At the time of writing, four qualitative/mixed-methods papers from the CARMS workstreams which address efficacy and clinical research questions have been published.46,115–117 Summaries of these findings are presented. More detailed information is available from the relevant cited CARMS publications.
Qualitative efficacy workstream: acceptability and implementation
There were two efficacy research questions which were addressed with qualitative research methods and analyses, both of which are published: (1) What is the acceptability to participants of a suicide-focused therapy?116; and (2) What are the barriers and facilitators of implementing a suicide-focused therapy in NHS mental health services to both participants and mental health professionals?117
1. ‘Acceptability of a novel suicide prevention psychological therapy for people who experience non-affective psychosis’116
Overall goal and rationale
The main goal of this work was to probe issues of acceptability of the suicide-focused therapy (CBSPp) used in CARMS as perceived by people with psychotic and suicidal experiences. A problem at the outset was that although many stakeholders agree that the boundaries of acceptability of any intervention are important to establish scientifically, clinically and pragmatically, the operationalisation of acceptability lacks precision.131 In an attempt to address this problem, the Theoretical Framework of Acceptability (TFA) was developed in the domain of health psychology. The TFA views acceptability as multicomponential, with a need to consider not only perceived effectiveness but also other factors, such as burden, ethical issues, emotional and cognitive responses elicited to the intervention, self-efficacy, opportunity costs and intervention coherence. Given the inherent complexity in developing efficacious psychological interventions for those experiencing psychosis and suicidal thoughts/acts, the TFA seemed an appropriate tool for assessing the acceptability of CBSPp.
Analyses
Qualitative one-to-one interviews from 20 SU participants who had been randomised to receive therapy were analysed both inductively using thematic analysis (TA)132,133 and deductively using the seven domains of the TFA, that is, affective attitude, burden, ethicality, intervention coherence, opportunity costs, perceived effectiveness and self-efficacy. The TFA coding manual131 was adapted so that it was applicable to data from people with psychotic and suicidal experiences. Specifically, the therapeutic alliance was an added construct following the inductive analysis, which also indicated that ethicality and opportunity costs had limited utility for this CPSPp intervention.
Findings
All participants had completed therapy (with an average of 17 sessions). Interviews lasted around 45 minutes (range 27–63 minutes). There were six main themes of affective attitude, burden, alliance, intervention coherence, perceived effectiveness and self-efficacy (Figure 4).
FIGURE 4 The main themes and subthemes expressed by participants when asked about the acceptability of the CARMS suicide-focused CBSPp therapy. (Figure not reproduced here; see the original article.)
Conclusions
There were three main conclusions. First, talking about suicide during therapy was perceived as sometimes difficult, because it could evoke powerful emotions and required trust to be placed in both the therapist and the therapeutic process. However, it was also apparent that there was a mismatch between the apprehensions of participants about the emotional demands of therapy and their actual experiences. These findings were consistent with other work in the area investigating ‘suicide talk’2,115,134 but also expanded it because participants who were acutely suicidal and also had psychotic experiences were able to speak directly about their recent experiences of a suicide-focused therapy which is rare in the extant literature.
Second, CBSPp could be perceived as burdensome especially with respect to managing potential distress and ‘homework’ tasks. That said, there was also a perception that although therapy involved hard work, especially in the initial stages, it was balanced by experiencing increased and evolving understanding.
Third, even though the focus of the therapy on suicide was welcome, it could also lead to some participants excluding themselves from therapy because they did not think that their experiences were severe enough to warrant it. This underscores a wider problem, which is that suicidal experiences can manifest in many ways, such as fleeting thoughts, urges, compulsions, images, concrete plans and a more passive desire to not be here or to die. These different experiences can also fluctuate and change, sometimes hour by hour.27,135 Communicating and raising awareness about these different types of experience may be of benefit to potential clients.
2. ‘What is important to service users and staff when implementing suicide-focused psychological therapies for people with psychosis into mental health services?’117
Overall goal and rationale
Despite a clear need for psychological therapies which focus on suicidal experience in people with severe mental health problems,1 the development and implementation of such therapies is relatively sparse. One such therapy, CBSPp,1 has a growing evidence base2,42,53,64 and the potential to alleviate suicidal thoughts, plans and acts. However, a key stage in assessing the potential of this therapy is to understand the factors which are necessary for it to be implemented within NHS services, not only from the perspectives of SUs with experiences of suicide and psychosis but also from the perspectives of a range of mental health professionals and mental health service commissioners. Actualising this understanding was the goal of the current study.
