. The fi 3.3. Suicidal ideation and negative self current ndings suggest that there are several different ways in which negative self-cognitions associated with persecutory delu- fi Correlational analyses were conducted in order to investigate sions may be framed. No single construct was identi ed as the key potential associations between suicidal ideation and the measures treatment direction, however, negative self-schema, self-compas- of self-cognitions (see Table 4). Suicidal ideation was highly asso- sion, self-esteem and negative self-comparisons to others were ciated with low self-compassion, low self-esteem, negative self- found to be most important. All constructs were highly inter-cor- schema, and negative self-comparisons to others. Fears of madness related, with the exception of self-stigma, which did not correlate fl and depression were also significantly related to suicidal ideation. with negative self-comparisons to others. Therefore, a exible, fi Self-stigma was not related to current levels of suicidal ideation. It individually tailored approach to improving self con dence in this can also be seen in Table 4 that all the self-concept assessments, patient group may be possible. We suspect that at this stage of apart from self-stigma, were highly correlated with each other. knowledge, the language and concepts that best resonate with an individual patient may prove best to use in treatment.

4. Discussion Acknowledgements The current study investigated a number of different con- ceptualisations of the negative self in patients with persecutory DF is supported by a National Institute for Health Research delusions, and also examined associations with the clinically im- (NIHR) Research Professorship (Grant Number: RP-2014-05-003). portant problem of suicidal ideation in this population. Findings suggested that negative self-cognitions in those with persecutory delusions are marked and globally low, however they are con- References ceptualised. The findings in the delusions group are in line with previous studies (e.g. Fowler et al., 2012; Hutton et al., 2013). Pa- Addington, J., Tran, L., 2009. Using the brief core schema scales with individuals at – tients with persecutory delusions experience severe feelings of clinical High risk of psychosis. Behav. Cogn. Psychother. 37, 227 230. Atherton, S., Antley, A., Evans, N., Cernis, E., Lister, R., Dunn, G., Slater, M., Freeman, D., being inferior to others, worry that they are mad, and have lower 2016. Self-confidence and paranoia: an experimental study using an immersive self-compassion. They have increased negative beliefs about virtual reality social situation. Behav. Cogn. Psychother. 44 (1), 56–64. themselves, decreased positive beliefs about themselves, low self- Allan, S., Gilbert, P., 1995. A social comparison scale: psychometric properties and relationship to psychopathology. Pers. Indiv. Diff 19, 293–299. esteem, and low mood. The only construct which did not sig- Bassett, M., Sperlinger, D., Freeman, D., 2009. Fear of madness and persecutory nificantly differ between the two groups was self-stigma, which delusions: preliminary investigation of a new scale. Psychosis 1, 39–50. was an unexpected finding, although the relatively small sample Beck, A.T., 1967. Depression: Clinical Experimental, and Theoretical Aspects. Harper and Row, New York. size limited the power of the study to the detection of large effect Beck, A.T., Rush, A.J., Shaw, B.F., Emery, G., 1979. Cognitive Theory of Depression. size differences only. We conclude that negative self-concept in Guilford, New York. patients with persecutory delusions are a clear treatment target, Beck, A.T., Steer, R.A., Carbin, M.G., 1988. Psychometric properties of the beck de- pression inventory: twenty five years of evaluation. Clin. Psychol. Rev. 8, and may well lead to consequent improvement in the paranoia. 77–100. The levels of suicidal ideation in the group with persecutory Beck, A.T., Steer, R.A., 1991. Use of the beck scale for suicidal ideation with 84 N. Collett et al. / Psychiatry Research 239 (2016) 79–84

