Introduction
Loss of life through suicide is both a global problem and an individual tragedy. It is estimated that around the world over 700,000 people a year die by suicide (World Health Organization, 2021) and that for each loss of life 135 people are affected to some degree (Cerel et al., Reference Cerel, Brown, Maple, Singleton, Venne, van de Moore and Flaherty2018). There is a recognised association between suicide risk and mental health problems, but it is important to keep this in context. Whilst the majority of people in Western countries who take their own lives have a diagnosable mental health condition, the vast majority of people with a mental health condition do not die by suicide (O’Connor, Reference O’Connor2021). Identifying those at risk of taking their own lives has proven to be challenging and the reasons for this are numerous and much debated (for a fuller discussion, please see Zortea et al., Reference Zortea, Cleare, Melson, Wetherall and O’Connor2020). Two key factors are first, that suicide, whilst tragic and of wide-ranging impact, is statistically relatively rare. For example, worldwide approximately nine people per 100,000 of the population die by suicide each year (World Health Organization, 2019). Second, suicide risk is dynamic; it waxes and wanes, with risk potentially escalating rapidly and without detection. Mental health practitioners who work with people with common mental problems such as depression or anxiety disorders, would not generally be expected to be working with people who are in suicidal crisis or at high risk of engaging in suicidal behaviours. These practitioners are usually based in primary health care settings without the remit, resources or training required to manage such level of risk. However, given the high volume of people treated in primary care and the fact that suicide risk may either not be identified or may escalate rapidly, those who work in primary care mental health services are regularly in contact with people who die by suicide or are at risk of doing so. This includes people whose deaths occur either in treatment or at some point post-discharge. In England, for instance, nearly a third of people who die by suicide were in contact with mental health services in the 12 months prior to their death (National Confidential Inquiry into Suicide and Homicide in Mental Health, 2017), whereas three-quarters are estimated to have been in contact with their GP within the same time scale (National Confidential Inquiry into Suicide and Homicide in Mental Health, 2016).
Whilst primary care staff are not usually expected to provide therapy to acutely suicidal individuals or work directly on suicidal thoughts, all staff should be vigilant about suicide risk, potentially helping to reduce vulnerabilities to future suicide risk as part of their routine work. As this paper describes, many of the risk factors and processes hypothesised to play a role in the development of suicidal thinking and behaviours are likely to be evident within the usual content of standard evidence-based protocols for depression or anxiety disorders. In this paper we are not proposing a change in the remit of primary care mental health services as has been explored elsewhere; for example, the piloting of low intensity interventions for patients presenting in Emergency Departments (Bastiampillai et al., Reference Bastiampillai, Jones, Furber, Moreau, Healey, Watson and Battersby2014). Rather, we are suggesting that even within the current primary care remit, (i) an increased awareness of suicide risk vulnerability factors and (ii) using knowledge of a psychological model of suicidal behaviour to inform clinical care are likely to be extremely helpful in structuring clinical formulation and informing interventions.
Improving Access to Psychological Therapy services
In England access to primary mental healthcare has been widened greatly by the introduction over the past decade of Improving Access to Psychological therapies services (IAPT; Clark, Reference Clark2011). These services have also informed programme delivery in Australia (Baigent et al., Reference Baigent, Smith, Battersby, Lawn, Redpath and McCoy2020), Canada (Naeem et al., Reference Naeem, Pikard, Rao, Ayub and Munshi2017), Japan (Kobori et al., Reference Kobori, Nakazato, Yoshinaga, Shiraishi, Takaoka, Nakagawa, Iyo and Shimizu2014), New Zealand (Haarhoff and Williams, Reference Haarhoff and Williams2017) and Norway (Knapstad et al., Reference Knapstad, Nordgreen and Smith2018). IAPT services follow a stepped care structure to deliver evidence-based interventions at scale (Wakefield et al., Reference Wakefield, Kellett, Simmonds-Buckley, Stockton, Bradbury and Delgadillo2021). Although IAPT services aim to offer a range of therapeutic modalities (e.g. counselling for depression, interpersonal therapy, couples therapy for depression), the predominant approach utilised to support people experiencing common mental health problems such as depression or anxiety disorders is cognitive behavioural therapy (CBT).
In the original implementation plan for IAPT, no reference was made to suicide, because the intention was for IAPT services to support patients without significant suicide risk (Care Services Improvement Partnership, 2007; p. 6); with the guidance that high-risk patients should be urgently referred to the appropriate Community Mental Health Team or secondary care mental health provider without delay. IAPT services were designed for people who would be deemed appropriate for low or high intensity psychological interventions and individuals at significant risk of suicide would require a wider and more intensive package of care such as that provided by a multi-disciplinary team (NICE, 2011). The current IAPT manual reiterates this, stating ‘if someone currently has clear plans or intent to act on these thoughts, IAPT services are not best placed to meet the patient’s needs’ (National Collaborating Centre for Mental Health, 2021; p. 9).
Nonetheless, we believe that risk assessment, formulation and the subsequent management of any identified risk is a key role for IAPT staff (and other primary care-based CBT therapists) for several reasons. First, a key feature of IAPT services is self-referral and even with very clear communication around the role of IAPT services, some individuals who receive an assessment will present with significant risk of suicide. Second, as already mentioned, risk is dynamic and can change during the course of treatment. An individual with depression may enter treatment with low risk, but this may fluctuate over the course of sessions due to range of potential factors. Third, a person may be reticent to discuss thoughts of suicide at initial and early contact, but they may disclose suicidal thoughts once a therapeutic relationship has been established.
There have been a number of contemporary psychological models developed to understand suicidality (O’Connor and Nock, Reference O’Connor and Nock2014); the focus here is on one, the Integrated Motivational Volitional (IMV) model (O’Connor and Kirtley, Reference O’Connor and Kirtley2018). The IMV model is an ideation-to-action model that has been influenced by previous theoretical developments in the field including Joiner’s Interpersonal Theory (IPT) (Van Orden et al., Reference Van Orden, Witte, Cukrowicz, Braithwaite, Selby and Joiner2010) and Williams’ ‘Cry of Pain’ model (Williams, Reference Williams1997).