Analyses
One-to-one qualitative interviews were conducted with 20 mental health professionals and 18 SUs (4/18 were randomised to therapy). Mental health professionals included those working in the NHS on inpatient wards, in early intervention services, in community teams, and those responsible for commissioning mental health services. Stakeholders also included those responsible for mental health service provision outside of the NHS, namely mental health charities.
Findings
As depicted in Figure 5, there were four main themes which coalesced in capturing the views of mental health professionals and SUs, which were (1) creating safe spaces to be understood, (2) gaining a voice, (3) accessing therapy at the right time and (4) ensuring a straightforward pathway to accessing therapy.
FIGURE 5 The four themes embodying the views of SUs and mental health professionals about implementing a suicide-focused talking therapy in mental health services. (Figure not reproduced here; see the original article.)
Conclusions
Participants were in agreement, be they SUs or mental healthcare professionals, that a suicide-focused talking therapy was valuable and should be offered to people with severe mental health problems, such as schizophrenia. This, in itself, was an important finding, because some trials of therapeutic interventions actively exclude people with psychosis, for example.136 For successful implementation of a suicide-focused talking therapy, three factors were highlighted as needing to be in place.
First, many dimensions of safety need to be met, ranging from the provision of a safe geographical location to ensuring an emotionally safe listening environment. Mental health professionals emphasised that interactive suicide training was essential to safety.
Second, enabling SUs to feel actively empowered in decisions about their own mental health care, including therapy, was considered fundamental. It was important that SUs could perceive concrete changes to their mental health care resulting from their input.
Third, discussions need to be opened up about when therapy should be offered so that it is maximally effective. Views of some mental health professionals were that therapy was not helpful during times of acute psychotic crises. Countering this was the perception that for many people, mental health experiences fluctuate in severity and duration, meaning that therapy could, indeed, be very helpful to SUs in understanding those experiences and in feeling supported through those difficulties.
Qualitative mechanistic workstream
3. ‘The interplay between suicidal experiences, psychotic experiences and interpersonal relationships: a qualitative study’46
Overall goals and rationale
Recent psychological models of suicidal thoughts and behaviours21,22,54,137 converge on the central role of perceptions of destructive interpersonal relationships and a feeling of not belonging or being a burden in pathways to suicidal experiences. However, there are three problems to date with work in this area. First, the complexities of interpersonal relationships and social dynamics in people with suicidal thoughts, urges, plans and attempts have not been examined in depth. Second, the ways in which psychotic experiences impact this dynamic are poorly understood. Third, how people who live with suicidal experiences counter those experiences using different aspects of interpersonal relationships have been minimally explored. The overarching goal of the current study was to use qualitative interviews to address these three issues.
Analyses
An inductive reflexive TA132,133 was used on data generated from semistructured, one-to-one qualitative interviews from 22 participants who had been randomised to CBSPp therapy. A critical realist perspective was adopted and steps taken to ensure trustworthiness of the final analysis, including researcher triangulation.118
Findings
The data were understood in the form of a novel conceptual model, in which there was an overall context of a dynamic interplay between psychotic experiences, suicidal experiences and experiences of interpersonal social relationships. Three themes fed into this interplay, which were (1) not mattering and mattering, (2) being connected and being disconnected and (3) constraints versus freedom of talking and sharing, each of which comprised two subthemes (Figure 6).
FIGURE 6 Themes and subthemes which fed into a complex dynamic between psychosis, suicide and interpersonal relationships. (Figure not reproduced here; see the original article.)
Conclusions
This is the first study to examine a complex interplay between non-affective psychosis, myriad suicidal experiences and social interpersonal relationships. A novel dynamic and interactive conceptual model is presented (see Figure 6), which goes beyond simple pathways or polarisations.
The theme of not mattering versus mattering captured a continuum, whereon, at times, participants had felt unimportant, not valued and inconsequential. These sorts of thoughts and feelings were at the heart of suicidal experiences for many participants. Mattering and not mattering, sometimes called ‘anti-mattering’, has been investigated in a number of mental health domains138–140 but thus far only to a limited extent in relation to suicidal experiences.139,141,142 From a clinical perspective, understanding perceptions of not mattering, and exploring changes in those perceptions, appear fundamental to combatting suicidality.
Feelings and perceptions of disconnection and connection were multifaceted. For some individuals, becoming disconnected felt preferable compared to having to live with an apparent constancy of being watched, observed and judged or mocked. Acting against this was a knowledge that social isolation can make mental health problems worse. Acknowledging and attempting to understand these intricate dynamics seem paramount.
