Examining Veterans’ Interactions with the UK Social Security System through a Trauma-Informed Lens

Abstract This paper uses the principles of trauma-informed care – safety, collaboration, choice, trustworthiness, and respect – to reflect on the quality of veterans’ treatment within the UK social security system. Drawing upon new data from qualitative longitudinal research with veterans in four geographical locations across England, UK, it explores their experiences within the social security system, highlighting specific issues relating to their interactions with the Work Capability Assessment (WCA) but also the conditionality inherent within the UK benefits system. Overall, it is evident that there is a lack of understanding of the impact of trauma on people’s psychosocial functioning and, as a result, veterans are treated in ways which are variously perceived as disrespectful, unfair or disempowering and in some cases exacerbate existing mental health problems. We propose that the application of trauma-informed care principles to the UK social security system could improve interactions within this system and avoid re-traumatising those experiencing on-going or unresolved trauma. The paradigm of trauma-informed care has been used internationally to examine health, homelessness, prison and childcare services, but ours is the first exploration of its application to the delivery of social security.


Introduction
Originating in the US and subsequently moving to Australia, New Zealand and Canada, among other countries, a growing trauma-informed care movement is seeking to apply what is known about trauma to inform the design of human services (Quadara and Hunter, ). This movement draws on neuro-scientific insights into the long-lasting effects of overwhelming, life-threatening experiences on the structure and functioning of the human brain, as well as insights from Jnl. Soc. Pol. (2022), 51, 1, 96-113 © The Author(s), 2021. Published by Cambridge University Press. This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http:// creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited. doi:10.1017/S0047279420000719 benefits system with the introduction of Universal Credit (UC). Such reforms have attracted significant concerns about the disproportionate impacts on those with complex needs (Whitworth and Griggs, ; Geiger, ; Reeves and Loopstra, ; Scullion, ; Jones, ), with evidence that interactions with the social security system can exacerbate mental ill health (Griffiths and Patterson, ; Dwyer et al., ). These concerns have highlighted the need to 'rebalance' the system towards one that is based on the prioritisation of 'personalised, negotiated packages of support' (Dwyer et al., : ).
Building on these existing calls for more effective and personalised support within the social security system, this paper uses the case of military veterans  to explore how the principles of trauma-informed caresafety, collaboration, choice, trustworthiness and respect (Hopper et al., ; Bowen and Murshid, ; Quadara and Hunter, )could improve interactions within the benefits system. The article proceeds by outlining the methods of the study. This is followed by two findings sections that explore veterans' experiences of social security, with specific reference to () the Work Capability Assessment (WCA); and () the conditionality within the UK benefits system. We find that veterans were treated in ways that were variously perceived as disrespectful, unfair or disempowering, and in some cases exacerbated existing mental health problems. Our paper concludes with a discussion of how the application of trauma-informed care principles could improve the quality of support within the social security system and avoid re-traumatising those experiencing ongoing or unresolved trauma. Although our analysis focuses on the experiences of veterans, these principles would benefit other social security claimants who have experienced trauma in their lives.

Methods
Our analysis draws upon the first substantive research to focus on veterans within the UK social security system (Scullion et al., ; ). The research involved two main methods: () two waves of repeat qualitative longitudinal interviews with veterans; and () interviews and focus groups with policy and practice stakeholders. Information on each is provided briefly below.

Qualitative longitudinal research (QLR) with veterans
QLR is a methodological approach that moves away from providing a 'snapshot' of experiences, enabling a more nuanced understanding of people's interactions with welfare services over time (Corden and Millar, ). Two waves of interviews were carried out across four locations in England. Purposive non-random sampling techniques (Mason, ) were used to recruit participants with the help of a range of organisations, including Armed Forces charities, other third-sector organisations, local authorities, and housing/       accommodation providers. The inclusion criteria were: identifying as British Armed Forces veterans; living within our specified geographical fieldwork areas; and claiming one of the following UK social security benefits at the time of the first interview: Employment and Support Allowance (ESA), Jobseeker's Allowance (JSA) or Universal Credit (UC). The organisations facilitated the recruitment in two ways: () disseminating a recruitment flyer across their networks; and () speaking directly to veterans who were using their services about the research. Across the project, a total of  veterans were interviewed at Wave A (interviewed between June and November ). While acknowledging debates about a priori sample size determination (Sim et al., ), the initial aim was to recruit - participants, in line with similar studies that sample specific groups within the benefits system (Dwyer et al., ). Towards the end of the fieldwork period, the interviews began highlighting similar issues (Guest et al., ); however, we wanted to give voice to all those who had come forward to participate, so the sample was increased to .