Overview of the IMV model
An important consideration – and one central to the IMV model – is the utility of conceptualising suicide as a behaviour. This may appear obvious but was in fact a significant departure from historical thinking that has often viewed suicide as an epiphenomenon of a psychiatric condition (O’Connor, Reference O’Connor2011). Understanding suicide as a behaviour has afforded researchers the opportunity to draw upon a rich field of knowledge and theories from areas including health psychology, such as the theory of planned behaviour (Ajzen, Reference Ajzen1991). The IMV model therefore draws upon a range of established theories from suicide research and more broadly from health psychology. The overall structure of the model, as the title suggests, is based on a motivational–volitional framework. As shown in Fig. 1, it consists of three main parts or phases; the pre-motivational (background vulnerability, diathesis/stress factors), the motivational (risk related thinking and perceptions) and volitional (suicidal behaviours or actions) phases.
Figure 1. The Integrated Motivational-Volitional (IMV) model of suicide behaviour.
The diathesis–stress approach is used to describe the background (pre-motivational) and relatively fixed social, psychological and biological factors that might influence how a person responds to recent and current stressful life events.
The central pathway of the motivational section of the model is drawn from Williams’ (Reference Williams1997) cry of pain model. This describes how a person may experience negative life event(s) as being defeating and/or humiliating. If they then feel trapped by the situation and believe that they have insufficient rescue or support available to cope or escape, they may develop suicidal ideation and intent.
An example of a particular psychological factor relevant to the first part of the model, the pre-motivational phase, is that of socially prescribed perfectionism. If a person is high on this type of perfectionism, they tend to believe that those around them have high expectations of them and are much more likely to experience feelings of defeat and humiliation in the face of a negative life event, such as loss of employment. The IMV model describes a number of other psychological factors or moderators that may influence the likelihood of suicidal thoughts emerging. It also draws on Joiner’s interpersonal model (Van Orden et al., Reference Van Orden, Witte, Cukrowicz, Braithwaite, Selby and Joiner2010), highlighting the role of feeling a burden on others and thwarted belongingness as moderators contributing to the development of suicidal ideation and intent (Dhingra et al., Reference Dhingra, Boduszek and O’Connor2015; O’Connor and Portzky, Reference O’Connor and Kirtley2018).
The final part of the model, the volitional phase, describes the key factors associated with the transition from suicidal thoughts to suicidal behaviours. This phase includes Joiner’s concept of having the capability for suicide (having high levels of physical pain tolerance and fearlessness about dying) which acts as a moderator, facilitating the transition from suicidal ideation towards suicidal behaviour. Further examples of moderators here include the access to means, impulsivity, and mental imagery about dying or death.
Although there have been previous books and chapters on applying CBT to suicidal behaviour (e.g. Bryan and Rudd, Reference Bryan and Rudd2018; Wenzel et al., Reference Wenzel, Brown and Beck2009), this current paper aims to do something novel in terms of the target audience and the target patient group. This paper focuses on CBT therapists who work in a range of settings, not purely in acute care, to consider factors affecting suicide risk in those in all stages of the pathway towards suicidal behaviour. To do so, we use the IMV model as the underpinning framework. Although the IMV model may not be known to all CBT therapists, it is likely that many of elements of the model are concepts that are very familiar to most CBT therapists. First, however, Section 1 examines the core processes and skills utilised within risk assessment and formulation and links these to standard CBT practice. Section 2 then details the concepts within the IMV model and makes clear their role in general CBT.
Section 1: Core CBT skills and processes utilised in risk assessment and formulation
Many of the skills that CBT therapists already possess are highly transferable to assessing, formulating and intervening with factors that could lead to suicidal behaviour. Before examining specific knowledge and skills, however, it is worth mentioning that the philosophy behind contemporary models of risk formulation is also a good fit for CBT therapists
Empirically driven philosophy (individualised to the patient)
The generic model of CBT and the various disorder-specific models are based on experimental and clinical research (Westbrook et al., Reference Westbrook, Kennerley and Kirk2012) (e.g. the principle of how certain behaviours can maintain beliefs) but CBT, when delivered well, is applied in a highly personalised and individual way. This ensures that the personal relevance and meaning of factors is considered for each individual when applying evidence-based models. For example, CBT treatment for panic disorder is usually based on an empirically grounded model of panic disorder (Clark, Reference Clark1986) and yet remains focused on an individual’s reported physical experiences and the personal meanings attached to these. The same applies in the field of risk assessment; Lewis and Doyle (Reference Lewis and Doyle2009) describe ‘a shift of focus away from simply identifying risk factors to thinking about how the key variables that are unique to the individual (predisposing, precipitating, prolonging, and protective) interact to cause them to act’ (p. 288). So rather than adopting an actuarial predictive approach as an insurance company might do when assessing risk for large numbers of car drivers, someone formulating risk might be aware of the evidence base around certain risk factors but would consider its relevance to each individual and formulate how it might act to increase or even decrease risk for that individual (and in what circumstance).
Assessment skills
The gathering of information around the presenting problem is central to CBT [both low intensity CBT (Papworth, Reference Papworth, Papworth and Marrinan2019) and high intensity CBT (Westbrook et al., Reference Westbrook, Kennerley and Kirk2012)], in order to understand the patient’s experiences and to develop hypotheses around problem development and maintenance. Many of the domains that we might ask about in a standard CBT assessment are also ones that we might also ask about when considering factors which may impact on risk. For example, we might enquire about context, thoughts, behaviours, physical sensations and feelings and then start to hypothesise and explore potential triggers and maintaining behaviours. Risk assessment would equally be considered an ongoing process rather than something that occurs as a one-off during an initial appointment. Similarly, in standard CBT assessments and specific risk assessments we may reflect on information gathered via questionnaires but use this as a prompt for exploration with the individual.
