Work package 2: Qualitative study – grounded theory 2

Parts of this chapter have been reproduced from Causer et al.115 and Spiers et al.116 These are Open Access articles distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) licence, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The text below includes minor additions and formatting changes to the original text.

The study and findings reported in this chapter respond to the following research objectives:

• To explore the impact of colleague suicide on staff well-being and grief reactions to such an event.
• To explore staff views about risk factors which may have contributed to the suicide of their NHS colleague as well as any ‘warning signs’ the individual may have displayed.
• To identify what helps and/or hinders bereaved colleagues to seek support, to characterise supportive work cultures, and to identify staff preferences for future support.
• To explore how managers and other staff tasked with delivering postvention respond to and support their employees and colleagues following a death by suicide, and to identify current postvention activity.

Grounded theory 2: Negotiating the postvention situation: a grounded theory of National Health Service staff experiences when supporting their coworkers following a colleague’s suicide

Supporter participants

We recruited a total of 51 NHS workers, 22 of whom had supported their coworkers following the suicide of a colleague (N = 29). Participants held a variety of posts and had different levels of experience. Further demographic details for participants from both groups can be seen in Table 7.

Data analysis

Data collection and analysis took place simultaneously. Nine interviews were analysed by JS, using classic grounded theory.135 Transcripts were coded for behavioural incidents which illustrated conceptual categories. Constant comparison was used to revise codes so they best represented participants’ core concerns. Memos were used to record analytic thinking about concerns and behaviours. Codes and memos were used to develop the theory.123

Early analysis informed later interviews.135 For example, in response to early codes and memos, JS asked subsequent participants about the capacity their team had to handle their workload at the time of the suicide. After analysing 14 interviews, theoretical sampling was used to recruit more participants.136 Specifically, we sought supporters who did not know the person who died by suicide, as we were interested in exploring whether this influenced their ability to deliver postvention. Additionally, existing participants tended to report delivering good support, so we sought supporters who had experienced challenges in delivering postvention. However, we were unable to recruit participants to meet this criterion.

Analysis continued while the final seven supporter participants were recruited. Once all analysis was completed and theoretical saturation was achieved,122 the theory of ‘Negotiating postvention situations’ was developed (Figure 4).

FIGURE 4 Negotiating the postvention situation. Reproduced from Spiers et al.116 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) licence, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The figure includes minor additions and formatting changes to the original text. (Figure not reproduced here; see the original article.)

Negotiating postvention situations

The model below (see Figure 4) illustrates the theory of ‘Negotiating postvention situations’, demonstrating how the elements which comprise a supporter’s postvention situation inform the behaviours they can choose to engage in, which lead to the delivery of supportive or unsupportive postvention. Delivering postvention has an emotional outcome for supporters. All of this may lead to learning which can improve future postvention, meaning that the negotiation of the postvention situation is a cyclical process. Additionally, supporters themselves need support that may or may not be forthcoming. Being supported, or not, will have an emotional outcome. The green arrows indicate the situation/behaviours/outcomes of the supporters, whereas the orange arrows refer to the situation and behaviours of those supporting the supporters. We present this theory below. Please note that […], used within quotations, indicates editorial elision, often the removal of interviewer interjections.

Fixed workplace elements create supporters’ postvention situations

National Health Service workers who support their colleagues following a coworker’s suicide must do so from a contextual situation comprised of enabling and disabling elements. These elements, which supporters cannot control, create a unique ‘postvention situation’ which defines and/or restricts their ability to deliver support, be supported, and learn.

Disabling elements

Disabling elements make it harder for a supporter to deliver postvention. These might be structural, cultural, personal (such as a supporter’s level of experience), emotional (how the suicide has emotionally impacted the supporter), or staff perceptions (beliefs held by staff that make delivering postvention harder).

While structural and cultural factors as well as staff perceptions are clearly fixed aspects of the postvention situation, personal and emotional elements might be argued to be within a supporter’s control. Individuals are, of course, responsible for their job roles and the years of experience they have acquired. However, when it is discovered that a colleague has died by suicide, supporters cannot decide to have more experience, nor can they switch role. Similarly, supporters may feel affected by the suicide. Thus, all elements of the postvention situation are usually out of a supporter’s control.