The Wave A interviews provided a comprehensive baseline of people's experiences of the benefits system, but also explored other aspects of participants' transitions to civilian life (e.g. employment, finances, housing, relationships), as well as existing health conditions (both physical and mental), the nature and impacts of these conditions and whether they attributed them to their service in the Armed Forces. Although these are self-reported health conditions, the majority described a formal diagnosis and/or interactions with healthcare professionals in relation to their condition(s).
Research participants were informed at the recruitment stage that the study was longitudinal. This was reiterated at the Wave A interview, and all  participants gave written consent to be re-contacted and additional contact details (e.g. email address, postal address, mobile number) were taken (if not already provided). Invitations to take part in the second interview (Wave B) were sent - months later. A total of  veterans took part in a Wave B interview (between July  and January ). The Wave B interviews explored participants' lives since the first interview, focusing specifically on their social security benefit claims, movements into work and any support received.
The majority of the interviews were face-to-face; however, a smaller number were undertaken via telephone or Skype, particularly during Wave B, to provide flexibility for those who had relocated or moved into employment. In a small number of cases, interviews were carried out jointly with spouses who were undertaking a caring role. All participants (including spouses in joint interviews) received a £ shopping voucher, at each wave of interview, as a thank you for their time.
Given our focus on trauma-informed care, this article draws upon analysis of the accounts of veterans with self-reported mental health impairments that they attributed to service in the Armed Forces;  participants at Wave A and  at Wave B ( interviews in total). Of the  participants, the majority were male (one female veteran). The majority had served in the British Army ( participants), the remainder in the Royal Air Force or Royal Navy. With regards to the social security benefits they were claiming at Wave A,  were claiming ESA;  were claiming UC; and two were claiming JSA. Three people had very recently moved out of the benefits system; two into paid work and one into full time study (however, this person subsequently re-engaged with the benefits system as an ESA claimant by Wave B).
Interviews/focus groups with policy and practice stakeholders A range of policy and practice stakeholders were also consulted. Firstly,  interviews were carried out with stakeholders representing a range of Armed Forces charities and other third sector organisations. Secondly, the research included consultation with  DWP staff through three focus groups (each in a different geographical area). The policy and practice stakeholder consultation focused on their perceptions of the issues veterans face when navigating the social security benefits system, and how they approach providing support to veterans and their families.
All interviews and focus groups were audio recorded, with permission, and transcribed verbatim. Transcripts were analysed thematically, using a QSR NVivo software package to aid storage and retrieval of data. Ethical approval was granted by the University of Salford School of Health & Society Research Ethics Panel. The sections that follow present the findings of the research focusing specifically on veterans' experiences of two aspects of the social security system: () the Work Capability Assessment; and () their interactions with the conditionality within the system. We then present a discussion of how trauma-informed principles could improve experiences within the social security system.