Whilst the skills and methods of assessment are very similar, it needs acknowledging that the assessment of risk factors for suicide would also include specific areas not typically covered within a standard CBT assessment. Examples detailed in a recent self-harm and suicide prevention competence framework for work with adults and older adults (National Collaborating Centre for Mental Health, 2018) include a range of areas that CBT therapists would routinely ask about (e.g. negative expectations based on prior experiences with the health or social care system, a sense of social isolation). However, it would also include a range of content areas that are perhaps less routinely asked about, for example, a sense of burdensomeness to others, markers that indicate the development of the capability to carry out suicide or self-harm, protective factors that may be associated with decreased thoughts of suicide or feelings that life is not worth living. These areas will be covered in detail within the remainder of this paper. One key thing to note is that this risk formulation framework emphasises ‘a collaborative assessment of risk, needs and strengths, which engages a person in a meaningful dialogue that helps them to consider their difficulties, the context in which these difficulties arise and the resources available to help keep them safe’ (National Collaborating Centre for Mental Health, 2018; p. 15) rather than a reliance on formal tools, scales or attempts to predict risk. Westbrook and colleagues describe CBT assessment in similar terms where it is viewed as ‘not a simple matter of ticking off a checklist of symptoms or completing a standard life history. Rather, it is an active and flexible process of repeatedly building and testing hypotheses’ (Westbrook et al., Reference Westbrook, Kennerley and Kirk2012; p. 67).
Formulation-based
Similarly, CBT therapists have extensive declarative knowledge around the role of formulation, but they are also well practised in the skill of formulating psychological problems. These skills are highly transferable to the current best practice in terms of formulating suicide risk (National Confidential Inquiry into Suicide and Homicide in Mental Health, 2018) including to the development of individualised safety plans (Stanley and Brown, Reference Stanley and Brown2012). The more recent emphasis on risk formulation in the world of suicide prevention (and also forensic psychology) is based on the established literature around the formulation of presenting problems within mental health (Hart et al., Reference Hart, Sturmey, Logan and McMurran2011), and as such, it is a natural fit for CBT therapists. Risk formulation plays a similar role to CBT problem formulations in terms of providing a shared understanding, road map for interventions, etc. Interestingly, evaluative criteria for a good risk formulation would largely apply directly to standard CBT problem formulations (e.g. external coherence, factual foundation, internal coherence, explanatory breadth, acceptability, treatment utility) (Hart et al., Reference Hart, Sturmey, Logan and McMurran2011).
Emphasis on the interpersonal relationship
The therapeutic relationship in CBT is widely agreed to be essential to effective CBT. It is defined as ‘as an exchange between therapist and client that develops for the purpose of sharing intimate, thoughts, beliefs, and emotions in an endeavour to facilitate change. This relationship is characterised by a safe, open, non-judgmental atmosphere that imbues trust and confidence’ (Kazantzis et al., Reference Kazantzis, Dattilio and Dobson2017; p. 17). Key features of the therapeutic relationship in CBT include its collaborative nature, its structure and active approach, the use of Socratic dialogue and the use of the therapeutic relationship to test out interpersonal beliefs (Bennett-Levy et al., Reference Bennett-Levy, Butler, Fennell, Hackman, Mueller and Westbrook2004). Almost all the knowledge and skills that CBT therapists employ in developing effective therapeutic relationships will be useful in both gathering relevant risk information but also in intervening to reduce these risk factors. It has been suggested that key transferable skills include communicating complex things clearly and simply (e.g. around suicidal behaviour), instilling hope, and providing a clear framework for treatment with active engagement by both parties. Helping the patient feel listened to and understood is also vital (Michel, Reference Michel2021). Again, just as in CBT, therapists are encouraged to reflect on their own thoughts and feelings ahead of, during, and after sessions (Haarhoff and Thwaites, Reference Haarhoff, Thwaites, Haarhoff and Thwaites2016). It is suggested this happens particularly when working with risk, to be aware of our own thoughts and feelings specifically around suicide in order to guard against avoidance or anxiety that may impede treatment (Foster et al., Reference Foster, Alderman, Safin, Aponte, McCoy, Caughey and Galynker2021).
Cultural awareness and adaptations
Although there are gaps in many core CBT training courses around cultural inclusivity, many CBT organisations are now becoming more explicit that CBT therapists need to be trained in how to consider the cultures and context of the patient and adapt CBT accordingly (British Association for Behavioural & Cognitive Psychotherapies, 2021). Everyone possesses a personal culture and belief system (Tribe, Reference Tribe, Moodley and Icampo2014) despite culture often being misunderstood as something only had by ethnically or racially minoritised others (Patallo, Reference Patallo2019). Cultural identities have various intersecting aspects including ethnicity or racialised identity, religion, nationality, gender identity, transgender, sexual orientation, socio-economic status, physical and intellectual ableism, neurodiversity and age. Each part of an identity plays a part in how an individual views and experiences themselves and the world. This includes how mental health problems are experienced, expressed, responded to, and views about help seeking and treatment interventions.
