Trauma-informed mental healthcare in the UK: what is it and how can we further its development?

It is known that many people in contact with mental health services have experienced physical or sexual trauma (Mauritz et al., 2013), that there is a strong link between childhood trauma and adult mental distress (Bentall et al., 2014), and that experiences of marginalisation, poverty, racism and violence are correlated with poor mental health (Paradies, 2006). This has led to a call for services to acknowledge psychological and social factors in the development of extreme mental distress (Read et al., 2009). The hope is that such models would minimise the risk that people presenting to services have their symptoms disconnected from the context of their lives. In this paper, we will describe the concept of trauma-informed approaches (TIAs) which were developed in North America but have relatively few published models from public services across Europe. TIAs are based on the understanding that most people in contact with human services have experienced trauma, and this understanding needs to permeate service relationships and delivery (Harris and Fallot, 2001). We begin by examining the theoretical basis for TIAs including the link between trauma and mental distress and institutional retraumatisation. We will argue for a more systematic transformation of mental health services that acknowledges the role of trauma in people’s lives and consequently reconceptualises relationships between survivors (people who have experienced trauma and mental distress and who may use mental health services) and service providers. Finally, we present a narrative overview of literature on effectiveness of TIAs, map current TIA activity, explore why TIAs have not impacted on mainstream UK practice and discuss what might be needed to bring TIAs to the UK.

Definitions of trauma vary, but broadly, trauma refers to events or circumstances that are experienced as harmful or life-threatening and that have lasting impacts on mental, physical, emotional and/or social well-being (SAMHSA, 2014). Trauma can be a single event or multiple events compounded over time. The concept of trauma encompasses experiences of interpersonal violence, such as rape or domestic violence. Complex childhood and developmental traumas include community violence (e.g. bullying, gang culture, sexual assault, homicide, war), abuse, neglect, abandonment and family separation (Van der Kolk, 2005; www.nctsnet.org/trauma-types/complex-trauma). Lesser understood forms of trauma include social trauma, such as inequality, marginalisation, racism and poverty, and historical trauma, the trauma legacy of violence having been committed against entire groups, including slavery, genocide and the Holocaust (Blanch et al., 2012). Lenore Terr (1991) has conceptualised two basic types of childhood trauma: Type I trauma involves witnessing or experiencing a single event such as a serious accident or rape. Type II trauma results from repeated exposure to extreme external events, such as ongoing sexual abuse.

The Adverse Childhood Experiences (ACE) study investigated the association between childhood trauma and adult health in over 17,000 people (predominantly white, middle class Americans; www.cdc.gov/violenceprevention/acestudy/prevalence.html). Childhood trauma was common: 30 per cent of respondents reported substance use in their household; 27 per cent reported physical abuse; 25 per cent reported sexual abuse; 13 per cent reported emotional abuse; 17 per cent reported emotional neglect; 9 per cent reported physical neglect; and 14 per cent reported seeing their mother treated violently (www.cdc.gov/violenceprevention/acestudy/prevalence.html).

Research has demonstrated that people in contact with the mental health system have experienced higher rates of interpersonal violence than the general population. A systematic review estimated that half of those in the mental health system had experienced physical abuse (range 25-72 per cent) and more than one-third had experienced sexual abuse (range 24-49 per cent) in childhood or adulthood, significantly higher than in the general population (Mauritz et al., 2013). Similarly, survey research has found that people using mental health services are substantially more likely to have experienced domestic and sexual violence in the previous year compared to the general population (27 per cent of women and 13 per cent of men had experienced domestic violence compared to 9 and 5 per cent, respectively, of the general population; 10 per cent of women had experienced sexual violence compared to 2 per cent of the general population; Khalifeh et al., 2015).