Re-traumatisation? Veterans' experiences of the Work Capability Assessment
There were significant and complex needs within the sample. The mental health conditions referred to most frequently were PTSD, anxiety and depression. Some had been living with unresolved trauma for many years before reaching a 'crisis' point: I suffer from PTSD : : : For eight years I was coping with it on my own, sort of ignoring it. One day I started crying in front of my little boy, the next minute I'm jumping behind the couch : : : by that time it was too late, there was no stopping it, it was just every day. (ESA claimant, Wave A)

     
Consistent with other studies of veterans both in the UK (Iversen et al., ) and elsewhere (Van Voorhees et al., ), in addition to mental ill health attributed to the Armed Forces, some participants had pre-existing trauma often related to adverse childhood experiences. Participants often described the intermittent nature of their mental ill health, referring to 'good days' and 'bad days': So yesterday I was all like, 'Oh my God! What's this? Can't wait for today [to] end' : : : Just from the moment I opened my eyes I just wanted to go back to bed. But some days, like today, I woke up this morning feeling fresh and, you know, like good day (ESA claimant, Wave A) Mental illness impacted significantly on people's post-Service labour market experiences. Although many described being able to find employment relatively quickly after leaving the Armed Forces, subsequent difficulties sustaining employment were common and resulted in social security benefit claims. Given that their benefits claims were due to ill health, many participants had therefore been required to undergo a Work Capability Assessment (WCA), which assesses how a person's health condition/disability affects their ability to complete a range of functional activities. Within ESA, claimants are classified as either 'fit for work' (i.e. with the expectation of engagement in a full range of work-related activities); having 'limited capability for work' but deemed likely to become capable of work in the future (and placed within the Work Related Activity Group of ESA); or having 'limited capability for work and limited capability for work-related activity' (and placed within the Support Group of ESA). Although ESA is being phased out and replaced by UC, the WCA remains the process of determining people's ability to work. For the veteran participants, the WCA was problematic for two key reasons: the inability to appropriately assess mental health issues that were attributed to service in the Armed Forces; and the potential for re-traumatisation through the assessment process.
The WCA has been widely criticised for the inadequate consideration of mental health (Griffiths and Patterson, ; Barr et al., ; Maclean et al., ). Indeed, many veterans reported being assessed as 'fit for work' because they felt their WCA had focused primarily on physical capabilities. However, more significant concerns were raised about whether assessors were qualified to assess Service-related mental health impairments. A number of veterans were being treated by specialist mental health practitioners. Consequently, they struggled to understand how the WCA had assessed them as 'fit for work' despite these professionals having certified them as unable to work due to the severity of their mental health issues. In some extreme cases, it was alleged that assessors had made inappropriate comments, suggesting that they were ignorant of the psychosocial impacts of trauma, but also revealing a lack of trust in the accounts of veterans: Respondent: the people that do the assessments, they're not medically trained people. I remember I had one particular incident when I went along, and I said to the guy, 'Look, if I start to feel unwell or if we need to stop, can we stop the interview and can I walk out for five minutes?' The guy says, 'No'. He says, 'I've got  minutes. It's got to be done in  minutes'. He said, 'To be honest, all you veterans that say you've got PTSD and everything, it's just a crock of s * * * '. (ESA claimant, Wave A) Like many other benefit claimants (Garthwaite, ), veterans described the WCA as something to be feared. For some respondents, the WCA created a level of anxiety that undermined the progress made in the management of their on-going mental ill health. One veteran described the devastating consequences of this: I had a letter come through the letterbox : : : wanted me to go in for an assessment : : : I rang them up and I say, 'I'm unfit to travel to an assessment', and they said to me, 'No, but you've got to come in for an assessment : : : You've got to provide evidence that you've got PTSD'. I said, 'Doesn't my War Pension evidence count?' He says, 'No, because you're claiming for a different benefit'. Unfortunately, I put the phone down, and my anxiety levels were so high I tried popping a couple of diazepam and that wouldn't work : : : I took a serrated knife to my arm : : : A couple of days later I had another phone call : : : same sort of rigmarole, 'We're waiting for evidence' : : : Unfortunately, I put the phone down and hacked at my left arm, my right arm. Same situation again (ESA claimant, Wave B) Another respondent described a similar deterioration in mental health following the WCA. He described having PTSD, which had led to an earlier suicide attempt. He was currently being supported by a number of Armed Forces charities, as well as receiving counselling. He had been advised to apply for ESA on the basis of his mental health impairment but had been assessed as 'fit for work'. He appealed the outcome and at his Wave B interview he reported he had subsequently been placed in the WRAG, where there is still an expectation to prepare for