Introduction
Many people with psychosis report childhood adversity, and early adversity substantially increases risk of psychosis, with a likely dose–response relationship (Varese et al., Reference Varese, Smeets, Drukker, Lieverse, Lataster and Viechtbauer2012). It is unsurprising then that three-quarters of people with psychosis report recurrent, distressing intrusions, often linked to trauma memories (Morrison et al., Reference Morrison, Beck, Glenworth, Dunn, Reid, Larkin and Williams2002).
While we have well-evidenced trauma-focused interventions, such as imagery rescripting (IR), most people with psychosis are not offered these, partly due to clinical caution – we are concerned about exacerbating people’s psychosis and associated risks (Sin et al., Reference Sin, Spain, Futura, Murrelis and Norman2017). This is in contrast to the growing number of studies examining cognitive behavioural interventions targeting distressing intrusions in psychosis, including IR (e.g. Ison et al., Reference Ison, Medoro, Keen and Kuipers2014; Keen et al., Reference Keen, Hunter and Peters2017; Paulik et al., Reference Paulik, Steel and Arntz2019; Sheaves et al., Reference Sheaves, Onwumere, Keen and Kuipers2015; Taylor et al., Reference Taylor, Bee, Kelly and Haddock2018), and evidence that IR affects the same cognitive-affective processes in non-clinical paranoia as has been shown in people with depression and anxiety (Newman-Taylor et al., Reference Newman-Taylor, McSherry and Stopa2019). If we are reluctant to offer effective interventions for fear of causing harm, the question is: how can we offer trauma-focused CBT safely and effectively to people with psychosis?
There is good evidence for the effectiveness of IR in treating post-traumatic stress disorder (PTSD) and intrusions cross-diagnostically (Arntz, Reference Arntz2012; Morina et al., Reference Morina, Lancee and Arntz2017). Additionally, it may be that IR is more suitable than prolonged exposure for people with psychosis given (i) the shorter duration of exposure to distressing memories (which is likely to be linked to drop-out rates), (ii) the focus on change to meaning and affect linked to memories rather than the perceptual experience of intrusions, which may be particularly important in interpersonal trauma, and (iii) the requirement for fewer sessions per memory, making this approach appropriate for people with multiple traumas, as is often the case for this group (Paulik et al., Reference Paulik, Steel and Arntz2019). Despite initial evidence for IR with psychosis, take-up remains low. Certainly, we need to support clinicians to develop these skills. In addition, it may be that some people with psychosis would benefit from learning to manage increases in affect, to ensure that IR is acceptable, and can be used safely and effectively.
The concept of ‘felt security’ was developed in the attachment and social psychology literature, and describes the sense of interpersonal safety associated with secure, protective relationships, that in childhood form the basis for healthy psychosocial trajectories. Recent advances in CBT for psychosis draw on the attachment literature to formulate psychosis in developmental context (Berry et al., Reference Berry, Varese and Bucci2017; Gumley et al., Reference Gumley, Schwannauer, MacBeth and Read2008), and preliminary experimental studies show that attachment-based imagery designed to strengthen felt security reduces paranoia and anxiety, and improves mood and self-esteem in analogue groups (e.g. Bullock et al., Reference Bullock, Newman-Taylor and Stopa2016; Newman-Taylor et al., Reference Newman-Taylor, Kemp, Potter and Au-Yeung2018).
In this case study, we sought to prime felt security as a means of facilitating engagement in IR following the three-stage protocol set out by Arntz and Weertman (Reference Arntz and Weertman1999), which involves recalling traumatic memories in a particular way to allow reprocessing of intrusive material. We predicted that priming felt security would enable the person to manage the temporary distress elicited by IR, engage fully in the process, and so benefit from the intervention.
Method
Participant and formulation
Kip presented with frequent derogatory voices, paranoia and intrusive trauma memories. She was 36 years old and had been assessed by her psychiatrist to have psychosis at the time of referral. A past diagnosis of emotionally unstable personality disorder (EUPD) had led to psychological therapy to manage her emotions more effectively, and she now sought help to address her psychosis and trauma symptoms. Kip had a history of hurting herself (through cutting) and others (through attempted strangulation) from adolescence, and had been arrested once and admitted to hospital on several occasions due to her risks.