Over the last decade, research evidence has increasingly supported the notion that trauma is linked to adult psychosis and a wide range of other forms of mental distress (e.g. Bentall et al., 2014; Fisher et al., 2010; Kessler et al., 2010; Paradies, 2006; Varese et al., 2012). The ACE study found that the more adverse life events people experience prior to the age of 18, the greater the impact on health and well-being over the lifespan, including poor mental health, severe physical health problems, sexual and reproductive health issues, engaging in health-risk activities and premature death (Anda et al., 2010). Similarly, Shevlin et al. (2008) found that experiencing two or more trauma types significantly increased the likelihood of experiencing psychosis. Dillon et al. (2012) report evidence of a dose-dependent relationship between the severity, frequency and range of adverse experiences and subsequent impact on mental health. Interestingly, research has also demonstrated that the general public share the notion that trauma and adverse life events play a causal role in mental health difficulties (e.g. Read et al., 2013; Angermeyer and Dietrich, 2006).

Contemporary neuroscience is exploring the link between childhood trauma and neurological development. This research is informing TIAs which typically adopt a whole systems view of people and their environments, including an understanding of the role and impact of neurological damage. For instance, research has demonstrated that trauma has an impact on developing brains in childhood which can go on to affect the structure and function of adult brains (Perry 1995, 2005). This has led to the development of a traumagenic neurodevelopmental understanding of the link between childhood adversity and adult psychosis, which now has a large body of supporting evidence (Read et al., 2014). The neurological damage caused by trauma suggests that survivors can be “primed” to respond to current situations that replicate the experience of loss of power, choice, control and safety in ways that may appear extreme, or even abnormal, when a history of past adverse events is not taken into account. However, research has also indicated the healing potential of current relationships (Perry, 2005). To find out more see, for example, Van der Kolk (2005), Read et al. (2014) and Dillon et al. (2012).

Research has also demonstrated that traumatic events are more frequently experienced by people in low-socioeconomic groups and from minority ethnic communities (e.g. Hatch and Dohrenwend, 2007). It has further been argued that poverty is the most powerful predictor of mental distress because it predicts so many other causes (Read, 2010). Moreover, black people are over-represented in the mental health system, are more likely to experience negative or adversarial pathways to care, to be diagnosed with psychotic disorders and to receive compulsory treatment (e.g. Mohan et al., 2006; Morgan et al., 2004). Yet, there is little discussion of the potential role of historical and cultural trauma in this. Indeed, social trauma, including poverty, racism and urbanicity, is so prevalent it is often not recognised as integral to poor mental health by clinicians or those experiencing it.

Notably, people in contact with mental health services who have been sexually or physically abused in childhood typically have longer and more frequent hospital admissions, are prescribed more medication, are more likely to self-harm and are more likely to attempt to kill themselves than people without experiences of childhood abuse (Read et al., 2007).

Retraumatisation essentially means to be traumatised again. It occurs when a person experiences something in the present that is reminiscent of a past traumatic event. This current event or trigger often evokes the same emotional and physiological responses associated with the original event. People are not always aware that their current distress is rooted in past events, nor do all people relive the original event in a logical, coherent manner (Durant, 2011).

The mental health system can retraumatise survivors through its fundamental operating principles of coercion and control (Bloom and Farragher, 2010). Retraumatisation includes overt acts, such as restraining and forcibly medicating a rape victim, as well as less palpable retraumatisation, such as pressure to accept medication which mimics prior experiences of powerlessness. Empirical research indicates that traumatic experiences (e.g. physical assault, seclusion, restraint) are widespread in inpatient settings (Freuh et al., 2005). Mental health services can also contribute to historical and cultural trauma by recasting responses to racism as individual pathology (Jackson, 2003), recasting women’s attempts to resist domestic control as hysteria (St-Amand and LeBlanc, 2013) and recasting homosexuality as sexual deviance in need of corrective treatment (Friedman, 2014).

Jennings believes that whilst retraumatisation can be unintentional and unanticipated, it will remain whilst mental health systems fails to acknowledge the role of trauma in people’s lives and their consequent need for safety, mutuality, collaboration and empowerment (www.theannainstitute.com). Current services and supports that do not take these impacts into account may inadvertently retraumatise, further reinforcing survivors’ needs for coping strategies such as illicit drug use or self-harm.