For CBT therapists using a specific model, for example, best practice might be to reflect on the culture of the patient, potential cultural differences with the therapist, how aspects of the therapist’s own identities may influence their understanding of the patient’s difficulties (Patel, Reference Patel, Fleming and Steen2004) and demonstrate curiosity and humility in terms of gaps in knowledge and identify where adaptations may need to be made to the model (Haarhoff and Thwaites, Reference Haarhoff, Thwaites, Haarhoff and Thwaites2016). A useful definition of cultural humility to keep in mind for the delivery of CBT and risk assessment and formulation is:
‘cultural humility involves the ability to maintain an interpersonal stance that is other-oriented (or open to the other) in relation to aspects of cultural identity that are most important to the client. Cultural humility is especially apparent when a therapist is able to express respect and a lack of superiority even when cultural differences threaten to weaken the therapy alliance’
Hook et al., Reference Hook, Davis, Owen, Worthington and Utsey2013 (p. 354)Cultural humility also involves therapists creating opportunities to invite discussion about a patient’s cultural identity and tolerating any discomfort that engaging in these conversations may evoke (Davis et al., Reference Davis, DeBlaere, Owen, Hook, Rivera, Choe, Van Tongeren, Worthington and Placeres2018). Cultural discussions are an essential element of therapy and risk assessments (Drinane et al., Reference Drinane, Owen and Tao2018) and it is important that therapists both initiate these discussions and create a safe and welcoming space that supports patients in addressing the aspects of their identities related to their presenting problems (Gordon, Reference Gordon, Ade-Serrano and Nkansa-Dwamena2020; Hook et al., Reference Hook, Davis, Owen and DeBlaere2017). Such opportunities need to be provided for patients, as those with marginalised identities may have had impactful life experiences involving oppression and discrimination and been socialised not to talk about such experiences or their emotions voluntarily (Foy et al., Reference Foy, Morris, Fernandes and Rimes2019; Graham et al., Reference Graham, Sorenson and Hayes-Skelton2013).
For example, if a therapist is working with someone from a different culture to their own, they might consult a more generic text around cultural adaptations (e.g. Beck, Reference Beck2016) or a specific paper looking at cultural adaptations of a disorder-specific model (e.g. SAD: Chapman et al., Reference Chapman, DeLapp and Williams2013; PTSD: Williams et al., Reference Williams, Printz, Ching and Wetterneck2018; OCD: Williams et al., Reference Williams, Rouleau, La Torre and Sharif2020). These same principles of cultural sensitivity and a willingness to discuss the impact of culture should similarly be applied to understanding cultural aspects of risk presentation when working with patients who may be suicidal. A useful and simple empirically derived framework to reflect on cultural aspects of risk has identified four main areas where culture can intersect with risk and where consideration needs to be focused (Chu et al., Reference Chu, Goldblum, Floyd and Bongar2010). Care needs to be taken not to pathologise specific aspects of culture where the issues are more likely to relate to lack of an understanding by staff or systemic discrimination rather than anything intrinsic to a specific culture. The four main areas of the model are:
-
(i) Cultural sanctions (messages of approval or acceptability within an individual’s culture) (e.g. the acceptability of suicide as an option has been found to differ between people from different ethnicities, cultures and religions and therapists need to be aware of this and not make any assumptions based on their own culture).
-
(ii) Idioms of distress (cultural variations in how psychological distress is displayed) (e.g. one study has found that individuals from minoritised ethnicities are less likely to express suicidal ideation (Morrison and Downey, Reference Morrison and Downey2000), externalised expression of distress will differ between people based on their experiences and may be missed as potential flags for increased risk of suicide unless the therapist is able to display cultural humility and curiosity to explore the meaning of specific expressions of distress).
-
(iii) Minority stress (the specific stresses that cultural minorities experience because of their identity) (e.g. social disadvantage and discrimination have been found to increase suicide risk). For example, UK-based therapists will need to be aware of increases in racist incidents following Brexit (Komaromi, Reference Komaromi2016) and be willing to explicitly ask about racism and include this within formulations of both the presenting problem and suicide risk when relevant (Beck, Reference Beck2019).
-
(iv) Social discord (cultural variations in the social factors that play a role in suicide risk for minority groups) [e.g. for some ethnicities familial conflict can be particularly associated with increased suicide risk (Cheng et al., Reference Cheng, Fancher, Ratanasen, Conner, Duberstein, Sue and Takeuchi2010), for LGBTQ individuals, familial rejection or alienation can also play a role in increased suicidality (D’Augelli et al., Reference D’Augelli, Grossman, Salter, Vasey, Starks and Sinclair2005)]. CBT therapists need to be willing to suspend assumptions based on their culture and life experience and ask questions to try and understand experiences and meanings from their patient’s perspective.
In summary, as for culturally sensitive or culturally adapted CBT, culturally sensitive risk assessments and formulations should also embody cultural humility, involve therapist reflection on personal assumptions and gaps in knowledge and be individualised to the specific patient. When working with patients with minoritised statuses, therapists should inquire about any negative consequences that they may experience linked to aspects of their diverse identities (Pantalone et al., Reference Pantalone, Iwamasa, Martell and Dobson2010). Therapists should also conduct an exploration of any contextual factors in society at large that may be linked to their presenting problems and may have potential to increase risk.
Section 2: Using CBT to address the biopsychosocial context in which suicidal ideation and behaviour may emerge (pre-motivational phase variables)
Within the following sections we move from reflecting on how core CBT skills equip therapists to gather information and formulate suicide risk to the potentially less familiar area of exploring key features of the IMV model, with respect to the routine work of CBT therapists working with common mental health problems.
Firstly, we will flag up some of the key biopsychosocial background factors that CBT therapists may well be aware of and consider their potential role in the emergence of suicidal ideation. The IMV model does not suggest that these factors necessarily lead to suicidal behaviour, or operate in the same way for all individuals; they are seen as functioning within a diathesis–environment–life events triad (with diatheses including biological, genetic or cognitive vulnerability factors which may increase risk of suicide) (O’Connor and Kirtley, Reference O’Connor and Kirtley2018). The suggestion is that CBT therapists may reflect upon these areas as potential factors that can be considered during ongoing assessment and any potential interactions formulated and addressed.