Structural disabling elements reported by participants included having no capacity. This could be physical (no space to sit and grieve), staff-related, financial or connected to the structure of the service – for example, those working in 9–5, Monday–Friday well-being teams cannot support staff out-of-hours: ‘everybody can see us and they can check in. But they [busy team] are isolated behind their doors, so there is no good time to capture them’ (N27).

Many participants reported that their entire Trust had a lack of capacity: ‘You need to be able to let people have that breathing space. […] But nobody’s got it because we’re on this treadmill of the contract and the KPIs [key performance indicators]’ (SS04). Additionally, teams of workers constantly shift with the NHS, meaning it can be hard for supporters to reach all colleagues who may have been impacted by a suicide.

Most participants reported a lack of training or postvention guidance. Further, issues of confidentiality restricted supporters. For example, if a family did not consent to share that the death was a suicide, this could hinder postvention.

Supporters may also work in blaming, unrelenting, hostile or static cultures. Stigma about suicide specifically and mental health in general were often reported. Again, all these elements hinder effective postvention.

Personal disabling elements included a lack of experience in individuals:

[W]e don’t have the expertise in place, we can’t rely on one clinical psychologist who may or may not be available.

Disabling emotional elements included supporters’ reactions following a colleague suicide. Shock, sadness, anger and regret were reported. Supporters may also have to juggle their own grief with postvention:

I didn’t wanna go off sick […] I wanted to be there with the team in it but I just also I was just sitting at my desk staring at a screen crying.

If supporters have experienced the death, suicide, or suicide attempt of several colleagues, the compound effect may make postvention harder:

[E]ven though my rational mind was saying you’re not personally responsible, […] I really felt the weight of that. […] And it knocked my confidence.

Additionally, staff’s perceptions of supporters could act as a disabling element. For example, a chaplain reported that NHS workers ‘won’t even look at me’ (SS09) due to (incorrect) preconceptions that chaplains only deliver religious care. Additionally, one well-being lead (SS12) felt that coming to see her might be perceived as stigmatising.

Enabling elements

Enabling elements make it easier for a supporter to deliver effective postvention. These can also be structural, cultural, personal, or emotional factors or staff perceptions.

Enabling structural elements included having enough capacity within a team, helped by adequate staffing. Teams that include chaplains, psychologists, well-being leads and managers have a better capacity for dealing with postvention.

Participants shared that useful guidance or training enabled postvention. While specific postvention training was only mentioned by one, some had access to trauma, well-being, or psychological training/guidance which could be adapted for postvention.

A caring culture also enabled postvention efforts. Care manifested through relationships, communications, meetings, or actions. Additionally, an enabling culture can adapt, for example by shifting from management leadership to compassionate leadership.

Not just putting the corporate political spin out there […] But actually, backing it up and being prepared to have the tough conversations […] the organisation that I work for […] has come on leaps and bounds, and I think if this [suicide] happened now, it would be a completely different story.

Personal enabling elements relate to supporters’ experience and role. Role seniority, expertise in mental health and suicide, or experiential/professional learning derived from having lived through the suicides of other staff or family members were helpful. Various job roles were more suited to delivering postvention. For example, well-being leads and chaplains perceived themselves to be well-positioned for postvention:

[W]e [chaplains] can do the listening, the supporting, […] the creating a memorial service […] and then […] if we do all that and somebody is still struggling, then we signpost them.

Several supporters who did not know the person who died felt they personally did not need support. This also may be an enabling element as it might facilitate some distance.

Staff can also have enabling perceptions of supporters. For example, staff may be more likely to turn to supporters they consider friends, those who make conversations about mental health more acceptable, or those who are familiar:

If you keep showing your face, and people start going oh you know they’re alright, […] even if you’re associated with bringing biscuits.

One well-being lead reported that staff were more likely to come back for further support after an initial individual session, suggesting that building trust and psychological safety were key:

[W]e know once people have used this [well-being] service, they’re much more likely to come back and use it again

Negotiating the postvention situation

Supporters negotiate their unique postvention situation by engaging (or not) in various behaviours. These are leading, caring, educating and mis-stepping.

Leading

Supporters engaged in a variety of leadership behaviours to deliver postvention. Successful communication about suicide required leadership. The need to deliver the news quickly and in an emotionally congruent fashion was reported:

I said ‘you know what I’m like, usually I’m all about happy and everything but right now there’s no way I can dress this up. What I’m gonna tell you is shit. […] You’re gonna be sad, I’m feeling really sad, but I’ve got to tell you now’.