psychiatric inpatients diagnosed with schizophrenia, schizoaffective or bipolar (PSYRATS). Psychol. Med. 29, 879–889. disorders. Behav. Res. Ther. 40 (9), 1071–1079. Hall, P., Tarrier, N., 2003. The cognitive-behavioural treatment of low self-esteem in Bentall, R.P., Corcoran, R., Howard, R., Blackwood, N., Kinderman, P., 2001. Persec- psychotic patients: a pilot study. Behav. Res. Ther. 41, 317–332. utory delusions: a review and theoretical integration. Clin. Psych. Rev. 21 (8), Hor, K., Taylor, M., 2010. Suicide and schizophrenia: a systematic review of rates 1143–1192. and risk factors. J. Psychopharmacol. 24 (4), 81–90. Braehler, C., Gumley, A., Harper, J., Wallace, S., Norrie, J., Gilbert, P., 2013. Exploring Hutton, P., Kelly, J., Lowens, I., Taylor, P., Tai, S., 2013. Self-attacking and self-re- change processes in compassion focussed therapy in psychosis: results of a assurance in persecutory delusions: a comparison of healthy, depressed and feasibility randomised controlled trial. Br. J. Clin. Psychol. 52, 199–214. paranoid individuals. Psychiatry Res. 205, 127–136. Combs, D.R., Penn, D.L., Michael, C.O., Basso, M.R., Wiedeman, R., Siebenmorgan, M., IBM Corp, 2011. IBM SPSS Statistics for Windows, Version 20. IBM Corp, Armonk, Chapman, D., 2009. Perceptions of hostility by persons with and without per- NY, p. 0. secutory delusions. Cogn. Neuropsychiatry 14 (1), 30–52. Kay, S.R., Flszbein, A., Opfer, L.A., 1987. The positive and negative syndrome scale Corrigan, P.W., Watson, A.C., 2002. The paradox of self-stigma and mental illness. (PANSS) for schizophrenia. Schizophr. Bull. 13 (2), 261. Clin. Psychol. Sci. Pract. 9, 35–53. Kay, S.R., 1991. Positive and Negative Syndromes in Schizophrenia: Assessment and Corrigan, P.W., Watson, A.C., Barr, L., 2006. The self-stigma of mental illness: im- Research (No. 5). Psychology Press, UK. plications for self-esteem and self-efficacy. J. Soc. Clin. Psychol. 25 (8), 875–884. Kesting, M.L., Lincoln, T.M., 2013. The relevance of self-esteem and self-schemas to Corrigan, P.W., Rafcaz, J., Rusch, N., 2011. Examining a progressive model of self- persecutory delusions: a systematic review. Compr. Psychiatry 54 (7), 766–789. stigma and its impact on people with serious mental illness. Psychiatry Res. 89, King, E.A., Baldwin, D.S., Sinclair, J.M., 2001. The Wessex recent in-patient suicide 340–345. study, I. Case—control study of 234 recently discharged psychiatric patient Drake, R., Haddock, G., Tarrier, N., Bentall, R., Lewis, S., 2007. The psychotic suicides. Br. J. Psychiatry 178, 531–536. symptoms rating scales (PSYRATS): their usefulness and properties in first Klonsky, D.E., Kotov, R., Bakst, S., Rabinowitz, J., Bromet, E.J., 2012. Hopelesness as a episode psychosis. Schizophr. Res. 89, 119–122. predictor of attempted suicide among first admission patients with psychosis: a Eicher, A.C., Davis, L.W., Lysaker, P.H., 2013. Self-compassion: a novel link with 10 year cohort study. Suicide Life Threat. Behav. 42, 1–10. – symptoms in schizophrenia? J. Nerv. Ment. Dis. 201 (5), 1 5. Lecomte, T., Cyr, M., Lesage, A., Wilde, J., Leclere, C., Ricard, N., 1999. Efficacy of a fl Faul, F., Erdfelder, E., Lang, A.G., Buchner, A., 2007. G*Power 3: A exible statistical self-esteem module in the empowerment of individuals with chronic schizo- power analysis program for the social, behavioral, and biomedical sciences. phrenia. J. Nerv. Ment. Dis. 187, 406–413. – Behav. Res. Methods 39 (2), 175 191. Lincoln, T.M., Hohenhaus, F., Hartmann, M., 2013. Can paranoid thoughts be re- Fialko, L., Freeman, D., Bebbington, P.E., Kuipers, E., Garety, P.A., Dunn, G., Fowler, D., duced by targeting negative emotions and self-esteem? An experimental in- fi 2006. Understanding suicidal ideation in psychosis: ndings from the psy- vestigation of a brief compassion-focused intervention. Cogn. Ther. Res. 37 (2), chological prevention of relapse in psychosis (PRP) trial. Acta Psychiatr. Scand. 