خشم بعنوان پیش بینی کننده پریشانی روان شناختی و افکار خودآسیبی در زندانیان: مطالعه خاطرات ساختار خود ارزیابی

Abstract Suicidal ideation and behaviour are common among inmates. Anger is found at exaggerated levels and has been associated with suicidal ideation and behaviour in inmate samples suggesting its possible salience in the prediction of suicide. The study investigated relationships between anger, psychological distress, and self-harm/suicidal ideation among inmates. The principles of Ecological Momentary Assessment were considered and a structured self-assessment diary was utilised to examine relationships between the variables of interest. Participants completed a structured self-assessment diary for six consecutive days which included momentary ratings of items describing psychological states of concurrent affects, thoughts, and appraisals related to anger, psychological distress, and self-harm/suicidal ideation. Psychometric assessment measures were also conducted. Temporal associations between predictors and outcomes were investigated. Multilevel modelling analyses were performed. Increased anger was significantly associated with concurrent high levels of self-harm ideation in inmates, when controlling for depression and hopelessness. Temporal analyses also revealed that anger at one time point did not predict suicidal ideation at the next time point. Elucidating the temporal nature of the relationship between anger, psychological distress, and self-harm/suicidal ideation has advanced understanding of the mechanisms of suicidal behaviour, by demonstrating an increased risk of suicide when a male inmate is angry.

. Introduction Compared to the general population, suicidal ideation and behaviour is significantly more prevalent amongst inmates (Jenkins et al., 2005 and Sarchiapone et al., 2009). A high association exists between recent suicidal ideation and suicide risk in inmates, and suicidal thoughts are likely to precede acts in this population (Fazel et al., 2008). Being suicidal in prison is considered to be a dynamic process with the interplay of various social, biological, and psychological factors (Bonner, 1992, Borrill et al., 2005 and Rivlin et al., 2011). Psychological distress is thought to be on the pathway towards suicidal ideation and behaviour including factors such as depression, hopelessness, and loss of relational or social support, which can lead to perceptions of loneliness or isolation (Móscicki, 1997 and Hawton and van Heeringen, 2009). Such concepts reportedly involved in this overarching experience of psychological distress have previously been described by inmates who have experienced suicidal ideation or have engaged in suicidal behaviour (Biggam and Power, 1997, Chapman et al., 2005, Ivanoff and Jang, 1991, Jenkins et al., 2005 and Palmer and Connelly, 2005; Larney et al., 2012; Liebling, 1992, Marzano et al., 2011a, Marzano et al., 2011b, Rivlin et al., 2011 and Suto and Arnaut, 2010) and as such they may be related to such outcomes. The Cry of Pain theoretical model of suicide suggests psychosocial factors thought to be involved in such adverse outcomes (Williams, 1997, Williams and Pollock, 2001 and Williams et al., 2005) by proposing the psychological conditions under which suicide is likely to occur. These conditions include feelings of depression and hopelessness, negative appraisals, appraisals of defeat and entrapment, and the perception of a lack of rescue such as social support. Previous literature lends support to this model both in general community (Gilbert and Allan, 1998, O’Connor, 2003, Rasmussen et al., 2010, Taylor et al., 2010 and Taylor et al., 2011) and inmate samples (Marzano et al., 2011a, Marzano et al., 2011b, Rivlin et al., 2011 and Suto and Arnaut, 2010). Alongside self-harm/suicidal ideation, psychological distress warranted further investigation as an outcome for male inmates at risk of self-harm/suicide. For male inmates, being angry and the expression of this anger may be a particularly important factor in those who are suicidal as studies of the characteristics of those who die by suicide in correctional settings have found that over-representation of histories of violence and violent offences suggesting anger may be a prominent feature of such individuals (Blaauw et al., 2005, Fazel et al., 2008, Fruehwald et al., 2004, Humber et al., 2011 and Sarchiapone et al., 2009). Anger has been suggested to be associated with suicidal ideation and behaviour in inmate samples (Carli et al., 2011; Chapman and Dixon-Gordon, 2007, Lekka et al., 2006, Liebling, 1992, Marzano et al., 2011b, Medlicott, 2000, Penn et al., 2003, Rivlin et al., 2011, Sarchiapone et al., 2009, Snow, 2002 and Suto and Arnaut, 2010). However, although previous research has suggested an association, the delineation between anger experience, its expression, and the type of expression has not been specified. Is it the increasing experience of being in an angry ‘state’ that is associated with psychological distress and self-harm/suicidal ideation