Supporters need to acknowledge the death rather than remain silent about it. Additionally, they sometimes need to control the news by ensuring it is rolled out in the best way possible, such as by highlighting their strong social media policy.

The ability to take charge and act immediately was beneficial. Supporters with seniority or mental health expertise felt able to take charge, switch into ‘professional mode’ and thus offer psychological safety to colleagues in the uncertain aftermath of suicide. However, this might make supporters more vulnerable to harm.

I had to do all that sort of manager mode if you know what I mean. […] Had to do the process and then think of the staff. […] at that time […] you can’t always think of your own feelings, that comes later […] You have to keep face and do.

Some supporters engaged in ‘carrying on’, that is, helping the team or themselves to get back to ‘business as usual’, a perceived necessity in high-demand, patient-facing work.

Caring

Supporters who engaged in caring behaviour employed emotional labour, offering empathic and compassionate responses. Examples included taking affected colleagues out for lunch, making efforts to contact all affected colleagues (including those who had moved on to a new team), and legitimising colleagues’ feelings. Supporters who are aware of their staff’s individual needs (an enabling element) could tailor postvention:

…we did go and [break the news to] another colleague just before that meeting. And this is a lady whose [family member] very tragically took her life […] she said ‘I won’t come into that meeting now thank you for telling me individually’.

Several supporters reported on activities that contained colleagues’ difficult emotions by offering psychological safety or enabled emotional authenticity. Activities included offering reflective spaces for chatting ‘organically and naturally’ (SS06) about feelings. Offering availability by being vocal about the support on offer was also beneficial.

Some supporters engaged in self-care. Reported examples included self-reflection, exercise, asking for support and establishing boundaries.

Educating

Several supporters used psychoeducation to teach staff about suicide or signs of mental ill health. Some fought stigma by talking about suicide or acknowledging the death without judgement. Others pushed for changes to hospital policies and procedures:

I went to the director of people within the NHS […] And then I went to [chief nursing officer] […] And I said we need to do something about this

Mis-stepping

The disabling elements that supporters negotiate in their postvention situations create barriers to providing good enough postvention. For instance, silence around a suicide (perhaps underpinned by stigma) resulted in an absence of support:

…so it was never sort of mentioned to myself as wellbeing lead, that […] [name] had ended his life through suicide. […] and so no formal processes were put in place. And so we didn’t go out and make sure that we were visible.

Participants reported support plans that had to be abandoned due to red tape or a lack of capacity. Other mis-steps might be linked to not knowing what to do:

[I]t was just left there and no one really knew how to deal with it. So you know, cause no one had an idea it wasn’t dealt with

In contrast, doing too much could also be a mis-step:

Often there’s a bit of a race to go out and firefighting, do some early support. And the worst thing people – is to have loads people going in.

The helpful and unhelpful behaviours that are engaged in as a response to the postvention situation create a range of ‘postvention moments’. These are individual instances of the postvention that are delivered to, and experienced by, affected staff.

Postvention moments

Our data suggest that support must be sustained to be effective. Participant accounts evidenced that it must be on offer promptly, and for an extended period, so affected colleagues can access it when they need it. However, postvention is made up of individual moments that staff might interpret as either supportive or unsupportive. We theorise that a supporter might deliver moments that are experienced as supportive by some staff and unsupportive by others.

Supportive postvention moments

While postvention support must be adaptive to respond to complex needs and contexts following a suicide, postvention tools can be simple. Talking and listening were frequently cited. Affected colleagues need to be able to gather in a group so that they can talk and share stories about the person who died by suicide; supporters need to listen, either to groups of staff or in one-to-one sessions.

One chaplain described how a memorial activity enabled colleagues to talk about the person who died by suicide:

That seemed to be quite cathartic. Everyone then talked about what have you done [to memorialise the person who died by suicide]? Oh I remember him doing that. What have you done? Oh yeah, I’ve done that.

The power of talking was demonstrated by the many participants who commented during or after the interview that it was good to talk about the person who had died:

[I]t was nice talking about [name] actually, I can picture her again.