work. Reflecting on his experience of the WCA, he stated: The [benefits] system has made me five times as worse as when I first went to the doctor for help (ESA claimant, Wave B) The harmful effect of the WCA was not just felt by the individual claimant, but also by the wider family. One striking example was described during a joint interview with a veteran and his wife. They explained how he would usually ensure that he was accompanied to his appointments/assessments but had to go alone to his most recent one. His wife described the impact that the 'stressful environment' of the WCA had on her husband's mental health, but also on their family when he had returned home: Veteran: So, two days before [the WCA], they were insistent that I turned up. [Spouse] was not able to go with me, because of the kids. I drove myself across. They asked me in that interview five times how I managed to get myself to [the assessment]. I said, well, I drove. I did drive. I present       very well : : : It ended up turning in to a complete fiasco. They kept me waiting for two hours : : : which I think was deliberate Spouse: : : : just comprehend the level of risk and threat that come along with that, for the family. So, the [benefits assessors] have no concerns in having [him] waiting in an incredibly stressful environment, with somebody that wasn't qualified to manage him : : : This is somebody that's heavily medicated for a serious mental health condition, and that had repercussions within my family unit. When he came home. His behaviour does sometimes become unmanageable. We have come very close to [him] having to be sectioned : : : I'm not asking for special treatment. I'm just asking for somebody to think, 'Is this the most appropriate course of action with this person, and what are the possible repercussions and ramifications for that?' (ESA claimant and spouse, Wave B) Some veterans' anxiety also manifested as frustration or even, on occasion, anger or aggression towards those involved in the benefits assessment process. For example, some talked about 'getting into arguments' during face-to-face appointments or over the phone. One veteran described his reaction to being denied the opportunity to be assessed at home: I got a letter from my GP stating exactly that [I was unfit to travel]. I sent that off to them, and they said, 'You've got to attend.' I said, 'That's a load of bullshit.' I had an argument with the person on the phone. He said, 'We don't send assessors out to your home address' (ESA claimant, Wave A).
Many concerns raised by the veterans were also reiterated by practitioner respondents who had accompanied veterans to assessments and provided support with subsequent appeals. Overall, practitioners felt that assessors often had limited understanding of Service-related conditions (mental or physical) or the impact of trauma. Indeed, some practitioners stated that a significant proportion of their working week was spent in tribunals representing veterans who they felt had been inappropriately assessed as 'fit for work'.
Trauma-blind? Conditionality, the Claimant Commitment and interactions with Jobcentre staff So far, we have described the difficulties faced by veterans in navigating the assessment process that determines the type and level of benefit they will receive. Now we focus on their experiences as they navigate the rules and responsibilities associated with their benefit claims. Although a significant proportion of the sample were ESA claimants, there were some veterans who were claiming UC or JSA and were subject to varying degrees of conditionality, despite their on-going mental ill health. As highlighted earlier, much has been written about the ineffectiveness of conditionality for people with disabilities or health conditions (Patrick, ; Weston, ; Lindsay and Houston, ; Pickles et al., ; Dwyer, ), with suggestions that it can actually trigger, or exacerbate, mental ill health (Dwyer et al., ).
It was evident that many veterans experienced difficulties meeting the conditions of their claim. When discussing what was expected of them, people referred to their Claimant Commitment, which sets out their obligations and what will happen if they fail to comply (i.e. the possibility of being sanctioned). The Claimant Commitment is produced during the initial interview with the Jobcentre Plus (JCP) Work Coach, with work-related requirements supposedly tailored to an individual's circumstances (DWP, ). However, it has been described as a 'paternalist tool' that forms one of the 'strategies of surveillance' of claimants (Fletcher and Wright, : ). Many participants felt they had no control over the content of their Claimant Commitment; rather, they viewed it as something that they had to agree to, or they would lose financial support. Frustration over this lack of control sometimes led to disagreements with Work Coaches during JCP appointments. For example, one veteran described a dispute that required the advocacy of a third sector organisation: We had a bit of a disagreement over it : : : I was sat there with the support worker [from NGO] because they [Work Coach] were telling me that I had to job search in order to receive the benefits. I had to turn up to their appointments every two weeks : : : and some other things. I didn't agree with any of them because [of] the mental state that I was going through, I already told her that I'd struggled going there, so why was I going to go there and sit on day courses and stuff like that, when I'm trying to sort myself out : : : why would I be looking for work, when I know for a fact that I've got severe anxiety issues, severe PTSD (UC claimant, Wave A) Similar to the distrust of the WCA process above, such interactions demonstrated limited understanding or consideration