Socioeconomic deprivation
A criticism often directed at CBT is that it fails to consider society, culture and context in sufficient detail within problem formulations (Hays, Reference Hays, Iwamasa and Hays2019). Socioeconomic deprivation is already identified as a factor within the IMV model due to its link with suicide, with those living in the most deprived areas and in the lowest socio-economic group being ten times more at risk of suicide than those in the most affluent groups (Platt, Reference Platt, O’Connor, Platt and Gordon2011). For this reason alone, CBT therapists need to consider this within assessment and formulation (whilst not minimising risks to people in higher socio-economic groups). However, there is also growing data that socio-economic status also impacts on the access to, and clinical outcomes from, CBT. Studies have found that individuals from lower socio-economic groups have worse mental health, a higher need for mental health care (Finegan et al., Reference Finegan, Firth, Wojnarowski and Delgadillo2018), less likelihood of accessing psychological therapy (Delgadillo et al., Reference Delgadillo, Farnfield and North2018), worse outcomes within IAPT services (Delgadillo et al., Reference Delgadillo, Asaria, Ali and Gilbody2016), and require a greater number of sessions to benefit from therapy (Finegan et al., Reference Finegan, Firth and Delgadillo2019).
Within their roles, CBT therapists (rather than as voters or activists), are limited in their ability to influence the socio-economic status of their patients or the areas in which they live but this does not mean that these factors can be ignored in assessment, formulation or interventions. CBT therapists can ask about relevant factors such as safety, housing, debt, racism and poverty. They can acknowledge and validate any role these may play in their formulation and signpost to other sources of support whether this is food banks, employment advisors, debt advice or housing support. It is important that any acknowledgement or signposting related to, for example, financial difficulties is normalised so as not to inadvertently engender any sense of shame or humiliation that could heighten risk. Therapists can ensure that they do not make assumptions within their therapeutic interventions and ensure these are adapted to take account of each individual’s circumstances, e.g. when working on social anxiety, by not suggesting that the individual goes into their local high street coffee shop for a coffee when they actually cannot afford to put their heating on or buy regular food for their children.
Social perfectionism
Perfectionism is a transdiagnostic factor that most CBT therapists will be aware of, with respect to its role in the development and maintenance of a range of disorders including eating disorders, depression, bipolar disorder, OCD, social anxiety, panic disorder, worry and some personality disorders (Egan et al., Reference Egan, Wade and Shafran2011). CBT therapists are also likely to have experienced the negative impact that perfectionism can exert on treatment outcomes.
Of particular note in the IMV model and suicide risk literature is socially prescribed perfectionism which is a specific subtype of interpersonal perfectionism involving ‘the perceived need to attain standards and expectations prescribed by significant others. Socially prescribed perfectionism entails people’s belief or perception that significant others have unrealistic standards for them, evaluate them stringently, and exert pressure on them to be perfect’ (Hewitt and Flett, Reference Hewitt and Flett1991; p. 457). This is thought to be particularly important in the journey towards suicidal behaviour because social perfectionism often involves a strong need for approval and sensitivity to criticism (Flett et al., Reference Flett, Hewitt, Sherry, Zeigler-Hill and Marcus2016) and can also therefore lead to disconnection and isolation from others (Flett and Hewitt, Reference Flett and Hewitt2020) and hopelessness (Smith et al., Reference Smith, Vidovic, Sherry, Saklofske and Kumar2017). Those who are rated highly for socially prescribed perfectionism also tend to report high levels of defeat and entrapment (O’Connor, Reference O’Connor2021). The description below will not be unfamiliar to most CBT therapists:
‘Perfectionism can certainly be a problem in living for the person who is experiencing extreme socially prescribed perfectionism and who has an overarching concern with making mistakes but he or she has a life situation characterized by low personal control, exceedingly high demands, and low support’
(Flett and Hewitt, Reference Flett and Hewitt2020)Perfectionism, especially social prescribed perfectionism is a complex construct and can be difficult to identify and assess. There is also some initial work on how this can lead to ‘perfectionistic self-presentation’ which involves the need to present a flawless front and subsequent self-concealment behaviour on the part of individuals (Flett et al., 2014). CBT therapists can utilise their standard CBT questioning skills to identify emotional experiences linked with dysfunctional beliefs and behaviours, which can then be combined with their knowledge and understanding of transdiagnostic factors, disorder-specific models, and the IMV model to formulate how socially prescribed perfectionism might potentially function to increase risk for specific patients.
There are validated measures that can be used to assess perfectionism (that include subscales measuring socially prescribed perfectionism including items such as ‘The better I do, the better I am expected to do’ and ‘My family expects me to be perfect’) (Hewitt and Flett, Reference Hewitt and Flett1991). Standard cognitive and behavioural techniques can be adapted to explore and test out dysfunctional beliefs just as for other treatment areas. Can the individual identify and put the belief into words (this might be the first time they have become aware of it)? Can it be tested out, whether in discussions with those around them or in more formal behavioural experiments? Does the belief lead to cognitive or attentional biases that might need addressing? (Egan et al., Reference Egan, Wade and Shafran2011). At times the focus might need to switch to treating the perfectionism itself and there are specific CBT standalone treatments for this (both low and high intensity) that have been found to be effective (Shafran et al., Reference Shafran, Coughtrey and Kothari2016).
Section 3: Using CBT to address the factors that lead to the emergence of suicidal ideation: threat to self-moderators (TSMs) and motivational moderators (MMs)
Within the IMV model, a range of factors are hypothesised to contribute to the emergence of suicidal ideation. Although these are not exclusive to individuals experiencing mental health problems, they will be extremely familiar to CBT therapists from the presentations that they encounter in their routine work. This section details their role in the development of suicidal ideation and helps CBT therapists to be alert to these, and also consider how intervening with these may impact on suicidal risk.