While impacted colleagues need to come together to talk, supporters perceived their job was to listen: to be able to sit, non-judgementally, with another’s grief, whether in group or individual settings. For example:

Int: What was it that you were providing during that time?

SS01: A listening ear more than anything. A reflective, responsive, supportive presence I think I would say.

Togetherness was a key aspect of postvention. Supporters who had the resources provided time for colleagues to be together in a group safe space. Supportive postvention also included practicalities such as time off work and shifts being covered.

Effective postvention moments also had an emotional element. Supporters emphasised the importance of congruency: being present with the difficult emotions suicide creates. Additionally, supportive postvention might trigger a release of emotion. Affected colleagues may need to be able to cry or express frustration, anger, or shock:

[W]hen we hear shocking news, it’s almost like we go ‘Oh!’ and hold our breath. So, for me, the debrief is just about starting to breathe out.

Supporters also facilitated the emotional process of making sense of the death by sharing professional knowledge about suicide.

Remembrance is another key aspect of supportive postvention. This involves recalling the person who died by suicide through stories or memorialising them through – for example – gardens or donations. Participants gave examples of supporting colleagues to attend funerals or facilitating memorial services.

One participant shared how supportive postvention also operated as prevention by empowering staff members to voice suicidal thoughts and access support:

[A] staff member had killed themselves and then others […] were suicidal, so I then contacted their line manager to say ‘you have a team that is really struggling’ and then we got psychological services and wellbeing involved.

Suggestions for this preventative element of postvention included asking staff if they were having suicidal thoughts, finding out new recruits’ mental health needs, and access to more resources. This final suggestion demonstrates that having more enabling elements (e.g. greater resources) in a postvention situation may lead to better postvention, thus saving lives.

Unsupportive postvention moments

If supporters are unavailable, postvention will be ineffective. Supporters might be unable to help due to – for example – a lack of capacity. Unsupportive postvention moments might also include a lack of emotional congruency, silence or uncertainty about the suicide:

[T]he implication is that he died by suicide but no one […] has ever confirmed or denied that and no one’s been allowed to talk about that.

Remembrance which colleagues do not feel sufficiently represents the person who died or their feelings about the death could also be seen to be damaging. One chaplain said:

People need to feel safe; they need to know that somebody’s in charge […] And you can feel the difference, if you get somebody leading a service who can’t do that, you will feel the anxiety in the congregation.

Support must be sustained and available over a short, medium and longer time frame. Delays are not supportive. While some participants were unable to support staff immediately after the suicide for structural reasons (e.g. the suicide happening at a weekend), one was unable to access talking therapy for herself for 6 months:

I got referred to [company] through work but they said they can’t assist with grief for six months, post […] death.

By the time she could access this support, she no longer felt she needed it.

Outcomes of being (un)able to offer support

Delivering postvention had an emotional outcome for supporters. While some outcomes only appeared to come from delivering supportive postvention and others from unsupportive postvention, there was also some crossover.

Delivering supportive postvention led to satisfaction for some supporters and appreciative feedback from staff. For example, when one participant ran a postvention session for their colleagues, they said staff ‘grabbed it with both hands’ (N08). However, the emotional costs of offering postvention were more challenging. These included emotional exhaustion, anxiety, sadness and isolation:

Everyone thinks that I’m very kind of calm and get on with things. But they don’t really know about the stuff inside.

Delivering supportive postvention can also lead to trauma and burnout, or growth and learning:

[I]f anything it’s given me a kind of strength […] there’s kind of a spiritual component to my response to it.

Supporters were operating from a fixed postvention situation which, at times, meant they could not provide the postvention or culture they would like. Yet they carried the frustration, anger, professional doubt and moral injury that arose from being unable to deliver supportive postvention or make a long-term difference following the death. Here are two examples:

[M]y own sense of, you know, am I actually up to the job (scoffs). You know, am I the right person to be doing this.

[W]e’d sort of (pause) promised ourselves and promised [the deceased’s] family we would make a difference. And (pause) and it didn’t feel like we really had much to show for it.

These outcomes demonstrate the need for supporters themselves to be supported. Delivering supportive postvention is complex and challenging; it is not its own reward.

Being (un)supported as a supporter

Supporters shared that, like staff, they needed support when a coworker died by suicide. For a more in-depth look at the impact of suicide and the support needs of impacted colleagues, read our sister paper.115 However, these NHS workers also need support in their supportive roles.