of the impact of trauma, and for veterans elicited a lack of trust in those tasked with administering social security benefits. This mistrust was particularly acute where people had experienced inconsistency in their interactions. The interviews were carried out in different locations across England and it was evident that there was significant variation both across and within JCP, in how staff responded to the disclosure of Armed Forces attributed mental health issues. For example, one veteran's initial positive experience with his UC Work Coach (at Wave A) contrasted with his experience of being allocated a new Work Coach (at Wave B): Veteran: The fella I was dealing with is really good actually (Wave A) Veteran: it was quite brutal actually : : : we were talking about ex-Forces : : : and then that's where she said, 'Well, surely to God you know what you signed up for?' She went, 'This is what I don't understand, you lads all know what you're signing up for.' (Wave B) These interactions were also narrated as displaying a lack of respect towards their Armed Forces background, with perceptions that some staff lacked       empathy in relation to their Service-attributed mental ill health and how this can impact on their ability to navigate the benefits system: I chose a different career path to you. It's impacted on me, so just be respectful of that : : : I'm not asking to bow down to me or anything like that, but just general respect for people : : : I haven't chosen to be on benefits, it's just how things have happened with my mental health (ESA claimant, Wave B) Although the majority of veterans did not want, or expect, a separate system to be in place for them, it was evident that a 'one size fits all' approach was not appropriate to address their needs: Everybody is just treated the same down at the Jobcentre do you know what I mean? : : : But when you're ex-Army your life is structured : : : So, to go from that to then when you come out of the Army and you have to go to sign on you might be anxious, do you know what I mean? Because I know I am : : : especially around strangers and stuff. I hate it : : : As soon as they know that you're ex-forces, personally I think they should enquire about are you all right in this environment or would you like to be in a room on your own, do you know what I mean? So, we've got none of this hustle and bustle and stuff like that and if you prefer one-to-one rather than sitting with the other  people in there waiting for your name to be called out (ESA claimant, Wave A) However, it is unfair to suggest that all interactions with the benefits system were negative, and there was evidence of significant good practice, often in geographical areas where staff were more regularly engaging with Service/ex-Service personnel. For example, one participant described the flexible approach of his Work Coach, where telephone appointments were offered to those experiencing periods of mental ill health: Obviously, with [this Jobcentre] being right next to [a] Garrison, these are more aware : : : The Jobcentre here are understanding of, like, 'Right, fair enough, yes. We'll give you a phone call interview instead, if you want, if you can't come out your flat.' (UC claimant, Wave B) Another example of good practice related to a DWP Armed Forces Champion who acted as the first point of contact for veterans within their district and had taken on the responsibility of bringing together the other relevant statutory and specialist organisations: Very often, [veterans] come in, they've got no idea. They just get told to go to the Jobcentre : : : Basically, you pick up the pieces, trying to put a support network round them. I work very, very closely with [local authority] with [specialist third-sector organisation]. We usually try and tackle everything within that first interview. It's an awful lot to take on. It usually means finding out about their health, their housing, family : : : It could even mean taking the person over to the provider to try and tackle their housing and then trying to get the person from the council to come in as well, and try and put everything in place (DWP focus group respondent) However, such examples appeared to relate to the approaches and discretion of individual workers, and consultation with staff did not suggest that traumainformed care principles were a motivating factor.

Discussion: Towards a trauma-informed social security system
Internationally the experiences of combat veterans have been crucial to the development of effective approaches to supporting trauma survivors (Herman, ), with trauma-informed care employed as a framework for improving veteran's access to healthcare in the US (Kelly et al., ; Currier et al., ; Gerber, ). The issues that veterans with mental ill health experienced in accessing social security in the UK are similar to those described in the US literature on trauma-informed healthcare for veterans (Kelly et al., ; Currier et al., ; Gerber, ), and on trauma-informed human services more broadly (Harris and Fallot, ; Bloom and Farragher, a, b). Indeed, as highlighted previously, trauma-informed approaches are being effectively adopted by a range of human services (Ko et al., ; Hopper et al., ; Muskett, ; Quadara and Hunter, ; Levenson, ; Sweeney et al., ), but the social security system appears to be absent from this movement. In this section we propose that the five key principles of trauma-informed caresafety, trustworthiness, choice, collaboration and respect -provide a useful framework for identifying opportunities for improving interactions with the social security system. We will provide a brief overview of each principle, with reflections on how they relate to the experiences of veterans outlined in the previous findings sections.
