Resilience and risk factors are seen as at two ends of a pole in the unidimensional model. For example, a risk factor for suicidality may be feeling socially excluded whereas a resilience factor may be feeling socially connected and part of a community. There are two problems with this approach ^[1][35].

First, if a measure of social connectedness is found to be inversely related to a measure of suicidality, then it is sometimes concluded that social connectedness is a resilience factor. Indeed, social connectedness may form part of a resilience mechanism, but such a mechanism needs to be developed and tested. Unfortunately, there are many examples of this kind of assumption throughout the suicide literature which are too numerous to document.

The second and related problem is that not experiencing a suicide risk factor does not equate with a resilience factor. For example, hopelessness is a strong predictor of suicidal thoughts and acts including in those with psychosis ^[2][77]. However, not experiencing hopelessness is not synonymous with feeling hopeful about the future. Similarly, a negative perception of having serious financial problems is often one of the predictors of suicidal thoughts and acts ^[3][8]. To not have negative perceptions of having financial problems cannot be considered an aspect of resilience without further examination because: i. the financial problems may have resolved perhaps because someone close offered a loan, and/or ii. the relationship between negative perceptions of financial pressures and suicidality may have been buffered and weakened by other personal resources, such as finding other ways to alleviate financial pressures. In this case, the latter situation may represent a source of resilience which could be further explored.

2. The Two-Dimensional, Buffering, Model: Two Independent Dimensions

The two-dimensional approach to understanding resilience expands on the unidimensional approach. In the two-dimensional approach, risk factors and resilience factors are seen as independent dimensions. So, if a risk factor is present, such as intense stressors, it does not imply a lack of resilience. Indeed, if we go back to some of our qualitative findings, participants with psychosis who also had suicidal thoughts and acts were still able to get through their day, even though it took the most immense amount of effort for individuals to exist with mental health problems, and, in some cases they did this whilst accepting what they were going through and feeling that there were still reasons to live ^[4][73]. The buffering model takes the idea of two dimensions a little further. Resilience as a buffer examines the extent to which resilience factors can counter negative internal stressors (e.g., hallucinations) and external stressors (e.g., late disability payments) by weakening associations between such stressors and a negative outcome, such as, suicidal thoughts and behaviours ^[1][5][34,35]. To take a concrete example, a strong and positive association may be found between experiencing financial pressures and, because of those pressures, developing suicidal thoughts, but feeling meaningfully connected to others may counter, weaken or buffer, that association between financial pressures and suicidal thoughts and/or plans ^[6][78]. One cross-sectional study which recruited people who were suicidal and had psychotic symptoms found that high levels of resilience weakened the association between perceptions of stressful life events and suicidal thoughts and acts ^[7][67]. Another prospective study over a 3-month time span reported that over that time, defeat and entrapment predicted suicidal thoughts/acts only when psychotic experiences and the ensuing distress were high, and resilience was low. When resilience was low, this relationship between defeat, entrapment and suicidal thoughts/behaviours was amplified ^[8][79]. Hence, individuals reporting low levels of resilience may be more likely to experience suicidal thoughts and behaviours because of defeat, entrapment, psychosis, and the associated distress. Resilience in both studies was conceptualised as positive self-appraisals of emotional coping, social support and situation specific problem solving based on a theoretical model of psychological pathways to suicidal thoughts and acts ^[7][67]. It is important to note that many studies in the area of understanding pathways to suicidal experiences and pathways to resilience to suicide are cross-sectional. When examining the role of buffers to pathways to suicidality, prospective qualitative and quantitative designs, including those that are micro-longitudinal ^[9][80] are to be encouraged although not exclusively so ^[10][81].