Support for supporters might come from managers and supervisors. Importantly, those who are being asked to offer compassion and empathy to the supporters should be a good fit for the role. Having supportive and reliable teammates is also beneficial:

[T]here was me and the senior nurse, I think and we did it [offering support] together.

Supported supporters felt held and accepted, expressed admiration for their colleagues, and moved towards healing.

The provision of support for supporters can be undermined by the narrative that ‘carers don’t need care’ within organisational cultures.137 Also, supporters can only be supported if there is capacity. One participant explained that they were often the only chaplain on shift, so if something difficult happened, they were unsupported.

Reported features of poor postvention for supporters included blame and an absence of support:

I reached out to everybody and offered it to them but nobody actually offered it to me.

Some supporters cited difficulty in accepting support, which might have come from the internalisation of the ‘carers don’t need care’ narrative or perceptions that others were not interested in helping. A lack of support for supporters led to upset, anger, a sense of abandonment and burnout.

I think that’s part of the reason I ended up burning out. Because a lot of it I just thought I’ll just say I’m fine cause that’s easier than the reality because I didn’t feel they were interested anyway.

Some disillusioned supporters felt empowered to speak truth to power, informing managers that the provision of support was inadequate and needed improvement.

Learning

All postvention experiences could lead to learning, helping supporters to provide better postvention, in the present and future. Learning can take several forms. Supporters in this study reported more postvention and professional skills as well as increased awareness of mental health issues and their own boundaries and needs.

Teams or Trusts might also learn from postvention. For example, gaps in service might be spotted. One supporter found that their team became more open to talking about mental health.

… it also probably made our […] hospital executive team sign up to the idea that wellbeing is important. […] It is really difficult to make it properly meaningful in a massive organisation. But there definitely was a willingness rather than it not being important and I think [the deceased] contributed to that.

Any formal learning or evaluation activities must be carried out with care, involving those who were close to the situation. One participant related that a factually inaccurate report about a suicide in which they had been a supporter damaged their mental health.

… this report had been written without any consultation with myself or anybody […] who knew [the person who died by suicide] […] I felt it was worded in such a way it felt like it was trying to apportion some blame? […] which really is very disturbing and making [me] very very unhappy.

When done well, learning can change policy. Examples from participants included a better approach to complaints about healthcare professionals, time off for funerals, and more well-being resources in general. Several participants who had delivered postvention without guidance had contributed to writing their own:

[My colleague] and I developed the post-incident psychological support service, which is for […] all staff following any sort of traumatic incident, but generally that’s suicide, death of a colleague.

Some participants also reported that opportunities to learn had been missed, perhaps due to stigma or a lack of capacity, demonstrating how disabling elements of a postvention situation can impact postvention delivery all the way to the final stage of learning.

When I asked for a suicide plan to be put onto the agenda, the response was ‘we’ll talk about it but I don’t want the words suicide plan put on the agenda. We’ll call it any other business’. And I found that bizarre, because why are we not talking about this?

Discussion

National Health Service workers who deliver postvention following a colleague suicide are operating within a combination of enabling and disabling elements (their postvention situation), which makes their ability to deliver effective postvention, via various behaviours, easier or harder. Whether postvention is supportive or not, its delivery has an emotional impact on supporters which is often damaging yet can also result in satisfaction or growth. All postvention can lead to learning for individuals and employers.

Our theory of negotiating postvention situations demonstrates that, like suicide 60, postvention does not happen in a vacuum. While each suicide will depend on individual factors impacting the person who dies by suicide, structural contributors must be included when considering prevention 60. We argue that the same is true for postvention. Since effective postvention can act as prevention,50 all potentially relevant elements must be accounted for when delivering postvention support. Critical suicidologist White59 has argued that, for example, rather than suicidal queer people having to re-assess heterosexist violence, violence itself should be addressed. Similarly, we suggest that rather than those who struggle to offer supportive postvention in a disabling situation having to reassess their own worth or abilities, the systems themselves need to change. This systemic approach to postvention is novel since critical suicidologists have, thus far, only critiqued the pathologisation and individualisation of suicide itself rather than postvention.59,60

Given the above, a key question is how the elements of a postvention situation can be facilitated to ensure they are enabling. Supporters must have the capacity to offer postvention and be supported themselves. This is a challenge in the NHS, where many professionals, including GPs,7 junior doctors138 and nurses139 are leaving the profession due to poor working conditions. We found that supporters who worked in an ‘enabling role’ (i.e. well-being leads and chaplains) tended to have more resources for postvention than managers. Therefore, we recommend utilising workers in these roles when delivering postvention. Further, a lack of training or guidance inhibits the delivery of postvention. Some guidance in this area now exists53,140 and evidence suggests that training is beneficial.141 Hence, we propose that evidence-based training and guidelines, including information on how to lead, care and educate when supporting affected colleagues, be offered to NHS workers who may be expected to deliver postvention.