The policies, procedures and practices that staff may be required to perform in “trauma-organised systems” (Bloom and Farragher, 2010) can conflict with personal and ethical codes of conduct. For example, the use of seclusion and restraint as an institutional practice erodes the very meaning of compassion and care, the primary reasons most staff enter their chosen field. Staff who experience conflicts between job duties and their moral code are under chronic stress for which they must learn to cope and adapt. Those coping strategies may include “shutting off” the ability to empathise, and viewing people receiving services as “other” thereby disqualifying their humanity and basic human rights. Pessimism – rather than enthusiasm and hope – may buffer staff from their own feelings of helplessness (Chambers et al., 2014).

Staff may also engage in “power over” relationships when organisations place a higher priority on risk management than human relationships. A nurse who is required to perform a personal search may become frustrated by a service user’s resistance, failing to recognise that she/he is a stranger who is placing hands on the body of another who may be a rape survivor. Organisational cultures may become corrupted, paving the way to power over relationships that reinforce people’s helplessness and hopelessness. In these “corrupted cultures”, the basic values of the organisation are no longer driving practice; instead, the needs of service users become secondary to the needs of staff, and restraint and coercion may be used widely even when less restrictive options are available. This and other working practices and routines (such as rigid professional hierarchies and a lack of supervision) can dehumanise both staff and service users and lead to human rights violations (for an account of corrupted cultures and the impact on coercion see, Paterson et al., 2013; Wardhaugh and Wilding, 1993). The National Institute for Clinical Excellence (NICE) has expressed frustration at first resort to coercive practices even where other approaches are indicated (NICE, 2005). The impact of trauma-organised services on workers is analogous to the impact of trauma on survivors – it reshapes and re-constructs self-identity and can shatter individual meaning and purpose (Knight, 2015).

The development of TIA can be traced to the USA and to Harris and Fallot’s (2001) seminal text, Using Trauma Theory to Design Service Systems. Bloom (2013), also from the USA, who developed the Sanctuary Model outlines the development of TIA from the era of moral treatment, through social psychiatry and finally the concept of the Therapeutic Community (Bloom and Norton, 2004) which includes developments in the UK. TIAs can be defined as “a system development model that is grounded in and directed by a complete understanding of how trauma exposure affects service user’s neurological, biological, psychological and social development” (Paterson, 2014). Consequently, TIAs are informed by neuroscience, psychology and social science as well as attachment and trauma theories, and give central prominence to the complex and pervasive impact trauma has on a person’s worldview and interrelationships.

TIAs are applicable to all human services, including physical health, education and schools, forensic, housing and social care (Schachter et al., 2008; Havig, 2008; Cole et al., 2013). In a trauma-informed service, it is assumed that people have experienced trauma and may consequently find it difficult to develop trusting relationships with providers and feel safe within services. Accordingly, services are structured, organised and delivered in ways that engender safety and trust and do not retraumatise. Thus, trauma-informed services can be distinguished from trauma-specific services which aim to treat the impacts of trauma using specific therapies and other approaches. The key principles underlying TIAs can be found in Table I, adapted from SAMHSA (2014), Elliot et al. (2005) and Bloom (2006).

Whilst it may seem that principles such as safety and collaboration define any good service for any service user, Elliot et al. (2005) have argued that if these principles are not adhered to, trauma survivors may be unable to use services. It is striking that these general principles have strong resonance with the values that psychiatric survivors have historically called for, and underpin much peer support practice (e.g. Mead and MacNeil, 2006).

The potential benefits of TIAs to survivors are myriad, including hope, empowerment, support that does not retraumatise and access to trauma-specific services. Moreover, the medicalisation of human suffering has created a divide between people receiving services and those offering support; this divide can create tenuous bonds that are inadequate, at times, to protect the human and civil rights of people viewed as other (Filson and Mead, 2016). But trauma is something that many of us experience, and indeed, a small number of studies suggest that workers in human services have a high prevalence of ACE scores (e.g. Esaki and Larkin, 2013). In recognising trauma as a shared event, healing too becomes something we do together.