Our initial CBT sessions focused on making sense of Kip’s current experiences in the context of her trauma history. She had been physically and sexually abused by her father from the age of 3 to 15 years, and bullied at secondary school. From an early age, she learnt to see herself as worthless and disgusting, and others as unpredictable and threatening. During adolescence, she started to hear voices echoing her fears, and became convinced that demons had taken over her body. When we met, Kip described a pervasive sense of threat, indicating that others were planning to harm her and that she was unable to protect herself. These beliefs were reiterated by a powerful male voice that issued commands to kill herself. She also heard a jeering crowd of voices, which left her feeling terrified although she was unsure why. Kip managed her fears by withdrawing socially, listening for the voices, and either ruminating or trying to push difficult thoughts and voices out of mind. She was isolated, lonely, and low in mood.
Following formulation, Kip was able to start to reconsider her beliefs about herself, others and the voices in developmental context. She was prompted to extend her use of emotion regulation skills (learnt in her previous therapy) when feeling fearful of others or distressed by the voices. She learnt to recognise unhelpful thinking patterns, articulate previously unspoken feelings including fury, shame and disgust, and engage with valued activities and safe relationships. This work resulted in Kip being able to respond more calmly and kindly to her voices and herself, with the result that her paranoia and the distress associated with the voices reduced. However, frequent intrusive memories persisted, linked to her early abuse, bullying at school, an arrest as a young woman, and events while an in-patient. At this point, Kip decided she was ready to address her trauma memories directly, something she had wanted to do for several years, although she was fearful about the likely distress of recalling past events, and her ability to manage the feelings elicited without becoming unwell or suicidal.
Measure
The Felt Security Scale (Luke et al., Reference Luke, Sedikides and Carnelley2012) assesses the degree to which a person feels safe and unthreatened, as would be expected in a secure attachment relationship. Ten descriptors (e.g. safe, protected) are rated on a 6-point scale (1: not at all, to 6: very much). For the purpose of clinical use, the descriptors were rated ‘right now’ to assess state felt security. Scores range from 10 to 60, with higher scores indicating greater felt security. The scale has excellent internal consistency (α = .97).
Felt security imagery task
The felt security imagery (FSI) task was developed from an imagery priming script for social anxiety (Hirsch et al., Reference Hirsch, Clark, Mathews and Williams2003) and secure attachment priming instructions (Bartz and Lydon, Reference Bartz and Lydon2004), and adapted for people with paranoia (available on request). The task takes approximately three minutes to complete and involves recalling a specific interpersonal memory of a time when the person felt safe and relaxed, and knew that s/he could trust and rely on the other person or people present.
Procedure
Kip’s fears about addressing past trauma were consistent with her formulation, which indicated that recall of certain memories increased her paranoia and linked distress. To enable her to engage in IR of the intrusions, we agreed to use the FSI task to engender a sense of felt security immediately before the reprocessing, and then again after IR so that she would leave sessions feeling safe. One trauma memory was rescripted per session, over six sessions. Kip attended therapy weekly, and completed IR on each occasion she felt able to do so; occasionally, her level of distress was such that she declined. IR targeted memories of being bullied and assaulted, and took approximately 30 minutes per session. Memories were rescripted until Kip reported a reduction in intrusions and linked affect over the following week. Over the six sessions reported, four key memories were rescripted. Kip rated the frequency and distress (0–10) of intrusions pre- and post-intervention.
Results
Felt security
Figure 1 shows the first six felt security ratings pre- and post-FSI task, which was used before and after IR in each of these sessions.
Figure 1. Felt security before and after imagery rescripting.
The figure indicates a similar pattern for each session; felt security increased over the period of the FSI task, then fell during IR (when key trauma memories were recalled and rescripted) and increased again over the period of the second FSI task.