Our findings demonstrate the influence that personal elements such as experience and expertise can have on postvention. Enabling personal elements might include being in a well-being/chaplaincy role, seniority/years of job experience, mental health expertise and professional or personal experience of suicide. Given the variation in this list, postvention should ideally be delivered by a trained team of NHS workers, rather than solitary professionals.1,33,86 The utilisation of such a team may also enable trust in those staff being supported following a suicide and mean that staff with different support needs can approach different people for support.

Stigma leads to inadequate postvention delivery.33 It is harder for supporters to deliver postvention if they feel unable to discuss suicide at work. Many authors reporting on colleague suicide have stated that the death was not properly acknowledged,57,75,76 including in healthcare settings.77,86 Given that talking about what happened and being together to share grief and memories appear to lead to effective postvention, we encourage a culture in which mental health and suicide can be discussed, as this helps to challenge stigma140–142 and provide better support.

Additionally, we found that their emotional response to the suicide may impact NHS workers’ ability to deliver postvention. Quantitative researchers may wish to explore any links between the level of grief a supporter is feeling and their ability to support others.

Remembrance is another key aspect of postvention. Various authors suggested caution here, with some advising against the erection of permanent memorials to those who have died by suicide.1,86,140 However, we have found no evidence to underpin this advice. Given the positive feedback that many of the supporters in this study had from conducting memorialising activities – including creating memorial items – we suggest that this area needs further investigation.

Supporters also talked about the need to ‘carry on’ their team’s patient-facing work despite bereavement. This narrative is potentially harmful33 yet manager supporters are often in an impossible situation where patient care delivery must continue despite grief. Further research is needed to address this thorny issue.

Several supporter participants either reported that they could not access talking therapy right away or felt concerned that support which was offered too soon would not be effective. While we searched the literature for confirmation that talking therapies following bereavement should not be offered too quickly, we could not find any evidence to support this idea, although it may have grown out of the suggestion that one-off debrief sessions following trauma have the potential for long-term negative outcomes.143 We suggest that effective postvention – which differs from bereavement counselling – needs to be delivered in a timely, sustained manner, so that as many sessions of support are available for those who need it, when they need it.

Given that many NHS workers will be delivering support from a disabling postvention situation, it is unsurprising that many will mis-step, offering inadequate or non-existent support. Our sister paper on the experiences of NHS staff who have lost a colleague to suicide illustrates this.115 Given that NHS supporters are working from a postvention situation that is largely out of their control, we propose that better resourcing and education are more useful responses to a lack of support than blame.

Delivering postvention in a work setting, including the NHS, has an emotional impact.33 Therefore, guidance and training need to include information about how to care for the carers, as well as impacted staff.

Finally, given that all postvention may lead to learning, we propose it should always be evaluated and the results used to improve future support,61 something which currently does not happen often.33

Strengths and limitations

This is the first study to investigate the experiences of, and impact on, NHS workers who offer postvention to colleagues. It also presents the first theorisation of enabling and disabling elements and contexts in the delivery of postvention support. Given that NHS workers are vulnerable to suicide,144 we suggest that this is important work and that our novel use of a critical suicidology lens60 is necessary since supporters are operating from a postvention situation they cannot control.

Notwithstanding these strengths, this study has limitations. Firstly, despite our efforts to connect with NHS workers from ethnically diverse populations, our sample was almost exclusively White British. Given that Black and minority ethnic people comprise almost a quarter of that workforce,120 this means our findings are unlikely to represent the experiences of all staff members. We suggest that future researchers engage with community gatekeepers to mitigate similar problems.145 Secondly, most participants in this study were delivering supportive postvention. Although we proactively sought participants who were delivering less optimal support, we could not recruit these people. Thus, the paper reflects the views and experiences of those providing supportive postvention and may give the impression that NHS workers are more fully supported following the suicide of a colleague than in reality. We direct readers to our sister paper,115 in which many colleagues impacted by suicide describe the impact of a considerable lack of support. However, we believe that the views and experiences of the participants in this study can provide informative learning opportunities for Trusts who wish to deliver effective postvention.