Because TIAs are premised on the understanding that most of the people who come into contact with mental health services have been impacted by trauma, training, supervision and support for staff are seen as essential. This attention to staff support has the potential to decrease burnout and reduce staff turnover. For example, research suggests that supervisors who feel that their organisation values them and cares about their well-being are more likely to be supportive towards the people they are responsible for (Shanock and Eisenberger, 2006).

There are complex interactions between service users, practitioners and organisations that can come to mirror one another through “parallel processes” (Bloom, 2006). Trauma survivors’ lives may be organised around the trauma experience, just as systems can come to be organised around models that are inadequate for responding to survivors. This means that, for example, in trauma-organised systems, survivors may feel and be unsafe, leading to aggression towards staff. Experiencing aggression from survivors may cause staff to become wary and hostile, with organisations responding with greater punitive and risk-averse measures. This increases survivors’ sense of unsafety and aggression. Becoming trauma-informed has the potential to break these negative parallel processes and create positive interactions.

Trauma carries a heavy economic cost. Dolezal et al. (2009) have reviewed US research evidence on the economic impacts of violence and abuse and estimate a cost of between 17 and 37.5 per cent of the total spend on healthcare. They believe that a compassionate healthcare system that understands the impacts of violence and abuse and offers appropriate support may avoid many of these costs. In the UK, the Department of Health (DH) has estimated that:

There is also some evidence that a reduction in seclusion and restraint has large cost savings (e.g. a 92 per cent reduction in the costs linked to restraint, LeBel and Goldstein, 2006).

Trauma-informed mental health services are strengths based: they reframe complex behaviour in terms of its function in helping survival and as a response to situational or relational triggers. Reframing refers to looking at, presenting, and thinking about a phenomenon in a new and different way, and replaces traditional individual/medical model approaches to madness and distress with a social perspective, somewhat akin to the Social Model of Disability (Wilson and Beresford, 2002). Reframing behaviour as meaningful allows providers to address underlying needs and utilise less intrusive strategies. We have fictionalised a trauma-informed response to a woman who self-harms in Box 1.

In a trauma-informed mental health service, all staff – clinical and non-clinical – understand the impact of trauma on a person’s ability to survive in the present moment. Crucially, this entails a shift from thinking “what is wrong with you” to “what happened to you” (Harris and Fallot, 2001). The critical roles of racism, sexism, homophobia, ageism, poverty and their intersectionalities are recognised. Survivors in crisis are not viewed as manipulative, attention-seeking or destructive, but as trying to cope in the present moment using any available resource.

Providers do not fear asking about trauma, yet do so in ways that are respectful of potential retraumatisation; the power of telling one’s story but also the impotency of telling it where nothing changes (Filson, 2011); the need to move at the survivor’s pace; the need to truly listen; and the need for post-disclosure support. Survivors are forewarned about trauma questions, and can choose not to answer. Trauma information is integrated into treatment plans so that people can be referred to trauma-specific services (if wanted) (see Read et al., 2007 for a full account of why, when and how to ask about abuse).

The basic safety of environments is prioritised – physical, psychological, social and moral – with organisations making a commitment to non-violence (Bloom, 2006). Staff receive support to help them focus on trauma, and steps are taken to build a sense of community and shared responsibility between staff and survivors (Bloom, 2006). This means that services prioritise building trusting, mutual relationships between staff and survivors. When relationships are prioritised, policies and procedures (such as time limited sessions with a therapist) can be re-evaluated in light of whether or not they support TIAs.

TIAs in mental health aim to reduce or eradicate coercion and control, including medication as restraint, verbal coercion, threats of enforced detention, withholding information, restrictive risk-aversive practices, disrespectful and infantilising interactions and Community Treatment Orders (see, for instance O’Hagan, 2003). Clinicians understand the revictimisation that “power over” relationships reinforce. Training and supervision provide staff with the tools to attend to potential relational and situational triggers and to use trust-based, collaborative relationships to support people.