The principle of safety is relevant to our discussion and involves ensuring that clients are physically and emotionally safe. From the perspective of traumainformed care, it is the clients' perception of safety that matters, and this is promoted through awareness of clients' triggers, and by respecting privacy and personal boundaries (Kelly et al, ). This is important because people who have experienced trauma can be acutely sensitive to threats in their environment; consequently, small acts of disrespect can sometimes be met with violent outbursts (Bloom and Farragher, b: ). As such, the first step in recovering from trauma is regaining a sense of safety and the approach of trauma-informed services is to try to create predictable environments for clients that are absent environmental stresses such as loud noises and crowding (Gerber, ). The impact of the physical environment of JCP offices/WCA centres was referred to by participants, some of whom expressed feelings of significant anxiety when faced with busy waiting rooms/offices or when having to wait for extended periods of time at these venues. Additionally, the fluctuating nature of people's mental ill health (described by some as 'bad days' and 'good days') meant that people's capabilities varied over time. Although some staff did appear to provide       flexibilityfor example, offering telephone appointments rather than an expectation of coming into the officethis appeared to be the exception rather that the rule.
It was also evident that many veterans experienced difficulties with emotional safety when navigating other aspects of their claim. For example, the potential of having their benefits stoppedeither as a result of 'failing' a WCA or through a benefit sanctionwas another source of anxiety that could undermine their feelings of safety. The experiences presented in the findings sections demonstrate that some people were intensely fearful, and this appeared to be a key trigger for both self-harm but also aggression towards staff. However, it is not surprising that these potential threats to their means of subsistence created high levels of anxiety. Indeed, the provision of an effective social safety net has been identified in the trauma-informed literature as a way of increasing people's sense of safety, and thereby their opportunity for recovery (Bowen and Murshid, : -).
With regards to the principle of trustworthiness, people who have experienced trauma have difficulty trusting, so services need to help them to learn to trust again by being trustworthy themselves; for example, through consistency and transparency. Unfortunately, the accounts of the veterans often demonstrated a lack of trust in a benefits system they found, at times, inconsistent, confusing and hard to navigate. For example, the disparity between the outcomes of the WCA and the judgements made by specialist mental health practitioners were a concern, with participants articulating a significant lack of trust in the skills and qualifications of the benefits assessors. A further source of mistrust was the imposition, via the Claimant Commitment, of expectations that didn't fit their needs and priorities (Currier et al., : ), which made some veterans sceptical about the motivations of Work Coaches. This mistrust was particularly acute where people had experienced inconsistency in the approaches of different Work Coaches.
This also links to the principle of collaborationservices can help trauma survivors rebuild their sense of self-efficacy by doing things with their clients rather than to them. In practice, this means working with clients towards goals they value and believe are achievable (Harris and Fallot, ). Feelings of helplessness and powerlessness are central to traumatic experiences so recovery entails learning to exercise agency again (Hopper et al, ; Levenson, ). As such, trauma-informed services aim to give clients a sense of control and choice over what happens to them. Here, interactions with staff, particularly in relation to the Claimant Commitment and on-going expectations of the claim, were often perceived as perfunctory and disempowering, and veterans described a lack of control over the requirements that were enshrined within their benefit claim. Moving away from a 'one size fits all' approach and allowing claimants to have more input into the Claimant Commitment, but also a greater say over the pace of JCP meetings or the WCA process, as well as the opportunity to take breaks if they felt overwhelmed, would enable people to feel more in control.
Shame is also a powerful part of the experience of trauma, so traumainformed services work to discourage shame and stigma (Harris and Fallot, ; Levenson, ). Existing research shows that such feelings are common amongst benefit claimants (Shildrick et al., ; Patrick, ) and it was evident that veterans were grappling with an intense sense of shame at moving from a position of respect in the Armed Forces to having to ask for financial support. This was amplified when staff appeared to demonstrate a lack of respect towards people's experiences. Treating clients with respect and empathy is a central part of trauma-informed care because people recover from trauma through connection with others (Herman, ). The interviews suggested that some JCP staff and benefits assessors failed to see a connection between veterans' traumatic life histories and their current difficulties in seeking and sustaining employment. Better understanding of the symptoms of psychological trauma would therefore enable staff to understand veterans' difficulties in taking steps towards employment not as moral failure but as a predictable effect of overwhelming, life-threatening experiences (van der Kolk, ).