The motivational phase of the IMV model details two sets of psychological factors that may explain the development of suicidal ideation and intent. Threat to self-moderators (TSMs) influence the likelihood of an individual developing a perception of being trapped in their current circumstances with an increased sense of hopelessness for any escape. This sense of entrapment can be subdivided into external entrapment and internal entrapment (Gilbert and Allan, Reference Gilbert and Allan1998), concepts that may already be familiar to CBT therapists. External entrapment relates to external events or circumstances such as loss of employment or long-term health problems (‘I have no chance of finding work’, ‘My breathing is getting worse, I’ll never be able to exercise again’), whilst internal entrapment is a sense of being unable to resolve distressing emotions or manage aversive thoughts (‘I can’t stand feeling like this any longer’, ‘I’m losing my mind’). Once a sense of entrapment has become established, further psychological moderators, the motivational moderators (MMs), are potential explanatory factors for the development of suicidal ideation and intent. These can further intensify the level of distress and lack of hope of escape.
Entrapment is largely driven by personal appraisals of one’s current situation and as such, the latter can at times be highly realistic (e.g. ‘I am trapped living with my current critical partner until we manage to sell our flat’) and as such they can be acknowledged, validated and then problem-solved. At other times these appraisals might be distorted or exaggerated and therefore can be addressed using a range of standard CBT interventions (Roth and Pilling, Reference Roth and Pilling2007) from specific mini-formulations to increase understanding of the problem and what maintains it, through to thought diaries, the management of rumination or chair work (Pugh, Reference Pugh2019). The defeat behaviours (e.g. giving up on job applications to avoid further setbacks, self-isolation for fear of shame-inducing conversations, non-concordance with health care regime or resignation to a living situation due to loss of hope) that can develop as a result of perceptions of entrapment (and which maintains them) can also be addressed via standard behavioural interventions from behavioural activation or activity scheduling through to behavioural experiments.
Threat to self-moderators
Social problem-solving
Social problem-solving difficulties can relate to a perceived inability to deal with challenging situations (often termed as problem orientation) as well as to an actual skill deficit. The ability to problem-solve can also be compromised by memory biases (also identified as a TSM) that are activated during periods of depression. The generation of potential solutions to a specific problem is aided by drawing upon experience of overcoming similar challenges in the past. However, memories can become overgeneralised and therefore lacking the necessary detail to be useful, and memories of negative events can also be more readily recalled than memories of challenges that were successfully overcome. Problem-solving training (Nezu et al., Reference Nezu, Maguth Nezu and D’Zurilla2013), which is routinely employed in both low and high intensity CBT interventions for depression and anxiety disorders, is known to be effective for both improving problem-solving ability and increasing confidence in facing potential future challenges. Addressing barriers to problem-solving during the treatment of depression or an anxiety disorder can help an individual actively address the social problems that contribute to their sense of entrapment. In the previous example, following the loss of employment, a sense of entrapment was related to the belief in the impossibility of finding work again. Problem-solving might help the person to reflect on how they had coped with a similar experience in a previous recession. They could be helped to generate possible action plans and select the option most likely to be accessible and successful. For instance, engaging with an employment advisor to help with curriculum bitae (CV) writing and interview practice may be beneficial. Furthermore, such support can help the person envisage a more positive future (i.e. strengthen positive future thinking which is a MM) and most crucially help the individual envisage an alternative to suicidal behaviour.
Rumination
Rumination can be directly addressed in CBT for depression (Watkins, Reference Watkins2009) in a similar way to the management of problematic worry in generalised anxiety disorder treatment protocols (Dugas and Koerner, Reference Dugas and Koerner2005). Rumination is conceptualised as behaviour over which control can be enacted. Treatment generally starts with monitoring to increasing awareness of rumination. This aids the recognition of any triggers and the identification of possible intended functions of rumination, e.g. to try to identify ways to stop aversive events in the future, to uncover acts of wrong doing, or to avoid current necessary but aversive tasks. Such maladaptive beliefs about the benefits of rumination can then be identified and reappraised. This approach can be productively employed by CBT therapists to address rumination when it acts as a moderator in the development of suicidal ideation. Specifically, reduction in rumination can contribute to an individual experiencing less internal entrapment and help mobilise their efforts to address problems related to a sense of external entrapment.
Motivational moderators
Thwarted belongingness
Thwarted belongingness is one of the components identified in Joiner’s interpersonal theory (IPT; Van Orden et al., Reference Van Orden, Witte, Cukrowicz, Braithwaite, Selby and Joiner2010) of suicidal behaviour. The theory posits that a feeling of belonging is a basic human need, and a sense of loneliness or lack of social connectivity can be risk factors for the development of a desire for suicide. A recent systematic review (McClelland et al., Reference McClelland, Evans, Nowland, Ferguson and O’Connor2020) reported on the evidence that loneliness is a predictor of later suicidal ideation and behaviour. The Interpersonal Needs Questionnaire (INQ) (Van Orden et al., Reference Van Orden, Cukrowicz, Witte and Joiner2012) was developed to measure two constructs of the IPT model, perceived burdensomeness and thwarted belongingness. Items specific to thwarted belongingness include ‘These days, I feel disconnected from other people’, ‘I feel that there are people I can turn to in times of need’ and ‘I have at least one satisfying interaction every day’. Behavioural activation, which is a recommended first-line treatment for depression (NICE, 2022), may be utilised to specifically address these perceptions of disconnection and loneliness by planning interactions with other people. The focus of this intervention is to increase engagement in pleasurable and rewarding activities and to help schedule necessary and routine tasks. Patients are also guided to reduce the time spent on unproductive and depressogenic rumination over past losses and perceived failures. For individuals experiencing suicidal ideation, the MM of perceived lack of belongingness could be specifically targeted by careful scheduling of activities that promoted engagement with significant others or with community or voluntary groups. Support to engage in necessary and routine tasks could focus on overcoming avoidant coping styles which act as a moderator within the motivational phase of the model. Reduction in time spent ruminating (a TSM) would also serve a protective function as discussed previously.
Goals
Research has indicated that an individual’s self-regulation when encountering unattainable goals may increase their risk of developing suicidal ideation (O’Connor et al., 2012). Two specific components have been identified, goal disengagement and goal re-engagement. Generally, the ability to disengage from unattainable goals whilst then re-engaging in more achievable ones is seen to be adaptive. However, there is evidence of variation across the lifespan. For younger people the inability to disengage is most clearly linked to emotional disturbance, whilst for older people the most significant problem is if there is a lack of accessible new goals following successful disengagement.