Survivors often encounter numerous human services across their lives. To be trauma-informed, each service within and beyond the local mental health system should operate according to TIA principles. This includes primary care, A&E, talking therapies, mental health teams, crisis care, the police, social services and voluntary sector services (such as trauma-specific service providers).

To provide an overview of the current state of evidence on the effectiveness of TIAs in mental health we searched nine electronic databases (Medline, Embase, PsycInfo, CINAHL, Cochrane Library, Sociologicial Abstracts, Social Policy and Practice, Global Health and Maternity and Infant Care) using the title-word search “trauma and informed”. Searches were from the earliest date of each database to August 2014 (searches run on 5 August 2014). This yielded 129 unique publications. One author (SC) assessed the items against the following inclusion criteria: participants – adults and/or youth in receipt of mental health services; intervention – any form of trauma-informed care; control (any or none); outcomes – any, including those relating to staff as well as service users; and study design – any design providing evidence on effectiveness, including qualitative evaluation studies.

Eight studies (Azeem et al., 2011; Chandler, 2008; Domino et al., 2006; Gatz et al., 2007; Greenwald et al., 2012; Messina et al., 2014; Morrissey et al., 2005; Weissbecker and Clark, 2007) met the inclusion criteria. The findings are presented in Table II.

All studies were conducted in the USA, and four were evaluations of Women and Co-Occurring Disorders services. Four studies were controlled pre-post-studies, two were pre-post-studies and one was a qualitative study. Beneficial effects noted in these studies included reduction in seclusion, reduced post-traumatic stress symptoms and general mental health symptoms, increased coping skills, improved physical health, greater treatment retention and shorter inpatient stays (see Table II, main findings). Other outcomes did not change such as substance misuse, emergency room use, imprisonment and shelter use.

Several of the studies were large, multi-site and quasi-experimental. The evidence-base is limited, however, by the relatively small number of studies, restricted to one country, with half the studies evaluating one particular TIA model. We did not locate any randomised controlled trials of trauma-informed mental healthcare. The findings cannot necessarily be generalised beyond the USA where the evidence is located. Also, as noted by Greenwald et al. (2012), the inclusion of numerous interventions makes it difficult to precisely identify the causes of the improvement. Integrating trauma-specific services within a broader trauma-informed service is advocated from a theoretical and service perspective; therefore, as is the case with complex interventions, research may not be able to pinpoint the key active ingredients.

This narrative overview of the evidence has limitations. It is not a systematic review as to conduct one was beyond the scope and intention of this paper. Consequently, although the number of bibliographic databases searched was large, the search strategy was basic, we did not search the grey literature or use reference checking or consultation with experts, and the selection of studies was undertaken by one investigator. A future systematic review on the effectiveness of trauma-informed care generally or in a mental health context would be welcome and illuminating.

Future research will hopefully provide a fuller picture of what TIAs are able to achieve. Given the sound theoretical and ethical underpinnings of TIAs and the extensive developmental work undertaken, coupled with the current promising evidence to date, there is certainly a strong case for the wider implementation and evaluation of TIAs.

The concept of TIAs originated in the USA and in 2005 the United States Federal Substance Abuse and Mental Health Services Administration (SAMHSA) established a National Centre for Trauma-Informed Care. Many human service providers, including those in mental health, are now familiar with the concept of TIAs. However, the US mental health system remains heavily biomedical and despite conceptual familiarity, implementation of TIAs across sectors is patchy (Cheryl Sharp, Senior Advisor for Trauma-Informed Services, National Council for Behavioural Health, USA, personal communication). In terms of strategies and implementation, the International Initiative for Mental Health Leadership has produced a brief report on key national and regional activities in TIAs across nations involved with the IIMHL as part of their “Make it so” series (IIMHL, 2012). As the aim of this series is to rapidly share information, it is likely that some key activities have been missed. Notwithstanding this caveat, countries where trauma-informed practices were identified were the USA, Canada, Australia and New Zealand. Nascent strategies and practices were described for Scotland and Ireland with nothing noted for England. The USA was described as leading the world in TIAs, and as the only nation to have national policy relating to trauma. Specifically, in 2011-2014 the eight strategic objectives of the SAMHSA included “trauma and justice”. This objective aimed to reduce the impacts of violence and trauma by integrating TIAs throughout health, behavioural health and related systems.