390–402. – 114 (3), 177 186. Link, B.G., Phelan, J.C., 2001. Conceptualising stigma. Annu. Rev. Sociol. 27, 363–385. Fowler, D., Freeman, D., Smith, B., Kuipers, E., Bebbington, P., Bashforth, H., Coker, S., MacBeth, A., Gumley, A., 2012. Exploring compassion: a meta-analysisof the asso- Hodgekins, J., Gracie, A., Dunn, G., Garety, P., 2006. The brief core schema scale ciation between self-compassion and psychopathology. Clin. Psychol. Rev. 32 (BCSS): psychometric properties and associations with paranoia and grandi- (6), 545–552. – osity in non-clinical and psychosis samples. Psychol. Med. 36, 749 759. Mayhew, S.L., Gilbert, P., 2008. Compassionate mind training with people WHO Fowler, D., Hodgekins, J., Garety, P., Freeman, D., Kuipers, E., Dunn, G., Smith, B., hear malevolent voices: a case series report. Clin. Psychol. Psychother. 11, Bebbington, P.E., 2011. Negative cognition, depressed mood, and paranoia: a 113–138. longitudinal pathway analysis using structural equation modelling. Schizophr. Melo, S., Corcoran, R., Shryane, N., Bentall, R.P., 2009. The persecution and de- Bull. 38, 1063–1073. servedness scale. Psychol. Psychother. Theory Res. Pract. 82 (3), 247–260. Freeman, D., Evans, N., Lister, A., Antley, A., Dunn, G., Slater, M., 2014a. Height, social Melo, S., Bentall, R.P., 2010. Coping in subclinical paranoia: a two nations study. comparison and paranoia: an immersive virtual reality experimental study. Psychol. Psychother. Theo. Res. Pract. 83, 407–420. Psychiatry Res. 218 (37), 348–362. Miller, G.A., Chapman, J.P., 2001. Misunderstanding analysis of covariance. J. Ab- Freeman, D., Garety, P.A., 2000. Comments on the content of persecutory delusions: norm. Psychol. 110 (1), 40–48. does the definition need clarification? Br. J. Clin. Psychol. 39, 407–414. Mills, A., Gilbert, P., Bellew, R., McEwan, K., Gale, C., 2007. Paranoid beliefs and self Freeman, D., Garety, P.A., Bebbington, P.E., Smith, B., Rollinson, R., Fowler, D., Kui- criticism in students. Clin. Psychol. Psychother. 14, 358–364. pers, E., Ray, K., Dunn, G., 2005. Psychological investigation of the structure of Neff, K., 2003a. Development and validation of a scale to measure self-compassion. paranoia in a non-clinical population. Br. J. Psychiatry 186 (5), 427–435. Self Identity 2, 223–250. Freeman, D., Garety, P.A., Fowler, D., Kuipers, E., Dunn, G., Bebbington, P., Hadley, C., Neff, K., 2003b. Self-compassion: an alternative conceptualisation of a healthy at- 1998. The London-East Anglia randomised controlled trial of cognitive-beha- titude towards oneself. Self Identity 2, 85–102. viour therapy for psychosis IV: self-esteem and persecutory delusions. Br. J. Oliver, J.E., O’Connor, J.A., Jose, P.E., McLachlan, K., Peters, E., 2012. The impact of Clin. Psychol. 37 (4), 415–430. negative schemas, mood and psychological flexibility on delusional ideation- Freeman, D., Garety, P.A., Kuipers, E., Fowler, D., Bebbington, P.E., 2002. A cognitive mediating and moderating effects. Psychosis: Psychol., Soc. Integr. Approach 4, model of persecutory delusions. Br. J. Clin. Psychol. 41, 331–347. – Freeman, D., McManus, S., Brugha, T., Meltzer, H., Jenkins, R., Bebbington, P., 2011. 6 18. Pinninti, N., Steer, R.A., Rissmiller, D.J., Nelson, S., Beck, A.T., 2002. Use of the beck Concomitants of paranoia in the general population. Psychol. Med. 41, 923–936. Freeman, D., Morrison, P.D., Murray, R.M., Evans, N., Lister, R., Dunn, G., 2013a. scale for suicide ideation with psychiatric inpatients diagnosed with schizo- Persecutory ideation and a history of cannabis use. Schizophr. Res. 148, phrenia, schizoaffective or bipolar disorders. Behav. Res. Ther. 40 (9), – 122–125. 