Goal setting is a key component of CBT-based interventions and is often based on the SMART goal approach (Specific, Measurable, Achievable, Realistic and Time bound) (Doran, Reference Doran1981). However little attention is paid to the impact of self-regulation of goals. In the context of suicide prevention, CBT practitioners would be advised to consider extending goal planning to include acceptance that goals may not be achieved and to encourage a self-compassionate response that supports flexibility to reset unhelpful goals (Barton et al., Reference Barton, Armstrong, Freeston and Twaddle2008). Giving consideration of a different focus across the lifespan, e.g. helping young people to disengage from unattainable goals or helping older people access new achievable goals, maybe most beneficial.
Section 4: Using CBT to address the factors associated with the transition from suicidal ideation to suicidal behaviour (volitional moderators: VMs)
The final phase of the IMV model describes the factors that are associated with the transition from thoughts of suicide to suicidal behaviour. Although much of the work to address this will occur in crisis teams, there is a strong rationale for CBT therapists working in primary care and other secondary care settings to have a knowledge of the volitional moderators. First, knowledge of these factors is important in the development of a risk formulation and second, it will inform safety planning.
Capability
This moderator is also derived from Joiner’s interpersonal theory model (Van Orden et al., Reference Van Orden, Witte, Cukrowicz, Braithwaite, Selby and Joiner2010). This theory proposes that the desire for suicide develops through the interaction of perceptions of lack of belonging and of burdensomeness. However, this desire (i.e. suicidal ideation) will only transition to behaviour in the presence of the capability to act on that desire. To engage in suicidal behaviour, one has to overcome the natural human instinct to survive. Prior to enacting a suicide plan a person must first have acquired the capability to carry out that specific behaviour. Capability to engage in suicidal acts is thought to consist of two components: fearlessness about death and increased tolerance to physical pain. The innately powerful and protective nature of these components can be diminished through habituation (exposure that reduces fear of death and of pain) and by strengthening of the opposing behaviour (e.g. reinforcement of self-harm by the experience of relief from emotional distress or distraction from aversive thoughts). Habituation to the fear of dying and the physical pain often associated with a suicidal act, can occur through acts of preparation as well as previous suicide attempts, and through self-harm (Van Orden et al., Reference Van Orden, Witte, Cukrowicz, Braithwaite, Selby and Joiner2010). This may explain why these two past behaviours are amongst the most significant of the known risk factors for suicide. It is important to note the evidence that all self-injurious behaviour regardless of whether there is suicidal intent is associated with later suicide attempts including those described as non-suicidal self-injury (NSSI) (Kiekens et al., Reference Kiekens, Hasking, Boyes, Claes, Mortier, Auerbach, Cuijpers, Demyttenaere, Green, Kessler, Myin-Germeys, Nock and Bruffaerts2018). Habituation can also occur through planning, as this can act as mental preparation, through exposure to suicide in others and through a vocation that exposes a person to death or the risk of death. People with a trait of impulsivity may be more exposed to risk through their lifespan which also may increase the chance of habituation and the acquired capability to suicide (Van Orden et al., Reference Van Orden, Witte, Cukrowicz, Braithwaite, Selby and Joiner2010).
From a clinical perspective, an understanding of the role of acquired capability will aid assessment and help in developing a shared understanding of risk with the individual you are working with. For example, it may help in explaining the concern that non-suicidal self-harm may elevate suicide risk and in engaging an individual in efforts to reduce self-harm and put measures in place to reduce access to means. Examples of reducing access to means are enlisting the help of a trusted other to control medication supplies, changing prescriptions so smaller quantities of medication are available, or encouraging the disposal of rope or of tools used for cutting. Reducing episodes of self-harm (e.g. teaching more benign ways to distract from emotional distress as employed in the safety planning described later) may in turn act to mediate the effects of habituation and reinforcement.
Impulsivity
Anecdotally, practitioners often fear that a tendency towards impulsivity in people at risk of suicide negates the benefit of safety planning and severely limits the opportunities to intervene to prevent suicide. Studies have indeed concluded that the time elapsed between the decision to act and the suicide attempt often indicates the impulsivity of the behaviour. For example, Deisenhammer et al. (Reference Deisenhammer, Ing, Strauss, Kemmler, Hinterhuber and Weiss2009) found that nearly half of all suicide attempts in their study occurred within 10 minutes of the decision to act and Simon et al. (Reference Simon, Swann, Powell, Potter, Kresnow and O’Carroll2001) reported that a quarter were within 5 minutes of making the decision. However, it is also worth noting that a history of impulsivity, or trait impulsivity, is not directly associated with suicide ideation and the indirect link to suicidal behaviour is through the habituation effect described earlier (Hadzic et al., Reference Hadzic, Spangenberg, Hallensleben, Forkmann, Rath, Strauß, Kersting and Glaesmer2020).
Given the impulsive nature of much suicidal behaviour the importance of restricting or delaying access to means as part of safety planning becomes apparent.
Access to means
When suicidal ideation has developed to the point of the formation of a suicide plan, collaborative efforts to restrict access to the preferred method are paramount. Arguably this is best agreed at a point in time when the level of intention to act on the plan is low. This underpins the rationale for reduction in access to means being the last stage in the preparation of a safety plan. A person is more likely to agree to putting barriers between themselves and the chosen suicide method (e.g. asking a family member to look after their medication) if they have confidence in a pre-prepared and detailed list of agreed, alternative ways to manage their suicidal ideation.