or is the relationship more readily explained by how the individual expresses such anger, i.e. by suppressing or internalising it (‘anger-in’), or externalising it (‘anger-out’) (Spielberger, 1999 and STATA, 2011). In addition, the temporal nature of the relationship between anger experience and expression in relation to psychological distress and self-harm/suicidal ideation in inmates has not been elucidated. Are inmates simultaneously angry and suicidal or do those who experience anger become suicidal in a delayed response (Carli et al., 2009; Chapman and Dixon-Gordon, 2007; Lekka et al., 2006; Marzano et al., 2011; Rivlin et al., 2011; Suto and Arnaut, 2010)? This warranted further investigation. Previous research has often relied on retrospective methods using casenotes and clinical records (Fazel et al., 2008) which suffer from well-known limitations (Liebling, 1999 and Marzano et al., 2011b). The use of psychometric assessment tools and semi-structured interviews post-suicidal event have also previously been employed (Liebling, 1992, Magaletta et al., 2008, Marzano et al., 2011b and Medlicott, 2000), and although useful, these methods are hindered by recall and memory biases that do not allow for the assessment of fluctuation in mood states and their relation to thought content and appraisals during the actual episode. The accurate assessment of psychological factors and their interaction with the environment in terms of suicidal ideation and behaviour in inmates has presented great challenges to scientific research (Liebling, 1999). Some have therefore advocated the need for empirical and methodological advancement to enable a theoretical understanding of the mechanisms of this behaviour to be elucidated (Chapman and Dixon-Gordon, 2007, Felthous, 2011, Marzano et al., 2009 and Snow, 2002). Over the last few decades, Ecological Momentary Assessment (EMA) methods have developed (Bolger et al., 2003, Csikszentmihalyi and Larson, 1987 and Shiffman et al., 2008) to capture psychopathological processes such as affects, thoughts, and appraisals as they happen in real-time (Conner Christensen et al., 2003, Myin-Germeys et al., 2009 and Palmier-Claus et al., 2011). The value of EMA methods is that they are conducted as part of the individual's daily routine to ensure it is ecologically valid, giving reliable insight into their real-time psychological processes (Reis and Gable, 2000). Compared to an individual's retrospective recall of experiences, which is often accompanied with memory and recall biases, the real-time ESM method provides a more accurate measure of affective and cognitive states (Bolger et al., 2003, Palmier-Claus et al., 2011 and Stone et al., 1998). The EMA technique has been successfully utilised with different clinical populations (Anestis et al., 2012, Ben-Zeev et al., 2012 and Nock et al., 2009) including secure forensic settings (Hillbrand and Waite, 1992, Hillbrand and Waite, 1994, Hillbrand et al., 2000, Phelps et al., 1998 and Waite, 1994) but to the authors' knowledge has not been conducted in a penitentiary. The principles of this innovative methodology were therefore deemed viable to adopt. The EMA approach was significantly modified and adapted using a structured self-assessment diary to base a novel investigation into and understanding of the associations between anger, psychological distress, and self-harm/suicidal ideation within an inmate sample.

3. Results 3.1. Descriptive Twenty-one male inmates participated and ages ranged from 22 to 58 years (M=36 years). Descriptive statistics of the psychometric assessment battery can be found in Table 3. Table 3. Descriptive statistics for psychometric measures (n=21). Assessment measure Range Mean Standard deviation (S.D.) State-Trait Anger Expression Inventory-Version 2 (STAXI-II) Total 91–153 123.29 14.92 State 4–42 20.29 7.52 Trait 2–33 21.48 8.12 Expression-In 10–32 20.71 5.67 Expression-Out 8–28 18.90 5.96 Control-In 8–32 20.33 7.55 Control-Out 9–30 20.00 7.42 Barratt Impulsivity Scale (BIS) 46–95 74.81 12.08 Beck Anxiety Inventory (BAI) 0–32 14.38 8.13 Beck Hopelessness Scale (BHS) 0–20 10.33 6.19 Beck Depression Inventory-Version 2 (BDI-II) 10–55 29.48 12.55 Beck Scale for Suicidal Ideation (BSSI) 1–27 10.48 9.43 Defeat scale 21–73 49.10 15.61 Entrapment scale 18–67 47.05 14.67 Brief Psychiatric Rating Scale-Expanded Version 28–49 34.14 4.64 Table options 3.2. Diary protocol Five hundred and sixty-five diary entries (each entry represents a single ‘entry point’ questionnaire that was completed) were made in total, which was an average of 27 per participant. The range of diary entries for participants was between 13 and 54. The data utilised for analysis in this article was restricted to the first six entries in the diary that is from ‘waking up’ to ‘going to sleep’, in which the maximum number of entries that could be completed by participants was 36. The data analysed was restricted to six entries as the majority of participants provided limited or no data for the ‘optional’ three night entries (only 31 were recorded), and these entries were potentially qualitatively different. Six entries over six consecutive days was regarded as an adequate representation over an extended period for reliable and valid data analysis. The range of diary entries for participants was between 13 and 36. Assuming the total sample participants completed all entries for 6 days, they could have responded a total of 756 times in total. The sample completed 534 entries in total which represented an average compliance of 69% which is less than the average response frequency of 80% reported by Csikszentmihalyi and Larson (1987) whose study required a similar entry response from participants as that of the current study. At the end of each day, participants were asked to rate the day and whether completing the diary had affected their mood with ‘today has been an ordinary day’ (1–10) and ‘completing the diary has affected my mood’ (1–10) respectively. Overall the ratings suggested that during the six days of study completion, these were average days for the majority of participants (n=21; M=4.87) with it suggested that the diary measure was not perceived as too intrusive in affecting their mood (n=21; M=3.60). 3.3. Multi-level modelling Table 4 and Table 5 present the statistical results for the temporal concurrent and delayed associations between summed predictor and outcome variables to test hypothesis 1 and hypothesis 2 respectively. The tables also present the effect of internalised and externalised anger on the respective outcomes when both are added simultaneously. Table 4. Hypothesis 1—Temporal (concurrent) association between summed predictor and outcome variables. Outcome Predictors (coefficient (SE), p-value) Predictors entered together (coefficient (SE), p-value) Anger affect Anger ex in Anger ex out Anger ex in Anger ex out I want to hurt myself 0.034 (0.008), <0.001 0.031 (0.013), 0.018 0.040 (0.009), <0.001 0.015 (0.014), 0.280 0.037 (0.009), <0.001 Life is worth living 0.003 (0.009), 0.779 0.037 (0.015), 0.012 0.006 (0.010), 0.585 0.049 (0.015), 0.002 −0.006 (0.011), 0.548 I want to live −0.010 (0.009), 0.262 −0.001 (0.014), 0.933 −0.016 (0.009), 0.078 0.002 (0.015), 0.882 −0.017 (0.010), 0.090 Psychological distress (affects) 0.220 (0.035), <0.001 0.225 (0.058), <0.001 0.127 (0.040), 0.001 0.190 (0.062), 0.002 0.083 (0.042), 0.049 Psychological distress (appraisals) 0.309 (0.039), <0.001 0.441 (0.063), <0.001 0.301 (0.043), <0.001 0.330 (0.066), <0.001 0.224 (0.045), <0.001 Table options Table 5. Hypothesis 2—Temporal (delayed) association between summed predictor and outcome variables. Outcome Predictors (coefficient (SE), p-value) Predictors entered together (coefficient (SE), p-value) Anger affect Anger ex in Anger ex out Anger ex in Anger ex out I want to hurt myself 0.013 (0.008), 0.109 -0.001 (0.014), 0.957 0.026 (0.009), 0.004 −0.014 (0.015), 0.336 0.030 (0.009), 0.002 Life is worth living 0.001 (0.010), 0.997 0.021 (0.016), 0.206 −0.001 (0.011), 0.904 0.015 (0.018), 0.392 −0.005 (0.012), 0.679 I want to live 0.002 (0.010), 0.818 0.056 (0.016), <0.001 −0.002 (0.010), 0.816 0.065 (0.016), <0.001 −0.018 (0.011), 0.086 Psychological distress (affects) 0.073 (0.039), 0.060 0.013 (0.066), 0.846 0.077 (0.042), 0.073 −0.027 (0.072), 0.707 0.084 (0.047), 0.071 Psychological distress (appraisals) −0.032 (0.042), 0.447 0.012 (0.073), 0.873 −0.048 (0.047), 0.304 0.067 (0.079), 0.390 −0.063 (0.050), 0.212 Table options 3.3.1. Hypothesis 1: temporal (concurrent) associations between anger, psychological distress, and suicidal ideation Being angry was concurrently associated with self-harm ideation [‘I want to hurt myself’] (0.03; p<0.001) and psychological distress (affects) (0.22; p<0.001). Items positively skewed to measure suicidal thoughts [‘I want to live’ and ‘life is worth living’] were not concurrently associated with anger experience or expression. High externalised anger predicted concurrent self-harm ideation [‘I want to hurt myself’] (0.04; p<0.001). High internalised anger was significantly associated with concurrent psychological distress (affects) (0.23; p<0.001). The relationships between internalised and externalised anger and the psychological distress (appraisals) outcome, when considering the predictors separately, still held when both anger expression variables were included in the model together (0.33; p<0.001 and 0.22; p<0.001, respectively), implying that the anger expression variables had independent effects. 3.3.2. Hypothesis 2: temporal (delayed) associations between anger, psychological distress, and suicidal ideation There were less significant effects for lagged predictors. Concurrent predictors of self-harm ideation and psychological distress outcomes were no longer significantly associated in the temporal delay. High internalised anger was associated with thoughts of wanting to live (0.06; p<0.001). High internalised anger remained associated with thoughts of wanting to live when both anger expression variables were included in the model (0.07; p<0.001).