1071 1079. Freeman, D., Pugh, K., Antley, A., Slater, M., Bebbington, P., Gittins, M., Dunn, G., Rodrigues, S., Serper, M., Novak, S., Corrigan, P., Hobart, M., Ziedonis, M., Smelson, Kuipers, E., Fowler, D., Garety, P., 2008. Virtual reality study of paranoid D., 2013. Self-Stigma, self-esteem, and co-occurring disorders. J. Dual Diagn. 9 – thinking in the general population. Br. J. Psychiatry 192 (4), 258–263. (2), 129 133. Freeman, D., Pugh, K., Dunn, G., Evans, N., Sheaves, B., Waite, F., Cernis, E., Lister, R., Rosenberg, M., 1965. Rosenberg self-esteem scale (RSES). Acceptance and Com- Fowler, D., 2014b. An early phase II randomised controlled trial testing the mitment Therapy. Measures Package, pp. 61. effect on persecutory delusions of using CBT to reduce negative cognitions Smith, B., Fowler, D.G., Freeman, D., Bebbington, P., Bashforth, H., Garety, P., Dunn, about the self: the potential benefits of enhancing self confidence. Schizophr. G., Kuipers, E., 2006. Emotion and psychosis: links between depression, self Res. 160 (1–3), 186–192. esteem, negative schematic beliefs and delusions and hallucinations. Schizophr. Freeman, D., Startup, H., Dunn, G., Wingham, G., Cernis, E., Evans, N., Lister, R., Pugh, Res. 86 (1–3), 181–188. K., Cordwell, J., Kingdon, D., 2014c. Persecutory delusions and psychological Smith, S., Smith, R., Smith, J., Orgill, J., 2011. Qualtrics. Provo, Utah, USA. 〈www. well-being. Soc. Psychiatry Psychiatr. Epidemiol. 49 (7), 1045–1050. qualtrics.com〉. Freeman, D., Thompson, C., Vorontsova, N., Dunn, G., Carter, L.A., Garety, P., Kuipers, Thewissen, V., Bentall, R.P., Lecomte, T., van Os, J., Myin-Germeys, I., 2008. Fluc- E., Slater, M., Antley, A., Glucksman, E., Ehlers, A., 2013b. Paranoia and post- tuations in self-esteem and paranoia in the context of daily life. J. Ab. Psych. 117, traumatic stress disorder in the months after a physical assault: a longitudinal 143–153. study examining shared and differential predictors. Psychol. MEd. 43 (12), Tiernan, B., Tracey, R., Ciaran, S., 2014. Paranoia and self-concepts in psychosis: a 2673–2684. systematic review of the literature. Psychiatry Res. 216 (3), 303–313. Garety, P.A., Freeman, D., 2013. The past and future of delusions research: from the Turner, M.H., Bernard, M., Birchwood, M., Jackson, C., Jones, C., 2012. The con- inexplicable to the treatable. Br. J. Psychiatry 203 (5), 327–333. tribution of shame to post-psychotic trauma. Br. J. Clin. Psychol. 52 (2), Gracie, A., Freeman, D., Green, S., Garety, P.A., Kuipers, E., Hardy, A., Ray, K., Dunn, 162–182. G., Bebbington, P., Fowler, D., 2007. The association between traumatic ex- Waite, F., Knight, M.T.D., Lee, D., 2015. Self-compassion and self-criticism in re- perience, paranoia and hallucinations: a test of the predictions of psychological covery in psychosis: an interpretative phenomenological analysis study. J. Clin. models. Acta Psychiatr. Scand. 116 (4), 280–289. Psychol. 71 (12), 1201–1217. Gumley, A.I., MacBeth, A., Reilly, J.D., O’Grady, M., White, R.G., McLeod, H., White, R.G., Gumley, A.I., 2009. Post-psychotic PTSD: associations with fear of re- Schwannauer, M., Power, K.G., 2015. Fear of recurrence: results of a randomized currence and intolerance of uncertainty. J. Nerv. Ment. Dis. 197, 841–849. trial of relapse detection in schizophrenia. Br. J. Clin. Psychol. 54, 49–62. Zessin, U., Dickhauser, O., Garbade, O., 2015. The relationship between self-com- Haddock, G., McCarron, J., Tarrier, N., Faragher, E.B., 1999. Scales to measure di- passion and well-being: a meta-analysis. Appl. Psychol. Health Well Being 7 (3), mensions of hallucinations and delusions: the psychotic symptom rating scales 340–364.