Restricting access to the means of suicide is among the most effective public health interventions to prevent suicide (Hawton, Reference Hawton2007). There are two factors that support the importance and effectiveness of restriction of access to means. First is that for many people the period of crisis can pass in a matter of minutes or hours (Hawton, Reference Hawton2007) so that restricting access to means can give time for the acute crisis to pass before fatal action is taken (World Health Organization, 2021) and second there is evidence that when access to means is restricted the substitution to other means is limited (Zalsman et al., Reference Zalsman, Hawton, Wasserman, van Heeringen, Arensman, Sarchiapone, Carli, Höschl, Barzilay, Balazs and Zohar2016).
The role of preparation and planning in habituation to fear of suicide noted in the previous section is also thought to explain, in part, why method substitution does not frequently occur. The reduction in fear is specific to the chosen method and not necessarily transferable. Similarly, an increased tolerance to physical pain is thought to be method-specific and therefore not conducive to method substitution (Hadzic et al., Reference Hadzic, Spangenberg, Hallensleben, Forkmann, Rath, Strauß, Kersting and Glaesmer2020).
Safety plans
A core competency for all mental health practitioners (including CBT therapists) should be the ability to work with a person collaboratively and compassionately to produce a safety plan (National Collaborating Centre for Mental Health, 2018). Together they establish how to identify an impending suicidal crisis as early as possible. The five areas approach that is familiar to CBT practitioners is key here to recognise situations (e.g. relationship problems), thoughts (e.g. suicidal ideation including images, rumination), emotions (including shame or guilt), behaviours (e.g. arguments, poor sleep) or physical sensations (including urges to self-harm) that may act as warning signs and a trigger to implement the safety plan. The safety plan (Stanley and Brown, Reference Stanley and Brown2012) is organised in five stages which detail actions that can be taken to manage the suicide crisis. The individual is advised to select the stage most appropriate to the level of risk and to escalate through the stages as necessary.
The first stages are:
-
(1) Activities the person can engage in to distract themselves.
-
(2) People they can contact to help with distraction when a crisis is escalating.
-
(3) People who can help more specifically with the crisis.
-
(4) Professional and emergency contacts to help resolve the crisis.
Once these stages have been agreed, the final task in safety planning is then to work to reduce access to the means identified in the individual’s suicidal ideation and plan. The safety plan is recorded either on paper or electronically; with multiple copies given to the individual to be kept where they will be most readily accessible. If appropriate, copies can also be given to those who provide support, e.g. close friends or family members.
In-session behavioural experiments can be an effective way to strengthen the utility of safety plans. For example, they can be used to check the practicality of the plan (e.g. is this person available at the time most needed?) or to test negative predictions related to help seeking (e.g. phoning the Samaritans to overcome fear of feeling judged). The aim is to help the person think about their safety plan as a crisis plan which can help interrupt their suicidal thoughts so that they are less likely to cross the precipice from suicidal thoughts to suicidal acts (O’Connor, Reference O’Connor2021).
Conclusions
Within this paper we have introduced the IMV model of suicidal behaviour and demonstrated how it can be used as a framework for CBT therapists to conceptualise, assess, formulate and address potential factors that may increase the risk of suicidal behaviour.
Space does not permit a full discussion of all potential training and supervision implications of the IMV model for CBT therapists; however, some key points are flagged here. Anecdotal reports suggest that training on models of suicidal behaviour for CBT therapists can be anxiety-provoking and at times overwhelming. We hope that this paper makes it clear that CBT therapists already possess a range of knowledge and skills that can be integrated with new specific information (from, for example, the IMV model) in a way that can feel less overwhelming. The key message is that the task is not to learn a brand-new set of skills, but to learn how to apply the already acquired skills of assessment, formulation and intervention techniques to (a) areas they may already be considering that also impact on suicide risk (e.g. perfectionism) and to (b) new areas that disorder-specific CBT models may not necessarily include (e.g. socio-economic deprivation, safety planning).
Clinical supervisors and therapists ideally need to have a shared framework for considering and formulating suicidal behaviour and potential factors that may make this more likely. We think that the IMV model is a natural fit for CBT therapists and is broad enough to provide structure when considering suicidal risk at all stages, Regular discussion of suicide risk factors is likely to maintain a focus on their consideration and how they can be addressed. It also serves the important function of normalising such discussion and preparing CBT therapists for a potential incident when one of their patients does take their own life, an experience which evidence tells us can exert a profound effect on therapists (Sandford et al., Reference Sandford, Kirtley, Thwaites and O’Connor2020).
Key practice points
-
(1) The IMV model is an evidence-based tripartite model that can be used to formulate and understand the dynamic nature of the risk of suicidal behaviour.
-
(2) Although the majority of CBT therapists may not view themselves as experts in suicide prevention, suicide prevention is relevant to all therapists (and indeed all individuals in society), and CBT therapists have a range of core skills and knowledge that they can use to assess and formulate risk factors for suicidal behaviour.
-
(3) CBT therapists can intervene (and often already are) to address key features that can increase the risk of an individual progressing to suicidal behaviour, and the IMV model provides a framework to understand this and potentially fine-tune the formulation and interventions.
Data availability statement
Not applicable – no new data were generated.
Acknowledgements
Thanks to Carla Quan-Soon for substantial input into the section on cultural considerations and to Carla, Sylvia Bradley and Katrina Haines-Myers for many other helpful discussions.
Author contributions
David Sandford: Conceptualization (equal), Writing – original draft (equal); Richard Thwaites: Conceptualization (equal), Writing – original draft (equal); Olivia Kirtley: Writing – review & editing (equal); Rory O’Connor: Writing – review & editing (equal).
Financial support
This paper received no specific grant from any funding agency, commercial or not-for-profit sectors.
Conflicts of interest
Richard Thwaites is Editor of the Cognitive Behaviour Therapist. He was not involved in the review or editorial process for this paper, on which he is listed as an author. The other authors have no declarations.
Ethical standards
The authors have abided by the Ethical Principles of Psychologists and Code of Conduct as set out by the BABCP and BPS. The authors can confirm that no ethical approval was needed for this clinical guidance paper.
Open access
Comments
No Comments have been published for this article.