Subjective feedback on trauma intrusions and psychosis
For each of the memories rescripted, Kip reported a reduction in frequency (from up to several times a day to nil most days) and distress (from 10/10 to 3/10) of related intrusions. She had not heard the jeering crowd since rescripting a memory of being bullied at school. She hears one remaining voice less frequently, which she now understands as an expression of her mood and worst fears (‘inner me’), and uses as a stress gauge and to prompt self-care and help seeking. She also reported a further reduction in her paranoia over this time.
Discussion
Our collective caution regarding trauma-focused CBT for people with psychosis is partly due to concerns about causing harm (Sin et al., Reference Sin, Spain, Futura, Murrelis and Norman2017). This case study illustrates the use of a brief imagery (FSI) task to facilitate imagery rescripting of trauma memories linked to one woman’s voices and paranoia. We found that the FSI task resulted in an increase in felt security when used before and after IR. Despite fears about engaging in IR, and managing the temporary distress elicited, Kip was able to do both and this proved effective in reducing the frequency and distress of her intrusions.
As a case study, we cannot draw conclusions about whether the imagery task was either instrumental or necessary in facilitating IR. It may be that Kip would have benefited from standard IR with no adverse effects. However, Kip and her care team were reluctant to pursue this option. Additionally, the lack of other measures means we cannot determine how specific the task was in increasing Kip’s felt sense of security as opposed to general anxiety. A single case experimental study would be the next step in assessing the impact of the felt security task for people who decline or have concerns about engaging in standard IR.
With an increasing recognition of the role of early adversity in the development of psychosis, and evidence that the majority of people with psychosis report related intrusions, we need to find ways to offer trauma-focused CBT that are acceptable, safe and effective. If some people with psychosis are concerned about tolerating IR, the FSI task may provide a simple means of regulating emotion and therefore increasing the acceptability of IR. If their formulation indicates that heightened arousal exacerbates voices, paranoia or risk, the FSI task may enable IR to be used safely. Where clinicians are hesitant, the task may increase the likelihood that IR is offered.
Acknowledgements
Thanks go to Kip for agreeing to me writing up an anonymous summary of our work together.
Financial support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Conflicts of interest
The author has no conflicts of interest with respect to this publication.
Ethical statements
The author has abided by the Ethical Principles of Psychologists and Code of Conduct as set out by the APA: http://www.apa.org/ethics/code/. The Trust Research and Development Department confirmed that formal ethical approval was not required because the work was completed as part of routine clinical practice.
Key practice points
(1) People with psychosis can engage in trauma-focused CBT.
(2) A brief attachment-based imagery task may provide a simple means of facilitating ‘felt security’.
(3) This in turn may enable people with psychosis to engage in imagery rescripting.
Further reading
Ison, R., Medoro, L., Keen, N., & Kuipers, E. (2014). The use of rescripting imagery for people with psychosis who hear voices. Behavioural and Cognitive Psychotherapy, 42, 129–142.
Keen, N., Hunter, E. C. M., & Peters, E. (2017). Integrated trauma-focused cognitive-behavioural therapy for post-traumatic stress and psychotic symptoms: a case-series study using imaginal reprocessing strategies. Frontiers in Psychiatry, 8. doi:10.3389/fpsyt.2017.00092
Newman-Taylor, K., McSherry, P., & Stopa, L. (2019). Imagery rescripting in non-clinical paranoia: a pilot study of the impact on key cognitive and affective processes. Behavioural and Cognitive Psychotherapy, 48, 54–66. doi:10.1017/S1352465819000419
Paulik, G., Steel, C., & Arntz, A. (2019). Imagery rescripting for the treatment of trauma in voice hearers: a case series. Behavioural and Cognitive Psychotherapy, 47, 709–725.
Sin, J., Spain, D., Futura, M., Murrelis, T., & Norman, I. (2017). Psychological interventions for PTSD in people with severe mental illness. Cochrane Database of Systematic Reviews, issue 1: CD011464.
Taylor, C. D. J., Bee, P. E., Kelly, J., & Haddock, G. (2018). iMAgery focused therapy for persecutory delusions in PSychosis (iMAPS): A novel treatment approach. Cognitive and Behavioural Practice, 26, 575–588.
Comments
No Comments have been published for this article.