Referral to appropriate services (e.g. therapeutic treatment) is also a key component of trauma-informed care (Harris and Fallot, ). The interviews revealed that some staff are successfully providing support that meets the needs of veterans by working collaboratively with statutory and third sector organisations (even though they may not be familiar with the terminology of traumainformed care). A key challenge was the inconsistency in this support. There is a need to build capacity of all staff who are interacting with veterans to ensure that those delivering social security are aware of the relevant services and are referring people appropriately. One approach would be to emulate the good practice of those Jobcentres that appeared to be effectively supporting veterans. However, a vital step in making social security interactions more trauma informedeven in those areas where good practice was evidentis through the provision of appropriate staff training on how traumatic experiences can affect individual functioning. This appears to be particularly urgent for those responsible for conducting benefits assessments.

Conclusions
Our analysis suggests that there is much to be gained from the application of trauma-informed approaches to the social security system. Although some JCP and WCA staff have an understanding of mental health issues linked to Service in the Armed Forcesand are consequently able to deliver services which are experienced as safe, supportive, understanding and empowering -this       appears, at present, to be the exception rather than the rule; and there was no evidence that this was driven by trauma-informed care principles. Indeed, overall, the current approach appeared to be 'trauma-blind' (Quadara and Hunter, ). A trauma-informed approach would improve not only the social and emotional well-being of the individual veteran but also family members such as spouses and children. Additionally, there are potential employment gains, as veterans themselves identify unresolved trauma as a barrier to sustaining paid work. Making social security services more trauma-informed is also likely to improve the workplace safety of staff by reducing the level of aggression they may encounter from clients.
However, 'trauma-informed approaches are a process of organisational change' (Sweeney et al., : ) and we need to recognise that there are some significant barriers to making social security more trauma-informed. JCP staff have large caseloads and appointments are short. More broadly, the 'any job' approach to labour market activation pursued in the UK (Taylor, ) would appear to be in tension with addressing the needs of people with unresolved trauma, with many of the veterans interviewed reporting that the pressure to apply for jobs that were poorly matched to their skills contributed to their on-going mental health challenges. Furthermore, the focus in the UK and other modern welfare states on monitoring compliance with behavioural conditions (Mead, ; Dwyer, ; Clasen and Clegg, ; Wright, ; Trenz and Grasso, ) and punishing noncompliance through benefit sanctions (Pavetti, Derr and Hesketh, ; Hofmann, ; Griggs and Evans, ; Dwyer, ) is difficult to reconcile with the principles of traumainformed care.
It is important for us to reflect upon the limitations of the research. With the exception of a study that explores benefit usage through an existing quantitative cohort dataset (Burdett et al., ), there is currently insufficient data available in relation to veterans who are claiming benefits from which to draw a sample frame. As such, further research is required to understand the size of the veteran benefit claimant population. We recognise that our sample includes a significant proportion of veterans who were experiencing Service-attributed mental ill health, and that this may not be reflective of the wider UK veteran population for whom rates of PTSD, for example, are reported as 'surprisingly low' (Hunt et al., ). Furthermore, although multiple access points were used to recruit participants, it is likely that those who came forward to participate were those who had experienced difficulties navigating the social security system and were therefore more likely to articulate negative experiences of their interactions.
Although our paper focuses on the case of veterans, the challenges outlined here are similar to those of other people experiencing mental ill health (see, for example, Dwyer et al., ). Furthermore, service in the Armed Forces is not the only source of potential exposure to trauma. Indeed, it is not the only source of trauma referred to by participants in this study. Child abuse, sexual assault, domestic violence, surgery, accidents and natural disasters are all capable of producing overwhelming experiences, leaving psychological scars from which it can take years, or even decades, to recover (Hopper et al, ; van der Kolk ). As such, future research could potentially explore how social security interactions affect other benefit claimants with exposure to trauma. A further question is whether those JCP locations providing veterans with appropriate support deliver similarly effective support to claimants with other trauma backgrounds.