3. The Recovery Model

Many conceptualizations of recovery have followed a medical model in which: i. negative events and negative internal and external stressors were identified; ii. the subsequent detriment to functioning caused by these events and stressors was assessed; and then iii. the degree to which these “deviations from normality” could be rectified signifying a return to normality or remission from the perceived ‘medical problem’ was determined ^[11][12][88,89]. The medical problem was defined in terms of symptoms. Health was defined as an easing of those symptoms ^{[12][13][14][15]}[89,90,91,92]. The idea of recovery, derived from more contemporary frameworks, focuses on how recovery can be best defined and understood from the personal perspectives of experts-by-experience (sometimes referred to as consumers) and not exclusively from psychiatric symptoms nomenclature. That is, a distinction is made between clinical recovery and personal recovery ^[16][83]. This grounding of recovery on the personal perspectives of people who are experts-by-experience is considered fundamental if advances in understanding recovery processes and mechanisms are to be made in ways that make a meaningful difference to experts-by-experience, scientists, clinicians and policy makers ^{[17][18][19][20][21]}[93,94,95,96,97]. Personal recovery embraces and explores experiences of personal growth as a result of coming through, or better still, being able to simply ‘be with’ severe mental health problems ^{[19][20][21][22][23]}[68,95,96,97,98]. There are extensive evidence-based examples of personal recovery in people with psychosis ^[11][16][24][82,83,88]. Yet, the question arises concerning the extent of the evidence for personal recovery in people experiencing both psychosis and suicidal thoughts/acts in tandem. Research that aims to understand recovery based on the views of a diverse range of people experiencing psychosis is expanding ^{[17][18][19][20][21][23][25][26][27][28][29]}[93,94,95,96,97,98,99,100,101,102,103]. As we have seen Ifrom some of the qualitative work that we have presented, it is not necessarily the case that the only way of finding meaning in life whilst having mental health problems is to escape from those problems ^[4][73]. It is potentially enlightening, especially from a therapeutic point of view, that some people have found that for them, suicidal experiences can co-exist in parallel with other aspects of their lives.

4. The Maintenance Model

The maintenance model is most closely allied with the positive psychology movement, one of the goals of which is to understand how people prosper and grow over time and across different realms of their lives ^{[28][29][30][31]}[102,103,104,105]. Being founded on the Psychological Flexibility Model ^[31][105], the Broaden and Build theory ^[30][31][104,105], and the Self-Determination Theory ^[32][33][106,107], this model describes a set of abilities which enable people to maintain a sense of positivity when facing negative experiences, stressors, and health conditions some of which are chronic and occur over long time periods. Being able to identify and engage with meaningful activities and values, despite significant adversity, is central to the maintenance model which was developed in the context of coping with chronic pain ^[11][88]. An important point is that this approach to resilience is not characterized by recovery nor a resolution/partial resolution of health problems due to buffering factors. Embracing personally meaningful activities and developing values which are important to an individual may be observed even when there is no improvement in health problems. One suggestion is that aspects of these kind of positive thoughts and behaviours may serve to deflect foci away from health problems in addition to strengthening psychological flexibility and the experience of positive emotions together with monitoring perceptions about the extent to which important personal needs have been both identified and fulfilled ^[11][88]. It can be argued that qualitative methodologies are best suited to exploring ways that people with psychosis who are suicidal are able to identify and enact personal values in the face of severe mental health problems, whilst also accepting those mental health problems at least to some extent.

5. Conclusions

There are two over-arching ways of conceptualising psychological resilience. The first was as a process and the second was as personal characteristics or resources of an individual. There were three concepts: i. immunity, such as, being able to resist the effects of stressors; ii. bouncing back, defined as returning to a state of cognitive-emotional-behavioural stability; and iii. growth in which positive transformation was harnessed as a result of adversity. Characteristics of individuals included: i. having personal resources and ii. having social resources, both of which were seen as having the potential to build and sustain resilience. Perhaps, thinking about resilience as a set of interacting processes which includes the development of multi-dimensional personal resources but also encompasses accessibility to, together with a feeling of personal ownership of social, practical and societal resources, offers the greatest potential for nurturing resilience across different levels, e.g., societal, community and individual. In other words, resilience may be affected by a range of modifiable personal, psychosocial and political factors ^[34][35][36][74,108,109]. Of the four models identified; buffering (two-dimensional), recovery and maintenance, seem to offer the most potential with respect to investigating resilience to suicidality. That the psychosocial factors focused on in these models are adaptable, perhaps, offers pathways for developing psychological approaches and techniques which can maximise and sustain resilience to suicidal experiences.