Of the 129 academic publications with “trauma” and “informed” in the title found when we conducted our overview of the evidence of effectiveness, the vast majority of publications (86 per cent) were from the USA. Five other countries, including the UK, had a small amount of academic activity. The UK work comprised two discussion papers (Rose et al., 2012; Ardino, 2014), a book (Taylor, 2012) and two reviews (Harragan, 2013; Steckley, 2013) of a non-UK-authored book on TIA. The focus was typically on residential treatment or mental health services for children and adolescents, with two general mental health papers (Ardino, 2014; Rose et al., 2012).

Academic publication rates may have little bearing on the provision of services, therefore to gain a wider perspective on the current UK situation, we conducted a Google search on UK pdf documents with “trauma-informed” in the title. This yielded information about:

• a trauma-informed foster care service in North Wales, Chester and the Wirral (www.newfocas.co.uk/includes/spaw2/uploads/files/trauma.pdf);

• a guide to trauma-informed resettlement for people leaving youth custody (www.beyondyouthcustody.net/wp-content/uploads/BYC-Developing-trauma-resettlement-youth-custody-practitioners-guide.pdf);

• workshops led by visiting North American experts (http://stephaniecovington.com/assets/files/Brochure-Flyer-dated-2-26-14-Woman-at-Risk-Flyer-V1.pdf; www.vptorg.co.uk/whoslistening/docs/conf%20speeches/Aimee_Neri_Presentation.pdf; www.stir.ac.uk/trauma14/);

• conferences, including two on mental health (www.makingwaves.org/news/trauma-informed-services-conference/; www.stir.ac.uk/trauma14/; www.bild.org.uk/our-services/events/supportingpeople/); and

• training with a trauma-informed approach (www.ukpts.co.uk/site/assets/CALM-Final.pdf; www.calmtraining.co.uk/index.php; www.ewrasac.org.uk/Training/; www.avaproject.org.uk).

One of us, (AK), has played a key role in introducing TIAs to Tees, Esk and Wear Valleys NHS Foundation Trust, and describes her experiences in Box 2.

It is clear that TIAs are beginning to reach the UK, although often in settings beyond mental health. However, the two conferences on trauma-informed mental healthcare in 2014 – with speakers from psychology, mental health nursing, psychiatry and the survivor movement – indicate the beginnings of a sea-change. Scotland’s Mental Health Strategy 2012-2015 includes psychological trauma as a key priority (Scottish Government, 2012). The strategy states that “General Services should be Trauma Aware”, and aims to improve recognition and awareness of trauma in Primary Care and Mental Health Services, encourage staff to make appropriate referrals for trauma survivors, and roll out trauma training. Although TIAs are not named, this is nevertheless a welcome development.

Similarly, the National Mental Health Development Unit (2010) and DH (2011) have released strategy documents on gender sensitive services that include trauma awareness. The DH published recommendations regarding routine enquiry of abuse in mental health settings over a decade ago (DH, 2003) and a programme of work was undertaken to train staff, which demonstrated changes in skill (McNeish and Scott, 2008). This focused on changing the emphasis from “What is wrong with this person?” to “What has happened to this person?”. Asking the basic question: “Have you ever experienced physical, sexual or emotional abuse at any time in your life?” has now become mandatory for UK services. However, current evidence that staff do this in practice is scant and this suggests that good practice that goes beyond this question is not widespread (Hepworth and McGowan, 2013; Brooker et al., 2016). One significant change that may prompt responses from services is the inclusion of trauma in some NICE guidelines, for example, the recently updated guidance for the management of schizophrenia (NICE, 2014). Some early intervention services for psychosis, in particular, are attempting to be more trauma-informed. Toner et al. (2013) showed that having a formulation-driven approach to understanding psychosis was more important in creating staff that were empowered to address trauma than having the ability to enquire about it. There is something very important about the model of mental health that staff bring with them to the role.