Even when frightening, participants expressed the importance of talking and sharing experiences of mental health problems, including suicidality. However, there were many influences which made talking and sharing less than straightforward, including unwanted and overbearing advice giving, being judged, feeling stigmatised and quite simply not being listened to. There were clear messages about communication emanating from these findings which apply not only to mental health professionals who work with people who have severe mental health problems but also to friends, family and wider social network communities.
Qualitative and mixed-methods experiences of participating in a suicide randomised controlled trial workstream
4. ‘A systematic investigation of the short and long term experiences of taking part in research about suicidal thoughts and acts in the context of a randomised control trial’115
Overall goal and rationale
Even though there are numerous sources of evidence showing that talking about suicidal experiencers can be helpful, concerns have been raised by ethical review committees, mental health professionals and researchers that explicitly discussing suicidal thoughts, urges, plans and acts may be triggering and amplify those thoughts and feelings. Up until recently, ensuring that suicide work programmes involve a component which seeks the views of participants about being involved in suicide research has been largely neglected. However, more currently, this is being redressed. The evidence which is available shows that a range of positive experiences are frequently endorsed and that although negative thoughts and feelings can result from talking about suicidal experiences, these are often expected by participants and short-lived. While the existing work is promising, it still requires robust expansion in two domains. The first is an examination of consequences of participating in suicide research work not only in the shorter term but also in the longer term. The second is the need to investigate experiences of taking part in RCTs about suicidal thoughts and acts in which participants may be asked about their experiences intensely, repetitively and with different methods (e.g. questionnaires, clinical and qualitative interviews). Hence, the goal of this study was to contribute to a better understanding of these two domains using a mixed-methods approach.
Analyses
At baseline and 6- and 12-month follow-up time points, CARMS participants were qualitatively interviewed at the end of assessment sessions about their experiences of taking part in research about suicide. In addition, participants were asked to complete a VAS mood scale at the beginning and end of each assessment session and to complete a 16-item adjective checklist when the session finished.
Analyses examining the short-term effects of participating in CARMS were based on 100 participants. Longitudinal analysis from 32 participants was restricted to those in the TAU group so that experiences of suicide research were not conflated with experiences of therapy. Qualitative data were thematically analysed. Quantitative data were examined for changes in mood before and after an assessment; and relative frequency data of positive and negative adjectives to describe participants experience and the extent to which suicidal state (ASIQ) impacted upon their experience of taking part in research about suicide.
Findings
There were four key findings. First, the mood of participants measured by the VAS either stayed the same or improved from the beginning to the end of an assessment session. Improvements were reported by 86% of participants.
Second, most participants selected a mixture of positive and negative adjectives, using the adjective checklist with more positive than negative adjectives being picked. The most frequent positive adjective was ‘worthwhile’, and the most frequent negative adjective was ‘anxiety-provoking’. Only 5/100 participants chose negative adjectives to describe their experience.
The third finding related to the TA of the interviews, which produced themes of (1) ‘Suicide is difficult to talk about’, (2) ‘Participation benefited me’, (3) ‘Participation is worth it’ and (4) ‘Contextual factors influence the experience of participation’ (Figure 7).
FIGURE 7 Themes and subthemes related to the experiences of participating in a RCT about psychotic and suicidal experiences. (Figure not reproduced here; see the original article.)
Fourth, for some participants, there was a dip in mood, which was experienced as short-lived, lasting up to a few hours at most. However, for others, there was a more enduring dip in mood, which lasted across 3 or 4 days. Participants felt that they had the resources to be able to cope with these decrements in mood, especially if their state of mind was stable and largely positive. In contrast to mood dips, participants also reported a type of longer-term catharsis, which lasted beyond a few hours or a few days. This was perceived to be, at least in part, due to being able to talk about and share their suicidal experiences and having the space to be able to ‘process’ their thoughts and feelings during research assessments.
Conclusions
The main convergent finding across qualitative and quantitative methods of data collection was that participation in a RCT investigating psychotic and suicidal experiences was largely positive in both the shorter and longer terms (up to 6 and 12 months). This convergent finding was both arresting and reassuring. It was arresting for two reasons. First, the participants were recruited because they had recent experiences of suicidal thoughts and behaviours in addition to psychotic experiences, most of which were in the form of hallucinations and/or delusions. Hence, a more negative outcome might have been expected. Second, convergence from quite different methodologies is infrequently documented. In the current study, the methodologies included short qualitative interviews, adjective checklists and a simple VAS rating scale. Reassurance was gained from this convergence but also from the way that these findings augment similar work in the suicide literature.134,143–151 This work goes a considerable way to assuage concerns about conducting research which asks participants to share their suicidal experiences.