Work package 2: Findings pertaining to objective 3: to explore staff views about risk factors, which may have contributed to the suicide of their National Health Service colleague and ‘warning signs’ the individual may have displayed

Data that pertained to the following interview questions were extracted from 47 interview transcripts (these being a subset of the 51 interviews undertaken in WP2. The remaining transcripts did not contain any data pertaining to these particular interview questions): ‘Were you aware of any warning signs prior to [name]’s death?’ and ‘Was there any sense that [name] was under any particular pressure or struggling in any way?’, or to the follow-up prompts: ‘Had anyone mentioned any concerns about [name]?’, ‘Were you aware if [name] sought any support leading up to their death?’, ‘Did they tell you they were struggling to cope?’. The extracted data related to 37 NHS staff who had died by suicide, of whom n = 19 were male, n = 18 were female.

Data analysis

In addition to the grounded theory analyses undertaken (reported in Chapter 3), we undertook an additional analysis of this subset of data. Data were analysed following a critical, interpretive approach,146 in order to examine the social context in which participant experiences and perceptions were situated. Analysis followed the six phases set out by Braun and Clarke.147 Initial semantic and latent codes146 were generated following a deep reading of the data. Themes were developed, reviewed, defined and named. Theme names were developed to reflect the ‘patterns of shared meaning’148 held by participants, and as such call on participants’ phraseology as a means of ensuring their presence in the reporting of findings. In six themes, we present our interpretation of the stories told by participants about the reasons for, or the causes of, their colleague’s death by suicide.

There were signs

All participants told stories that related to signs, clues, or indicators that they related to their colleague’s death by suicide. In some cases, participants were clear that these ‘signs’ pointed towards the outcome of suicide; in other cases, the ‘signs’ appeared more apparent after the event when the participant was reflecting; for others, the signs made sense only when they spoke with other colleagues and put the pieces of the jigsaw together. Participants spoke about warning signs such as colleagues sharing their struggles, talking about suicide or making threats about suicide, previous suicidal behaviour or attempts, noticeable changes in behaviour or appearance or well-being, taking more sick leave than usual, and in a minority of instances, substance mis/use or illegal behaviour.

They were only human

The person who died was seen by their colleagues as being human, in that, like everyone, they had specific traits, vulnerabilities, or weaknesses which in their case may have led to suicidal behaviour. There were also stories of the efforts that their colleague made to stay engaged with life.

Insurmountable life events

Participants identified the normality of the hurdles and challenges that life presents to all of us, for instance, events such as bereavement, relationship breakdown or significant life changes, or challenging experiences such as experiences of domestic violence, exposure to suicide, or childhood trauma. They told stories of how, for their colleagues, these challenges may have been particularly complex or multifaceted, and how their colleagues may have felt particularly harmed by them, or that their challenges were insurmountable.

Challenged by workplace events

Stories about the work environment, cultures and practices were threaded throughout participants’ accounts and formed a significant part of the narrative surrounding their colleagues’ deaths. Often, critical work events related to a range of experiences were perceived to be out of their colleague’s control, and which their colleague felt challenged by, or there were perceptions of unfairness to such events. Examples included workplace changes, redeployment, retirement, critical workplace incidents, experiences of official proceedings or professional failure and the ongoing impact of COVID-19.

Harmed by workplace cultures

Participants shared how their colleagues experienced struggles due to cultures, working conditions or employment status, which were perceived as unhelpful, harmful, abandoning, or contributing to perceptions of invisibility for their colleagues. Such experiences were perceived to be connected with their colleague’s sense of isolation.

Failed by workplace systems

Further stories of workplace experiences focused on the organisational or systemic processes that participants perceived ought to have supported and ‘saved’ their colleagues but did not. These included examples of missed opportunities to respond compassionately and provide support, poorly executed interventions, and failures to report concerns or protect their colleague from harm, for example, bullying by other colleagues.