We have identified a number of potential explanations for the slow implementation of TIAs in the UK, although our list is not exhaustive. Many of these implementation barriers are applicable to settings beyond the UK. First, despite compelling evidence, there remains strong resistance to the notion that trauma and childhood abuse plays a causal role in psychosis and mental distress. Historically, such claims have been seen as “family blaming”, and have been vehemently opposed, e.g. historic opposition to Freud and Laing. Instead there is a focus on the biological basis of mental distress, with genes and neurology seen as causal and trauma relegated to a trigger at best (Moskowitz, 2011). Thus, mental distress is understood as a scientific, medical and pharmacological problem, rather than a human, familial or social issue.

Second, Western societies have strongly resisted notions of historical and cultural violence and their consequent trauma legacies. Jackson (2003), an African American survivor and therapist, has produced a powerful research account of scientific racism, slavery and colonialism and the impact this has had on survivors generationally and today. Focussing on the social and systemic causes of trauma places practitioners in opposition to powerful groups and consequently is often avoided (Coles, 2014).

Third, Coles (2014) has described “horror” as a barrier to practitioners embracing notions of trauma: “to stand as witness to the extent and horror of people’s accounts of pain and suffering is to encounter and experience fear, despair, loss and rage”.

Fourth, UK public services face continuous change and upheaval, making many wary and weary of new initiatives. Consequently, introducing new conceptualisations of care can be challenging, and this is particularly acute with TIAs because the role and prevalence of trauma is disputed (e.g. the DH and NICE focus on diagnostic categories, rarely referring to trauma). Compounding this, UK austerity means that resources are scarcer and morale lower. This context makes it harder to engage with new initiatives.

Fifth, TIAs are a relatively complex and involved approach to service provision, and are easily confused with trauma-specific services. Muskett (2014) has described how mental health nurses in Australia struggle to translate TIA principles into their everyday practice beyond reducing seclusion and restraint.

Sixth, there have been a number of initiatives aimed at improving mental health services and relationships between service providers and users. For example, in the UK, Star Wards aims to support excellence on inpatient psychiatric wards (www.starwards.org.uk), Safewards aims to reduce conflict and containment and increase safety on inpatient mental health wards (www.safewards.net/), and Compassion in Practice centralises the six Cs of nursing and midwifery (care, compassion, competence, communication, courage and commitment). Whilst such initiatives are compatible with TIA, they are nonetheless another way to conceptualise and implement care for providers to grapple with.

Seventh, many UK mental health staff have no access to regular structured supervision, and this is a serious barrier to the implementation of TIA. In our case study (Box 3) we cite trauma-specific supervision groups as a way of supporting therapists to respond to issues of complex trauma.

Finally, once a concept starts to take hold it can gain momentum. Debate, training opportunities, champions, mentors and networking all perpetuate thinking and practice. Our mapping work suggests that despite evidence of increasing interest in TIAs in the UK, we have not yet achieved the critical mass needed for frontline TIA implementation.

Addressing each of the issues outlined above will go some way towards contributing to the development of UK TIA practice. First and foremost, a paradigm shift in collective thinking about the causes of mental distress is vital (Harris and Fallot, 2001). Practitioners must move from asking “what is wrong with you” to “what happened to you” (Harris and Fallot, 2001). In other words, practitioners must understand the critical and primary role of trauma and fundamentally change their practice as a result. Without this, TIAs are at high risk of co-optation (as has arguably occurred with the concepts of recovery and peer support), meaning that mainstream implementation could be tokenistic, fragmentary and divorced from TIAs core principles.

Although individual practitioners can engage with people in trauma-informed ways, this will be inadequate without system-wide change as systems, as well as individual managers, can block what is needed to effect change (Kotter and Cohen, 2003). Rose et al. (2012) assessed DH documentation for its fit with TIA principles. They found that it is policy that trauma and abuse is discussed and documented in all mental health assessments. The principles of TIA are also consistent with policy advocating partnership working between survivors and providers, such as choice and coproduction. However, further policies are needed so that services can move away from force, coercion and risk-aversive practice and towards TIAs, with trusts incentivised to implement change.

AK played a key role in introducing TIAs to her mental health trust, and Box 3 explores some of the key factors that made this successful. As can be seen, it was particularly important to demonstrate that TIAs fitted with and furthered key organisational objectives.

Alongside a paradigm shift, there must be discussion and acknowledgement of the critical roles of historical and cultural violence, including ethnically and socio-demographically based differential experiences within the mental health system (e.g. Morgan et al., 2004; McKenzie and Bhui, 2007).

We must combat the view that TIAs are utopian because survivors are dangerous and in need of compulsion (e.g. Muskett, 2014). We must recognise the cycle of crisis that a focus on risk management perpetuates as people struggle for personal agency, choice and control over their lives (Samele et al., 2007). In TEWV, positive risk-taking is policy, including understanding some risky behaviours as survival strategies.

TIA is a complex concept, and opportunities to acquire and develop knowledge are needed. The two 2014 UK conferences on trauma-informed mental healthcare have helped create such opportunities. Yet knowledge alone is insufficient: as Kotter and Cohen (2003) argue, “Without conviction that you can make change happen, you will not act, even if you see the vision”. Support for implementation is also crucial. It is estimated that it takes 10-15 years for new healthcare innovations to be incorporated into routine clinical practice (e.g. Proctor et al., 2009). This gap between research and its implementation is referred to as the translational gap (Tansella and Thornicroft, 2009). Implementation science aims to bridge this translational divide. Thus, in bringing TIA to the UK, implementation science will have an important role to play in supporting individuals and organisations to enact change. Whilst implementation science is beyond the scope of this paper to describe (see, e.g. Damschroder et al., 2009), key ingredients might include implementing a rewards and recognition scheme for staff (Kotter and Cohen, 2003; Tansella and Thornicroft, 2009); understanding current organisational culture and the shifts needed to achieve change (Damschroder et al., 2009); and providing case studies of successful implementation to combat hopelessness and bolster confidence that change is possible (Kotter and Cohen, 2003). Specific to TIA, Harris and Fallot (2009) have developed a self-assessment and planning protocol which supports an organisation’s implementation of TIAs. They argue that if the principles of TIA are reflected in the culture of an organisation for each contact, physical setting, relationship and activity for survivors and staff, the organisation is trauma-informed. Several key steps for moving towards this culture are described in Table III, and Table IV contains a summary of useful resources. In Box 2, AK described her experiences of bringing TIA to TEWV. Key steps in implementing TIA in TEWV have included designing a trauma-informed pathway, training staff, conducting evaluations, developing written guidelines for stakeholders and promoting ownership at senior levels; this implementation can take ten years (Brooker et al., 2016).

Once a critical mass develops, it will become easier for people to model TIA, mentor others, create networks, identify trauma champions and share ideas (Turner, 1990). We believe that trauma survivors have a pivotal role to play in this. We live the impact of trauma everyday. We understand its devastating effects, the damage inflicted by the current mental health system, the need for mutual relationships based on safety and cooperation, the need for personal control, and the vital support of peers. In bringing TIAs to the UK, we need survivor leaders and champions advocating for values-based system change with passion and commitment. Our hope is that this vision will become a reality and that this discussion paper will have contributed to this.

Angela Sweeney is funded by a National Institute for Health Research Post-Doctoral Fellowship. This paper presents independent research partially funded by the National Institute for Health Research (NIHR). The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health.